CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 Basic Concepts 2

03 Chapter 1-Coloring 3–6

04 Chapter 2-Chromatic Number 7 – 16

05 Chapter 3-Applications Of Graph Colouring 17 – 19

06 Conclusion 20
07 Bibliography 21
 INTRODUCTION

A proper coloring of a graph is an assignment of colors to the vertices of the graph so that

no two adjacent vertices have the same color.

Usually we drop the word “proper” unless other types of coloring are also under

discussion. Of course, the “colors” don’t have to be actual colors ; may can be any distinct

labels - integers ,for examples , if a graph is not connected , each connected component can

be colored independently; except where otherwise noted , we assume graphs are

connected. We also assume graphs are simple in this section. Graph coloring has many

applications in addition to its intrinsic interest.

In the same way the most important concept of graph coloring is utilized in

resource allocation, scheduling. Also, paths, walks and circuits in graph theory are used in

tremendous applications say travelling salesman problem, database design concepts,

resource networking.

This project deals with coloring which is one of the most important topics in

graph theory. In this project there are three chapters. First chapter is coloring . The second

chapter is chromatic number. The last chapter deals with application of graph coloring.

1
 BASIC CONCEPTS

1. GRAPH
A graph is an ordered triplet. G=(V(G), E(G), I(G)); V(G) is a non empty set, E(G) is a set

disjoint from V(G) and I(G) is an incidence map that associates each element of E(G) and

unrecorded pair of element of V(G). The elements of V(G) are called vertices (or nodes or

points) of G and the elements of E(G) are Called edges or lines of G.

2. MULTIPLE EDGE / PARALLEL EDGE

A set of 2 or more edges of a graph G is called a multiple edge or parallel edge if they

have the same end vertices.

3. LOOP
An edge for which the 2 end vertices are same is called a loop.

4. SIMPLE GRAPH
A graph is simple if it has no loop and no multiple edges.

5. DEGREE
Let G be a graph and v € V the number of edge incident at V in G is called the degree or

vacancy of the vertex v in G.

2
 CHAPTER - 1

COLORING

Graph coloring is nothing but a simple way of labeling graph components such as

vertices , edges and regions under some constraints. In a graph, no two adjacent vertices,

adjacent edges , or adjacent regions are colored with minimum number of colors .This

number is called the chromatic number and the graph is called properly colored graph.

In graph theory coloring is a special case of graph labeling; it is

an assignment of labels traditionally called “colors” to elements of a graph subject to certain

constraints. In it simplest form, it is a way of coloring the vertices of a graph such that no

two adjacent vertices share the same color, it is called vertex coloring. Similarly, edge

coloring assigns a color to each edge so that no two adjacent edges share the common

color.

While graph coloring , the constraints that are set on the graph are

colors , order of coloring , the way of assigning color , etc. A coloring is given to a vertex or a

particular region . Thus, the vertices or regions having same colors form independent sets.

3
VERTEX COLORING
Vertex coloring is an assignment of colors to the vertices of a graph ‘G ’ such

that no two adjacent vertices have the same color .Simply put , no two vertices of an edge

should be of the same color.

The most common type of vertex coloring seeks to minimize the

number of colors for a given graph . Such a coloring is known as a minimum vertex coloring ,

and the minimum number of colors which with the vertices of a graph may be colored is

called the chromatic number .

CHROMATIC NUMBER:
The minimum number of colors required for vertex coloring of graph ‘ G ’

is called as the chromatic number of G , denoted by X (G) .

X(G) = 1 iff ‘ G ’ is a null graph. If ‘G ’ is not a null graph , then X(G) ≥ 2.

EXAMPLES;

1. 2.

Null Graph ( X (G) = 1 ) Not Null Graph ( X (G) = 2 )

4
EDGE COLORING
An edge coloring of a graph G is a coloring of the edges of G such that adjacent

edges ( or the edges bounding different regions ) receive different colors. An edge coloring

containing the smallest possible number of colors for a given graph is known as a minimum

edge coloring.

The edge chromatic number gives the minimum number of colours with which
graph’s edges can be colored.

CHROMATIC INDEX
The minimum number of colors required for proper edge coloring of graph is

called chromatic index.

A complete graph is the one in which each vertex is directly connected with all

other vertices with an edge. If the number of vertices of a complete graph is n, then the

chromatic index for an odd number of vertices will be n and the chromatic index for even

number of vertices will be n-1.

5
EXAMPLES;
1.

The given graph will require 3 unique colors so that no two incident edges have the

Same color. So its chromatic index will be 3.

2.

The given graph will require 2 unique colors so that no two incident edges have

the same color. So its chromatic index will be 2.

6
 CHAPTER 2

Chromatic Number

The chromatic number of a graph is the smallest number of colors needed to color the vertices
of so that no two adjacent vertices share the same color. That is the smallest value of possible
to obtain a k-coloring.

• Graph Coloring is a process of assigning colors to the vertices of a graph.

• It ensures that no two adjacent vertices of the graph are colored with the same color.
• Chromatic Number is the minimum number of colors required to properly color any graph.

Graph Coloring Algorithm

• There exists no efficient algorithm for coloring a graph with minimum number of colors.

However, a following greedy algorithm is known for finding the chromatic number of any given
graph.

Greedy Algorithm

Step-01:

Color first vertex with the first color.

7
Step-02:

Now, consider the remaining (V-1) vertices one by one and do the following-

• Color the currently picked vertex with the lowest numbered color if it has not been used to
color any of its adjacent vertices.
• If it has been used, then choose the next least numbered color.
• If all the previously used colors have been used, then assign a new color to the currently
picked vertex.

Problems Based On Finding Chromatic Number of a Graph

Problem-01:

Find chromatic number of the following graph-

8
Solution-

Applying Greedy Algorithm, we have

Vertex a B C d e f

Color C1 C2 C1 C2 C1 C2

From here,

• Minimum numbers of colors used to color the given graph are 2.

• Therefore, Chromatic Number of the given graph = 2.

The given graph may be properly colored using 2 colors as shown below-

9
Problem-02:

Find chromatic number of the following graph-

Solution-

Applying Greedy Algorithm, we have-

Vertex a b C d e f

Color C1 C2 C2 C3 C3 C1

From here,

• Minimum numbers of colors used to color the given graph are 3.

• Therefore, Chromatic Number of the given graph = 3.

10
The given graph may be properly colored using 3 colors as shown below-

Chromatic Number of Graphs

Chromatic Number of some common types of graphs are as follows-

1. Cycle Graph-

• A simple graph of ‘n’ vertices (n>=3) and ‘n’ edges forming a cycle of length ‘n’ is called as a
cycle graph.
• In a cycle graph, all the vertices are of degree 2.

Chromatic Number
• If number of vertices in cycle graph is even, then its chromatic number = 2.
• If number of vertices in cycle graph is odd, then its chromatic number = 3.

11
Examples-

2. Planar Graphs-

A planar graph is a graph that can be embedded in the plane, that is it can be drawn on the
plane in such a way that its edges intersect only at their endpoint. In other words, it can be
drawn in such a way that no edges cross each other.

12
A Planar Graph is a graph that can be drawn in a plane such that none of its edges cross each
other.

Chromatic Number
Chromatic Number of any Planar Graph is less than or equal to 4

Examples-
+

• All the above cycle graphs are also planar graphs.

• Chromatic number of each graph is less than or equal to 4.

13
3. Complete Graphs-

• A complete graph is a graph in which every two distinct vertices are joined by exactly one
edge.
• In a complete graph, each vertex is connected with every other vertex.
• So to properly it, as many different colors are needed as there are number of vertices in the
given graph.

Chromatic Number
Chromatic Number of any Complete Graph

= Number of vertices in that Complete Graph

Examples-

14
4. Bipartite Graphs-

A bipartite graph is a graph whose vertices can be divided into two disjoint and independent
sets U and V such that every edge connects a vertex in U to one in V. Vertex sets U and V are
usually called the parts of the graph.

• A Bipartite Graph consists of two sets of vertices X and Y.

• The edges only join vertices in X to vertices in Y, not vertices within a set.

Chromatic Number
Chromatic Number of any Bipartite Graph

=2

Example-

15
5. Trees-

A tree is an undirected graph in which any two vertices are connected by exactly one path, or
equivalently a connected acyclic undirected graph.

• A Tree is a special type of connected graph in which there are no circuits.

• Every tree is a bipartite graph.
• So, chromatic number of a tree with any number of vertices = 2.

Chromatic Number
Chromatic Number of any tree

=2

Examples-

16
 CHAPTER-3
APPLICATIONS OF GRAPH COLORING

1) Making Schedule or Time Table:

Suppose we want to make an exam schedule for a university. We have list
different subjects and students enrolled in every subject. Many subjects would have common
students (of same batch, some backlog students, etc). How do we schedule the exam so that no
two exams with a common student are scheduled at same time? How many minimum time
slots are needed to schedule all exams? This problem can be represented as a graph where
every vertex is a subject and an edge between two vertices mean there is a common student.
So this is a graph coloring problem where minimum number of time slots is equal to the
chromatic number of the graph.

2) Mobile Radio Frequency Assignment:

When frequencies are assigned to towers, frequencies assigned to all towers at the
same location must be different. How to assign frequencies with this constraint? What is the
minimum number of frequencies needed? This problem is also an instance of graph coloring
problem where every tower represents a vertex and an edge between two towers represents
that they are in range of each other.

3) Register Allocation:
In compiler optimization, register allocation is the process of assigning a large number
of target program variables onto a small number of CPU registers. This problem is also a graph
coloring problem.

4) Sudoku:
Sudoku is also a variation of Graph coloring problem where every cell represents a
vertex. There is an edge between two vertices if they are in same row or same column or same
block.

17
 5) Map Coloring:

Geographical maps of countries or states where no two adjacent cities cannot be

assigned same color. Four colors are sufficient to color any map.

6) Bipartite Graphs:

We can check if a graph is bipartite or not by coloring the graph using two colors. If a
given graph is 2-colorable, then it is Bipartite, otherwise not. See this for more details.

Explanation;

Algorithm:
A bipartite graph is possible if it is possible to assign a color to each vertex such that no
two neighbour vertices are assigned the same color. Only two colors can be used in this
process.

Steps:
1. Assign a color (say red) to the source vertex.

2. Assign all the neighbours of the above vertex another color (say blue).

3. Taking one neighbour at a time, assign all the neighbour's neighbours the color red.

4. Continue in this manner till all the vertices have been assigned a color.

5. If at any stage, we find a neighbour which has been assigned the same color as that of the
current vertex, stop the process. The graph cannot be colored using two colors. Thus the graph
is not bipartite.

18
Example:

19
 CONCLUSION

This project aims to provide a solid background in the basic topics of graph coloring. Graph
coloring problem is to assign colors to certain elements of a graph subject to certain
constraints. The nature of coloring problem depends on the number of colors but not on what
they are.

The study of this topic gives excellent introduction to the subject called “Graph
Coloring”.

This project includes two important topics such as vertex coloring and edge coloring and came
to know about different ways and importance of coloring.

Graph coloring enjoys many practical applications as well as theoretical challenges.

Besides the applications, different limitations can also be set on the graph or on the away a color
is assigned or even on the color itself. It has been reached popularity with the general public in
the form of the popular number puzzle Sudoku and it is also use in the making of time
management which is an important application of coloring. So graph coloring is still a very
active field of research.

20
 BIBLIOGRAPHY

1) V.K Balakrishnan, outlines, graph theory.

2) J.A Bondy and U.S.R Murty, graph theory with applications.

3) S.Arumugam, S.Ramachandran, invitation to graph theory.

21
PLANARITY IN GRAPH THEORY

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
AMAL THOMAS
DB18CMSR11
Under the guidance of
Mrs. PRIJA V

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu
June 2021
 CERTIFICATE

Certified that this project ‘Planar Graph’ is a bona fide project of Amal Thomas
carried out the project work under my supervision.

Mrs.
Mrs. Riya
Riya Baby
Baby Mrs. Prija V
Head
Head of
of Department
Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu

I
 DECLARATION

I Amal Thomas hereby declare that the project ‘Planar Graph’ is an original
record of studies and bona fide project carried out by me during the period of
2018 – 2021 under the guidance of Mrs. Prija V, Department of Mathematics,
Don Bosco Arts and Science College, Angadikkadavu and has not submitted by
me elsewhere for the award of my degree, diploma, title, or recognition, before.

Amal Thomas

DBI8CMSR11

Department of Mathematics

Don Bosco Arts and Science College,

Angadikkadavu

II
 ACKNOWLEDGEMENT

Introduction is the proper place to begin. But first I bow my head before
the Almighty who is always with me. Also, I must express my deepest gratitude
to people along the way.

No words can adequately express the sense of gratitude; still, I try to

express my heartfelt thanks through words. At the outset, I am deeply indebted to
my project supervisor Mrs. Prija V, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement, and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the principal
and the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the
successful completion of this project. It is my pleasure and duty to thank each and
every one of them who walked with me.

My greatest debt is always, to God Almighty.

III
 CONTENTS

Sl No Title Page No

1 Introduction 1

2 Chapter 1 - Basic Concepts 2-9

3 Chapter 2 – Planar Graph 10 - 17

4 Chapter 3 – Euler’s Formula 18 – 19

5 Chapter 4 – Dual of a Plane Graph 20

6 Conclusion 21

7 Bibliography 22

IV
 INTRODUCTION

In recent years, Graph Theory has established itself as an important

mathematical tool in a wide variety of subjects, ranging from Operational
Research and Chemistry to Genetics and Linguistics, and from Electrical
Engineering and Geography to Sociology and Architecture. At the same time, it
has also emerged as a worthwhile mathematical discipline in its own right.

A great mathematician, Euler become the Father of Graph Theory, when

in 1736, he solved a famous unsolved problem of his days, called Konigsberg
Bridge Problem. This is today, called as the First Problem of the Graph theory.
This problem leads to the concept of the planar graph as well as Eulerian Graphs,
while planar graphs were introduced for practical reasons, they pose many
remarkable mathematical properties. In 1936, the psychologist Lewin used planar
graphs to represent the life space of an individual.

1
Chapter 1

BASIC CONCEPTS

Graph

A graph is an ordered triple 𝐺 = {𝑉(𝐺), 𝐸(𝐺), 𝐼𝐺 } where 𝑉(𝐺) is a non-

empty set, 𝐸(𝐺) is a set disjoint from 𝑉(𝐺) and 𝐼(𝐺) is an incidence map that
associates each element of 𝐸(𝐺) and unordered pair of elements of 𝑉(𝐺). The
elements of 𝑉(𝐺) are called vertices (or nodes or points) of 𝐺 and the element
of 𝐸(𝐺) are called edges or lines of 𝐺.

Example:

Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 }

𝐸(𝐺) = {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 }
𝐼𝐺 (𝑒1 ) = {𝑣1 , 𝑣2 } or {𝑣2 , 𝑣1 }
I𝐺 (𝑒2 ) = {𝑣2 , 𝑣3 } or {𝑣3 , 𝑣2 }
𝐼𝐺 (𝑒3 ) = {𝑣3 , 𝑣4 } or {𝑣4 , 𝑣3 }
𝐼𝐺 (𝑒4 ) = {𝑣4 , 𝑣1 } or {𝑣1 , 𝑣4 }

Multiple edges

A set of two or more edges of a graph 𝐺 is called multiple edges or parallel

edges if they have the same end vertices.

2
Loop

An edge for which the two end vertices are same is called a loop.

Here {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 } form the parallel edges.

𝑒7 is the Loop.
Simple Graph

A graph is simple if it has no loops and no multiple edges.

Finite & Infinite Graphs

A graph is called finite if both 𝑉(𝐺) & 𝐸(𝐺) are finite. A graph that is not
finite is called infinite graph.

Adjacent Vertices

Two vertices 𝑢 and 𝑣 are said to be adjacent vertices if and only if there is
an edge with 𝑢 and 𝑣 as its end vertices.

3
Adjacent Edges

Two distinct edges are said to be adjacent edges if and only if they have a
continuous end vertex.

Complete Graph

A simple graph 𝐺 is said to be complete if every pair of distinct vertices of

𝐺 are adjacent in 𝐺. A complete graph with n vertices is denoted by 𝐾𝑛 .

Bipartite Graph

A graph is bipartite if its vertex set can be partitioned into two non-empty
subsets 𝑋 and 𝑌 such that each edge of 𝐺 has one end in 𝑋 and the other in 𝑌. The
pair (𝑋, 𝑌) is called a bipartition of the bipartite graph 𝐺. The bipartite graph 𝐺
with bipartition (𝑋, 𝑌) denoted by 𝐺 (𝑋, 𝑌).

4
Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }
The Bipartition is
𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }

Complete Bipartite Graph

A simple bipartite graph 𝐺 (𝑋, 𝑌) is complete if each vertex 𝑋 is adjacent

to all the vertices of 𝑌.

Here 𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 }

Subgraph

A graph 𝐻 is called subgraph of 𝐺 if 𝑉(𝐻) ⊆ 𝑉(𝐺), 𝐸(𝐻) ⊆ 𝐸(𝐺) and IH

is the restriction of 𝐼𝐺 to 𝐸(𝐻) [ie, 𝐼𝐻 (𝑒) = 𝐼𝐺 (𝑒) whenever 𝑒 ∈ 𝐸(𝐻).

5
Degrees of Vertices

The number of edges incident with vertex 𝑉 is called degree of a vertex

or valency of a vertex and it is denoted by 𝑑(𝑣).

Isomorphism of Graph

A graph isomorphism from a graph 𝐺 to a graph 𝐻 is a pair (𝜙, 𝜃), where

𝜙 ∶ 𝑉(𝐺) → 𝑉(𝐻) and 𝜃 ∶ 𝐸(𝐺) → 𝐸(𝐻) are bijection with a property that
𝐼𝐺 (𝑒) = {𝑢, 𝑣} and 𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}.

Walk

A walk in a graph 𝐺 is an alternative sequence 𝑊=

𝑣0 𝑣1 𝑒1 𝑣2 𝑒2 … 𝑣𝑛 𝑒𝑛 vertices and edges, beginning and ending with vertices where
𝑣0 is the origin and 𝑣𝑛 is the terminus of 𝑊.

𝑊 = 𝑣6 𝑒8 𝑣1 𝑒1 𝑣2 𝑒2 𝑣3 𝑒3 𝑣2 𝑒1 𝑣1

6
Closed Walk

A walk to begin and ends at the same vertex is called a closed walk. That
is, the walk 𝑊 is closed if 𝑣0 = 𝑣𝑛 .

Open Walk

If the origin of the walk and terminus of the walk are different vertices,
then it is called an open walk.

Trail

A walk is called a trail if all the edges in the walk are distinct.

Path

A walk is called a path if all the vertices are distinct.

Example:

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒6 𝑣1 → A trail

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 → A path

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 𝑒5 𝑣1 → A trail, but not a path

Euler’s Theorem

The sum of the degrees of the vertices of a graph is equal to the twice the
number of edges.

ie: ∑𝑛𝑖=1 𝑑(𝑣𝑖 ) = 2𝑚

7
Isomorphic Graph

𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}

A graph 𝐺1 = (𝑉1 , 𝐸1 ) is said to be isomorphic to graph 𝐺2 = (𝑉2 , 𝐸2 ) if

there is a one-to-one correspondence between the edge sets 𝐸1 and 𝐸2 in such a
way that if 𝑒1 is an edge with end vertices 𝑢1 and 𝑣1 in 𝐺1 then the corresponding
edge 𝑒2 in 𝐺2 has its end vertices 𝑢2 and 𝑣2 in 𝐺2 . This correspondence is called
a graph isomorphism.

Example:

𝐺=

𝐻=

ie: G and H are isomorphic.

Components

A connected component of a graph is a maximal connected subgraph. The

term is also used for maximal subgraph or subset of a graph 's vertices that have
some higher order of connectivity, including bi-connected components, tri-
connected components and strongly connected components.

Tree

A connected graph without cycles is called a tree.

Vertex Cut

Let 𝐺 be a connected graph. The set 𝑉՛ subset of 𝑉(𝐺) is called a Vertex

cut of 𝐺, if 𝐺 − 𝑉՛ is a disconnected graph.

8
Cut Vertex

If 𝑉՛ = {𝑣} is a Vertex cut of the connected Graph 𝐺, then the vertex 𝒗 is

called a Cut vertex.

Edge Cut

Let 𝐺 be a non-trivial connected graph with vertex set 𝑉 and let 𝑆 be a non-
empty subset of 𝑉 and 𝑆̅ = 𝑉 − 𝑆. Let 𝐸՛ = [𝑆, 𝑆̅] denote the set of all edges of 𝐺
that have one end vertex is 𝑆 and the other is 𝑆̅. Then 𝐺 − 𝐸՛ is a disconnected
graph and 𝐸՛ = [𝑆, 𝑆̅] is called an edge cut of 𝐺.

Cut Edge

If 𝐸՛ = {𝑒} is an edge cut of 𝐺 then 𝑒 is called a cut edge of 𝐺.

Block

A block is a Connected graph without any cut vertices.

Eg:

Graph 𝐺 Blocks of 𝐺

9
Chapter 2

PLANAR GRAPHS

Plane Graph

A plane graph is a graph drawn in the plane, such a way that any pair of
edges meet only at their end vertices.

Example:

Planar Graph

A planar graph is a graph which is isomorphic to a plane graph, ie: it can

be drawn as a plane graph.

A plane graph is a graph that can be drawn in the plane without any edge crossing.

10
Example of Planar graph:

Planar Representation

The pictorial representation of a planar graph as a plane graph is called a

planar representation.

Eg: Is Q3 shown below, planar?

The graph Q3

Planar representation of Q3 is:

11
Jordan Curve

A Jordan Curve in the plane is a continuous non-self-intersecting curve

where Origin and Terminals coincide.

Example:

Jordan Curves

Non-Jordan Curves

Remark

If J is a Jordan Curve in the plane, then the part of the plane enclosed by J
is called interior of J and is denoted by ‘int J’. We exclude from ‘int J’ the points
actually lying on J. Similarly, the part of the plane lying outside J is called the
exterior of J and is denoted by ‘ext J’.

Example:

Arc connecting point 𝑥 in int J with point 𝑦 in ext J.

12
Theorem

Let J be a Jordan Curve, if 𝑥 is a point in int J and 𝑦 is a point in ext J then

any line joining 𝑥 to 𝑦 must meet J at some point, ie: must cross J. this is called
Jordan Curve Theorem.

Boundary

The set of edges that bound a region is called its boundary.

Definition

A graph which is not planar is known as non-planar graph or a graph that

cannot be drawn in the plane without any edge crossing is known as non-planar
graph.

Theorem

K5 is nonplanar:

Every drawing of the complex graph K5 in the plane (or sphere) contains
at least one edge crossing.

Proof:

Label the vertices 0, 1, 2, 3, 4. By the Jordan Curve theorem any drawing of the
cycle (1, 2, 3, 4, 1) separates the plane into two regions. Consider the region with

13
vertex 0 in its interior as the ‘inside’ of the circle. By the Jordan Curve theorem,
the edges joining vertex 0 to each of its vertices 1, 2, 3 and 4 must also lie entirely
inside the cycle, as illustrated below.

Drawing most of the K5 in the plane

Moreover, each of the 3-cycles {0, 1, 2, 0}, {0, 2, 3, 0}, {0, 3, 4, 0} and {0, 4, 1, 0}
also separates the plane and hence the edges (2, 4) must also lie to the exterior of
the cycle {1, 2, 3, 4} as shown. It follows that the cycle formed by edges (2, 4),
(4, 0) and (0, 2) separates the vertices 1 and 3, again by Jordan Curve theorem.
Thus, it is impossible to draw edge (1, 3) without crossing an edge of that cycle.
So, it is proven that the drawing of the K5 in the plane contains at least one edge-
crossing.

Theorem

K33 is nonplanar:

Every drawing of the complete bipartite graph K33 in the plane (or sphere)
contains at least one edge crossing.

Proof:

Label the vertices of one partite set 0, 2, 4 and of the order 1, 3, 5. By the
Jordan Curve theorem, cycle {2, 3, 4, 5, 2} separates the plane into two regions,

14
and as in the previous proof (K5), we regard the region containing the vertex 0 as
the ‘inside’ of the cycle. By the Jordan Curve theorem, the edges joining vertex
0 to each of the vertices 3 and 5 lie entirely inside that cycle, and each of the cycle
{0, 3, 2, 5, 0} and {0, 3. 4, 5, 0} separates the plane, as illustrated below.

Drawing most of the K33 in the plane

Thus, there are 3 regions: the exterior of cycles {2, 3, 4, 5, 2} and the inside
of each of the other two cycles. It follows that no matter which region contains
vertex 1, there must be some even numbered vertex that is not in that region, and
hence the edge from vertex 1 to that even-numbered vertex would have to cross
some cycle edge.

Corollary

Subgraph of a planar graph is planar.

Definition

A plane graph partitions the plane into number of regions called faces.

Let G be plane graph. If x is a point on the plane which is not in G, ie: 𝑥 is not a
vertex of G or a point on any edge of G, then we define the faces of G containing
𝑥 to be the set of all points on the plane which can be reached from 𝑥 by a line
which does not cross any edge of G or go through any vertex of G.

15
The number of faces of a plane graph G denoted by 𝑓(𝑎) or simply 𝑓.

Each plane graph has exactly one unbounded face called the exterior face.

Here 𝑓(𝐺) = 4

Degree of faces

The degree 𝑑(𝑓) of a face 𝑓 is the number of edges with which it is

incident, that is the number of edges in the boundary of a face.

Cut edge being counted twice.

Eg:

𝑑(𝑓2 ) = 3
𝑑(𝑓1 ) = 4
𝑑(𝑓3 ) = 3

Theorem

A graph is planar if and only if each of its blocks is planar.

Proof:

If G is planar, then each of its blocks is planar since a subgraph of planar

graph is planar.

Conversely, suppose that each block of G is planar. We now use induction

on the number of blocks of G to prove the result. Without loss of generality, we
16
assume that G is connected. If G has only one block, then G itself is a block, and
hence G is planar.

Now suppose G has k planar blocks and that the result has been proved for
all connected graph having (k-1) planar blocks. Choose any end block B0 of G
and delete from G all the vertices of B0 except the unique cut vertex, say 𝑣0 of G
in B0. The resulting connected graph G` of G contains (k-1) planar blocks. Hence,
by the induction hypothesis G` is planar. Let G~` be plane embedded of G` such
that 𝑣0 belongs to the boundary of unbounded face, say 𝑓 `. Let B0~ be a plane
embedding of B0 in 𝑓 `, so that 𝑣0 is in the exterior face of B0~. Then G~` and B0~
is a plane embedding of G.

17
Chapter 3

EULER’S FORMULA

Theorems

Euler Formula:

For a connected plain graph 𝐺, 𝑛 − 𝑚 + 𝑓 = 2 where 𝑛, 𝑚, and 𝑓 denote

the number of vertices, edges and faces of 𝐺 respectively.

Proof:

We apply the induction on 𝑓.

If 𝑓 = 1 the 𝐺 is a tree and 𝑚 = 𝑛 − 1.

Hence 𝑛 − 𝑚 + 𝑓 = 2 and suppose that 𝐺 has 𝑓 faces.

Since 𝑓 ≥ 2, 𝐺 is not a tree and hence contains a cycle 𝐶. Let 𝑒 be an edge of 𝐶.

Then 𝑒 belongs to exactly 2 faces, say 𝑓1 and 𝑓2 and the deletion of 𝑒 from 𝐺
results in the formation of a single face from 𝑓1 and 𝑓2 . Also, since 𝑒 is not a cut
edge of 𝐺. 𝐺 − 𝑒 is connected.

Further the number of faces of 𝐺 − 𝑒 is 𝑓 − 1, number of edges in 𝐺 − 𝑒 is 𝑚 −

1 and number of vertices in 𝐺 − 𝑒 is 𝑛. So, applying induction to 𝐺 − 𝑒, we get
𝑛 − (𝑚 − 1) + (𝑓 − 1) = 2 and this implies that 𝑛 − 𝑚 + 𝑓 = 2. This
completes the proof of theorem.

Corollary 1

All plane embedding of a planar graph have the same number of faces.

18
Proof:

Since 𝑓 = 𝑚 − 𝑛 + 2 the number of faces depends only on 𝑛 and 𝑚 and not on

the particular embedding.

Corollary 2

If 𝐺 is a simple planar graph with at least 3 vertices, then 𝑚 ≤ 3𝑛 − 6.

Proof:

Without the generality we can assume that 𝐺 is a simple connected plane graph.
Since 𝐺 is simple and 𝑛 ≥ 3, each face of 𝐺 has degree at least 3. Hence if 𝑓
denote the set of faces of 𝐺 ∑𝑓𝜖𝐹 𝑑(𝑓) ≥ 3𝑓. But ∑𝑓𝜖𝐹 𝑑(𝑓) = 2𝑚.

2𝑚
Consequently 2𝑚 ≥ 3𝑓 so that 𝑓 ≤ .
3

2𝑚 2m
By the Euler formula 𝑚 = 𝑛 + 𝑓 − 2 now 𝑓 ≤ implies m ≤ n + ( ) − 2.
3 3

This gives. 𝑚 ≤ 3𝑛 − 6.

19
Chapter 4

DUAL OF A PLANE GRAPH

Definition

Let G be a plane graph. One can form out of G a new graph H in the
following way corresponding to each face f(g), take the vertex f* and
corresponding to each edge e(g), take an edge e*. Then edge e* joins vertices f*
and g* in H iff edge e is common to the boundaries of faces f and g in G. The
graph H is then called dual of G.

Example:

Plane graph and its Dual

20
 CONCLUSION

In this project we discussed the topic planar graph in graph theory.

We discussed about Euler formula and verified that some graphs are planar, and
some are non-planar. A related important property of planar graphs, maps and
triangulations is that they can be enumerated very nicely.

We also discussed about duality of a graph.in mathematical discipline of graph

theory, the dual graph of a plane graph G is a graph that has a vertex of each face
of G .it has many applications in mathematical and computational study.

In fact, graph theory is being used in our so many routine activities. For eg; using
GPS or google maps to determine a route based on used settings.

21
 BIBLIOGRAPHY

1) R. Balakrishnan and K. Ranganathan, A textbook of Graph Theory,

University Text

2) J. A. Bondy and U. S. R. Murty, Graph Theory with Great Britain: The

Macmillian Press Ltd. 1976

22
POWER SERIES SOLUTIONS AND
SPECIAL FUNCTIONS
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
ANAINA MANIYATH
DB18CMSR17
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Power Series’ is a bona fide project of ANAINA
MANIYATH carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I ANAINA MANIYATH hereby declare that the project ‘Power Series’ is an

original record of studies and bona fide project carried out by me during the
period of 2018 – 2021 under the guidance of Ms. Athulya P, Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu and has not
submitted by me elsewhere for the award of my degree, diploma, title or
recognition, before.

Name
ANAINA MANIYATH

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express my deepest gratitude to
people along the way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my
project supervisor Ms. Athulya P, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and
the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every
one of them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- SERIES SOLUTION OF FIRST 4-9

ORDER EQUATION

04 Chapter 2- SECOND ORDER LINEAR 9-18

EQUATION, ORDINARY POINTS

05 Conclusion 19

06 Bibliography 20
 INTRODUCTION
A power series is a type of series with terms involving a variable. Power series
are often used by calculators and computers to evaluate trigonometric,
hyperbolic, exponential and logarithm functions. So any application of these
kind of functions is a possible application of power series. Many interesting and
important differential equations can be found in power series.

1
 PRELIMINERY

A. An infinite series of the form

∞

∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ (1)
𝑛=0

is called a power series in x. The series

∞

∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛 = 𝑎0 + 𝑎1 (𝑥 − 𝑥0 ) + 𝑎2 (𝑥 − 𝑥0 )2 + ⋯
𝑛=0

is a power series in x – x0.

B. The series (1) is said to converge at a point x if the limit

𝑚

𝑙𝑖𝑚 ∑ 𝑎𝑛 𝑥 𝑛
𝑚→∞
𝑛=0

exists, and in this case the sum of the series is the value of this limit.
Radius of convergence: Series in 𝑥 has a radius of convergence 𝑅, where
0 ≤ 𝑅 ≤ ∞, with the property that the series converges if |𝑥| < 𝑅 and
diverges if |𝑥| > 𝑅. It should be noted that if 𝑅 = 0 then no 𝑥 satisfies
|𝑥| < 𝑅, and if 𝑅 = ∞ then no 𝑥 satisfies |𝑥| > 𝑅
𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | | , if the limit exists.
𝑛→∞ 𝑎𝑛+1

C. Suppose that (1) converges for |𝑥| < 𝑅 with 𝑅 > 0, and denote its sum
by f(x):
∞

𝑓(𝑥) = ∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑛=0

Then f(x) is automatically continuous and has derivatives of all orders for
|𝑥| < 𝑅.

2
D. Let f(x) be a continuous function that has derivatives of all orders for
|x|< R with R > 0. f(x) be represented as power series using Taylor’s
formula:
𝑛
𝑓 (𝑘) (0) 𝑘
𝑓(𝑥) = ∑ 𝑥 + 𝑅𝑛 (𝑥)
𝑘!
𝑘=0

where the remainder Rn (x) is given by

𝑓 (𝑛+1) (𝑥̅ ) 𝑛+1
𝑅𝑛 (𝑥) = 𝑥
(𝑛 + 1)!
for some point 𝑥̅ between 0 and x.

E. A function f(x) with the property that a power series expansion of

the form

𝑓(𝑥) = ∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛
𝑛=0

is valid in some neighbourhood of the point x0 is said to be analytic at

x0. In this case the an are necessarily given by
𝑓 (𝑛) (𝑥0 )
𝑎𝑛 =
𝑛!
and is called the Taylor series of f(x) at x0.

Analytic functions: A function f defined on some open subset U of R or C is

called analytic if it is locally given by a convergent power series. This means
that every a ∈ U has an open neighbourhood V ⊆ U, such that there exists
a power series with centre a that converges to f(x) for every x ∈ V.

3
 CHAPTER 1
SERIES SOLUTION OF FIRST ORDER EQUATION
We have studied to solve linear equations with constants coefficient but with
variable coefficient only specific cases are discussed. Now we turn to these
latter cases and try to find a general method to solve this. The idea is to assume
that the unknown function y can be explained into a power series. Our purpose
in this section is to explain the procedures by showing how it works in the case
of first order equation that are easy to solve by elementary methods.

Example 1: we consider the equation

𝑦ʹ = 𝑦
Consider the above equation as (1). Assume that y has a power series solution
of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |x| < R, R > 0

That is we assume that 𝑦 ʹ = 𝑦 has a solution that is analytic at origin. We have
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ ⋯
then
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ … ….
∴ (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 ⋯
= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
⇒ 𝑎1 = 𝑎0
𝑎1 𝑎0
2𝑎2 = 𝑎1 ⇒ 𝑎2 = =
2 2

4
 𝑎2 𝑎0 𝑎0
3𝑎3 = 𝑎2 ⇒ 𝑎3 = = =
3 2 ∙ 3 3!
𝑎3 𝑎0 𝑎0
4𝑎4 = 𝑎3 ⇒ 𝑎4 = = =
4 2 ⋅ 3 ⋅ 4 4!
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0 𝑎0
= 𝑎0 + 𝑎0 𝑥 + 𝑥 2 + 𝑥 3 + 𝑥 4 + ⋯
2 3! 4!
𝑥2 𝑥3 𝑥4
= 𝑎0 (1 + 𝑥 + + + + ⋯ )
2! 3! 4!
𝑦 = 𝑎0 𝑒 𝑥
To find the actual function we have 𝑦 ʹ = 𝑦
𝑑𝑦 𝑑𝑦
i.e., =𝑦 ⇒ = 𝑑𝑥
𝑑𝑥 𝑦
integrating
log 𝑦 = 𝑥 + 𝑐
i.e., 𝑦 = 𝑒 𝑥+𝑐 = 𝑒 𝑥 ⋅ 𝑒 𝑐
𝑦 = 𝑎0 𝑒 𝑥 , where a0 = ec , a constant.

Example 2: solve 𝑦 ′ = 2𝑥𝑦. Also find its actual solution.

Solution: 𝑦 ′ = 2𝑥𝑦 (1)
Assume that y has a power series of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |𝑥| < 𝑅, 𝑅 > 0

∞

We have 𝑦 = ∑ a𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

5
 = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 2𝑥(𝑎0 + 𝑎1 𝑥 +𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯ )
= 2𝑥𝑎0 + 2𝑥𝑎1 𝑥 + 2𝑥𝑎2 𝑥 2 + 2𝑥𝑎3 𝑥 3 + ⋯
= 2𝑥𝑎0 + 2𝑎1 𝑥 2 + 2𝑎2 𝑥 3 + 2𝑎3 𝑥 4 + ⋯ … ..
2𝑎0
⇒ 𝑎1 = 0 2𝑎2 = 2𝑎0 ⇒ 𝑎2 = = 𝑎0
𝑧
2𝑎1
3. 𝑎3 = 2𝑎1 ⇒ 𝑎3 = =0
3
2𝑎2 𝑎0
4𝑎4 = 2𝑎2 ⇒ 𝑎4 = =
42 2

5𝑎5 = 2𝑎3 = 0 ⇒ 𝑎5 = 0
2𝑎4 𝑎4 𝑎0 𝑎0
6𝑎6 = 2𝑎4 ⇒ 𝑎6 = = = =
6 3 2⋅3 3!

We get,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0
= 𝑎0 + 0 + 𝑎0 𝑥 2 + 0𝑥 3 + 𝑥 4 + ⋯
2
𝑎 0
= 𝑎0 + 𝑎0 𝑥 2 + 𝑥 4 + ⋯
2
2
𝑥4 𝑥6
= 𝑎0 (1 + 𝑥 + + + ⋯ )
2! 3!
2
𝑦 = 𝑎0 𝑒 𝑥
To find an actual solution

𝑦 ′ = 2𝑥𝑦
𝑑𝑦
= 2𝑥𝑦
𝑑𝑥
𝑑𝑦
⇒ = 2𝑥 ⋅ 𝑑𝑥
𝑦
log 𝑦 = 𝑥 2 + 𝑐
2
𝑦 = 𝑒𝑥 + 𝑐
2
⇒ 𝑦 = 𝑎0 𝑒 𝑥 , where 𝑎0 = 𝑒 𝑐

6
Example 3: Consider 𝑦 = (1 + 𝑥)𝑝 where p is an arbitrary constant. Construct a
differential equation from this and then find the solution using power series
method.
Solution
First, we construct a differential equation
i.e. 𝑦 = (1 + 𝑥)𝑝
𝑝(1+𝑥)𝑝 𝑝𝑦
𝑦 ′ = 𝑝(1 + 𝑥)𝑝−1 = =
1+𝑥 1+𝑥

∴ (1 + 𝑥)𝑦 ′ = 𝑝𝑦, 𝑦(0) = 𝑟

Assume that y has a power series solution of the form,
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎̇ 2 𝑥 2 + ⋯ … …
Which converges for |𝑥| < 𝑅̇, 𝑅 > 0
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ … … ..
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1 + 𝑥)𝑦 ′ = 𝑝𝑦
⇒ (1 + 𝑥)𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 𝑝(𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
⇒ (𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) + (𝑎1 𝑥 + 2𝑎2 𝑥 2 + 3𝑎3 𝑥 3 + ⋯ )
= 𝑎0 𝑝 + 𝑎1 𝑝𝑥 + 𝑎2 𝑝𝑥 2 + ⋯
Equating the coefficients of 𝑥, 𝑥 2 , …
𝑎1 = 𝑎0 𝑝 i.e. 𝑎1 = 𝑝, (since 𝑎0 = 1)
⇒ 2𝑎2 = 𝑎1 (p − 1)
𝑎1 (p − 1) 𝑎0 𝑃(𝑝 − 1)
𝑎2 = =
2 2

7
 3𝑎3 + 2𝑎2 = 𝑎2 𝑝
𝑠𝑎3 = 𝑎2 𝑝 − 2𝑎2
= 𝑎2 (𝑝 − 2)
𝑎2 (𝑝 − 2) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)
𝑎3 = =
3 2⋅3
4𝑎4 + 3𝑎3 = 𝑎3 𝑝
4𝑎4 = 𝑎3 𝑝 − 3𝑎3
= 𝑎3 (𝑝 − 3)
𝑎3 (𝑝 − 3) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3)
𝑎4 = =
4 2⋅3⋅4
∴ we get,
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0 𝑝(𝑝 − 1) 2 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 𝑎0 + 𝑎0 𝑝𝑥 + 𝑥 + 𝑥 + ⋯…
2 2⋅3
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 1 + 𝑝𝑥 + 𝑥 + 𝑥 +
2! 3!
𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3) 4 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − (𝑛 − 1)) 𝑛
𝑥 + ⋯+ 𝑥
4! 𝑛!
Since the initial problem y(0) = 1 has one solution the series converges for |x|<1
So this is a power solution,
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1) ⋯ (𝑝 − (𝑛 − 1)) 𝑛
(1 + 𝑥)𝑝 = 1 + 𝑝𝑥 + 𝑥 + ⋯+ 𝑥
2! 𝑛!
Which is binomial series.

Example 4: Solve the equation 𝑦 ′ = 𝑥 − 𝑦, 𝑦 (0) = 0

Solution: Assume that y has a power series solution of the form
∞

𝑦 = ∑ an 𝑥 𝑛
𝑛=0

which converges for |𝑥| < 𝑅, 𝑅 > 0

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯

8
 Now 𝑦 ′ = 𝑥 − 𝑦
(𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) = 𝑥 − (𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
Equating the coefficients of 𝑥, 𝑥 2 ,
𝑎1 = −𝑎0 = 0, Since 𝑦(0) = 0
2𝑎2 = 1 − 𝑎1
=1−0
1
⇒ 𝑎2 =
2
3𝑎3 = −𝑎2
−𝑎2 1
𝑎3 = =−
3 2⋅3
4𝑎4 = −𝑎3
1
⇒ 𝑎4 =
2⋅3⋅4
𝑥2 𝑥3 𝑥4
∴ 𝑦 = 0 + 0 + − + − ⋯……
2! 3! 4!
𝑥2 𝑥3
= (1 − 𝑥 + − + ⋯ ) + 𝑥 − 1
2! 3!
= 𝑒 −𝑥 + 𝑥 − 1
By direct method
𝑦′ = 𝑥 − 𝑦 𝑑𝑦
𝑑𝑦 𝑑𝑦 ( + 𝑝𝑦 = 𝑄 𝑓𝑜𝑟𝑚)
=𝑥−𝑦 ⇒ +𝑦 =𝑥 𝑑𝑥
𝑑𝑥 𝑑𝑥
here 𝑃(𝑥) = 1, integrating factor
= 𝑒 ∫ 𝑝(𝑥)⋅𝑑𝑥
= 𝑒𝑥
∴ 𝑦𝑒 𝑥 = ∫ 𝑥𝑒 𝑥 ⋅ 𝑑𝑥
𝑦𝑒 𝑥 = 𝑥 ⋅ 𝑒 𝑥 − ∫ 𝑒 𝑥 ⋅ 𝑑𝑥
= 𝑥𝑒 𝑥 − 𝑒 𝑥
𝑦𝑒 𝑥 = 𝑒 𝑥 (𝑥 − 1) + 𝑐
𝑒 𝑥 (𝑥 − 1) + 𝑐 𝑐
𝑦= = 𝑥 − 1 + 𝑥 = 𝑐𝑒 −𝑥 + (𝑥 − 1)
𝑑𝑥 𝑒
∴ 𝑦 = (𝑥 − 1) + 𝑐𝑒 −𝑥

9
 CHAPTER 2
SECOND ORDER LINEAR EQUATION, ORDINARY POINTS

Consider the general homogeneous second order linear equation,

𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0 (1)
As we know, it is occasionally possible to solve such an equation in terms of
familiar elementary functions. This is true, for instance, when P(x) and Q(x)
are constants, and in a few other cases as well. For the most part, however,
the equations of this type having the greatest significance in both pure and
applied mathematics are beyond the reach of elementary methods, and can
only be solved by means of power series.
P(x) and Q(x) are called coefficients of the equation. The behaviour of its
solutions near a point x0 depends on the behaviour of its coefficient functions
P(x) and Q(x) near this point. we confine ourselves to the case in which P(x) and
Q(x) are well behaved in the sense of being analytic at x0, which means that
each has a power series expansion valid in some neighbourhood of this point. In
this case x0 is called an ordinary point of equation (1). Any point that is not an
ordinary point of (1) is called a singular point.

Consider the equation,

𝑦 ′′ + 𝑦 = 0 (2)

the coefficient functions are P(x) = 0 and Q(x) = 1, These functions are analytic
at all points, so we seek a solution of the form,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ + 𝑎𝑛 𝑥 𝑛 + ⋯ (3)

Differentiating (3) we get,

𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ + (𝑛 + 1)𝑎𝑛+1 𝑥 𝑛 + ⋯ (4)

10
And

𝑦 ′′ = 2𝑎2 + 2 ⋅ 3𝑎3 𝑥 + 3 ⋅ 4𝑎4 𝑥 2 + ⋯ + (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ⋯ (5)

If we substitute (5) and (3) into (2) and add the two series term by term, we get

(2𝑎2 + 𝑎0 ) + (2 ⋅ 3𝑎3 + 𝑎1 )𝑥 + (3 ⋅ 4𝑎4 + 𝑎2 )𝑥 2 +

𝑦 ′′ + 𝑦 = =0
(4 ⋅ 5𝑎5 + 𝑎3 )𝑥 3 + ⋯ + [(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 ]𝑥 𝑛 + ⋯
and equating to zero the coefficients of successive powers of x gives
2𝑎2 + 𝑎0 = 0, 2 ⋅ 3𝑎3 + 𝑎1 = 0, 3 ⋅ 4𝑎4 + 𝑎2 = 0
4 ⋅ 5𝑎5 + 𝑎3 = 0, … … , (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 = 0, …
By means of these equations we can express an in terms of a0 or a0, according
as n is even or odd:
𝑎0 𝑎1 𝑎2 𝑎0
𝑎2 = − , 𝑎3 = − , 𝑎4 = − =
2 2⋅3 3⋅4 2⋅3⋅4
𝑎3 𝑎1
𝑎5 = − = ,⋯
4⋅5 2⋅3⋅4⋅5
With these coefficients, (3) becomes
𝑎0 𝑎1 3 𝑎0 𝑎1
𝑦 = 𝑎0 + 𝑎1 𝑥 − 𝑥 2 − 𝑥 + 𝑥4 + 𝑥5 − ⋯
2 2⋅3 2⋅3⋅4 2⋅3⋅4⋅5
𝑥2 𝑥4 𝑥3 𝑥5
= 𝑎0 (1 − + − ⋯ ) + 𝑎1 (𝑥 − + − ⋯ ) (6)
2! 4! 3! 5!
𝑖. 𝑒, 𝑦 = 𝑎0 cos 𝑥 + 𝑎1 sin 𝑥
Since each of the series in the parenthesis converges for all x. This implies the
series (2) for all x.

Solve the legenders equation,

(1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0
Solution
Consider (1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
11
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2

put 𝑛 = 𝑛 + 2 (Since 𝑦 ′′ is not 𝑥 𝑛 form )

∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛+2 𝑥 𝑛+2−2

𝑛=0
∞

∴ 𝑦 ′′ = ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

Now (1) ⇒ 𝑦 ′′ − 𝑥 2 𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0

⇒ ∑(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 − ∑𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛 − ∑2𝑛𝑎𝑛 𝑥 𝑛 + ∑𝑝(𝑝 + 1)𝑎𝑛 𝑥 𝑛 = 0
∞

⇒ ∑ [((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 )𝑥 𝑛 ] = 0

𝑛=0

for n = 0,1,2,3…….
⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 = 0
[𝑛(𝑛 − 1) + 2𝑛 − 𝑝(𝑝 + 1)]
⇒ 𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(𝑛2 − 𝑛 + 2𝑛 − 𝑝2 − 𝑝)𝑎𝑛
=
(𝑛 + 1)(𝑛 + 2)
(𝑛 + 𝑛 − 𝑝2 − 𝑝)𝑎𝑛
2
=
(𝑛 + 1)(𝑛 + 2)

12
 −(𝑝 − 𝑛)(𝑝 + 𝑛 + 1)
∴ 𝑎𝑛+2 = 𝑎𝑛 , 𝑛 = 0,1,2 …
(𝑛 + 1)(𝑛 + 2)

This is an Recursion formula

−𝑝(𝑝 + 1)
put 𝑛 = 0, 𝑎2 = 𝑎0
1⋅2
−(𝑝 − 1)(𝑝 + 2)
𝑛 = 1, 𝑎3 = ⋅ 𝑎1
2⋅3
−(𝑝 − 2)(𝑝 + 3)
𝑛 = 2, 𝑎4 = 𝑎2
3𝑖4
𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3)
= 𝑎0
4!
−(𝑝 − 3)[𝑝 + 4)
𝑛 = 3, 𝑎5 = 𝑎3
4⋅5
(𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4)
= 𝑎1
5!
−(𝑝 − 4)(𝑝 + 5)
𝑛 = 4, 𝑎6 = 𝑎4
5⋅6
−𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5)
= 𝑎0
6!
(𝑝 − 5)(𝑝 + 6)
𝑛 = 5, 𝑎7 = − 𝑎5
6⋅7
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6)
=− 𝑎1
7!

𝑝(𝑝 + 1) 2 𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3) 4

𝑦 = 𝑎0 [1 − 𝑥 + 𝑥
2! 4!
𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5) 6
− 𝑥 + ⋯]
6!
(𝑝 − 1)(𝑝 + 2) 3 (𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4) 5
+𝑎1 [𝑥 − 𝑥 + 𝑥
3! 5!
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6) 7
− 𝑥 + ⋯]
7!

13
Find the general solution of (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 in terms of power
series in 𝑥. Can you express this solution by means of elementary functions?
Solution
Consider the equation (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

(1 + 𝑥 2 )𝑦 ′′ = 𝑦 ′′ + 𝑥 2 𝑦 ′′
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2
∞

Now 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

put 𝑛 = 𝑛 + 2
∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛 + 2𝑥 𝑛+2=2

𝑛=0
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

14
 ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥𝑛

𝑛=0 𝑛=2
∞ ∞

+ ∑ 2𝑛𝑎𝑛 𝑥 𝑛 − ∑ 2𝑎𝑛 𝑥 𝑛 = 0
𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 = 0

[−𝑛(𝑛 − 1) − 2𝑛 + 2]
𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(−𝑛2 + 𝑛 − 2𝑛 + 2)
= 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)

2 2𝑎0
put 𝑛 = 0, 𝑎2 = 𝑎0 = = 𝑎0
1⋅2 2!
(1 − 1 − 2 + 2)
𝑛 = 1, 𝑎3 = 𝑎1 = 0
2⋅3
2−4−4+2 −4 −𝑎0
𝑛 = 2, 𝑎4 = 𝑎2 = 𝑎0 =
3⋅4 3⋅4 3
3 − 9 − 16 + 2
𝑛 = 3, 𝑎5 = 𝑎3 = 0
4.5
4 − 16 − 8 + 2 −3 3𝑎0 𝑎0
𝑛 = 4, 𝑎6 = 𝑎4 = 𝑎4 = =
5.6 5 3.5 5

∴ 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0
= 𝑎0 + 𝑎1 𝑥 + 𝑎0 𝑥 2 − 𝑥 4 + 𝑥 6 … .
3 5
4 6
𝑥 𝑥
= 𝑎0 [1 + 𝑥 2 − + − ⋯ ] + 𝑎1 𝑥
3 5
𝑥3 𝑥5
= 𝑎0 [1 + 𝑥 (𝑥 − + ⋯ )] + 𝑎1 𝑥
3 5
= 𝑎0 (1 + 𝑥tan−1 𝑥) + 𝑎1 𝑥

15
Consider the equation 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0
(a) Find its general solution 𝑦 = ∑𝑎𝑛 𝑥 𝑛 in the form
𝑦 = 𝑎0 𝑦1 (𝑥) + 𝑎1 𝑦2 (𝑥) where 𝑦1 (𝑥) and 𝑦2 (𝑥) are power series
(b) use the ratio test to verify that the two series 𝑦1 (𝑥) and 𝑦2 (𝑥) converges
. for all x.
Solution:
Given 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ (1)
Assume that y has a power series solution the form ∑a𝑛 𝑥 𝑛 which converges
for |𝑥| 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛 ⋅ 𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0
∞

𝑥𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞ ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)a𝑛+2 𝑥 + ∑ 𝑛𝑎𝑛 𝑥 + ∑ 𝑎𝑛 𝑥 𝑛 = 0

𝑛 𝑛

𝑛=0 𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 = 0

(−𝑛 − 1)𝑎𝑛 −𝑎𝑛
⇒ 𝑎𝑛+2 = =
(𝑛 + 1)(𝑛 + 2) 𝑛 + 2
𝑎0
put 𝑛 = 0, 𝑎2 = −
2
−2𝑎1 −𝑎1
𝑛 = 1, 𝑎3 = =
2⋅3 3
16
 −3𝑎2 −𝑎2 𝑎0
𝑛 = 2, 𝑎4 = = =
3⋅4 4 8
−4𝑎3 𝑎1
𝑛 = 3, 𝑎5 = =
4⋅5 15
−5𝑎4 −𝑎0
𝑛 = 4, 𝑎6 = =
5⋅6 48
𝑎0 𝑎1 𝑎0 𝑎1 𝑎0
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + − 𝑥 2 − 𝑥 3 + 𝑥 4 + 𝑥 5 − 𝑥 6 + ⋯
2 3 8 15 48

𝑥2 𝑥4 𝑥6 𝑥3 𝑥5
= 𝑎0 [1 − + − + ⋯ ] + 𝑎1 [𝑥 − + + ⋯]
2 2⋅4 2⋅4⋅6 3 3.5

𝑥2 𝑥4 𝑥̇ 2
𝑤ℎ𝑒𝑟𝑒 𝑦1 (𝑥) = 1 − + − +
2 2⋅4 2⋅4⋅6

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + +⋯
3 3⋅5

𝑥2 𝑥4 𝑥6
(b) 𝑦1 (𝑥) = 1 − + − +⋯
2 2⋅4 2⋅4⋅6

𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | |
𝑛→∞ 𝑎𝑛+1
(−1)𝑛 (−1)𝑛+1
= 𝑙𝑖𝑚 | / |
𝑛→∞ 2 ⋅ 4 ⋅ (2𝑛) 2 ⋅ 4 ⋅⋅ (2𝑛 + 2)

2(𝑛 + 1)
= 𝑙𝑖𝑚 | |
𝑛→∞ −1
1
= 𝑙𝑖𝑚 | − 2𝑛(1 + )| = ∞
𝑛→∞ 𝑛

∴ 𝑦1 (𝑥) converges for all 𝑥

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + −⋯
3 3⋅5

17
 (−1)𝑛 (−1)𝑛+1
𝑅 = 𝑙𝑖𝑚 | ⁄ |
𝑛→∞ 3 ⋅ 5 ⋯ (2𝑛 + 1) 3 ⋅ 5 ⋅ ⋯ (2𝑛 + 3)

(−1) ⋅ 3 ⋅ 5 ⋯ (2𝑛 + 1)(2𝑛 + 3)

= 𝑙𝑖𝑚 | |
𝑛→∞ 3 ⋅ 5 ⋯ ⋅ (2𝑛 + 1)
= 𝑙𝑖𝑚 |(−1)𝑛(2 + 3/𝑛)| = ∞
𝑛→∞

∴ 𝑦2 (𝑥) converges for all 𝑥

REGULAR SINGULAR POINTS

A singular point 𝑥0 of equation
𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0
is said to be regular if the functions (𝑥 − 𝑥0 )𝑃(𝑥) and (𝑥 − 𝑥0 )2 𝑄(𝑥) are
analytic, and irregular otherwise. Roughly speaking, this means that the
singularity in 𝑃(𝑥) cannot be worse than 1/(𝑥 − 𝑥0 ), and that in 𝑄(𝑥) cannot
be worse than 1/(𝑥 − 𝑥0 )2 .
If we consider Legendre’s equation in the form
2𝑥 𝑝(𝑝 + 1)
𝑦 ′′ − 𝑦 ′
+ 𝑦=0
1 − 𝑥2 1 − 𝑥2
it is clear that x = 1 and x = −1 are singular points. The first is regular because
2𝑥 (𝑥 − 1)𝑝(𝑝 + 1)
(𝑥 − 1)𝑃(𝑥) = and (𝑥 − 1)2 𝑄(𝑥) = −
𝑥+1 𝑥+1
are analytic at x = 1, and the second is also regular for similar reasons.
Example: Bessel’s equation of order p, where p is a nonnegative constant:
𝑥 2 𝑦 ′′ + 𝑥𝑦 ′ + (𝑥 2 − 𝑝2 )𝑦 = 0
If this is written in the form
1 ′ 𝑥 2 − 𝑝2
′′
𝑦 + 𝑦 + 𝑦 = 0,
𝑥 𝑥2
it is apparent that the origin is a regular singular point because𝑥𝑃(𝑥) = 1 and
𝑥 2 𝑄(𝑥) = 𝑥 2 − 𝑝2 are analytic at x = 0.

18
 CONCLUSION
The purpose of this project gives a simple account of series solution of first
order equation, second order linear equation, ordinary points. The study of these
topics given excellent introduction to the subject called ‘POWER SERIES’
we used application of power series extensively throughout this project. We
take it for granted that most readers are reasonably well acquainted with these
series from an earlier course in calculus. Nevertheless, for the benefit of those
whose familiarity with this topic may have faded slightly, we presented a brief
review of the main facts of power series.

19
 BIBLIOGRAPHY
George F. Simmons - Differential Equations With Applications and Historical
. Notes
G.Birkoff and G.C Rota - Ordinary Differential Equations; Wiley and Sons; 3rd
. Edition ( 1978 )
P.Hartmon - Ordinary Differential Equations; John Wiley and Sons

20
DON BOSCO ARTS & SCIENCE COLLEGE
ANGADIKADAVU

DEPARTMENT OF MATHEMATICS
2018-2021

Project Report on

INNER PRODUCT SPACES

 DEPARTMENT OF MATHEMATICS

DON BOSCO ARTS & SCIENCE COLLEGE

ANGADIKADAVU

MARCH 2021

Project Report on

INNER PRODUCT SPACES

Dissertation submitted in the partial fulfilment

of the requirement for the award of

Bachelor of Science in Mathematics of

Kannur University

Name : ANASWARA SASIDHARAN

Roll No. : DB18CMSR02

Examiners: 1.

2.
 𝐊𝐀𝐍𝐍𝐔𝐑 𝐔𝐍𝐈𝐕𝐄𝐑𝐒𝐈𝐓𝐘

𝐁𝐎𝐍𝐀𝐅𝐈𝐃𝐄 𝐂𝐄𝐑𝐓𝐈𝐅𝐈𝐂𝐀𝐓𝐄

Certified that this project report on " INNER PRODUCT SPACES” is the bonafide
work of ANASWARA SASIDHARAN who carried out the project work under
my supervision.

Mrs. Riya Baby Mr. Anil M V

Head of Department Supervisor
 𝐃𝐄𝐂𝐋𝐀𝐑𝐀𝐓𝐈𝐎𝐍

I, ANASWARA SASIDHARAN hereby declare that the project work entitled

‘INNER PRODUCT SPACES′ has been prepared by me and submitted to Kannur
University in partial fulfilment of requirement for the award of Bachelor of
Science is a record of original work done by me under the supervision of Mr. ANIL
M V, Assistant Professor, Department of Mathematics, Don Bosco Arts and Science
College, Angadikadavu.

I, also declare that this Project work has been submitted by me fully or partially for
the award of any Degree, Diploma, Title or recognition before any authority.

Place ∶ Angadikadavu
ANASWARA SASIDHARAN
Date ∶ DB18CMSR02
 𝐀𝐂𝐊𝐍𝐎𝐖𝐋𝐄𝐃𝐆𝐄𝐌𝐄𝐍𝐓

Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express mydeepest gratitude to
people along the way.
No words can adequately express the sense of gratitude, still I try to express my
heartfelt thanks through words. The outset, I am deeply indebted to my project
supervisor Mr. ANIL M.V, Assistant Professor, Department of Mathematics, Don
Bosco Arts and ScienceCollege, Angadikadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career.I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikadavu.
I could not name many who sincerely supported and helped for the successful
completion of this Project. It is my pleasure and duty to thank each and every one
of them who walked with me.

ANASWARA SASIDHARAN
 Contents

1. INTRODUCTION 1

2. PRELIMINARIES 2

3. CHAPTER 1 : INNER PRODUCT SPACES 6

4. CHAPTER 2 : ORTHOGONAL SETS 9

5. BIBLIOGRAPHY 16
INTODUCTION

In linear algebra, an inner product space is a vector space with an additional structure

called an inner product. This additional structure associates

each pair of vectors in the space with a scalar quantity known as the inner product of the

vectors. Inner products allow the rigorous introduction of intuitive geometrical notions

such as the length of a vector or the angle between two vectors. They also provide the

means of defining orthogonality between vectors (zero inner product). Inner product

spaces generalize Euclidean spaces (in which the inner product is the dot product, also

known as the scalar product) to vector spaces of any (possibly infinite) dimension and are

studied in functional analysis. The first usage of the concept of a vector space with an

inner product is due to Peano, in 1898.

An inner product naturally induces an associated norm, thus an inner product space is also

a normed vector space. A complete space with an inner product is called a Hilbert space.

An (incomplete) space with an inner product is called a pre-Hilbert space.

1
 PRELIMINARIES

LINEAR SPACES

Definition 1: A linear (vector) space X over a field F is a set of elements

together with a function, called addition, from X ×X into X and a function

called scalar multiplication, from F × X into X which satisfy the following

conditions for all x, y, z ∈ X and α, β ∈ F;

i. (x + y) + z = x + (y + z)

ii. x +y = y +x

iii. There is an element 0 in X such that x + 0 = x for all x ∈ X.

iv. For each x ∈ X there is an element −x ∈ X such that x + (−x) = 0.

v. (x + y) = αx + αy

vi. (α + β)x = αx + βx

vii. α(βx) = (αβ)x

viii. 1 · x = x.

Properties i to iv imply that X is an abelian group under addition and v to vi

relate the operation of scalar multiplication to addition X and to addition and

multiplication in F.

Examples:

(a) Vn(R). The vectors are n-tuples of real numbers and the scalars are real

2
 numbers with addition and scalar multiplication defined by

(1 ,···,  n ) + ( 1 ,···,  n ) = (1 + 1 ,···,  n +  n ) (1)

 (1 ,···,  n ) = ( 1 ,···,  n ) (2)

Vn(R) is a linear space over R. Similarly, the set of all n-tuples of complex

numbers with the above definition of addition and multiplication is a linear

space over C and is denoted as Vn(C).

(b) The set of all functions from a nonempty set X into a field F with addition and

scalar multiplication defined by

[f + g](t) = f (t) + g(t) and [αf ](t) = αf (t); f, g ∈ X, t ∈ T (3)

is a linear space.

Let T = N the set of all positive integers and X is the set of all sequences of

elements F with addition and scalar multiplication defined by

( n +  n ) = ( n +  n ) (4)

 ( n ) = (  n ) (5)

denoted as V∞(F), form a linear space.

3
METRIC SPACES

Remember the distance function in the Euclidean space Rn.

Let x, y, z ∈ Rn, then

(1) |x − y| ≥ 0; |x − y| = 0 if and only if x = y ;

(2) |x − y| = |y − x|;

(3) |x − y| ≤ |x − z| + z − y|.

Definition 2: A metric or distance function on a set X is a real valued function

d defined on X × X which has the following properties: for all x, y, z ∈ X.

(1) d(x, y) ≥ 0; d(x, y) = 0 if and only if x = y;

(2) d(x, y) = d(y, x);

(3) d(x, y) ≤ d(x, z) + d(z, y)

A metric space (X, d) is a nonempty set X and a metric d defined on X.

Examples: In addition to the Euclidean spaces let us have the following examples.

Here all functions are assumed to be continuous. Let Lp denotes a set of complex

p
valued functions in Rn such that f is integrable. Let us recall some results

concerning such functions.

Höder’s Inequality: If p > 1, 1/q = 1 − 1/p

 | fg |  [ | f | ] [  | g |q ]1/ q .
p 1/ p

Minkowski’s Inequality: If p ≥ 1,

[ | f + g | p ]1/ p  [ | f | p ]1/ p + [  | g | p ]1/p

4
If xk and yk for k = 1, … , m are complex numbers, let f (t) = |xk| and g(t) =

|yk| for t ∈ [k, k + 1] and f (t) = 0 = g(t) for t∈ [1, m + 1]. Then we obtain the

summation form of the above inequalities from the integral form

Hölder’s Inequality

1/ p 1/ q
m m p
 m q

 xk yk   xk    yk 
k =1  k =1   k =1 

Minkowski’s Inequality:

p 1/ p p 1/ p p 1/p
m  m  m 
  xk + yk     xk  +   yk 
 k =1   k =1   k =1 

NORMED LINEAR SPACES

Definition 3. A norm on X is a real valued function, whose value at x is denoted

by ||x||, satisfying the following conditions for all x, y ∈ X and α ∈ F;

(1) ||x|| > 0 if x ≠ 0

(2) ||αx|| = |α|||x||

(3) ||x + y|| ≤ ||x|| + ||y||.

A linear space X with a norm defined on it is called a normed linear space.

Example: l p space. On the linear space Vn(F), define

n
x = [ |  i | p ]1/ p
k =1

where p ≥ 1 is any real number and x = (1 ,···,  n ) . This defines a norm (called p-

p
norm) on Vn(F). This space is called l space .
5
 CHAPTER 1

INNER PRODUCT SPACES

INNER PRODUCTS

Let 𝐹 be the field of real numbers or the field of complex numbers, and V a vector space over

F an inner product on V is a function which assigns to each ordered’ pair of vectors 𝛼, 𝛽 in V

a scalar (𝛼|𝛽) in 𝐹 in such a way that for all 𝛼, 𝛽, γ in V and all scalars c.

(a) (𝛼 + 𝛽|𝛾) = (𝛼|𝛾) + (𝛽|𝛾) ;

(b) (c𝛼|𝛽) = 𝑐(𝛼|𝛽) ;

̅̅̅̅̅ ), the bar denoting complex conjugation

(c) (𝛽|𝛼) = (𝛼|𝛽

(d) (𝛼|𝛼) > 0 if 𝛼 ≠ 0

It should be observed that conditions (a), (b) and (c) implies that

(𝑒) = (𝛼 ∣ 𝑐𝛽 + 𝛾) = (𝑐̅(𝛼|𝛽) + (𝛼|𝛾)

One other point should be made. When 𝐹 is the field 𝑅 of real nunbers. The complex conjugates

appearing in (c) and (e) are superflom. However, in the complex case they are necessary for

the consistency of the conditions. Without these complex conjugates we would have the

contradiction

(𝛼|𝛼) > 0 and (𝑖𝛼 ∣ 𝑖𝛼) = −1(𝛼|𝛼)

Example 1:

On F 𝑛 there is an inner product which we call the standard inner product. It is defined on 𝛼 =

(𝑥1 , ⋯ 𝑥𝑛 ) and 𝛽 = (𝑦1 , … , 𝑦𝑛 ), by

6
 (𝛼|𝛽) = ∑ 𝑥𝑖 𝑦̅𝑖
𝑖

When F is R this may be also written as

(𝛼|𝛽) = ∑ 𝑥𝑖 𝑦𝑖
𝑖

In the real case, the standard inner product is often called the dot or scalar product and denoted

by 𝛼 ⋅ 𝛽.

INNER PRODUCTS SPACES

An inner product space is a real or complex vector space together with a specified inner product

on that space.

• A finite-dimensional real inner product space is often called a Euclidean spare. A

complex inner product spare often referred to as a unitary spare.

• Every inner product space is a normed linear space and every normed space is a metric

space. Hence , every inner product space is a metric space.

Theorem

If V is an inner product space, then for any vector’s 𝛼, 𝛽 in 𝑉 and any scalar c

(1) ||𝑐𝛼|| = |𝑐|||𝛼|| ;

(ii) ||𝛼|| > 0 for 𝛼 ≠ 0

(iii) |(𝛼 ∣ 𝛽)| ⩽ ||𝛼|| ||𝛽||

(iv) ∥ 𝛼 + 𝛽|| ⩽∥ 𝛼 ∥ +∥ 𝛽||

Proof:

Statements (i) and (ii) follow almost immediately form the various definitions

involved. The inequality in (iii) is clearly valid when 𝛼 = 0. if 𝛼 ≠ 0, put

(𝛽|𝛼)
𝛾=𝛽− 𝛼
∥ 𝛼 ∥2

7
 Then, (𝛾 ∣ 𝛼) = 0 and

(𝛽|𝛼) (𝛽|𝛼)
0 ⩽∥ 𝛾 ∥2 = (𝛽 − 2
𝛼 ⁄𝛽 − 𝛼)
∥𝛼∥ ∥ 𝛼 ∥2

(𝛽|𝛼)(𝛼|𝛽)
= (𝛽|𝛽) −
∥ 𝛼 ∥2

|(𝛼|𝛽)|2
=∥ 𝛽 ∥2 −
∥ 𝛼 ∥2

Hence,

|(𝛼 ∣ 𝛽)|2 ⩽∥ 𝛼 ∥2 ∥ 𝛽 ∥2

Now using (c) we find that

∥ 𝛼 + 𝛽 ∥2 = ∥ 𝛼 ∥2 + (𝛼 ∣ 𝛽) + (𝛽 ∣ 𝛼)+∥ 𝛽 ∥2

=∥ 𝛼 ∥2 + 2Re (𝛼 ∣ 𝛽)+∥ 𝛽 ∥2
⩽∥ 𝛼 ∥2 + 2 ∥ 𝛼 ∥∥ 𝛽 ∥ +∥ 𝛽 ∥2
= (∥ 𝛼 ∥ +∥ 𝛽 ∥)2

Thus,

∥ 𝛼 + 𝛽 ∥ ⩽ ∥ 𝛼 ∥ +∥ 𝛽 ∥

the inequality (iii) is called the Cauchy -Schwarz inequality. It has a wide variety of application

the proof shows that if 𝛼is non-zero then

(( 𝛼 ∣ 𝛽 )) <∥ 𝛼 ∥∥ 𝛽 ∥, unless

(𝛽|𝛼)
𝛽= 𝛼
∥ 𝛼 ∥2

Then equality occurs in (iii) if and only if 𝛼 and 𝛽 are linearly independent.

8
 CHAPTER 2

ORTHOGONAL SETS

Definition

Let α and β be the vectors in an inner product space V. Then α is orthogonal to β if

(α | β) = 0. We simply say that and are orthogonal.

Definition

If S is a set of vectors in V, S is called an orthogonal set provided all set pairs of

distinct vectors in S are orthogonal.

Definition

An orthogonal set is an orthogonal set S with the additional property that ∥ 𝛼 ∥= 1 for

every 𝛼 in S.

• The zero vectors are orthogonal to every vector in V and is the only vector with this

property.

• It is an appropriate to think of an orthonormal set as a set of mutually perpendicular

vectors each having length l.

Example: the vector (x , y) is 𝑅 2 is orthogonal to (−y , x) with respect to the standard inner

product, for,

((x , y)|(−y , x)) = −xy + yx = 0

• The standard basis of either 𝑅 𝑛 or 𝐶 𝑛 is an orthonormal set with respect to the standard

inner product.

9
Theorem : An orthogonal set of nonzero vectors is linearly independent.

Proof:

Let S be a finite or infinite orthogonal set of nonzero vectors in a given inner product space

suppose 𝛼1, 𝛼2 , … 𝛼𝑛 are distinct vectors in S and that β=𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛

Then (β|𝛼𝑘 )=( 𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛 |𝛼𝑘 )

= 𝑐1 (𝛼1 |𝛼𝑘 ) + 𝑐2 ( 𝛼2 |𝛼𝑘 )+…+𝑐𝑛 (𝛼𝑛 |𝛼𝑘 )

= 𝑐𝑘 (𝛼𝑛 |𝛼𝑘 ) , since (𝛼𝑖 |𝛼𝑗 ) = 0,if i ≠ j and (𝛼𝑖 |𝛼𝑗 ) = 1,if i=j

Hence, 𝑐𝑘 = (β |𝛼𝑘 ) /(𝛼𝑘 , 𝛼𝑘 ) )

𝑐𝑘 =(β |𝛼𝑘 )/||𝛼𝑘 ||2,1≤ k ≤ m

Thus, when β=0 each 𝑐𝑘 =0; so S is a linearly independent set.

Corollary:

If {𝛼1, 𝛼2 , … 𝛼𝑚 } is an orthogonal set of nonzero vectors in a finite dimensional inner product

space V, then m ≤ dimV.

That is number of mutually orthogonal vectors in V cannot exceed the dimensional V.

Corollary:

If a vector β is linear combination of an orthogonal of nonzero vectors 𝛼1, 𝛼2 , … 𝛼𝑛 , then β is

the particular linear combination

(𝛽 |𝛼𝑘 )
β =∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘
𝑘 ||

Proof :

Since β is the linear combination of an orthogonal sequence of nonzero vectors

𝛼1, 𝛼2 , … 𝛼𝑛 , we can write β =𝑐1 𝛼1 + ⋯ 𝑐𝑛 𝛼𝑛 .

(𝛽|𝛼𝑘 )
Where 𝑐𝑘 = , 1 ≤ k ≤ m (ref. by previous theorem)
||𝛼𝑘 ||2

( 𝛽 | 𝛼𝑘 )
Hence, β= ∑𝑚
𝑘=1 ||𝛼𝑘 ||2

10
Theorem (Gram Schmidt Orthogonalization Process)

Let V be an inner product space and {𝛽1, … , 𝛽𝑛 } be any linearly independent vectors in V. Then

one may construct orthogonal vectors {𝛼1, 𝛼2 , … 𝛼𝑛 } in V, such that for each k = 1, 2, …n, the

set {𝛼1, 𝛼2 , … 𝛼𝑘 }is an orthogonal basis for the subspace of V spanned by 𝛽1, … , 𝛽𝑛 .

Proof :

The vectors are obtained by means of a construction known as the Gram Schmidt

orthogonalization process.

First let 𝛼1 =𝛽1 The other vectors are then given inductively as follows:

Suppose 𝛼1, 𝛼2 , … 𝛼𝑚 (1 ≤ m ≤ n) have been chosen so that for every k

{𝛼1, 𝛼2 , … 𝛼𝑘 } (1≤k≤m)

is an orthogonal basis for the space of v that is spanned by 𝛽1, … , 𝛽𝑛

To construct the next vector 𝛼𝑚+1, let

( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1, = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
Then  m +1  0 . For otherwise, 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘 =0, implies,
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
𝛽𝑚+1= 𝛽𝑚+1 -∑𝑚
𝑘=1 𝛼𝑘 ,implies,  m+1 is a linear combination of 𝛼1, 𝛼2 , … 𝛼𝑚 and
||𝛼𝑘 ||2

hence a linear combination of 1 ,  2 ,...,  m , a contradiction.

Furthermore, if 1≤j≤m, then,

(𝛽𝑚+1 |𝛼𝑘 )
(𝛼𝑚+1 | 𝛼𝑗 ) = ( 𝛽𝑚+1 | 𝛼𝑗 ) -∑𝑚
𝑘=1 (𝛼𝑘 |𝛼𝑗 )
||𝛼𝑘 ||2

= ( 𝛽𝑚+1 | 𝛼𝑚 ) – ( 𝛽𝑚+1 | 𝛼𝑗 ) , using the orthonormality of {𝛼1, 𝛼2 , … 𝛼𝑚 }.

Therefore {𝛼1, 𝛼2 , …,𝛼𝑚+1 } is an orthogonal set consisting of m+1 nonzero vectors in the

subspace spanned by 𝛽1, … , 𝛽𝑚+1 . Hence by an earlier Theorem , it is a basis for this subspace

.Thus the vectors , 𝛼1, 𝛼2 , … 𝛼𝑛 may be constructed using the formula

11
 ( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1 = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

In particular, when n=3 ,we have

𝛼1 =𝛽1

(𝛼2 |𝛽2 )
𝛼2 =𝛽2- α1
||𝛼𝑘 ||2

(𝛽3 |𝛼1 ) (𝛼2 |𝛽3 )

𝛼3 =𝛽3 - α1 - 𝛼
||𝛼1 ||2 ||𝛼𝑘 ||2 2

Corollary :

Every finite dimensional inner product space has an orthonormal basis.

Proof :

Let V be a finite dimensional inner product space and { 𝛽1, … , 𝛽𝑛 } a basis for V. Apply the

gram Schmidt orthogonalization process to construct an orthogonal basis , simply replace each
𝛼𝑘
vector 𝛼𝑛 by .
||𝛼𝑘 ||

Gram-Schmidt process can be used to test for linear dependence . For suppose 𝛽1, … , 𝛽𝑛 are

linearly independent vectors in an inner product space; to exclude a trivial case , assume that

β≠0. Let m be largest integers for which 𝛽1, … , 𝛽𝑚 are independent. Then 1≤m˂n.

Let 𝛼1 , 𝛼2 , … 𝛼𝑚 be the vectors obtained by applying the orthogonalization process to

(𝛽𝑚+1 |𝛼𝑘 )
𝛽1, … , 𝛽𝑚 . Then the vector 𝛼𝑚+1 given by 𝛼𝑚+1 = 𝛽𝑚+1–∑𝑚
𝑘=1 𝛼𝑘 is necessarily 0.
||𝛼𝑘 ||2

For 𝛼𝑚+1 is in the subspace spanned by 𝛼1 , 𝛼2 , … 𝛼𝑚 and orthogonal to each of the vectors ,

(β|𝛼𝑘 )
hence it is 0 as β=∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘 . Conversely, if 𝛼1 , 𝛼2 , … 𝛼𝑚 are different from 0 and 𝛼𝑚+1 =0,
𝑘 ||

then 𝛽1, … , 𝛽𝑚+1 are linearly independent .

Definition:

A best approximation to β  V by vectors in a subspace W of V is a vector α  W such that

 −    −  for every vector   W .

12
Theorem

Let W be a subspace of an inner product space V and let  V .

1. The vector  W is a best approximation to  V by vectors in W if and

only if  − is orthogonal to every vector in W .

2. If a best approximation to  V by vectors in W exists, it is unique.

3. If W is finite-dimensional and {𝛼1, 𝛼2 , … 𝛼𝑛 } is any orthonormal basis for W ,

then the vector

 =
n
(  |  k )
k
2 k
k =1

is the (unique) best approximation to  by vectors in W.

Definition:

Let V be an inner product space and S be any set of vectors in V. The orthogonal complement

of S is the set S ⊥ of all vectors in V which are orthogonal to every vector in S.

That is, S ⊥ =  V : ( |  ) = 0,    S

Definition:

Whenever the vector α in the above theorem exists it is called the orthogonal projection of

β on W. If every vector in V has an orthogonal projection on W, the mapping that assigns to

each vector in V its orthogonal projection on W is called the orthogonal projection of V on W.

Corollary :

Let V be an inner product space and W a finite dimensional subspace and E be the

orthogonal projection of V on W. Then the mapping

β →β – Eβ

is the orthogonal projection of V on W ⊥ .

13
Proof :

Let β  V . Then β – Eβ  W ⊥ , and for any γ  W ⊥ , β – γ = E β+(β – Eβ – γ)

Since Eβ  W and β – Eβ – γ  W ⊥ ,

It follows that

||𝛽 – 𝛾||2 = (Eβ+(β – Eβ – γ) ,E β+(β –Eβ – γ))

= ||𝐸𝛽||2+||𝛽 – 𝐸𝛽 – 𝛾||2

≥ ||𝛽 – (𝛽 – 𝐸𝛽)||2

with strict inequality when γ≠ β – Eβ . Therefore, β – Eβ is the best approximation to β by

vectors in W ⊥ .

Theorem

Let W be a finite dimensional subspace of an inner product space V and let E be the orthogonal

projection of V on W. Then E is an idempotent linear transformation of V onto W , W ⊥ is the

null space of E , and V= W ⨁ W ⊥ .

Proof

Let β be an arbitrary vector in V. Then Eβ is the best approximation to β that lies in W .

In particular, Eβ =β when β is in W . Therefore, E(Eβ) =Eβ for every β in V; that is, E is

idempotent : 𝐸 2 = E . To prove that E is linear transformation, let α and β be any vectors in V

and c an arbitrary scalar ,Then by theorem,

α-Eα and β-Eβ are each orthogonal to every vector in W . Hence the vector

c(α-Eα)+(β-Eβ)=(cα +β)-(cEα +Eβ)

Also belongs to W ⊥ . Since cEα+ Eβ is a vector in W , it follows from theorem that

E(cα+ β)= cEα+ Eβ.

Again let β be any vector in V. Then Eβ is the unique vector in W such that β-Eβ is in W ⊥ .

Thus Eβ=0 when β is in W ⊥ .

Conversely, β is in W ⊥ when Eβ=0. Thus W ⊥ is the null space of E.

14
The equation ,

β = E β+β – Eβ

shows that V = W + W ⊥ ; moreover W  W ⊥ = {0} ; for if α is a vector in W  W ⊥ ,then

( |  ) =0. Therefore, α=0 and V is the direct sum of W and W ⊥ .

Corollary :

Under the conditions of theorem, I − E is the orthogonal projection of V on W ⊥ .

It is an independent linear transformation of V onto W ⊥ with null space W .

Proof :

We have seen that the mapping β →β- E β is the orthogonal projection of V on W ⊥ .

Since E is a linear transformation , this projection W ⊥ is the linear transformation I − E from

its geometric properties one sees that I − E is an idempotent .Transformation of V onto W .

This also follows from the computation ( I − E )( I − E )= I − E - E +𝐸 2

=I −E

Moreover , ( I − E )β =0 If and only if β = Eβ , and this is the case if and only if β is in W .

Therefore W is the null space of I − E .

15
 BIBLIOGRAPHY

● Balmohan V .Limaye :Functional analysis

● Kenneth Hoffman,Ray Kunze:Linear Algebra,second edition

● Stephen H.Friedberg, Arnold J.Insel, Lawrence E.Spence : Linear Algebra, Fourth

edition

16
DIRECT PRODUCTS AND FINITELY GENERATED ABELIAN
GROUPS

Project report submitted to

The Kannur University

for the award of the degree

of

Bachelor of Science

by

ANILA BIJU

DB18CMSR18

Under the guidance of

MS. Sneha P Sebastian

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report ‘DIRECT PRODUCTS AND FINITELY

GENERATED ABELIAN GROUPS’ is the bonafide project of ANILA BIJU who
carried out the project under my supervision.

Mrs.Riya Baby Ms.Sneha P Sebastian

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, ANILA BIJU, hereby declare that this project report entitled ‘DIRECT PRODUCTS
AND FINITELY GENERATED ABELIAN GROUPS’ is an original record of studies and
bonafide project carried out by me during the period from November 2019 to March 2020, under
the guidance of Ms.Sneha P Sebastian, Department of Mathematics, Don Bosco Arts and
Science College, Angadikadavu and has not been submitted by me elsewhere for the award of
any degree, diploma, title or recognition, before.

ANILA BIJU
Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great help to me
during the course of my dissertation. First and foremost I thank the almighty, for his blessing and
protection during the period of this work. I express my thanks to Dr. Fr. Francis Karakkatt,
Principal, for support in the completion of this dissertation. I express my gratitude to Ms.Sneha P
Sebastian, my project supervisor, for the constant encouragement, valuable guidance and timely
corrections, which made this work a success.

I am also indebted to all my classmates and friends who supported me throughout the
study. I would like to express my thanks to my parents and dear ones for their constant
encouragement and support. I also thank all those who helped me directly or indirectly to
complete this project.

ANILA BIJU
 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 preliminary 2–3
03 Chapter 1 4 – 12
04 Chapter 2 13 – 17
05 Conclusion 18
06 Bibliography 19
 INTRODUCTION

In mathematics, a group is a set equipped with a binary operation that combines any two
elements to form a third element in such a way that the three conditions called group axioms are
satisfied , namely associativity , identity and invertability.

Let us take a moment to review our present stockpile of groups. Starting with finite
groups, we have the cyclic group ℤ𝑛 ,the symmetric group 𝑆𝑛 , and the alternating group 𝐴𝑛 for
each positive integer n. We also have the dihedral group 𝐷𝑛 and klein 4-group . Of course we
know that subgroups of these groups exists. Turning to infinite groups , we have ℤ, ℝ, ℂ under
addition , and their non zero elements under multiplication we also have the group 𝑆𝐴 of all
permutation of an infinite set 𝐴 , as well as various groups formed from matrices .

One purpose of this section is to show a way to use known groups as building blocks to
form more groups. Given two groups 𝐺 and 𝐻, it is possible to construct a new group from the
cartesian product of 𝐺 and 𝐻 . Conversely , given a large group , it is sometimes possible to
decompose the group ; that is , a group is sometimes isomorphic to the direct product of two
smaller groups. Rather than studying a large group , it is often easier to study the component
group of that group.

1
 PRELIMINARY

Groups : A non empty set 𝐺 together with an operation ∗ is said to be a group , denote by
(𝐺 ,∗) , if it satisfy the following axioms.

• Closure property
• Associative property
• Existence of identity
• Existence of inverse

Abelian group

A group (𝐺 ,∗) is said to be abelian if it satisfies commutative law .

Finite group

If the underlying set G of the group (𝐺 ,∗) consist of finite number of elements , then the
group is finite group .

Infinite group

A group that is not finite is an infinite group .

Order of a group : The number of elements in a finite group is called the order of the group ,
denoted by 𝑂(𝐺 ) .

Example

Show that the set of integers ℤ is a group with respect to the operation of addition of
integers.

ℤ = {… … … . −3, −2, −1,0,1,2,3, … … … }

Since the addition of two integers gives an integer , it satisfy closure property .

2
If 𝑎, 𝑏, 𝑐 𝜖 ℤ then the (𝑎 + 𝑏) + 𝑐 = 𝑎 + (𝑏 + 𝑐) , hence associativity holds .

There is a number 0 𝜖 ℤ such that 0 + 𝑎 = 𝑎 + 0 , hence identity exists

If 𝑎 𝜖 ℤ then there exists – 𝑎 𝜖 ℤ , such that −𝑎 + 𝑎 = 0 = 𝑎 + −𝑎

Therefore inverse exist .

Therefore ℤ is a group under addition .

Subgroup

A subset 𝐻 of 𝐺 is said to be a subgroup of 𝐺 if 𝐻 itself is a group under the same operation in

𝐺.

There are two different types of group structure of order 4 .

ℤ4 = { 0,1,2,3}

Klein 4 – group , 𝑉 = {𝑒, 𝑎, 𝑏, 𝑐}

Cyclic group

A group 𝐺 is cyclic if there is some element ‘𝑎’ in 𝐺 that generate 𝐺. And the element ‘𝑎’ is
called generator of 𝐺.

Group Homomorphism

A function Ѱ: 𝐺 → 𝐺′ is a group homomorphism ( or simply homomorphism ).

If Ѱ(𝑎𝑏) = Ѱ(𝑎) Ѱ(𝑏) hold for all 𝑎 , 𝑏 ∈ 𝐺 , is called homomorphism property .

Isomorphism

A one to one and onto homomorphism Ѱ: 𝐺 → 𝐺′ is called an isomorphism .

3
 CHAPTER – 1

DIRECT PRODUCT OF GROUPS

Definition

The Cartesian product of sets 𝑆, 𝑆2 , … … . , 𝑆𝑛 is the set of all ordered n-tuples (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ),
where 𝑎𝑖 ∈ 𝑆𝑖 for 𝑖 = 1,2,3, … … . , 𝑛. The Cartesian product is denoted by either

𝑛
𝑆1 × 𝑆2 × … … .× 𝑆𝑛 or by Π𝑖=1 𝑆𝑖 .

Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups and let us use multiplicative notation for all the group operations.

If we consider 𝐺𝑖 as a set , 𝑖 = 1,2, … … . 𝑛 . we have the products 𝐺1 × 𝐺2 × … … . ,× 𝐺𝑛 we

𝑛 𝑛
denote it by Π𝑖=1 𝐺𝑖 . This product is called direct-product of groups. We can make Π𝑖=1 𝐺𝑖 into a
group by means of a binary operation of multiplication by components.

Theorem
𝑛
Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups. For (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) and (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) in Π𝑖=1 𝐺𝑖 define ;

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

𝑛
Then Π𝑖=1 𝐺𝑖 is a group.

4
Proof

We have ,

𝑛
Π𝑖=1 𝐺𝑖 = {(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∶ 𝑎𝑖 ∈ 𝐺𝑖 }

(1) Closure property

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

And we have ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

Here 𝑎𝑖 ∈ 𝐺𝑖 and 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

∵ 𝐺𝑖 is a group , 𝑎𝑖 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

𝑛
⇒ (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

𝑛
i.e. Π𝑖=1 𝐺𝑖 is closed under the binary operation.

(2) Associativity
𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ), (𝑐1, 𝑐2 , … … . , 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

We have,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

𝑛
= (𝑎1 𝑏1 𝑐1 , 𝑎2 𝑏2 𝑐2 , … … . , 𝑎𝑛 𝑏𝑛 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

[(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

= [𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )
= [(𝑎1 𝑏1 )𝑐1 , (𝑎2 𝑏2 )𝑐2, … … . , (𝑎𝑛 𝑏𝑛 )𝑐𝑛 ]
= [𝑎1 (𝑏1 𝑐1 ), 𝑎2 (𝑏2 𝑐2), … … . , 𝑎𝑛 (𝑏𝑛 𝑐𝑛 )]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[𝑏1 𝑐1, 𝑏2 𝑐2 , … … . , 𝑏𝑛 𝑐𝑛 ]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )]

5
Hence associativity holds.

(3) Existence of identity

If 𝑒𝑖 is the identity element in 𝐺𝑖 .

Then,

𝑛
(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Also for ,

𝑛
(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖 ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) = (𝑎1 𝑒1 , 𝑎2 𝑒2 , … … . , 𝑎𝑛 𝑒𝑛 )

= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )

𝑛
∴ (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) is the identity element ‘𝑒’ in Π𝑖=1 𝐺𝑖

(4) Existence of inverse

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Here 𝑎𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛.

Since 𝐺𝑖 is a group ,

∃ an inverse element 𝑎𝑖−1 in 𝐺𝑖 : 𝑎𝑖 𝑎𝑖−1 = 𝑒𝑖 𝑖 = 1,2, … … . , 𝑛

𝑛
Clearly, (𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) ∈ Π𝑖=1 𝐺𝑖 &

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 )

𝑛
Hence Π𝑖=1 𝐺𝑖 is a group.

6
Note
𝑛
If the operation of each 𝐺𝑖 is a commutative. We sometimes use additive notation in Π𝑖=1 𝐺𝑖 and
𝑛
refer to Π𝑖=1 𝐺𝑖 as the direct sum of the group 𝐺𝑖 . The notation ⨁𝑛𝑖=1 𝐺𝑖 , especially with abelian
groups with operation +.

The direct sum of abelian groups 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 may be written 𝐺1 ⨁𝐺2 ⨁ … … ⨁𝐺𝑛

• Direct product of abelian group is abelian

Example

Q. Check whether ℤ2 × ℤ3 is cyclic or not.

ℤ2 = {0,1}

ℤ3 = {0,1,2}

ℤ2 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2)}

Consider,

1(1,1) = (1,1)

2(1,1) = (1,1) + (1,1) = (0,2)

3(1,1) = (1,1) + (1,1) + (1,1) = (1,0)

4(1,1) = 3(1,1) + (1,1) = (1,0) + (1,1) = (0,1)

5(1,1) = 4(1,1) + (1,1) = (0,1) + (1,1) = (1,2)

6(1,1) = 5(1,1) + (1,1) = (1,2) + (1,1) = (0,0)

∴ (1,1) is a generator of ℤ2 × ℤ3

∴ ℤ2 × ℤ3 is a cyclic group generated by (1,1).

7
Q. Check whether ℤ3 × ℤ3 is cyclic or not.

ℤ3 = {0,1,2}

ℤ3 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2), (2,0), (2,1), (2,2)}

1(0,1) = (0,1)

2(0,1) = (0,2)

3(0,1) = (0,3) = (0,0) ∴ order (0,1) = 3

1(0,2) = (0,2)

2(0,2) = (0,4) = (0,1)

3(0,2) = (0,6) = (0,0) ∴ order (0,2) = 3

Every element added to itself three times gives the identity. Thus no element can
generate the group. Hence ℤ3 × ℤ3 is not cyclic.

similarly ℤ𝑚 × ℤ𝑚 is not cyclic for any 𝑚.

Theorem

The group ℤ𝑚 × ℤ𝑛 is cyclic and is isomorphic to ℤ𝑚𝑛 if and only if 𝑚 and 𝑛 are relatively
prime, that is, the gcd of 𝑚 and 𝑛 is 1.

Proof

Suppose ℤ𝑚 × ℤ𝑛 is cyclic and isomorphic to ℤ𝑚𝑛 .

To show that 𝑚 and 𝑛 are relatively prime.

Suppose not, let d be the 𝑔𝑐𝑑 of 𝑚 and 𝑛.

So that 𝑑 > 1

8
 𝑚𝑛
Consider , which is an integer since 𝑑|𝑚 and 𝑑|𝑛
𝑑

𝑚𝑛
Let (𝑟, 𝑠) be an arbitrary element of ℤ𝑚 × ℤ𝑛 , add (𝑟, 𝑠) repeatedly times
𝑑

(𝑟, 𝑠) + (𝑟, 𝑠)+, … … . , +(𝑟, 𝑠) } 𝑚𝑛

𝑑
𝑡𝑖𝑚𝑒𝑠 = (0,0)

∴ no element of ℤ𝑚 × ℤ𝑛 having order 𝑚𝑛. ∴ no element of ℤ𝑚 × ℤ𝑛 can generate ℤ𝑚 × ℤ𝑛

which is not possible. ∵ ℤ𝑚 × ℤ𝑛 is cyclic. Hence 𝑔𝑐𝑑(𝑚, 𝑛) = 1.

i.e. 𝑚 and 𝑛 are relatively prime.

Conversely, suppose 𝑚 and 𝑛 are relatively prime, i.e. gcd(𝑚, 𝑛) = 1

To show that ℤ𝑚 × ℤ𝑛 is cyclic.

If ℤ𝑚 × ℤ𝑛 is cyclic, then it is isomorphic to ℤ𝑚𝑛 , ∵ ℤ𝑚 × ℤ𝑛 has 𝑚𝑛 elements.

Consider the cyclic subgroup of ℤ𝑚 × ℤ𝑛 generated by the element (1,1).The order of this
cyclic subgroup is the smallest power of (1,1),that gives the identity (0,0). Here taking a power
of (1,1) in our additive notation will involve adding (1,1) to itself repeatedly.

Consider (1,1) + (1,1)+, … … . , +(1,1)

If we add first coordinates 𝑚 times , we get zero.

∴ order of first coordinate = 𝑚.

Similarly , Order of second coordinate = 𝑛.

The two coordinates together become zero. If we add them 𝑙𝑐𝑚(𝑚, 𝑛) times.

∵ gcd(𝑚, 𝑛) = 1, We get the 𝑙𝑐𝑚 = 𝑚𝑛.

i.e. (1,1) generates a cyclic subgroup of ℤ𝑚 × ℤ𝑛 of order 𝑚𝑛 , which is the order of the whole
group.

⇒ ℤ𝑚 × ℤ𝑛 =< (1,1) >

9
 ⇒ ℤ𝑚 × ℤ𝑛 is cyclic.

Corollary

The group ⨅𝑛𝑖=1 ℤ𝑚𝑖 is cyclic and isomorphic to ℤ𝑚1 𝑚2 …….𝑚𝑛 if and only if the numbers 𝑚𝑖 for
𝑖 = 1,2, … … . , 𝑛 are such that the 𝑔𝑐𝑑 of any two of them is 1.

Example

If 𝑛 is written as a product of powers of distinct prime numbers , as in ,

𝑛 = (𝑝1 )𝑛1 . (𝑝2 )𝑛2 … … . (𝑝𝑛 )𝑛𝑟

Then ℤ𝑛 is isomorphic to ℤ(𝑝1 )𝑛1 × ℤ(𝑝2 )𝑛2 × … … .× ℤ(𝑝𝑟 )𝑛𝑟 .

In particular , ℤ72 is isomorphic to ℤ8 × ℤ9 .

Consider set of integers ℤ, cyclic subgroup of ℤ is of the form 𝑛ℤ , 𝑛 ∈ ℤ. Consider 2ℤ and 3ℤ ,

then < 2 > ∩ < 3 > = < 6 >

∴ if we take 𝑟ℤ , 𝑠ℤ of ℤ , then the 𝑙𝑐𝑚(𝑟, 𝑠) =generator of < 𝑟 > ∩ < 𝑠 >

Using this we can define the 𝑙𝑐𝑚 of the positive integers.

Definition

Let 𝑟1 , 𝑟2 , … … . , 𝑟𝑛 be positive integers. Their least common multiple (abbreviated lcm ) is the
positive generator of the cyclic group of all common multiples of the 𝑟𝑖 , that is the cyclic group
of all integers divisible by each 𝑟𝑖 for 𝑖 = 1,2, … … . , 𝑛.

10
Theorem

Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ ⨅𝑛𝑖=1 𝐺𝑖 .

If 𝑎𝑖 is of finite order 𝑟𝑖 in 𝐺𝑖 , then the order of (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) in ⨅𝑛𝑖=1 𝐺𝑖 is equal to the least
common multiple of all the 𝑟𝑖 .

Proof

Given,

𝑟
order of 𝑎1 = 𝑟1 ⇒ 𝑎11 = 𝑒1 in 𝐺1

𝑟
order of 𝑎2 = 𝑟2 ⇒ 𝑎22 = 𝑒2 in 𝐺2

.
.
.
𝑟
order of 𝑎𝑛 = 𝑟𝑛 ⇒ 𝑎𝑛𝑛 = 𝑒𝑛 in 𝐺𝑛 .

We have to find a power 𝑘 for (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ).

So that (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )𝑘 = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ).

The power must simultaneously be a multiple of 𝑟1 , multiple of 𝑟2 and so on. But 𝑘 is the least
positive integers having the above property.

∴ 𝑘 = 𝑙𝑐𝑚(𝑟1 , 𝑟2 , … … . , 𝑟𝑛 ).

Q. Find the order of (8,4,10) in the group ℤ12 × ℤ60 × ℤ24 .

𝑂(8) = 3 in 𝑍12

11
 𝑂(4) = 15 in 𝑍60

𝑂(10) = 12 in 𝑍24

𝑂(8,4,10) = 𝑙𝑐𝑚(3,15,12) = 60

Q. Find a generator of ℤ × ℤ2

ℤ × ℤ2 = {(𝑛, 0), (𝑛, 1): 𝑛 ∈ ℤ}

(𝑛, 0) = 𝑛(1,0)
(𝑛, 1) = (𝑛, 0) + (0,1) = 𝑛(1,0) + (0,1)

∴ ℤ × ℤ2 is generated by {(1,0), (0,1)}

In general , ℤ × ℤ𝑛 is generated by ,

{(1,0,0, … … . ,0), (0,1,0, … … . ,0), … … . , (0,0, … … . ,1)}

Q. Find the order of (3,10,9) in (ℤ4 , ℤ12 , ℤ15 )

𝑂(3) = 4 in ℤ4

𝑂(10) = 6 in ℤ12

𝑂(9) = 5 in ℤ15

∴ 𝑂(3,10,9) = 𝑙𝑐𝑚(4,6,5)
= 60

∴ order of (3,10,9) in ℤ4 × ℤ12 × ℤ15 is 60.

12
 CHAPTER-2

FUNDAMENTAL THEOREM OF FINITELY GENERATED ABELIAN

GROUPS

Every finitely generated abelian group 𝐺 is isomorphic to a direct product of cyclic groups in the
form,

ℤ(𝑝1 )𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛 × ℤ × ℤ × ℤ × … … .× ℤ

Where the 𝑝𝑖 are primes , not necessarily distinct and the 𝑟𝑖 are positive integers.

Remark

• The direct product is unique except for possible rearrangement of the factors.
• The number of factors ℤ is unique and this number is called Betti number.

Example

Find all abelian groups , upto isomorphism of order

1)8 , 2)16 , 3)360

(1) Order 8

8=1×8

8 = 2 × 4 = 2 × 22

8 = 2×2×2

3 non-isomorphic groups are ℤ8 , ℤ2 × ℤ4 ,

ℤ2 × ℤ2 × ℤ2
13
(2) Order 16

16 = 1 × 16 = 1 × 24

16 = 2 × 8 = 2 × 23

16 = 4 × 4 = 22 × 22

16 = 2 × 2 × 2 × 2

16 = 2 × 2 × 22

ℤ16 , ℤ2 × ℤ8 , ℤ4 × ℤ4 , ℤ2 × ℤ2 × ℤ2 × ℤ2 , ℤ2 × ℤ2 × ℤ4

(3) Order 360

360 = 22 ⋅ 32 ⋅ 5

Possibilities are,

1) ℤ8 × ℤ9 × ℤ5

2) ℤ2 × ℤ4 × ℤ9 × ℤ5

3) ℤ2 × ℤ2 × ℤ2 × ℤ9 × ℤ5

4) ℤ8 × ℤ3 × ℤ3 × ℤ5

5) ℤ2 × ℤ4 × ℤ3 × ℤ3 × ℤ5

6) ℤ2 × ℤ2 × ℤ2 × ℤ3 × ℤ3 × ℤ5

Definition

A group 𝐺 is decomposable if it is isomorphic to a direct product of two proper non-trivial

subgroups , otherwise 𝐺 is indecomposable.

14
Example

ℤ6 is decomposable while ℤ5 is indecomposable.

ℤ6 is isomorphic to ℤ2 × ℤ3

ℤ𝑚𝑛 is isomorphic to ℤ𝑚 × ℤ𝑛 , if 𝑚 and 𝑛 are prime.

Theorem

The finite indecomposable abelian groups are exactly the cyclic groups with order a power of a
prime.

Proof

Let 𝐺 be a finite indecomposable abelian group ∵ 𝐺 is finitely generated , we can apply

fundamental theorem of finitely generated abelian groups.

∴ 𝐺 ≅ ℤ(𝑝)𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

∵ 𝐺 is indecomposable and ℤ(𝑝𝑖 )𝑟𝑖 ’s are proper subgroups we get in the above , there is only one
factor say ℤ(𝑝𝑖 )𝑟𝑖 which is cyclic group with order a prime power.

Theorem

If 𝑚 divides the order of a finite abelian group , then 𝐺 has a subgroup of order 𝑚.

Proof

Given 𝐺 is a finite abelian group.

∴ we can apply Fundamental Theorem ,

15
Hence,

𝐺 ≅ ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛

1 2 𝑛

Here all primes need not be distinct.

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 . 𝑝22 … … 𝑝𝑛𝑛

Let 𝑚 is a +𝑣𝑒 integer which divides 𝑂(𝐺).

0 ≤ 𝑠𝑖 ≤ 𝑟𝑖 By theorem , “ let 𝐺 be a cyclic group with 𝑛 elements and generated by 𝑎. Let

𝑛
𝑏 ∈ 𝐺 , 𝑏 = 𝑎𝑠 , then ‘ 𝑏’ generates a cyclic subgroup 𝐻 of 𝐺 containing elements , where
𝑑

𝑑 = gcd(𝑛, 𝑠).”

𝑟
𝑟 −𝑠 𝑝𝑖 𝑖
𝑝𝑖 𝑖 𝑖 generates a cyclic subgroup of ℤ𝑝𝑟𝑖 having order 𝑟 𝑟 −𝑠𝑖
𝑖 𝑔𝑐𝑑(𝑝𝑖 𝑖 ,𝑝𝑖 𝑖 )

𝑟
𝑝𝑖 𝑖 𝑠
= 𝑟 −𝑠𝑖 = 𝑝𝑖 𝑖
𝑝𝑖 𝑖

𝑟 −𝑠𝑖 𝑠
∴ 𝑂(< 𝑝𝑖 𝑖 >) = 𝑝𝑖 𝑖

𝑟 −𝑠1 𝑠
i.e. < 𝑝11 > is a subgroup of ℤ𝑝𝑟1 having order 𝑝11 .
1

𝑟 −𝑠2 𝑠
< 𝑝22 > is a subgroup of ℤ𝑝𝑟2 having order 𝑝22 .
2

…………………………………………………………

𝑟 −𝑠𝑛 𝑠
< 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟𝑛 having order 𝑝𝑛𝑛 .
𝑛

𝑟 −𝑠1 𝑟 −𝑠2 𝑟 −𝑠𝑛

∴ < 𝑝11 > × < 𝑝22 > × … … .× < 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛
1 2 𝑛
𝑠 𝑠 𝑠
having order 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛 = 𝑚.

16
Theorem

If 𝑚 is a square free integer , that is 𝑚 is not divisible by the square of any prime . Then every
abelian group of order 𝑚 is cyclic.

Proof

Let 𝑚 be a square free integer , then 𝑝𝑖 ⫮ 𝑚 for every 𝑖 greater than 1 for a prime 𝑝.

Given 𝐺 is a finite abelian group having order 𝑚 , by fundamental theorem , then

𝐺 ≅ ℤ(𝑝1 )𝑟1 × ℤ(𝑝2 )𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛

∵ 𝑂(𝐺) is a square free integer , the only possibility

𝑟1 = 𝑟2 = … … . . = 𝑟𝑛 = 1

Then,

𝐺 ≅ ℤ𝑝1 × ℤ𝑝2 × … … .× ℤ𝑝𝑛

≅ ℤ𝑝1 ,𝑝2 ,…….,𝑝𝑛 , which is cyclic.

Example

15 is a square free integer. So an abelian group of order 15 is cyclic.

17
 CONCLUSION

Direct product of groups is the product 𝐺1 × 𝐺2 , … … . 𝐺𝑛 ,where each 𝐺𝑖 is a set. We have

discussed about definition and some properties related to the direct product of groups. The
fundamental theorem of finitely generated abelian group helped us to get a deeper understanding
about the topic. The theorems gives us complete structural information about abelian group, in
particular finite abelian group. We have also discussed some examples in order to develope more
intrest in algebra.

18
 BIBLIOGRAPHY

• A first course in Abstract Algebra – 7th edition.

John . B . Fraleigh

• Contomporary Abstract Algebra – 7th edition.

Joseph . A . Gallian

• Topics In Algebra – 2nd edition

I . N . Herstein

19
DON BOSCO ARTS & SCIENCE COLLEGE
ANGADIKADAVU

DEPARTMENT OF MATHEMATICS
2018-2021

Project Report on

INNER PRODUCT SPACES

 DEPARTMENT OF MATHEMATICS

DON BOSCO ARTS & SCIENCE COLLEGE

ANGADIKADAVU

MARCH 2021

Project Report on

INNER PRODUCT SPACES

Dissertation submitted in the partial fulfilment

of the requirement for the award of

Bachelor of Science in Mathematics of

Kannur University

Name : ANJALY BABY

Roll No. : DB18CMSR19

Examiners: 1.

2.
 𝐊𝐀𝐍𝐍𝐔𝐑 𝐔𝐍𝐈𝐕𝐄𝐑𝐒𝐈𝐓𝐘

𝐁𝐎𝐍𝐀𝐅𝐈𝐃𝐄 𝐂𝐄𝐑𝐓𝐈𝐅𝐈𝐂𝐀𝐓𝐄

Certified that this project report on " INNER PRODUCT SPACES” is the bonafide
work of ANJALY BABY who carried out the project work under my supervision.

Mrs. Riya Baby Mr. Anil M V

Head of Department Supervisor
 𝐃𝐄𝐂𝐋𝐀𝐑𝐀𝐓𝐈𝐎𝐍

I, ANJALY BABY hereby declare that the project work entitled ‘INNER
PRODUCT SPACES′ has been prepared by me and submitted to Kannur University in
partial fulfilment of requirement for the award of Bachelor of Science is a record of
original work done by me under the supervision of Mr. ANIL M V, Assistant Professor,
Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I, also declare that this Project work has been submitted by me fully or partially for
the award of any Degree, Diploma, Title or recognition before any authority.

Place ∶ Angadikadavu
ANJALY BABY
Date ∶ DB18CMSR19
 𝐀𝐂𝐊𝐍𝐎𝐖𝐋𝐄𝐃𝐆𝐄𝐌𝐄𝐍𝐓

Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express mydeepest gratitude to
people along the way.
No words can adequately express the sense of gratitude, still I try to express my
heartfelt thanks through words. The outset, I am deeply indebted to my project
supervisor Mr. ANIL M.V, Assistant Professor, Department of Mathematics, Don
Bosco Arts and ScienceCollege, Angadikadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career.I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikadavu.
I could not name many who sincerely supported and helped for the successful
completion of this Project. It is my pleasure and duty to thank each and every one
of them who walked with me.

ANJALY BABY
 Contents

1. INTRODUCTION 1

2. PRELIMINARIES 2

3. CHAPTER 1 : INNER PRODUCT SPACES 6

4. CHAPTER 2 : ORTHOGONAL SETS 9

5. BIBLIOGRAPHY 16
INTODUCTION

In linear algebra, an inner product space is a vector space with an additional structure

called an inner product. This additional structure associates

each pair of vectors in the space with a scalar quantity known as the inner product of the

vectors. Inner products allow the rigorous introduction of intuitive geometrical notions

such as the length of a vector or the angle between two vectors. They also provide the

means of defining orthogonality between vectors (zero inner product). Inner product

spaces generalize Euclidean spaces (in which the inner product is the dot product, also

known as the scalar product) to vector spaces of any (possibly infinite) dimension and are

studied in functional analysis. The first usage of the concept of a vector space with an

inner product is due to Peano, in 1898.

An inner product naturally induces an associated norm, thus an inner product space is also

a normed vector space. A complete space with an inner product is called a Hilbert space.

An (incomplete) space with an inner product is called a pre-Hilbert space.

1
 PRELIMINARIES

LINEAR SPACES

Definition 1: A linear (vector) space X over a field F is a set of elements

together with a function, called addition, from X ×X into X and a function

called scalar multiplication, from F × X into X which satisfy the following

conditions for all x, y, z ∈ X and α, β ∈ F;

i. (x + y) + z = x + (y + z)

ii. x +y = y +x

iii. There is an element 0 in X such that x + 0 = x for all x ∈ X.

iv. For each x ∈ X there is an element −x ∈ X such that x + (−x) = 0.

v. (x + y) = αx + αy

vi. (α + β)x = αx + βx

vii. α(βx) = (αβ)x

viii. 1 · x = x.

Properties i to iv imply that X is an abelian group under addition and v to vi

relate the operation of scalar multiplication to addition X and to addition and

multiplication in F.

Examples:

(a) Vn(R). The vectors are n-tuples of real numbers and the scalars are real

2
 numbers with addition and scalar multiplication defined by

(1 ,···,  n ) + ( 1 ,···,  n ) = (1 + 1 ,···,  n +  n ) (1)

 (1 ,···,  n ) = ( 1 ,···,  n ) (2)

Vn(R) is a linear space over R. Similarly, the set of all n-tuples of complex

numbers with the above definition of addition and multiplication is a linear

space over C and is denoted as Vn(C).

(b) The set of all functions from a nonempty set X into a field F with addition and

scalar multiplication defined by

[f + g](t) = f (t) + g(t) and [αf ](t) = αf (t); f, g ∈ X, t ∈ T (3)

is a linear space.

Let T = N the set of all positive integers and X is the set of all sequences of

elements F with addition and scalar multiplication defined by

( n +  n ) = ( n +  n ) (4)

 ( n ) = (  n ) (5)

denoted as V∞(F), form a linear space.

3
METRIC SPACES

Remember the distance function in the Euclidean space Rn.

Let x, y, z ∈ Rn, then

(1) |x − y| ≥ 0; |x − y| = 0 if and only if x = y ;

(2) |x − y| = |y − x|;

(3) |x − y| ≤ |x − z| + z − y|.

Definition 2: A metric or distance function on a set X is a real valued function

d defined on X × X which has the following properties: for all x, y, z ∈ X.

(1) d(x, y) ≥ 0; d(x, y) = 0 if and only if x = y;

(2) d(x, y) = d(y, x);

(3) d(x, y) ≤ d(x, z) + d(z, y)

A metric space (X, d) is a nonempty set X and a metric d defined on X.

Examples: In addition to the Euclidean spaces let us have the following examples.

Here all functions are assumed to be continuous. Let Lp denotes a set of complex

p
valued functions in Rn such that f is integrable. Let us recall some results

concerning such functions.

Höder’s Inequality: If p > 1, 1/q = 1 − 1/p

 | fg |  [ | f | ] [  | g |q ]1/ q .
p 1/ p

Minkowski’s Inequality: If p ≥ 1,

[ | f + g | p ]1/ p  [ | f | p ]1/ p + [  | g | p ]1/p

4
If xk and yk for k = 1, … , m are complex numbers, let f (t) = |xk| and g(t) =

|yk| for t ∈ [k, k + 1] and f (t) = 0 = g(t) for t∈ [1, m + 1]. Then we obtain the

summation form of the above inequalities from the integral form

Hölder’s Inequality

1/ p 1/ q
m m p
 m q

 xk yk   xk    yk 
k =1  k =1   k =1 

Minkowski’s Inequality:

p 1/ p p 1/ p p 1/p
m  m  m 
  xk + yk     xk  +   yk 
 k =1   k =1   k =1 

NORMED LINEAR SPACES

Definition 3. A norm on X is a real valued function, whose value at x is denoted

by ||x||, satisfying the following conditions for all x, y ∈ X and α ∈ F;

(1) ||x|| > 0 if x ≠ 0

(2) ||αx|| = |α|||x||

(3) ||x + y|| ≤ ||x|| + ||y||.

A linear space X with a norm defined on it is called a normed linear space.

Example: l p space. On the linear space Vn(F), define

n
x = [ |  i | p ]1/ p
k =1

where p ≥ 1 is any real number and x = (1 ,···,  n ) . This defines a norm (called p-

p
norm) on Vn(F). This space is called l space .
5
 CHAPTER 1

INNER PRODUCT SPACES

INNER PRODUCTS

Let 𝐹 be the field of real numbers or the field of complex numbers, and V a vector space over

F an inner product on V is a function which assigns to each ordered’ pair of vectors 𝛼, 𝛽 in V

a scalar (𝛼|𝛽) in 𝐹 in such a way that for all 𝛼, 𝛽, γ in V and all scalars c.

(a) (𝛼 + 𝛽|𝛾) = (𝛼|𝛾) + (𝛽|𝛾) ;

(b) (c𝛼|𝛽) = 𝑐(𝛼|𝛽) ;

̅̅̅̅̅ ), the bar denoting complex conjugation

(c) (𝛽|𝛼) = (𝛼|𝛽

(d) (𝛼|𝛼) > 0 if 𝛼 ≠ 0

It should be observed that conditions (a), (b) and (c) implies that

(𝑒) = (𝛼 ∣ 𝑐𝛽 + 𝛾) = (𝑐̅(𝛼|𝛽) + (𝛼|𝛾)

One other point should be made. When 𝐹 is the field 𝑅 of real nunbers. The complex conjugates

appearing in (c) and (e) are superflom. However, in the complex case they are necessary for

the consistency of the conditions. Without these complex conjugates we would have the

contradiction

(𝛼|𝛼) > 0 and (𝑖𝛼 ∣ 𝑖𝛼) = −1(𝛼|𝛼)

Example 1:

On F 𝑛 there is an inner product which we call the standard inner product. It is defined on 𝛼 =

(𝑥1 , ⋯ 𝑥𝑛 ) and 𝛽 = (𝑦1 , … , 𝑦𝑛 ), by

6
 (𝛼|𝛽) = ∑ 𝑥𝑖 𝑦̅𝑖
𝑖

When F is R this may be also written as

(𝛼|𝛽) = ∑ 𝑥𝑖 𝑦𝑖
𝑖

In the real case, the standard inner product is often called the dot or scalar product and denoted

by 𝛼 ⋅ 𝛽.

INNER PRODUCTS SPACES

An inner product space is a real or complex vector space together with a specified inner product

on that space.

• A finite-dimensional real inner product space is often called a Euclidean spare. A

complex inner product spare often referred to as a unitary spare.

• Every inner product space is a normed linear space and every normed space is a metric

space. Hence , every inner product space is a metric space.

Theorem

If V is an inner product space, then for any vector’s 𝛼, 𝛽 in 𝑉 and any scalar c

(1) ||𝑐𝛼|| = |𝑐|||𝛼|| ;

(ii) ||𝛼|| > 0 for 𝛼 ≠ 0

(iii) |(𝛼 ∣ 𝛽)| ⩽ ||𝛼|| ||𝛽||

(iv) ∥ 𝛼 + 𝛽|| ⩽∥ 𝛼 ∥ +∥ 𝛽||

Proof:

Statements (i) and (ii) follow almost immediately form the various definitions

involved. The inequality in (iii) is clearly valid when 𝛼 = 0. if 𝛼 ≠ 0, put

(𝛽|𝛼)
𝛾=𝛽− 𝛼
∥ 𝛼 ∥2

7
 Then, (𝛾 ∣ 𝛼) = 0 and

(𝛽|𝛼) (𝛽|𝛼)
0 ⩽∥ 𝛾 ∥2 = (𝛽 − 2
𝛼 ⁄𝛽 − 𝛼)
∥𝛼∥ ∥ 𝛼 ∥2

(𝛽|𝛼)(𝛼|𝛽)
= (𝛽|𝛽) −
∥ 𝛼 ∥2

|(𝛼|𝛽)|2
=∥ 𝛽 ∥2 −
∥ 𝛼 ∥2

Hence,

|(𝛼 ∣ 𝛽)|2 ⩽∥ 𝛼 ∥2 ∥ 𝛽 ∥2

Now using (c) we find that

∥ 𝛼 + 𝛽 ∥2 = ∥ 𝛼 ∥2 + (𝛼 ∣ 𝛽) + (𝛽 ∣ 𝛼)+∥ 𝛽 ∥2

=∥ 𝛼 ∥2 + 2Re (𝛼 ∣ 𝛽)+∥ 𝛽 ∥2
⩽∥ 𝛼 ∥2 + 2 ∥ 𝛼 ∥∥ 𝛽 ∥ +∥ 𝛽 ∥2
= (∥ 𝛼 ∥ +∥ 𝛽 ∥)2

Thus,

∥ 𝛼 + 𝛽 ∥ ⩽ ∥ 𝛼 ∥ +∥ 𝛽 ∥

the inequality (iii) is called the Cauchy -Schwarz inequality. It has a wide variety of application

the proof shows that if 𝛼is non-zero then

(( 𝛼 ∣ 𝛽 )) <∥ 𝛼 ∥∥ 𝛽 ∥, unless

(𝛽|𝛼)
𝛽= 𝛼
∥ 𝛼 ∥2

Then equality occurs in (iii) if and only if 𝛼 and 𝛽 are linearly independent.

8
 CHAPTER 2

ORTHOGONAL SETS

Definition

Let α and β be the vectors in an inner product space V. Then α is orthogonal to β if

(α | β) = 0. We simply say that and are orthogonal.

Definition

If S is a set of vectors in V, S is called an orthogonal set provided all set pairs of

distinct vectors in S are orthogonal.

Definition

An orthogonal set is an orthogonal set S with the additional property that ∥ 𝛼 ∥= 1 for

every 𝛼 in S.

• The zero vectors are orthogonal to every vector in V and is the only vector with this

property.

• It is an appropriate to think of an orthonormal set as a set of mutually perpendicular

vectors each having length l.

Example: the vector (x , y) is 𝑅 2 is orthogonal to (−y , x) with respect to the standard inner

product, for,

((x , y)|(−y , x)) = −xy + yx = 0

• The standard basis of either 𝑅 𝑛 or 𝐶 𝑛 is an orthonormal set with respect to the standard

inner product.

9
Theorem : An orthogonal set of nonzero vectors is linearly independent.

Proof:

Let S be a finite or infinite orthogonal set of nonzero vectors in a given inner product space

suppose 𝛼1, 𝛼2 , … 𝛼𝑛 are distinct vectors in S and that β=𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛

Then (β|𝛼𝑘 )=( 𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛 |𝛼𝑘 )

= 𝑐1 (𝛼1 |𝛼𝑘 ) + 𝑐2 ( 𝛼2 |𝛼𝑘 )+…+𝑐𝑛 (𝛼𝑛 |𝛼𝑘 )

= 𝑐𝑘 (𝛼𝑛 |𝛼𝑘 ) , since (𝛼𝑖 |𝛼𝑗 ) = 0,if i ≠ j and (𝛼𝑖 |𝛼𝑗 ) = 1,if i=j

Hence, 𝑐𝑘 = (β |𝛼𝑘 ) /(𝛼𝑘 , 𝛼𝑘 ) )

𝑐𝑘 =(β |𝛼𝑘 )/||𝛼𝑘 ||2,1≤ k ≤ m

Thus, when β=0 each 𝑐𝑘 =0; so S is a linearly independent set.

Corollary:

If {𝛼1, 𝛼2 , … 𝛼𝑚 } is an orthogonal set of nonzero vectors in a finite dimensional inner product

space V, then m ≤ dimV.

That is number of mutually orthogonal vectors in V cannot exceed the dimensional V.

Corollary:

If a vector β is linear combination of an orthogonal of nonzero vectors 𝛼1, 𝛼2 , … 𝛼𝑛 , then β is

the particular linear combination

(𝛽 |𝛼𝑘 )
β =∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘
𝑘 ||

Proof :

Since β is the linear combination of an orthogonal sequence of nonzero vectors

𝛼1, 𝛼2 , … 𝛼𝑛 , we can write β =𝑐1 𝛼1 + ⋯ 𝑐𝑛 𝛼𝑛 .

(𝛽|𝛼𝑘 )
Where 𝑐𝑘 = , 1 ≤ k ≤ m (ref. by previous theorem)
||𝛼𝑘 ||2

( 𝛽 | 𝛼𝑘 )
Hence, β= ∑𝑚
𝑘=1 ||𝛼𝑘 ||2

10
Theorem (Gram Schmidt Orthogonalization Process)

Let V be an inner product space and {𝛽1, … , 𝛽𝑛 } be any linearly independent vectors in V. Then

one may construct orthogonal vectors {𝛼1, 𝛼2 , … 𝛼𝑛 } in V, such that for each k = 1, 2, …n, the

set {𝛼1, 𝛼2 , … 𝛼𝑘 }is an orthogonal basis for the subspace of V spanned by 𝛽1, … , 𝛽𝑛 .

Proof :

The vectors are obtained by means of a construction known as the Gram Schmidt

orthogonalization process.

First let 𝛼1 =𝛽1 The other vectors are then given inductively as follows:

Suppose 𝛼1, 𝛼2 , … 𝛼𝑚 (1 ≤ m ≤ n) have been chosen so that for every k

{𝛼1, 𝛼2 , … 𝛼𝑘 } (1≤k≤m)

is an orthogonal basis for the space of v that is spanned by 𝛽1, … , 𝛽𝑛

To construct the next vector 𝛼𝑚+1, let

( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1, = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
Then  m +1  0 . For otherwise, 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘 =0, implies,
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
𝛽𝑚+1= 𝛽𝑚+1 -∑𝑚
𝑘=1 𝛼𝑘 ,implies,  m+1 is a linear combination of 𝛼1, 𝛼2 , … 𝛼𝑚 and
||𝛼𝑘 ||2

hence a linear combination of 1 ,  2 ,...,  m , a contradiction.

Furthermore, if 1≤j≤m, then,

(𝛽𝑚+1 |𝛼𝑘 )
(𝛼𝑚+1 | 𝛼𝑗 ) = ( 𝛽𝑚+1 | 𝛼𝑗 ) -∑𝑚
𝑘=1 (𝛼𝑘 |𝛼𝑗 )
||𝛼𝑘 ||2

= ( 𝛽𝑚+1 | 𝛼𝑚 ) – ( 𝛽𝑚+1 | 𝛼𝑗 ) , using the orthonormality of {𝛼1, 𝛼2 , … 𝛼𝑚 }.

Therefore {𝛼1, 𝛼2 , …,𝛼𝑚+1 } is an orthogonal set consisting of m+1 nonzero vectors in the

subspace spanned by 𝛽1, … , 𝛽𝑚+1 . Hence by an earlier Theorem , it is a basis for this subspace

.Thus the vectors , 𝛼1, 𝛼2 , … 𝛼𝑛 may be constructed using the formula

11
 ( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1 = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

In particular, when n=3 ,we have

𝛼1 =𝛽1

(𝛼2 |𝛽2 )
𝛼2 =𝛽2- α1
||𝛼𝑘 ||2

(𝛽3 |𝛼1 ) (𝛼2 |𝛽3 )

𝛼3 =𝛽3 - α1 - 𝛼
||𝛼1 ||2 ||𝛼𝑘 ||2 2

Corollary :

Every finite dimensional inner product space has an orthonormal basis.

Proof :

Let V be a finite dimensional inner product space and { 𝛽1, … , 𝛽𝑛 } a basis for V. Apply the

gram Schmidt orthogonalization process to construct an orthogonal basis , simply replace each
𝛼𝑘
vector 𝛼𝑛 by .
||𝛼𝑘 ||

Gram-Schmidt process can be used to test for linear dependence . For suppose 𝛽1, … , 𝛽𝑛 are

linearly independent vectors in an inner product space; to exclude a trivial case , assume that

β≠0. Let m be largest integers for which 𝛽1, … , 𝛽𝑚 are independent. Then 1≤m˂n.

Let 𝛼1 , 𝛼2 , … 𝛼𝑚 be the vectors obtained by applying the orthogonalization process to

(𝛽𝑚+1 |𝛼𝑘 )
𝛽1, … , 𝛽𝑚 . Then the vector 𝛼𝑚+1 given by 𝛼𝑚+1 = 𝛽𝑚+1–∑𝑚
𝑘=1 𝛼𝑘 is necessarily 0.
||𝛼𝑘 ||2

For 𝛼𝑚+1 is in the subspace spanned by 𝛼1 , 𝛼2 , … 𝛼𝑚 and orthogonal to each of the vectors ,

(β|𝛼𝑘 )
hence it is 0 as β=∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘 . Conversely, if 𝛼1 , 𝛼2 , … 𝛼𝑚 are different from 0 and 𝛼𝑚+1 =0,
𝑘 ||

then 𝛽1, … , 𝛽𝑚+1 are linearly independent .

Definition:

A best approximation to β  V by vectors in a subspace W of V is a vector α  W such that

 −    −  for every vector   W .

12
Theorem

Let W be a subspace of an inner product space V and let  V .

1. The vector  W is a best approximation to  V by vectors in W if and

only if  − is orthogonal to every vector in W .

2. If a best approximation to  V by vectors in W exists, it is unique.

3. If W is finite-dimensional and {𝛼1, 𝛼2 , … 𝛼𝑛 } is any orthonormal basis for W ,

then the vector

 =
n
(  |  k )
k
2 k
k =1

is the (unique) best approximation to  by vectors in W.

Definition:

Let V be an inner product space and S be any set of vectors in V. The orthogonal complement

of S is the set S ⊥ of all vectors in V which are orthogonal to every vector in S.

That is, S ⊥ =  V : ( |  ) = 0,    S

Definition:

Whenever the vector α in the above theorem exists it is called the orthogonal projection of

β on W. If every vector in V has an orthogonal projection on W, the mapping that assigns to

each vector in V its orthogonal projection on W is called the orthogonal projection of V on W.

Corollary :

Let V be an inner product space and W a finite dimensional subspace and E be the

orthogonal projection of V on W. Then the mapping

β →β – Eβ

is the orthogonal projection of V on W ⊥ .

13
Proof :

Let β  V . Then β – Eβ  W ⊥ , and for any γ  W ⊥ , β – γ = E β+(β – Eβ – γ)

Since Eβ  W and β – Eβ – γ  W ⊥ ,

It follows that

||𝛽 – 𝛾||2 = (Eβ+(β – Eβ – γ) ,E β+(β –Eβ – γ))

= ||𝐸𝛽||2+||𝛽 – 𝐸𝛽 – 𝛾||2

≥ ||𝛽 – (𝛽 – 𝐸𝛽)||2

with strict inequality when γ≠ β – Eβ . Therefore, β – Eβ is the best approximation to β by

vectors in W ⊥ .

Theorem

Let W be a finite dimensional subspace of an inner product space V and let E be the orthogonal

projection of V on W. Then E is an idempotent linear transformation of V onto W , W ⊥ is the

null space of E , and V= W ⨁ W ⊥ .

Proof

Let β be an arbitrary vector in V. Then Eβ is the best approximation to β that lies in W .

In particular, Eβ =β when β is in W . Therefore, E(Eβ) =Eβ for every β in V; that is, E is

idempotent : 𝐸 2 = E . To prove that E is linear transformation, let α and β be any vectors in V

and c an arbitrary scalar ,Then by theorem,

α-Eα and β-Eβ are each orthogonal to every vector in W . Hence the vector

c(α-Eα)+(β-Eβ)=(cα +β)-(cEα +Eβ)

Also belongs to W ⊥ . Since cEα+ Eβ is a vector in W , it follows from theorem that

E(cα+ β)= cEα+ Eβ.

Again let β be any vector in V. Then Eβ is the unique vector in W such that β-Eβ is in W ⊥ .

Thus Eβ=0 when β is in W ⊥ .

Conversely, β is in W ⊥ when Eβ=0. Thus W ⊥ is the null space of E.

14
The equation ,

β = E β+β – Eβ

shows that V = W + W ⊥ ; moreover W  W ⊥ = {0} ; for if α is a vector in W  W ⊥ ,then

( |  ) =0. Therefore, α=0 and V is the direct sum of W and W ⊥ .

Corollary :

Under the conditions of theorem, I − E is the orthogonal projection of V on W ⊥ .

It is an independent linear transformation of V onto W ⊥ with null space W .

Proof :

We have seen that the mapping β →β- E β is the orthogonal projection of V on W ⊥ .

Since E is a linear transformation , this projection W ⊥ is the linear transformation I − E from

its geometric properties one sees that I − E is an idempotent .Transformation of V onto W .

This also follows from the computation ( I − E )( I − E )= I − E - E +𝐸 2

=I −E

Moreover , ( I − E )β =0 If and only if β = Eβ , and this is the case if and only if β is in W .

Therefore W is the null space of I − E .

15
 BIBLIOGRAPHY

● Balmohan V .Limaye :Functional analysis

● Kenneth Hoffman,Ray Kunze:Linear Algebra,second edition

● Stephen H.Friedberg, Arnold J.Insel, Lawrence E.Spence : Linear Algebra, Fourth

edition

16
DIRECT PRODUCTS AND FINITELY GENERATED ABELIAN
GROUPS

Project report submitted to

The Kannur University

for the award of the degree

of

Bachelor of Science

by

ARJUN MOHANAN P V

DB18CMSR26

Under the guidance of

MS. Sneha P Sebastian

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report ‘DIRECT PRODUCTS AND FINITELY

GENERATED ABELIAN GROUPS’ is the bonafide project of ARJUN MOHANAN
P V who carried out the project under my supervision.

Mrs.Riya Baby Ms.Sneha P Sebastian

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, ARJUN MOHANAN P V, hereby declare that this project report entitled ‘DIRECT
PRODUCTS AND FINITELY GENERATED ABELIAN GROUPS’ is an original record of
studies and bonafide project carried out by me during the period from November 2019 to March
2020, under the guidance of Ms.Sneha P Sebastian, Department of Mathematics, Don Bosco
Arts and Science College, Angadikadavu and has not been submitted by me elsewhere for the
award of any degree, diploma, title or recognition, before.

ARJUN MOHANAN P V
Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great help to me
during the course of my dissertation. First and foremost I thank the almighty, for his blessing and
protection during the period of this work. I express my thanks to Dr. Fr. Francis Karakkatt,
Principal, for support in the completion of this dissertation. I express my gratitude to Ms.Sneha P
Sebastian, my project supervisor, for the constant encouragement, valuable guidance and timely
corrections, which made this work a success.

I am also indebted to all my classmates and friends who supported me throughout the
study. I would like to express my thanks to my parents and dear ones for their constant
encouragement and support. I also thank all those who helped me directly or indirectly to
complete this project.

ARJUN MOHANAN P V
 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 preliminary 2–3
03 Chapter 1 4 – 12
04 Chapter 2 13 – 17
05 Conclusion 18
06 Bibliography 19
 INTRODUCTION

In mathematics, a group is a set equipped with a binary operation that combines any two
elements to form a third element in such a way that the three conditions called group axioms are
satisfied , namely associativity , identity and invertability.

Let us take a moment to review our present stockpile of groups. Starting with finite
groups, we have the cyclic group ℤ𝑛 ,the symmetric group 𝑆𝑛 , and the alternating group 𝐴𝑛 for
each positive integer n. We also have the dihedral group 𝐷𝑛 and klein 4-group . Of course we
know that subgroups of these groups exists. Turning to infinite groups , we have ℤ, ℝ, ℂ under
addition , and their non zero elements under multiplication we also have the group 𝑆𝐴 of all
permutation of an infinite set 𝐴 , as well as various groups formed from matrices .

One purpose of this section is to show a way to use known groups as building blocks to
form more groups. Given two groups 𝐺 and 𝐻, it is possible to construct a new group from the
cartesian product of 𝐺 and 𝐻 . Conversely , given a large group , it is sometimes possible to
decompose the group ; that is , a group is sometimes isomorphic to the direct product of two
smaller groups. Rather than studying a large group , it is often easier to study the component
group of that group.

1
 PRELIMINARY

Groups : A non empty set 𝐺 together with an operation ∗ is said to be a group , denote by
(𝐺 ,∗) , if it satisfy the following axioms.

• Closure property
• Associative property
• Existence of identity
• Existence of inverse

Abelian group

A group (𝐺 ,∗) is said to be abelian if it satisfies commutative law .

Finite group

If the underlying set G of the group (𝐺 ,∗) consist of finite number of elements , then the
group is finite group .

Infinite group

A group that is not finite is an infinite group .

Order of a group : The number of elements in a finite group is called the order of the group ,
denoted by 𝑂(𝐺 ) .

Example

Show that the set of integers ℤ is a group with respect to the operation of addition of
integers.

ℤ = {… … … . −3, −2, −1,0,1,2,3, … … … }

Since the addition of two integers gives an integer , it satisfy closure property .

2
If 𝑎, 𝑏, 𝑐 𝜖 ℤ then the (𝑎 + 𝑏) + 𝑐 = 𝑎 + (𝑏 + 𝑐) , hence associativity holds .

There is a number 0 𝜖 ℤ such that 0 + 𝑎 = 𝑎 + 0 , hence identity exists

If 𝑎 𝜖 ℤ then there exists – 𝑎 𝜖 ℤ , such that −𝑎 + 𝑎 = 0 = 𝑎 + −𝑎

Therefore inverse exist .

Therefore ℤ is a group under addition .

Subgroup

A subset 𝐻 of 𝐺 is said to be a subgroup of 𝐺 if 𝐻 itself is a group under the same operation in

𝐺.

There are two different types of group structure of order 4 .

ℤ4 = { 0,1,2,3}

Klein 4 – group , 𝑉 = {𝑒, 𝑎, 𝑏, 𝑐}

Cyclic group

A group 𝐺 is cyclic if there is some element ‘𝑎’ in 𝐺 that generate 𝐺. And the element ‘𝑎’ is
called generator of 𝐺.

Group Homomorphism

A function Ѱ: 𝐺 → 𝐺′ is a group homomorphism ( or simply homomorphism ).

If Ѱ(𝑎𝑏) = Ѱ(𝑎) Ѱ(𝑏) hold for all 𝑎 , 𝑏 ∈ 𝐺 , is called homomorphism property .

Isomorphism

A one to one and onto homomorphism Ѱ: 𝐺 → 𝐺′ is called an isomorphism .

3
 CHAPTER – 1

DIRECT PRODUCT OF GROUPS

Definition

The Cartesian product of sets 𝑆, 𝑆2 , … … . , 𝑆𝑛 is the set of all ordered n-tuples (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ),
where 𝑎𝑖 ∈ 𝑆𝑖 for 𝑖 = 1,2,3, … … . , 𝑛. The Cartesian product is denoted by either

𝑛
𝑆1 × 𝑆2 × … … .× 𝑆𝑛 or by Π𝑖=1 𝑆𝑖 .

Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups and let us use multiplicative notation for all the group operations.

If we consider 𝐺𝑖 as a set , 𝑖 = 1,2, … … . 𝑛 . we have the products 𝐺1 × 𝐺2 × … … . ,× 𝐺𝑛 we

𝑛 𝑛
denote it by Π𝑖=1 𝐺𝑖 . This product is called direct-product of groups. We can make Π𝑖=1 𝐺𝑖 into a
group by means of a binary operation of multiplication by components.

Theorem
𝑛
Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups. For (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) and (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) in Π𝑖=1 𝐺𝑖 define ;

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

𝑛
Then Π𝑖=1 𝐺𝑖 is a group.

4
Proof

We have ,

𝑛
Π𝑖=1 𝐺𝑖 = {(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∶ 𝑎𝑖 ∈ 𝐺𝑖 }

(1) Closure property

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

And we have ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

Here 𝑎𝑖 ∈ 𝐺𝑖 and 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

∵ 𝐺𝑖 is a group , 𝑎𝑖 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

𝑛
⇒ (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

𝑛
i.e. Π𝑖=1 𝐺𝑖 is closed under the binary operation.

(2) Associativity
𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ), (𝑐1, 𝑐2 , … … . , 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

We have,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

𝑛
= (𝑎1 𝑏1 𝑐1 , 𝑎2 𝑏2 𝑐2 , … … . , 𝑎𝑛 𝑏𝑛 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

[(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

= [𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )
= [(𝑎1 𝑏1 )𝑐1 , (𝑎2 𝑏2 )𝑐2, … … . , (𝑎𝑛 𝑏𝑛 )𝑐𝑛 ]
= [𝑎1 (𝑏1 𝑐1 ), 𝑎2 (𝑏2 𝑐2), … … . , 𝑎𝑛 (𝑏𝑛 𝑐𝑛 )]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[𝑏1 𝑐1, 𝑏2 𝑐2 , … … . , 𝑏𝑛 𝑐𝑛 ]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )]

5
Hence associativity holds.

(3) Existence of identity

If 𝑒𝑖 is the identity element in 𝐺𝑖 .

Then,

𝑛
(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Also for ,

𝑛
(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖 ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) = (𝑎1 𝑒1 , 𝑎2 𝑒2 , … … . , 𝑎𝑛 𝑒𝑛 )

= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )

𝑛
∴ (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) is the identity element ‘𝑒’ in Π𝑖=1 𝐺𝑖

(4) Existence of inverse

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Here 𝑎𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛.

Since 𝐺𝑖 is a group ,

∃ an inverse element 𝑎𝑖−1 in 𝐺𝑖 : 𝑎𝑖 𝑎𝑖−1 = 𝑒𝑖 𝑖 = 1,2, … … . , 𝑛

𝑛
Clearly, (𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) ∈ Π𝑖=1 𝐺𝑖 &

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 )

𝑛
Hence Π𝑖=1 𝐺𝑖 is a group.

6
Note
𝑛
If the operation of each 𝐺𝑖 is a commutative. We sometimes use additive notation in Π𝑖=1 𝐺𝑖 and
𝑛
refer to Π𝑖=1 𝐺𝑖 as the direct sum of the group 𝐺𝑖 . The notation ⨁𝑛𝑖=1 𝐺𝑖 , especially with abelian
groups with operation +.

The direct sum of abelian groups 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 may be written 𝐺1 ⨁𝐺2 ⨁ … … ⨁𝐺𝑛

• Direct product of abelian group is abelian

Example

Q. Check whether ℤ2 × ℤ3 is cyclic or not.

ℤ2 = {0,1}

ℤ3 = {0,1,2}

ℤ2 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2)}

Consider,

1(1,1) = (1,1)

2(1,1) = (1,1) + (1,1) = (0,2)

3(1,1) = (1,1) + (1,1) + (1,1) = (1,0)

4(1,1) = 3(1,1) + (1,1) = (1,0) + (1,1) = (0,1)

5(1,1) = 4(1,1) + (1,1) = (0,1) + (1,1) = (1,2)

6(1,1) = 5(1,1) + (1,1) = (1,2) + (1,1) = (0,0)

∴ (1,1) is a generator of ℤ2 × ℤ3

∴ ℤ2 × ℤ3 is a cyclic group generated by (1,1).

7
Q. Check whether ℤ3 × ℤ3 is cyclic or not.

ℤ3 = {0,1,2}

ℤ3 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2), (2,0), (2,1), (2,2)}

1(0,1) = (0,1)

2(0,1) = (0,2)

3(0,1) = (0,3) = (0,0) ∴ order (0,1) = 3

1(0,2) = (0,2)

2(0,2) = (0,4) = (0,1)

3(0,2) = (0,6) = (0,0) ∴ order (0,2) = 3

Every element added to itself three times gives the identity. Thus no element can
generate the group. Hence ℤ3 × ℤ3 is not cyclic.

similarly ℤ𝑚 × ℤ𝑚 is not cyclic for any 𝑚.

Theorem

The group ℤ𝑚 × ℤ𝑛 is cyclic and is isomorphic to ℤ𝑚𝑛 if and only if 𝑚 and 𝑛 are relatively
prime, that is, the gcd of 𝑚 and 𝑛 is 1.

Proof

Suppose ℤ𝑚 × ℤ𝑛 is cyclic and isomorphic to ℤ𝑚𝑛 .

To show that 𝑚 and 𝑛 are relatively prime.

Suppose not, let d be the 𝑔𝑐𝑑 of 𝑚 and 𝑛.

So that 𝑑 > 1

8
 𝑚𝑛
Consider , which is an integer since 𝑑|𝑚 and 𝑑|𝑛
𝑑

𝑚𝑛
Let (𝑟, 𝑠) be an arbitrary element of ℤ𝑚 × ℤ𝑛 , add (𝑟, 𝑠) repeatedly times
𝑑

(𝑟, 𝑠) + (𝑟, 𝑠)+, … … . , +(𝑟, 𝑠) } 𝑚𝑛

𝑑
𝑡𝑖𝑚𝑒𝑠 = (0,0)

∴ no element of ℤ𝑚 × ℤ𝑛 having order 𝑚𝑛. ∴ no element of ℤ𝑚 × ℤ𝑛 can generate ℤ𝑚 × ℤ𝑛

which is not possible. ∵ ℤ𝑚 × ℤ𝑛 is cyclic. Hence 𝑔𝑐𝑑(𝑚, 𝑛) = 1.

i.e. 𝑚 and 𝑛 are relatively prime.

Conversely, suppose 𝑚 and 𝑛 are relatively prime, i.e. gcd(𝑚, 𝑛) = 1

To show that ℤ𝑚 × ℤ𝑛 is cyclic.

If ℤ𝑚 × ℤ𝑛 is cyclic, then it is isomorphic to ℤ𝑚𝑛 , ∵ ℤ𝑚 × ℤ𝑛 has 𝑚𝑛 elements.

Consider the cyclic subgroup of ℤ𝑚 × ℤ𝑛 generated by the element (1,1).The order of this
cyclic subgroup is the smallest power of (1,1),that gives the identity (0,0). Here taking a power
of (1,1) in our additive notation will involve adding (1,1) to itself repeatedly.

Consider (1,1) + (1,1)+, … … . , +(1,1)

If we add first coordinates 𝑚 times , we get zero.

∴ order of first coordinate = 𝑚.

Similarly , Order of second coordinate = 𝑛.

The two coordinates together become zero. If we add them 𝑙𝑐𝑚(𝑚, 𝑛) times.

∵ gcd(𝑚, 𝑛) = 1, We get the 𝑙𝑐𝑚 = 𝑚𝑛.

i.e. (1,1) generates a cyclic subgroup of ℤ𝑚 × ℤ𝑛 of order 𝑚𝑛 , which is the order of the whole
group.

⇒ ℤ𝑚 × ℤ𝑛 =< (1,1) >

9
 ⇒ ℤ𝑚 × ℤ𝑛 is cyclic.

Corollary

The group ⨅𝑛𝑖=1 ℤ𝑚𝑖 is cyclic and isomorphic to ℤ𝑚1 𝑚2 …….𝑚𝑛 if and only if the numbers 𝑚𝑖 for
𝑖 = 1,2, … … . , 𝑛 are such that the 𝑔𝑐𝑑 of any two of them is 1.

Example

If 𝑛 is written as a product of powers of distinct prime numbers , as in ,

𝑛 = (𝑝1 )𝑛1 . (𝑝2 )𝑛2 … … . (𝑝𝑛 )𝑛𝑟

Then ℤ𝑛 is isomorphic to ℤ(𝑝1 )𝑛1 × ℤ(𝑝2 )𝑛2 × … … .× ℤ(𝑝𝑟 )𝑛𝑟 .

In particular , ℤ72 is isomorphic to ℤ8 × ℤ9 .

Consider set of integers ℤ, cyclic subgroup of ℤ is of the form 𝑛ℤ , 𝑛 ∈ ℤ. Consider 2ℤ and 3ℤ ,

then < 2 > ∩ < 3 > = < 6 >

∴ if we take 𝑟ℤ , 𝑠ℤ of ℤ , then the 𝑙𝑐𝑚(𝑟, 𝑠) =generator of < 𝑟 > ∩ < 𝑠 >

Using this we can define the 𝑙𝑐𝑚 of the positive integers.

Definition

Let 𝑟1 , 𝑟2 , … … . , 𝑟𝑛 be positive integers. Their least common multiple (abbreviated lcm ) is the
positive generator of the cyclic group of all common multiples of the 𝑟𝑖 , that is the cyclic group
of all integers divisible by each 𝑟𝑖 for 𝑖 = 1,2, … … . , 𝑛.

10
Theorem

Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ ⨅𝑛𝑖=1 𝐺𝑖 .

If 𝑎𝑖 is of finite order 𝑟𝑖 in 𝐺𝑖 , then the order of (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) in ⨅𝑛𝑖=1 𝐺𝑖 is equal to the least
common multiple of all the 𝑟𝑖 .

Proof

Given,

𝑟
order of 𝑎1 = 𝑟1 ⇒ 𝑎11 = 𝑒1 in 𝐺1

𝑟
order of 𝑎2 = 𝑟2 ⇒ 𝑎22 = 𝑒2 in 𝐺2

.
.
.
𝑟
order of 𝑎𝑛 = 𝑟𝑛 ⇒ 𝑎𝑛𝑛 = 𝑒𝑛 in 𝐺𝑛 .

We have to find a power 𝑘 for (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ).

So that (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )𝑘 = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ).

The power must simultaneously be a multiple of 𝑟1 , multiple of 𝑟2 and so on. But 𝑘 is the least
positive integers having the above property.

∴ 𝑘 = 𝑙𝑐𝑚(𝑟1 , 𝑟2 , … … . , 𝑟𝑛 ).

Q. Find the order of (8,4,10) in the group ℤ12 × ℤ60 × ℤ24 .

𝑂(8) = 3 in 𝑍12

11
 𝑂(4) = 15 in 𝑍60

𝑂(10) = 12 in 𝑍24

𝑂(8,4,10) = 𝑙𝑐𝑚(3,15,12) = 60

Q. Find a generator of ℤ × ℤ2

ℤ × ℤ2 = {(𝑛, 0), (𝑛, 1): 𝑛 ∈ ℤ}

(𝑛, 0) = 𝑛(1,0)
(𝑛, 1) = (𝑛, 0) + (0,1) = 𝑛(1,0) + (0,1)

∴ ℤ × ℤ2 is generated by {(1,0), (0,1)}

In general , ℤ × ℤ𝑛 is generated by ,

{(1,0,0, … … . ,0), (0,1,0, … … . ,0), … … . , (0,0, … … . ,1)}

Q. Find the order of (3,10,9) in (ℤ4 , ℤ12 , ℤ15 )

𝑂(3) = 4 in ℤ4

𝑂(10) = 6 in ℤ12

𝑂(9) = 5 in ℤ15

∴ 𝑂(3,10,9) = 𝑙𝑐𝑚(4,6,5)
= 60

∴ order of (3,10,9) in ℤ4 × ℤ12 × ℤ15 is 60.

12
 CHAPTER-2

FUNDAMENTAL THEOREM OF FINITELY GENERATED ABELIAN

GROUPS

Every finitely generated abelian group 𝐺 is isomorphic to a direct product of cyclic groups in the
form,

ℤ(𝑝1 )𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛 × ℤ × ℤ × ℤ × … … .× ℤ

Where the 𝑝𝑖 are primes , not necessarily distinct and the 𝑟𝑖 are positive integers.

Remark

• The direct product is unique except for possible rearrangement of the factors.
• The number of factors ℤ is unique and this number is called Betti number.

Example

Find all abelian groups , upto isomorphism of order

1)8 , 2)16 , 3)360

(1) Order 8

8=1×8

8 = 2 × 4 = 2 × 22

8 = 2×2×2

3 non-isomorphic groups are ℤ8 , ℤ2 × ℤ4 ,

ℤ2 × ℤ2 × ℤ2
13
(2) Order 16

16 = 1 × 16 = 1 × 24

16 = 2 × 8 = 2 × 23

16 = 4 × 4 = 22 × 22

16 = 2 × 2 × 2 × 2

16 = 2 × 2 × 22

ℤ16 , ℤ2 × ℤ8 , ℤ4 × ℤ4 , ℤ2 × ℤ2 × ℤ2 × ℤ2 , ℤ2 × ℤ2 × ℤ4

(3) Order 360

360 = 22 ⋅ 32 ⋅ 5

Possibilities are,

1) ℤ8 × ℤ9 × ℤ5

2) ℤ2 × ℤ4 × ℤ9 × ℤ5

3) ℤ2 × ℤ2 × ℤ2 × ℤ9 × ℤ5

4) ℤ8 × ℤ3 × ℤ3 × ℤ5

5) ℤ2 × ℤ4 × ℤ3 × ℤ3 × ℤ5

6) ℤ2 × ℤ2 × ℤ2 × ℤ3 × ℤ3 × ℤ5

Definition

A group 𝐺 is decomposable if it is isomorphic to a direct product of two proper non-trivial

subgroups , otherwise 𝐺 is indecomposable.

14
Example

ℤ6 is decomposable while ℤ5 is indecomposable.

ℤ6 is isomorphic to ℤ2 × ℤ3

ℤ𝑚𝑛 is isomorphic to ℤ𝑚 × ℤ𝑛 , if 𝑚 and 𝑛 are prime.

Theorem

The finite indecomposable abelian groups are exactly the cyclic groups with order a power of a
prime.

Proof

Let 𝐺 be a finite indecomposable abelian group ∵ 𝐺 is finitely generated , we can apply

fundamental theorem of finitely generated abelian groups.

∴ 𝐺 ≅ ℤ(𝑝)𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

∵ 𝐺 is indecomposable and ℤ(𝑝𝑖 )𝑟𝑖 ’s are proper subgroups we get in the above , there is only one
factor say ℤ(𝑝𝑖 )𝑟𝑖 which is cyclic group with order a prime power.

Theorem

If 𝑚 divides the order of a finite abelian group , then 𝐺 has a subgroup of order 𝑚.

Proof

Given 𝐺 is a finite abelian group.

∴ we can apply Fundamental Theorem ,

15
Hence,

𝐺 ≅ ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛

1 2 𝑛

Here all primes need not be distinct.

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 . 𝑝22 … … 𝑝𝑛𝑛

Let 𝑚 is a +𝑣𝑒 integer which divides 𝑂(𝐺).

0 ≤ 𝑠𝑖 ≤ 𝑟𝑖 By theorem , “ let 𝐺 be a cyclic group with 𝑛 elements and generated by 𝑎. Let

𝑛
𝑏 ∈ 𝐺 , 𝑏 = 𝑎𝑠 , then ‘ 𝑏’ generates a cyclic subgroup 𝐻 of 𝐺 containing elements , where
𝑑

𝑑 = gcd(𝑛, 𝑠).”

𝑟
𝑟 −𝑠 𝑝𝑖 𝑖
𝑝𝑖 𝑖 𝑖 generates a cyclic subgroup of ℤ𝑝𝑟𝑖 having order 𝑟 𝑟 −𝑠𝑖
𝑖 𝑔𝑐𝑑(𝑝𝑖 𝑖 ,𝑝𝑖 𝑖 )

𝑟
𝑝𝑖 𝑖 𝑠
= 𝑟 −𝑠𝑖 = 𝑝𝑖 𝑖
𝑝𝑖 𝑖

𝑟 −𝑠𝑖 𝑠
∴ 𝑂(< 𝑝𝑖 𝑖 >) = 𝑝𝑖 𝑖

𝑟 −𝑠1 𝑠
i.e. < 𝑝11 > is a subgroup of ℤ𝑝𝑟1 having order 𝑝11 .
1

𝑟 −𝑠2 𝑠
< 𝑝22 > is a subgroup of ℤ𝑝𝑟2 having order 𝑝22 .
2

…………………………………………………………

𝑟 −𝑠𝑛 𝑠
< 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟𝑛 having order 𝑝𝑛𝑛 .
𝑛

𝑟 −𝑠1 𝑟 −𝑠2 𝑟 −𝑠𝑛

∴ < 𝑝11 > × < 𝑝22 > × … … .× < 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛
1 2 𝑛
𝑠 𝑠 𝑠
having order 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛 = 𝑚.

16
Theorem

If 𝑚 is a square free integer , that is 𝑚 is not divisible by the square of any prime . Then every
abelian group of order 𝑚 is cyclic.

Proof

Let 𝑚 be a square free integer , then 𝑝𝑖 ⫮ 𝑚 for every 𝑖 greater than 1 for a prime 𝑝.

Given 𝐺 is a finite abelian group having order 𝑚 , by fundamental theorem , then

𝐺 ≅ ℤ(𝑝1 )𝑟1 × ℤ(𝑝2 )𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛

∵ 𝑂(𝐺) is a square free integer , the only possibility

𝑟1 = 𝑟2 = … … . . = 𝑟𝑛 = 1

Then,

𝐺 ≅ ℤ𝑝1 × ℤ𝑝2 × … … .× ℤ𝑝𝑛

≅ ℤ𝑝1 ,𝑝2 ,…….,𝑝𝑛 , which is cyclic.

Example

15 is a square free integer. So an abelian group of order 15 is cyclic.

17
 CONCLUSION

Direct product of groups is the product 𝐺1 × 𝐺2 , … … . 𝐺𝑛 ,where each 𝐺𝑖 is a set. We have

discussed about definition and some properties related to the direct product of groups. The
fundamental theorem of finitely generated abelian group helped us to get a deeper understanding
about the topic. The theorems gives us complete structural information about abelian group, in
particular finite abelian group. We have also discussed some examples in order to develope more
intrest in algebra.

18
 BIBLIOGRAPHY

• A first course in Abstract Algebra – 7th edition.

John . B . Fraleigh

• Contomporary Abstract Algebra – 7th edition.

Joseph . A . Gallian

• Topics In Algebra – 2nd edition

I . N . Herstein

19
NORMED LINEAR SPACES
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
ARUN K A
DB18CMSR27
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE
It is to certify that this project report ‘Normed Linear Spaces’ is the bonafide project of

Arun K A carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head Of Department Supervisor

Department Of Mathematics
Don Bosco Arts And Science College
Angadikadavu

2
 DECLARATION
I Arun K A hereby declare that the project ‘Normed Linear Space’ is an original record of
studies and bona fide project carried out by me during the period of 2018 – 2021 under the
guidance of Ms. Athulya P , Department of Mathematics, Don Bosco Arts and Science College,
Angadikadavu and has not submitted by me elsewhere for the award of my degree, diploma, title
or recognition, before.

Arun K A

DB18CMSR27

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu

3
 ACKNOWLEDGEMENT

First and foremost, praises and thanks to God, the Almighty, for His showers of blessings
throughout my work to complete the project successfully. I had finally managed to finish up
this project with great enthusiasm and determination. All the time spent to search and discuss
ideas as well as justifying theoretical clues to drive the output were worth my effort and time.

Therefore, I would like to express my sincere gratitude to my supervisor Ms. Athulya P,

Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu, for providing
invaluable guidance, comments and suggestions throughout the course of the project. I also
acknowledge with a deep sense of reverence, my gratitude towards all the faculty members of
the Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I owe and respectfully offer my thanks to the principal and staff of Don Bosco Arts and
Science College, Angadikadavu for their constant moral support and mellifluous affection
provided to me.

I express my gratitude to all who directly or indirectly helped me to complete this project
successfully. Their guidance and support was very helpful in bringing this work to
conclusion.

4
 CONTENTS

SI no Title Page no

1 Introduction 6

2 Preliminaries 7-9

3 Chapter 1 10 - 18

4 Chapter 2 19 - 25

5 Conclusion 26

6 Bibliography 27

5
 INTRODUCTION

This chapter gives an introduction to the theory of normed linear spaces. A skeptical reader may
wonder why this topic in pure mathematics is useful in applied mathematics. The reason is quite
simple: Many problems of applied mathematics can be formulated as a search for a certain
function, such as the function that solves a given differential equation. Usually the function
sought must belong to a definite family of acceptable functions that share some useful properties.
For example, perhaps it must possess two continuous derivatives. The families that arise
naturally in formulating problems are often linear spaces. This means that any linear combination
of functions in the family will be another member of the family. It is common, in addition, that
there is an appropriate means of measuring the “distance” between two functions in the family.
This concept comes into play when the exact solution to a problem is inaccessible, while
approximate solutions can be computed. We often measure how far apart the exact and
approximate solutions are by using a norm. In this process we are led to a normed linear space,
presumably one appropriate to the problem at hand. Some normed linear spaces occur over and
over again in applied mathematics, and these, at least, should be familiar to the practitioner.
Examples are the space of continuous functions on a given domain and the space of functions
whose squares have a finite integral on a given domain.

6
 PRELIMINARIES

1) LINEAR SPACES
We introduce an algebraic structure on a set 𝑋and study functions on X which are well behaved
with respect to this structure. From now onwards , K will denote either R , the set of all real
numbers or C, the set of all complex numbers. For k ∈ C , Re k and Im k will denote the real and
imaginary part of k.

A linear space(or a vector space) over K is a non-empty set 𝑋 along with a function
+ : 𝑋 × 𝑋 → 𝑋, called addition and a function ·: K× 𝑋 → 𝑋 called scalar multiplication, such
that for all 𝑥 , 𝑦 , 𝑧 ∈ 𝑋and 𝑘 , 𝑙 ∈ K , we have

𝑥+ 𝑦 = 𝑦 + 𝑥

𝑥 + (𝑦 + 𝑧) = (𝑥 + 𝑦) + 𝑧

∃0∈𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + 0 = 𝑥,

∃ − 𝑥 ∈ 𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + (− 𝑥) = 0 ,

𝑘 · (𝑥 + 𝑦) = 𝑘 · 𝑥 + 𝑘 · 𝑦,

(𝑘 + 𝑙)⋅𝑥 = 𝑘 · 𝑥 + 𝑙 · 𝑥,

(𝑘𝑙)⋅𝑥 = 𝑘 · (𝑙 · 𝑥),

1⋅𝑥 = 𝑥.

We shall write 𝑘𝑥 in place of 𝑘 · 𝑥. We shall also adopt the following notations. For
𝑥, 𝑦 ∈ 𝑋, 𝑘 ∈ K and subsets 𝐸, 𝐹of 𝑋 ,

𝑥 + 𝐹 = {𝑥 + 𝑦: 𝑦 ∈ 𝐹},

𝐸 + 𝐹 = {𝑥 + 𝑦: 𝑥 ∈ 𝐸, 𝑦 ∈ 𝐹},`

𝑘𝐸 = {𝑘𝑥: 𝑥 ∈ 𝐸}.

2) BASIS

A nonempty subset 𝐸 of 𝑋 is said to be a subspace of 𝑋 if 𝑘𝑥 + 𝑙𝑦 ∈ 𝐸 whenever 𝑥, 𝑦 ∈ 𝐸 and

𝑘, 𝑙 ∈ K . If ∅≠𝐸 ⊂ 𝑋, then the smallest subspace of 𝑋 containing 𝐸 is

7
 {
𝑠𝑝𝑎𝑛⁡𝐸 = 𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛: 𝑥1, …, 𝑥𝑛 ∈ 𝐸 , 𝑘1, …, 𝑘𝑛 ∈ 𝐾 }
It is called the span of 𝐸. If span 𝐸 = 𝑋, we say that 𝐸 spans 𝑋. A subset 𝐸 of 𝑋 is said to be
linearly independent if for all 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈K , the equation
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0implies that 𝑘1 = ⋯ = 𝑘𝑛 = 0. It is called linearly dependent if it is not
linearly independent, that is, if there exist 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈ K such that
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0, where at least one 𝑘𝑗is nonzero.

A subset E of X is called a Hamel basis or simply basis for X if span of E = X and E is

linearly independent .

3) DIMENSION

If a linear space 𝑋 has a basis consisting of a finite number of elements , then X is called finite
dimensional and the number of elements in a basis for 𝑋 is called the dimension of 𝑋, denoted as
dimX . Every basis for a finite dimensional linear space has the same (finite) number of elements
and hence the dimension is well-defined. The space {0} is said to have zero dimension. Note that
it has no basis !

If a linear space contains an infinite linearly independent subset, then it is said to be infinite
dimensional.

4)METRIC SPACE

We introduce a distance structure on a set 𝑋 and study functions on 𝑋 which are well-behaved
with respect to this structure.

A metric 𝑑 on a nonempty set 𝑋 is a function 𝑑: 𝑋 × 𝑋 →R

such that for all 𝑥, 𝑦, 𝑧 ∈ 𝑋

d(x, y) ≥0 and d(x , y) = 0 iff x=y

d(y , x) = d(x , y)

d(x , y) ≤ d(x , z) + d(z , y) .

The last condition is known as the triangle inequality. A metric space is a nonempty set 𝑋 along
with a metric on it.

8
5)CONTINUOUS FUNCTIONS

Roughly speaking, a function from a metric space to a metric space is continuous if it sends
‘nearby’ points to ‘nearby’ points. If 𝑋 and 𝑌 are metric spaces with metrics 𝑑 and 𝑒 respectively,
then a function 𝐹: 𝑋 → 𝑌 is said to be continuous at 𝑥0 ∈ 𝑋 if for every ϵ⟩ 0 , there is some
(
δ > 0 (possibly depending on ϵ and 𝑥0 ) such that 𝑒 𝐹(𝑥), 𝐹 𝑥0 ( )) < ϵ for all 𝑥 ∈ 𝑋 satisfying
( )
𝑑 𝑥, 𝑥0 < δ. Further, 𝐹 is said to be continuous on 𝑋 if it is continuous at every point of 𝑋. It is
easy to see that 𝐹 is continuous on 𝑋 if and only if the set F -1(E) is open in X whenever the set E
is open inY. Also , this happens iff F(xn) →F(x) in Y whenever xn →x in X.

6) UNIFORM CONTINUITY

We note that a continuous function 𝐹: 𝑇 → 𝑆 is, in fact, uniformly

continuous, that is, for every ϵ > 0, there exists some δ > 0
such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < ϵ whenever 𝑑 𝑡, 𝑢( )< δ. This can be seen as follows. Let 𝑡 ∈ 𝑇. By
ε
the continuity of 𝐹 at 𝑡 ∈ 𝑇, there is some δ𝑡, such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < 2 whenever
𝑑(𝑡, 𝑢) < δ𝑡.

7) FIELD

A ring is a set R together with two binary operations + and ·( which we call addition and
multiplication ) such that the following axioms are satisfied .

➢ R is an abelian group with respect to addition

➢ Multiplication is associative
➢ ∀𝑎 , 𝑏, 𝑐 ∈ 𝑅the left distributive law a(b + c) = (a·b) + (a·c) and the right distributive
law (a + b)c = (a·c) + (b·c) , hold .

A field is a commutative division ring

9
 CHAPTER 1

NORMED LINEAR SPACE

Let X be a linear space over K . A norm on X is the function || || from 𝑋 to R such that ∀
x,y ∈ X and k ∈ K ,

||𝑥||≥0 and ||𝑥|| = 0 if and only if x = 0 ,

||x + y|| ≤||x|| + ||y|| ,

||kx|| =|k| ||x|| .

A norm is the formalization and generalization to real vector spaces of the intuitive
notion of “ length” in the real world .

A normed space is a linear space with norm on it .

For x and y in X , let

d(x,y) = ||x - y||

Then d is a metric on X so that (X,d) is a metric space , thus every normed space is a metric
space

➢ Every normed linear space is a metric space . But converse may not be true .

Example :
|𝑥 − 𝑦|
d(x,y) = , ∀ x , y ∈X
1 + |𝑥 − 𝑦|

|𝑥 − 𝑦|
⇒ ||x - y|| =
1 + |𝑥 − 𝑦|

|𝑧|
⇒ ||z|| = , z = x - y ∈X
1 + |𝑧|

10
 |α𝑧|
||αz|| =
1 +|α𝑧|

|α| |𝑧|
=
1 + |α| |𝑧|

= |α| ( |𝑧|
1 + |α| |𝑧| )
≠ |α| ||z|| .

➢ Result

Let X be a normed linear space . Then ,

| ||x|| - ||y|| | ≤ ||x - y|| , ∀x , y ∈ X

Proof :

||x|| = || ( x - y ) + y|| ≤ ||x - y|| + ||y||

⇒||x|| - ||y|| ≤||x - y|| → (1)

x↔y

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||→(2)

From (1) and (2)

|||x|| - ||y||| ≤ ||x - y||

➢ Norm is a continuous function

Let xn →x , as n → ∞

11
 ⇒ xn - x →0 as n →∞

⇒ ||xn - x||→0 as n→∞

| ||xn|| - ||x|| | ≤ ||xn - x|| →0 as n→∞

⇒||xn|| - ||x|| →0 , as n →∞

⇒||x|| is continuous

➢ Norm is a uniformly continuous function

We have , || || :X→R . Let x,y ∈ X and ε> 0

Then ||x|| = ||x - y + y ||

≤||x - y|| + ||y||

⇒ ||x|| - ||y|| ≤ ||x - y||→ (1)

Interchanging x and y ,

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||

⇒ ||x|| - ||y|| ≥ - ||x - y|| → (2)

Combining (1) and (2)

- ||x - y|| ≤ ||x|| - ||y|| ≤ ||x - y||

That is ,

| ||x ||- ||y|| | ≤ ||x - y||

Take δ = ε , then whenever ||x - y|| < δ, | ||x|| -|| y|| |< ε

12
 Therefore || || is a uniformly continuous function .

➢ Continuity of addition and scalar multiplication

To show that + : X×X →X and ·: K ×X →X are continuous functions.

Let (x,y) ∈ X×X . To show that + is continuous at (x ,y) , that is , to show

that for each (x,y) ∈ X×X if xn→x and yn→y in X , then

+(xn , yn) →+(x , y) ;

That is ,
xn + yn →x + y .

Consider
||( xn + yn) - (x + y )|| = ||xn - x + yn - y||

≤||xn - x|| + ||yn - y||

Given xn→x and yn→y , for each ϵ > 0, ∃ N1 ∋

ε
||xn - x|| < 2
∀ n ≥N1 , and ∃ N2 ∋

ε
||yn - y|| < 2
∀ n ≥N2

\
Take N = max { N1, N2}

ε ε
Then ||xn - x|| < 2
and ||yn - y|| < 2
∀n≥N

ε ε
Therefore ||(xn + yn) - (x + y)|| ≤ 2
+ 2
= ε ∀n ≥N

That is , xn + yn →x + y

Now to show that ·: K ×X→X is continuous

Let (k , x) ∈ K×X

13
 To show that if kn →k and xn →x , then knxn →kx

Since kn →k , ∀ ε > 0 ∃ N1 ∋ |kn - k| < ε ∀ n ≥ N1

2

ε
Since xn →x , ∀ ε > 0 ∃ N2 ∋ ||xn - x|| < ∀ n ≥ N2
2

Consider ||knxn - kx|| = ||knxn - kx + xnk - xnk ||

= ||xn (kn - k) + k(xn - x)||

≤ ||xn(kn - k)|| + ||k(xn - x)||

= ||xn|| |kn- k| + |k| ||xn - x||

ε ε
≤ ||xn|| 2 + |k| 2

∴ knxn→kx

➢ Examples of normed space

1) Spaces Kn (K = R or C)

For n = 1 , the absolute value of function | | is a norm on K , since ∀k ∈K

We have ,

||k|| = ||k · 1|| = |k| ||1|| , by definition .

But ||1|| is a positive scalar .

∴ ||k|| is a positive scalar multiple of the absolute value function .

∴ any norm on K is a positive scalar multiple of the absolute value

function

For n > 1 , let p≥ 1 be a real number

14
 Kn = { ( x(1) , x(2) , . . . , x(n) ) : x(i) ∈K , i = 1 , 2 , . . . , n }

For x ∈Kn , that is , x = ( x(1) , x(2) , . . . , x(n) ) , define

𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (|𝑥(1)| +. . . + |𝑥(𝑛)| )

Then || ||p is a norm on Kn

When p = 1 ,

Then , ||x||1 = |x(1)| + |x(2)| + . . . + |x(n)|

Since |x(i)| ≥0 ∀ i = 1 , 2 , . . . , n , ||x||1 ≥0

And ||x||1 = 0 ⇔ |x(1)| + . . . +|x(n)| = 0

⇔ |x(i)| = 0 ∀ i

⇔ x(i) = 0 ∀ i

⇔ x = ( x(1) , . . . , x(n) ) = 0

Now ||kx||1 = |kx(1)| + |kx(2)| + . . . + |kx(n)|

= |k| |x(1)| + . . . + |k| |x(n)|

= |k| ( |x(1)| + . . . + |x(n)| )

= |k| ||x||1

||x + y||1 = |(x + y)(1)| + . . . + |(x + y)(n)|

= |x(1) + y(1)| + . . . + |x(n) + y(n)|

≤|x(1)| + |y(1)| + . . . + |x(n)| + |y(n)|

= |x(1)| + . . . + |x(n)| + |y(1)| + . . . + |y(n)|

= ||x||1 + ||y||1

15
 Consider 1<p<∞

𝑝 𝑝 1/𝑝
Now , ||𝑥||𝑝 = ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

𝑝
Since |x(i)| ≥0 ∀i , we have ||x||p ≥0

𝑝 𝑝 1/𝑝
And ||𝑥||𝑝 = 0⇔( |𝑥(1)| +. . . + |𝑥(𝑛)| ) = 0

𝑝
⇔ |𝑥(𝑖)| = 0 ∀i

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 ∀i

⇔ x = ( x(1) , . . . , x(n) ) = 0 .
Now
𝑝 𝑝 1/𝑝
||𝑘𝑥||𝑝 = ( |𝑘𝑥(1)| +. . . + |𝑘𝑥(𝑛)| )

𝑝 𝑝 𝑝 𝑝 1/𝑝
= ( |𝑘| |𝑥(1)| +. . . + |𝑘| |𝑥(𝑛)| )

𝑝 𝑝 1/𝑝
= |𝑘| ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

= |𝑘| ||𝑥||𝑝 .

𝑝 𝑝 1/𝑝
||𝑥 + 𝑦||𝑝 = ( |𝑥(1) + 𝑦(1)| +. . . + |𝑥(𝑛) + 𝑦(𝑛)| )

We have by Minkowski’s inequality ,

1/𝑝 1/𝑝 1/𝑝

( ) ( ) ( )
𝑛 𝑛 𝑛
𝑝 𝑝 𝑝
∑ |𝑥(𝑖) + 𝑦(𝑖)| ≤ ∑ |𝑥(𝑖)| + ∑ |𝑦(𝑖)|
𝑖=1 𝑖=1 𝑖=1
Then

16
 𝑝 𝑝 1/𝑝 𝑝 𝑝 1/𝑝
(
||𝑥 + 𝑦||𝑝 ≤ |𝑥(1)| +. . . + |𝑥(𝑛)| ) (
+ |𝑦(1)| +. . . + |𝑦(𝑛)| )
= ||𝑥||𝑝 + ||𝑦||𝑝

Then , for 1≤p<∞ ,|| ||𝑝 is a norm on Kn

When p = ∞, define ||𝑥||∞ = 𝑚𝑎𝑥 { |𝑥(1)| , |𝑥(2)| ,. . . , |𝑥(𝑛)| }

Then it is a norm on Kn

||𝑥||𝑝 ≥ 0 since each values |x(i)|≥0

So that

max {|x(i)| , i=1, . . . , n} ≥ 0

||𝑥||∞ = 0 ⇔𝑚𝑎𝑥 { |𝑥(𝑖)| : 𝑖 = 1,. . . , 𝑛 }= 0

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 , ∀i

⇔x = 0

||𝑘𝑥||∞ = 𝑚𝑎𝑥 { |𝑘𝑥(1)| ,. . . , |𝑘𝑥(𝑛)| }

= max { |k| |x(1)| , . . . , |k| |x(n)|}

= |k| max {|x(1)| , . . . , |x(n)|}

= |k| ||x||∞

||x + y||∞ = max { |x(1) + y(1)| , . . . , |x(n) + y(n)| }

≤max { |x(1)| + |y(1)| , . . . , |x(n)| + |y(n)| }

17
 ≤max { |x(1)| , . . . , |x(n)| } + max { |y(1)| , . . . , |y(n)| }

= ||x||∞ + ||y||∞

2) Sequence space

∞
𝑝 𝑝 𝑝
Let 1≤p < ∞, 𝑙 = { x = ( x(1) , x(2) , . . . ) ; x(i) ∈K and ∑ |x(j)| <∞} , that is , 𝑙 is the
𝑗=1
𝑝
space of p-summable scalar sequences in K . For x = (x(1) , x(2) , . . . ) ∈ 𝑙 ,

let ||x||p = ( |x(1)|p + |x(2)|p + . . . )1/p . Then it is a norm on lp.

That is , || ||p is a function from lp to R .

If p = 1 , then l1 is a linear space and ||x||1 = ( |x(1)| + |x(2)| + . . . ) is a norm on l1

∞
Let p = ∞. Then 𝑙 is the linear space of all bounded scalar sequences . And ,

||x||∞ = sup { |x(j)| : j = 1, 2, 3, . . . }

∞
Then || ||∞ is a norm on 𝑙

18
 CHAPTER 2

THEOREMS ON NORMED SPACES

a) Let Y be a subspace of a normed space X , then Y and its closure 𝑌 are normed spaces with
the induced norm.
b) Let Y be a closed subspace of a normed space X , for x +Y in the quotient space X/Y, let
|||x +Y||| = inf { ||x+y|| : y ∈Y} . Then ||| ||| is a norm on X/Y , called the quotient norm.

A sequence (xn + Y) converges to x + Y in X/Y iff there is a sequence (yn) in Y , (xn+ yn)
converges to x in X.

c) Let || ||pbe a norm on the linear space Xp , j = 1,2,…. . Fix p such that 1 ≤p ≤∞

For x = (x(1) , x(2) , … , x(m)) that is the product space X =X1× X2 × …× Xm ,

1/𝑝
Let ||𝑥||𝑝 = ( ||𝑥(1)|| 𝑝
1
𝑝
+ ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚
𝑝
) , if 1≤ p <∞

||x||p = max { ||x(1)||1 , … , ||x(m)||m } , if p = ∞.

Then || ||p is a norm on X.

A sequence (xn) converges to x in X ⇔ (xn(j)) converges to x(j) in Xj ∀ j=1,2,…,m.

Proof:
a) Since X is a normed space, there is a norm on X to Y . Since Y is a subspace of X,

|| ||y: Y → R is a function. To show that || ||y is a norm on Y.

For y∈ Y , || y||Y = ||y|| , then

||y||Y ≥ 0 ( ∵||y||≥ 0 ) and ||y||Y = 0 ⇔ y = 0

||ky||Y = ||ky|| = |k| ||y|| = |k| ||y||y .

Let y1 , y2 ∈Y. Then ,

||𝑦1 + 𝑦2||𝑦 = ||𝑦1 + 𝑦2|| ≤ ||𝑦1|| + ||𝑦2|| = ||𝑦1||𝑦 + ||𝑦2||𝑦

Now the continuity of addition and scalar multiplication shows that 𝑌is a subspace of X, since if
xn →x and yn → y , xn , yn ∈ 𝑌 , then

xn + yn →x + y (by continuity of addition) and

19
 kxn →kx (by continuity of scalar Xn) .

Since 𝑌is closed , x + y ∈ 𝑌 and kx ∈ 𝑌 . Therefore 𝑌 ≤ X.

∴ norm on X induces a norm on Y and 𝑌

b) X/Y , the quotient space equals X/Y={ x + Y : x ∈X }.

|||x + y||| = inf { ||x + y|| : y ∈Y }

Claim: ||| ||| is a norm on X/Y , called quotient norm

• Let x ∈X ,

|||x + Y||| = inf { ||x + y|| : y ∈Y } ≥ 0.

∴ |||x + Y||| ≥0 .

If |||x + y||| = 0 ( 0 in X/Y is Y) , then there is a sequence (yn) in Y ∋

||x + yn || →0

⇒ x + yn →0

⇒ yn →-x

Since yn ∈Y and Y is closed

-x ∈Y ⇔ x ∈Y (∵ Y is a subspace)

⇔x + Y = Y , zero in X/Y.

• For k ∈K ,

|||k(x + Y)||| = |||kx + Y|||

= inf { ||k(x + y)|| : y ∈Y}

= inf { |k| ||x + y|| : y ∈Y}

= |k| inf { ||x + y|| : y ∈Y}

= |k| |||x + Y||| .

• Let x1 , x2 ∈X . Then

|||x1 + Y||| = inf { ||x1 + y|| : y ∈Y } . Then ∃ y1 ∈ Y ∋

ε
|||x1 + Y||| + > ||x1 + y1|| , and
2

20
 |||x2 + Y||| = inf { ||x2 + y|| : y ∈Y} , Then ∃y2 ∈Y ∋
ε
|||x2 + Y||| + 2
> ||x2 + y2|| .

||x1 + y1 + x2 + y2 || ≤ ||x1 + y1|| + ||x2 + y2||

ε ε
≤ |||x1 + Y||| + 2 + |||x2 + Y||| + 2

Let y = y1 + y2 ∈Y . Then ,

||(x1+x2) + y|| ≤ |||x1 + Y||| + |||x2 + Y||| + ℇ —(1)

Now , |||(x1 + Y) + (x2 + Y)||| = |||x1 + x2 + Y|||

=inf { ||x1 + x2 + y|| : y ∈Y }

< ||x1 + x2 + y||

≤ |||x1 + Y||| + |||x2 + Y||| + ℇ (by (1) )

since ℇ is arbitrary , we have

|||(x1 + Y) + (x2 + Y)||| ≤|||x1 + Y||| + |||x2 + Y|||

∴ ||| ||| is a norm on X/Y.

Let (xn + Y) be a sequence in X/Y . Assume that (yn) is a sequence in Y ∋ (xn + yn) converges
to x in X.
That is , (xn - x + yn) converges to 0 . —(1)
Claim: (xn + Y) converges to x + Y.
Consider
|||xn + Y - (x+Y)||| = |||(xn - x) + Y|||

= inf { ||xn - x + yn|| : y ∈Y }

≤ ||xn - x + yn|| ∀yn ∈Y .

Then by (1) , xn + Y converges to x + Y in X/Y.

Conversely assume that the sequence (xn + Y) → x + Y in X/Y.

Consider |||xn + Y - (x + Y)||| = |||xn - x + Y|||

= inf { ||xn -x + y|| : y ∈Y }

Then we can choose yn ∈Y ∋

21
 1
||xn - x + yn|| < |||(xn - x) + Y||| + 𝑛 , n=1,2,3,….

Since xn+Y →x+Y , we get

(xn - x + yn) converges to zero as n →∞

That is , (xn + yn) converges to x in X as n →∞

c) Consider 1≤ p < ∞

Given that
𝑝 𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (||𝑥(1)||1 + ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚)

Clearly , ||x||p ≥0 .
𝑝
Since each ||𝑥(𝑖)||𝑖 ≥ 0 .

𝑝
||x||p = 0 ⇔ |𝑥(𝑗)|𝑗 = 0 ∀ j = 1, . . . ,m

⇔ x(j) = 0 ∀ j.

⇔ x = (x(1), . . . ,x(m)) = 0
1/𝑝
( 𝑝
||kx||p = ||𝑘𝑥(1)||1 +. . . + ||𝑘𝑥(𝑚)||𝑚
𝑝
)
1/𝑝
= ( |𝑘| ||𝑥(1)|| )
𝑝 𝑝 𝑝 𝑝
1
+. . . + |𝑘| ||𝑥(𝑚)||𝑚

1/𝑝
( 𝑝
= |𝑘| ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= |𝑘| ||𝑥||𝑝 , k∈K and x∈X

1/𝑝
(
Now, ||𝑥 + 𝑦||𝑝 = ||𝑥(1) + 𝑦(1)||1 +. . . + ||𝑥(𝑚) + 𝑦(𝑚)||𝑚
𝑝 𝑝
)
(by Minkowski’s inequality)

≤
1/𝑝
( (||𝑥(1)||1
+ ||𝑦(1)||1 )𝑝
(
+. . . + ||𝑥(𝑚)||𝑚 + ||𝑦(𝑚)||𝑚 ))
𝑝

1/𝑝 1/𝑝

( ) ( )
𝑚 𝑚
𝑝 𝑝
≤ ∑ ||𝑥(𝑗)||𝑗 + ∑ ||𝑦(𝑗)||𝑗 (Minkowski’s inequality )
𝑗=1 𝑗=1

22
 1/𝑝
( 𝑝
= ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= ||x||p + ||y||p

Now suppose p =∞

||x||∞ = max { ||x(1)||1 , . . . , ||x(m)|| m }

||x||∞ ≥0 Since ||x(j)|| ≥0 , ∀j

||x||∞ = 0 ⇔ ||x(m)|| = 0 ∀m

⇔ x(m) = 0 ∀m

⇔x=0

||kx||∞ = max { ||kx(1)||1 , . . . , ||kx(m)||m }

= |k| max { ||x(1)||1 , . . . , ||x(m)||m }

= |k| ||x||∞

||x + y||∞ = max { ||x(1) + y(1)||1, . . . , ||x(m) + y(m)||m }

≤max { ||x(1)||1 + ||y(1)||1 , . . . , ||x(m)||m + ||y(m)||m }

= max { ||x(1)||1 , . . . , ||x(m)||m } + max { ||y(1)||1 , . . . , ||y(m)||m }

= ||x||∞ + ||y||∞

We now consider ,

𝑝 𝑝 1/𝑝
( )
||𝑥𝑛 − 𝑥(1)||𝑝 = ||𝑥𝑛(1) − 𝑥(1 ||1 +. . . + ||𝑥𝑛(𝑚) − 𝑥(𝑚)||𝑚)

Then

xn →x in X ⇔ ||xn - x ||p → 0
𝑝
⇔ ||xn(j) - x(j)||𝑗 → 0

⇔ xn(j) - x(j) → 0

⇔ xn(j) → x(j) in X ∀j .

23
 RIESZ LEMMA

Let 𝑋 be a normed space . 𝑌be a closed subspace of 𝑋and 𝑋 ≠ 𝑌 . Let 𝑟be a real number
such that 0 < 𝑟 < 1 . Then there exist some xr ∈X such that ||xr|| = 1 and

r<dist ( xr , Y )≤1

Proof :
We have ,

dist (x , Y) = inf { d(x , y) : y∈Y}

= inf { ||x - y|| : y ∈Y}

Since Y ≠X , consider x ∈X ∋ x∉Y.

If dist(x , Y) = 0 , then ||x - y|| = 0 ⇒ x∈𝑌= Y ( ∵ Y is closed )

Therefore ,

dist (x , Y) ≠0

That is ,
dist (x , Y) > 0
1
Since 0 < r < 1 , >1
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
⇒ > dist (x , Y)
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
That is , is not a lower bound of { ||x - y|| : y ∈Y }
𝑟

𝑑𝑖𝑠𝑡(𝑥 , 𝑌)
Then ∃ y0 ∈Y ∋ ||x - y0|| <
𝑟
→(1)
𝑥 − 𝑦0
Let xr = . Then xr ∈X
||𝑥 − 𝑦0||

( ∵y0 ∈Y , x∉Y ⇒x - y0 ∈X and ||x - y0|| ≠0 )

24
 𝑥 − 𝑦0 ||𝑥 − 𝑦0||
Then ||xr|| = || || = =1
||𝑥 − 𝑦 ||0
||𝑥 − 𝑦0||

Now to prove r < dist(𝑥𝑟, 𝑌)≤1

We have dist(xr , Y) = inf { ||xr - y|| : y∈Y }

≤||xr - y|| ∀y∈Y

In particular, 0∈ 𝑌, so that dist(xr , Y) ≤||xr - 0|| = 1

That is ,

dist (xr , Y) ≤1

Now ,
𝑥 − 𝑦0
dist (xr , Y) = dist ( ,Y )
||𝑥−𝑦0||

1
= dist ( x - y0 , Y)
||𝑥−𝑦0||

1
= inf { ||x - y0 - y|| : y∈ 𝑌}
||𝑥−𝑦0||

1
= inf {||x - (y0+ y)|| : y0 + y∈ 𝑌}
||𝑥−𝑦0||

1
= dist (x , Y)
||𝑥−𝑦0||

𝑟
> dist (x , Y) by (1)
𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)

⇒ dist (xr , Y) > r

That is ,

r < dist (xr , Y) ≤ 1

25
 CONCLUSION

This project discusses the concept of normed linear space that is fundamental to
functional analysis . A normed linear space is a vector space over a real or complex
numbers ,on which the norm is defined . A norm is a formalization and generalization to
real vector spaces of the intuitive notion of “length” in real world
In this project , the concept of a norm on a linear space is introduced and thus
illustrated . It mostly includes the properties of normed linear spaces and different proofs
related to the topic.

26
 BIBLIOGRAPHY

➢ Balmohan V Limaye , Functional Analysis

➢ S.H Friedberg , Arnold J.Insel and Lawrence E.Spence , Linear Algebra , 2 nd Edition, PH
Inc .
➢ S. Narayanan and Mittai , A Text Book of Matrices , Revised Edition, S. Chand

27
DIRECT PRODUCTS AND FINITELY GENERATED ABELIAN
GROUPS

Project report submitted to

The Kannur University

for the award of the degree

of

Bachelor of Science

by

ASWATHI BABU P B

DB18CMSR20

Under the guidance of

MS. Sneha P Sebastian

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report ‘DIRECT PRODUCTS AND FINITELY

GENERATED ABELIAN GROUPS’ is the bonafide project of ASWATHI BABU P B
who carried out the project under my supervision.

Mrs.Riya Baby Ms.Sneha P Sebastian

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, ASWATHI BABU P B, hereby declare that this project report entitled ‘DIRECT
PRODUCTS AND FINITELY GENERATED ABELIAN GROUPS’ is an original record of
studies and bonafide project carried out by me during the period from November 2019 to March
2020, under the guidance of Ms.Sneha P Sebastian, Department of Mathematics, Don Bosco
Arts and Science College, Angadikadavu and has not been submitted by me elsewhere for the
award of any degree, diploma, title or recognition, before.

ASWATHI BABU P B
Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great help to me
during the course of my dissertation. First and foremost I thank the almighty, for his blessing and
protection during the period of this work. I express my thanks to Dr. Fr. Francis Karakkatt,
Principal, for support in the completion of this dissertation. I express my gratitude to Ms.Sneha P
Sebastian, my project supervisor, for the constant encouragement, valuable guidance and timely
corrections, which made this work a success.

I am also indebted to all my classmates and friends who supported me throughout the
study. I would like to express my thanks to my parents and dear ones for their constant
encouragement and support. I also thank all those who helped me directly or indirectly to
complete this project.

ASWATHI BABU P B
 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 preliminary 2–3
03 Chapter 1 4 – 12
04 Chapter 2 13 – 17
05 Conclusion 18
06 Bibliography 19
 INTRODUCTION

In mathematics, a group is a set equipped with a binary operation that combines any two
elements to form a third element in such a way that the three conditions called group axioms are
satisfied , namely associativity , identity and invertability.

Let us take a moment to review our present stockpile of groups. Starting with finite
groups, we have the cyclic group ℤ𝑛 ,the symmetric group 𝑆𝑛 , and the alternating group 𝐴𝑛 for
each positive integer n. We also have the dihedral group 𝐷𝑛 and klein 4-group . Of course we
know that subgroups of these groups exists. Turning to infinite groups , we have ℤ, ℝ, ℂ under
addition , and their non zero elements under multiplication we also have the group 𝑆𝐴 of all
permutation of an infinite set 𝐴 , as well as various groups formed from matrices .

One purpose of this section is to show a way to use known groups as building blocks to
form more groups. Given two groups 𝐺 and 𝐻, it is possible to construct a new group from the
cartesian product of 𝐺 and 𝐻 . Conversely , given a large group , it is sometimes possible to
decompose the group ; that is , a group is sometimes isomorphic to the direct product of two
smaller groups. Rather than studying a large group , it is often easier to study the component
group of that group.

1
 PRELIMINARY

Groups : A non empty set 𝐺 together with an operation ∗ is said to be a group , denote by
(𝐺 ,∗) , if it satisfy the following axioms.

• Closure property
• Associative property
• Existence of identity
• Existence of inverse

Abelian group

A group (𝐺 ,∗) is said to be abelian if it satisfies commutative law .

Finite group

If the underlying set G of the group (𝐺 ,∗) consist of finite number of elements , then the
group is finite group .

Infinite group

A group that is not finite is an infinite group .

Order of a group : The number of elements in a finite group is called the order of the group ,
denoted by 𝑂(𝐺 ) .

Example

Show that the set of integers ℤ is a group with respect to the operation of addition of
integers.

ℤ = {… … … . −3, −2, −1,0,1,2,3, … … … }

Since the addition of two integers gives an integer , it satisfy closure property .

2
If 𝑎, 𝑏, 𝑐 𝜖 ℤ then the (𝑎 + 𝑏) + 𝑐 = 𝑎 + (𝑏 + 𝑐) , hence associativity holds .

There is a number 0 𝜖 ℤ such that 0 + 𝑎 = 𝑎 + 0 , hence identity exists

If 𝑎 𝜖 ℤ then there exists – 𝑎 𝜖 ℤ , such that −𝑎 + 𝑎 = 0 = 𝑎 + −𝑎

Therefore inverse exist .

Therefore ℤ is a group under addition .

Subgroup

A subset 𝐻 of 𝐺 is said to be a subgroup of 𝐺 if 𝐻 itself is a group under the same operation in

𝐺.

There are two different types of group structure of order 4 .

ℤ4 = { 0,1,2,3}

Klein 4 – group , 𝑉 = {𝑒, 𝑎, 𝑏, 𝑐}

Cyclic group

A group 𝐺 is cyclic if there is some element ‘𝑎’ in 𝐺 that generate 𝐺. And the element ‘𝑎’ is
called generator of 𝐺.

Group Homomorphism

A function Ѱ: 𝐺 → 𝐺′ is a group homomorphism ( or simply homomorphism ).

If Ѱ(𝑎𝑏) = Ѱ(𝑎) Ѱ(𝑏) hold for all 𝑎 , 𝑏 ∈ 𝐺 , is called homomorphism property .

Isomorphism

A one to one and onto homomorphism Ѱ: 𝐺 → 𝐺′ is called an isomorphism .

3
 CHAPTER – 1

DIRECT PRODUCT OF GROUPS

Definition

The Cartesian product of sets 𝑆, 𝑆2 , … … . , 𝑆𝑛 is the set of all ordered n-tuples (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ),
where 𝑎𝑖 ∈ 𝑆𝑖 for 𝑖 = 1,2,3, … … . , 𝑛. The Cartesian product is denoted by either

𝑛
𝑆1 × 𝑆2 × … … .× 𝑆𝑛 or by Π𝑖=1 𝑆𝑖 .

Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups and let us use multiplicative notation for all the group operations.

If we consider 𝐺𝑖 as a set , 𝑖 = 1,2, … … . 𝑛 . we have the products 𝐺1 × 𝐺2 × … … . ,× 𝐺𝑛 we

𝑛 𝑛
denote it by Π𝑖=1 𝐺𝑖 . This product is called direct-product of groups. We can make Π𝑖=1 𝐺𝑖 into a
group by means of a binary operation of multiplication by components.

Theorem
𝑛
Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups. For (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) and (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) in Π𝑖=1 𝐺𝑖 define ;

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

𝑛
Then Π𝑖=1 𝐺𝑖 is a group.

4
Proof

We have ,

𝑛
Π𝑖=1 𝐺𝑖 = {(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∶ 𝑎𝑖 ∈ 𝐺𝑖 }

(1) Closure property

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

And we have ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

Here 𝑎𝑖 ∈ 𝐺𝑖 and 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

∵ 𝐺𝑖 is a group , 𝑎𝑖 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

𝑛
⇒ (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

𝑛
i.e. Π𝑖=1 𝐺𝑖 is closed under the binary operation.

(2) Associativity
𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ), (𝑐1, 𝑐2 , … … . , 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

We have,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

𝑛
= (𝑎1 𝑏1 𝑐1 , 𝑎2 𝑏2 𝑐2 , … … . , 𝑎𝑛 𝑏𝑛 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

[(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

= [𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )
= [(𝑎1 𝑏1 )𝑐1 , (𝑎2 𝑏2 )𝑐2, … … . , (𝑎𝑛 𝑏𝑛 )𝑐𝑛 ]
= [𝑎1 (𝑏1 𝑐1 ), 𝑎2 (𝑏2 𝑐2), … … . , 𝑎𝑛 (𝑏𝑛 𝑐𝑛 )]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[𝑏1 𝑐1, 𝑏2 𝑐2 , … … . , 𝑏𝑛 𝑐𝑛 ]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )]

5
Hence associativity holds.

(3) Existence of identity

If 𝑒𝑖 is the identity element in 𝐺𝑖 .

Then,

𝑛
(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Also for ,

𝑛
(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖 ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) = (𝑎1 𝑒1 , 𝑎2 𝑒2 , … … . , 𝑎𝑛 𝑒𝑛 )

= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )

𝑛
∴ (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) is the identity element ‘𝑒’ in Π𝑖=1 𝐺𝑖

(4) Existence of inverse

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Here 𝑎𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛.

Since 𝐺𝑖 is a group ,

∃ an inverse element 𝑎𝑖−1 in 𝐺𝑖 : 𝑎𝑖 𝑎𝑖−1 = 𝑒𝑖 𝑖 = 1,2, … … . , 𝑛

𝑛
Clearly, (𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) ∈ Π𝑖=1 𝐺𝑖 &

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 )

𝑛
Hence Π𝑖=1 𝐺𝑖 is a group.

6
Note
𝑛
If the operation of each 𝐺𝑖 is a commutative. We sometimes use additive notation in Π𝑖=1 𝐺𝑖 and
𝑛
refer to Π𝑖=1 𝐺𝑖 as the direct sum of the group 𝐺𝑖 . The notation ⨁𝑛𝑖=1 𝐺𝑖 , especially with abelian
groups with operation +.

The direct sum of abelian groups 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 may be written 𝐺1 ⨁𝐺2 ⨁ … … ⨁𝐺𝑛

• Direct product of abelian group is abelian

Example

Q. Check whether ℤ2 × ℤ3 is cyclic or not.

ℤ2 = {0,1}

ℤ3 = {0,1,2}

ℤ2 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2)}

Consider,

1(1,1) = (1,1)

2(1,1) = (1,1) + (1,1) = (0,2)

3(1,1) = (1,1) + (1,1) + (1,1) = (1,0)

4(1,1) = 3(1,1) + (1,1) = (1,0) + (1,1) = (0,1)

5(1,1) = 4(1,1) + (1,1) = (0,1) + (1,1) = (1,2)

6(1,1) = 5(1,1) + (1,1) = (1,2) + (1,1) = (0,0)

∴ (1,1) is a generator of ℤ2 × ℤ3

∴ ℤ2 × ℤ3 is a cyclic group generated by (1,1).

7
Q. Check whether ℤ3 × ℤ3 is cyclic or not.

ℤ3 = {0,1,2}

ℤ3 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2), (2,0), (2,1), (2,2)}

1(0,1) = (0,1)

2(0,1) = (0,2)

3(0,1) = (0,3) = (0,0) ∴ order (0,1) = 3

1(0,2) = (0,2)

2(0,2) = (0,4) = (0,1)

3(0,2) = (0,6) = (0,0) ∴ order (0,2) = 3

Every element added to itself three times gives the identity. Thus no element can
generate the group. Hence ℤ3 × ℤ3 is not cyclic.

similarly ℤ𝑚 × ℤ𝑚 is not cyclic for any 𝑚.

Theorem

The group ℤ𝑚 × ℤ𝑛 is cyclic and is isomorphic to ℤ𝑚𝑛 if and only if 𝑚 and 𝑛 are relatively
prime, that is, the gcd of 𝑚 and 𝑛 is 1.

Proof

Suppose ℤ𝑚 × ℤ𝑛 is cyclic and isomorphic to ℤ𝑚𝑛 .

To show that 𝑚 and 𝑛 are relatively prime.

Suppose not, let d be the 𝑔𝑐𝑑 of 𝑚 and 𝑛.

So that 𝑑 > 1

8
 𝑚𝑛
Consider , which is an integer since 𝑑|𝑚 and 𝑑|𝑛
𝑑

𝑚𝑛
Let (𝑟, 𝑠) be an arbitrary element of ℤ𝑚 × ℤ𝑛 , add (𝑟, 𝑠) repeatedly times
𝑑

(𝑟, 𝑠) + (𝑟, 𝑠)+, … … . , +(𝑟, 𝑠) } 𝑚𝑛

𝑑
𝑡𝑖𝑚𝑒𝑠 = (0,0)

∴ no element of ℤ𝑚 × ℤ𝑛 having order 𝑚𝑛. ∴ no element of ℤ𝑚 × ℤ𝑛 can generate ℤ𝑚 × ℤ𝑛

which is not possible. ∵ ℤ𝑚 × ℤ𝑛 is cyclic. Hence 𝑔𝑐𝑑(𝑚, 𝑛) = 1.

i.e. 𝑚 and 𝑛 are relatively prime.

Conversely, suppose 𝑚 and 𝑛 are relatively prime, i.e. gcd(𝑚, 𝑛) = 1

To show that ℤ𝑚 × ℤ𝑛 is cyclic.

If ℤ𝑚 × ℤ𝑛 is cyclic, then it is isomorphic to ℤ𝑚𝑛 , ∵ ℤ𝑚 × ℤ𝑛 has 𝑚𝑛 elements.

Consider the cyclic subgroup of ℤ𝑚 × ℤ𝑛 generated by the element (1,1).The order of this
cyclic subgroup is the smallest power of (1,1),that gives the identity (0,0). Here taking a power
of (1,1) in our additive notation will involve adding (1,1) to itself repeatedly.

Consider (1,1) + (1,1)+, … … . , +(1,1)

If we add first coordinates 𝑚 times , we get zero.

∴ order of first coordinate = 𝑚.

Similarly , Order of second coordinate = 𝑛.

The two coordinates together become zero. If we add them 𝑙𝑐𝑚(𝑚, 𝑛) times.

∵ gcd(𝑚, 𝑛) = 1, We get the 𝑙𝑐𝑚 = 𝑚𝑛.

i.e. (1,1) generates a cyclic subgroup of ℤ𝑚 × ℤ𝑛 of order 𝑚𝑛 , which is the order of the whole
group.

⇒ ℤ𝑚 × ℤ𝑛 =< (1,1) >

9
 ⇒ ℤ𝑚 × ℤ𝑛 is cyclic.

Corollary

The group ⨅𝑛𝑖=1 ℤ𝑚𝑖 is cyclic and isomorphic to ℤ𝑚1 𝑚2 …….𝑚𝑛 if and only if the numbers 𝑚𝑖 for
𝑖 = 1,2, … … . , 𝑛 are such that the 𝑔𝑐𝑑 of any two of them is 1.

Example

If 𝑛 is written as a product of powers of distinct prime numbers , as in ,

𝑛 = (𝑝1 )𝑛1 . (𝑝2 )𝑛2 … … . (𝑝𝑛 )𝑛𝑟

Then ℤ𝑛 is isomorphic to ℤ(𝑝1 )𝑛1 × ℤ(𝑝2 )𝑛2 × … … .× ℤ(𝑝𝑟 )𝑛𝑟 .

In particular , ℤ72 is isomorphic to ℤ8 × ℤ9 .

Consider set of integers ℤ, cyclic subgroup of ℤ is of the form 𝑛ℤ , 𝑛 ∈ ℤ. Consider 2ℤ and 3ℤ ,

then < 2 > ∩ < 3 > = < 6 >

∴ if we take 𝑟ℤ , 𝑠ℤ of ℤ , then the 𝑙𝑐𝑚(𝑟, 𝑠) =generator of < 𝑟 > ∩ < 𝑠 >

Using this we can define the 𝑙𝑐𝑚 of the positive integers.

Definition

Let 𝑟1 , 𝑟2 , … … . , 𝑟𝑛 be positive integers. Their least common multiple (abbreviated lcm ) is the
positive generator of the cyclic group of all common multiples of the 𝑟𝑖 , that is the cyclic group
of all integers divisible by each 𝑟𝑖 for 𝑖 = 1,2, … … . , 𝑛.

10
Theorem

Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ ⨅𝑛𝑖=1 𝐺𝑖 .

If 𝑎𝑖 is of finite order 𝑟𝑖 in 𝐺𝑖 , then the order of (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) in ⨅𝑛𝑖=1 𝐺𝑖 is equal to the least
common multiple of all the 𝑟𝑖 .

Proof

Given,

𝑟
order of 𝑎1 = 𝑟1 ⇒ 𝑎11 = 𝑒1 in 𝐺1

𝑟
order of 𝑎2 = 𝑟2 ⇒ 𝑎22 = 𝑒2 in 𝐺2

.
.
.
𝑟
order of 𝑎𝑛 = 𝑟𝑛 ⇒ 𝑎𝑛𝑛 = 𝑒𝑛 in 𝐺𝑛 .

We have to find a power 𝑘 for (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ).

So that (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )𝑘 = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ).

The power must simultaneously be a multiple of 𝑟1 , multiple of 𝑟2 and so on. But 𝑘 is the least
positive integers having the above property.

∴ 𝑘 = 𝑙𝑐𝑚(𝑟1 , 𝑟2 , … … . , 𝑟𝑛 ).

Q. Find the order of (8,4,10) in the group ℤ12 × ℤ60 × ℤ24 .

𝑂(8) = 3 in 𝑍12

11
 𝑂(4) = 15 in 𝑍60

𝑂(10) = 12 in 𝑍24

𝑂(8,4,10) = 𝑙𝑐𝑚(3,15,12) = 60

Q. Find a generator of ℤ × ℤ2

ℤ × ℤ2 = {(𝑛, 0), (𝑛, 1): 𝑛 ∈ ℤ}

(𝑛, 0) = 𝑛(1,0)
(𝑛, 1) = (𝑛, 0) + (0,1) = 𝑛(1,0) + (0,1)

∴ ℤ × ℤ2 is generated by {(1,0), (0,1)}

In general , ℤ × ℤ𝑛 is generated by ,

{(1,0,0, … … . ,0), (0,1,0, … … . ,0), … … . , (0,0, … … . ,1)}

Q. Find the order of (3,10,9) in (ℤ4 , ℤ12 , ℤ15 )

𝑂(3) = 4 in ℤ4

𝑂(10) = 6 in ℤ12

𝑂(9) = 5 in ℤ15

∴ 𝑂(3,10,9) = 𝑙𝑐𝑚(4,6,5)
= 60

∴ order of (3,10,9) in ℤ4 × ℤ12 × ℤ15 is 60.

12
 CHAPTER-2

FUNDAMENTAL THEOREM OF FINITELY GENERATED ABELIAN

GROUPS

Every finitely generated abelian group 𝐺 is isomorphic to a direct product of cyclic groups in the
form,

ℤ(𝑝1 )𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛 × ℤ × ℤ × ℤ × … … .× ℤ

Where the 𝑝𝑖 are primes , not necessarily distinct and the 𝑟𝑖 are positive integers.

Remark

• The direct product is unique except for possible rearrangement of the factors.
• The number of factors ℤ is unique and this number is called Betti number.

Example

Find all abelian groups , upto isomorphism of order

1)8 , 2)16 , 3)360

(1) Order 8

8=1×8

8 = 2 × 4 = 2 × 22

8 = 2×2×2

3 non-isomorphic groups are ℤ8 , ℤ2 × ℤ4 ,

ℤ2 × ℤ2 × ℤ2
13
(2) Order 16

16 = 1 × 16 = 1 × 24

16 = 2 × 8 = 2 × 23

16 = 4 × 4 = 22 × 22

16 = 2 × 2 × 2 × 2

16 = 2 × 2 × 22

ℤ16 , ℤ2 × ℤ8 , ℤ4 × ℤ4 , ℤ2 × ℤ2 × ℤ2 × ℤ2 , ℤ2 × ℤ2 × ℤ4

(3) Order 360

360 = 22 ⋅ 32 ⋅ 5

Possibilities are,

1) ℤ8 × ℤ9 × ℤ5

2) ℤ2 × ℤ4 × ℤ9 × ℤ5

3) ℤ2 × ℤ2 × ℤ2 × ℤ9 × ℤ5

4) ℤ8 × ℤ3 × ℤ3 × ℤ5

5) ℤ2 × ℤ4 × ℤ3 × ℤ3 × ℤ5

6) ℤ2 × ℤ2 × ℤ2 × ℤ3 × ℤ3 × ℤ5

Definition

A group 𝐺 is decomposable if it is isomorphic to a direct product of two proper non-trivial

subgroups , otherwise 𝐺 is indecomposable.

14
Example

ℤ6 is decomposable while ℤ5 is indecomposable.

ℤ6 is isomorphic to ℤ2 × ℤ3

ℤ𝑚𝑛 is isomorphic to ℤ𝑚 × ℤ𝑛 , if 𝑚 and 𝑛 are prime.

Theorem

The finite indecomposable abelian groups are exactly the cyclic groups with order a power of a
prime.

Proof

Let 𝐺 be a finite indecomposable abelian group ∵ 𝐺 is finitely generated , we can apply

fundamental theorem of finitely generated abelian groups.

∴ 𝐺 ≅ ℤ(𝑝)𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

∵ 𝐺 is indecomposable and ℤ(𝑝𝑖 )𝑟𝑖 ’s are proper subgroups we get in the above , there is only one
factor say ℤ(𝑝𝑖 )𝑟𝑖 which is cyclic group with order a prime power.

Theorem

If 𝑚 divides the order of a finite abelian group , then 𝐺 has a subgroup of order 𝑚.

Proof

Given 𝐺 is a finite abelian group.

∴ we can apply Fundamental Theorem ,

15
Hence,

𝐺 ≅ ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛

1 2 𝑛

Here all primes need not be distinct.

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 . 𝑝22 … … 𝑝𝑛𝑛

Let 𝑚 is a +𝑣𝑒 integer which divides 𝑂(𝐺).

0 ≤ 𝑠𝑖 ≤ 𝑟𝑖 By theorem , “ let 𝐺 be a cyclic group with 𝑛 elements and generated by 𝑎. Let

𝑛
𝑏 ∈ 𝐺 , 𝑏 = 𝑎𝑠 , then ‘ 𝑏’ generates a cyclic subgroup 𝐻 of 𝐺 containing elements , where
𝑑

𝑑 = gcd(𝑛, 𝑠).”

𝑟
𝑟 −𝑠 𝑝𝑖 𝑖
𝑝𝑖 𝑖 𝑖 generates a cyclic subgroup of ℤ𝑝𝑟𝑖 having order 𝑟 𝑟 −𝑠𝑖
𝑖 𝑔𝑐𝑑(𝑝𝑖 𝑖 ,𝑝𝑖 𝑖 )

𝑟
𝑝𝑖 𝑖 𝑠
= 𝑟 −𝑠𝑖 = 𝑝𝑖 𝑖
𝑝𝑖 𝑖

𝑟 −𝑠𝑖 𝑠
∴ 𝑂(< 𝑝𝑖 𝑖 >) = 𝑝𝑖 𝑖

𝑟 −𝑠1 𝑠
i.e. < 𝑝11 > is a subgroup of ℤ𝑝𝑟1 having order 𝑝11 .
1

𝑟 −𝑠2 𝑠
< 𝑝22 > is a subgroup of ℤ𝑝𝑟2 having order 𝑝22 .
2

…………………………………………………………

𝑟 −𝑠𝑛 𝑠
< 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟𝑛 having order 𝑝𝑛𝑛 .
𝑛

𝑟 −𝑠1 𝑟 −𝑠2 𝑟 −𝑠𝑛

∴ < 𝑝11 > × < 𝑝22 > × … … .× < 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛
1 2 𝑛
𝑠 𝑠 𝑠
having order 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛 = 𝑚.

16
Theorem

If 𝑚 is a square free integer , that is 𝑚 is not divisible by the square of any prime . Then every
abelian group of order 𝑚 is cyclic.

Proof

Let 𝑚 be a square free integer , then 𝑝𝑖 ⫮ 𝑚 for every 𝑖 greater than 1 for a prime 𝑝.

Given 𝐺 is a finite abelian group having order 𝑚 , by fundamental theorem , then

𝐺 ≅ ℤ(𝑝1 )𝑟1 × ℤ(𝑝2 )𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛

∵ 𝑂(𝐺) is a square free integer , the only possibility

𝑟1 = 𝑟2 = … … . . = 𝑟𝑛 = 1

Then,

𝐺 ≅ ℤ𝑝1 × ℤ𝑝2 × … … .× ℤ𝑝𝑛

≅ ℤ𝑝1 ,𝑝2 ,…….,𝑝𝑛 , which is cyclic.

Example

15 is a square free integer. So an abelian group of order 15 is cyclic.

17
 CONCLUSION

Direct product of groups is the product 𝐺1 × 𝐺2 , … … . 𝐺𝑛 ,where each 𝐺𝑖 is a set. We have

discussed about definition and some properties related to the direct product of groups. The
fundamental theorem of finitely generated abelian group helped us to get a deeper understanding
about the topic. The theorems gives us complete structural information about abelian group, in
particular finite abelian group. We have also discussed some examples in order to develope more
intrest in algebra.

18
 BIBLIOGRAPHY

• A first course in Abstract Algebra – 7th edition.

John . B . Fraleigh

• Contomporary Abstract Algebra – 7th edition.

Joseph . A . Gallian

• Topics In Algebra – 2nd edition

I . N . Herstein

19
 GRAPH COLORING

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
CHRISTY JOY
DB18CMSR21
Under the guidance of
MRS. Riya Baby

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report ‘GRAPH COLORING’ is the bona fide
project of CHRISTY JOY who carried out the project under my supervision.

Mrs. Riya Baby

Supervisor, HOD

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, CHRISTY JOY, hereby declare that this project report entitled “GRAPH
COLORING” is an original record of studies and bona fide project carried out by
me during the period from November 2019 to March 2020, under the guidance of
Mrs. Riya Baby, Department of Mathematics, Don Bosco Arts and Science College,
Angadikadavu and has not been submitted by me elsewhere for the award of any
degree, diploma, title or recognition, before.

CHRISTY JOY
Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great
help to me during the course of my dissertation. First and foremost I thank the
almighty, for his blessing and protection during the period of this work. I express
my thanks to Dr. Fr. Francies Karakkatt, Principal, for support in the completion of
this dissertation. I express my gratitude to Mrs. Riya Baby, my project supervisor,
for the constant encouragement, valuable guidance and timely corrections, which
made this work a success.

I am also indebted to all my classmates and friends who supported me

throughout the study. I would like to express my thanks to my parents and dear ones
for their constant encouragement and support. I also thank all those who helped me
directly or indirectly to complete this project.

CHRISTY JOY
 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 Basic Concepts 2

03 Chapter 1-Coloring 3–6

04 Chapter 2-Chromatic Number 7 – 16

05 Chapter 3-Applications Of Graph Colouring 17 – 19

06 Conclusion 20
07 Bibliography 21
 INTRODUCTION

A proper coloring of a graph is an assignment of colors to the vertices of the graph so that

no two adjacent vertices have the same color.

Usually we drop the word “proper” unless other types of coloring are also under

discussion. Of course, the “colors” don’t have to be actual colors ; may can be any distinct

labels - integers ,for examples , if a graph is not connected , each connected component can

be colored independently; except where otherwise noted , we assume graphs are

connected. We also assume graphs are simple in this section. Graph coloring has many

applications in addition to its intrinsic interest.

In the same way the most important concept of graph coloring is utilized in

resource allocation, scheduling. Also, paths, walks and circuits in graph theory are used in

tremendous applications say travelling salesman problem, database design concepts,

resource networking.

This project deals with coloring which is one of the most important topics in

graph theory. In this project there are three chapters. First chapter is coloring . The second

chapter is chromatic number. The last chapter deals with application of graph coloring.

1
 BASIC CONCEPTS

1. GRAPH
A graph is an ordered triplet. G=(V(G), E(G), I(G)); V(G) is a non empty set, E(G) is a set

disjoint from V(G) and I(G) is an incidence map that associates each element of E(G) and

unrecorded pair of element of V(G). The elements of V(G) are called vertices (or nodes or

points) of G and the elements of E(G) are Called edges or lines of G.

2. MULTIPLE EDGE / PARALLEL EDGE

A set of 2 or more edges of a graph G is called a multiple edge or parallel edge if they

have the same end vertices.

3. LOOP
An edge for which the 2 end vertices are same is called a loop.

4. SIMPLE GRAPH
A graph is simple if it has no loop and no multiple edges.

5. DEGREE
Let G be a graph and v € V the number of edge incident at V in G is called the degree or

vacancy of the vertex v in G.

2
 CHAPTER - 1

COLORING

Graph coloring is nothing but a simple way of labeling graph components such as

vertices , edges and regions under some constraints. In a graph, no two adjacent vertices,

adjacent edges , or adjacent regions are colored with minimum number of colors .This

number is called the chromatic number and the graph is called properly colored graph.

In graph theory coloring is a special case of graph labeling; it is

an assignment of labels traditionally called “colors” to elements of a graph subject to certain

constraints. In it simplest form, it is a way of coloring the vertices of a graph such that no

two adjacent vertices share the same color, it is called vertex coloring. Similarly, edge

coloring assigns a color to each edge so that no two adjacent edges share the common

color.

While graph coloring , the constraints that are set on the graph are

colors , order of coloring , the way of assigning color , etc. A coloring is given to a vertex or a

particular region . Thus, the vertices or regions having same colors form independent sets.

3
VERTEX COLORING
Vertex coloring is an assignment of colors to the vertices of a graph ‘G ’ such

that no two adjacent vertices have the same color .Simply put , no two vertices of an edge

should be of the same color.

The most common type of vertex coloring seeks to minimize the

number of colors for a given graph . Such a coloring is known as a minimum vertex coloring ,

and the minimum number of colors which with the vertices of a graph may be colored is

called the chromatic number .

CHROMATIC NUMBER:
The minimum number of colors required for vertex coloring of graph ‘ G ’

is called as the chromatic number of G , denoted by X (G) .

X(G) = 1 iff ‘ G ’ is a null graph. If ‘G ’ is not a null graph , then X(G) ≥ 2.

EXAMPLES;

1. 2.

Null Graph ( X (G) = 1 ) Not Null Graph ( X (G) = 2 )

4
EDGE COLORING
An edge coloring of a graph G is a coloring of the edges of G such that adjacent

edges ( or the edges bounding different regions ) receive different colors. An edge coloring

containing the smallest possible number of colors for a given graph is known as a minimum

edge coloring.

The edge chromatic number gives the minimum number of colours with which
graph’s edges can be colored.

CHROMATIC INDEX
The minimum number of colors required for proper edge coloring of graph is

called chromatic index.

A complete graph is the one in which each vertex is directly connected with all

other vertices with an edge. If the number of vertices of a complete graph is n, then the

chromatic index for an odd number of vertices will be n and the chromatic index for even

number of vertices will be n-1.

5
EXAMPLES;
1.

The given graph will require 3 unique colors so that no two incident edges have the

Same color. So its chromatic index will be 3.

2.

The given graph will require 2 unique colors so that no two incident edges have

the same color. So its chromatic index will be 2.

6
 CHAPTER 2

Chromatic Number

The chromatic number of a graph is the smallest number of colors needed to color the vertices
of so that no two adjacent vertices share the same color. That is the smallest value of possible
to obtain a k-coloring.

• Graph Coloring is a process of assigning colors to the vertices of a graph.

• It ensures that no two adjacent vertices of the graph are colored with the same color.
• Chromatic Number is the minimum number of colors required to properly color any graph.

Graph Coloring Algorithm

• There exists no efficient algorithm for coloring a graph with minimum number of colors.

However, a following greedy algorithm is known for finding the chromatic number of any given
graph.

Greedy Algorithm

Step-01:

Color first vertex with the first color.

7
Step-02:

Now, consider the remaining (V-1) vertices one by one and do the following-

• Color the currently picked vertex with the lowest numbered color if it has not been used to
color any of its adjacent vertices.
• If it has been used, then choose the next least numbered color.
• If all the previously used colors have been used, then assign a new color to the currently
picked vertex.

Problems Based On Finding Chromatic Number of a Graph

Problem-01:

Find chromatic number of the following graph-

8
Solution-

Applying Greedy Algorithm, we have

Vertex a b C d e f

Color C1 C2 C1 C2 C1 C2

From here,

• Minimum numbers of colors used to color the given graph are 2.

• Therefore, Chromatic Number of the given graph = 2.

The given graph may be properly colored using 2 colors as shown below-

9
Problem-02:

Find chromatic number of the following graph-

Solution-

Applying Greedy Algorithm, we have-

Vertex a b C d e f

Color C1 C2 C2 C3 C3 C1

From here,

• Minimum numbers of colors used to color the given graph are 3.

• Therefore, Chromatic Number of the given graph = 3.

10
The given graph may be properly colored using 3 colors as shown below-

Chromatic Number of Graphs

Chromatic Number of some common types of graphs are as follows-

1. Cycle Graph-

• A simple graph of ‘n’ vertices (n>=3) and ‘n’ edges forming a cycle of length ‘n’ is called as a
cycle graph.
• In a cycle graph, all the vertices are of degree 2.

Chromatic Number
• If number of vertices in cycle graph is even, then its chromatic number = 2.
• If number of vertices in cycle graph is odd, then its chromatic number = 3.

11
Examples-

2. Planar Graphs-

A planar graph is a graph that can be embedded in the plane, that is it can be drawn on the
plane in such a way that its edges intersect only at their endpoint. In other words, it can be
drawn in such a way that no edges cross each other.

12
A Planar Graph is a graph that can be drawn in a plane such that none of its edges cross each
other.

Chromatic Number
Chromatic Number of any Planar Graph is less than or equal to 4

Examples-
+

• All the above cycle graphs are also planar graphs.

• Chromatic number of each graph is less than or equal to 4.

13
3. Complete Graphs-

• A complete graph is a graph in which every two distinct vertices are joined by exactly one
edge.
• In a complete graph, each vertex is connected with every other vertex.
• So to properly it, as many different colors are needed as there are number of vertices in the
given graph.

Chromatic Number
Chromatic Number of any Complete Graph

= Number of vertices in that Complete Graph

Examples-

14
4. Bipartite Graphs-

A bipartite graph is a graph whose vertices can be divided into two disjoint and independent
sets U and V such that every edge connects a vertex in U to one in V. Vertex sets U and V are
usually called the parts of the graph.

• A Bipartite Graph consists of two sets of vertices X and Y.

• The edges only join vertices in X to vertices in Y, not vertices within a set.

Chromatic Number
Chromatic Number of any Bipartite Graph

=2

Example-

15
5. Trees-

A tree is an undirected graph in which any two vertices are connected by exactly one path, or
equivalently a connected acyclic undirected graph.

• A Tree is a special type of connected graph in which there are no circuits.

• Every tree is a bipartite graph.
• So, chromatic number of a tree with any number of vertices = 2.

Chromatic Number
Chromatic Number of any tree

=2

Examples-

16
 CHAPTER-3
APPLICATIONS OF GRAPH COLORING

1) Making Schedule or Time Table:

Suppose we want to make an exam schedule for a university. We have list
different subjects and students enrolled in every subject. Many subjects would have common
students (of same batch, some backlog students, etc). How do we schedule the exam so that no
two exams with a common student are scheduled at same time? How many minimum time
slots are needed to schedule all exams? This problem can be represented as a graph where
every vertex is a subject and an edge between two vertices mean there is a common student.
So this is a graph coloring problem where minimum number of time slots is equal to the
chromatic number of the graph.

2) Mobile Radio Frequency Assignment:

When frequencies are assigned to towers, frequencies assigned to all towers at the
same location must be different. How to assign frequencies with this constraint? What is the
minimum number of frequencies needed? This problem is also an instance of graph coloring
problem where every tower represents a vertex and an edge between two towers represents
that they are in range of each other.

3) Register Allocation:
In compiler optimization, register allocation is the process of assigning a large number
of target program variables onto a small number of CPU registers. This problem is also a graph
coloring problem.

4) Sudoku:
Sudoku is also a variation of Graph coloring problem where every cell represents a
vertex. There is an edge between two vertices if they are in same row or same column or same
block.

17
 5) Map Coloring:

Geographical maps of countries or states where no two adjacent cities cannot be

assigned same color. Four colors are sufficient to color any map.

6) Bipartite Graphs:

We can check if a graph is bipartite or not by coloring the graph using two colors. If a
given graph is 2-colorable, then it is Bipartite, otherwise not. See this for more details.

Explanation;

Algorithm:
A bipartite graph is possible if it is possible to assign a color to each vertex such that no
two neighbour vertices are assigned the same color. Only two colors can be used in this
process.

Steps:
1. Assign a color (say red) to the source vertex.

2. Assign all the neighbours of the above vertex another color (say blue).

3. Taking one neighbour at a time, assign all the neighbour's neighbours the color red.

4. Continue in this manner till all the vertices have been assigned a color.

5. If at any stage, we find a neighbour which has been assigned the same color as that of the
current vertex, stop the process. The graph cannot be colored using two colors. Thus the graph
is not bipartite.

18
Example:

19
 CONCLUSION

This project aims to provide a solid background in the basic topics of graph coloring. Graph
coloring problem is to assign colors to certain elements of a graph subject to certain
constraints. The nature of coloring problem depends on the number of colors but not on what
they are.

The study of this topic gives excellent introduction to the subject called “Graph
Coloring”.

This project includes two important topics such as vertex coloring and edge coloring and came
to know about different ways and importance of coloring.

Graph coloring enjoys many practical applications as well as theoretical challenges.

Besides the applications, different limitations can also be set on the graph or on the away a color
is assigned or even on the color itself. It has been reached popularity with the general public in
the form of the popular number puzzle Sudoku and it is also use in the making of time
management which is an important application of coloring. So graph coloring is still a very
active field of research.

20
 BIBLIOGRAPHY

1) V.K Balakrishnan, outlines, graph theory.

2) J.A Bondy and U.S.R Murty, graph theory with applications.

3) S.Arumugam, S.Ramachandran, invitation to graph theory.

21
POWER SERIES SOLUTIONS AND
SPECIAL FUNCTIONS
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
CYRIL SAJI
DB18CMSR28
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Power Series’ is a bona fide project of

CYRIL SAJI carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I CYRIL SAJI hereby declare that the project ‘Power Series’ is an original
record of studies and bona fide project carried out by me during the period of
2018 – 2021 under the guidance of Ms. Athulya P, Department of Mathematics,
Don Bosco Arts and Science College, Angadikkadavu and has not submitted by
me elsewhere for the award of my degree, diploma, title or recognition, before.

Name
CYRIL SAJI

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express my deepest gratitude to
people along the way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my
project supervisor Ms. Athulya P, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and
the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every
one of them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- SERIES SOLUTION OF FIRST 4-9

ORDER EQUATION

04 Chapter 2- SECOND ORDER LINEAR 9-18

EQUATION, ORDINARY POINTS

05 Conclusion 19

06 Bibliography 20
 INTRODUCTION
A power series is a type of series with terms involving a variable. Power series
are often used by calculators and computers to evaluate trigonometric,
hyperbolic, exponential and logarithm functions. So any application of these
kind of functions is a possible application of power series. Many interesting and
important differential equations can be found in power series.

1
 PRELIMINERY

A. An infinite series of the form

∞

∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ (1)
𝑛=0

is called a power series in x. The series

∞

∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛 = 𝑎0 + 𝑎1 (𝑥 − 𝑥0 ) + 𝑎2 (𝑥 − 𝑥0 )2 + ⋯
𝑛=0

is a power series in x – x0.

B. The series (1) is said to converge at a point x if the limit

𝑚

𝑙𝑖𝑚 ∑ 𝑎𝑛 𝑥 𝑛
𝑚→∞
𝑛=0

exists, and in this case the sum of the series is the value of this limit.
Radius of convergence: Series in 𝑥 has a radius of convergence 𝑅, where
0 ≤ 𝑅 ≤ ∞, with the property that the series converges if |𝑥| < 𝑅 and
diverges if |𝑥| > 𝑅. It should be noted that if 𝑅 = 0 then no 𝑥 satisfies
|𝑥| < 𝑅, and if 𝑅 = ∞ then no 𝑥 satisfies |𝑥| > 𝑅
𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | | , if the limit exists.
𝑛→∞ 𝑎𝑛+1

C. Suppose that (1) converges for |𝑥| < 𝑅 with 𝑅 > 0, and denote its sum
by f(x):
∞

𝑓(𝑥) = ∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑛=0

Then f(x) is automatically continuous and has derivatives of all orders for
|𝑥| < 𝑅.

2
D. Let f(x) be a continuous function that has derivatives of all orders for
|x|< R with R > 0. f(x) be represented as power series using Taylor’s
formula:
𝑛
𝑓 (𝑘) (0) 𝑘
𝑓(𝑥) = ∑ 𝑥 + 𝑅𝑛 (𝑥)
𝑘!
𝑘=0

where the remainder Rn (x) is given by

𝑓 (𝑛+1) (𝑥̅ ) 𝑛+1
𝑅𝑛 (𝑥) = 𝑥
(𝑛 + 1)!
for some point 𝑥̅ between 0 and x.

E. A function f(x) with the property that a power series expansion of

the form

𝑓(𝑥) = ∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛
𝑛=0

is valid in some neighbourhood of the point x0 is said to be analytic at

x0. In this case the an are necessarily given by
𝑓 (𝑛) (𝑥0 )
𝑎𝑛 =
𝑛!
and is called the Taylor series of f(x) at x0.

Analytic functions: A function f defined on some open subset U of R or C is

called analytic if it is locally given by a convergent power series. This means
that every a ∈ U has an open neighbourhood V ⊆ U, such that there exists
a power series with centre a that converges to f(x) for every x ∈ V.

3
 CHAPTER 1
SERIES SOLUTION OF FIRST ORDER EQUATION
We have studied to solve linear equations with constants coefficient but with
variable coefficient only specific cases are discussed. Now we turn to these
latter cases and try to find a general method to solve this. The idea is to assume
that the unknown function y can be explained into a power series. Our purpose
in this section is to explain the procedures by showing how it works in the case
of first order equation that are easy to solve by elementary methods.

Example 1: we consider the equation

𝑦ʹ = 𝑦
Consider the above equation as (1). Assume that y has a power series solution
of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |x| < R, R > 0

That is we assume that 𝑦 ʹ = 𝑦 has a solution that is analytic at origin. We have
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ ⋯
then
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ … ….
∴ (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 ⋯
= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
⇒ 𝑎1 = 𝑎0
𝑎1 𝑎0
2𝑎2 = 𝑎1 ⇒ 𝑎2 = =
2 2

4
 𝑎2 𝑎0 𝑎0
3𝑎3 = 𝑎2 ⇒ 𝑎3 = = =
3 2 ∙ 3 3!
𝑎3 𝑎0 𝑎0
4𝑎4 = 𝑎3 ⇒ 𝑎4 = = =
4 2 ⋅ 3 ⋅ 4 4!
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0 𝑎0
= 𝑎0 + 𝑎0 𝑥 + 𝑥 2 + 𝑥 3 + 𝑥 4 + ⋯
2 3! 4!
𝑥2 𝑥3 𝑥4
= 𝑎0 (1 + 𝑥 + + + + ⋯ )
2! 3! 4!
𝑦 = 𝑎0 𝑒 𝑥
To find the actual function we have 𝑦 ʹ = 𝑦
𝑑𝑦 𝑑𝑦
i.e., =𝑦 ⇒ = 𝑑𝑥
𝑑𝑥 𝑦
integrating
log 𝑦 = 𝑥 + 𝑐
i.e., 𝑦 = 𝑒 𝑥+𝑐 = 𝑒 𝑥 ⋅ 𝑒 𝑐
𝑦 = 𝑎0 𝑒 𝑥 , where a0 = ec , a constant.

Example 2: solve 𝑦 ′ = 2𝑥𝑦. Also find its actual solution.

Solution: 𝑦 ′ = 2𝑥𝑦 (1)
Assume that y has a power series of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |𝑥| < 𝑅, 𝑅 > 0

∞

We have 𝑦 = ∑ a𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

5
 = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 2𝑥(𝑎0 + 𝑎1 𝑥 +𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯ )
= 2𝑥𝑎0 + 2𝑥𝑎1 𝑥 + 2𝑥𝑎2 𝑥 2 + 2𝑥𝑎3 𝑥 3 + ⋯
= 2𝑥𝑎0 + 2𝑎1 𝑥 2 + 2𝑎2 𝑥 3 + 2𝑎3 𝑥 4 + ⋯ … ..
2𝑎0
⇒ 𝑎1 = 0 2𝑎2 = 2𝑎0 ⇒ 𝑎2 = = 𝑎0
𝑧
2𝑎1
3. 𝑎3 = 2𝑎1 ⇒ 𝑎3 = =0
3
2𝑎2 𝑎0
4𝑎4 = 2𝑎2 ⇒ 𝑎4 = =
42 2

5𝑎5 = 2𝑎3 = 0 ⇒ 𝑎5 = 0
2𝑎4 𝑎4 𝑎0 𝑎0
6𝑎6 = 2𝑎4 ⇒ 𝑎6 = = = =
6 3 2⋅3 3!

We get,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0
= 𝑎0 + 0 + 𝑎0 𝑥 2 + 0𝑥 3 + 𝑥 4 + ⋯
2
𝑎 0
= 𝑎0 + 𝑎0 𝑥 2 + 𝑥 4 + ⋯
2
2
𝑥4 𝑥6
= 𝑎0 (1 + 𝑥 + + + ⋯ )
2! 3!
2
𝑦 = 𝑎0 𝑒 𝑥
To find an actual solution

𝑦 ′ = 2𝑥𝑦
𝑑𝑦
= 2𝑥𝑦
𝑑𝑥
𝑑𝑦
⇒ = 2𝑥 ⋅ 𝑑𝑥
𝑦
log 𝑦 = 𝑥 2 + 𝑐
2
𝑦 = 𝑒𝑥 + 𝑐
2
⇒ 𝑦 = 𝑎0 𝑒 𝑥 , where 𝑎0 = 𝑒 𝑐

6
Example 3: Consider 𝑦 = (1 + 𝑥)𝑝 where p is an arbitrary constant. Construct a
differential equation from this and then find the solution using power series
method.
Solution
First, we construct a differential equation
i.e. 𝑦 = (1 + 𝑥)𝑝
𝑝(1+𝑥)𝑝 𝑝𝑦
𝑦 ′ = 𝑝(1 + 𝑥)𝑝−1 = =
1+𝑥 1+𝑥

∴ (1 + 𝑥)𝑦 ′ = 𝑝𝑦, 𝑦(0) = 𝑟

Assume that y has a power series solution of the form,
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎̇ 2 𝑥 2 + ⋯ … …
Which converges for |𝑥| < 𝑅̇, 𝑅 > 0
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ … … ..
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1 + 𝑥)𝑦 ′ = 𝑝𝑦
⇒ (1 + 𝑥)𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 𝑝(𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
⇒ (𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) + (𝑎1 𝑥 + 2𝑎2 𝑥 2 + 3𝑎3 𝑥 3 + ⋯ )
= 𝑎0 𝑝 + 𝑎1 𝑝𝑥 + 𝑎2 𝑝𝑥 2 + ⋯
Equating the coefficients of 𝑥, 𝑥 2 , …
𝑎1 = 𝑎0 𝑝 i.e. 𝑎1 = 𝑝, (since 𝑎0 = 1)
⇒ 2𝑎2 = 𝑎1 (p − 1)
𝑎1 (p − 1) 𝑎0 𝑃(𝑝 − 1)
𝑎2 = =
2 2

7
 3𝑎3 + 2𝑎2 = 𝑎2 𝑝
𝑠𝑎3 = 𝑎2 𝑝 − 2𝑎2
= 𝑎2 (𝑝 − 2)
𝑎2 (𝑝 − 2) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)
𝑎3 = =
3 2⋅3
4𝑎4 + 3𝑎3 = 𝑎3 𝑝
4𝑎4 = 𝑎3 𝑝 − 3𝑎3
= 𝑎3 (𝑝 − 3)
𝑎3 (𝑝 − 3) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3)
𝑎4 = =
4 2⋅3⋅4
∴ we get,
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0 𝑝(𝑝 − 1) 2 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 𝑎0 + 𝑎0 𝑝𝑥 + 𝑥 + 𝑥 + ⋯…
2 2⋅3
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 1 + 𝑝𝑥 + 𝑥 + 𝑥 +
2! 3!
𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3) 4 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − (𝑛 − 1)) 𝑛
𝑥 + ⋯+ 𝑥
4! 𝑛!
Since the initial problem y(0) = 1 has one solution the series converges for |x|<1
So this is a power solution,
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1) ⋯ (𝑝 − (𝑛 − 1)) 𝑛
(1 + 𝑥)𝑝 = 1 + 𝑝𝑥 + 𝑥 + ⋯+ 𝑥
2! 𝑛!
Which is binomial series.

Example 4: Solve the equation 𝑦 ′ = 𝑥 − 𝑦, 𝑦 (0) = 0

Solution: Assume that y has a power series solution of the form
∞

𝑦 = ∑ an 𝑥 𝑛
𝑛=0

which converges for |𝑥| < 𝑅, 𝑅 > 0

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯

8
 Now 𝑦 ′ = 𝑥 − 𝑦
(𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) = 𝑥 − (𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
Equating the coefficients of 𝑥, 𝑥 2 ,
𝑎1 = −𝑎0 = 0, Since 𝑦(0) = 0
2𝑎2 = 1 − 𝑎1
=1−0
1
⇒ 𝑎2 =
2
3𝑎3 = −𝑎2
−𝑎2 1
𝑎3 = =−
3 2⋅3
4𝑎4 = −𝑎3
1
⇒ 𝑎4 =
2⋅3⋅4
𝑥2 𝑥3 𝑥4
∴ 𝑦 = 0 + 0 + − + − ⋯……
2! 3! 4!
𝑥2 𝑥3
= (1 − 𝑥 + − + ⋯ ) + 𝑥 − 1
2! 3!
= 𝑒 −𝑥 + 𝑥 − 1
By direct method
𝑦′ = 𝑥 − 𝑦 𝑑𝑦
𝑑𝑦 𝑑𝑦 ( + 𝑝𝑦 = 𝑄 𝑓𝑜𝑟𝑚)
=𝑥−𝑦 ⇒ +𝑦 =𝑥 𝑑𝑥
𝑑𝑥 𝑑𝑥
here 𝑃(𝑥) = 1, integrating factor
= 𝑒 ∫ 𝑝(𝑥)⋅𝑑𝑥
= 𝑒𝑥
∴ 𝑦𝑒 𝑥 = ∫ 𝑥𝑒 𝑥 ⋅ 𝑑𝑥
𝑦𝑒 𝑥 = 𝑥 ⋅ 𝑒 𝑥 − ∫ 𝑒 𝑥 ⋅ 𝑑𝑥
= 𝑥𝑒 𝑥 − 𝑒 𝑥
𝑦𝑒 𝑥 = 𝑒 𝑥 (𝑥 − 1) + 𝑐
𝑒 𝑥 (𝑥 − 1) + 𝑐 𝑐
𝑦= = 𝑥 − 1 + 𝑥 = 𝑐𝑒 −𝑥 + (𝑥 − 1)
𝑑𝑥 𝑒
∴ 𝑦 = (𝑥 − 1) + 𝑐𝑒 −𝑥

9
 CHAPTER 2
SECOND ORDER LINEAR EQUATION, ORDINARY POINTS

Consider the general homogeneous second order linear equation,

𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0 (1)
As we know, it is occasionally possible to solve such an equation in terms of
familiar elementary functions. This is true, for instance, when P(x) and Q(x)
are constants, and in a few other cases as well. For the most part, however,
the equations of this type having the greatest significance in both pure and
applied mathematics are beyond the reach of elementary methods, and can
only be solved by means of power series.
P(x) and Q(x) are called coefficients of the equation. The behaviour of its
solutions near a point x0 depends on the behaviour of its coefficient functions
P(x) and Q(x) near this point. we confine ourselves to the case in which P(x) and
Q(x) are well behaved in the sense of being analytic at x0, which means that
each has a power series expansion valid in some neighbourhood of this point. In
this case x0 is called an ordinary point of equation (1). Any point that is not an
ordinary point of (1) is called a singular point.

Consider the equation,

𝑦 ′′ + 𝑦 = 0 (2)

the coefficient functions are P(x) = 0 and Q(x) = 1, These functions are analytic
at all points, so we seek a solution of the form,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ + 𝑎𝑛 𝑥 𝑛 + ⋯ (3)

Differentiating (3) we get,

𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ + (𝑛 + 1)𝑎𝑛+1 𝑥 𝑛 + ⋯ (4)

10
And

𝑦 ′′ = 2𝑎2 + 2 ⋅ 3𝑎3 𝑥 + 3 ⋅ 4𝑎4 𝑥 2 + ⋯ + (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ⋯ (5)

If we substitute (5) and (3) into (2) and add the two series term by term, we get

(2𝑎2 + 𝑎0 ) + (2 ⋅ 3𝑎3 + 𝑎1 )𝑥 + (3 ⋅ 4𝑎4 + 𝑎2 )𝑥 2 +

𝑦 ′′ + 𝑦 = =0
(4 ⋅ 5𝑎5 + 𝑎3 )𝑥 3 + ⋯ + [(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 ]𝑥 𝑛 + ⋯
and equating to zero the coefficients of successive powers of x gives
2𝑎2 + 𝑎0 = 0, 2 ⋅ 3𝑎3 + 𝑎1 = 0, 3 ⋅ 4𝑎4 + 𝑎2 = 0
4 ⋅ 5𝑎5 + 𝑎3 = 0, … … , (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 = 0, …
By means of these equations we can express an in terms of a0 or a0, according
as n is even or odd:
𝑎0 𝑎1 𝑎2 𝑎0
𝑎2 = − , 𝑎3 = − , 𝑎4 = − =
2 2⋅3 3⋅4 2⋅3⋅4
𝑎3 𝑎1
𝑎5 = − = ,⋯
4⋅5 2⋅3⋅4⋅5
With these coefficients, (3) becomes
𝑎0 𝑎1 3 𝑎0 𝑎1
𝑦 = 𝑎0 + 𝑎1 𝑥 − 𝑥 2 − 𝑥 + 𝑥4 + 𝑥5 − ⋯
2 2⋅3 2⋅3⋅4 2⋅3⋅4⋅5
𝑥2 𝑥4 𝑥3 𝑥5
= 𝑎0 (1 − + − ⋯ ) + 𝑎1 (𝑥 − + − ⋯ ) (6)
2! 4! 3! 5!
𝑖. 𝑒, 𝑦 = 𝑎0 cos 𝑥 + 𝑎1 sin 𝑥
Since each of the series in the parenthesis converges for all x. This implies the
series (2) for all x.

Solve the legenders equation,

(1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0
Solution
Consider (1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
11
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2

put 𝑛 = 𝑛 + 2 (Since 𝑦 ′′ is not 𝑥 𝑛 form )

∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛+2 𝑥 𝑛+2−2

𝑛=0
∞

∴ 𝑦 ′′ = ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

Now (1) ⇒ 𝑦 ′′ − 𝑥 2 𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0

⇒ ∑(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 − ∑𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛 − ∑2𝑛𝑎𝑛 𝑥 𝑛 + ∑𝑝(𝑝 + 1)𝑎𝑛 𝑥 𝑛 = 0
∞

⇒ ∑ [((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 )𝑥 𝑛 ] = 0

𝑛=0

for n = 0,1,2,3…….
⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 = 0
[𝑛(𝑛 − 1) + 2𝑛 − 𝑝(𝑝 + 1)]
⇒ 𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(𝑛2 − 𝑛 + 2𝑛 − 𝑝2 − 𝑝)𝑎𝑛
=
(𝑛 + 1)(𝑛 + 2)
(𝑛 + 𝑛 − 𝑝2 − 𝑝)𝑎𝑛
2
=
(𝑛 + 1)(𝑛 + 2)

12
 −(𝑝 − 𝑛)(𝑝 + 𝑛 + 1)
∴ 𝑎𝑛+2 = 𝑎𝑛 , 𝑛 = 0,1,2 …
(𝑛 + 1)(𝑛 + 2)

This is an Recursion formula

−𝑝(𝑝 + 1)
put 𝑛 = 0, 𝑎2 = 𝑎0
1⋅2
−(𝑝 − 1)(𝑝 + 2)
𝑛 = 1, 𝑎3 = ⋅ 𝑎1
2⋅3
−(𝑝 − 2)(𝑝 + 3)
𝑛 = 2, 𝑎4 = 𝑎2
3𝑖4
𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3)
= 𝑎0
4!
−(𝑝 − 3)[𝑝 + 4)
𝑛 = 3, 𝑎5 = 𝑎3
4⋅5
(𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4)
= 𝑎1
5!
−(𝑝 − 4)(𝑝 + 5)
𝑛 = 4, 𝑎6 = 𝑎4
5⋅6
−𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5)
= 𝑎0
6!
(𝑝 − 5)(𝑝 + 6)
𝑛 = 5, 𝑎7 = − 𝑎5
6⋅7
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6)
=− 𝑎1
7!

𝑝(𝑝 + 1) 2 𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3) 4

𝑦 = 𝑎0 [1 − 𝑥 + 𝑥
2! 4!
𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5) 6
− 𝑥 + ⋯]
6!
(𝑝 − 1)(𝑝 + 2) 3 (𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4) 5
+𝑎1 [𝑥 − 𝑥 + 𝑥
3! 5!
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6) 7
− 𝑥 + ⋯]
7!

13
Find the general solution of (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 in terms of power
series in 𝑥. Can you express this solution by means of elementary functions?
Solution
Consider the equation (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

(1 + 𝑥 2 )𝑦 ′′ = 𝑦 ′′ + 𝑥 2 𝑦 ′′
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2
∞

Now 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

put 𝑛 = 𝑛 + 2
∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛 + 2𝑥 𝑛+2=2

𝑛=0
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

14
 ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥𝑛

𝑛=0 𝑛=2
∞ ∞

+ ∑ 2𝑛𝑎𝑛 𝑥 𝑛 − ∑ 2𝑎𝑛 𝑥 𝑛 = 0
𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 = 0

[−𝑛(𝑛 − 1) − 2𝑛 + 2]
𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(−𝑛2 + 𝑛 − 2𝑛 + 2)
= 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)

2 2𝑎0
put 𝑛 = 0, 𝑎2 = 𝑎0 = = 𝑎0
1⋅2 2!
(1 − 1 − 2 + 2)
𝑛 = 1, 𝑎3 = 𝑎1 = 0
2⋅3
2−4−4+2 −4 −𝑎0
𝑛 = 2, 𝑎4 = 𝑎2 = 𝑎0 =
3⋅4 3⋅4 3
3 − 9 − 16 + 2
𝑛 = 3, 𝑎5 = 𝑎3 = 0
4.5
4 − 16 − 8 + 2 −3 3𝑎0 𝑎0
𝑛 = 4, 𝑎6 = 𝑎4 = 𝑎4 = =
5.6 5 3.5 5

∴ 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0
= 𝑎0 + 𝑎1 𝑥 + 𝑎0 𝑥 2 − 𝑥 4 + 𝑥 6 … .
3 5
4 6
𝑥 𝑥
= 𝑎0 [1 + 𝑥 2 − + − ⋯ ] + 𝑎1 𝑥
3 5
𝑥3 𝑥5
= 𝑎0 [1 + 𝑥 (𝑥 − + ⋯ )] + 𝑎1 𝑥
3 5
= 𝑎0 (1 + 𝑥tan−1 𝑥) + 𝑎1 𝑥

15
Consider the equation 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0
(a) Find its general solution 𝑦 = ∑𝑎𝑛 𝑥 𝑛 in the form
𝑦 = 𝑎0 𝑦1 (𝑥) + 𝑎1 𝑦2 (𝑥) where 𝑦1 (𝑥) and 𝑦2 (𝑥) are power series
(b) use the ratio test to verify that the two series 𝑦1 (𝑥) and 𝑦2 (𝑥) converges
. for all x.
Solution:
Given 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ (1)
Assume that y has a power series solution the form ∑a𝑛 𝑥 𝑛 which converges
for |𝑥| 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛 ⋅ 𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0
∞

𝑥𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞ ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)a𝑛+2 𝑥 + ∑ 𝑛𝑎𝑛 𝑥 + ∑ 𝑎𝑛 𝑥 𝑛 = 0

𝑛 𝑛

𝑛=0 𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 = 0

(−𝑛 − 1)𝑎𝑛 −𝑎𝑛
⇒ 𝑎𝑛+2 = =
(𝑛 + 1)(𝑛 + 2) 𝑛 + 2
𝑎0
put 𝑛 = 0, 𝑎2 = −
2
−2𝑎1 −𝑎1
𝑛 = 1, 𝑎3 = =
2⋅3 3
16
 −3𝑎2 −𝑎2 𝑎0
𝑛 = 2, 𝑎4 = = =
3⋅4 4 8
−4𝑎3 𝑎1
𝑛 = 3, 𝑎5 = =
4⋅5 15
−5𝑎4 −𝑎0
𝑛 = 4, 𝑎6 = =
5⋅6 48
𝑎0 𝑎1 𝑎0 𝑎1 𝑎0
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + − 𝑥 2 − 𝑥 3 + 𝑥 4 + 𝑥 5 − 𝑥 6 + ⋯
2 3 8 15 48

𝑥2 𝑥4 𝑥6 𝑥3 𝑥5
= 𝑎0 [1 − + − + ⋯ ] + 𝑎1 [𝑥 − + + ⋯]
2 2⋅4 2⋅4⋅6 3 3.5

𝑥2 𝑥4 𝑥̇ 2
𝑤ℎ𝑒𝑟𝑒 𝑦1 (𝑥) = 1 − + − +
2 2⋅4 2⋅4⋅6

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + +⋯
3 3⋅5

𝑥2 𝑥4 𝑥6
(b) 𝑦1 (𝑥) = 1 − + − +⋯
2 2⋅4 2⋅4⋅6

𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | |
𝑛→∞ 𝑎𝑛+1
(−1)𝑛 (−1)𝑛+1
= 𝑙𝑖𝑚 | / |
𝑛→∞ 2 ⋅ 4 ⋅ (2𝑛) 2 ⋅ 4 ⋅⋅ (2𝑛 + 2)

2(𝑛 + 1)
= 𝑙𝑖𝑚 | |
𝑛→∞ −1
1
= 𝑙𝑖𝑚 | − 2𝑛(1 + )| = ∞
𝑛→∞ 𝑛

∴ 𝑦1 (𝑥) converges for all 𝑥

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + −⋯
3 3⋅5

17
 (−1)𝑛 (−1)𝑛+1
𝑅 = 𝑙𝑖𝑚 | ⁄ |
𝑛→∞ 3 ⋅ 5 ⋯ (2𝑛 + 1) 3 ⋅ 5 ⋅ ⋯ (2𝑛 + 3)

(−1) ⋅ 3 ⋅ 5 ⋯ (2𝑛 + 1)(2𝑛 + 3)

= 𝑙𝑖𝑚 | |
𝑛→∞ 3 ⋅ 5 ⋯ ⋅ (2𝑛 + 1)
= 𝑙𝑖𝑚 |(−1)𝑛(2 + 3/𝑛)| = ∞
𝑛→∞

∴ 𝑦2 (𝑥) converges for all 𝑥

REGULAR SINGULAR POINTS

A singular point 𝑥0 of equation
𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0
is said to be regular if the functions (𝑥 − 𝑥0 )𝑃(𝑥) and (𝑥 − 𝑥0 )2 𝑄(𝑥) are
analytic, and irregular otherwise. Roughly speaking, this means that the
singularity in 𝑃(𝑥) cannot be worse than 1/(𝑥 − 𝑥0 ), and that in 𝑄(𝑥) cannot
be worse than 1/(𝑥 − 𝑥0 )2 .
If we consider Legendre’s equation in the form
2𝑥 𝑝(𝑝 + 1)
𝑦 ′′ − 𝑦 ′
+ 𝑦=0
1 − 𝑥2 1 − 𝑥2
it is clear that x = 1 and x = −1 are singular points. The first is regular because
2𝑥 (𝑥 − 1)𝑝(𝑝 + 1)
(𝑥 − 1)𝑃(𝑥) = and (𝑥 − 1)2 𝑄(𝑥) = −
𝑥+1 𝑥+1
are analytic at x = 1, and the second is also regular for similar reasons.
Example: Bessel’s equation of order p, where p is a nonnegative constant:
𝑥 2 𝑦 ′′ + 𝑥𝑦 ′ + (𝑥 2 − 𝑝2 )𝑦 = 0
If this is written in the form
1 ′ 𝑥 2 − 𝑝2
′′
𝑦 + 𝑦 + 𝑦 = 0,
𝑥 𝑥2
it is apparent that the origin is a regular singular point because𝑥𝑃(𝑥) = 1 and
𝑥 2 𝑄(𝑥) = 𝑥 2 − 𝑝2 are analytic at x = 0.

18
 CONCLUSION
The purpose of this project gives a simple account of series solution of first
order equation, second order linear equation, ordinary points. The study of these
topics given excellent introduction to the subject called ‘POWER SERIES’
we used application of power series extensively throughout this project. We
take it for granted that most readers are reasonably well acquainted with these
series from an earlier course in calculus. Nevertheless, for the benefit of those
whose familiarity with this topic may have faded slightly, we presented a brief
review of the main facts of power series.

19
 BIBLIOGRAPHY
George F. Simmons - Differential Equations With Applications and Historical
. Notes
G.Birkoff and G.C Rota - Ordinary Differential Equations; Wiley and Sons; 3rd
. Edition ( 1978 )
P.Hartmon - Ordinary Differential Equations; John Wiley and Sons

20
NORMED LINEAR SPACES
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
DELNA JOMY
DB18CMSR22
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE
It is to certify that this project report ‘Normed Linear Spaces’ is the bonafide project of

Delna Jomy carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head Of Department Supervisor

Department Of Mathematics
Don Bosco Arts And Science College
Angadikadavu

2
 DECLARATION
I Delna Jomy hereby declare that the project ‘Normed Linear Space’ is an original record of
studies and bona fide project carried out by me during the period of 2018 – 2021 under the
guidance of Ms. Athulya P , Department of Mathematics, Don Bosco Arts and Science College,
Angadikadavu and has not submitted by me elsewhere for the award of my degree, diploma, title
or recognition, before.

Delna Jomy

DB18CMSR22

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu

3
 ACKNOWLEDGEMENT

First and foremost, praises and thanks to God, the Almighty, for His showers of blessings
throughout my work to complete the project successfully. I had finally managed to finish up
this project with great enthusiasm and determination. All the time spent to search and discuss
ideas as well as justifying theoretical clues to drive the output were worth my effort and time.

Therefore, I would like to express my sincere gratitude to my supervisor Ms. Athulya P,

Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu, for providing
invaluable guidance, comments and suggestions throughout the course of the project. I also
acknowledge with a deep sense of reverence, my gratitude towards all the faculty members of
the Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I owe and respectfully offer my thanks to the principal and staff of Don Bosco Arts and
Science College, Angadikadavu for their constant moral support and mellifluous affection
provided to me.

I express my gratitude to all who directly or indirectly helped me to complete this project
successfully. Their guidance and support was very helpful in bringing this work to
conclusion.

4
 CONTENTS

SI no Title Page no

1 Introduction 6

2 Preliminaries 7-9

3 Chapter 1 10 - 18

4 Chapter 2 19 - 25

5 Conclusion 26

6 Bibliography 27

5
 INTRODUCTION

This chapter gives an introduction to the theory of normed linear spaces. A skeptical reader may
wonder why this topic in pure mathematics is useful in applied mathematics. The reason is quite
simple: Many problems of applied mathematics can be formulated as a search for a certain
function, such as the function that solves a given differential equation. Usually the function
sought must belong to a definite family of acceptable functions that share some useful properties.
For example, perhaps it must possess two continuous derivatives. The families that arise
naturally in formulating problems are often linear spaces. This means that any linear combination
of functions in the family will be another member of the family. It is common, in addition, that
there is an appropriate means of measuring the “distance” between two functions in the family.
This concept comes into play when the exact solution to a problem is inaccessible, while
approximate solutions can be computed. We often measure how far apart the exact and
approximate solutions are by using a norm. In this process we are led to a normed linear space,
presumably one appropriate to the problem at hand. Some normed linear spaces occur over and
over again in applied mathematics, and these, at least, should be familiar to the practitioner.
Examples are the space of continuous functions on a given domain and the space of functions
whose squares have a finite integral on a given domain.

6
 PRELIMINARIES

1) LINEAR SPACES
We introduce an algebraic structure on a set 𝑋and study functions on X which are well behaved
with respect to this structure. From now onwards , K will denote either R , the set of all real
numbers or C, the set of all complex numbers. For k ∈ C , Re k and Im k will denote the real and
imaginary part of k.

A linear space(or a vector space) over K is a non-empty set 𝑋 along with a function
+ : 𝑋 × 𝑋 → 𝑋, called addition and a function ·: K× 𝑋 → 𝑋 called scalar multiplication, such
that for all 𝑥 , 𝑦 , 𝑧 ∈ 𝑋and 𝑘 , 𝑙 ∈ K , we have

𝑥+ 𝑦 = 𝑦 + 𝑥

𝑥 + (𝑦 + 𝑧) = (𝑥 + 𝑦) + 𝑧

∃0∈𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + 0 = 𝑥,

∃ − 𝑥 ∈ 𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + (− 𝑥) = 0 ,

𝑘 · (𝑥 + 𝑦) = 𝑘 · 𝑥 + 𝑘 · 𝑦,

(𝑘 + 𝑙)⋅𝑥 = 𝑘 · 𝑥 + 𝑙 · 𝑥,

(𝑘𝑙)⋅𝑥 = 𝑘 · (𝑙 · 𝑥),

1⋅𝑥 = 𝑥.

We shall write 𝑘𝑥 in place of 𝑘 · 𝑥. We shall also adopt the following notations. For
𝑥, 𝑦 ∈ 𝑋, 𝑘 ∈ K and subsets 𝐸, 𝐹of 𝑋 ,

𝑥 + 𝐹 = {𝑥 + 𝑦: 𝑦 ∈ 𝐹},

𝐸 + 𝐹 = {𝑥 + 𝑦: 𝑥 ∈ 𝐸, 𝑦 ∈ 𝐹},`

𝑘𝐸 = {𝑘𝑥: 𝑥 ∈ 𝐸}.

2) BASIS

A nonempty subset 𝐸 of 𝑋 is said to be a subspace of 𝑋 if 𝑘𝑥 + 𝑙𝑦 ∈ 𝐸 whenever 𝑥, 𝑦 ∈ 𝐸 and

𝑘, 𝑙 ∈ K . If ∅≠𝐸 ⊂ 𝑋, then the smallest subspace of 𝑋 containing 𝐸 is

7
 {
𝑠𝑝𝑎𝑛⁡𝐸 = 𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛: 𝑥1, …, 𝑥𝑛 ∈ 𝐸 , 𝑘1, …, 𝑘𝑛 ∈ 𝐾 }
It is called the span of 𝐸. If span 𝐸 = 𝑋, we say that 𝐸 spans 𝑋. A subset 𝐸 of 𝑋 is said to be
linearly independent if for all 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈K , the equation
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0implies that 𝑘1 = ⋯ = 𝑘𝑛 = 0. It is called linearly dependent if it is not
linearly independent, that is, if there exist 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈ K such that
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0, where at least one 𝑘𝑗is nonzero.

A subset E of X is called a Hamel basis or simply basis for X if span of E = X and E is

linearly independent .

3) DIMENSION

If a linear space 𝑋 has a basis consisting of a finite number of elements , then X is called finite
dimensional and the number of elements in a basis for 𝑋 is called the dimension of 𝑋, denoted as
dimX . Every basis for a finite dimensional linear space has the same (finite) number of elements
and hence the dimension is well-defined. The space {0} is said to have zero dimension. Note that
it has no basis !

If a linear space contains an infinite linearly independent subset, then it is said to be infinite
dimensional.

4)METRIC SPACE

We introduce a distance structure on a set 𝑋 and study functions on 𝑋 which are well-behaved
with respect to this structure.

A metric 𝑑 on a nonempty set 𝑋 is a function 𝑑: 𝑋 × 𝑋 →R

such that for all 𝑥, 𝑦, 𝑧 ∈ 𝑋

d(x, y) ≥0 and d(x , y) = 0 iff x=y

d(y , x) = d(x , y)

d(x , y) ≤ d(x , z) + d(z , y) .

The last condition is known as the triangle inequality. A metric space is a nonempty set 𝑋 along
with a metric on it.

8
5)CONTINUOUS FUNCTIONS

Roughly speaking, a function from a metric space to a metric space is continuous if it sends
‘nearby’ points to ‘nearby’ points. If 𝑋 and 𝑌 are metric spaces with metrics 𝑑 and 𝑒 respectively,
then a function 𝐹: 𝑋 → 𝑌 is said to be continuous at 𝑥0 ∈ 𝑋 if for every ϵ⟩ 0 , there is some
(
δ > 0 (possibly depending on ϵ and 𝑥0 ) such that 𝑒 𝐹(𝑥), 𝐹 𝑥0 ( )) < ϵ for all 𝑥 ∈ 𝑋 satisfying
( )
𝑑 𝑥, 𝑥0 < δ. Further, 𝐹 is said to be continuous on 𝑋 if it is continuous at every point of 𝑋. It is
easy to see that 𝐹 is continuous on 𝑋 if and only if the set F -1(E) is open in X whenever the set E
is open inY. Also , this happens iff F(xn) →F(x) in Y whenever xn →x in X.

6) UNIFORM CONTINUITY

We note that a continuous function 𝐹: 𝑇 → 𝑆 is, in fact, uniformly

continuous, that is, for every ϵ > 0, there exists some δ > 0
such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < ϵ whenever 𝑑 𝑡, 𝑢( )< δ. This can be seen as follows. Let 𝑡 ∈ 𝑇. By
ε
the continuity of 𝐹 at 𝑡 ∈ 𝑇, there is some δ𝑡, such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < 2 whenever
𝑑(𝑡, 𝑢) < δ𝑡.

7) FIELD

A ring is a set R together with two binary operations + and ·( which we call addition and
multiplication ) such that the following axioms are satisfied .

➢ R is an abelian group with respect to addition

➢ Multiplication is associative
➢ ∀𝑎 , 𝑏, 𝑐 ∈ 𝑅the left distributive law a(b + c) = (a·b) + (a·c) and the right distributive
law (a + b)c = (a·c) + (b·c) , hold .

A field is a commutative division ring

9
 CHAPTER 1

NORMED LINEAR SPACE

Let X be a linear space over K . A norm on X is the function || || from 𝑋 to R such that ∀
x,y ∈ X and k ∈ K ,

||𝑥||≥0 and ||𝑥|| = 0 if and only if x = 0 ,

||x + y|| ≤||x|| + ||y|| ,

||kx|| =|k| ||x|| .

A norm is the formalization and generalization to real vector spaces of the intuitive
notion of “ length” in the real world .

A normed space is a linear space with norm on it .

For x and y in X , let

d(x,y) = ||x - y||

Then d is a metric on X so that (X,d) is a metric space , thus every normed space is a metric
space

➢ Every normed linear space is a metric space . But converse may not be true .

Example :
|𝑥 − 𝑦|
d(x,y) = , ∀ x , y ∈X
1 + |𝑥 − 𝑦|

|𝑥 − 𝑦|
⇒ ||x - y|| =
1 + |𝑥 − 𝑦|

|𝑧|
⇒ ||z|| = , z = x - y ∈X
1 + |𝑧|

10
 |α𝑧|
||αz|| =
1 +|α𝑧|

|α| |𝑧|
=
1 + |α| |𝑧|

= |α| ( |𝑧|
1 + |α| |𝑧| )
≠ |α| ||z|| .

➢ Result

Let X be a normed linear space . Then ,

| ||x|| - ||y|| | ≤ ||x - y|| , ∀x , y ∈ X

Proof :

||x|| = || ( x - y ) + y|| ≤ ||x - y|| + ||y||

⇒||x|| - ||y|| ≤||x - y|| → (1)

x↔y

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||→(2)

From (1) and (2)

|||x|| - ||y||| ≤ ||x - y||

➢ Norm is a continuous function

Let xn →x , as n → ∞

11
 ⇒ xn - x →0 as n →∞

⇒ ||xn - x||→0 as n→∞

| ||xn|| - ||x|| | ≤ ||xn - x|| →0 as n→∞

⇒||xn|| - ||x|| →0 , as n →∞

⇒||x|| is continuous

➢ Norm is a uniformly continuous function

We have , || || :X→R . Let x,y ∈ X and ε> 0

Then ||x|| = ||x - y + y ||

≤||x - y|| + ||y||

⇒ ||x|| - ||y|| ≤ ||x - y||→ (1)

Interchanging x and y ,

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||

⇒ ||x|| - ||y|| ≥ - ||x - y|| → (2)

Combining (1) and (2)

- ||x - y|| ≤ ||x|| - ||y|| ≤ ||x - y||

That is ,

| ||x ||- ||y|| | ≤ ||x - y||

Take δ = ε , then whenever ||x - y|| < δ, | ||x|| -|| y|| |< ε

12
 Therefore || || is a uniformly continuous function .

➢ Continuity of addition and scalar multiplication

To show that + : X×X →X and ·: K ×X →X are continuous functions.

Let (x,y) ∈ X×X . To show that + is continuous at (x ,y) , that is , to show

that for each (x,y) ∈ X×X if xn→x and yn→y in X , then

+(xn , yn) →+(x , y) ;

That is ,
xn + yn →x + y .

Consider
||( xn + yn) - (x + y )|| = ||xn - x + yn - y||

≤||xn - x|| + ||yn - y||

Given xn→x and yn→y , for each ϵ > 0, ∃ N1 ∋

ε
||xn - x|| < 2
∀ n ≥N1 , and ∃ N2 ∋

ε
||yn - y|| < 2
∀ n ≥N2

\
Take N = max { N1, N2}

ε ε
Then ||xn - x|| < 2
and ||yn - y|| < 2
∀n≥N

ε ε
Therefore ||(xn + yn) - (x + y)|| ≤ 2
+ 2
= ε ∀n ≥N

That is , xn + yn →x + y

Now to show that ·: K ×X→X is continuous

Let (k , x) ∈ K×X

13
 To show that if kn →k and xn →x , then knxn →kx

Since kn →k , ∀ ε > 0 ∃ N1 ∋ |kn - k| < ε ∀ n ≥ N1

2

ε
Since xn →x , ∀ ε > 0 ∃ N2 ∋ ||xn - x|| < ∀ n ≥ N2
2

Consider ||knxn - kx|| = ||knxn - kx + xnk - xnk ||

= ||xn (kn - k) + k(xn - x)||

≤ ||xn(kn - k)|| + ||k(xn - x)||

= ||xn|| |kn- k| + |k| ||xn - x||

ε ε
≤ ||xn|| 2 + |k| 2

∴ knxn→kx

➢ Examples of normed space

1) Spaces Kn (K = R or C)

For n = 1 , the absolute value of function | | is a norm on K , since ∀k ∈K

We have ,

||k|| = ||k · 1|| = |k| ||1|| , by definition .

But ||1|| is a positive scalar .

∴ ||k|| is a positive scalar multiple of the absolute value function .

∴ any norm on K is a positive scalar multiple of the absolute value

function

For n > 1 , let p≥ 1 be a real number

14
 Kn = { ( x(1) , x(2) , . . . , x(n) ) : x(i) ∈K , i = 1 , 2 , . . . , n }

For x ∈Kn , that is , x = ( x(1) , x(2) , . . . , x(n) ) , define

𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (|𝑥(1)| +. . . + |𝑥(𝑛)| )

Then || ||p is a norm on Kn

When p = 1 ,

Then , ||x||1 = |x(1)| + |x(2)| + . . . + |x(n)|

Since |x(i)| ≥0 ∀ i = 1 , 2 , . . . , n , ||x||1 ≥0

And ||x||1 = 0 ⇔ |x(1)| + . . . +|x(n)| = 0

⇔ |x(i)| = 0 ∀ i

⇔ x(i) = 0 ∀ i

⇔ x = ( x(1) , . . . , x(n) ) = 0

Now ||kx||1 = |kx(1)| + |kx(2)| + . . . + |kx(n)|

= |k| |x(1)| + . . . + |k| |x(n)|

= |k| ( |x(1)| + . . . + |x(n)| )

= |k| ||x||1

||x + y||1 = |(x + y)(1)| + . . . + |(x + y)(n)|

= |x(1) + y(1)| + . . . + |x(n) + y(n)|

≤|x(1)| + |y(1)| + . . . + |x(n)| + |y(n)|

= |x(1)| + . . . + |x(n)| + |y(1)| + . . . + |y(n)|

= ||x||1 + ||y||1

15
 Consider 1<p<∞

𝑝 𝑝 1/𝑝
Now , ||𝑥||𝑝 = ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

𝑝
Since |x(i)| ≥0 ∀i , we have ||x||p ≥0

𝑝 𝑝 1/𝑝
And ||𝑥||𝑝 = 0⇔( |𝑥(1)| +. . . + |𝑥(𝑛)| ) = 0

𝑝
⇔ |𝑥(𝑖)| = 0 ∀i

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 ∀i

⇔ x = ( x(1) , . . . , x(n) ) = 0 .
Now
𝑝 𝑝 1/𝑝
||𝑘𝑥||𝑝 = ( |𝑘𝑥(1)| +. . . + |𝑘𝑥(𝑛)| )

𝑝 𝑝 𝑝 𝑝 1/𝑝
= ( |𝑘| |𝑥(1)| +. . . + |𝑘| |𝑥(𝑛)| )

𝑝 𝑝 1/𝑝
= |𝑘| ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

= |𝑘| ||𝑥||𝑝 .

𝑝 𝑝 1/𝑝
||𝑥 + 𝑦||𝑝 = ( |𝑥(1) + 𝑦(1)| +. . . + |𝑥(𝑛) + 𝑦(𝑛)| )

We have by Minkowski’s inequality ,

1/𝑝 1/𝑝 1/𝑝

( ) ( ) ( )
𝑛 𝑛 𝑛
𝑝 𝑝 𝑝
∑ |𝑥(𝑖) + 𝑦(𝑖)| ≤ ∑ |𝑥(𝑖)| + ∑ |𝑦(𝑖)|
𝑖=1 𝑖=1 𝑖=1
Then

16
 𝑝 𝑝 1/𝑝 𝑝 𝑝 1/𝑝
(
||𝑥 + 𝑦||𝑝 ≤ |𝑥(1)| +. . . + |𝑥(𝑛)| ) (
+ |𝑦(1)| +. . . + |𝑦(𝑛)| )
= ||𝑥||𝑝 + ||𝑦||𝑝

Then , for 1≤p<∞ ,|| ||𝑝 is a norm on Kn

When p = ∞, define ||𝑥||∞ = 𝑚𝑎𝑥 { |𝑥(1)| , |𝑥(2)| ,. . . , |𝑥(𝑛)| }

Then it is a norm on Kn

||𝑥||𝑝 ≥ 0 since each values |x(i)|≥0

So that

max {|x(i)| , i=1, . . . , n} ≥ 0

||𝑥||∞ = 0 ⇔𝑚𝑎𝑥 { |𝑥(𝑖)| : 𝑖 = 1,. . . , 𝑛 }= 0

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 , ∀i

⇔x = 0

||𝑘𝑥||∞ = 𝑚𝑎𝑥 { |𝑘𝑥(1)| ,. . . , |𝑘𝑥(𝑛)| }

= max { |k| |x(1)| , . . . , |k| |x(n)|}

= |k| max {|x(1)| , . . . , |x(n)|}

= |k| ||x||∞

||x + y||∞ = max { |x(1) + y(1)| , . . . , |x(n) + y(n)| }

≤max { |x(1)| + |y(1)| , . . . , |x(n)| + |y(n)| }

17
 ≤max { |x(1)| , . . . , |x(n)| } + max { |y(1)| , . . . , |y(n)| }

= ||x||∞ + ||y||∞

2) Sequence space

∞
𝑝 𝑝 𝑝
Let 1≤p < ∞, 𝑙 = { x = ( x(1) , x(2) , . . . ) ; x(i) ∈K and ∑ |x(j)| <∞} , that is , 𝑙 is the
𝑗=1
𝑝
space of p-summable scalar sequences in K . For x = (x(1) , x(2) , . . . ) ∈ 𝑙 ,

let ||x||p = ( |x(1)|p + |x(2)|p + . . . )1/p . Then it is a norm on lp.

That is , || ||p is a function from lp to R .

If p = 1 , then l1 is a linear space and ||x||1 = ( |x(1)| + |x(2)| + . . . ) is a norm on l1

∞
Let p = ∞. Then 𝑙 is the linear space of all bounded scalar sequences . And ,

||x||∞ = sup { |x(j)| : j = 1, 2, 3, . . . }

∞
Then || ||∞ is a norm on 𝑙

18
 CHAPTER 2

THEOREMS ON NORMED SPACES

a) Let Y be a subspace of a normed space X , then Y and its closure 𝑌 are normed spaces with
the induced norm.
b) Let Y be a closed subspace of a normed space X , for x +Y in the quotient space X/Y, let
|||x +Y||| = inf { ||x+y|| : y ∈Y} . Then ||| ||| is a norm on X/Y , called the quotient norm.

A sequence (xn + Y) converges to x + Y in X/Y iff there is a sequence (yn) in Y , (xn+ yn)
converges to x in X.

c) Let || ||pbe a norm on the linear space Xp , j = 1,2,…. . Fix p such that 1 ≤p ≤∞

For x = (x(1) , x(2) , … , x(m)) that is the product space X =X1× X2 × …× Xm ,

1/𝑝
Let ||𝑥||𝑝 = ( ||𝑥(1)|| 𝑝
1
𝑝
+ ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚
𝑝
) , if 1≤ p <∞

||x||p = max { ||x(1)||1 , … , ||x(m)||m } , if p = ∞.

Then || ||p is a norm on X.

A sequence (xn) converges to x in X ⇔ (xn(j)) converges to x(j) in Xj ∀ j=1,2,…,m.

Proof:
a) Since X is a normed space, there is a norm on X to Y . Since Y is a subspace of X,

|| ||y: Y → R is a function. To show that || ||y is a norm on Y.

For y∈ Y , || y||Y = ||y|| , then

||y||Y ≥ 0 ( ∵||y||≥ 0 ) and ||y||Y = 0 ⇔ y = 0

||ky||Y = ||ky|| = |k| ||y|| = |k| ||y||y .

Let y1 , y2 ∈Y. Then ,

||𝑦1 + 𝑦2||𝑦 = ||𝑦1 + 𝑦2|| ≤ ||𝑦1|| + ||𝑦2|| = ||𝑦1||𝑦 + ||𝑦2||𝑦

Now the continuity of addition and scalar multiplication shows that 𝑌is a subspace of X, since if
xn →x and yn → y , xn , yn ∈ 𝑌 , then

xn + yn →x + y (by continuity of addition) and

19
 kxn →kx (by continuity of scalar Xn) .

Since 𝑌is closed , x + y ∈ 𝑌 and kx ∈ 𝑌 . Therefore 𝑌 ≤ X.

∴ norm on X induces a norm on Y and 𝑌

b) X/Y , the quotient space equals X/Y={ x + Y : x ∈X }.

|||x + y||| = inf { ||x + y|| : y ∈Y }

Claim: ||| ||| is a norm on X/Y , called quotient norm

• Let x ∈X ,

|||x + Y||| = inf { ||x + y|| : y ∈Y } ≥ 0.

∴ |||x + Y||| ≥0 .

If |||x + y||| = 0 ( 0 in X/Y is Y) , then there is a sequence (yn) in Y ∋

||x + yn || →0

⇒ x + yn →0

⇒ yn →-x

Since yn ∈Y and Y is closed

-x ∈Y ⇔ x ∈Y (∵ Y is a subspace)

⇔x + Y = Y , zero in X/Y.

• For k ∈K ,

|||k(x + Y)||| = |||kx + Y|||

= inf { ||k(x + y)|| : y ∈Y}

= inf { |k| ||x + y|| : y ∈Y}

= |k| inf { ||x + y|| : y ∈Y}

= |k| |||x + Y||| .

• Let x1 , x2 ∈X . Then

|||x1 + Y||| = inf { ||x1 + y|| : y ∈Y } . Then ∃ y1 ∈ Y ∋

ε
|||x1 + Y||| + > ||x1 + y1|| , and
2

20
 |||x2 + Y||| = inf { ||x2 + y|| : y ∈Y} , Then ∃y2 ∈Y ∋
ε
|||x2 + Y||| + 2
> ||x2 + y2|| .

||x1 + y1 + x2 + y2 || ≤ ||x1 + y1|| + ||x2 + y2||

ε ε
≤ |||x1 + Y||| + 2 + |||x2 + Y||| + 2

Let y = y1 + y2 ∈Y . Then ,

||(x1+x2) + y|| ≤ |||x1 + Y||| + |||x2 + Y||| + ℇ —(1)

Now , |||(x1 + Y) + (x2 + Y)||| = |||x1 + x2 + Y|||

=inf { ||x1 + x2 + y|| : y ∈Y }

< ||x1 + x2 + y||

≤ |||x1 + Y||| + |||x2 + Y||| + ℇ (by (1) )

since ℇ is arbitrary , we have

|||(x1 + Y) + (x2 + Y)||| ≤|||x1 + Y||| + |||x2 + Y|||

∴ ||| ||| is a norm on X/Y.

Let (xn + Y) be a sequence in X/Y . Assume that (yn) is a sequence in Y ∋ (xn + yn) converges
to x in X.
That is , (xn - x + yn) converges to 0 . —(1)
Claim: (xn + Y) converges to x + Y.
Consider
|||xn + Y - (x+Y)||| = |||(xn - x) + Y|||

= inf { ||xn - x + yn|| : y ∈Y }

≤ ||xn - x + yn|| ∀yn ∈Y .

Then by (1) , xn + Y converges to x + Y in X/Y.

Conversely assume that the sequence (xn + Y) → x + Y in X/Y.

Consider |||xn + Y - (x + Y)||| = |||xn - x + Y|||

= inf { ||xn -x + y|| : y ∈Y }

Then we can choose yn ∈Y ∋

21
 1
||xn - x + yn|| < |||(xn - x) + Y||| + 𝑛 , n=1,2,3,….

Since xn+Y →x+Y , we get

(xn - x + yn) converges to zero as n →∞

That is , (xn + yn) converges to x in X as n →∞

c) Consider 1≤ p < ∞

Given that
𝑝 𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (||𝑥(1)||1 + ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚)

Clearly , ||x||p ≥0 .
𝑝
Since each ||𝑥(𝑖)||𝑖 ≥ 0 .

𝑝
||x||p = 0 ⇔ |𝑥(𝑗)|𝑗 = 0 ∀ j = 1, . . . ,m

⇔ x(j) = 0 ∀ j.

⇔ x = (x(1), . . . ,x(m)) = 0
1/𝑝
( 𝑝
||kx||p = ||𝑘𝑥(1)||1 +. . . + ||𝑘𝑥(𝑚)||𝑚
𝑝
)
1/𝑝
= ( |𝑘| ||𝑥(1)|| )
𝑝 𝑝 𝑝 𝑝
1
+. . . + |𝑘| ||𝑥(𝑚)||𝑚

1/𝑝
( 𝑝
= |𝑘| ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= |𝑘| ||𝑥||𝑝 , k∈K and x∈X

1/𝑝
(
Now, ||𝑥 + 𝑦||𝑝 = ||𝑥(1) + 𝑦(1)||1 +. . . + ||𝑥(𝑚) + 𝑦(𝑚)||𝑚
𝑝 𝑝
)
(by Minkowski’s inequality)

≤
1/𝑝
( (||𝑥(1)||1
+ ||𝑦(1)||1 )𝑝
(
+. . . + ||𝑥(𝑚)||𝑚 + ||𝑦(𝑚)||𝑚 ))
𝑝

1/𝑝 1/𝑝

( ) ( )
𝑚 𝑚
𝑝 𝑝
≤ ∑ ||𝑥(𝑗)||𝑗 + ∑ ||𝑦(𝑗)||𝑗 (Minkowski’s inequality )
𝑗=1 𝑗=1

22
 1/𝑝
( 𝑝
= ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= ||x||p + ||y||p

Now suppose p =∞

||x||∞ = max { ||x(1)||1 , . . . , ||x(m)|| m }

||x||∞ ≥0 Since ||x(j)|| ≥0 , ∀j

||x||∞ = 0 ⇔ ||x(m)|| = 0 ∀m

⇔ x(m) = 0 ∀m

⇔x=0

||kx||∞ = max { ||kx(1)||1 , . . . , ||kx(m)||m }

= |k| max { ||x(1)||1 , . . . , ||x(m)||m }

= |k| ||x||∞

||x + y||∞ = max { ||x(1) + y(1)||1, . . . , ||x(m) + y(m)||m }

≤max { ||x(1)||1 + ||y(1)||1 , . . . , ||x(m)||m + ||y(m)||m }

= max { ||x(1)||1 , . . . , ||x(m)||m } + max { ||y(1)||1 , . . . , ||y(m)||m }

= ||x||∞ + ||y||∞

We now consider ,

𝑝 𝑝 1/𝑝
( )
||𝑥𝑛 − 𝑥(1)||𝑝 = ||𝑥𝑛(1) − 𝑥(1 ||1 +. . . + ||𝑥𝑛(𝑚) − 𝑥(𝑚)||𝑚)

Then

xn →x in X ⇔ ||xn - x ||p → 0
𝑝
⇔ ||xn(j) - x(j)||𝑗 → 0

⇔ xn(j) - x(j) → 0

⇔ xn(j) → x(j) in X ∀j .

23
 RIESZ LEMMA

Let 𝑋 be a normed space . 𝑌be a closed subspace of 𝑋and 𝑋 ≠ 𝑌 . Let 𝑟be a real number
such that 0 < 𝑟 < 1 . Then there exist some xr ∈X such that ||xr|| = 1 and

r<dist ( xr , Y )≤1

Proof :
We have ,

dist (x , Y) = inf { d(x , y) : y∈Y}

= inf { ||x - y|| : y ∈Y}

Since Y ≠X , consider x ∈X ∋ x∉Y.

If dist(x , Y) = 0 , then ||x - y|| = 0 ⇒ x∈𝑌= Y ( ∵ Y is closed )

Therefore ,

dist (x , Y) ≠0

That is ,
dist (x , Y) > 0
1
Since 0 < r < 1 , >1
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
⇒ > dist (x , Y)
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
That is , is not a lower bound of { ||x - y|| : y ∈Y }
𝑟

𝑑𝑖𝑠𝑡(𝑥 , 𝑌)
Then ∃ y0 ∈Y ∋ ||x - y0|| <
𝑟
→(1)
𝑥 − 𝑦0
Let xr = . Then xr ∈X
||𝑥 − 𝑦0||

( ∵y0 ∈Y , x∉Y ⇒x - y0 ∈X and ||x - y0|| ≠0 )

24
 𝑥 − 𝑦0 ||𝑥 − 𝑦0||
Then ||xr|| = || || = =1
||𝑥 − 𝑦 ||0
||𝑥 − 𝑦0||

Now to prove r < dist(𝑥𝑟, 𝑌)≤1

We have dist(xr , Y) = inf { ||xr - y|| : y∈Y }

≤||xr - y|| ∀y∈Y

In particular, 0∈ 𝑌, so that dist(xr , Y) ≤||xr - 0|| = 1

That is ,

dist (xr , Y) ≤1

Now ,
𝑥 − 𝑦0
dist (xr , Y) = dist ( ,Y )
||𝑥−𝑦0||

1
= dist ( x - y0 , Y)
||𝑥−𝑦0||

1
= inf { ||x - y0 - y|| : y∈ 𝑌}
||𝑥−𝑦0||

1
= inf {||x - (y0+ y)|| : y0 + y∈ 𝑌}
||𝑥−𝑦0||

1
= dist (x , Y)
||𝑥−𝑦0||

𝑟
> dist (x , Y) by (1)
𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)

⇒ dist (xr , Y) > r

That is ,

r < dist (xr , Y) ≤ 1

25
 CONCLUSION

This project discusses the concept of normed linear space that is fundamental to
functional analysis . A normed linear space is a vector space over a real or complex
numbers ,on which the norm is defined . A norm is a formalization and generalization to
real vector spaces of the intuitive notion of “length” in real world
In this project , the concept of a norm on a linear space is introduced and thus
illustrated . It mostly includes the properties of normed linear spaces and different proofs
related to the topic.

26
 BIBLIOGRAPHY

➢ Balmohan V Limaye , Functional Analysis

➢ S.H Friedberg , Arnold J.Insel and Lawrence E.Spence , Linear Algebra , 2 nd Edition, PH
Inc .
➢ S. Narayanan and Mittai , A Text Book of Matrices , Revised Edition, S. Chand

27
NUMBER THEORETIC FUNCTION
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
DILNA TERECE JOSE
DB18CMSR03
Under the guidance of
Ms. Ajeena joseph

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Number Theoretic Function’ is a bona fide

project of DILNA TERECE JOSE carried out the project work under my
supervision.

Mrs. Riya Baby Ms. Ajeena joseph

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I DILNA TERECE JOSE hereby declare that the project ‘Number Theoretic
Function’ is an original record of studies and bona fide project carried out by me
during the period of 2018 – 2021 under the guidance of Ms. Ajeena joseph,
Department of Mathematics, Don Bosco Arts and Science College, Angadikkadavu
and has not submitted by me elsewhere for the award of my degree, diploma, title or
recognition, before.

Name
DILNA TERECE JOSE

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the Almighty
who is always with me. Also, I must express my deepest gratitude to people along the
way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my project
supervisor Ms. Ajeena joseph, Department of Mathematics, Don Bosco Arts and
Science College, Angadikkadavu, for the invaluable guidance, loving encouragement
and meticulous care towards me throughout my career. I express my deep sense of
gratitude to all the faculty members of the Department of Mathematics, Don Bosco
Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and the
staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every one of
them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- ARITHMETIC FUNCTION 4-10

04 Chapter 2- EULER’S 𝜙 FUNCTION 11-20

05 Conclusion 21

06 Bibliography 22
 INTRODUCTION

A Number Theoretic Function is a complex valued function defined for all positive
integers. In Number Theory, there exist many number theoretic functions. This
includes Divisor Function, Sigma Function, Euler’s-Phi Function and Mobius
Function. All these functions play a very important role in the field of Number
Theory.

In the first chapter we will discuss about Arithmetic Function. In the second chapter
we will introduce Euler’s-Phi Function and Mobius Function.

1
 PRELIMINARY

Let n be a fixed positive integer. Two integers a and b are said to be congruent
modulo n, symbolized by
a ≡ b (mod n)
if n divides the difference a − b; that is, provided that a − b = kn for some integer k.
Example:
To fix the idea, consider n = 7. It is routine to check that
3 ≡ 24 (mod 7) − 31 ≡ 11 (mod 7) − 15 ≡ −64 (mod 7)
Because 3 − 24 = (−3)7, −31 − 11 = (−6)7 and −15 − (−64) = 77. When
n does not divide (a − b), we say that a is incongruent to b modulo n, and in this case
we write
a ≢ b (mod n). For a simple example: 25 ≢ 12 (mod 7), because 7 fails to divide

25 − 12 = 13.
It is to be noted that any two integers are congruent modulo 1, whereas two integers
are congruent modulo 2 when they are both even or both odd. In as much as
congruence modulo 1 is not particularly interesting, the usual practice is to assume
that n > 1.
Remark:
Given an integer a, let q and r be its quotient and remainder upon division by n,
so that
a = qn + r 0≤r<n
Then, by definition of congruence, a ≡ r (mod n). Because there are n choices for
r , we see that every integer is congruent modulo n to exactly one of the values
0, 1, 2, . . . , n − 1; in particular, a ≡ 0 (mod n) if and only if n | a.

2
Fundamental Theorem of Arithmetic
is Every integer 𝑛 > 1 can be represented as Product of prime factor in only one way,
apart from the order of the factors.

Residue

If a is an integer and 𝑚 is a positive integer then the residue class of a modulo 𝑚 is

denoted by 𝑎ˆ and is given by

𝑎ˆ = {𝑥: 𝑥 ≡ 𝑎(𝑚𝑜𝑑𝑚)}
= {𝑥: 𝑥 = 𝑎 + 𝑚𝑘, 𝑘 = 0, ±1, ±2, ⋯ }

3
 CHAPTER 1
ARITHMETIC FUNCTION
An arithmetic Function is a function defined on the positive integers which take
values in the real or complex numbers. i.e., A function f: N→ C is called an
arithmetic function.
An arithmetic function is called multiplicative if f(mn) = f(m)f(n) for all coprime
natural numbers m and n.
Examples
a) Sum of divisors 𝜎(n)
b) Number of divisors 𝜏(n)
c) Euler’s function 𝜙(n)
d) Mobius function 𝜇(n)

Definition 1.1

Given a positive integer n, let τ (n) denote the number of positive divisors of n and

σ(n) denote the sum of positive divisors of n.

Example

Consider n = 12. Since 12 has the positive divisors 1, 2, 3, 4, 6, 12, we find that

τ (12) = 6 and σ(12) = 1 + 2 + 3 + 4 + 6 + 12 = 28

For the first few integers,

τ (1) = 1 τ (2) = 2 τ (3) = 2 τ (4) = 3 τ (5) = 2 τ (6) = 4, . . .

σ(1) = 1, σ(2) = 3, σ(3) = 4, σ(4) = 7 , σ(5) = 6, σ(6) = 12, . . .

It is not difficult to see that τ (n) = 2 if and only if n is a prime number; also,

σ(n) = n + 1 if and only if n is a prime.

4
Theorem 1.1
𝑘 𝑘
If n = 𝑝1 1 … … … … . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

(a) τ (n) = (k1+ 1)(k2 + 1) ・ ・ ・ (kr + 1), and

𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
(b) σ(n) = ………………….
𝑝1 −1 𝑝𝑟 −1

Proof

The positive divisors of n are precisely those integers

𝑎 𝑎 𝑎
d = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟

where 0 ≤ ai ≤ ki . There are k1 + 1 choices for the exponent a1; k2 + 1 choices for a2, .

. . ; and kr + 1 choices for ar . Hence, there are

(k1 + 1)(k2 + 1) · · · (kr + 1)

possible divisors of n.

To evaluate σ(n), consider the product

𝐾 𝐾
(1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Each positive divisor of n appears once and only once as a term in the expansion of

this product, so that

𝐾 𝐾
σ(n) = (1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Applying the formula for the sum of a finite geometric series to the ith factor on the

right-hand side, we get

𝑘 +1
𝐾 𝑝𝑖 𝑖 − 1
(1 + 𝑝𝑖 + 𝑃𝑖2 + ⋯ … … … 𝑃𝑖 𝑖 ) =
𝑝𝑖 − 1

5
It follows that
𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
σ(n) = 𝑝1 −1
………………….
𝑝𝑟 −1
.

Corresponding to the ∑ notation for sums, the notation for products may be

defined using ∏ , the Greek capital letter pi. The restriction delimiting the numbers

over which the product is to be made is usually put under the ∏

sign.

Examples

With this convention, the conclusion to Theorem 1.1 takes the compact form: if
𝑘 𝑘 𝑘
n = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

and

6
Theorem 1.2

The functions τ and σ are both multiplicative functions

Proof

Let m and n be relatively prime integers. Because the result is trivially true if

either m or n is equal to 1, we may assume that m > 1 and n > 1. If

are the prime factorizations of m and n . It follows that the prime factorization of

the product mn is given by

Applying to theorem 1.1, we obtain

In a similar fashion, theorem 1.1 gives

Thus, τ and σ are multiplicative functions.

Theorem 1.3

If f is a multiplicative function and F is defined by

then F is also multiplicative.

7
Proof

Let m and n be relatively prime positive integers. Then

because every divisor d of mn can be uniquely written as a product of a divisor d1

of m and a divisor d2 of n, where gcd(d1, d2) = 1. By the definition of a

multiplicative function,

f (d1d2) = f (d1) f (d2)

It follows that

It might be helpful to take time out and run through the proof of Theorem 1.3

in a concrete case. Letting m = 8 and n = 3, we have

= f (1) + f (2) + f (3) + f (4) + f (6) + f (8) + f (12) + f (24)

= f (1 · 1) + f (2 · 1) + f (1 · 3) + f (4 · 1) + f (2 · 3)+ f (8 · 1) + f (4 · 3) + f (8 · 3)

= f (1) f (1) + f (2) f (1) + f (1) f (3) + f (4) f (1) + f (2) f (3)+ f (8) f (1)

. + f (4)f(3)+ f (8) f (3)

8
= [ f (1) + f (2) + f (4) + f (8)][ f (1) + f (3)]

= F(8)F(3)

Theorem 1.3 provides a deceptively short way of drawing the conclusion that τ

and σ are multiplicative

The Mangoldt function 𝚲(𝒏)

Definition 1.2

For every integer 𝑛 ≥ 1 we define

log 𝑝 if 𝑛 = 𝑝𝑚 for some prime 𝑝 and some 𝑚 ≥ 1 ,

Λ(𝑛) = {
0 otherwise.

Here is a short table of values of Λ(𝑛) :

𝑛: 1 2 3 4 5 6 7 8 9 10
Λ(𝑛): 0 log 2 log 3 log 2 log 5 0 log 7 log 2 log 3 0

The proof of the next theorem shows how this function arises naturally from the

fundamental theorem of arithmetic.

Theorem 1.4

If 𝑛 ≥ 1 we have

log 𝑛 = ∑ Λ(𝑑 ) … … … … … … … … … … … … (1)

𝑑∣𝑛

Proof

The theorem is true if 𝑛 = 1 since both members are 0 . Therefore, assume that 𝑛 > 1

and write

9
 𝑟

𝑛 = ∏ 𝑝𝑘 𝑎𝑘
𝑘=1

Taking logarithms we have

𝑟

log 𝑛 = ∑ 𝑎𝑘 log 𝑝𝑘
𝑘=1

Now consider the sum on the right of (1). The only nonzero terms in the sum come

from those divisors 𝑑 of the form 𝑝𝑘 𝑚 for 𝑚 = 1,2, … , 𝑎𝑘 and 𝑘 = 1,2, … , 𝑟. Hence

𝑟 𝑎𝑘 𝑟 𝑎𝑘 𝑟

∑ Λ(𝑑) = ∑ ∑ Λ(𝑝𝑘𝑚 ) = ∑ ∑ log 𝑝𝑘 = ∑ 𝑎𝑘 log 𝑝𝑘 = log 𝑛

𝑑∣𝑛 𝑘=1 𝑚=1 𝑘=1 𝑚=1 𝑘=1

which proves (1).

10
 CHAPTER 2
EULER’S 𝝓 FUNCTION
Let n be positive integer. Let Un denote the set of all positive integers less than n and
coprime to it

For example,

U6 = {1,5}

U10 = {1,3,7,9}

U18 = {1,5,7,11,13,17}

Definition 2.1
Euler’s 𝜙 function is a function 𝜙: N→N such that for any n ∈ N, 𝜙 (n) is the number
of integers less than n and coprime to it

In other words

‘Euler’s 𝜙 function counts the number of elements in Un’

For example,

𝜙(1) = 1, 𝜙(2) = 1, 𝜙(3) = 2, 𝜙(4) = 2, 𝜙(5) = 4

𝜙(6) = 2 … .

Theorem 2.1
Let p be a prime. Then 𝜙 (p) = p-1

Proof:
By definition, any natural number strictly less than p is coprime to p, hence

𝜙 (p) = p-1

Theorem 2.2

If 𝑝 is a prime and 𝑘 > 0, then

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1)

11
Proof:

Consider the successive pk natural numbers not greater than pk arranged in the
following rectangular array of p columns and pk-1 rows

1 2 . . p

p+1 p+2 . . 2p

. . . . .

. . . . .

pk-p+1 pk-p+2 . . pk

among these numbers only the ones at the rightmost sides are not coprime to pk and
there are pk-1 members in that column. So

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1).

For example, 𝜙(8) = 23 − 22 = 4 which counts the number of elements in the set
U8 = {1,3,5,7}

By the fundamental theorem of arithmetic, we can write any natural number n as

𝑘 𝑘
n= 𝑝1 1 … … … . . 𝑝𝑟 𝑟

where 𝑃𝑖 ‘s are distinct prime and k 𝑖 ≥ 1 are integers. We already know how to find
𝑘 𝑘
𝜙(𝑝𝑖 𝑖 ) we would lie to see how 𝜙(𝑛) is related to 𝜙(𝑝𝑖 𝑖 ). This follows from a very
important property of Euler’s 𝜙 Function

Multiplicativity of Euler’s 𝝓 Function

Theorem 2.3

𝜙(mn) = 𝜙(m)𝜙(n) if m and n are coprime natural numbers.

Proof:

12
Consider the array of natural numbers not greater than mn arranged in m columns and
n rows in the following manner

1 2 ⋯ 𝑟 ⋯ 𝑚
𝑚+1 𝑚+2 𝑚+𝑟 2𝑚
2𝑚 + 1 2𝑚 + 2 2𝑚 + 𝑟 3𝑚
⋮ ⋮ ⋮ ⋮
(𝑛 − 1)𝑚 + 1 (𝑛 − 1)𝑚 + 2 (𝑛 − 1)𝑚 + 𝑟 𝑛𝑚

Clearly each row of the above array has m distinct residues modulo m. Each column
has n distinct residues modulo n: for 1 ≤ 𝑖, 𝑖 ≤ 𝑛 − 1

im +j ≡ im + j (mod n)

⇒ im ≡ im (mod n)

⇒ i ≡ i (mod n) (as gcd(m,n) = 1)

⇒i≡i

Each row has 𝜙(m) residues coprime to m, and each column has 𝜙(n) residues
coprime to n. Hence in total 𝜙(m)𝜙(n) elements in the above array which are
coprime to both m and n, it follows that

𝜙(mn) = 𝜙(m)𝜙(n)

Theorem 2.4
Let n be any natural numbers, then
1 1 1
𝜙(𝑛) = 𝑛 (1 − ) (1 − ) ⋯ (1 − )
𝑝1 𝑝2 𝑝𝑟

Proof:
By fundamental theorem of arithmetic, we can write
𝑘 𝑘 𝑘
𝑛 = 𝑃1 1 𝑃2 2 … … . … 𝑃𝑟 𝑟

Where 𝑝𝑖 are the distinct prime factor of n, and 𝑘𝑖 are the non negative integers. By
previous theorem and proposition,
𝑘 𝑘
𝜙(𝑛) = 𝜙(𝑝1 1 ) ⋅ … , 𝜙(𝑝𝑟 𝑟 )
𝑘 −1 𝑘
= 𝑃1 1 (𝑃1 − 1) ⋯ 𝑃𝑟 𝑟−1 (𝑃𝑟 − 1)

13
 𝑘 1 𝑘 1
= 𝑝1 1 (1 − ) ⋯ 𝑃𝑟 𝑟 (1 − )
𝑃1 𝑃𝑟

1 1
= 𝑛 (1 − ) ⋯ ⋅ (1 − )
𝑝1 𝑝𝑟

Theorem 2.5
For n > 2, 𝜙 (n) is an even integer.

Proof:

First, assume that n is a power of 2, let us say that n = 2k , with k ≥ 2. By

theorem 2.2,
1
𝜙(𝑛) = 𝜙(2𝑘 ) = 2𝑘 (1 − ) = 2𝑘−1
2
an even integer. If 𝑛 does not happen to be a power of 2, then it is divisible by an odd
prime 𝑝; we therefore may write 𝑛 as 𝑛 = 𝑝𝑘 𝑚, where 𝑘 ≥ 1 and gcd (𝑝𝑘 , 𝑚) = 1.
Exploiting the multiplicative nature of the phi-function, we obtain

𝜙(𝑛) = 𝜙(𝑝𝑘 )𝜙(𝑚) = 𝑝𝑘−1 (𝑝 − 1)𝜙(𝑚)

which again is even because 2 | p – 1.

Theorem 2.6
For each positive integer n,

𝑛 = ∑ 𝜙(𝑑)
𝑑∣𝑛

Proof:
Let us partition the set {1,2,…….,n} into mutually disjoint subsets Sd for each d/n,
where

𝑆𝑑 = {1 ≤ 𝑚 ≤ 𝑛 ∣ gcd (𝑚, 𝑛) = 𝑑}
𝑚 𝑛 𝑚 𝑛
= {1 ≤ ≤ ∣ gcd ( , ) = 1}
𝑑 𝑑 𝑑 𝑑

Then

14
 {1,2, … … . , n} = ∑ 𝑆𝑑
𝑑∣𝑛

𝑛
⇒ 𝑛 = ∑ 𝜙( )
𝑑
𝑑∣𝑛

= ∑ 𝜙(𝑑)
𝑑∣𝑛

As for each divisor of n, n/d is also a divisor of n

MOBIUS FUNCTION

Definition 2.2
The Mobius function 𝜇: 𝑁 ⟶ {0, ±1} is defined as

1 if 𝑛 = 1
𝜇(𝑛) = {0 if 𝑝2 ⁄𝑛 for some prime 𝑝
(−1)𝑟 if 𝑛 = 𝑝1 𝑝2 ⋯ 𝑝𝑟 , where 𝑝𝑖 are distinct primes

For example,

μ(1) = 1 μ(2) = −1 μ(3) = −1

μ(4) = 0 μ(5) = −1 μ(6) = 1

If 𝑝 is a prime number, it is clear that 𝜇(𝑝) = −1; in addition, 𝜇(𝑝𝑒 ) = 0 for 𝑒 ≥ 2.

Theorem 2.7
The Mobius function is a multiplicative function i.e.

μ(mn) = μ(m)μ(n), if m and n are relatively prime

Proof:
Let m and n be coprime integers, we can consider the following to cases
2
Case 1: let μ(mn) = 0 then there is a prime p such that 𝑝 ⁄𝑚𝑛. As m and n are
𝑝2⁄ 𝑝2⁄
coprime p cannot divide both m and n hence either 𝑚 or 𝑛 . Therefore either
μ(m) = 0 or μ(n) = 0 and we have μ(mn) = μ(m)μ(n)

Case 2: suppose that μ(mn) ≠ 0 then mn is square free, hence so are m and n. let
15
𝑚 = 𝑝1 … … … 𝑝𝑟 and 𝑛 = 𝑞1 … … … 𝑞𝑠 where 𝑝𝑖 𝑎𝑛𝑑 𝑞𝑗 are all distinct primes then
mn = 𝑝1 … … … 𝑝𝑟 𝑞1 … … … 𝑞𝑠 where all the primes occurring in the factorization of
mn are distinct. Hence

𝜇(𝑚𝑛) = (−1)𝑟+𝑠

= (−1)𝑟 (−1)𝑠

= μ(m)μ(n)

Theorem 2.8

1 if 𝑛 = 1
∑ 𝜇(𝑑) = {
0 if 𝑛 > 1
𝑑∣𝑛

Where d runs through all the positive divisors of n.

Proof:

𝐿𝑒𝑡 𝐹(𝑛) = ∑ 𝜇(𝑑)

𝑑∣𝑛

As μ is multiplicative, so is F(n) by the theorem (F be a multiplicative arithmetic

function 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ) then F is also a multiplicative arthmetic function)
𝑑∣𝑛

Clearly

𝐹 (1) = ∑ 𝜇 (𝑑 )
𝑑∣𝑛

= μ(1)

=1

For integers which are prime power, i.e. of the form pk for some k ≥ 1

𝐹 (𝑝2 ) = μ(1) + μ(p) + μ(p2 ) + ⋯ … … … . . +μ(p𝑘 )

= 1 + (-1) + 0……………+ 0

=0

Now consider any integer n, and consider its prime factorization. Then
𝑘 𝑘
𝑛 = 𝑝1 1 … … … … … . 𝑝𝑟 𝑟 , 𝑘𝑖 ≥ 1

16
 𝑘
⇒ 𝐹 (𝑛) = ∏𝐹(𝑝𝑖 𝑖 )

=0

Mobius inversion formula

The following theorem is known as Mobius inversion formula

Theorem 2.9
Let F and f be two function from the set N of natural number to the field complex
number C such that

𝐹(𝑛) = ∑ 𝑓(𝑑)
𝑑∣𝑛

Then we can express f(n) as

𝑛
𝑓 (𝑛 ) = ∑ 𝜇 (𝑑 )𝐹 ( )
𝑑
𝑑∣𝑛

𝑛
= ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Proof:
First observe that if d is divisor of n so is n/d. Hence both the summation in the last
line of the theorem are same. Now

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

The crucial step in the proof is to observe that the set of S of pairs of integers (c,d)
with d|n and c|n/d is the same as the set T of pairs (c,d) with c/n and d|n/c.

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

= ∑ ( ∑ 𝜇(𝑑)𝑓(𝑐))
𝑑∣𝑛 𝑐∣(𝑛/𝑑)

17
 = ∑ 𝑓(𝑐)𝜇(𝑑)
(𝑐,𝑑)∈𝑆

= ∑ 𝑓 (𝑐 ) 𝜇 (𝑑 )
(𝑐,𝑑)∈𝑇

= ∑ (𝑓(𝑐) ∑ 𝜇(𝑑))
𝑐∣𝑛 𝑑∣(𝑛/𝑐)

= F(n)
𝐴𝑠 ∑ 𝜇(𝑑 ) = 0 𝑢𝑛𝑙𝑒𝑠𝑠 𝑛⁄𝑐 = 1, which happens when c = n
𝑑∣𝑛

Let us demonstrate this with n = 15

15
∑ 𝜇 (𝑑 )𝐹 ( ) = 𝜇 (1)[ 𝑓 (1) + 𝑓 (3) + 𝑓 (5) + 𝑓 (15)] + 𝜇(3)[ 𝑓 (1) + 𝑓 (5)]
𝑑
𝑑∣15

+𝜇(5)[ 𝑓 (1) + 𝑓 (3)] + 𝜇(15)[ 𝑓 (1)]

= f (1)[μ(1) + μ(3) + μ(5) + μ(15)] + f (3)[μ(1) + μ(5)] + f (5)[μ(1) +

μ (5)] + f(15) μ(1)

= f(1).0 + f(3).0 + f(5).0 + f(15)

= f(15)

The above theorem leads to the following interesting identities

1. we know that for any positive integer n,

∑ 𝜙 (𝑑 ) = 𝑛
𝑑∣𝑛

Where 𝜙(𝑛) is Euler’s 𝜙 function. Hence

𝑛
𝜙 (𝑛 ) = ∑ 𝜇 ( ) 𝑑
𝑑
𝑑∣𝑛

For example,

𝜙 (10) = μ(1)10 + μ(2)5 + μ(5)2 + μ(10)1

18
 = 10 - 5 – 2 + 1

=4

2. similarly

𝜎(𝑛) = ∑ 𝑑
𝑑∣𝑛
𝑛
𝑛 = ∑ 𝜇 ( ) 𝜎(𝑑)
𝑑
𝑑∣𝑛

For example,

With n = 10

μ(10). 1 + μ(2)(1 + 5) + μ(5)(1 + 3) + μ(1)(1 + 3 + 5 + 10)

= 1 – 1 – 5 – 1 – 3 + 1 + 3 +5 + 10

= 10

We have seen before that if multiplicative so is 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ). But we can now

𝑑∣𝑛

Prove that converse applying the Mobius inversion formula

Theorem 2.10
If F is a multiplicative function and

𝐹 (𝑛 ) = ∑ 𝑓 (𝑑 )
𝑑∣𝑛

then f is also multiplicative.

Proof:
By the Mobius inversion formula we know that
𝑛
𝑓 (𝑛) = ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Let m and n be relatively prime positive integers. We recall that any divisor

d of mn can be uniquely written as d = d1, d2, where d1 |m, d2 | n, and

𝑚 𝑛
gcd(d1, d2) = 1 = gcd( , ).
𝑑1 𝑑2

19
Conversely if d1/m and d2/n then d1d2/mn thus,
𝑚𝑛
𝑓(𝑚𝑛) = ∑ 𝜇(𝑑)𝐹 ( )
𝑑
𝑑∣𝑚𝑛
𝑚𝑛
= ∑ 𝜇 (𝑑1 𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇 (𝑑1 )𝜇(𝑑2 )𝐹 ( )𝐹( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇(𝑑1 )𝐹 ( ) ∑ 𝜇 (𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 ∣𝑚 𝑑2 ∣𝑛
= 𝑓(𝑚)𝑓(𝑛)

In view of the above theorem we can say that as N(n) = n is a multiplicative function
so is 𝜙(𝑛) because

∑ 𝜙(𝑑 ) = 𝑛 = N(n)
𝑑∣𝑛

20
 CONCLUSION

The purpose of this project gives a simple account of Arithmetic function, Euler’s phi
function and Mobius Function. The study of these topics given excellent introduction
to the subject called ‘NUMBER THEORETIC FUNCTION’

Number Theoretic Function demands a high standard of rigor. Thus, our presentation
necessarily has its formal aspect with care taken to present clear and detailed
argument. An understanding of the statement of the theorem, number theory proof is
the important issue. In the first chapter we discuss about function τ and σ are both
multiplicative function. If f is a multiplicative function and F is defined by

𝐹(𝑛) = ∑𝑑∣𝑛 𝑓(𝑑 ), then F is also multiplicative. In the second chapter 2 we discuss
about that if p is prime the 𝜙(𝑝) = 𝑝 − 1, 𝜙(𝑚𝑛) = 𝜙(𝑚)𝜙(𝑛). The Mobius
function is multiplicative function if f is multiplicative function and 𝐹(𝑛)=∑𝑑∣𝑛 𝑓(𝑑 ),

then F is also multiplicative.

21
 BIBLIOGRAPHY
1 DAVID M. BURTON - ELEMENTRY NUMBER THEORY
2 TOM.M. APOSTOL - INTRODUCTION TO ANALYTIC NUMBER THEORY
3 https://en.m.wikipedia.org/wiki/Euler%27s_totient_function
4 https://en.m.wikipedia.org/wiki/Arithmetic_function

22
 GRAPH COLORING

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
DONA ROSE STEPHEN
DB18CMSR21
Under the guidance of
MRS. Riya Baby

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report „GRAPH COLORING‟ is the bona fide
project of DONA ROSE STEPHEN who carried out the project under my
supervision.

Mrs. Riya Baby

Supervisor, HOD

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, DONA ROSE STEPHEN, hereby declare that this project report entitled
“GRAPH COLORING” is an original record of studies and bona fide project
carried out by me during the period from November 2019 to March 2020, under
the guidance of Mrs. Riya Baby, Department of Mathematics, Don Bosco Arts
and Science College, Angadikadavu and has not been submitted by me elsewhere
for the award of any degree, diploma, title or recognition, before.

DONA ROSE STEPHEN

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great
help to me during the course of my dissertation. First and foremost I thank the
almighty, for his blessing and protection during the period of this work. I express
my thanks to Dr. Fr. Francies Karakkatt, Principal, for support in the completion of
this dissertation. I express my gratitude to Mrs. Riya Baby, my project supervisor,
for the constant encouragement, valuable guidance and timely corrections, which
made this work a success.

I am also indebted to all my classmates and friends who supported me

throughout the study. I would like to express my thanks to my parents and dear
ones for their constant encouragement and support. I also thank all those who
helped me directly or indirectly to complete this project.

DONA ROSE STEPHEN

 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 Basic Concepts 2

03 Chapter 1-Coloring 3–6

04 Chapter 2-Chromatic Number 7 – 16

05 Chapter 3-Applications Of Graph Colouring 17 – 19

06 Conclusion 20
07 Bibliography 21
 INTRODUCTION

A proper coloring of a graph is an assignment of colors to the vertices of the graph so that

no two adjacent vertices have the same color.

Usually we drop the word “proper” unless other types of coloring are also under

discussion. Of course, the “colors” don’t have to be actual colors ; may can be any distinct

labels - integers ,for examples , if a graph is not connected , each connected component can

be colored independently; except where otherwise noted , we assume graphs are

connected. We also assume graphs are simple in this section. Graph coloring has many

applications in addition to its intrinsic interest.

In the same way the most important concept of graph coloring is utilized in

resource allocation, scheduling. Also, paths, walks and circuits in graph theory are used in

tremendous applications say travelling salesman problem, database design concepts,

resource networking.

This project deals with coloring which is one of the most important topics in

graph theory. In this project there are three chapters. First chapter is coloring . The second

chapter is chromatic number. The last chapter deals with application of graph coloring.

1
 BASIC CONCEPTS

1. GRAPH
A graph is an ordered triplet. G=(V(G), E(G), I(G)); V(G) is a non empty set, E(G) is a set

disjoint from V(G) and I(G) is an incidence map that associates each element of E(G) and

unrecorded pair of element of V(G). The elements of V(G) are called vertices (or nodes or

points) of G and the elements of E(G) are Called edges or lines of G.

2. MULTIPLE EDGE / PARALLEL EDGE

A set of 2 or more edges of a graph G is called a multiple edge or parallel edge if they

have the same end vertices.

3. LOOP
An edge for which the 2 end vertices are same is called a loop.

4. SIMPLE GRAPH
A graph is simple if it has no loop and no multiple edges.

5. DEGREE
Let G be a graph and v € V the number of edge incident at V in G is called the degree or

vacancy of the vertex v in G.

2
 CHAPTER - 1

COLORING

Graph coloring is nothing but a simple way of labeling graph components such as

vertices , edges and regions under some constraints. In a graph, no two adjacent vertices,

adjacent edges , or adjacent regions are colored with minimum number of colors .This

number is called the chromatic number and the graph is called properly colored graph.

In graph theory coloring is a special case of graph labeling; it is

an assignment of labels traditionally called “colors” to elements of a graph subject to certain

constraints. In it simplest form, it is a way of coloring the vertices of a graph such that no

two adjacent vertices share the same color, it is called vertex coloring. Similarly, edge

coloring assigns a color to each edge so that no two adjacent edges share the common

color.

While graph coloring , the constraints that are set on the graph are

colors , order of coloring , the way of assigning color , etc. A coloring is given to a vertex or a

particular region . Thus, the vertices or regions having same colors form independent sets.

3
VERTEX COLORING
Vertex coloring is an assignment of colors to the vertices of a graph ‘G ’ such

that no two adjacent vertices have the same color .Simply put , no two vertices of an edge

should be of the same color.

The most common type of vertex coloring seeks to minimize the

number of colors for a given graph . Such a coloring is known as a minimum vertex coloring ,

and the minimum number of colors which with the vertices of a graph may be colored is

called the chromatic number .

CHROMATIC NUMBER:
The minimum number of colors required for vertex coloring of graph ‘ G ’

is called as the chromatic number of G , denoted by X (G) .

X(G) = 1 iff ‘ G ’ is a null graph. If ‘G ’ is not a null graph , then X(G) ≥ 2.

EXAMPLES;

1. 2.

Null Graph ( X (G) = 1 ) Not Null Graph ( X (G) = 2 )

4
EDGE COLORING
An edge coloring of a graph G is a coloring of the edges of G such that adjacent

edges ( or the edges bounding different regions ) receive different colors. An edge coloring

containing the smallest possible number of colors for a given graph is known as a minimum

edge coloring.

The edge chromatic number gives the minimum number of colours with which
graph’s edges can be colored.

CHROMATIC INDEX
The minimum number of colors required for proper edge coloring of graph is

called chromatic index.

A complete graph is the one in which each vertex is directly connected with all

other vertices with an edge. If the number of vertices of a complete graph is n, then the

chromatic index for an odd number of vertices will be n and the chromatic index for even

number of vertices will be n-1.

5
EXAMPLES;
1.

The given graph will require 3 unique colors so that no two incident edges have the

Same color. So its chromatic index will be 3.

2.

The given graph will require 2 unique colors so that no two incident edges have

the same color. So its chromatic index will be 2.

6
 CHAPTER 2

Chromatic Number

The chromatic number of a graph is the smallest number of colors needed to color the vertices
of so that no two adjacent vertices share the same color. That is the smallest value of possible
to obtain a k-coloring.

 Graph Coloring is a process of assigning colors to the vertices of a graph.

 It ensures that no two adjacent vertices of the graph are colored with the same color.
 Chromatic Number is the minimum number of colors required to properly color any graph.

Graph Coloring Algorithm

 There exists no efficient algorithm for coloring a graph with minimum number of colors.

However, a following greedy algorithm is known for finding the chromatic number of any given
graph.

Greedy Algorithm

Step-01:

Color first vertex with the first color.

7
Step-02:

Now, consider the remaining (V-1) vertices one by one and do the following-

 Color the currently picked vertex with the lowest numbered color if it has not been used to
color any of its adjacent vertices.
 If it has been used, then choose the next least numbered color.
 If all the previously used colors have been used, then assign a new color to the currently
picked vertex.

Problems Based On Finding Chromatic Number of a Graph

Problem-01:

Find chromatic number of the following graph-

8
Solution-

Applying Greedy Algorithm, we have

Vertex a b C d e f

Color C1 C2 C1 C2 C1 C2

From here,

 Minimum numbers of colors used to color the given graph are 2.

 Therefore, Chromatic Number of the given graph = 2.

The given graph may be properly colored using 2 colors as shown below-

9
Problem-02:

Find chromatic number of the following graph-

Solution-

Applying Greedy Algorithm, we have-

Vertex a b C d e f

Color C1 C2 C2 C3 C3 C1

From here,

 Minimum numbers of colors used to color the given graph are 3.

 Therefore, Chromatic Number of the given graph = 3.

10
The given graph may be properly colored using 3 colors as shown below-

Chromatic Number of Graphs

Chromatic Number of some common types of graphs are as follows-

1. Cycle Graph-

 A simple graph of ‘n’ vertices (n>=3) and ‘n’ edges forming a cycle of length ‘n’ is called as a
cycle graph.
 In a cycle graph, all the vertices are of degree 2.

Chromatic Number
 If number of vertices in cycle graph is even, then its chromatic number = 2.
 If number of vertices in cycle graph is odd, then its chromatic number = 3.

11
Examples-

2. Planar Graphs-

A planar graph is a graph that can be embedded in the plane, that is it can be drawn on the
plane in such a way that its edges intersect only at their endpoint. In other words, it can be
drawn in such a way that no edges cross each other.

12
A Planar Graph is a graph that can be drawn in a plane such that none of its edges cross each
other.

Chromatic Number
Chromatic Number of any Planar Graph is less than or equal to 4

Examples-
+

 All the above cycle graphs are also planar graphs.

 Chromatic number of each graph is less than or equal to 4.

13
3. Complete Graphs-

 A complete graph is a graph in which every two distinct vertices are joined by exactly one
edge.
 In a complete graph, each vertex is connected with every other vertex.
 So to properly it, as many different colors are needed as there are number of vertices in the
given graph.

Chromatic Number
Chromatic Number of any Complete Graph

= Number of vertices in that Complete Graph

Examples-

14
4. Bipartite Graphs-

A bipartite graph is a graph whose vertices can be divided into two disjoint and independent
sets U and V such that every edge connects a vertex in U to one in V. Vertex sets U and V are
usually called the parts of the graph.

 A Bipartite Graph consists of two sets of vertices X and Y.

 The edges only join vertices in X to vertices in Y, not vertices within a set.

Chromatic Number
Chromatic Number of any Bipartite Graph

=2

Example-

15
5. Trees-

A tree is an undirected graph in which any two vertices are connected by exactly one path, or
equivalently a connected acyclic undirected graph.

 A Tree is a special type of connected graph in which there are no circuits.

 Every tree is a bipartite graph.
 So, chromatic number of a tree with any number of vertices = 2.

Chromatic Number
Chromatic Number of any tree

=2

Examples-

16
 CHAPTER-3
APPLICATIONS OF GRAPH COLORING

1) Making Schedule or Time Table:

Suppose we want to make an exam schedule for a university. We have list
different subjects and students enrolled in every subject. Many subjects would have common
students (of same batch, some backlog students, etc). How do we schedule the exam so that no
two exams with a common student are scheduled at same time? How many minimum time
slots are needed to schedule all exams? This problem can be represented as a graph where
every vertex is a subject and an edge between two vertices mean there is a common student.
So this is a graph coloring problem where minimum number of time slots is equal to the
chromatic number of the graph.

2) Mobile Radio Frequency Assignment:

When frequencies are assigned to towers, frequencies assigned to all towers at the
same location must be different. How to assign frequencies with this constraint? What is the
minimum number of frequencies needed? This problem is also an instance of graph coloring
problem where every tower represents a vertex and an edge between two towers represents
that they are in range of each other.

3) Register Allocation:
In compiler optimization, register allocation is the process of assigning a large number
of target program variables onto a small number of CPU registers. This problem is also a graph
coloring problem.

4) Sudoku:
Sudoku is also a variation of Graph coloring problem where every cell represents a
vertex. There is an edge between two vertices if they are in same row or same column or same
block.

17
 5) Map Coloring:

Geographical maps of countries or states where no two adjacent cities cannot be

assigned same color. Four colors are sufficient to color any map.

6) Bipartite Graphs:

We can check if a graph is bipartite or not by coloring the graph using two colors. If a
given graph is 2-colorable, then it is Bipartite, otherwise not. See this for more details.

Explanation;

Algorithm:
A bipartite graph is possible if it is possible to assign a color to each vertex such that no
two neighbour vertices are assigned the same color. Only two colors can be used in this
process.

Steps:
1. Assign a color (say red) to the source vertex.

2. Assign all the neighbours of the above vertex another color (say blue).

3. Taking one neighbour at a time, assign all the neighbour's neighbours the color red.

4. Continue in this manner till all the vertices have been assigned a color.

5. If at any stage, we find a neighbour which has been assigned the same color as that of the
current vertex, stop the process. The graph cannot be colored using two colors. Thus the graph
is not bipartite.

18
Example:

19
 CONCLUSION

This project aims to provide a solid background in the basic topics of graph coloring. Graph
coloring problem is to assign colors to certain elements of a graph subject to certain
constraints. The nature of coloring problem depends on the number of colors but not on what
they are.

The study of this topic gives excellent introduction to the subject called “Graph
Coloring”.

This project includes two important topics such as vertex coloring and edge coloring and came
to know about different ways and importance of coloring.

Graph coloring enjoys many practical applications as well as theoretical challenges.

Besides the applications, different limitations can also be set on the graph or on the away a color
is assigned or even on the color itself. It has been reached popularity with the general public in
the form of the popular number puzzle Sudoku and it is also use in the making of time
management which is an important application of coloring. So graph coloring is still a very
active field of research.

20
 BIBLIOGRAPHY

1) V.K Balakrishnan, outlines, graph theory.

2) J.A Bondy and U.S.R Murty, graph theory with applications.

3) S.Arumugam, S.Ramachandran, invitation to graph theory.

21
NORMED LINEAR SPACES
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
FREDIN JOSHY
DB18CMSR13
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE
It is to certify that this project report ‘Normed Linear Spaces’ is the bonafide project of

Fredin Joshy carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head Of Department Supervisor

Department Of Mathematics
Don Bosco Arts And Science College
Angadikadavu

2
 DECLARATION
I Fredin Joshy hereby declare that the project ‘Normed Linear Space’ is an original record of
studies and bona fide project carried out by me during the period of 2018 – 2021 under the
guidance of Ms. Athulya P , Department of Mathematics, Don Bosco Arts and Science College,
Angadikadavu and has not submitted by me elsewhere for the award of my degree, diploma, title
or recognition, before.

Fredin Joshy

DB18CMSR13

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu

3
 ACKNOWLEDGEMENT

First and foremost, praises and thanks to God, the Almighty, for His showers of blessings
throughout my work to complete the project successfully. I had finally managed to finish up
this project with great enthusiasm and determination. All the time spent to search and discuss
ideas as well as justifying theoretical clues to drive the output were worth my effort and time.

Therefore, I would like to express my sincere gratitude to my supervisor Ms. Athulya P,

Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu, for providing
invaluable guidance, comments and suggestions throughout the course of the project. I also
acknowledge with a deep sense of reverence, my gratitude towards all the faculty members of
the Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I owe and respectfully offer my thanks to the principal and staff of Don Bosco Arts and
Science College, Angadikadavu for their constant moral support and mellifluous affection
provided to me.

I express my gratitude to all who directly or indirectly helped me to complete this project
successfully. Their guidance and support was very helpful in bringing this work to
conclusion.

4
 CONTENTS

SI no Title Page no

1 Introduction 6

2 Preliminaries 7-9

3 Chapter 1 10 - 18

4 Chapter 2 19 - 25

5 Conclusion 26

6 Bibliography 27

5
 INTRODUCTION

This chapter gives an introduction to the theory of normed linear spaces. A skeptical reader may
wonder why this topic in pure mathematics is useful in applied mathematics. The reason is quite
simple: Many problems of applied mathematics can be formulated as a search for a certain
function, such as the function that solves a given differential equation. Usually the function
sought must belong to a definite family of acceptable functions that share some useful properties.
For example, perhaps it must possess two continuous derivatives. The families that arise
naturally in formulating problems are often linear spaces. This means that any linear combination
of functions in the family will be another member of the family. It is common, in addition, that
there is an appropriate means of measuring the “distance” between two functions in the family.
This concept comes into play when the exact solution to a problem is inaccessible, while
approximate solutions can be computed. We often measure how far apart the exact and
approximate solutions are by using a norm. In this process we are led to a normed linear space,
presumably one appropriate to the problem at hand. Some normed linear spaces occur over and
over again in applied mathematics, and these, at least, should be familiar to the practitioner.
Examples are the space of continuous functions on a given domain and the space of functions
whose squares have a finite integral on a given domain.

6
 PRELIMINARIES

1) LINEAR SPACES
We introduce an algebraic structure on a set 𝑋and study functions on X which are well behaved
with respect to this structure. From now onwards , K will denote either R , the set of all real
numbers or C, the set of all complex numbers. For k ∈ C , Re k and Im k will denote the real and
imaginary part of k.

A linear space(or a vector space) over K is a non-empty set 𝑋 along with a function
+ : 𝑋 × 𝑋 → 𝑋, called addition and a function ·: K× 𝑋 → 𝑋 called scalar multiplication, such
that for all 𝑥 , 𝑦 , 𝑧 ∈ 𝑋and 𝑘 , 𝑙 ∈ K , we have

𝑥+ 𝑦 = 𝑦 + 𝑥

𝑥 + (𝑦 + 𝑧) = (𝑥 + 𝑦) + 𝑧

∃0∈𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + 0 = 𝑥,

∃ − 𝑥 ∈ 𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + (− 𝑥) = 0 ,

𝑘 · (𝑥 + 𝑦) = 𝑘 · 𝑥 + 𝑘 · 𝑦,

(𝑘 + 𝑙)⋅𝑥 = 𝑘 · 𝑥 + 𝑙 · 𝑥,

(𝑘𝑙)⋅𝑥 = 𝑘 · (𝑙 · 𝑥),

1⋅𝑥 = 𝑥.

We shall write 𝑘𝑥 in place of 𝑘 · 𝑥. We shall also adopt the following notations. For
𝑥, 𝑦 ∈ 𝑋, 𝑘 ∈ K and subsets 𝐸, 𝐹of 𝑋 ,

𝑥 + 𝐹 = {𝑥 + 𝑦: 𝑦 ∈ 𝐹},

𝐸 + 𝐹 = {𝑥 + 𝑦: 𝑥 ∈ 𝐸, 𝑦 ∈ 𝐹},`

𝑘𝐸 = {𝑘𝑥: 𝑥 ∈ 𝐸}.

2) BASIS

A nonempty subset 𝐸 of 𝑋 is said to be a subspace of 𝑋 if 𝑘𝑥 + 𝑙𝑦 ∈ 𝐸 whenever 𝑥, 𝑦 ∈ 𝐸 and

𝑘, 𝑙 ∈ K . If ∅≠𝐸 ⊂ 𝑋, then the smallest subspace of 𝑋 containing 𝐸 is

7
 {
𝑠𝑝𝑎𝑛⁡𝐸 = 𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛: 𝑥1, …, 𝑥𝑛 ∈ 𝐸 , 𝑘1, …, 𝑘𝑛 ∈ 𝐾 }
It is called the span of 𝐸. If span 𝐸 = 𝑋, we say that 𝐸 spans 𝑋. A subset 𝐸 of 𝑋 is said to be
linearly independent if for all 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈K , the equation
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0implies that 𝑘1 = ⋯ = 𝑘𝑛 = 0. It is called linearly dependent if it is not
linearly independent, that is, if there exist 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈ K such that
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0, where at least one 𝑘𝑗is nonzero.

A subset E of X is called a Hamel basis or simply basis for X if span of E = X and E is

linearly independent .

3) DIMENSION

If a linear space 𝑋 has a basis consisting of a finite number of elements , then X is called finite
dimensional and the number of elements in a basis for 𝑋 is called the dimension of 𝑋, denoted as
dimX . Every basis for a finite dimensional linear space has the same (finite) number of elements
and hence the dimension is well-defined. The space {0} is said to have zero dimension. Note that
it has no basis !

If a linear space contains an infinite linearly independent subset, then it is said to be infinite
dimensional.

4)METRIC SPACE

We introduce a distance structure on a set 𝑋 and study functions on 𝑋 which are well-behaved
with respect to this structure.

A metric 𝑑 on a nonempty set 𝑋 is a function 𝑑: 𝑋 × 𝑋 →R

such that for all 𝑥, 𝑦, 𝑧 ∈ 𝑋

d(x, y) ≥0 and d(x , y) = 0 iff x=y

d(y , x) = d(x , y)

d(x , y) ≤ d(x , z) + d(z , y) .

The last condition is known as the triangle inequality. A metric space is a nonempty set 𝑋 along
with a metric on it.

8
5)CONTINUOUS FUNCTIONS

Roughly speaking, a function from a metric space to a metric space is continuous if it sends
‘nearby’ points to ‘nearby’ points. If 𝑋 and 𝑌 are metric spaces with metrics 𝑑 and 𝑒 respectively,
then a function 𝐹: 𝑋 → 𝑌 is said to be continuous at 𝑥0 ∈ 𝑋 if for every ϵ⟩ 0 , there is some
(
δ > 0 (possibly depending on ϵ and 𝑥0 ) such that 𝑒 𝐹(𝑥), 𝐹 𝑥0 ( )) < ϵ for all 𝑥 ∈ 𝑋 satisfying
( )
𝑑 𝑥, 𝑥0 < δ. Further, 𝐹 is said to be continuous on 𝑋 if it is continuous at every point of 𝑋. It is
easy to see that 𝐹 is continuous on 𝑋 if and only if the set F -1(E) is open in X whenever the set E
is open inY. Also , this happens iff F(xn) →F(x) in Y whenever xn →x in X.

6) UNIFORM CONTINUITY

We note that a continuous function 𝐹: 𝑇 → 𝑆 is, in fact, uniformly

continuous, that is, for every ϵ > 0, there exists some δ > 0
such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < ϵ whenever 𝑑 𝑡, 𝑢( )< δ. This can be seen as follows. Let 𝑡 ∈ 𝑇. By
ε
the continuity of 𝐹 at 𝑡 ∈ 𝑇, there is some δ𝑡, such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < 2 whenever
𝑑(𝑡, 𝑢) < δ𝑡.

7) FIELD

A ring is a set R together with two binary operations + and ·( which we call addition and
multiplication ) such that the following axioms are satisfied .

➢ R is an abelian group with respect to addition

➢ Multiplication is associative
➢ ∀𝑎 , 𝑏, 𝑐 ∈ 𝑅the left distributive law a(b + c) = (a·b) + (a·c) and the right distributive
law (a + b)c = (a·c) + (b·c) , hold .

A field is a commutative division ring

9
 CHAPTER 1

NORMED LINEAR SPACE

Let X be a linear space over K . A norm on X is the function || || from 𝑋 to R such that ∀
x,y ∈ X and k ∈ K ,

||𝑥||≥0 and ||𝑥|| = 0 if and only if x = 0 ,

||x + y|| ≤||x|| + ||y|| ,

||kx|| =|k| ||x|| .

A norm is the formalization and generalization to real vector spaces of the intuitive
notion of “ length” in the real world .

A normed space is a linear space with norm on it .

For x and y in X , let

d(x,y) = ||x - y||

Then d is a metric on X so that (X,d) is a metric space , thus every normed space is a metric
space

➢ Every normed linear space is a metric space . But converse may not be true .

Example :
|𝑥 − 𝑦|
d(x,y) = , ∀ x , y ∈X
1 + |𝑥 − 𝑦|

|𝑥 − 𝑦|
⇒ ||x - y|| =
1 + |𝑥 − 𝑦|

|𝑧|
⇒ ||z|| = , z = x - y ∈X
1 + |𝑧|

10
 |α𝑧|
||αz|| =
1 +|α𝑧|

|α| |𝑧|
=
1 + |α| |𝑧|

= |α| ( |𝑧|
1 + |α| |𝑧| )
≠ |α| ||z|| .

➢ Result

Let X be a normed linear space . Then ,

| ||x|| - ||y|| | ≤ ||x - y|| , ∀x , y ∈ X

Proof :

||x|| = || ( x - y ) + y|| ≤ ||x - y|| + ||y||

⇒||x|| - ||y|| ≤||x - y|| → (1)

x↔y

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||→(2)

From (1) and (2)

|||x|| - ||y||| ≤ ||x - y||

➢ Norm is a continuous function

Let xn →x , as n → ∞

11
 ⇒ xn - x →0 as n →∞

⇒ ||xn - x||→0 as n→∞

| ||xn|| - ||x|| | ≤ ||xn - x|| →0 as n→∞

⇒||xn|| - ||x|| →0 , as n →∞

⇒||x|| is continuous

➢ Norm is a uniformly continuous function

We have , || || :X→R . Let x,y ∈ X and ε> 0

Then ||x|| = ||x - y + y ||

≤||x - y|| + ||y||

⇒ ||x|| - ||y|| ≤ ||x - y||→ (1)

Interchanging x and y ,

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||

⇒ ||x|| - ||y|| ≥ - ||x - y|| → (2)

Combining (1) and (2)

- ||x - y|| ≤ ||x|| - ||y|| ≤ ||x - y||

That is ,

| ||x ||- ||y|| | ≤ ||x - y||

Take δ = ε , then whenever ||x - y|| < δ, | ||x|| -|| y|| |< ε

12
 Therefore || || is a uniformly continuous function .

➢ Continuity of addition and scalar multiplication

To show that + : X×X →X and ·: K ×X →X are continuous functions.

Let (x,y) ∈ X×X . To show that + is continuous at (x ,y) , that is , to show

that for each (x,y) ∈ X×X if xn→x and yn→y in X , then

+(xn , yn) →+(x , y) ;

That is ,
xn + yn →x + y .

Consider
||( xn + yn) - (x + y )|| = ||xn - x + yn - y||

≤||xn - x|| + ||yn - y||

Given xn→x and yn→y , for each ϵ > 0, ∃ N1 ∋

ε
||xn - x|| < 2
∀ n ≥N1 , and ∃ N2 ∋

ε
||yn - y|| < 2
∀ n ≥N2

\
Take N = max { N1, N2}

ε ε
Then ||xn - x|| < 2
and ||yn - y|| < 2
∀n≥N

ε ε
Therefore ||(xn + yn) - (x + y)|| ≤ 2
+ 2
= ε ∀n ≥N

That is , xn + yn →x + y

Now to show that ·: K ×X→X is continuous

Let (k , x) ∈ K×X

13
 To show that if kn →k and xn →x , then knxn →kx

Since kn →k , ∀ ε > 0 ∃ N1 ∋ |kn - k| < ε ∀ n ≥ N1

2

ε
Since xn →x , ∀ ε > 0 ∃ N2 ∋ ||xn - x|| < ∀ n ≥ N2
2

Consider ||knxn - kx|| = ||knxn - kx + xnk - xnk ||

= ||xn (kn - k) + k(xn - x)||

≤ ||xn(kn - k)|| + ||k(xn - x)||

= ||xn|| |kn- k| + |k| ||xn - x||

ε ε
≤ ||xn|| 2 + |k| 2

∴ knxn→kx

➢ Examples of normed space

1) Spaces Kn (K = R or C)

For n = 1 , the absolute value of function | | is a norm on K , since ∀k ∈K

We have ,

||k|| = ||k · 1|| = |k| ||1|| , by definition .

But ||1|| is a positive scalar .

∴ ||k|| is a positive scalar multiple of the absolute value function .

∴ any norm on K is a positive scalar multiple of the absolute value

function

For n > 1 , let p≥ 1 be a real number

14
 Kn = { ( x(1) , x(2) , . . . , x(n) ) : x(i) ∈K , i = 1 , 2 , . . . , n }

For x ∈Kn , that is , x = ( x(1) , x(2) , . . . , x(n) ) , define

𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (|𝑥(1)| +. . . + |𝑥(𝑛)| )

Then || ||p is a norm on Kn

When p = 1 ,

Then , ||x||1 = |x(1)| + |x(2)| + . . . + |x(n)|

Since |x(i)| ≥0 ∀ i = 1 , 2 , . . . , n , ||x||1 ≥0

And ||x||1 = 0 ⇔ |x(1)| + . . . +|x(n)| = 0

⇔ |x(i)| = 0 ∀ i

⇔ x(i) = 0 ∀ i

⇔ x = ( x(1) , . . . , x(n) ) = 0

Now ||kx||1 = |kx(1)| + |kx(2)| + . . . + |kx(n)|

= |k| |x(1)| + . . . + |k| |x(n)|

= |k| ( |x(1)| + . . . + |x(n)| )

= |k| ||x||1

||x + y||1 = |(x + y)(1)| + . . . + |(x + y)(n)|

= |x(1) + y(1)| + . . . + |x(n) + y(n)|

≤|x(1)| + |y(1)| + . . . + |x(n)| + |y(n)|

= |x(1)| + . . . + |x(n)| + |y(1)| + . . . + |y(n)|

= ||x||1 + ||y||1

15
 Consider 1<p<∞

𝑝 𝑝 1/𝑝
Now , ||𝑥||𝑝 = ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

𝑝
Since |x(i)| ≥0 ∀i , we have ||x||p ≥0

𝑝 𝑝 1/𝑝
And ||𝑥||𝑝 = 0⇔( |𝑥(1)| +. . . + |𝑥(𝑛)| ) = 0

𝑝
⇔ |𝑥(𝑖)| = 0 ∀i

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 ∀i

⇔ x = ( x(1) , . . . , x(n) ) = 0 .
Now
𝑝 𝑝 1/𝑝
||𝑘𝑥||𝑝 = ( |𝑘𝑥(1)| +. . . + |𝑘𝑥(𝑛)| )

𝑝 𝑝 𝑝 𝑝 1/𝑝
= ( |𝑘| |𝑥(1)| +. . . + |𝑘| |𝑥(𝑛)| )

𝑝 𝑝 1/𝑝
= |𝑘| ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

= |𝑘| ||𝑥||𝑝 .

𝑝 𝑝 1/𝑝
||𝑥 + 𝑦||𝑝 = ( |𝑥(1) + 𝑦(1)| +. . . + |𝑥(𝑛) + 𝑦(𝑛)| )

We have by Minkowski’s inequality ,

1/𝑝 1/𝑝 1/𝑝

( ) ( ) ( )
𝑛 𝑛 𝑛
𝑝 𝑝 𝑝
∑ |𝑥(𝑖) + 𝑦(𝑖)| ≤ ∑ |𝑥(𝑖)| + ∑ |𝑦(𝑖)|
𝑖=1 𝑖=1 𝑖=1
Then

16
 𝑝 𝑝 1/𝑝 𝑝 𝑝 1/𝑝
(
||𝑥 + 𝑦||𝑝 ≤ |𝑥(1)| +. . . + |𝑥(𝑛)| ) (
+ |𝑦(1)| +. . . + |𝑦(𝑛)| )
= ||𝑥||𝑝 + ||𝑦||𝑝

Then , for 1≤p<∞ ,|| ||𝑝 is a norm on Kn

When p = ∞, define ||𝑥||∞ = 𝑚𝑎𝑥 { |𝑥(1)| , |𝑥(2)| ,. . . , |𝑥(𝑛)| }

Then it is a norm on Kn

||𝑥||𝑝 ≥ 0 since each values |x(i)|≥0

So that

max {|x(i)| , i=1, . . . , n} ≥ 0

||𝑥||∞ = 0 ⇔𝑚𝑎𝑥 { |𝑥(𝑖)| : 𝑖 = 1,. . . , 𝑛 }= 0

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 , ∀i

⇔x = 0

||𝑘𝑥||∞ = 𝑚𝑎𝑥 { |𝑘𝑥(1)| ,. . . , |𝑘𝑥(𝑛)| }

= max { |k| |x(1)| , . . . , |k| |x(n)|}

= |k| max {|x(1)| , . . . , |x(n)|}

= |k| ||x||∞

||x + y||∞ = max { |x(1) + y(1)| , . . . , |x(n) + y(n)| }

≤max { |x(1)| + |y(1)| , . . . , |x(n)| + |y(n)| }

17
 ≤max { |x(1)| , . . . , |x(n)| } + max { |y(1)| , . . . , |y(n)| }

= ||x||∞ + ||y||∞

2) Sequence space

∞
𝑝 𝑝 𝑝
Let 1≤p < ∞, 𝑙 = { x = ( x(1) , x(2) , . . . ) ; x(i) ∈K and ∑ |x(j)| <∞} , that is , 𝑙 is the
𝑗=1
𝑝
space of p-summable scalar sequences in K . For x = (x(1) , x(2) , . . . ) ∈ 𝑙 ,

let ||x||p = ( |x(1)|p + |x(2)|p + . . . )1/p . Then it is a norm on lp.

That is , || ||p is a function from lp to R .

If p = 1 , then l1 is a linear space and ||x||1 = ( |x(1)| + |x(2)| + . . . ) is a norm on l1

∞
Let p = ∞. Then 𝑙 is the linear space of all bounded scalar sequences . And ,

||x||∞ = sup { |x(j)| : j = 1, 2, 3, . . . }

∞
Then || ||∞ is a norm on 𝑙

18
 CHAPTER 2

THEOREMS ON NORMED SPACES

a) Let Y be a subspace of a normed space X , then Y and its closure 𝑌 are normed spaces with
the induced norm.
b) Let Y be a closed subspace of a normed space X , for x +Y in the quotient space X/Y, let
|||x +Y||| = inf { ||x+y|| : y ∈Y} . Then ||| ||| is a norm on X/Y , called the quotient norm.

A sequence (xn + Y) converges to x + Y in X/Y iff there is a sequence (yn) in Y , (xn+ yn)
converges to x in X.

c) Let || ||pbe a norm on the linear space Xp , j = 1,2,…. . Fix p such that 1 ≤p ≤∞

For x = (x(1) , x(2) , … , x(m)) that is the product space X =X1× X2 × …× Xm ,

1/𝑝
Let ||𝑥||𝑝 = ( ||𝑥(1)|| 𝑝
1
𝑝
+ ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚
𝑝
) , if 1≤ p <∞

||x||p = max { ||x(1)||1 , … , ||x(m)||m } , if p = ∞.

Then || ||p is a norm on X.

A sequence (xn) converges to x in X ⇔ (xn(j)) converges to x(j) in Xj ∀ j=1,2,…,m.

Proof:
a) Since X is a normed space, there is a norm on X to Y . Since Y is a subspace of X,

|| ||y: Y → R is a function. To show that || ||y is a norm on Y.

For y∈ Y , || y||Y = ||y|| , then

||y||Y ≥ 0 ( ∵||y||≥ 0 ) and ||y||Y = 0 ⇔ y = 0

||ky||Y = ||ky|| = |k| ||y|| = |k| ||y||y .

Let y1 , y2 ∈Y. Then ,

||𝑦1 + 𝑦2||𝑦 = ||𝑦1 + 𝑦2|| ≤ ||𝑦1|| + ||𝑦2|| = ||𝑦1||𝑦 + ||𝑦2||𝑦

Now the continuity of addition and scalar multiplication shows that 𝑌is a subspace of X, since if
xn →x and yn → y , xn , yn ∈ 𝑌 , then

xn + yn →x + y (by continuity of addition) and

19
 kxn →kx (by continuity of scalar Xn) .

Since 𝑌is closed , x + y ∈ 𝑌 and kx ∈ 𝑌 . Therefore 𝑌 ≤ X.

∴ norm on X induces a norm on Y and 𝑌

b) X/Y , the quotient space equals X/Y={ x + Y : x ∈X }.

|||x + y||| = inf { ||x + y|| : y ∈Y }

Claim: ||| ||| is a norm on X/Y , called quotient norm

• Let x ∈X ,

|||x + Y||| = inf { ||x + y|| : y ∈Y } ≥ 0.

∴ |||x + Y||| ≥0 .

If |||x + y||| = 0 ( 0 in X/Y is Y) , then there is a sequence (yn) in Y ∋

||x + yn || →0

⇒ x + yn →0

⇒ yn →-x

Since yn ∈Y and Y is closed

-x ∈Y ⇔ x ∈Y (∵ Y is a subspace)

⇔x + Y = Y , zero in X/Y.

• For k ∈K ,

|||k(x + Y)||| = |||kx + Y|||

= inf { ||k(x + y)|| : y ∈Y}

= inf { |k| ||x + y|| : y ∈Y}

= |k| inf { ||x + y|| : y ∈Y}

= |k| |||x + Y||| .

• Let x1 , x2 ∈X . Then

|||x1 + Y||| = inf { ||x1 + y|| : y ∈Y } . Then ∃ y1 ∈ Y ∋

ε
|||x1 + Y||| + > ||x1 + y1|| , and
2

20
 |||x2 + Y||| = inf { ||x2 + y|| : y ∈Y} , Then ∃y2 ∈Y ∋
ε
|||x2 + Y||| + 2
> ||x2 + y2|| .

||x1 + y1 + x2 + y2 || ≤ ||x1 + y1|| + ||x2 + y2||

ε ε
≤ |||x1 + Y||| + 2 + |||x2 + Y||| + 2

Let y = y1 + y2 ∈Y . Then ,

||(x1+x2) + y|| ≤ |||x1 + Y||| + |||x2 + Y||| + ℇ —(1)

Now , |||(x1 + Y) + (x2 + Y)||| = |||x1 + x2 + Y|||

=inf { ||x1 + x2 + y|| : y ∈Y }

< ||x1 + x2 + y||

≤ |||x1 + Y||| + |||x2 + Y||| + ℇ (by (1) )

since ℇ is arbitrary , we have

|||(x1 + Y) + (x2 + Y)||| ≤|||x1 + Y||| + |||x2 + Y|||

∴ ||| ||| is a norm on X/Y.

Let (xn + Y) be a sequence in X/Y . Assume that (yn) is a sequence in Y ∋ (xn + yn) converges
to x in X.
That is , (xn - x + yn) converges to 0 . —(1)
Claim: (xn + Y) converges to x + Y.
Consider
|||xn + Y - (x+Y)||| = |||(xn - x) + Y|||

= inf { ||xn - x + yn|| : y ∈Y }

≤ ||xn - x + yn|| ∀yn ∈Y .

Then by (1) , xn + Y converges to x + Y in X/Y.

Conversely assume that the sequence (xn + Y) → x + Y in X/Y.

Consider |||xn + Y - (x + Y)||| = |||xn - x + Y|||

= inf { ||xn -x + y|| : y ∈Y }

Then we can choose yn ∈Y ∋

21
 1
||xn - x + yn|| < |||(xn - x) + Y||| + 𝑛 , n=1,2,3,….

Since xn+Y →x+Y , we get

(xn - x + yn) converges to zero as n →∞

That is , (xn + yn) converges to x in X as n →∞

c) Consider 1≤ p < ∞

Given that
𝑝 𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (||𝑥(1)||1 + ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚)

Clearly , ||x||p ≥0 .
𝑝
Since each ||𝑥(𝑖)||𝑖 ≥ 0 .

𝑝
||x||p = 0 ⇔ |𝑥(𝑗)|𝑗 = 0 ∀ j = 1, . . . ,m

⇔ x(j) = 0 ∀ j.

⇔ x = (x(1), . . . ,x(m)) = 0
1/𝑝
( 𝑝
||kx||p = ||𝑘𝑥(1)||1 +. . . + ||𝑘𝑥(𝑚)||𝑚
𝑝
)
1/𝑝
= ( |𝑘| ||𝑥(1)|| )
𝑝 𝑝 𝑝 𝑝
1
+. . . + |𝑘| ||𝑥(𝑚)||𝑚

1/𝑝
( 𝑝
= |𝑘| ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= |𝑘| ||𝑥||𝑝 , k∈K and x∈X

1/𝑝
(
Now, ||𝑥 + 𝑦||𝑝 = ||𝑥(1) + 𝑦(1)||1 +. . . + ||𝑥(𝑚) + 𝑦(𝑚)||𝑚
𝑝 𝑝
)
(by Minkowski’s inequality)

≤
1/𝑝
( (||𝑥(1)||1
+ ||𝑦(1)||1 )𝑝
(
+. . . + ||𝑥(𝑚)||𝑚 + ||𝑦(𝑚)||𝑚 ))
𝑝

1/𝑝 1/𝑝

( ) ( )
𝑚 𝑚
𝑝 𝑝
≤ ∑ ||𝑥(𝑗)||𝑗 + ∑ ||𝑦(𝑗)||𝑗 (Minkowski’s inequality )
𝑗=1 𝑗=1

22
 1/𝑝
( 𝑝
= ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= ||x||p + ||y||p

Now suppose p =∞

||x||∞ = max { ||x(1)||1 , . . . , ||x(m)|| m }

||x||∞ ≥0 Since ||x(j)|| ≥0 , ∀j

||x||∞ = 0 ⇔ ||x(m)|| = 0 ∀m

⇔ x(m) = 0 ∀m

⇔x=0

||kx||∞ = max { ||kx(1)||1 , . . . , ||kx(m)||m }

= |k| max { ||x(1)||1 , . . . , ||x(m)||m }

= |k| ||x||∞

||x + y||∞ = max { ||x(1) + y(1)||1, . . . , ||x(m) + y(m)||m }

≤max { ||x(1)||1 + ||y(1)||1 , . . . , ||x(m)||m + ||y(m)||m }

= max { ||x(1)||1 , . . . , ||x(m)||m } + max { ||y(1)||1 , . . . , ||y(m)||m }

= ||x||∞ + ||y||∞

We now consider ,

𝑝 𝑝 1/𝑝
( )
||𝑥𝑛 − 𝑥(1)||𝑝 = ||𝑥𝑛(1) − 𝑥(1 ||1 +. . . + ||𝑥𝑛(𝑚) − 𝑥(𝑚)||𝑚)

Then

xn →x in X ⇔ ||xn - x ||p → 0
𝑝
⇔ ||xn(j) - x(j)||𝑗 → 0

⇔ xn(j) - x(j) → 0

⇔ xn(j) → x(j) in X ∀j .

23
 RIESZ LEMMA

Let 𝑋 be a normed space . 𝑌be a closed subspace of 𝑋and 𝑋 ≠ 𝑌 . Let 𝑟be a real number
such that 0 < 𝑟 < 1 . Then there exist some xr ∈X such that ||xr|| = 1 and

r<dist ( xr , Y )≤1

Proof :
We have ,

dist (x , Y) = inf { d(x , y) : y∈Y}

= inf { ||x - y|| : y ∈Y}

Since Y ≠X , consider x ∈X ∋ x∉Y.

If dist(x , Y) = 0 , then ||x - y|| = 0 ⇒ x∈𝑌= Y ( ∵ Y is closed )

Therefore ,

dist (x , Y) ≠0

That is ,
dist (x , Y) > 0
1
Since 0 < r < 1 , >1
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
⇒ > dist (x , Y)
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
That is , is not a lower bound of { ||x - y|| : y ∈Y }
𝑟

𝑑𝑖𝑠𝑡(𝑥 , 𝑌)
Then ∃ y0 ∈Y ∋ ||x - y0|| <
𝑟
→(1)
𝑥 − 𝑦0
Let xr = . Then xr ∈X
||𝑥 − 𝑦0||

( ∵y0 ∈Y , x∉Y ⇒x - y0 ∈X and ||x - y0|| ≠0 )

24
 𝑥 − 𝑦0 ||𝑥 − 𝑦0||
Then ||xr|| = || || = =1
||𝑥 − 𝑦 ||0
||𝑥 − 𝑦0||

Now to prove r < dist(𝑥𝑟, 𝑌)≤1

We have dist(xr , Y) = inf { ||xr - y|| : y∈Y }

≤||xr - y|| ∀y∈Y

In particular, 0∈ 𝑌, so that dist(xr , Y) ≤||xr - 0|| = 1

That is ,

dist (xr , Y) ≤1

Now ,
𝑥 − 𝑦0
dist (xr , Y) = dist ( ,Y )
||𝑥−𝑦0||

1
= dist ( x - y0 , Y)
||𝑥−𝑦0||

1
= inf { ||x - y0 - y|| : y∈ 𝑌}
||𝑥−𝑦0||

1
= inf {||x - (y0+ y)|| : y0 + y∈ 𝑌}
||𝑥−𝑦0||

1
= dist (x , Y)
||𝑥−𝑦0||

𝑟
> dist (x , Y) by (1)
𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)

⇒ dist (xr , Y) > r

That is ,

r < dist (xr , Y) ≤ 1

25
 CONCLUSION

This project discusses the concept of normed linear space that is fundamental to
functional analysis . A normed linear space is a vector space over a real or complex
numbers ,on which the norm is defined . A norm is a formalization and generalization to
real vector spaces of the intuitive notion of “length” in real world
In this project , the concept of a norm on a linear space is introduced and thus
illustrated . It mostly includes the properties of normed linear spaces and different proofs
related to the topic.

26
 BIBLIOGRAPHY

➢ Balmohan V Limaye , Functional Analysis

➢ S.H Friedberg , Arnold J.Insel and Lawrence E.Spence , Linear Algebra , 2 nd Edition, PH
Inc .
➢ S. Narayanan and Mittai , A Text Book of Matrices , Revised Edition, S. Chand

27
DON BOSCO ARTS & SCIENCE COLLEGE
ANGADIKADAVU

DEPARTMENT OF MATHEMATICS
2018-2021

Project Report on

INNER PRODUCT SPACES

 DEPARTMENT OF MATHEMATICS

DON BOSCO ARTS & SCIENCE COLLEGE

ANGADIKADAVU

MARCH 2021

Project Report on

INNER PRODUCT SPACES

Dissertation submitted in the partial fulfilment

of the requirement for the award of

Bachelor of Science in Mathematics of

Kannur University

Name : JEROM DOMINIC

Roll No. : DB18CMSR14

Examiners: 1.

2.
 𝐊𝐀𝐍𝐍𝐔𝐑 𝐔𝐍𝐈𝐕𝐄𝐑𝐒𝐈𝐓𝐘

𝐁𝐎𝐍𝐀𝐅𝐈𝐃𝐄 𝐂𝐄𝐑𝐓𝐈𝐅𝐈𝐂𝐀𝐓𝐄

Certified that this project report on " INNER PRODUCT SPACES” is the bonafide
work of JEROM DOMINIC who carried out the project work under my supervision.

Mrs. Riya Baby Mr. Anil M V

Head of Department Supervisor
 𝐃𝐄𝐂𝐋𝐀𝐑𝐀𝐓𝐈𝐎𝐍

I, JEROM DOMINIC hereby declare that the project work entitled ‘INNER
PRODUCT SPACES′ has been prepared by me and submitted to Kannur University in
partial fulfilment of requirement for the award of Bachelor of Science is a record of
original work done by me under the supervision of Mr. ANIL M V, Assistant Professor,
Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I, also declare that this Project work has been submitted by me fully or partially for
the award of any Degree, Diploma, Title or recognition before any authority.

Place ∶ Angadikadavu
JEROM DOMINIC
Date ∶ DB18CMSR14
 𝐀𝐂𝐊𝐍𝐎𝐖𝐋𝐄𝐃𝐆𝐄𝐌𝐄𝐍𝐓

Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express mydeepest gratitude to
people along the way.
No words can adequately express the sense of gratitude, still I try to express my
heartfelt thanks through words. The outset, I am deeply indebted to my project
supervisor Mr. ANIL M.V, Assistant Professor, Department of Mathematics, Don
Bosco Arts and ScienceCollege, Angadikadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career.I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikadavu.
I could not name many who sincerely supported and helped for the successful
completion of this Project. It is my pleasure and duty to thank each and every one
of them who walked with me.

JEROM DOMINIC
 Contents

1. INTRODUCTION 1

2. PRELIMINARIES 2

3. CHAPTER 1 : INNER PRODUCT SPACES 6

4. CHAPTER 2 : ORTHOGONAL SETS 9

5. BIBLIOGRAPHY 16
INTODUCTION

In linear algebra, an inner product space is a vector space with an additional structure

called an inner product. This additional structure associates

each pair of vectors in the space with a scalar quantity known as the inner product of the

vectors. Inner products allow the rigorous introduction of intuitive geometrical notions

such as the length of a vector or the angle between two vectors. They also provide the

means of defining orthogonality between vectors (zero inner product). Inner product

spaces generalize Euclidean spaces (in which the inner product is the dot product, also

known as the scalar product) to vector spaces of any (possibly infinite) dimension and are

studied in functional analysis. The first usage of the concept of a vector space with an

inner product is due to Peano, in 1898.

An inner product naturally induces an associated norm, thus an inner product space is also

a normed vector space. A complete space with an inner product is called a Hilbert space.

An (incomplete) space with an inner product is called a pre-Hilbert space.

1
 PRELIMINARIES

LINEAR SPACES

Definition 1: A linear (vector) space X over a field F is a set of elements

together with a function, called addition, from X ×X into X and a function

called scalar multiplication, from F × X into X which satisfy the following

conditions for all x, y, z ∈ X and α, β ∈ F;

i. (x + y) + z = x + (y + z)

ii. x +y = y +x

iii. There is an element 0 in X such that x + 0 = x for all x ∈ X.

iv. For each x ∈ X there is an element −x ∈ X such that x + (−x) = 0.

v. (x + y) = αx + αy

vi. (α + β)x = αx + βx

vii. α(βx) = (αβ)x

viii. 1 · x = x.

Properties i to iv imply that X is an abelian group under addition and v to vi

relate the operation of scalar multiplication to addition X and to addition and

multiplication in F.

Examples:

(a) Vn(R). The vectors are n-tuples of real numbers and the scalars are real

2
 numbers with addition and scalar multiplication defined by

(1 ,···,  n ) + ( 1 ,···,  n ) = (1 + 1 ,···,  n +  n ) (1)

 (1 ,···,  n ) = ( 1 ,···,  n ) (2)

Vn(R) is a linear space over R. Similarly, the set of all n-tuples of complex

numbers with the above definition of addition and multiplication is a linear

space over C and is denoted as Vn(C).

(b) The set of all functions from a nonempty set X into a field F with addition and

scalar multiplication defined by

[f + g](t) = f (t) + g(t) and [αf ](t) = αf (t); f, g ∈ X, t ∈ T (3)

is a linear space.

Let T = N the set of all positive integers and X is the set of all sequences of

elements F with addition and scalar multiplication defined by

( n +  n ) = ( n +  n ) (4)

 ( n ) = (  n ) (5)

denoted as V∞(F), form a linear space.

3
METRIC SPACES

Remember the distance function in the Euclidean space Rn.

Let x, y, z ∈ Rn, then

(1) |x − y| ≥ 0; |x − y| = 0 if and only if x = y ;

(2) |x − y| = |y − x|;

(3) |x − y| ≤ |x − z| + z − y|.

Definition 2: A metric or distance function on a set X is a real valued function

d defined on X × X which has the following properties: for all x, y, z ∈ X.

(1) d(x, y) ≥ 0; d(x, y) = 0 if and only if x = y;

(2) d(x, y) = d(y, x);

(3) d(x, y) ≤ d(x, z) + d(z, y)

A metric space (X, d) is a nonempty set X and a metric d defined on X.

Examples: In addition to the Euclidean spaces let us have the following examples.

Here all functions are assumed to be continuous. Let Lp denotes a set of complex

p
valued functions in Rn such that f is integrable. Let us recall some results

concerning such functions.

Höder’s Inequality: If p > 1, 1/q = 1 − 1/p

 | fg |  [ | f | ] [  | g |q ]1/ q .
p 1/ p

Minkowski’s Inequality: If p ≥ 1,

[ | f + g | p ]1/ p  [ | f | p ]1/ p + [  | g | p ]1/p

4
If xk and yk for k = 1, … , m are complex numbers, let f (t) = |xk| and g(t) =

|yk| for t ∈ [k, k + 1] and f (t) = 0 = g(t) for t∈ [1, m + 1]. Then we obtain the

summation form of the above inequalities from the integral form

Hölder’s Inequality

1/ p 1/ q
m m p
 m q

 xk yk   xk    yk 
k =1  k =1   k =1 

Minkowski’s Inequality:

p 1/ p p 1/ p p 1/p
m  m  m 
  xk + yk     xk  +   yk 
 k =1   k =1   k =1 

NORMED LINEAR SPACES

Definition 3. A norm on X is a real valued function, whose value at x is denoted

by ||x||, satisfying the following conditions for all x, y ∈ X and α ∈ F;

(1) ||x|| > 0 if x ≠ 0

(2) ||αx|| = |α|||x||

(3) ||x + y|| ≤ ||x|| + ||y||.

A linear space X with a norm defined on it is called a normed linear space.

Example: l p space. On the linear space Vn(F), define

n
x = [ |  i | p ]1/ p
k =1

where p ≥ 1 is any real number and x = (1 ,···,  n ) . This defines a norm (called p-

p
norm) on Vn(F). This space is called l space .
5
 CHAPTER 1

INNER PRODUCT SPACES

INNER PRODUCTS

Let 𝐹 be the field of real numbers or the field of complex numbers, and V a vector space over

F an inner product on V is a function which assigns to each ordered’ pair of vectors 𝛼, 𝛽 in V

a scalar (𝛼|𝛽) in 𝐹 in such a way that for all 𝛼, 𝛽, γ in V and all scalars c.

(a) (𝛼 + 𝛽|𝛾) = (𝛼|𝛾) + (𝛽|𝛾) ;

(b) (c𝛼|𝛽) = 𝑐(𝛼|𝛽) ;

̅̅̅̅̅ ), the bar denoting complex conjugation

(c) (𝛽|𝛼) = (𝛼|𝛽

(d) (𝛼|𝛼) > 0 if 𝛼 ≠ 0

It should be observed that conditions (a), (b) and (c) implies that

(𝑒) = (𝛼 ∣ 𝑐𝛽 + 𝛾) = (𝑐̅(𝛼|𝛽) + (𝛼|𝛾)

One other point should be made. When 𝐹 is the field 𝑅 of real nunbers. The complex conjugates

appearing in (c) and (e) are superflom. However, in the complex case they are necessary for

the consistency of the conditions. Without these complex conjugates we would have the

contradiction

(𝛼|𝛼) > 0 and (𝑖𝛼 ∣ 𝑖𝛼) = −1(𝛼|𝛼)

Example 1:

On F 𝑛 there is an inner product which we call the standard inner product. It is defined on 𝛼 =

(𝑥1 , ⋯ 𝑥𝑛 ) and 𝛽 = (𝑦1 , … , 𝑦𝑛 ), by

6
 (𝛼|𝛽) = ∑ 𝑥𝑖 𝑦̅𝑖
𝑖

When F is R this may be also written as

(𝛼|𝛽) = ∑ 𝑥𝑖 𝑦𝑖
𝑖

In the real case, the standard inner product is often called the dot or scalar product and denoted

by 𝛼 ⋅ 𝛽.

INNER PRODUCTS SPACES

An inner product space is a real or complex vector space together with a specified inner product

on that space.

• A finite-dimensional real inner product space is often called a Euclidean spare. A

complex inner product spare often referred to as a unitary spare.

• Every inner product space is a normed linear space and every normed space is a metric

space. Hence , every inner product space is a metric space.

Theorem

If V is an inner product space, then for any vector’s 𝛼, 𝛽 in 𝑉 and any scalar c

(1) ||𝑐𝛼|| = |𝑐|||𝛼|| ;

(ii) ||𝛼|| > 0 for 𝛼 ≠ 0

(iii) |(𝛼 ∣ 𝛽)| ⩽ ||𝛼|| ||𝛽||

(iv) ∥ 𝛼 + 𝛽|| ⩽∥ 𝛼 ∥ +∥ 𝛽||

Proof:

Statements (i) and (ii) follow almost immediately form the various definitions

involved. The inequality in (iii) is clearly valid when 𝛼 = 0. if 𝛼 ≠ 0, put

(𝛽|𝛼)
𝛾=𝛽− 𝛼
∥ 𝛼 ∥2

7
 Then, (𝛾 ∣ 𝛼) = 0 and

(𝛽|𝛼) (𝛽|𝛼)
0 ⩽∥ 𝛾 ∥2 = (𝛽 − 2
𝛼 ⁄𝛽 − 𝛼)
∥𝛼∥ ∥ 𝛼 ∥2

(𝛽|𝛼)(𝛼|𝛽)
= (𝛽|𝛽) −
∥ 𝛼 ∥2

|(𝛼|𝛽)|2
=∥ 𝛽 ∥2 −
∥ 𝛼 ∥2

Hence,

|(𝛼 ∣ 𝛽)|2 ⩽∥ 𝛼 ∥2 ∥ 𝛽 ∥2

Now using (c) we find that

∥ 𝛼 + 𝛽 ∥2 = ∥ 𝛼 ∥2 + (𝛼 ∣ 𝛽) + (𝛽 ∣ 𝛼)+∥ 𝛽 ∥2

=∥ 𝛼 ∥2 + 2Re (𝛼 ∣ 𝛽)+∥ 𝛽 ∥2
⩽∥ 𝛼 ∥2 + 2 ∥ 𝛼 ∥∥ 𝛽 ∥ +∥ 𝛽 ∥2
= (∥ 𝛼 ∥ +∥ 𝛽 ∥)2

Thus,

∥ 𝛼 + 𝛽 ∥ ⩽ ∥ 𝛼 ∥ +∥ 𝛽 ∥

the inequality (iii) is called the Cauchy -Schwarz inequality. It has a wide variety of application

the proof shows that if 𝛼is non-zero then

(( 𝛼 ∣ 𝛽 )) <∥ 𝛼 ∥∥ 𝛽 ∥, unless

(𝛽|𝛼)
𝛽= 𝛼
∥ 𝛼 ∥2

Then equality occurs in (iii) if and only if 𝛼 and 𝛽 are linearly independent.

8
 CHAPTER 2

ORTHOGONAL SETS

Definition

Let α and β be the vectors in an inner product space V. Then α is orthogonal to β if

(α | β) = 0. We simply say that and are orthogonal.

Definition

If S is a set of vectors in V, S is called an orthogonal set provided all set pairs of

distinct vectors in S are orthogonal.

Definition

An orthogonal set is an orthogonal set S with the additional property that ∥ 𝛼 ∥= 1 for

every 𝛼 in S.

• The zero vectors are orthogonal to every vector in V and is the only vector with this

property.

• It is an appropriate to think of an orthonormal set as a set of mutually perpendicular

vectors each having length l.

Example: the vector (x , y) is 𝑅 2 is orthogonal to (−y , x) with respect to the standard inner

product, for,

((x , y)|(−y , x)) = −xy + yx = 0

• The standard basis of either 𝑅 𝑛 or 𝐶 𝑛 is an orthonormal set with respect to the standard

inner product.

9
Theorem : An orthogonal set of nonzero vectors is linearly independent.

Proof:

Let S be a finite or infinite orthogonal set of nonzero vectors in a given inner product space

suppose 𝛼1, 𝛼2 , … 𝛼𝑛 are distinct vectors in S and that β=𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛

Then (β|𝛼𝑘 )=( 𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛 |𝛼𝑘 )

= 𝑐1 (𝛼1 |𝛼𝑘 ) + 𝑐2 ( 𝛼2 |𝛼𝑘 )+…+𝑐𝑛 (𝛼𝑛 |𝛼𝑘 )

= 𝑐𝑘 (𝛼𝑛 |𝛼𝑘 ) , since (𝛼𝑖 |𝛼𝑗 ) = 0,if i ≠ j and (𝛼𝑖 |𝛼𝑗 ) = 1,if i=j

Hence, 𝑐𝑘 = (β |𝛼𝑘 ) /(𝛼𝑘 , 𝛼𝑘 ) )

𝑐𝑘 =(β |𝛼𝑘 )/||𝛼𝑘 ||2,1≤ k ≤ m

Thus, when β=0 each 𝑐𝑘 =0; so S is a linearly independent set.

Corollary:

If {𝛼1, 𝛼2 , … 𝛼𝑚 } is an orthogonal set of nonzero vectors in a finite dimensional inner product

space V, then m ≤ dimV.

That is number of mutually orthogonal vectors in V cannot exceed the dimensional V.

Corollary:

If a vector β is linear combination of an orthogonal of nonzero vectors 𝛼1, 𝛼2 , … 𝛼𝑛 , then β is

the particular linear combination

(𝛽 |𝛼𝑘 )
β =∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘
𝑘 ||

Proof :

Since β is the linear combination of an orthogonal sequence of nonzero vectors

𝛼1, 𝛼2 , … 𝛼𝑛 , we can write β =𝑐1 𝛼1 + ⋯ 𝑐𝑛 𝛼𝑛 .

(𝛽|𝛼𝑘 )
Where 𝑐𝑘 = , 1 ≤ k ≤ m (ref. by previous theorem)
||𝛼𝑘 ||2

( 𝛽 | 𝛼𝑘 )
Hence, β= ∑𝑚
𝑘=1 ||𝛼𝑘 ||2

10
Theorem (Gram Schmidt Orthogonalization Process)

Let V be an inner product space and {𝛽1, … , 𝛽𝑛 } be any linearly independent vectors in V. Then

one may construct orthogonal vectors {𝛼1, 𝛼2 , … 𝛼𝑛 } in V, such that for each k = 1, 2, …n, the

set {𝛼1, 𝛼2 , … 𝛼𝑘 }is an orthogonal basis for the subspace of V spanned by 𝛽1, … , 𝛽𝑛 .

Proof :

The vectors are obtained by means of a construction known as the Gram Schmidt

orthogonalization process.

First let 𝛼1 =𝛽1 The other vectors are then given inductively as follows:

Suppose 𝛼1, 𝛼2 , … 𝛼𝑚 (1 ≤ m ≤ n) have been chosen so that for every k

{𝛼1, 𝛼2 , … 𝛼𝑘 } (1≤k≤m)

is an orthogonal basis for the space of v that is spanned by 𝛽1, … , 𝛽𝑛

To construct the next vector 𝛼𝑚+1, let

( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1, = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
Then  m +1  0 . For otherwise, 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘 =0, implies,
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
𝛽𝑚+1= 𝛽𝑚+1 -∑𝑚
𝑘=1 𝛼𝑘 ,implies,  m+1 is a linear combination of 𝛼1, 𝛼2 , … 𝛼𝑚 and
||𝛼𝑘 ||2

hence a linear combination of 1 ,  2 ,...,  m , a contradiction.

Furthermore, if 1≤j≤m, then,

(𝛽𝑚+1 |𝛼𝑘 )
(𝛼𝑚+1 | 𝛼𝑗 ) = ( 𝛽𝑚+1 | 𝛼𝑗 ) -∑𝑚
𝑘=1 (𝛼𝑘 |𝛼𝑗 )
||𝛼𝑘 ||2

= ( 𝛽𝑚+1 | 𝛼𝑚 ) – ( 𝛽𝑚+1 | 𝛼𝑗 ) , using the orthonormality of {𝛼1, 𝛼2 , … 𝛼𝑚 }.

Therefore {𝛼1, 𝛼2 , …,𝛼𝑚+1 } is an orthogonal set consisting of m+1 nonzero vectors in the

subspace spanned by 𝛽1, … , 𝛽𝑚+1 . Hence by an earlier Theorem , it is a basis for this subspace

.Thus the vectors , 𝛼1, 𝛼2 , … 𝛼𝑛 may be constructed using the formula

11
 ( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1 = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

In particular, when n=3 ,we have

𝛼1 =𝛽1

(𝛼2 |𝛽2 )
𝛼2 =𝛽2- α1
||𝛼𝑘 ||2

(𝛽3 |𝛼1 ) (𝛼2 |𝛽3 )

𝛼3 =𝛽3 - α1 - 𝛼
||𝛼1 ||2 ||𝛼𝑘 ||2 2

Corollary :

Every finite dimensional inner product space has an orthonormal basis.

Proof :

Let V be a finite dimensional inner product space and { 𝛽1, … , 𝛽𝑛 } a basis for V. Apply the

gram Schmidt orthogonalization process to construct an orthogonal basis , simply replace each
𝛼𝑘
vector 𝛼𝑛 by .
||𝛼𝑘 ||

Gram-Schmidt process can be used to test for linear dependence . For suppose 𝛽1, … , 𝛽𝑛 are

linearly independent vectors in an inner product space; to exclude a trivial case , assume that

β≠0. Let m be largest integers for which 𝛽1, … , 𝛽𝑚 are independent. Then 1≤m˂n.

Let 𝛼1 , 𝛼2 , … 𝛼𝑚 be the vectors obtained by applying the orthogonalization process to

(𝛽𝑚+1 |𝛼𝑘 )
𝛽1, … , 𝛽𝑚 . Then the vector 𝛼𝑚+1 given by 𝛼𝑚+1 = 𝛽𝑚+1–∑𝑚
𝑘=1 𝛼𝑘 is necessarily 0.
||𝛼𝑘 ||2

For 𝛼𝑚+1 is in the subspace spanned by 𝛼1 , 𝛼2 , … 𝛼𝑚 and orthogonal to each of the vectors ,

(β|𝛼𝑘 )
hence it is 0 as β=∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘 . Conversely, if 𝛼1 , 𝛼2 , … 𝛼𝑚 are different from 0 and 𝛼𝑚+1 =0,
𝑘 ||

then 𝛽1, … , 𝛽𝑚+1 are linearly independent .

Definition:

A best approximation to β  V by vectors in a subspace W of V is a vector α  W such that

 −    −  for every vector   W .

12
Theorem

Let W be a subspace of an inner product space V and let  V .

1. The vector  W is a best approximation to  V by vectors in W if and

only if  − is orthogonal to every vector in W .

2. If a best approximation to  V by vectors in W exists, it is unique.

3. If W is finite-dimensional and {𝛼1, 𝛼2 , … 𝛼𝑛 } is any orthonormal basis for W ,

then the vector

 =
n
(  |  k )
k
2 k
k =1

is the (unique) best approximation to  by vectors in W.

Definition:

Let V be an inner product space and S be any set of vectors in V. The orthogonal complement

of S is the set S ⊥ of all vectors in V which are orthogonal to every vector in S.

That is, S ⊥ =  V : ( |  ) = 0,    S

Definition:

Whenever the vector α in the above theorem exists it is called the orthogonal projection of

β on W. If every vector in V has an orthogonal projection on W, the mapping that assigns to

each vector in V its orthogonal projection on W is called the orthogonal projection of V on W.

Corollary :

Let V be an inner product space and W a finite dimensional subspace and E be the

orthogonal projection of V on W. Then the mapping

β →β – Eβ

is the orthogonal projection of V on W ⊥ .

13
Proof :

Let β  V . Then β – Eβ  W ⊥ , and for any γ  W ⊥ , β – γ = E β+(β – Eβ – γ)

Since Eβ  W and β – Eβ – γ  W ⊥ ,

It follows that

||𝛽 – 𝛾||2 = (Eβ+(β – Eβ – γ) ,E β+(β –Eβ – γ))

= ||𝐸𝛽||2+||𝛽 – 𝐸𝛽 – 𝛾||2

≥ ||𝛽 – (𝛽 – 𝐸𝛽)||2

with strict inequality when γ≠ β – Eβ . Therefore, β – Eβ is the best approximation to β by

vectors in W ⊥ .

Theorem

Let W be a finite dimensional subspace of an inner product space V and let E be the orthogonal

projection of V on W. Then E is an idempotent linear transformation of V onto W , W ⊥ is the

null space of E , and V= W ⨁ W ⊥ .

Proof

Let β be an arbitrary vector in V. Then Eβ is the best approximation to β that lies in W .

In particular, Eβ =β when β is in W . Therefore, E(Eβ) =Eβ for every β in V; that is, E is

idempotent : 𝐸 2 = E . To prove that E is linear transformation, let α and β be any vectors in V

and c an arbitrary scalar ,Then by theorem,

α-Eα and β-Eβ are each orthogonal to every vector in W . Hence the vector

c(α-Eα)+(β-Eβ)=(cα +β)-(cEα +Eβ)

Also belongs to W ⊥ . Since cEα+ Eβ is a vector in W , it follows from theorem that

E(cα+ β)= cEα+ Eβ.

Again let β be any vector in V. Then Eβ is the unique vector in W such that β-Eβ is in W ⊥ .

Thus Eβ=0 when β is in W ⊥ .

Conversely, β is in W ⊥ when Eβ=0. Thus W ⊥ is the null space of E.

14
The equation ,

β = E β+β – Eβ

shows that V = W + W ⊥ ; moreover W  W ⊥ = {0} ; for if α is a vector in W  W ⊥ ,then

( |  ) =0. Therefore, α=0 and V is the direct sum of W and W ⊥ .

Corollary :

Under the conditions of theorem, I − E is the orthogonal projection of V on W ⊥ .

It is an independent linear transformation of V onto W ⊥ with null space W .

Proof :

We have seen that the mapping β →β- E β is the orthogonal projection of V on W ⊥ .

Since E is a linear transformation , this projection W ⊥ is the linear transformation I − E from

its geometric properties one sees that I − E is an idempotent .Transformation of V onto W .

This also follows from the computation ( I − E )( I − E )= I − E - E +𝐸 2

=I −E

Moreover , ( I − E )β =0 If and only if β = Eβ , and this is the case if and only if β is in W .

Therefore W is the null space of I − E .

15
 BIBLIOGRAPHY

● Balmohan V .Limaye :Functional analysis

● Kenneth Hoffman,Ray Kunze:Linear Algebra,second edition

● Stephen H.Friedberg, Arnold J.Insel, Lawrence E.Spence : Linear Algebra, Fourth

edition

16
PLANARITY IN GRAPH THEORY

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
JOYAL SIBY
DB18CMSR15
Under the guidance of
Mrs. PRIJA V

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu
June 2021
 CERTIFICATE

Certified that this project ‘Planar Graph’ is a bona fide project of Joyal Siby
carried out the project work under my supervision.

Mrs.
Mrs. Riya
Riya Baby
Baby Mrs. Prija V
Head
Head of
of Department
Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu

I
 DECLARATION

I Joyal Siby hereby declare that the project ‘Planar Graph’ is an original record
of studies and bona fide project carried out by me during the period of 2018 –
2021 under the guidance of Mrs. Prija V, Department of Mathematics, Don Bosco
Arts and Science College, Angadikkadavu and has not submitted by me
elsewhere for the award of my degree, diploma, title, or recognition, before.

Joyal Siby

DBI8CMSR15

Department of Mathematics

Don Bosco Arts and Science College,

Angadikkadavu

II
 ACKNOWLEDGEMENT

Introduction is the proper place to begin. But first I bow my head before
the Almighty who is always with me. Also, I must express my deepest gratitude
to people along the way.

No words can adequately express the sense of gratitude; still, I try to

express my heartfelt thanks through words. At the outset, I am deeply indebted to
my project supervisor Mrs. Prija V, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement, and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the principal
and the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the
successful completion of this project. It is my pleasure and duty to thank each and
every one of them who walked with me.

My greatest debt is always, to God Almighty.

III
 CONTENTS

Sl No Title Page No

1 Introduction 1

2 Chapter 1 - Basic Concepts 2-9

3 Chapter 2 – Planar Graph 10 - 17

4 Chapter 3 – Euler’s Formula 18 – 19

5 Chapter 4 – Dual of a Plane Graph 20

6 Conclusion 21

7 Bibliography 22

IV
 INTRODUCTION

In recent years, Graph Theory has established itself as an important

mathematical tool in a wide variety of subjects, ranging from Operational
Research and Chemistry to Genetics and Linguistics, and from Electrical
Engineering and Geography to Sociology and Architecture. At the same time, it
has also emerged as a worthwhile mathematical discipline in its own right.

A great mathematician, Euler become the Father of Graph Theory, when

in 1736, he solved a famous unsolved problem of his days, called Konigsberg
Bridge Problem. This is today, called as the First Problem of the Graph theory.
This problem leads to the concept of the planar graph as well as Eulerian Graphs,
while planar graphs were introduced for practical reasons, they pose many
remarkable mathematical properties. In 1936, the psychologist Lewin used planar
graphs to represent the life space of an individual.

1
Chapter 1

BASIC CONCEPTS

Graph

A graph is an ordered triple 𝐺 = {𝑉(𝐺), 𝐸(𝐺), 𝐼𝐺 } where 𝑉(𝐺) is a non-

empty set, 𝐸(𝐺) is a set disjoint from 𝑉(𝐺) and 𝐼(𝐺) is an incidence map that
associates each element of 𝐸(𝐺) and unordered pair of elements of 𝑉(𝐺). The
elements of 𝑉(𝐺) are called vertices (or nodes or points) of 𝐺 and the element
of 𝐸(𝐺) are called edges or lines of 𝐺.

Example:

Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 }

𝐸(𝐺) = {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 }
𝐼𝐺 (𝑒1 ) = {𝑣1 , 𝑣2 } or {𝑣2 , 𝑣1 }
I𝐺 (𝑒2 ) = {𝑣2 , 𝑣3 } or {𝑣3 , 𝑣2 }
𝐼𝐺 (𝑒3 ) = {𝑣3 , 𝑣4 } or {𝑣4 , 𝑣3 }
𝐼𝐺 (𝑒4 ) = {𝑣4 , 𝑣1 } or {𝑣1 , 𝑣4 }

Multiple edges

A set of two or more edges of a graph 𝐺 is called multiple edges or parallel

edges if they have the same end vertices.

2
Loop

An edge for which the two end vertices are same is called a loop.

Here {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 } form the parallel edges.

𝑒7 is the Loop.
Simple Graph

A graph is simple if it has no loops and no multiple edges.

Finite & Infinite Graphs

A graph is called finite if both 𝑉(𝐺) & 𝐸(𝐺) are finite. A graph that is not
finite is called infinite graph.

Adjacent Vertices

Two vertices 𝑢 and 𝑣 are said to be adjacent vertices if and only if there is
an edge with 𝑢 and 𝑣 as its end vertices.

3
Adjacent Edges

Two distinct edges are said to be adjacent edges if and only if they have a
continuous end vertex.

Complete Graph

A simple graph 𝐺 is said to be complete if every pair of distinct vertices of

𝐺 are adjacent in 𝐺. A complete graph with n vertices is denoted by 𝐾𝑛 .

Bipartite Graph

A graph is bipartite if its vertex set can be partitioned into two non-empty
subsets 𝑋 and 𝑌 such that each edge of 𝐺 has one end in 𝑋 and the other in 𝑌. The
pair (𝑋, 𝑌) is called a bipartition of the bipartite graph 𝐺. The bipartite graph 𝐺
with bipartition (𝑋, 𝑌) denoted by 𝐺 (𝑋, 𝑌).

4
Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }
The Bipartition is
𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }

Complete Bipartite Graph

A simple bipartite graph 𝐺 (𝑋, 𝑌) is complete if each vertex 𝑋 is adjacent

to all the vertices of 𝑌.

Here 𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 }

Subgraph

A graph 𝐻 is called subgraph of 𝐺 if 𝑉(𝐻) ⊆ 𝑉(𝐺), 𝐸(𝐻) ⊆ 𝐸(𝐺) and IH

is the restriction of 𝐼𝐺 to 𝐸(𝐻) [ie, 𝐼𝐻 (𝑒) = 𝐼𝐺 (𝑒) whenever 𝑒 ∈ 𝐸(𝐻).

5
Degrees of Vertices

The number of edges incident with vertex 𝑉 is called degree of a vertex

or valency of a vertex and it is denoted by 𝑑(𝑣).

Isomorphism of Graph

A graph isomorphism from a graph 𝐺 to a graph 𝐻 is a pair (𝜙, 𝜃), where

𝜙 ∶ 𝑉(𝐺) → 𝑉(𝐻) and 𝜃 ∶ 𝐸(𝐺) → 𝐸(𝐻) are bijection with a property that
𝐼𝐺 (𝑒) = {𝑢, 𝑣} and 𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}.

Walk

A walk in a graph 𝐺 is an alternative sequence 𝑊=

𝑣0 𝑣1 𝑒1 𝑣2 𝑒2 … 𝑣𝑛 𝑒𝑛 vertices and edges, beginning and ending with vertices where
𝑣0 is the origin and 𝑣𝑛 is the terminus of 𝑊.

𝑊 = 𝑣6 𝑒8 𝑣1 𝑒1 𝑣2 𝑒2 𝑣3 𝑒3 𝑣2 𝑒1 𝑣1

6
Closed Walk

A walk to begin and ends at the same vertex is called a closed walk. That
is, the walk 𝑊 is closed if 𝑣0 = 𝑣𝑛 .

Open Walk

If the origin of the walk and terminus of the walk are different vertices,
then it is called an open walk.

Trail

A walk is called a trail if all the edges in the walk are distinct.

Path

A walk is called a path if all the vertices are distinct.

Example:

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒6 𝑣1 → A trail

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 → A path

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 𝑒5 𝑣1 → A trail, but not a path

Euler’s Theorem

The sum of the degrees of the vertices of a graph is equal to the twice the
number of edges.

ie: ∑𝑛𝑖=1 𝑑(𝑣𝑖 ) = 2𝑚

7
Isomorphic Graph

𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}

A graph 𝐺1 = (𝑉1 , 𝐸1 ) is said to be isomorphic to graph 𝐺2 = (𝑉2 , 𝐸2 ) if

there is a one-to-one correspondence between the edge sets 𝐸1 and 𝐸2 in such a
way that if 𝑒1 is an edge with end vertices 𝑢1 and 𝑣1 in 𝐺1 then the corresponding
edge 𝑒2 in 𝐺2 has its end vertices 𝑢2 and 𝑣2 in 𝐺2 . This correspondence is called
a graph isomorphism.

Example:

𝐺=

𝐻=

ie: G and H are isomorphic.

Components

A connected component of a graph is a maximal connected subgraph. The

term is also used for maximal subgraph or subset of a graph 's vertices that have
some higher order of connectivity, including bi-connected components, tri-
connected components and strongly connected components.

Tree

A connected graph without cycles is called a tree.

Vertex Cut

Let 𝐺 be a connected graph. The set 𝑉՛ subset of 𝑉(𝐺) is called a Vertex

cut of 𝐺, if 𝐺 − 𝑉՛ is a disconnected graph.

8
Cut Vertex

If 𝑉՛ = {𝑣} is a Vertex cut of the connected Graph 𝐺, then the vertex 𝒗 is

called a Cut vertex.

Edge Cut

Let 𝐺 be a non-trivial connected graph with vertex set 𝑉 and let 𝑆 be a non-
empty subset of 𝑉 and 𝑆̅ = 𝑉 − 𝑆. Let 𝐸՛ = [𝑆, 𝑆̅] denote the set of all edges of 𝐺
that have one end vertex is 𝑆 and the other is 𝑆̅. Then 𝐺 − 𝐸՛ is a disconnected
graph and 𝐸՛ = [𝑆, 𝑆̅] is called an edge cut of 𝐺.

Cut Edge

If 𝐸՛ = {𝑒} is an edge cut of 𝐺 then 𝑒 is called a cut edge of 𝐺.

Block

A block is a Connected graph without any cut vertices.

Eg:

Graph 𝐺 Blocks of 𝐺

9
Chapter 2

PLANAR GRAPHS

Plane Graph

A plane graph is a graph drawn in the plane, such a way that any pair of
edges meet only at their end vertices.

Example:

Planar Graph

A planar graph is a graph which is isomorphic to a plane graph, ie: it can

be drawn as a plane graph.

A plane graph is a graph that can be drawn in the plane without any edge crossing.

10
Example of Planar graph:

Planar Representation

The pictorial representation of a planar graph as a plane graph is called a

planar representation.

Eg: Is Q3 shown below, planar?

The graph Q3

Planar representation of Q3 is:

11
Jordan Curve

A Jordan Curve in the plane is a continuous non-self-intersecting curve

where Origin and Terminals coincide.

Example:

Jordan Curves

Non-Jordan Curves

Remark

If J is a Jordan Curve in the plane, then the part of the plane enclosed by J
is called interior of J and is denoted by ‘int J’. We exclude from ‘int J’ the points
actually lying on J. Similarly, the part of the plane lying outside J is called the
exterior of J and is denoted by ‘ext J’.

Example:

Arc connecting point 𝑥 in int J with point 𝑦 in ext J.

12
Theorem

Let J be a Jordan Curve, if 𝑥 is a point in int J and 𝑦 is a point in ext J then

any line joining 𝑥 to 𝑦 must meet J at some point, ie: must cross J. this is called
Jordan Curve Theorem.

Boundary

The set of edges that bound a region is called its boundary.

Definition

A graph which is not planar is known as non-planar graph or a graph that

cannot be drawn in the plane without any edge crossing is known as non-planar
graph.

Theorem

K5 is nonplanar:

Every drawing of the complex graph K5 in the plane (or sphere) contains
at least one edge crossing.

Proof:

Label the vertices 0, 1, 2, 3, 4. By the Jordan Curve theorem any drawing of the
cycle (1, 2, 3, 4, 1) separates the plane into two regions. Consider the region with

13
vertex 0 in its interior as the ‘inside’ of the circle. By the Jordan Curve theorem,
the edges joining vertex 0 to each of its vertices 1, 2, 3 and 4 must also lie entirely
inside the cycle, as illustrated below.

Drawing most of the K5 in the plane

Moreover, each of the 3-cycles {0, 1, 2, 0}, {0, 2, 3, 0}, {0, 3, 4, 0} and {0, 4, 1, 0}
also separates the plane and hence the edges (2, 4) must also lie to the exterior of
the cycle {1, 2, 3, 4} as shown. It follows that the cycle formed by edges (2, 4),
(4, 0) and (0, 2) separates the vertices 1 and 3, again by Jordan Curve theorem.
Thus, it is impossible to draw edge (1, 3) without crossing an edge of that cycle.
So, it is proven that the drawing of the K5 in the plane contains at least one edge-
crossing.

Theorem

K33 is nonplanar:

Every drawing of the complete bipartite graph K33 in the plane (or sphere)
contains at least one edge crossing.

Proof:

Label the vertices of one partite set 0, 2, 4 and of the order 1, 3, 5. By the
Jordan Curve theorem, cycle {2, 3, 4, 5, 2} separates the plane into two regions,

14
and as in the previous proof (K5), we regard the region containing the vertex 0 as
the ‘inside’ of the cycle. By the Jordan Curve theorem, the edges joining vertex
0 to each of the vertices 3 and 5 lie entirely inside that cycle, and each of the cycle
{0, 3, 2, 5, 0} and {0, 3. 4, 5, 0} separates the plane, as illustrated below.

Drawing most of the K33 in the plane

Thus, there are 3 regions: the exterior of cycles {2, 3, 4, 5, 2} and the inside
of each of the other two cycles. It follows that no matter which region contains
vertex 1, there must be some even numbered vertex that is not in that region, and
hence the edge from vertex 1 to that even-numbered vertex would have to cross
some cycle edge.

Corollary

Subgraph of a planar graph is planar.

Definition

A plane graph partitions the plane into number of regions called faces.

Let G be plane graph. If x is a point on the plane which is not in G, ie: 𝑥 is not a
vertex of G or a point on any edge of G, then we define the faces of G containing
𝑥 to be the set of all points on the plane which can be reached from 𝑥 by a line
which does not cross any edge of G or go through any vertex of G.

15
The number of faces of a plane graph G denoted by 𝑓(𝑎) or simply 𝑓.

Each plane graph has exactly one unbounded face called the exterior face.

Here 𝑓(𝐺) = 4

Degree of faces

The degree 𝑑(𝑓) of a face 𝑓 is the number of edges with which it is

incident, that is the number of edges in the boundary of a face.

Cut edge being counted twice.

Eg:

𝑑(𝑓2 ) = 3
𝑑(𝑓1 ) = 4
𝑑(𝑓3 ) = 3

Theorem

A graph is planar if and only if each of its blocks is planar.

Proof:

If G is planar, then each of its blocks is planar since a subgraph of planar

graph is planar.

Conversely, suppose that each block of G is planar. We now use induction

on the number of blocks of G to prove the result. Without loss of generality, we
16
assume that G is connected. If G has only one block, then G itself is a block, and
hence G is planar.

Now suppose G has k planar blocks and that the result has been proved for
all connected graph having (k-1) planar blocks. Choose any end block B0 of G
and delete from G all the vertices of B0 except the unique cut vertex, say 𝑣0 of G
in B0. The resulting connected graph G` of G contains (k-1) planar blocks. Hence,
by the induction hypothesis G` is planar. Let G~` be plane embedded of G` such
that 𝑣0 belongs to the boundary of unbounded face, say 𝑓 `. Let B0~ be a plane
embedding of B0 in 𝑓 `, so that 𝑣0 is in the exterior face of B0~. Then G~` and B0~
is a plane embedding of G.

17
Chapter 3

EULER’S FORMULA

Theorems

Euler Formula:

For a connected plain graph 𝐺, 𝑛 − 𝑚 + 𝑓 = 2 where 𝑛, 𝑚, and 𝑓 denote

the number of vertices, edges and faces of 𝐺 respectively.

Proof:

We apply the induction on 𝑓.

If 𝑓 = 1 the 𝐺 is a tree and 𝑚 = 𝑛 − 1.

Hence 𝑛 − 𝑚 + 𝑓 = 2 and suppose that 𝐺 has 𝑓 faces.

Since 𝑓 ≥ 2, 𝐺 is not a tree and hence contains a cycle 𝐶. Let 𝑒 be an edge of 𝐶.

Then 𝑒 belongs to exactly 2 faces, say 𝑓1 and 𝑓2 and the deletion of 𝑒 from 𝐺
results in the formation of a single face from 𝑓1 and 𝑓2 . Also, since 𝑒 is not a cut
edge of 𝐺. 𝐺 − 𝑒 is connected.

Further the number of faces of 𝐺 − 𝑒 is 𝑓 − 1, number of edges in 𝐺 − 𝑒 is 𝑚 −

1 and number of vertices in 𝐺 − 𝑒 is 𝑛. So, applying induction to 𝐺 − 𝑒, we get
𝑛 − (𝑚 − 1) + (𝑓 − 1) = 2 and this implies that 𝑛 − 𝑚 + 𝑓 = 2. This
completes the proof of theorem.

Corollary 1

All plane embedding of a planar graph have the same number of faces.

18
Proof:

Since 𝑓 = 𝑚 − 𝑛 + 2 the number of faces depends only on 𝑛 and 𝑚 and not on

the particular embedding.

Corollary 2

If 𝐺 is a simple planar graph with at least 3 vertices, then 𝑚 ≤ 3𝑛 − 6.

Proof:

Without the generality we can assume that 𝐺 is a simple connected plane graph.
Since 𝐺 is simple and 𝑛 ≥ 3, each face of 𝐺 has degree at least 3. Hence if 𝑓
denote the set of faces of 𝐺 ∑𝑓𝜖𝐹 𝑑(𝑓) ≥ 3𝑓. But ∑𝑓𝜖𝐹 𝑑(𝑓) = 2𝑚.

2𝑚
Consequently 2𝑚 ≥ 3𝑓 so that 𝑓 ≤ .
3

2𝑚 2m
By the Euler formula 𝑚 = 𝑛 + 𝑓 − 2 now 𝑓 ≤ implies m ≤ n + ( ) − 2.
3 3

This gives. 𝑚 ≤ 3𝑛 − 6.

19
Chapter 4

DUAL OF A PLANE GRAPH

Definition

Let G be a plane graph. One can form out of G a new graph H in the
following way corresponding to each face f(g), take the vertex f* and
corresponding to each edge e(g), take an edge e*. Then edge e* joins vertices f*
and g* in H iff edge e is common to the boundaries of faces f and g in G. The
graph H is then called dual of G.

Example:

Plane graph and its Dual

20
 CONCLUSION

In this project we discussed the topic planar graph in graph theory.

We discussed about Euler formula and verified that some graphs are planar, and
some are non-planar. A related important property of planar graphs, maps and
triangulations is that they can be enumerated very nicely.

We also discussed about duality of a graph.in mathematical discipline of graph

theory, the dual graph of a plane graph G is a graph that has a vertex of each face
of G .it has many applications in mathematical and computational study.

In fact, graph theory is being used in our so many routine activities. For eg; using
GPS or google maps to determine a route based on used settings.

21
 BIBLIOGRAPHY

1) R. Balakrishnan and K. Ranganathan, A textbook of Graph Theory,

University Text

2) J. A. Bondy and U. S. R. Murty, Graph Theory with Great Britain: The

Macmillian Press Ltd. 1976

22
PLANARITY IN GRAPH THEORY

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
KEERTHANA M
DB18CMSR05
Under the guidance of
Mrs. PRIJA V

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu
June 2021
 CERTIFICATE

Certified that this project ‘Planar Graph’ is a bona fide project of Keerthana M
carried out the project work under my supervision.

Mrs.
Mrs. Riya
Riya Baby
Baby Mrs. Prija V
Head
Head of
of Department
Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu

I
 DECLARATION

I Keerthana M hereby declare that the project ‘Planar Graph’ is an original

record of studies and bona fide project carried out by me during the period of
2018 – 2021 under the guidance of Mrs. Prija V, Department of Mathematics,
Don Bosco Arts and Science College, Angadikkadavu and has not submitted by
me elsewhere for the award of my degree, diploma, title, or recognition, before.

Keerthana M

DBI8CMSR05

Department of Mathematics

Don Bosco Arts and Science College,

Angadikkadavu

II
 ACKNOWLEDGEMENT

Introduction is the proper place to begin. But first I bow my head before
the Almighty who is always with me. Also, I must express my deepest gratitude
to people along the way.

No words can adequately express the sense of gratitude; still, I try to

express my heartfelt thanks through words. At the outset, I am deeply indebted to
my project supervisor Mrs. Prija V, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement, and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the principal
and the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the
successful completion of this project. It is my pleasure and duty to thank each and
every one of them who walked with me.

My greatest debt is always, to God Almighty.

III
 CONTENTS

Sl No Title Page No

1 Introduction 1

2 Chapter 1 - Basic Concepts 2-9

3 Chapter 2 – Planar Graph 10 - 17

4 Chapter 3 – Euler’s Formula 18 – 19

5 Chapter 4 – Dual of a Plane Graph 20

6 Conclusion 21

7 Bibliography 22

IV
 INTRODUCTION

In recent years, Graph Theory has established itself as an important

mathematical tool in a wide variety of subjects, ranging from Operational
Research and Chemistry to Genetics and Linguistics, and from Electrical
Engineering and Geography to Sociology and Architecture. At the same time, it
has also emerged as a worthwhile mathematical discipline in its own right.

A great mathematician, Euler become the Father of Graph Theory, when

in 1736, he solved a famous unsolved problem of his days, called Konigsberg
Bridge Problem. This is today, called as the First Problem of the Graph theory.
This problem leads to the concept of the planar graph as well as Eulerian Graphs,
while planar graphs were introduced for practical reasons, they pose many
remarkable mathematical properties. In 1936, the psychologist Lewin used planar
graphs to represent the life space of an individual.

1
Chapter 1

BASIC CONCEPTS

Graph

A graph is an ordered triple 𝐺 = {𝑉(𝐺), 𝐸(𝐺), 𝐼𝐺 } where 𝑉(𝐺) is a non-

empty set, 𝐸(𝐺) is a set disjoint from 𝑉(𝐺) and 𝐼(𝐺) is an incidence map that
associates each element of 𝐸(𝐺) and unordered pair of elements of 𝑉(𝐺). The
elements of 𝑉(𝐺) are called vertices (or nodes or points) of 𝐺 and the element
of 𝐸(𝐺) are called edges or lines of 𝐺.

Example:

Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 }

𝐸(𝐺) = {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 }
𝐼𝐺 (𝑒1 ) = {𝑣1 , 𝑣2 } or {𝑣2 , 𝑣1 }
I𝐺 (𝑒2 ) = {𝑣2 , 𝑣3 } or {𝑣3 , 𝑣2 }
𝐼𝐺 (𝑒3 ) = {𝑣3 , 𝑣4 } or {𝑣4 , 𝑣3 }
𝐼𝐺 (𝑒4 ) = {𝑣4 , 𝑣1 } or {𝑣1 , 𝑣4 }

Multiple edges

A set of two or more edges of a graph 𝐺 is called multiple edges or parallel

edges if they have the same end vertices.

2
Loop

An edge for which the two end vertices are same is called a loop.

Here {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 } form the parallel edges.

𝑒7 is the Loop.
Simple Graph

A graph is simple if it has no loops and no multiple edges.

Finite & Infinite Graphs

A graph is called finite if both 𝑉(𝐺) & 𝐸(𝐺) are finite. A graph that is not
finite is called infinite graph.

Adjacent Vertices

Two vertices 𝑢 and 𝑣 are said to be adjacent vertices if and only if there is
an edge with 𝑢 and 𝑣 as its end vertices.

3
Adjacent Edges

Two distinct edges are said to be adjacent edges if and only if they have a
continuous end vertex.

Complete Graph

A simple graph 𝐺 is said to be complete if every pair of distinct vertices of

𝐺 are adjacent in 𝐺. A complete graph with n vertices is denoted by 𝐾𝑛 .

Bipartite Graph

A graph is bipartite if its vertex set can be partitioned into two non-empty
subsets 𝑋 and 𝑌 such that each edge of 𝐺 has one end in 𝑋 and the other in 𝑌. The
pair (𝑋, 𝑌) is called a bipartition of the bipartite graph 𝐺. The bipartite graph 𝐺
with bipartition (𝑋, 𝑌) denoted by 𝐺 (𝑋, 𝑌).

4
Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }
The Bipartition is
𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }

Complete Bipartite Graph

A simple bipartite graph 𝐺 (𝑋, 𝑌) is complete if each vertex 𝑋 is adjacent

to all the vertices of 𝑌.

Here 𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 }

Subgraph

A graph 𝐻 is called subgraph of 𝐺 if 𝑉(𝐻) ⊆ 𝑉(𝐺), 𝐸(𝐻) ⊆ 𝐸(𝐺) and IH

is the restriction of 𝐼𝐺 to 𝐸(𝐻) [ie, 𝐼𝐻 (𝑒) = 𝐼𝐺 (𝑒) whenever 𝑒 ∈ 𝐸(𝐻).

5
Degrees of Vertices

The number of edges incident with vertex 𝑉 is called degree of a vertex

or valency of a vertex and it is denoted by 𝑑(𝑣).

Isomorphism of Graph

A graph isomorphism from a graph 𝐺 to a graph 𝐻 is a pair (𝜙, 𝜃), where

𝜙 ∶ 𝑉(𝐺) → 𝑉(𝐻) and 𝜃 ∶ 𝐸(𝐺) → 𝐸(𝐻) are bijection with a property that
𝐼𝐺 (𝑒) = {𝑢, 𝑣} and 𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}.

Walk

A walk in a graph 𝐺 is an alternative sequence 𝑊=

𝑣0 𝑣1 𝑒1 𝑣2 𝑒2 … 𝑣𝑛 𝑒𝑛 vertices and edges, beginning and ending with vertices where
𝑣0 is the origin and 𝑣𝑛 is the terminus of 𝑊.

𝑊 = 𝑣6 𝑒8 𝑣1 𝑒1 𝑣2 𝑒2 𝑣3 𝑒3 𝑣2 𝑒1 𝑣1

6
Closed Walk

A walk to begin and ends at the same vertex is called a closed walk. That
is, the walk 𝑊 is closed if 𝑣0 = 𝑣𝑛 .

Open Walk

If the origin of the walk and terminus of the walk are different vertices,
then it is called an open walk.

Trail

A walk is called a trail if all the edges in the walk are distinct.

Path

A walk is called a path if all the vertices are distinct.

Example:

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒6 𝑣1 → A trail

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 → A path

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 𝑒5 𝑣1 → A trail, but not a path

Euler’s Theorem

The sum of the degrees of the vertices of a graph is equal to the twice the
number of edges.

ie: ∑𝑛𝑖=1 𝑑(𝑣𝑖 ) = 2𝑚

7
Isomorphic Graph

𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}

A graph 𝐺1 = (𝑉1 , 𝐸1 ) is said to be isomorphic to graph 𝐺2 = (𝑉2 , 𝐸2 ) if

there is a one-to-one correspondence between the edge sets 𝐸1 and 𝐸2 in such a
way that if 𝑒1 is an edge with end vertices 𝑢1 and 𝑣1 in 𝐺1 then the corresponding
edge 𝑒2 in 𝐺2 has its end vertices 𝑢2 and 𝑣2 in 𝐺2 . This correspondence is called
a graph isomorphism.

Example:

𝐺=

𝐻=

ie: G and H are isomorphic.

Components

A connected component of a graph is a maximal connected subgraph. The

term is also used for maximal subgraph or subset of a graph 's vertices that have
some higher order of connectivity, including bi-connected components, tri-
connected components and strongly connected components.

Tree

A connected graph without cycles is called a tree.

Vertex Cut

Let 𝐺 be a connected graph. The set 𝑉՛ subset of 𝑉(𝐺) is called a Vertex

cut of 𝐺, if 𝐺 − 𝑉՛ is a disconnected graph.

8
Cut Vertex

If 𝑉՛ = {𝑣} is a Vertex cut of the connected Graph 𝐺, then the vertex 𝒗 is

called a Cut vertex.

Edge Cut

Let 𝐺 be a non-trivial connected graph with vertex set 𝑉 and let 𝑆 be a non-
empty subset of 𝑉 and 𝑆̅ = 𝑉 − 𝑆. Let 𝐸՛ = [𝑆, 𝑆̅] denote the set of all edges of 𝐺
that have one end vertex is 𝑆 and the other is 𝑆̅. Then 𝐺 − 𝐸՛ is a disconnected
graph and 𝐸՛ = [𝑆, 𝑆̅] is called an edge cut of 𝐺.

Cut Edge

If 𝐸՛ = {𝑒} is an edge cut of 𝐺 then 𝑒 is called a cut edge of 𝐺.

Block

A block is a Connected graph without any cut vertices.

Eg:

Graph 𝐺 Blocks of 𝐺

9
Chapter 2

PLANAR GRAPHS

Plane Graph

A plane graph is a graph drawn in the plane, such a way that any pair of
edges meet only at their end vertices.

Example:

Planar Graph

A planar graph is a graph which is isomorphic to a plane graph, ie: it can

be drawn as a plane graph.

A plane graph is a graph that can be drawn in the plane without any edge crossing.

10
Example of Planar graph:

Planar Representation

The pictorial representation of a planar graph as a plane graph is called a

planar representation.

Eg: Is Q3 shown below, planar?

The graph Q3

Planar representation of Q3 is:

11
Jordan Curve

A Jordan Curve in the plane is a continuous non-self-intersecting curve

where Origin and Terminals coincide.

Example:

Jordan Curves

Non-Jordan Curves

Remark

If J is a Jordan Curve in the plane, then the part of the plane enclosed by J
is called interior of J and is denoted by ‘int J’. We exclude from ‘int J’ the points
actually lying on J. Similarly, the part of the plane lying outside J is called the
exterior of J and is denoted by ‘ext J’.

Example:

Arc connecting point 𝑥 in int J with point 𝑦 in ext J.

12
Theorem

Let J be a Jordan Curve, if 𝑥 is a point in int J and 𝑦 is a point in ext J then

any line joining 𝑥 to 𝑦 must meet J at some point, ie: must cross J. this is called
Jordan Curve Theorem.

Boundary

The set of edges that bound a region is called its boundary.

Definition

A graph which is not planar is known as non-planar graph or a graph that

cannot be drawn in the plane without any edge crossing is known as non-planar
graph.

Theorem

K5 is nonplanar:

Every drawing of the complex graph K5 in the plane (or sphere) contains
at least one edge crossing.

Proof:

Label the vertices 0, 1, 2, 3, 4. By the Jordan Curve theorem any drawing of the
cycle (1, 2, 3, 4, 1) separates the plane into two regions. Consider the region with

13
vertex 0 in its interior as the ‘inside’ of the circle. By the Jordan Curve theorem,
the edges joining vertex 0 to each of its vertices 1, 2, 3 and 4 must also lie entirely
inside the cycle, as illustrated below.

Drawing most of the K5 in the plane

Moreover, each of the 3-cycles {0, 1, 2, 0}, {0, 2, 3, 0}, {0, 3, 4, 0} and {0, 4, 1, 0}
also separates the plane and hence the edges (2, 4) must also lie to the exterior of
the cycle {1, 2, 3, 4} as shown. It follows that the cycle formed by edges (2, 4),
(4, 0) and (0, 2) separates the vertices 1 and 3, again by Jordan Curve theorem.
Thus, it is impossible to draw edge (1, 3) without crossing an edge of that cycle.
So, it is proven that the drawing of the K5 in the plane contains at least one edge-
crossing.

Theorem

K33 is nonplanar:

Every drawing of the complete bipartite graph K33 in the plane (or sphere)
contains at least one edge crossing.

Proof:

Label the vertices of one partite set 0, 2, 4 and of the order 1, 3, 5. By the
Jordan Curve theorem, cycle {2, 3, 4, 5, 2} separates the plane into two regions,

14
and as in the previous proof (K5), we regard the region containing the vertex 0 as
the ‘inside’ of the cycle. By the Jordan Curve theorem, the edges joining vertex
0 to each of the vertices 3 and 5 lie entirely inside that cycle, and each of the cycle
{0, 3, 2, 5, 0} and {0, 3. 4, 5, 0} separates the plane, as illustrated below.

Drawing most of the K33 in the plane

Thus, there are 3 regions: the exterior of cycles {2, 3, 4, 5, 2} and the inside
of each of the other two cycles. It follows that no matter which region contains
vertex 1, there must be some even numbered vertex that is not in that region, and
hence the edge from vertex 1 to that even-numbered vertex would have to cross
some cycle edge.

Corollary

Subgraph of a planar graph is planar.

Definition

A plane graph partitions the plane into number of regions called faces.

Let G be plane graph. If x is a point on the plane which is not in G, ie: 𝑥 is not a
vertex of G or a point on any edge of G, then we define the faces of G containing
𝑥 to be the set of all points on the plane which can be reached from 𝑥 by a line
which does not cross any edge of G or go through any vertex of G.

15
The number of faces of a plane graph G denoted by 𝑓(𝑎) or simply 𝑓.

Each plane graph has exactly one unbounded face called the exterior face.

Here 𝑓(𝐺) = 4

Degree of faces

The degree 𝑑(𝑓) of a face 𝑓 is the number of edges with which it is

incident, that is the number of edges in the boundary of a face.

Cut edge being counted twice.

Eg:

𝑑(𝑓2 ) = 3
𝑑(𝑓1 ) = 4
𝑑(𝑓3 ) = 3

Theorem

A graph is planar if and only if each of its blocks is planar.

Proof:

If G is planar, then each of its blocks is planar since a subgraph of planar

graph is planar.

Conversely, suppose that each block of G is planar. We now use induction

on the number of blocks of G to prove the result. Without loss of generality, we
16
assume that G is connected. If G has only one block, then G itself is a block, and
hence G is planar.

Now suppose G has k planar blocks and that the result has been proved for
all connected graph having (k-1) planar blocks. Choose any end block B0 of G
and delete from G all the vertices of B0 except the unique cut vertex, say 𝑣0 of G
in B0. The resulting connected graph G` of G contains (k-1) planar blocks. Hence,
by the induction hypothesis G` is planar. Let G~` be plane embedded of G` such
that 𝑣0 belongs to the boundary of unbounded face, say 𝑓 `. Let B0~ be a plane
embedding of B0 in 𝑓 `, so that 𝑣0 is in the exterior face of B0~. Then G~` and B0~
is a plane embedding of G.

17
Chapter 3

EULER’S FORMULA

Theorems

Euler Formula:

For a connected plain graph 𝐺, 𝑛 − 𝑚 + 𝑓 = 2 where 𝑛, 𝑚, and 𝑓 denote

the number of vertices, edges and faces of 𝐺 respectively.

Proof:

We apply the induction on 𝑓.

If 𝑓 = 1 the 𝐺 is a tree and 𝑚 = 𝑛 − 1.

Hence 𝑛 − 𝑚 + 𝑓 = 2 and suppose that 𝐺 has 𝑓 faces.

Since 𝑓 ≥ 2, 𝐺 is not a tree and hence contains a cycle 𝐶. Let 𝑒 be an edge of 𝐶.

Then 𝑒 belongs to exactly 2 faces, say 𝑓1 and 𝑓2 and the deletion of 𝑒 from 𝐺
results in the formation of a single face from 𝑓1 and 𝑓2 . Also, since 𝑒 is not a cut
edge of 𝐺. 𝐺 − 𝑒 is connected.

Further the number of faces of 𝐺 − 𝑒 is 𝑓 − 1, number of edges in 𝐺 − 𝑒 is 𝑚 −

1 and number of vertices in 𝐺 − 𝑒 is 𝑛. So, applying induction to 𝐺 − 𝑒, we get
𝑛 − (𝑚 − 1) + (𝑓 − 1) = 2 and this implies that 𝑛 − 𝑚 + 𝑓 = 2. This
completes the proof of theorem.

Corollary 1

All plane embedding of a planar graph have the same number of faces.

18
Proof:

Since 𝑓 = 𝑚 − 𝑛 + 2 the number of faces depends only on 𝑛 and 𝑚 and not on

the particular embedding.

Corollary 2

If 𝐺 is a simple planar graph with at least 3 vertices, then 𝑚 ≤ 3𝑛 − 6.

Proof:

Without the generality we can assume that 𝐺 is a simple connected plane graph.
Since 𝐺 is simple and 𝑛 ≥ 3, each face of 𝐺 has degree at least 3. Hence if 𝑓
denote the set of faces of 𝐺 ∑𝑓𝜖𝐹 𝑑(𝑓) ≥ 3𝑓. But ∑𝑓𝜖𝐹 𝑑(𝑓) = 2𝑚.

2𝑚
Consequently 2𝑚 ≥ 3𝑓 so that 𝑓 ≤ .
3

2𝑚 2m
By the Euler formula 𝑚 = 𝑛 + 𝑓 − 2 now 𝑓 ≤ implies m ≤ n + ( ) − 2.
3 3

This gives. 𝑚 ≤ 3𝑛 − 6.

19
Chapter 4

DUAL OF A PLANE GRAPH

Definition

Let G be a plane graph. One can form out of G a new graph H in the
following way corresponding to each face f(g), take the vertex f* and
corresponding to each edge e(g), take an edge e*. Then edge e* joins vertices f*
and g* in H iff edge e is common to the boundaries of faces f and g in G. The
graph H is then called dual of G.

Example:

Plane graph and its Dual

20
 CONCLUSION

In this project we discussed the topic planar graph in graph theory.

We discussed about Euler formula and verified that some graphs are planar, and
some are non-planar. A related important property of planar graphs, maps and
triangulations is that they can be enumerated very nicely.

We also discussed about duality of a graph.in mathematical discipline of graph

theory, the dual graph of a plane graph G is a graph that has a vertex of each face
of G .it has many applications in mathematical and computational study.

In fact, graph theory is being used in our so many routine activities. For eg; using
GPS or google maps to determine a route based on used settings.

21
 BIBLIOGRAPHY

1) R. Balakrishnan and K. Ranganathan, A textbook of Graph Theory,

University Text

2) J. A. Bondy and U. S. R. Murty, Graph Theory with Great Britain: The

Macmillian Press Ltd. 1976

22
DIRECT PRODUCTS AND FINITELY GENERATED ABELIAN
GROUPS

Project report submitted to

The Kannur University

for the award of the degree

of

Bachelor of Science

by

MEGHA A

DB18CMSR06

Under the guidance of

MS. Sneha P Sebastian

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report ‘DIRECT PRODUCTS AND FINITELY

GENERATED ABELIAN GROUPS’ is the bonafide project of MEGHA A who
carried out the project under my supervision.

Mrs.Riya Baby Ms.Sneha P Sebastian

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, MEGHA A, hereby declare that this project report entitled ‘DIRECT PRODUCTS
AND FINITELY GENERATED ABELIAN GROUPS’ is an original record of studies and
bonafide project carried out by me during the period from November 2019 to March 2020, under
the guidance of Ms.Sneha P Sebastian, Department of Mathematics, Don Bosco Arts and
Science College, Angadikadavu and has not been submitted by me elsewhere for the award of
any degree, diploma, title or recognition, before.

MEGHA A
Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great help to me
during the course of my dissertation. First and foremost I thank the almighty, for his blessing and
protection during the period of this work. I express my thanks to Dr. Fr. Francis Karakkatt,
Principal, for support in the completion of this dissertation. I express my gratitude to Ms.Sneha P
Sebastian, my project supervisor, for the constant encouragement, valuable guidance and timely
corrections, which made this work a success.

I am also indebted to all my classmates and friends who supported me throughout the
study. I would like to express my thanks to my parents and dear ones for their constant
encouragement and support. I also thank all those who helped me directly or indirectly to
complete this project.

MEGHA A
 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 preliminary 2–3
03 Chapter 1 4 – 12
04 Chapter 2 13 – 17
05 Conclusion 18
06 Bibliography 19
 INTRODUCTION

In mathematics, a group is a set equipped with a binary operation that combines any two
elements to form a third element in such a way that the three conditions called group axioms are
satisfied , namely associativity , identity and invertability.

Let us take a moment to review our present stockpile of groups. Starting with finite
groups, we have the cyclic group ℤ𝑛 ,the symmetric group 𝑆𝑛 , and the alternating group 𝐴𝑛 for
each positive integer n. We also have the dihedral group 𝐷𝑛 and klein 4-group . Of course we
know that subgroups of these groups exists. Turning to infinite groups , we have ℤ, ℝ, ℂ under
addition , and their non zero elements under multiplication we also have the group 𝑆𝐴 of all
permutation of an infinite set 𝐴 , as well as various groups formed from matrices .

One purpose of this section is to show a way to use known groups as building blocks to
form more groups. Given two groups 𝐺 and 𝐻, it is possible to construct a new group from the
cartesian product of 𝐺 and 𝐻 . Conversely , given a large group , it is sometimes possible to
decompose the group ; that is , a group is sometimes isomorphic to the direct product of two
smaller groups. Rather than studying a large group , it is often easier to study the component
group of that group.

1
 PRELIMINARY

Groups : A non empty set 𝐺 together with an operation ∗ is said to be a group , denote by
(𝐺 ,∗) , if it satisfy the following axioms.

• Closure property
• Associative property
• Existence of identity
• Existence of inverse

Abelian group

A group (𝐺 ,∗) is said to be abelian if it satisfies commutative law .

Finite group

If the underlying set G of the group (𝐺 ,∗) consist of finite number of elements , then the
group is finite group .

Infinite group

A group that is not finite is an infinite group .

Order of a group : The number of elements in a finite group is called the order of the group ,
denoted by 𝑂(𝐺 ) .

Example

Show that the set of integers ℤ is a group with respect to the operation of addition of
integers.

ℤ = {… … … . −3, −2, −1,0,1,2,3, … … … }

Since the addition of two integers gives an integer , it satisfy closure property .

2
If 𝑎, 𝑏, 𝑐 𝜖 ℤ then the (𝑎 + 𝑏) + 𝑐 = 𝑎 + (𝑏 + 𝑐) , hence associativity holds .

There is a number 0 𝜖 ℤ such that 0 + 𝑎 = 𝑎 + 0 , hence identity exists

If 𝑎 𝜖 ℤ then there exists – 𝑎 𝜖 ℤ , such that −𝑎 + 𝑎 = 0 = 𝑎 + −𝑎

Therefore inverse exist .

Therefore ℤ is a group under addition .

Subgroup

A subset 𝐻 of 𝐺 is said to be a subgroup of 𝐺 if 𝐻 itself is a group under the same operation in

𝐺.

There are two different types of group structure of order 4 .

ℤ4 = { 0,1,2,3}

Klein 4 – group , 𝑉 = {𝑒, 𝑎, 𝑏, 𝑐}

Cyclic group

A group 𝐺 is cyclic if there is some element ‘𝑎’ in 𝐺 that generate 𝐺. And the element ‘𝑎’ is
called generator of 𝐺.

Group Homomorphism

A function Ѱ: 𝐺 → 𝐺′ is a group homomorphism ( or simply homomorphism ).

If Ѱ(𝑎𝑏) = Ѱ(𝑎) Ѱ(𝑏) hold for all 𝑎 , 𝑏 ∈ 𝐺 , is called homomorphism property .

Isomorphism

A one to one and onto homomorphism Ѱ: 𝐺 → 𝐺′ is called an isomorphism .

3
 CHAPTER – 1

DIRECT PRODUCT OF GROUPS

Definition

The Cartesian product of sets 𝑆, 𝑆2 , … … . , 𝑆𝑛 is the set of all ordered n-tuples (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ),
where 𝑎𝑖 ∈ 𝑆𝑖 for 𝑖 = 1,2,3, … … . , 𝑛. The Cartesian product is denoted by either

𝑛
𝑆1 × 𝑆2 × … … .× 𝑆𝑛 or by Π𝑖=1 𝑆𝑖 .

Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups and let us use multiplicative notation for all the group operations.

If we consider 𝐺𝑖 as a set , 𝑖 = 1,2, … … . 𝑛 . we have the products 𝐺1 × 𝐺2 × … … . ,× 𝐺𝑛 we

𝑛 𝑛
denote it by Π𝑖=1 𝐺𝑖 . This product is called direct-product of groups. We can make Π𝑖=1 𝐺𝑖 into a
group by means of a binary operation of multiplication by components.

Theorem
𝑛
Let 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 be groups. For (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) and (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) in Π𝑖=1 𝐺𝑖 define ;

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

𝑛
Then Π𝑖=1 𝐺𝑖 is a group.

4
Proof

We have ,

𝑛
Π𝑖=1 𝐺𝑖 = {(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∶ 𝑎𝑖 ∈ 𝐺𝑖 }

(1) Closure property

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

And we have ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ) = (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 )

Here 𝑎𝑖 ∈ 𝐺𝑖 and 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

∵ 𝐺𝑖 is a group , 𝑎𝑖 𝑏𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛

𝑛
⇒ (𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ) ∈ Π𝑖=1 𝐺𝑖

𝑛
i.e. Π𝑖=1 𝐺𝑖 is closed under the binary operation.

(2) Associativity
𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ), (𝑏1 , 𝑏2 , … … . , 𝑏𝑛 ), (𝑐1, 𝑐2 , … … . , 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

We have,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

𝑛
= (𝑎1 𝑏1 𝑐1 , 𝑎2 𝑏2 𝑐2 , … … . , 𝑎𝑛 𝑏𝑛 𝑐𝑛 ) ∈ Π𝑖=1 𝐺𝑖

[(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )

= [𝑎1 𝑏1 , 𝑎2 𝑏2 , … … . , 𝑎𝑛 𝑏𝑛 ](𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )
= [(𝑎1 𝑏1 )𝑐1 , (𝑎2 𝑏2 )𝑐2, … … . , (𝑎𝑛 𝑏𝑛 )𝑐𝑛 ]
= [𝑎1 (𝑏1 𝑐1 ), 𝑎2 (𝑏2 𝑐2), … … . , 𝑎𝑛 (𝑏𝑛 𝑐𝑛 )]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[𝑏1 𝑐1, 𝑏2 𝑐2 , … … . , 𝑏𝑛 𝑐𝑛 ]
= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )[(𝑏1 , 𝑏2 , … … . , 𝑏𝑛 )(𝑐1 , 𝑐2 , … … . , 𝑐𝑛 )]

5
Hence associativity holds.

(3) Existence of identity

If 𝑒𝑖 is the identity element in 𝐺𝑖 .

Then,

𝑛
(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Also for ,

𝑛
(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖 ,

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) = (𝑎1 𝑒1 , 𝑎2 𝑒2 , … … . , 𝑎𝑛 𝑒𝑛 )

= (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )

𝑛
∴ (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ) is the identity element ‘𝑒’ in Π𝑖=1 𝐺𝑖

(4) Existence of inverse

𝑛
Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ Π𝑖=1 𝐺𝑖

Here 𝑎𝑖 ∈ 𝐺𝑖 for 𝑖 = 1,2, … … . , 𝑛.

Since 𝐺𝑖 is a group ,

∃ an inverse element 𝑎𝑖−1 in 𝐺𝑖 : 𝑎𝑖 𝑎𝑖−1 = 𝑒𝑖 𝑖 = 1,2, … … . , 𝑛

𝑛
Clearly, (𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) ∈ Π𝑖=1 𝐺𝑖 &

(𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )(𝑎1−1 , 𝑎2−1 , … … . , 𝑎𝑛−1 ) = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 )

𝑛
Hence Π𝑖=1 𝐺𝑖 is a group.

6
Note
𝑛
If the operation of each 𝐺𝑖 is a commutative. We sometimes use additive notation in Π𝑖=1 𝐺𝑖 and
𝑛
refer to Π𝑖=1 𝐺𝑖 as the direct sum of the group 𝐺𝑖 . The notation ⨁𝑛𝑖=1 𝐺𝑖 , especially with abelian
groups with operation +.

The direct sum of abelian groups 𝐺1 , 𝐺2 , … … . , 𝐺𝑛 may be written 𝐺1 ⨁𝐺2 ⨁ … … ⨁𝐺𝑛

• Direct product of abelian group is abelian

Example

Q. Check whether ℤ2 × ℤ3 is cyclic or not.

ℤ2 = {0,1}

ℤ3 = {0,1,2}

ℤ2 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2)}

Consider,

1(1,1) = (1,1)

2(1,1) = (1,1) + (1,1) = (0,2)

3(1,1) = (1,1) + (1,1) + (1,1) = (1,0)

4(1,1) = 3(1,1) + (1,1) = (1,0) + (1,1) = (0,1)

5(1,1) = 4(1,1) + (1,1) = (0,1) + (1,1) = (1,2)

6(1,1) = 5(1,1) + (1,1) = (1,2) + (1,1) = (0,0)

∴ (1,1) is a generator of ℤ2 × ℤ3

∴ ℤ2 × ℤ3 is a cyclic group generated by (1,1).

7
Q. Check whether ℤ3 × ℤ3 is cyclic or not.

ℤ3 = {0,1,2}

ℤ3 × ℤ3 = {(0,0), (0,1), (0,2), (1,0), (1,1), (1,2), (2,0), (2,1), (2,2)}

1(0,1) = (0,1)

2(0,1) = (0,2)

3(0,1) = (0,3) = (0,0) ∴ order (0,1) = 3

1(0,2) = (0,2)

2(0,2) = (0,4) = (0,1)

3(0,2) = (0,6) = (0,0) ∴ order (0,2) = 3

Every element added to itself three times gives the identity. Thus no element can
generate the group. Hence ℤ3 × ℤ3 is not cyclic.

similarly ℤ𝑚 × ℤ𝑚 is not cyclic for any 𝑚.

Theorem

The group ℤ𝑚 × ℤ𝑛 is cyclic and is isomorphic to ℤ𝑚𝑛 if and only if 𝑚 and 𝑛 are relatively
prime, that is, the gcd of 𝑚 and 𝑛 is 1.

Proof

Suppose ℤ𝑚 × ℤ𝑛 is cyclic and isomorphic to ℤ𝑚𝑛 .

To show that 𝑚 and 𝑛 are relatively prime.

Suppose not, let d be the 𝑔𝑐𝑑 of 𝑚 and 𝑛.

So that 𝑑 > 1

8
 𝑚𝑛
Consider , which is an integer since 𝑑|𝑚 and 𝑑|𝑛
𝑑

𝑚𝑛
Let (𝑟, 𝑠) be an arbitrary element of ℤ𝑚 × ℤ𝑛 , add (𝑟, 𝑠) repeatedly times
𝑑

(𝑟, 𝑠) + (𝑟, 𝑠)+, … … . , +(𝑟, 𝑠) } 𝑚𝑛

𝑑
𝑡𝑖𝑚𝑒𝑠 = (0,0)

∴ no element of ℤ𝑚 × ℤ𝑛 having order 𝑚𝑛. ∴ no element of ℤ𝑚 × ℤ𝑛 can generate ℤ𝑚 × ℤ𝑛

which is not possible. ∵ ℤ𝑚 × ℤ𝑛 is cyclic. Hence 𝑔𝑐𝑑(𝑚, 𝑛) = 1.

i.e. 𝑚 and 𝑛 are relatively prime.

Conversely, suppose 𝑚 and 𝑛 are relatively prime, i.e. gcd(𝑚, 𝑛) = 1

To show that ℤ𝑚 × ℤ𝑛 is cyclic.

If ℤ𝑚 × ℤ𝑛 is cyclic, then it is isomorphic to ℤ𝑚𝑛 , ∵ ℤ𝑚 × ℤ𝑛 has 𝑚𝑛 elements.

Consider the cyclic subgroup of ℤ𝑚 × ℤ𝑛 generated by the element (1,1).The order of this
cyclic subgroup is the smallest power of (1,1),that gives the identity (0,0). Here taking a power
of (1,1) in our additive notation will involve adding (1,1) to itself repeatedly.

Consider (1,1) + (1,1)+, … … . , +(1,1)

If we add first coordinates 𝑚 times , we get zero.

∴ order of first coordinate = 𝑚.

Similarly , Order of second coordinate = 𝑛.

The two coordinates together become zero. If we add them 𝑙𝑐𝑚(𝑚, 𝑛) times.

∵ gcd(𝑚, 𝑛) = 1, We get the 𝑙𝑐𝑚 = 𝑚𝑛.

i.e. (1,1) generates a cyclic subgroup of ℤ𝑚 × ℤ𝑛 of order 𝑚𝑛 , which is the order of the whole
group.

⇒ ℤ𝑚 × ℤ𝑛 =< (1,1) >

9
 ⇒ ℤ𝑚 × ℤ𝑛 is cyclic.

Corollary

The group ⨅𝑛𝑖=1 ℤ𝑚𝑖 is cyclic and isomorphic to ℤ𝑚1 𝑚2 …….𝑚𝑛 if and only if the numbers 𝑚𝑖 for
𝑖 = 1,2, … … . , 𝑛 are such that the 𝑔𝑐𝑑 of any two of them is 1.

Example

If 𝑛 is written as a product of powers of distinct prime numbers , as in ,

𝑛 = (𝑝1 )𝑛1 . (𝑝2 )𝑛2 … … . (𝑝𝑛 )𝑛𝑟

Then ℤ𝑛 is isomorphic to ℤ(𝑝1 )𝑛1 × ℤ(𝑝2 )𝑛2 × … … .× ℤ(𝑝𝑟 )𝑛𝑟 .

In particular , ℤ72 is isomorphic to ℤ8 × ℤ9 .

Consider set of integers ℤ, cyclic subgroup of ℤ is of the form 𝑛ℤ , 𝑛 ∈ ℤ. Consider 2ℤ and 3ℤ ,

then < 2 > ∩ < 3 > = < 6 >

∴ if we take 𝑟ℤ , 𝑠ℤ of ℤ , then the 𝑙𝑐𝑚(𝑟, 𝑠) =generator of < 𝑟 > ∩ < 𝑠 >

Using this we can define the 𝑙𝑐𝑚 of the positive integers.

Definition

Let 𝑟1 , 𝑟2 , … … . , 𝑟𝑛 be positive integers. Their least common multiple (abbreviated lcm ) is the
positive generator of the cyclic group of all common multiples of the 𝑟𝑖 , that is the cyclic group
of all integers divisible by each 𝑟𝑖 for 𝑖 = 1,2, … … . , 𝑛.

10
Theorem

Let (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) ∈ ⨅𝑛𝑖=1 𝐺𝑖 .

If 𝑎𝑖 is of finite order 𝑟𝑖 in 𝐺𝑖 , then the order of (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ) in ⨅𝑛𝑖=1 𝐺𝑖 is equal to the least
common multiple of all the 𝑟𝑖 .

Proof

Given,

𝑟
order of 𝑎1 = 𝑟1 ⇒ 𝑎11 = 𝑒1 in 𝐺1

𝑟
order of 𝑎2 = 𝑟2 ⇒ 𝑎22 = 𝑒2 in 𝐺2

.
.
.
𝑟
order of 𝑎𝑛 = 𝑟𝑛 ⇒ 𝑎𝑛𝑛 = 𝑒𝑛 in 𝐺𝑛 .

We have to find a power 𝑘 for (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 ).

So that (𝑎1 , 𝑎2 , … … . , 𝑎𝑛 )𝑘 = (𝑒1 , 𝑒2 , … … . , 𝑒𝑛 ).

The power must simultaneously be a multiple of 𝑟1 , multiple of 𝑟2 and so on. But 𝑘 is the least
positive integers having the above property.

∴ 𝑘 = 𝑙𝑐𝑚(𝑟1 , 𝑟2 , … … . , 𝑟𝑛 ).

Q. Find the order of (8,4,10) in the group ℤ12 × ℤ60 × ℤ24 .

𝑂(8) = 3 in 𝑍12

11
 𝑂(4) = 15 in 𝑍60

𝑂(10) = 12 in 𝑍24

𝑂(8,4,10) = 𝑙𝑐𝑚(3,15,12) = 60

Q. Find a generator of ℤ × ℤ2

ℤ × ℤ2 = {(𝑛, 0), (𝑛, 1): 𝑛 ∈ ℤ}

(𝑛, 0) = 𝑛(1,0)
(𝑛, 1) = (𝑛, 0) + (0,1) = 𝑛(1,0) + (0,1)

∴ ℤ × ℤ2 is generated by {(1,0), (0,1)}

In general , ℤ × ℤ𝑛 is generated by ,

{(1,0,0, … … . ,0), (0,1,0, … … . ,0), … … . , (0,0, … … . ,1)}

Q. Find the order of (3,10,9) in (ℤ4 , ℤ12 , ℤ15 )

𝑂(3) = 4 in ℤ4

𝑂(10) = 6 in ℤ12

𝑂(9) = 5 in ℤ15

∴ 𝑂(3,10,9) = 𝑙𝑐𝑚(4,6,5)
= 60

∴ order of (3,10,9) in ℤ4 × ℤ12 × ℤ15 is 60.

12
 CHAPTER-2

FUNDAMENTAL THEOREM OF FINITELY GENERATED ABELIAN

GROUPS

Every finitely generated abelian group 𝐺 is isomorphic to a direct product of cyclic groups in the
form,

ℤ(𝑝1 )𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛 × ℤ × ℤ × ℤ × … … .× ℤ

Where the 𝑝𝑖 are primes , not necessarily distinct and the 𝑟𝑖 are positive integers.

Remark

• The direct product is unique except for possible rearrangement of the factors.
• The number of factors ℤ is unique and this number is called Betti number.

Example

Find all abelian groups , upto isomorphism of order

1)8 , 2)16 , 3)360

(1) Order 8

8=1×8

8 = 2 × 4 = 2 × 22

8 = 2×2×2

3 non-isomorphic groups are ℤ8 , ℤ2 × ℤ4 ,

ℤ2 × ℤ2 × ℤ2
13
(2) Order 16

16 = 1 × 16 = 1 × 24

16 = 2 × 8 = 2 × 23

16 = 4 × 4 = 22 × 22

16 = 2 × 2 × 2 × 2

16 = 2 × 2 × 22

ℤ16 , ℤ2 × ℤ8 , ℤ4 × ℤ4 , ℤ2 × ℤ2 × ℤ2 × ℤ2 , ℤ2 × ℤ2 × ℤ4

(3) Order 360

360 = 22 ⋅ 32 ⋅ 5

Possibilities are,

1) ℤ8 × ℤ9 × ℤ5

2) ℤ2 × ℤ4 × ℤ9 × ℤ5

3) ℤ2 × ℤ2 × ℤ2 × ℤ9 × ℤ5

4) ℤ8 × ℤ3 × ℤ3 × ℤ5

5) ℤ2 × ℤ4 × ℤ3 × ℤ3 × ℤ5

6) ℤ2 × ℤ2 × ℤ2 × ℤ3 × ℤ3 × ℤ5

Definition

A group 𝐺 is decomposable if it is isomorphic to a direct product of two proper non-trivial

subgroups , otherwise 𝐺 is indecomposable.

14
Example

ℤ6 is decomposable while ℤ5 is indecomposable.

ℤ6 is isomorphic to ℤ2 × ℤ3

ℤ𝑚𝑛 is isomorphic to ℤ𝑚 × ℤ𝑛 , if 𝑚 and 𝑛 are prime.

Theorem

The finite indecomposable abelian groups are exactly the cyclic groups with order a power of a
prime.

Proof

Let 𝐺 be a finite indecomposable abelian group ∵ 𝐺 is finitely generated , we can apply

fundamental theorem of finitely generated abelian groups.

∴ 𝐺 ≅ ℤ(𝑝)𝑟1 × ℤ(𝑝2)𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

∵ 𝐺 is indecomposable and ℤ(𝑝𝑖 )𝑟𝑖 ’s are proper subgroups we get in the above , there is only one
factor say ℤ(𝑝𝑖 )𝑟𝑖 which is cyclic group with order a prime power.

Theorem

If 𝑚 divides the order of a finite abelian group , then 𝐺 has a subgroup of order 𝑚.

Proof

Given 𝐺 is a finite abelian group.

∴ we can apply Fundamental Theorem ,

15
Hence,

𝐺 ≅ ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛

1 2 𝑛

Here all primes need not be distinct.

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 . 𝑝22 … … 𝑝𝑛𝑛

Let 𝑚 is a +𝑣𝑒 integer which divides 𝑂(𝐺).

0 ≤ 𝑠𝑖 ≤ 𝑟𝑖 By theorem , “ let 𝐺 be a cyclic group with 𝑛 elements and generated by 𝑎. Let

𝑛
𝑏 ∈ 𝐺 , 𝑏 = 𝑎𝑠 , then ‘ 𝑏’ generates a cyclic subgroup 𝐻 of 𝐺 containing elements , where
𝑑

𝑑 = gcd(𝑛, 𝑠).”

𝑟
𝑟 −𝑠 𝑝𝑖 𝑖
𝑝𝑖 𝑖 𝑖 generates a cyclic subgroup of ℤ𝑝𝑟𝑖 having order 𝑟 𝑟 −𝑠𝑖
𝑖 𝑔𝑐𝑑(𝑝𝑖 𝑖 ,𝑝𝑖 𝑖 )

𝑟
𝑝𝑖 𝑖 𝑠
= 𝑟 −𝑠𝑖 = 𝑝𝑖 𝑖
𝑝𝑖 𝑖

𝑟 −𝑠𝑖 𝑠
∴ 𝑂(< 𝑝𝑖 𝑖 >) = 𝑝𝑖 𝑖

𝑟 −𝑠1 𝑠
i.e. < 𝑝11 > is a subgroup of ℤ𝑝𝑟1 having order 𝑝11 .
1

𝑟 −𝑠2 𝑠
< 𝑝22 > is a subgroup of ℤ𝑝𝑟2 having order 𝑝22 .
2

…………………………………………………………

𝑟 −𝑠𝑛 𝑠
< 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟𝑛 having order 𝑝𝑛𝑛 .
𝑛

𝑟 −𝑠1 𝑟 −𝑠2 𝑟 −𝑠𝑛

∴ < 𝑝11 > × < 𝑝22 > × … … .× < 𝑝𝑛𝑛 > is a subgroup of ℤ𝑝𝑟1 × ℤ𝑝𝑟2 × … … .× ℤ𝑝𝑟𝑛
1 2 𝑛
𝑠 𝑠 𝑠
having order 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛 = 𝑚.

16
Theorem

If 𝑚 is a square free integer , that is 𝑚 is not divisible by the square of any prime . Then every
abelian group of order 𝑚 is cyclic.

Proof

Let 𝑚 be a square free integer , then 𝑝𝑖 ⫮ 𝑚 for every 𝑖 greater than 1 for a prime 𝑝.

Given 𝐺 is a finite abelian group having order 𝑚 , by fundamental theorem , then

𝐺 ≅ ℤ(𝑝1 )𝑟1 × ℤ(𝑝2 )𝑟2 × … … .× ℤ(𝑝𝑛)𝑟𝑛

Then,

𝑟 𝑟 𝑟
𝑂(𝐺 ) = 𝑝11 ⋅ 𝑝22 ⋅⋅⋅⋅⋅⋅⋅ 𝑝𝑛𝑛

∵ 𝑂(𝐺) is a square free integer , the only possibility

𝑟1 = 𝑟2 = … … . . = 𝑟𝑛 = 1

Then,

𝐺 ≅ ℤ𝑝1 × ℤ𝑝2 × … … .× ℤ𝑝𝑛

≅ ℤ𝑝1 ,𝑝2 ,…….,𝑝𝑛 , which is cyclic.

Example

15 is a square free integer. So an abelian group of order 15 is cyclic.

17
 CONCLUSION

Direct product of groups is the product 𝐺1 × 𝐺2 , … … . 𝐺𝑛 ,where each 𝐺𝑖 is a set. We have

discussed about definition and some properties related to the direct product of groups. The
fundamental theorem of finitely generated abelian group helped us to get a deeper understanding
about the topic. The theorems gives us complete structural information about abelian group, in
particular finite abelian group. We have also discussed some examples in order to develope more
intrest in algebra.

18
 BIBLIOGRAPHY

• A first course in Abstract Algebra – 7th edition.

John . B . Fraleigh

• Contomporary Abstract Algebra – 7th edition.

Joseph . A . Gallian

• Topics In Algebra – 2nd edition

I . N . Herstein

19
POWER SERIES SOLUTIONS AND
SPECIAL FUNCTIONS
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
MERIN K JOHN
DB18CMSR23
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Power Series’ is a bona fide project of

MERIN K JOHN carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I MERIN K JOHN hereby declare that the project ‘Power Series’ is an

original record of studies and bona fide project carried out by me during the
period of 2018 – 2021 under the guidance of Ms. Athulya P, Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu and has not
submitted by me elsewhere for the award of my degree, diploma, title or
recognition, before.

Name
MERIN K JOHN

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express my deepest gratitude to
people along the way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my
project supervisor Ms. Athulya P, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and
the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every
one of them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- SERIES SOLUTION OF FIRST 4-9

ORDER EQUATION

04 Chapter 2- SECOND ORDER LINEAR 9-18

EQUATION, ORDINARY POINTS

05 Conclusion 19

06 Bibliography 20
 INTRODUCTION
A power series is a type of series with terms involving a variable. Power series
are often used by calculators and computers to evaluate trigonometric,
hyperbolic, exponential and logarithm functions. So any application of these
kind of functions is a possible application of power series. Many interesting and
important differential equations can be found in power series.

1
 PRELIMINERY

A. An infinite series of the form

∞

∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ (1)
𝑛=0

is called a power series in x. The series

∞

∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛 = 𝑎0 + 𝑎1 (𝑥 − 𝑥0 ) + 𝑎2 (𝑥 − 𝑥0 )2 + ⋯
𝑛=0

is a power series in x – x0.

B. The series (1) is said to converge at a point x if the limit

𝑚

𝑙𝑖𝑚 ∑ 𝑎𝑛 𝑥 𝑛
𝑚→∞
𝑛=0

exists, and in this case the sum of the series is the value of this limit.
Radius of convergence: Series in 𝑥 has a radius of convergence 𝑅, where
0 ≤ 𝑅 ≤ ∞, with the property that the series converges if |𝑥| < 𝑅 and
diverges if |𝑥| > 𝑅. It should be noted that if 𝑅 = 0 then no 𝑥 satisfies
|𝑥| < 𝑅, and if 𝑅 = ∞ then no 𝑥 satisfies |𝑥| > 𝑅
𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | | , if the limit exists.
𝑛→∞ 𝑎𝑛+1

C. Suppose that (1) converges for |𝑥| < 𝑅 with 𝑅 > 0, and denote its sum
by f(x):
∞

𝑓(𝑥) = ∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑛=0

Then f(x) is automatically continuous and has derivatives of all orders for
|𝑥| < 𝑅.

2
D. Let f(x) be a continuous function that has derivatives of all orders for
|x|< R with R > 0. f(x) be represented as power series using Taylor’s
formula:
𝑛
𝑓 (𝑘) (0) 𝑘
𝑓(𝑥) = ∑ 𝑥 + 𝑅𝑛 (𝑥)
𝑘!
𝑘=0

where the remainder Rn (x) is given by

𝑓 (𝑛+1) (𝑥̅ ) 𝑛+1
𝑅𝑛 (𝑥) = 𝑥
(𝑛 + 1)!
for some point 𝑥̅ between 0 and x.

E. A function f(x) with the property that a power series expansion of

the form

𝑓(𝑥) = ∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛
𝑛=0

is valid in some neighbourhood of the point x0 is said to be analytic at

x0. In this case the an are necessarily given by
𝑓 (𝑛) (𝑥0 )
𝑎𝑛 =
𝑛!
and is called the Taylor series of f(x) at x0.

Analytic functions: A function f defined on some open subset U of R or C is

called analytic if it is locally given by a convergent power series. This means
that every a ∈ U has an open neighbourhood V ⊆ U, such that there exists
a power series with centre a that converges to f(x) for every x ∈ V.

3
 CHAPTER 1
SERIES SOLUTION OF FIRST ORDER EQUATION
We have studied to solve linear equations with constants coefficient but with
variable coefficient only specific cases are discussed. Now we turn to these
latter cases and try to find a general method to solve this. The idea is to assume
that the unknown function y can be explained into a power series. Our purpose
in this section is to explain the procedures by showing how it works in the case
of first order equation that are easy to solve by elementary methods.

Example 1: we consider the equation

𝑦ʹ = 𝑦
Consider the above equation as (1). Assume that y has a power series solution
of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |x| < R, R > 0

That is we assume that 𝑦 ʹ = 𝑦 has a solution that is analytic at origin. We have
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ ⋯
then
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ … ….
∴ (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 ⋯
= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
⇒ 𝑎1 = 𝑎0
𝑎1 𝑎0
2𝑎2 = 𝑎1 ⇒ 𝑎2 = =
2 2

4
 𝑎2 𝑎0 𝑎0
3𝑎3 = 𝑎2 ⇒ 𝑎3 = = =
3 2 ∙ 3 3!
𝑎3 𝑎0 𝑎0
4𝑎4 = 𝑎3 ⇒ 𝑎4 = = =
4 2 ⋅ 3 ⋅ 4 4!
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0 𝑎0
= 𝑎0 + 𝑎0 𝑥 + 𝑥 2 + 𝑥 3 + 𝑥 4 + ⋯
2 3! 4!
𝑥2 𝑥3 𝑥4
= 𝑎0 (1 + 𝑥 + + + + ⋯ )
2! 3! 4!
𝑦 = 𝑎0 𝑒 𝑥
To find the actual function we have 𝑦 ʹ = 𝑦
𝑑𝑦 𝑑𝑦
i.e., =𝑦 ⇒ = 𝑑𝑥
𝑑𝑥 𝑦
integrating
log 𝑦 = 𝑥 + 𝑐
i.e., 𝑦 = 𝑒 𝑥+𝑐 = 𝑒 𝑥 ⋅ 𝑒 𝑐
𝑦 = 𝑎0 𝑒 𝑥 , where a0 = ec , a constant.

Example 2: solve 𝑦 ′ = 2𝑥𝑦. Also find its actual solution.

Solution: 𝑦 ′ = 2𝑥𝑦 (1)
Assume that y has a power series of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |𝑥| < 𝑅, 𝑅 > 0

∞

We have 𝑦 = ∑ a𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

5
 = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 2𝑥(𝑎0 + 𝑎1 𝑥 +𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯ )
= 2𝑥𝑎0 + 2𝑥𝑎1 𝑥 + 2𝑥𝑎2 𝑥 2 + 2𝑥𝑎3 𝑥 3 + ⋯
= 2𝑥𝑎0 + 2𝑎1 𝑥 2 + 2𝑎2 𝑥 3 + 2𝑎3 𝑥 4 + ⋯ … ..
2𝑎0
⇒ 𝑎1 = 0 2𝑎2 = 2𝑎0 ⇒ 𝑎2 = = 𝑎0
𝑧
2𝑎1
3. 𝑎3 = 2𝑎1 ⇒ 𝑎3 = =0
3
2𝑎2 𝑎0
4𝑎4 = 2𝑎2 ⇒ 𝑎4 = =
42 2

5𝑎5 = 2𝑎3 = 0 ⇒ 𝑎5 = 0
2𝑎4 𝑎4 𝑎0 𝑎0
6𝑎6 = 2𝑎4 ⇒ 𝑎6 = = = =
6 3 2⋅3 3!

We get,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0
= 𝑎0 + 0 + 𝑎0 𝑥 2 + 0𝑥 3 + 𝑥 4 + ⋯
2
𝑎 0
= 𝑎0 + 𝑎0 𝑥 2 + 𝑥 4 + ⋯
2
2
𝑥4 𝑥6
= 𝑎0 (1 + 𝑥 + + + ⋯ )
2! 3!
2
𝑦 = 𝑎0 𝑒 𝑥
To find an actual solution

𝑦 ′ = 2𝑥𝑦
𝑑𝑦
= 2𝑥𝑦
𝑑𝑥
𝑑𝑦
⇒ = 2𝑥 ⋅ 𝑑𝑥
𝑦
log 𝑦 = 𝑥 2 + 𝑐
2
𝑦 = 𝑒𝑥 + 𝑐
2
⇒ 𝑦 = 𝑎0 𝑒 𝑥 , where 𝑎0 = 𝑒 𝑐

6
Example 3: Consider 𝑦 = (1 + 𝑥)𝑝 where p is an arbitrary constant. Construct a
differential equation from this and then find the solution using power series
method.
Solution
First, we construct a differential equation
i.e. 𝑦 = (1 + 𝑥)𝑝
𝑝(1+𝑥)𝑝 𝑝𝑦
𝑦 ′ = 𝑝(1 + 𝑥)𝑝−1 = =
1+𝑥 1+𝑥

∴ (1 + 𝑥)𝑦 ′ = 𝑝𝑦, 𝑦(0) = 𝑟

Assume that y has a power series solution of the form,
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎̇ 2 𝑥 2 + ⋯ … …
Which converges for |𝑥| < 𝑅̇, 𝑅 > 0
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ … … ..
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1 + 𝑥)𝑦 ′ = 𝑝𝑦
⇒ (1 + 𝑥)𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 𝑝(𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
⇒ (𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) + (𝑎1 𝑥 + 2𝑎2 𝑥 2 + 3𝑎3 𝑥 3 + ⋯ )
= 𝑎0 𝑝 + 𝑎1 𝑝𝑥 + 𝑎2 𝑝𝑥 2 + ⋯
Equating the coefficients of 𝑥, 𝑥 2 , …
𝑎1 = 𝑎0 𝑝 i.e. 𝑎1 = 𝑝, (since 𝑎0 = 1)
⇒ 2𝑎2 = 𝑎1 (p − 1)
𝑎1 (p − 1) 𝑎0 𝑃(𝑝 − 1)
𝑎2 = =
2 2

7
 3𝑎3 + 2𝑎2 = 𝑎2 𝑝
𝑠𝑎3 = 𝑎2 𝑝 − 2𝑎2
= 𝑎2 (𝑝 − 2)
𝑎2 (𝑝 − 2) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)
𝑎3 = =
3 2⋅3
4𝑎4 + 3𝑎3 = 𝑎3 𝑝
4𝑎4 = 𝑎3 𝑝 − 3𝑎3
= 𝑎3 (𝑝 − 3)
𝑎3 (𝑝 − 3) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3)
𝑎4 = =
4 2⋅3⋅4
∴ we get,
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0 𝑝(𝑝 − 1) 2 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 𝑎0 + 𝑎0 𝑝𝑥 + 𝑥 + 𝑥 + ⋯…
2 2⋅3
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 1 + 𝑝𝑥 + 𝑥 + 𝑥 +
2! 3!
𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3) 4 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − (𝑛 − 1)) 𝑛
𝑥 + ⋯+ 𝑥
4! 𝑛!
Since the initial problem y(0) = 1 has one solution the series converges for |x|<1
So this is a power solution,
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1) ⋯ (𝑝 − (𝑛 − 1)) 𝑛
(1 + 𝑥)𝑝 = 1 + 𝑝𝑥 + 𝑥 + ⋯+ 𝑥
2! 𝑛!
Which is binomial series.

Example 4: Solve the equation 𝑦 ′ = 𝑥 − 𝑦, 𝑦 (0) = 0

Solution: Assume that y has a power series solution of the form
∞

𝑦 = ∑ an 𝑥 𝑛
𝑛=0

which converges for |𝑥| < 𝑅, 𝑅 > 0

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯

8
 Now 𝑦 ′ = 𝑥 − 𝑦
(𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) = 𝑥 − (𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
Equating the coefficients of 𝑥, 𝑥 2 ,
𝑎1 = −𝑎0 = 0, Since 𝑦(0) = 0
2𝑎2 = 1 − 𝑎1
=1−0
1
⇒ 𝑎2 =
2
3𝑎3 = −𝑎2
−𝑎2 1
𝑎3 = =−
3 2⋅3
4𝑎4 = −𝑎3
1
⇒ 𝑎4 =
2⋅3⋅4
𝑥2 𝑥3 𝑥4
∴ 𝑦 = 0 + 0 + − + − ⋯……
2! 3! 4!
𝑥2 𝑥3
= (1 − 𝑥 + − + ⋯ ) + 𝑥 − 1
2! 3!
= 𝑒 −𝑥 + 𝑥 − 1
By direct method
𝑦′ = 𝑥 − 𝑦 𝑑𝑦
𝑑𝑦 𝑑𝑦 ( + 𝑝𝑦 = 𝑄 𝑓𝑜𝑟𝑚)
=𝑥−𝑦 ⇒ +𝑦 =𝑥 𝑑𝑥
𝑑𝑥 𝑑𝑥
here 𝑃(𝑥) = 1, integrating factor
= 𝑒 ∫ 𝑝(𝑥)⋅𝑑𝑥
= 𝑒𝑥
∴ 𝑦𝑒 𝑥 = ∫ 𝑥𝑒 𝑥 ⋅ 𝑑𝑥
𝑦𝑒 𝑥 = 𝑥 ⋅ 𝑒 𝑥 − ∫ 𝑒 𝑥 ⋅ 𝑑𝑥
= 𝑥𝑒 𝑥 − 𝑒 𝑥
𝑦𝑒 𝑥 = 𝑒 𝑥 (𝑥 − 1) + 𝑐
𝑒 𝑥 (𝑥 − 1) + 𝑐 𝑐
𝑦= = 𝑥 − 1 + 𝑥 = 𝑐𝑒 −𝑥 + (𝑥 − 1)
𝑑𝑥 𝑒
∴ 𝑦 = (𝑥 − 1) + 𝑐𝑒 −𝑥

9
 CHAPTER 2
SECOND ORDER LINEAR EQUATION, ORDINARY POINTS

Consider the general homogeneous second order linear equation,

𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0 (1)
As we know, it is occasionally possible to solve such an equation in terms of
familiar elementary functions. This is true, for instance, when P(x) and Q(x)
are constants, and in a few other cases as well. For the most part, however,
the equations of this type having the greatest significance in both pure and
applied mathematics are beyond the reach of elementary methods, and can
only be solved by means of power series.
P(x) and Q(x) are called coefficients of the equation. The behaviour of its
solutions near a point x0 depends on the behaviour of its coefficient functions
P(x) and Q(x) near this point. we confine ourselves to the case in which P(x) and
Q(x) are well behaved in the sense of being analytic at x0, which means that
each has a power series expansion valid in some neighbourhood of this point. In
this case x0 is called an ordinary point of equation (1). Any point that is not an
ordinary point of (1) is called a singular point.

Consider the equation,

𝑦 ′′ + 𝑦 = 0 (2)

the coefficient functions are P(x) = 0 and Q(x) = 1, These functions are analytic
at all points, so we seek a solution of the form,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ + 𝑎𝑛 𝑥 𝑛 + ⋯ (3)

Differentiating (3) we get,

𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ + (𝑛 + 1)𝑎𝑛+1 𝑥 𝑛 + ⋯ (4)

10
And

𝑦 ′′ = 2𝑎2 + 2 ⋅ 3𝑎3 𝑥 + 3 ⋅ 4𝑎4 𝑥 2 + ⋯ + (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ⋯ (5)

If we substitute (5) and (3) into (2) and add the two series term by term, we get

(2𝑎2 + 𝑎0 ) + (2 ⋅ 3𝑎3 + 𝑎1 )𝑥 + (3 ⋅ 4𝑎4 + 𝑎2 )𝑥 2 +

𝑦 ′′ + 𝑦 = =0
(4 ⋅ 5𝑎5 + 𝑎3 )𝑥 3 + ⋯ + [(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 ]𝑥 𝑛 + ⋯
and equating to zero the coefficients of successive powers of x gives
2𝑎2 + 𝑎0 = 0, 2 ⋅ 3𝑎3 + 𝑎1 = 0, 3 ⋅ 4𝑎4 + 𝑎2 = 0
4 ⋅ 5𝑎5 + 𝑎3 = 0, … … , (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 = 0, …
By means of these equations we can express an in terms of a0 or a0, according
as n is even or odd:
𝑎0 𝑎1 𝑎2 𝑎0
𝑎2 = − , 𝑎3 = − , 𝑎4 = − =
2 2⋅3 3⋅4 2⋅3⋅4
𝑎3 𝑎1
𝑎5 = − = ,⋯
4⋅5 2⋅3⋅4⋅5
With these coefficients, (3) becomes
𝑎0 𝑎1 3 𝑎0 𝑎1
𝑦 = 𝑎0 + 𝑎1 𝑥 − 𝑥 2 − 𝑥 + 𝑥4 + 𝑥5 − ⋯
2 2⋅3 2⋅3⋅4 2⋅3⋅4⋅5
𝑥2 𝑥4 𝑥3 𝑥5
= 𝑎0 (1 − + − ⋯ ) + 𝑎1 (𝑥 − + − ⋯ ) (6)
2! 4! 3! 5!
𝑖. 𝑒, 𝑦 = 𝑎0 cos 𝑥 + 𝑎1 sin 𝑥
Since each of the series in the parenthesis converges for all x. This implies the
series (2) for all x.

Solve the legenders equation,

(1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0
Solution
Consider (1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
11
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2

put 𝑛 = 𝑛 + 2 (Since 𝑦 ′′ is not 𝑥 𝑛 form )

∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛+2 𝑥 𝑛+2−2

𝑛=0
∞

∴ 𝑦 ′′ = ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

Now (1) ⇒ 𝑦 ′′ − 𝑥 2 𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0

⇒ ∑(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 − ∑𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛 − ∑2𝑛𝑎𝑛 𝑥 𝑛 + ∑𝑝(𝑝 + 1)𝑎𝑛 𝑥 𝑛 = 0
∞

⇒ ∑ [((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 )𝑥 𝑛 ] = 0

𝑛=0

for n = 0,1,2,3…….
⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 = 0
[𝑛(𝑛 − 1) + 2𝑛 − 𝑝(𝑝 + 1)]
⇒ 𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(𝑛2 − 𝑛 + 2𝑛 − 𝑝2 − 𝑝)𝑎𝑛
=
(𝑛 + 1)(𝑛 + 2)
(𝑛 + 𝑛 − 𝑝2 − 𝑝)𝑎𝑛
2
=
(𝑛 + 1)(𝑛 + 2)

12
 −(𝑝 − 𝑛)(𝑝 + 𝑛 + 1)
∴ 𝑎𝑛+2 = 𝑎𝑛 , 𝑛 = 0,1,2 …
(𝑛 + 1)(𝑛 + 2)

This is an Recursion formula

−𝑝(𝑝 + 1)
put 𝑛 = 0, 𝑎2 = 𝑎0
1⋅2
−(𝑝 − 1)(𝑝 + 2)
𝑛 = 1, 𝑎3 = ⋅ 𝑎1
2⋅3
−(𝑝 − 2)(𝑝 + 3)
𝑛 = 2, 𝑎4 = 𝑎2
3𝑖4
𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3)
= 𝑎0
4!
−(𝑝 − 3)[𝑝 + 4)
𝑛 = 3, 𝑎5 = 𝑎3
4⋅5
(𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4)
= 𝑎1
5!
−(𝑝 − 4)(𝑝 + 5)
𝑛 = 4, 𝑎6 = 𝑎4
5⋅6
−𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5)
= 𝑎0
6!
(𝑝 − 5)(𝑝 + 6)
𝑛 = 5, 𝑎7 = − 𝑎5
6⋅7
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6)
=− 𝑎1
7!

𝑝(𝑝 + 1) 2 𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3) 4

𝑦 = 𝑎0 [1 − 𝑥 + 𝑥
2! 4!
𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5) 6
− 𝑥 + ⋯]
6!
(𝑝 − 1)(𝑝 + 2) 3 (𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4) 5
+𝑎1 [𝑥 − 𝑥 + 𝑥
3! 5!
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6) 7
− 𝑥 + ⋯]
7!

13
Find the general solution of (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 in terms of power
series in 𝑥. Can you express this solution by means of elementary functions?
Solution
Consider the equation (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

(1 + 𝑥 2 )𝑦 ′′ = 𝑦 ′′ + 𝑥 2 𝑦 ′′
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2
∞

Now 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

put 𝑛 = 𝑛 + 2
∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛 + 2𝑥 𝑛+2=2

𝑛=0
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

14
 ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥𝑛

𝑛=0 𝑛=2
∞ ∞

+ ∑ 2𝑛𝑎𝑛 𝑥 𝑛 − ∑ 2𝑎𝑛 𝑥 𝑛 = 0
𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 = 0

[−𝑛(𝑛 − 1) − 2𝑛 + 2]
𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(−𝑛2 + 𝑛 − 2𝑛 + 2)
= 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)

2 2𝑎0
put 𝑛 = 0, 𝑎2 = 𝑎0 = = 𝑎0
1⋅2 2!
(1 − 1 − 2 + 2)
𝑛 = 1, 𝑎3 = 𝑎1 = 0
2⋅3
2−4−4+2 −4 −𝑎0
𝑛 = 2, 𝑎4 = 𝑎2 = 𝑎0 =
3⋅4 3⋅4 3
3 − 9 − 16 + 2
𝑛 = 3, 𝑎5 = 𝑎3 = 0
4.5
4 − 16 − 8 + 2 −3 3𝑎0 𝑎0
𝑛 = 4, 𝑎6 = 𝑎4 = 𝑎4 = =
5.6 5 3.5 5

∴ 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0
= 𝑎0 + 𝑎1 𝑥 + 𝑎0 𝑥 2 − 𝑥 4 + 𝑥 6 … .
3 5
4 6
𝑥 𝑥
= 𝑎0 [1 + 𝑥 2 − + − ⋯ ] + 𝑎1 𝑥
3 5
𝑥3 𝑥5
= 𝑎0 [1 + 𝑥 (𝑥 − + ⋯ )] + 𝑎1 𝑥
3 5
= 𝑎0 (1 + 𝑥tan−1 𝑥) + 𝑎1 𝑥

15
Consider the equation 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0
(a) Find its general solution 𝑦 = ∑𝑎𝑛 𝑥 𝑛 in the form
𝑦 = 𝑎0 𝑦1 (𝑥) + 𝑎1 𝑦2 (𝑥) where 𝑦1 (𝑥) and 𝑦2 (𝑥) are power series
(b) use the ratio test to verify that the two series 𝑦1 (𝑥) and 𝑦2 (𝑥) converges
. for all x.
Solution:
Given 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ (1)
Assume that y has a power series solution the form ∑a𝑛 𝑥 𝑛 which converges
for |𝑥| 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛 ⋅ 𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0
∞

𝑥𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞ ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)a𝑛+2 𝑥 + ∑ 𝑛𝑎𝑛 𝑥 + ∑ 𝑎𝑛 𝑥 𝑛 = 0

𝑛 𝑛

𝑛=0 𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 = 0

(−𝑛 − 1)𝑎𝑛 −𝑎𝑛
⇒ 𝑎𝑛+2 = =
(𝑛 + 1)(𝑛 + 2) 𝑛 + 2
𝑎0
put 𝑛 = 0, 𝑎2 = −
2
−2𝑎1 −𝑎1
𝑛 = 1, 𝑎3 = =
2⋅3 3
16
 −3𝑎2 −𝑎2 𝑎0
𝑛 = 2, 𝑎4 = = =
3⋅4 4 8
−4𝑎3 𝑎1
𝑛 = 3, 𝑎5 = =
4⋅5 15
−5𝑎4 −𝑎0
𝑛 = 4, 𝑎6 = =
5⋅6 48
𝑎0 𝑎1 𝑎0 𝑎1 𝑎0
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + − 𝑥 2 − 𝑥 3 + 𝑥 4 + 𝑥 5 − 𝑥 6 + ⋯
2 3 8 15 48

𝑥2 𝑥4 𝑥6 𝑥3 𝑥5
= 𝑎0 [1 − + − + ⋯ ] + 𝑎1 [𝑥 − + + ⋯]
2 2⋅4 2⋅4⋅6 3 3.5

𝑥2 𝑥4 𝑥̇ 2
𝑤ℎ𝑒𝑟𝑒 𝑦1 (𝑥) = 1 − + − +
2 2⋅4 2⋅4⋅6

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + +⋯
3 3⋅5

𝑥2 𝑥4 𝑥6
(b) 𝑦1 (𝑥) = 1 − + − +⋯
2 2⋅4 2⋅4⋅6

𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | |
𝑛→∞ 𝑎𝑛+1
(−1)𝑛 (−1)𝑛+1
= 𝑙𝑖𝑚 | / |
𝑛→∞ 2 ⋅ 4 ⋅ (2𝑛) 2 ⋅ 4 ⋅⋅ (2𝑛 + 2)

2(𝑛 + 1)
= 𝑙𝑖𝑚 | |
𝑛→∞ −1
1
= 𝑙𝑖𝑚 | − 2𝑛(1 + )| = ∞
𝑛→∞ 𝑛

∴ 𝑦1 (𝑥) converges for all 𝑥

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + −⋯
3 3⋅5

17
 (−1)𝑛 (−1)𝑛+1
𝑅 = 𝑙𝑖𝑚 | ⁄ |
𝑛→∞ 3 ⋅ 5 ⋯ (2𝑛 + 1) 3 ⋅ 5 ⋅ ⋯ (2𝑛 + 3)

(−1) ⋅ 3 ⋅ 5 ⋯ (2𝑛 + 1)(2𝑛 + 3)

= 𝑙𝑖𝑚 | |
𝑛→∞ 3 ⋅ 5 ⋯ ⋅ (2𝑛 + 1)
= 𝑙𝑖𝑚 |(−1)𝑛(2 + 3/𝑛)| = ∞
𝑛→∞

∴ 𝑦2 (𝑥) converges for all 𝑥

REGULAR SINGULAR POINTS

A singular point 𝑥0 of equation
𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0
is said to be regular if the functions (𝑥 − 𝑥0 )𝑃(𝑥) and (𝑥 − 𝑥0 )2 𝑄(𝑥) are
analytic, and irregular otherwise. Roughly speaking, this means that the
singularity in 𝑃(𝑥) cannot be worse than 1/(𝑥 − 𝑥0 ), and that in 𝑄(𝑥) cannot
be worse than 1/(𝑥 − 𝑥0 )2 .
If we consider Legendre’s equation in the form
2𝑥 𝑝(𝑝 + 1)
𝑦 ′′ − 𝑦 ′
+ 𝑦=0
1 − 𝑥2 1 − 𝑥2
it is clear that x = 1 and x = −1 are singular points. The first is regular because
2𝑥 (𝑥 − 1)𝑝(𝑝 + 1)
(𝑥 − 1)𝑃(𝑥) = and (𝑥 − 1)2 𝑄(𝑥) = −
𝑥+1 𝑥+1
are analytic at x = 1, and the second is also regular for similar reasons.
Example: Bessel’s equation of order p, where p is a nonnegative constant:
𝑥 2 𝑦 ′′ + 𝑥𝑦 ′ + (𝑥 2 − 𝑝2 )𝑦 = 0
If this is written in the form
1 ′ 𝑥 2 − 𝑝2
′′
𝑦 + 𝑦 + 𝑦 = 0,
𝑥 𝑥2
it is apparent that the origin is a regular singular point because𝑥𝑃(𝑥) = 1 and
𝑥 2 𝑄(𝑥) = 𝑥 2 − 𝑝2 are analytic at x = 0.

18
 CONCLUSION
The purpose of this project gives a simple account of series solution of first
order equation, second order linear equation, ordinary points. The study of these
topics given excellent introduction to the subject called ‘POWER SERIES’
we used application of power series extensively throughout this project. We
take it for granted that most readers are reasonably well acquainted with these
series from an earlier course in calculus. Nevertheless, for the benefit of those
whose familiarity with this topic may have faded slightly, we presented a brief
review of the main facts of power series.

19
 BIBLIOGRAPHY
George F. Simmons - Differential Equations With Applications and Historical
. Notes
G.Birkoff and G.C Rota - Ordinary Differential Equations; Wiley and Sons; 3rd
. Edition ( 1978 )
P.Hartmon - Ordinary Differential Equations; John Wiley and Sons

20
NORMED LINEAR SPACES
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
NIVYA JOSEPH
DB18CMSR07
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE
It is to certify that this project report ‘Normed Linear Spaces’ is the bonafide project of

Nivya Joseph carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head Of Department Supervisor

Department Of Mathematics
Don Bosco Arts And Science College
Angadikadavu

2
 DECLARATION
I Nivya Joseph hereby declare that the project ‘Normed Linear Space’ is an original record of
studies and bona fide project carried out by me during the period of 2018 – 2021 under the
guidance of Ms. Athulya P , Department of Mathematics, Don Bosco Arts and Science College,
Angadikadavu and has not submitted by me elsewhere for the award of my degree, diploma, title
or recognition, before.

Nivya Joseph

DB18CMSR07

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu

3
 ACKNOWLEDGEMENT

First and foremost, praises and thanks to God, the Almighty, for His showers of blessings
throughout my work to complete the project successfully. I had finally managed to finish up
this project with great enthusiasm and determination. All the time spent to search and discuss
ideas as well as justifying theoretical clues to drive the output were worth my effort and time.

Therefore, I would like to express my sincere gratitude to my supervisor Ms. Athulya P,

Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu, for providing
invaluable guidance, comments and suggestions throughout the course of the project. I also
acknowledge with a deep sense of reverence, my gratitude towards all the faculty members of
the Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I owe and respectfully offer my thanks to the principal and staff of Don Bosco Arts and
Science College, Angadikadavu for their constant moral support and mellifluous affection
provided to me.

I express my gratitude to all who directly or indirectly helped me to complete this project
successfully. Their guidance and support was very helpful in bringing this work to
conclusion.

4
 CONTENTS

SI no Title Page no

1 Introduction 6

2 Preliminaries 7-9

3 Chapter 1 10 - 18

4 Chapter 2 19 - 25

5 Conclusion 26

6 Bibliography 27

5
 INTRODUCTION

This chapter gives an introduction to the theory of normed linear spaces. A skeptical reader may
wonder why this topic in pure mathematics is useful in applied mathematics. The reason is quite
simple: Many problems of applied mathematics can be formulated as a search for a certain
function, such as the function that solves a given differential equation. Usually the function
sought must belong to a definite family of acceptable functions that share some useful properties.
For example, perhaps it must possess two continuous derivatives. The families that arise
naturally in formulating problems are often linear spaces. This means that any linear combination
of functions in the family will be another member of the family. It is common, in addition, that
there is an appropriate means of measuring the “distance” between two functions in the family.
This concept comes into play when the exact solution to a problem is inaccessible, while
approximate solutions can be computed. We often measure how far apart the exact and
approximate solutions are by using a norm. In this process we are led to a normed linear space,
presumably one appropriate to the problem at hand. Some normed linear spaces occur over and
over again in applied mathematics, and these, at least, should be familiar to the practitioner.
Examples are the space of continuous functions on a given domain and the space of functions
whose squares have a finite integral on a given domain.

6
 PRELIMINARIES

1) LINEAR SPACES
We introduce an algebraic structure on a set 𝑋and study functions on X which are well behaved
with respect to this structure. From now onwards , K will denote either R , the set of all real
numbers or C, the set of all complex numbers. For k ∈ C , Re k and Im k will denote the real and
imaginary part of k.

A linear space(or a vector space) over K is a non-empty set 𝑋 along with a function
+ : 𝑋 × 𝑋 → 𝑋, called addition and a function ·: K× 𝑋 → 𝑋 called scalar multiplication, such
that for all 𝑥 , 𝑦 , 𝑧 ∈ 𝑋and 𝑘 , 𝑙 ∈ K , we have

𝑥+ 𝑦 = 𝑦 + 𝑥

𝑥 + (𝑦 + 𝑧) = (𝑥 + 𝑦) + 𝑧

∃0∈𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + 0 = 𝑥,

∃ − 𝑥 ∈ 𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + (− 𝑥) = 0 ,

𝑘 · (𝑥 + 𝑦) = 𝑘 · 𝑥 + 𝑘 · 𝑦,

(𝑘 + 𝑙)⋅𝑥 = 𝑘 · 𝑥 + 𝑙 · 𝑥,

(𝑘𝑙)⋅𝑥 = 𝑘 · (𝑙 · 𝑥),

1⋅𝑥 = 𝑥.

We shall write 𝑘𝑥 in place of 𝑘 · 𝑥. We shall also adopt the following notations. For
𝑥, 𝑦 ∈ 𝑋, 𝑘 ∈ K and subsets 𝐸, 𝐹of 𝑋 ,

𝑥 + 𝐹 = {𝑥 + 𝑦: 𝑦 ∈ 𝐹},

𝐸 + 𝐹 = {𝑥 + 𝑦: 𝑥 ∈ 𝐸, 𝑦 ∈ 𝐹},`

𝑘𝐸 = {𝑘𝑥: 𝑥 ∈ 𝐸}.

2) BASIS

A nonempty subset 𝐸 of 𝑋 is said to be a subspace of 𝑋 if 𝑘𝑥 + 𝑙𝑦 ∈ 𝐸 whenever 𝑥, 𝑦 ∈ 𝐸 and

𝑘, 𝑙 ∈ K . If ∅≠𝐸 ⊂ 𝑋, then the smallest subspace of 𝑋 containing 𝐸 is

7
 {
𝑠𝑝𝑎𝑛⁡𝐸 = 𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛: 𝑥1, …, 𝑥𝑛 ∈ 𝐸 , 𝑘1, …, 𝑘𝑛 ∈ 𝐾 }
It is called the span of 𝐸. If span 𝐸 = 𝑋, we say that 𝐸 spans 𝑋. A subset 𝐸 of 𝑋 is said to be
linearly independent if for all 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈K , the equation
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0implies that 𝑘1 = ⋯ = 𝑘𝑛 = 0. It is called linearly dependent if it is not
linearly independent, that is, if there exist 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈ K such that
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0, where at least one 𝑘𝑗is nonzero.

A subset E of X is called a Hamel basis or simply basis for X if span of E = X and E is

linearly independent .

3) DIMENSION

If a linear space 𝑋 has a basis consisting of a finite number of elements , then X is called finite
dimensional and the number of elements in a basis for 𝑋 is called the dimension of 𝑋, denoted as
dimX . Every basis for a finite dimensional linear space has the same (finite) number of elements
and hence the dimension is well-defined. The space {0} is said to have zero dimension. Note that
it has no basis !

If a linear space contains an infinite linearly independent subset, then it is said to be infinite
dimensional.

4)METRIC SPACE

We introduce a distance structure on a set 𝑋 and study functions on 𝑋 which are well-behaved
with respect to this structure.

A metric 𝑑 on a nonempty set 𝑋 is a function 𝑑: 𝑋 × 𝑋 →R

such that for all 𝑥, 𝑦, 𝑧 ∈ 𝑋

d(x, y) ≥0 and d(x , y) = 0 iff x=y

d(y , x) = d(x , y)

d(x , y) ≤ d(x , z) + d(z , y) .

The last condition is known as the triangle inequality. A metric space is a nonempty set 𝑋 along
with a metric on it.

8
5)CONTINUOUS FUNCTIONS

Roughly speaking, a function from a metric space to a metric space is continuous if it sends
‘nearby’ points to ‘nearby’ points. If 𝑋 and 𝑌 are metric spaces with metrics 𝑑 and 𝑒 respectively,
then a function 𝐹: 𝑋 → 𝑌 is said to be continuous at 𝑥0 ∈ 𝑋 if for every ϵ⟩ 0 , there is some
(
δ > 0 (possibly depending on ϵ and 𝑥0 ) such that 𝑒 𝐹(𝑥), 𝐹 𝑥0 ( )) < ϵ for all 𝑥 ∈ 𝑋 satisfying
( )
𝑑 𝑥, 𝑥0 < δ. Further, 𝐹 is said to be continuous on 𝑋 if it is continuous at every point of 𝑋. It is
easy to see that 𝐹 is continuous on 𝑋 if and only if the set F -1(E) is open in X whenever the set E
is open inY. Also , this happens iff F(xn) →F(x) in Y whenever xn →x in X.

6) UNIFORM CONTINUITY

We note that a continuous function 𝐹: 𝑇 → 𝑆 is, in fact, uniformly

continuous, that is, for every ϵ > 0, there exists some δ > 0
such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < ϵ whenever 𝑑 𝑡, 𝑢( )< δ. This can be seen as follows. Let 𝑡 ∈ 𝑇. By
ε
the continuity of 𝐹 at 𝑡 ∈ 𝑇, there is some δ𝑡, such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < 2 whenever
𝑑(𝑡, 𝑢) < δ𝑡.

7) FIELD

A ring is a set R together with two binary operations + and ·( which we call addition and
multiplication ) such that the following axioms are satisfied .

➢ R is an abelian group with respect to addition

➢ Multiplication is associative
➢ ∀𝑎 , 𝑏, 𝑐 ∈ 𝑅the left distributive law a(b + c) = (a·b) + (a·c) and the right distributive
law (a + b)c = (a·c) + (b·c) , hold .

A field is a commutative division ring

9
 CHAPTER 1

NORMED LINEAR SPACE

Let X be a linear space over K . A norm on X is the function || || from 𝑋 to R such that ∀
x,y ∈ X and k ∈ K ,

||𝑥||≥0 and ||𝑥|| = 0 if and only if x = 0 ,

||x + y|| ≤||x|| + ||y|| ,

||kx|| =|k| ||x|| .

A norm is the formalization and generalization to real vector spaces of the intuitive
notion of “ length” in the real world .

A normed space is a linear space with norm on it .

For x and y in X , let

d(x,y) = ||x - y||

Then d is a metric on X so that (X,d) is a metric space , thus every normed space is a metric
space

➢ Every normed linear space is a metric space . But converse may not be true .

Example :
|𝑥 − 𝑦|
d(x,y) = , ∀ x , y ∈X
1 + |𝑥 − 𝑦|

|𝑥 − 𝑦|
⇒ ||x - y|| =
1 + |𝑥 − 𝑦|

|𝑧|
⇒ ||z|| = , z = x - y ∈X
1 + |𝑧|

10
 |α𝑧|
||αz|| =
1 +|α𝑧|

|α| |𝑧|
=
1 + |α| |𝑧|

= |α| ( |𝑧|
1 + |α| |𝑧| )
≠ |α| ||z|| .

➢ Result

Let X be a normed linear space . Then ,

| ||x|| - ||y|| | ≤ ||x - y|| , ∀x , y ∈ X

Proof :

||x|| = || ( x - y ) + y|| ≤ ||x - y|| + ||y||

⇒||x|| - ||y|| ≤||x - y|| → (1)

x↔y

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||→(2)

From (1) and (2)

|||x|| - ||y||| ≤ ||x - y||

➢ Norm is a continuous function

Let xn →x , as n → ∞

11
 ⇒ xn - x →0 as n →∞

⇒ ||xn - x||→0 as n→∞

| ||xn|| - ||x|| | ≤ ||xn - x|| →0 as n→∞

⇒||xn|| - ||x|| →0 , as n →∞

⇒||x|| is continuous

➢ Norm is a uniformly continuous function

We have , || || :X→R . Let x,y ∈ X and ε> 0

Then ||x|| = ||x - y + y ||

≤||x - y|| + ||y||

⇒ ||x|| - ||y|| ≤ ||x - y||→ (1)

Interchanging x and y ,

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||

⇒ ||x|| - ||y|| ≥ - ||x - y|| → (2)

Combining (1) and (2)

- ||x - y|| ≤ ||x|| - ||y|| ≤ ||x - y||

That is ,

| ||x ||- ||y|| | ≤ ||x - y||

Take δ = ε , then whenever ||x - y|| < δ, | ||x|| -|| y|| |< ε

12
 Therefore || || is a uniformly continuous function .

➢ Continuity of addition and scalar multiplication

To show that + : X×X →X and ·: K ×X →X are continuous functions.

Let (x,y) ∈ X×X . To show that + is continuous at (x ,y) , that is , to show

that for each (x,y) ∈ X×X if xn→x and yn→y in X , then

+(xn , yn) →+(x , y) ;

That is ,
xn + yn →x + y .

Consider
||( xn + yn) - (x + y )|| = ||xn - x + yn - y||

≤||xn - x|| + ||yn - y||

Given xn→x and yn→y , for each ϵ > 0, ∃ N1 ∋

ε
||xn - x|| < 2
∀ n ≥N1 , and ∃ N2 ∋

ε
||yn - y|| < 2
∀ n ≥N2

\
Take N = max { N1, N2}

ε ε
Then ||xn - x|| < 2
and ||yn - y|| < 2
∀n≥N

ε ε
Therefore ||(xn + yn) - (x + y)|| ≤ 2
+ 2
= ε ∀n ≥N

That is , xn + yn →x + y

Now to show that ·: K ×X→X is continuous

Let (k , x) ∈ K×X

13
 To show that if kn →k and xn →x , then knxn →kx

Since kn →k , ∀ ε > 0 ∃ N1 ∋ |kn - k| < ε ∀ n ≥ N1

2

ε
Since xn →x , ∀ ε > 0 ∃ N2 ∋ ||xn - x|| < ∀ n ≥ N2
2

Consider ||knxn - kx|| = ||knxn - kx + xnk - xnk ||

= ||xn (kn - k) + k(xn - x)||

≤ ||xn(kn - k)|| + ||k(xn - x)||

= ||xn|| |kn- k| + |k| ||xn - x||

ε ε
≤ ||xn|| 2 + |k| 2

∴ knxn→kx

➢ Examples of normed space

1) Spaces Kn (K = R or C)

For n = 1 , the absolute value of function | | is a norm on K , since ∀k ∈K

We have ,

||k|| = ||k · 1|| = |k| ||1|| , by definition .

But ||1|| is a positive scalar .

∴ ||k|| is a positive scalar multiple of the absolute value function .

∴ any norm on K is a positive scalar multiple of the absolute value

function

For n > 1 , let p≥ 1 be a real number

14
 Kn = { ( x(1) , x(2) , . . . , x(n) ) : x(i) ∈K , i = 1 , 2 , . . . , n }

For x ∈Kn , that is , x = ( x(1) , x(2) , . . . , x(n) ) , define

𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (|𝑥(1)| +. . . + |𝑥(𝑛)| )

Then || ||p is a norm on Kn

When p = 1 ,

Then , ||x||1 = |x(1)| + |x(2)| + . . . + |x(n)|

Since |x(i)| ≥0 ∀ i = 1 , 2 , . . . , n , ||x||1 ≥0

And ||x||1 = 0 ⇔ |x(1)| + . . . +|x(n)| = 0

⇔ |x(i)| = 0 ∀ i

⇔ x(i) = 0 ∀ i

⇔ x = ( x(1) , . . . , x(n) ) = 0

Now ||kx||1 = |kx(1)| + |kx(2)| + . . . + |kx(n)|

= |k| |x(1)| + . . . + |k| |x(n)|

= |k| ( |x(1)| + . . . + |x(n)| )

= |k| ||x||1

||x + y||1 = |(x + y)(1)| + . . . + |(x + y)(n)|

= |x(1) + y(1)| + . . . + |x(n) + y(n)|

≤|x(1)| + |y(1)| + . . . + |x(n)| + |y(n)|

= |x(1)| + . . . + |x(n)| + |y(1)| + . . . + |y(n)|

= ||x||1 + ||y||1

15
 Consider 1<p<∞

𝑝 𝑝 1/𝑝
Now , ||𝑥||𝑝 = ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

𝑝
Since |x(i)| ≥0 ∀i , we have ||x||p ≥0

𝑝 𝑝 1/𝑝
And ||𝑥||𝑝 = 0⇔( |𝑥(1)| +. . . + |𝑥(𝑛)| ) = 0

𝑝
⇔ |𝑥(𝑖)| = 0 ∀i

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 ∀i

⇔ x = ( x(1) , . . . , x(n) ) = 0 .
Now
𝑝 𝑝 1/𝑝
||𝑘𝑥||𝑝 = ( |𝑘𝑥(1)| +. . . + |𝑘𝑥(𝑛)| )

𝑝 𝑝 𝑝 𝑝 1/𝑝
= ( |𝑘| |𝑥(1)| +. . . + |𝑘| |𝑥(𝑛)| )

𝑝 𝑝 1/𝑝
= |𝑘| ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

= |𝑘| ||𝑥||𝑝 .

𝑝 𝑝 1/𝑝
||𝑥 + 𝑦||𝑝 = ( |𝑥(1) + 𝑦(1)| +. . . + |𝑥(𝑛) + 𝑦(𝑛)| )

We have by Minkowski’s inequality ,

1/𝑝 1/𝑝 1/𝑝

( ) ( ) ( )
𝑛 𝑛 𝑛
𝑝 𝑝 𝑝
∑ |𝑥(𝑖) + 𝑦(𝑖)| ≤ ∑ |𝑥(𝑖)| + ∑ |𝑦(𝑖)|
𝑖=1 𝑖=1 𝑖=1
Then

16
 𝑝 𝑝 1/𝑝 𝑝 𝑝 1/𝑝
(
||𝑥 + 𝑦||𝑝 ≤ |𝑥(1)| +. . . + |𝑥(𝑛)| ) (
+ |𝑦(1)| +. . . + |𝑦(𝑛)| )
= ||𝑥||𝑝 + ||𝑦||𝑝

Then , for 1≤p<∞ ,|| ||𝑝 is a norm on Kn

When p = ∞, define ||𝑥||∞ = 𝑚𝑎𝑥 { |𝑥(1)| , |𝑥(2)| ,. . . , |𝑥(𝑛)| }

Then it is a norm on Kn

||𝑥||𝑝 ≥ 0 since each values |x(i)|≥0

So that

max {|x(i)| , i=1, . . . , n} ≥ 0

||𝑥||∞ = 0 ⇔𝑚𝑎𝑥 { |𝑥(𝑖)| : 𝑖 = 1,. . . , 𝑛 }= 0

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 , ∀i

⇔x = 0

||𝑘𝑥||∞ = 𝑚𝑎𝑥 { |𝑘𝑥(1)| ,. . . , |𝑘𝑥(𝑛)| }

= max { |k| |x(1)| , . . . , |k| |x(n)|}

= |k| max {|x(1)| , . . . , |x(n)|}

= |k| ||x||∞

||x + y||∞ = max { |x(1) + y(1)| , . . . , |x(n) + y(n)| }

≤max { |x(1)| + |y(1)| , . . . , |x(n)| + |y(n)| }

17
 ≤max { |x(1)| , . . . , |x(n)| } + max { |y(1)| , . . . , |y(n)| }

= ||x||∞ + ||y||∞

2) Sequence space

∞
𝑝 𝑝 𝑝
Let 1≤p < ∞, 𝑙 = { x = ( x(1) , x(2) , . . . ) ; x(i) ∈K and ∑ |x(j)| <∞} , that is , 𝑙 is the
𝑗=1
𝑝
space of p-summable scalar sequences in K . For x = (x(1) , x(2) , . . . ) ∈ 𝑙 ,

let ||x||p = ( |x(1)|p + |x(2)|p + . . . )1/p . Then it is a norm on lp.

That is , || ||p is a function from lp to R .

If p = 1 , then l1 is a linear space and ||x||1 = ( |x(1)| + |x(2)| + . . . ) is a norm on l1

∞
Let p = ∞. Then 𝑙 is the linear space of all bounded scalar sequences . And ,

||x||∞ = sup { |x(j)| : j = 1, 2, 3, . . . }

∞
Then || ||∞ is a norm on 𝑙

18
 CHAPTER 2

THEOREMS ON NORMED SPACES

a) Let Y be a subspace of a normed space X , then Y and its closure 𝑌 are normed spaces with
the induced norm.
b) Let Y be a closed subspace of a normed space X , for x +Y in the quotient space X/Y, let
|||x +Y||| = inf { ||x+y|| : y ∈Y} . Then ||| ||| is a norm on X/Y , called the quotient norm.

A sequence (xn + Y) converges to x + Y in X/Y iff there is a sequence (yn) in Y , (xn+ yn)
converges to x in X.

c) Let || ||pbe a norm on the linear space Xp , j = 1,2,…. . Fix p such that 1 ≤p ≤∞

For x = (x(1) , x(2) , … , x(m)) that is the product space X =X1× X2 × …× Xm ,

1/𝑝
Let ||𝑥||𝑝 = ( ||𝑥(1)|| 𝑝
1
𝑝
+ ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚
𝑝
) , if 1≤ p <∞

||x||p = max { ||x(1)||1 , … , ||x(m)||m } , if p = ∞.

Then || ||p is a norm on X.

A sequence (xn) converges to x in X ⇔ (xn(j)) converges to x(j) in Xj ∀ j=1,2,…,m.

Proof:
a) Since X is a normed space, there is a norm on X to Y . Since Y is a subspace of X,

|| ||y: Y → R is a function. To show that || ||y is a norm on Y.

For y∈ Y , || y||Y = ||y|| , then

||y||Y ≥ 0 ( ∵||y||≥ 0 ) and ||y||Y = 0 ⇔ y = 0

||ky||Y = ||ky|| = |k| ||y|| = |k| ||y||y .

Let y1 , y2 ∈Y. Then ,

||𝑦1 + 𝑦2||𝑦 = ||𝑦1 + 𝑦2|| ≤ ||𝑦1|| + ||𝑦2|| = ||𝑦1||𝑦 + ||𝑦2||𝑦

Now the continuity of addition and scalar multiplication shows that 𝑌is a subspace of X, since if
xn →x and yn → y , xn , yn ∈ 𝑌 , then

xn + yn →x + y (by continuity of addition) and

19
 kxn →kx (by continuity of scalar Xn) .

Since 𝑌is closed , x + y ∈ 𝑌 and kx ∈ 𝑌 . Therefore 𝑌 ≤ X.

∴ norm on X induces a norm on Y and 𝑌

b) X/Y , the quotient space equals X/Y={ x + Y : x ∈X }.

|||x + y||| = inf { ||x + y|| : y ∈Y }

Claim: ||| ||| is a norm on X/Y , called quotient norm

• Let x ∈X ,

|||x + Y||| = inf { ||x + y|| : y ∈Y } ≥ 0.

∴ |||x + Y||| ≥0 .

If |||x + y||| = 0 ( 0 in X/Y is Y) , then there is a sequence (yn) in Y ∋

||x + yn || →0

⇒ x + yn →0

⇒ yn →-x

Since yn ∈Y and Y is closed

-x ∈Y ⇔ x ∈Y (∵ Y is a subspace)

⇔x + Y = Y , zero in X/Y.

• For k ∈K ,

|||k(x + Y)||| = |||kx + Y|||

= inf { ||k(x + y)|| : y ∈Y}

= inf { |k| ||x + y|| : y ∈Y}

= |k| inf { ||x + y|| : y ∈Y}

= |k| |||x + Y||| .

• Let x1 , x2 ∈X . Then

|||x1 + Y||| = inf { ||x1 + y|| : y ∈Y } . Then ∃ y1 ∈ Y ∋

ε
|||x1 + Y||| + > ||x1 + y1|| , and
2

20
 |||x2 + Y||| = inf { ||x2 + y|| : y ∈Y} , Then ∃y2 ∈Y ∋
ε
|||x2 + Y||| + 2
> ||x2 + y2|| .

||x1 + y1 + x2 + y2 || ≤ ||x1 + y1|| + ||x2 + y2||

ε ε
≤ |||x1 + Y||| + 2 + |||x2 + Y||| + 2

Let y = y1 + y2 ∈Y . Then ,

||(x1+x2) + y|| ≤ |||x1 + Y||| + |||x2 + Y||| + ℇ —(1)

Now , |||(x1 + Y) + (x2 + Y)||| = |||x1 + x2 + Y|||

=inf { ||x1 + x2 + y|| : y ∈Y }

< ||x1 + x2 + y||

≤ |||x1 + Y||| + |||x2 + Y||| + ℇ (by (1) )

since ℇ is arbitrary , we have

|||(x1 + Y) + (x2 + Y)||| ≤|||x1 + Y||| + |||x2 + Y|||

∴ ||| ||| is a norm on X/Y.

Let (xn + Y) be a sequence in X/Y . Assume that (yn) is a sequence in Y ∋ (xn + yn) converges
to x in X.
That is , (xn - x + yn) converges to 0 . —(1)
Claim: (xn + Y) converges to x + Y.
Consider
|||xn + Y - (x+Y)||| = |||(xn - x) + Y|||

= inf { ||xn - x + yn|| : y ∈Y }

≤ ||xn - x + yn|| ∀yn ∈Y .

Then by (1) , xn + Y converges to x + Y in X/Y.

Conversely assume that the sequence (xn + Y) → x + Y in X/Y.

Consider |||xn + Y - (x + Y)||| = |||xn - x + Y|||

= inf { ||xn -x + y|| : y ∈Y }

Then we can choose yn ∈Y ∋

21
 1
||xn - x + yn|| < |||(xn - x) + Y||| + 𝑛 , n=1,2,3,….

Since xn+Y →x+Y , we get

(xn - x + yn) converges to zero as n →∞

That is , (xn + yn) converges to x in X as n →∞

c) Consider 1≤ p < ∞

Given that
𝑝 𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (||𝑥(1)||1 + ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚)

Clearly , ||x||p ≥0 .
𝑝
Since each ||𝑥(𝑖)||𝑖 ≥ 0 .

𝑝
||x||p = 0 ⇔ |𝑥(𝑗)|𝑗 = 0 ∀ j = 1, . . . ,m

⇔ x(j) = 0 ∀ j.

⇔ x = (x(1), . . . ,x(m)) = 0
1/𝑝
( 𝑝
||kx||p = ||𝑘𝑥(1)||1 +. . . + ||𝑘𝑥(𝑚)||𝑚
𝑝
)
1/𝑝
= ( |𝑘| ||𝑥(1)|| )
𝑝 𝑝 𝑝 𝑝
1
+. . . + |𝑘| ||𝑥(𝑚)||𝑚

1/𝑝
( 𝑝
= |𝑘| ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= |𝑘| ||𝑥||𝑝 , k∈K and x∈X

1/𝑝
(
Now, ||𝑥 + 𝑦||𝑝 = ||𝑥(1) + 𝑦(1)||1 +. . . + ||𝑥(𝑚) + 𝑦(𝑚)||𝑚
𝑝 𝑝
)
(by Minkowski’s inequality)

≤
1/𝑝
( (||𝑥(1)||1
+ ||𝑦(1)||1 )𝑝
(
+. . . + ||𝑥(𝑚)||𝑚 + ||𝑦(𝑚)||𝑚 ))
𝑝

1/𝑝 1/𝑝

( ) ( )
𝑚 𝑚
𝑝 𝑝
≤ ∑ ||𝑥(𝑗)||𝑗 + ∑ ||𝑦(𝑗)||𝑗 (Minkowski’s inequality )
𝑗=1 𝑗=1

22
 1/𝑝
( 𝑝
= ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= ||x||p + ||y||p

Now suppose p =∞

||x||∞ = max { ||x(1)||1 , . . . , ||x(m)|| m }

||x||∞ ≥0 Since ||x(j)|| ≥0 , ∀j

||x||∞ = 0 ⇔ ||x(m)|| = 0 ∀m

⇔ x(m) = 0 ∀m

⇔x=0

||kx||∞ = max { ||kx(1)||1 , . . . , ||kx(m)||m }

= |k| max { ||x(1)||1 , . . . , ||x(m)||m }

= |k| ||x||∞

||x + y||∞ = max { ||x(1) + y(1)||1, . . . , ||x(m) + y(m)||m }

≤max { ||x(1)||1 + ||y(1)||1 , . . . , ||x(m)||m + ||y(m)||m }

= max { ||x(1)||1 , . . . , ||x(m)||m } + max { ||y(1)||1 , . . . , ||y(m)||m }

= ||x||∞ + ||y||∞

We now consider ,

𝑝 𝑝 1/𝑝
( )
||𝑥𝑛 − 𝑥(1)||𝑝 = ||𝑥𝑛(1) − 𝑥(1 ||1 +. . . + ||𝑥𝑛(𝑚) − 𝑥(𝑚)||𝑚)

Then

xn →x in X ⇔ ||xn - x ||p → 0
𝑝
⇔ ||xn(j) - x(j)||𝑗 → 0

⇔ xn(j) - x(j) → 0

⇔ xn(j) → x(j) in X ∀j .

23
 RIESZ LEMMA

Let 𝑋 be a normed space . 𝑌be a closed subspace of 𝑋and 𝑋 ≠ 𝑌 . Let 𝑟be a real number
such that 0 < 𝑟 < 1 . Then there exist some xr ∈X such that ||xr|| = 1 and

r<dist ( xr , Y )≤1

Proof :
We have ,

dist (x , Y) = inf { d(x , y) : y∈Y}

= inf { ||x - y|| : y ∈Y}

Since Y ≠X , consider x ∈X ∋ x∉Y.

If dist(x , Y) = 0 , then ||x - y|| = 0 ⇒ x∈𝑌= Y ( ∵ Y is closed )

Therefore ,

dist (x , Y) ≠0

That is ,
dist (x , Y) > 0
1
Since 0 < r < 1 , >1
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
⇒ > dist (x , Y)
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
That is , is not a lower bound of { ||x - y|| : y ∈Y }
𝑟

𝑑𝑖𝑠𝑡(𝑥 , 𝑌)
Then ∃ y0 ∈Y ∋ ||x - y0|| <
𝑟
→(1)
𝑥 − 𝑦0
Let xr = . Then xr ∈X
||𝑥 − 𝑦0||

( ∵y0 ∈Y , x∉Y ⇒x - y0 ∈X and ||x - y0|| ≠0 )

24
 𝑥 − 𝑦0 ||𝑥 − 𝑦0||
Then ||xr|| = || || = =1
||𝑥 − 𝑦 ||0
||𝑥 − 𝑦0||

Now to prove r < dist(𝑥𝑟, 𝑌)≤1

We have dist(xr , Y) = inf { ||xr - y|| : y∈Y }

≤||xr - y|| ∀y∈Y

In particular, 0∈ 𝑌, so that dist(xr , Y) ≤||xr - 0|| = 1

That is ,

dist (xr , Y) ≤1

Now ,
𝑥 − 𝑦0
dist (xr , Y) = dist ( ,Y )
||𝑥−𝑦0||

1
= dist ( x - y0 , Y)
||𝑥−𝑦0||

1
= inf { ||x - y0 - y|| : y∈ 𝑌}
||𝑥−𝑦0||

1
= inf {||x - (y0+ y)|| : y0 + y∈ 𝑌}
||𝑥−𝑦0||

1
= dist (x , Y)
||𝑥−𝑦0||

𝑟
> dist (x , Y) by (1)
𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)

⇒ dist (xr , Y) > r

That is ,

r < dist (xr , Y) ≤ 1

25
 CONCLUSION

This project discusses the concept of normed linear space that is fundamental to
functional analysis . A normed linear space is a vector space over a real or complex
numbers ,on which the norm is defined . A norm is a formalization and generalization to
real vector spaces of the intuitive notion of “length” in real world
In this project , the concept of a norm on a linear space is introduced and thus
illustrated . It mostly includes the properties of normed linear spaces and different proofs
related to the topic.

26
 BIBLIOGRAPHY

➢ Balmohan V Limaye , Functional Analysis

➢ S.H Friedberg , Arnold J.Insel and Lawrence E.Spence , Linear Algebra , 2 nd Edition, PH
Inc .
➢ S. Narayanan and Mittai , A Text Book of Matrices , Revised Edition, S. Chand

27
POWER SERIES SOLUTIONS AND
SPECIAL FUNCTIONS
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
P K NIVEDITHA
DB18CMSR24
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Power Series’ is a bona fide project of

P K NIVEDITHA carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I P K NIVEDITHA hereby declare that the project ‘Power Series’ is an

original record of studies and bona fide project carried out by me during the
period of 2018 – 2021 under the guidance of Ms. Athulya P, Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu and has not
submitted by me elsewhere for the award of my degree, diploma, title or
recognition, before.

Name
P K NIVEDITHA

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express my deepest gratitude to
people along the way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my
project supervisor Ms. Athulya P, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and
the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every
one of them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- SERIES SOLUTION OF FIRST 4-9

ORDER EQUATION

04 Chapter 2- SECOND ORDER LINEAR 9-18

EQUATION, ORDINARY POINTS

05 Conclusion 19

06 Bibliography 20
 INTRODUCTION
A power series is a type of series with terms involving a variable. Power series
are often used by calculators and computers to evaluate trigonometric,
hyperbolic, exponential and logarithm functions. So any application of these
kind of functions is a possible application of power series. Many interesting and
important differential equations can be found in power series.

1
 PRELIMINERY

A. An infinite series of the form

∞

∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ (1)
𝑛=0

is called a power series in x. The series

∞

∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛 = 𝑎0 + 𝑎1 (𝑥 − 𝑥0 ) + 𝑎2 (𝑥 − 𝑥0 )2 + ⋯
𝑛=0

is a power series in x – x0.

B. The series (1) is said to converge at a point x if the limit

𝑚

𝑙𝑖𝑚 ∑ 𝑎𝑛 𝑥 𝑛
𝑚→∞
𝑛=0

exists, and in this case the sum of the series is the value of this limit.
Radius of convergence: Series in 𝑥 has a radius of convergence 𝑅, where
0 ≤ 𝑅 ≤ ∞, with the property that the series converges if |𝑥| < 𝑅 and
diverges if |𝑥| > 𝑅. It should be noted that if 𝑅 = 0 then no 𝑥 satisfies
|𝑥| < 𝑅, and if 𝑅 = ∞ then no 𝑥 satisfies |𝑥| > 𝑅
𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | | , if the limit exists.
𝑛→∞ 𝑎𝑛+1

C. Suppose that (1) converges for |𝑥| < 𝑅 with 𝑅 > 0, and denote its sum
by f(x):
∞

𝑓(𝑥) = ∑ 𝑎𝑛 𝑥 𝑛 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑛=0

Then f(x) is automatically continuous and has derivatives of all orders for
|𝑥| < 𝑅.

2
D. Let f(x) be a continuous function that has derivatives of all orders for
|x|< R with R > 0. f(x) be represented as power series using Taylor’s
formula:
𝑛
𝑓 (𝑘) (0) 𝑘
𝑓(𝑥) = ∑ 𝑥 + 𝑅𝑛 (𝑥)
𝑘!
𝑘=0

where the remainder Rn (x) is given by

𝑓 (𝑛+1) (𝑥̅ ) 𝑛+1
𝑅𝑛 (𝑥) = 𝑥
(𝑛 + 1)!
for some point 𝑥̅ between 0 and x.

E. A function f(x) with the property that a power series expansion of

the form

𝑓(𝑥) = ∑ 𝑎𝑛 (𝑥 − 𝑥0 )𝑛
𝑛=0

is valid in some neighbourhood of the point x0 is said to be analytic at

x0. In this case the an are necessarily given by
𝑓 (𝑛) (𝑥0 )
𝑎𝑛 =
𝑛!
and is called the Taylor series of f(x) at x0.

Analytic functions: A function f defined on some open subset U of R or C is

called analytic if it is locally given by a convergent power series. This means
that every a ∈ U has an open neighbourhood V ⊆ U, such that there exists
a power series with centre a that converges to f(x) for every x ∈ V.

3
 CHAPTER 1
SERIES SOLUTION OF FIRST ORDER EQUATION
We have studied to solve linear equations with constants coefficient but with
variable coefficient only specific cases are discussed. Now we turn to these
latter cases and try to find a general method to solve this. The idea is to assume
that the unknown function y can be explained into a power series. Our purpose
in this section is to explain the procedures by showing how it works in the case
of first order equation that are easy to solve by elementary methods.

Example 1: we consider the equation

𝑦ʹ = 𝑦
Consider the above equation as (1). Assume that y has a power series solution
of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |x| < R, R > 0

That is we assume that 𝑦 ʹ = 𝑦 has a solution that is analytic at origin. We have
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ ⋯
then
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ … ….
∴ (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 ⋯
= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
⇒ 𝑎1 = 𝑎0
𝑎1 𝑎0
2𝑎2 = 𝑎1 ⇒ 𝑎2 = =
2 2

4
 𝑎2 𝑎0 𝑎0
3𝑎3 = 𝑎2 ⇒ 𝑎3 = = =
3 2 ∙ 3 3!
𝑎3 𝑎0 𝑎0
4𝑎4 = 𝑎3 ⇒ 𝑎4 = = =
4 2 ⋅ 3 ⋅ 4 4!
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0 𝑎0
= 𝑎0 + 𝑎0 𝑥 + 𝑥 2 + 𝑥 3 + 𝑥 4 + ⋯
2 3! 4!
𝑥2 𝑥3 𝑥4
= 𝑎0 (1 + 𝑥 + + + + ⋯ )
2! 3! 4!
𝑦 = 𝑎0 𝑒 𝑥
To find the actual function we have 𝑦 ʹ = 𝑦
𝑑𝑦 𝑑𝑦
i.e., =𝑦 ⇒ = 𝑑𝑥
𝑑𝑥 𝑦
integrating
log 𝑦 = 𝑥 + 𝑐
i.e., 𝑦 = 𝑒 𝑥+𝑐 = 𝑒 𝑥 ⋅ 𝑒 𝑐
𝑦 = 𝑎0 𝑒 𝑥 , where a0 = ec , a constant.

Example 2: solve 𝑦 ′ = 2𝑥𝑦. Also find its actual solution.

Solution: 𝑦 ′ = 2𝑥𝑦 (1)
Assume that y has a power series of the form
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

Which converges for |𝑥| < 𝑅, 𝑅 > 0

∞

We have 𝑦 = ∑ a𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

5
 = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1) ⇒ 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 2𝑥(𝑎0 + 𝑎1 𝑥 +𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯ )
= 2𝑥𝑎0 + 2𝑥𝑎1 𝑥 + 2𝑥𝑎2 𝑥 2 + 2𝑥𝑎3 𝑥 3 + ⋯
= 2𝑥𝑎0 + 2𝑎1 𝑥 2 + 2𝑎2 𝑥 3 + 2𝑎3 𝑥 4 + ⋯ … ..
2𝑎0
⇒ 𝑎1 = 0 2𝑎2 = 2𝑎0 ⇒ 𝑎2 = = 𝑎0
𝑧
2𝑎1
3. 𝑎3 = 2𝑎1 ⇒ 𝑎3 = =0
3
2𝑎2 𝑎0
4𝑎4 = 2𝑎2 ⇒ 𝑎4 = =
42 2

5𝑎5 = 2𝑎3 = 0 ⇒ 𝑎5 = 0
2𝑎4 𝑎4 𝑎0 𝑎0
6𝑎6 = 2𝑎4 ⇒ 𝑎6 = = = =
6 3 2⋅3 3!

We get,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0
= 𝑎0 + 0 + 𝑎0 𝑥 2 + 0𝑥 3 + 𝑥 4 + ⋯
2
𝑎 0
= 𝑎0 + 𝑎0 𝑥 2 + 𝑥 4 + ⋯
2
2
𝑥4 𝑥6
= 𝑎0 (1 + 𝑥 + + + ⋯ )
2! 3!
2
𝑦 = 𝑎0 𝑒 𝑥
To find an actual solution

𝑦 ′ = 2𝑥𝑦
𝑑𝑦
= 2𝑥𝑦
𝑑𝑥
𝑑𝑦
⇒ = 2𝑥 ⋅ 𝑑𝑥
𝑦
log 𝑦 = 𝑥 2 + 𝑐
2
𝑦 = 𝑒𝑥 + 𝑐
2
⇒ 𝑦 = 𝑎0 𝑒 𝑥 , where 𝑎0 = 𝑒 𝑐

6
Example 3: Consider 𝑦 = (1 + 𝑥)𝑝 where p is an arbitrary constant. Construct a
differential equation from this and then find the solution using power series
method.
Solution
First, we construct a differential equation
i.e. 𝑦 = (1 + 𝑥)𝑝
𝑝(1+𝑥)𝑝 𝑝𝑦
𝑦 ′ = 𝑝(1 + 𝑥)𝑝−1 = =
1+𝑥 1+𝑥

∴ (1 + 𝑥)𝑦 ′ = 𝑝𝑦, 𝑦(0) = 𝑟

Assume that y has a power series solution of the form,
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0

= 𝑎0 + 𝑎1 𝑥 + 𝑎̇ 2 𝑥 2 + ⋯ … …
Which converges for |𝑥| < 𝑅̇, 𝑅 > 0
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ … … ..
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=0

= 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯
Then (1 + 𝑥)𝑦 ′ = 𝑝𝑦
⇒ (1 + 𝑥)𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ = 𝑝(𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
⇒ (𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) + (𝑎1 𝑥 + 2𝑎2 𝑥 2 + 3𝑎3 𝑥 3 + ⋯ )
= 𝑎0 𝑝 + 𝑎1 𝑝𝑥 + 𝑎2 𝑝𝑥 2 + ⋯
Equating the coefficients of 𝑥, 𝑥 2 , …
𝑎1 = 𝑎0 𝑝 i.e. 𝑎1 = 𝑝, (since 𝑎0 = 1)
⇒ 2𝑎2 = 𝑎1 (p − 1)
𝑎1 (p − 1) 𝑎0 𝑃(𝑝 − 1)
𝑎2 = =
2 2

7
 3𝑎3 + 2𝑎2 = 𝑎2 𝑝
𝑠𝑎3 = 𝑎2 𝑝 − 2𝑎2
= 𝑎2 (𝑝 − 2)
𝑎2 (𝑝 − 2) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)
𝑎3 = =
3 2⋅3
4𝑎4 + 3𝑎3 = 𝑎3 𝑝
4𝑎4 = 𝑎3 𝑝 − 3𝑎3
= 𝑎3 (𝑝 − 3)
𝑎3 (𝑝 − 3) 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3)
𝑎4 = =
4 2⋅3⋅4
∴ we get,
𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + 𝑎3 𝑥 3 + ⋯
𝑎0 𝑝(𝑝 − 1) 2 𝑎0 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 𝑎0 + 𝑎0 𝑝𝑥 + 𝑥 + 𝑥 + ⋯…
2 2⋅3
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1)(𝑝 − 2) 3
= 1 + 𝑝𝑥 + 𝑥 + 𝑥 +
2! 3!
𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − 3) 4 𝑝(𝑝 − 1)(𝑝 − 2)(𝑝 − (𝑛 − 1)) 𝑛
𝑥 + ⋯+ 𝑥
4! 𝑛!
Since the initial problem y(0) = 1 has one solution the series converges for |x|<1
So this is a power solution,
𝑝(𝑝 − 1) 2 𝑝(𝑝 − 1) ⋯ (𝑝 − (𝑛 − 1)) 𝑛
(1 + 𝑥)𝑝 = 1 + 𝑝𝑥 + 𝑥 + ⋯+ 𝑥
2! 𝑛!
Which is binomial series.

Example 4: Solve the equation 𝑦 ′ = 𝑥 − 𝑦, 𝑦 (0) = 0

Solution: Assume that y has a power series solution of the form
∞

𝑦 = ∑ an 𝑥 𝑛
𝑛=0

which converges for |𝑥| < 𝑅, 𝑅 > 0

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯

8
 Now 𝑦 ′ = 𝑥 − 𝑦
(𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ ) = 𝑥 − (𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ )
Equating the coefficients of 𝑥, 𝑥 2 ,
𝑎1 = −𝑎0 = 0, Since 𝑦(0) = 0
2𝑎2 = 1 − 𝑎1
=1−0
1
⇒ 𝑎2 =
2
3𝑎3 = −𝑎2
−𝑎2 1
𝑎3 = =−
3 2⋅3
4𝑎4 = −𝑎3
1
⇒ 𝑎4 =
2⋅3⋅4
𝑥2 𝑥3 𝑥4
∴ 𝑦 = 0 + 0 + − + − ⋯……
2! 3! 4!
𝑥2 𝑥3
= (1 − 𝑥 + − + ⋯ ) + 𝑥 − 1
2! 3!
= 𝑒 −𝑥 + 𝑥 − 1
By direct method
𝑦′ = 𝑥 − 𝑦 𝑑𝑦
𝑑𝑦 𝑑𝑦 ( + 𝑝𝑦 = 𝑄 𝑓𝑜𝑟𝑚)
=𝑥−𝑦 ⇒ +𝑦 =𝑥 𝑑𝑥
𝑑𝑥 𝑑𝑥
here 𝑃(𝑥) = 1, integrating factor
= 𝑒 ∫ 𝑝(𝑥)⋅𝑑𝑥
= 𝑒𝑥
∴ 𝑦𝑒 𝑥 = ∫ 𝑥𝑒 𝑥 ⋅ 𝑑𝑥
𝑦𝑒 𝑥 = 𝑥 ⋅ 𝑒 𝑥 − ∫ 𝑒 𝑥 ⋅ 𝑑𝑥
= 𝑥𝑒 𝑥 − 𝑒 𝑥
𝑦𝑒 𝑥 = 𝑒 𝑥 (𝑥 − 1) + 𝑐
𝑒 𝑥 (𝑥 − 1) + 𝑐 𝑐
𝑦= = 𝑥 − 1 + 𝑥 = 𝑐𝑒 −𝑥 + (𝑥 − 1)
𝑑𝑥 𝑒
∴ 𝑦 = (𝑥 − 1) + 𝑐𝑒 −𝑥

9
 CHAPTER 2
SECOND ORDER LINEAR EQUATION, ORDINARY POINTS

Consider the general homogeneous second order linear equation,

𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0 (1)
As we know, it is occasionally possible to solve such an equation in terms of
familiar elementary functions. This is true, for instance, when P(x) and Q(x)
are constants, and in a few other cases as well. For the most part, however,
the equations of this type having the greatest significance in both pure and
applied mathematics are beyond the reach of elementary methods, and can
only be solved by means of power series.
P(x) and Q(x) are called coefficients of the equation. The behaviour of its
solutions near a point x0 depends on the behaviour of its coefficient functions
P(x) and Q(x) near this point. we confine ourselves to the case in which P(x) and
Q(x) are well behaved in the sense of being analytic at x0, which means that
each has a power series expansion valid in some neighbourhood of this point. In
this case x0 is called an ordinary point of equation (1). Any point that is not an
ordinary point of (1) is called a singular point.

Consider the equation,

𝑦 ′′ + 𝑦 = 0 (2)

the coefficient functions are P(x) = 0 and Q(x) = 1, These functions are analytic
at all points, so we seek a solution of the form,

𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯ + 𝑎𝑛 𝑥 𝑛 + ⋯ (3)

Differentiating (3) we get,

𝑦 ′ = 𝑎1 + 2𝑎2 𝑥 + 3𝑎3 𝑥 2 + ⋯ + (𝑛 + 1)𝑎𝑛+1 𝑥 𝑛 + ⋯ (4)

10
And

𝑦 ′′ = 2𝑎2 + 2 ⋅ 3𝑎3 𝑥 + 3 ⋅ 4𝑎4 𝑥 2 + ⋯ + (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ⋯ (5)

If we substitute (5) and (3) into (2) and add the two series term by term, we get

(2𝑎2 + 𝑎0 ) + (2 ⋅ 3𝑎3 + 𝑎1 )𝑥 + (3 ⋅ 4𝑎4 + 𝑎2 )𝑥 2 +

𝑦 ′′ + 𝑦 = =0
(4 ⋅ 5𝑎5 + 𝑎3 )𝑥 3 + ⋯ + [(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 ]𝑥 𝑛 + ⋯
and equating to zero the coefficients of successive powers of x gives
2𝑎2 + 𝑎0 = 0, 2 ⋅ 3𝑎3 + 𝑎1 = 0, 3 ⋅ 4𝑎4 + 𝑎2 = 0
4 ⋅ 5𝑎5 + 𝑎3 = 0, … … , (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑎𝑛 = 0, …
By means of these equations we can express an in terms of a0 or a0, according
as n is even or odd:
𝑎0 𝑎1 𝑎2 𝑎0
𝑎2 = − , 𝑎3 = − , 𝑎4 = − =
2 2⋅3 3⋅4 2⋅3⋅4
𝑎3 𝑎1
𝑎5 = − = ,⋯
4⋅5 2⋅3⋅4⋅5
With these coefficients, (3) becomes
𝑎0 𝑎1 3 𝑎0 𝑎1
𝑦 = 𝑎0 + 𝑎1 𝑥 − 𝑥 2 − 𝑥 + 𝑥4 + 𝑥5 − ⋯
2 2⋅3 2⋅3⋅4 2⋅3⋅4⋅5
𝑥2 𝑥4 𝑥3 𝑥5
= 𝑎0 (1 − + − ⋯ ) + 𝑎1 (𝑥 − + − ⋯ ) (6)
2! 4! 3! 5!
𝑖. 𝑒, 𝑦 = 𝑎0 cos 𝑥 + 𝑎1 sin 𝑥
Since each of the series in the parenthesis converges for all x. This implies the
series (2) for all x.

Solve the legenders equation,

(1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0
Solution
Consider (1 − 𝑥 2 )𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
11
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2

put 𝑛 = 𝑛 + 2 (Since 𝑦 ′′ is not 𝑥 𝑛 form )

∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛+2 𝑥 𝑛+2−2

𝑛=0
∞

∴ 𝑦 ′′ = ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

Now (1) ⇒ 𝑦 ′′ − 𝑥 2 𝑦 ′′ − 2𝑥𝑦 ′ + 𝑝(𝑝 + 1)𝑦 = 0

⇒ ∑(𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 − ∑𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛 − ∑2𝑛𝑎𝑛 𝑥 𝑛 + ∑𝑝(𝑝 + 1)𝑎𝑛 𝑥 𝑛 = 0
∞

⇒ ∑ [((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 )𝑥 𝑛 ] = 0

𝑛=0

for n = 0,1,2,3…….
⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 − 𝑛(𝑛 − 1)𝑎𝑛 − 2𝑛𝑎𝑛 + 𝑝(𝑝 + 1)𝑎𝑛 = 0
[𝑛(𝑛 − 1) + 2𝑛 − 𝑝(𝑝 + 1)]
⇒ 𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(𝑛2 − 𝑛 + 2𝑛 − 𝑝2 − 𝑝)𝑎𝑛
=
(𝑛 + 1)(𝑛 + 2)
(𝑛 + 𝑛 − 𝑝2 − 𝑝)𝑎𝑛
2
=
(𝑛 + 1)(𝑛 + 2)

12
 −(𝑝 − 𝑛)(𝑝 + 𝑛 + 1)
∴ 𝑎𝑛+2 = 𝑎𝑛 , 𝑛 = 0,1,2 …
(𝑛 + 1)(𝑛 + 2)

This is an Recursion formula

−𝑝(𝑝 + 1)
put 𝑛 = 0, 𝑎2 = 𝑎0
1⋅2
−(𝑝 − 1)(𝑝 + 2)
𝑛 = 1, 𝑎3 = ⋅ 𝑎1
2⋅3
−(𝑝 − 2)(𝑝 + 3)
𝑛 = 2, 𝑎4 = 𝑎2
3𝑖4
𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3)
= 𝑎0
4!
−(𝑝 − 3)[𝑝 + 4)
𝑛 = 3, 𝑎5 = 𝑎3
4⋅5
(𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4)
= 𝑎1
5!
−(𝑝 − 4)(𝑝 + 5)
𝑛 = 4, 𝑎6 = 𝑎4
5⋅6
−𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5)
= 𝑎0
6!
(𝑝 − 5)(𝑝 + 6)
𝑛 = 5, 𝑎7 = − 𝑎5
6⋅7
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6)
=− 𝑎1
7!

𝑝(𝑝 + 1) 2 𝑝(𝑝 − 2)(𝑝 + 1)(𝑝 + 3) 4

𝑦 = 𝑎0 [1 − 𝑥 + 𝑥
2! 4!
𝑝(𝑝 − 2)(𝑝 − 4)(𝑝 + 1)(𝑝 + 3)(𝑝 + 5) 6
− 𝑥 + ⋯]
6!
(𝑝 − 1)(𝑝 + 2) 3 (𝑝 − 1)(𝑝 − 3)(𝑝 + 2)(𝑝 + 4) 5
+𝑎1 [𝑥 − 𝑥 + 𝑥
3! 5!
(𝑝 − 1)(𝑝 − 3)(𝑝 − 5)(𝑝 + 2)(𝑝 + 4)(𝑝 + 6) 7
− 𝑥 + ⋯]
7!

13
Find the general solution of (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 in terms of power
series in 𝑥. Can you express this solution by means of elementary functions?
Solution
Consider the equation (1 + 𝑥 2 )𝑦 ′′ + 2𝑥𝑦 ′ − 2𝑦 = 0 as equation (1)
Assume that y has a power series solution of the form
𝑦 = ∑𝑎𝑛 𝑥 𝑛
Which converges |𝑥| < 𝑅, 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

2𝑥𝑦 ′ = ∑ 2𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

(1 + 𝑥 2 )𝑦 ′′ = 𝑦 ′′ + 𝑥 2 𝑦 ′′
∞

𝑥 2 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛
𝑛=2
∞

Now 𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2

put 𝑛 = 𝑛 + 2
∞

⇒ ∑ (𝑛 + 2)(𝑛 + 2 − 1)𝑎𝑛 + 2𝑥 𝑛+2=2

𝑛=0
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0

14
 ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛 + ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥𝑛

𝑛=0 𝑛=2
∞ ∞

+ ∑ 2𝑛𝑎𝑛 𝑥 𝑛 − ∑ 2𝑎𝑛 𝑥 𝑛 = 0
𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛(𝑛 − 1)𝑎𝑛 + 2𝑛𝑎𝑛 − 2𝑎𝑛 = 0

[−𝑛(𝑛 − 1) − 2𝑛 + 2]
𝑎𝑛+2 = 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)
(−𝑛2 + 𝑛 − 2𝑛 + 2)
= 𝑎𝑛
(𝑛 + 1)(𝑛 + 2)

2 2𝑎0
put 𝑛 = 0, 𝑎2 = 𝑎0 = = 𝑎0
1⋅2 2!
(1 − 1 − 2 + 2)
𝑛 = 1, 𝑎3 = 𝑎1 = 0
2⋅3
2−4−4+2 −4 −𝑎0
𝑛 = 2, 𝑎4 = 𝑎2 = 𝑎0 =
3⋅4 3⋅4 3
3 − 9 − 16 + 2
𝑛 = 3, 𝑎5 = 𝑎3 = 0
4.5
4 − 16 − 8 + 2 −3 3𝑎0 𝑎0
𝑛 = 4, 𝑎6 = 𝑎4 = 𝑎4 = =
5.6 5 3.5 5

∴ 𝑦 = 𝑎0 + 𝑎1 𝑥 + 𝑎2 𝑥 2 + ⋯
𝑎0 𝑎0
= 𝑎0 + 𝑎1 𝑥 + 𝑎0 𝑥 2 − 𝑥 4 + 𝑥 6 … .
3 5
4 6
𝑥 𝑥
= 𝑎0 [1 + 𝑥 2 − + − ⋯ ] + 𝑎1 𝑥
3 5
𝑥3 𝑥5
= 𝑎0 [1 + 𝑥 (𝑥 − + ⋯ )] + 𝑎1 𝑥
3 5
= 𝑎0 (1 + 𝑥tan−1 𝑥) + 𝑎1 𝑥

15
Consider the equation 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0
(a) Find its general solution 𝑦 = ∑𝑎𝑛 𝑥 𝑛 in the form
𝑦 = 𝑎0 𝑦1 (𝑥) + 𝑎1 𝑦2 (𝑥) where 𝑦1 (𝑥) and 𝑦2 (𝑥) are power series
(b) use the ratio test to verify that the two series 𝑦1 (𝑥) and 𝑦2 (𝑥) converges
. for all x.
Solution:
Given 𝑦 ′′ + 𝑥𝑦 ′ + 𝑦 = 0 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ (1)
Assume that y has a power series solution the form ∑a𝑛 𝑥 𝑛 which converges
for |𝑥| 𝑅 > 0
∞

𝑦 = ∑ 𝑎𝑛 𝑥 𝑛
𝑛=0
∞

𝑦 ′ = ∑ 𝑛 ⋅ 𝑎𝑛 𝑥 𝑛−1
𝑛=1
∞

𝑦 ′′ = ∑ 𝑛(𝑛 − 1)𝑎𝑛 𝑥 𝑛−2

𝑛=2
∞

= ∑ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 𝑥 𝑛
𝑛=0
∞

𝑥𝑦 ′ = ∑ 𝑛𝑎𝑛 𝑥 𝑛
𝑛=1
∞ ∞ ∞

(1) ⇒ ∑ (𝑛 + 1)(𝑛 + 2)a𝑛+2 𝑥 + ∑ 𝑛𝑎𝑛 𝑥 + ∑ 𝑎𝑛 𝑥 𝑛 = 0

𝑛 𝑛

𝑛=0 𝑛=1 𝑛=0

⇒ ∑[((𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 )𝑥 𝑛 ] = 0

⇒ (𝑛 + 1)(𝑛 + 2)𝑎𝑛+2 + 𝑛𝑎𝑛 + 𝑎𝑛 = 0

(−𝑛 − 1)𝑎𝑛 −𝑎𝑛
⇒ 𝑎𝑛+2 = =
(𝑛 + 1)(𝑛 + 2) 𝑛 + 2
𝑎0
put 𝑛 = 0, 𝑎2 = −
2
−2𝑎1 −𝑎1
𝑛 = 1, 𝑎3 = =
2⋅3 3
16
 −3𝑎2 −𝑎2 𝑎0
𝑛 = 2, 𝑎4 = = =
3⋅4 4 8
−4𝑎3 𝑎1
𝑛 = 3, 𝑎5 = =
4⋅5 15
−5𝑎4 −𝑎0
𝑛 = 4, 𝑎6 = =
5⋅6 48
𝑎0 𝑎1 𝑎0 𝑎1 𝑎0
∴ we get 𝑦 = 𝑎0 + 𝑎1 𝑥 + − 𝑥 2 − 𝑥 3 + 𝑥 4 + 𝑥 5 − 𝑥 6 + ⋯
2 3 8 15 48

𝑥2 𝑥4 𝑥6 𝑥3 𝑥5
= 𝑎0 [1 − + − + ⋯ ] + 𝑎1 [𝑥 − + + ⋯]
2 2⋅4 2⋅4⋅6 3 3.5

𝑥2 𝑥4 𝑥̇ 2
𝑤ℎ𝑒𝑟𝑒 𝑦1 (𝑥) = 1 − + − +
2 2⋅4 2⋅4⋅6

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + +⋯
3 3⋅5

𝑥2 𝑥4 𝑥6
(b) 𝑦1 (𝑥) = 1 − + − +⋯
2 2⋅4 2⋅4⋅6

𝑎𝑛
𝑅 = 𝑙𝑖𝑚 | |
𝑛→∞ 𝑎𝑛+1
(−1)𝑛 (−1)𝑛+1
= 𝑙𝑖𝑚 | / |
𝑛→∞ 2 ⋅ 4 ⋅ (2𝑛) 2 ⋅ 4 ⋅⋅ (2𝑛 + 2)

2(𝑛 + 1)
= 𝑙𝑖𝑚 | |
𝑛→∞ −1
1
= 𝑙𝑖𝑚 | − 2𝑛(1 + )| = ∞
𝑛→∞ 𝑛

∴ 𝑦1 (𝑥) converges for all 𝑥

𝑥3 𝑥5
𝑦2 (𝑥) = 𝑥 − + −⋯
3 3⋅5

17
 (−1)𝑛 (−1)𝑛+1
𝑅 = 𝑙𝑖𝑚 | ⁄ |
𝑛→∞ 3 ⋅ 5 ⋯ (2𝑛 + 1) 3 ⋅ 5 ⋅ ⋯ (2𝑛 + 3)

(−1) ⋅ 3 ⋅ 5 ⋯ (2𝑛 + 1)(2𝑛 + 3)

= 𝑙𝑖𝑚 | |
𝑛→∞ 3 ⋅ 5 ⋯ ⋅ (2𝑛 + 1)
= 𝑙𝑖𝑚 |(−1)𝑛(2 + 3/𝑛)| = ∞
𝑛→∞

∴ 𝑦2 (𝑥) converges for all 𝑥

REGULAR SINGULAR POINTS

A singular point 𝑥0 of equation
𝑦 ′′ + 𝑃(𝑥)𝑦 ′ + 𝑄(𝑥)𝑦 = 0
is said to be regular if the functions (𝑥 − 𝑥0 )𝑃(𝑥) and (𝑥 − 𝑥0 )2 𝑄(𝑥) are
analytic, and irregular otherwise. Roughly speaking, this means that the
singularity in 𝑃(𝑥) cannot be worse than 1/(𝑥 − 𝑥0 ), and that in 𝑄(𝑥) cannot
be worse than 1/(𝑥 − 𝑥0 )2 .
If we consider Legendre’s equation in the form
2𝑥 𝑝(𝑝 + 1)
𝑦 ′′ − 𝑦 ′
+ 𝑦=0
1 − 𝑥2 1 − 𝑥2
it is clear that x = 1 and x = −1 are singular points. The first is regular because
2𝑥 (𝑥 − 1)𝑝(𝑝 + 1)
(𝑥 − 1)𝑃(𝑥) = and (𝑥 − 1)2 𝑄(𝑥) = −
𝑥+1 𝑥+1
are analytic at x = 1, and the second is also regular for similar reasons.
Example: Bessel’s equation of order p, where p is a nonnegative constant:
𝑥 2 𝑦 ′′ + 𝑥𝑦 ′ + (𝑥 2 − 𝑝2 )𝑦 = 0
If this is written in the form
1 ′ 𝑥 2 − 𝑝2
′′
𝑦 + 𝑦 + 𝑦 = 0,
𝑥 𝑥2
it is apparent that the origin is a regular singular point because𝑥𝑃(𝑥) = 1 and
𝑥 2 𝑄(𝑥) = 𝑥 2 − 𝑝2 are analytic at x = 0.

18
 CONCLUSION
The purpose of this project gives a simple account of series solution of first
order equation, second order linear equation, ordinary points. The study of these
topics given excellent introduction to the subject called ‘POWER SERIES’
we used application of power series extensively throughout this project. We
take it for granted that most readers are reasonably well acquainted with these
series from an earlier course in calculus. Nevertheless, for the benefit of those
whose familiarity with this topic may have faded slightly, we presented a brief
review of the main facts of power series.

19
 BIBLIOGRAPHY
George F. Simmons - Differential Equations With Applications and Historical
. Notes
G.Birkoff and G.C Rota - Ordinary Differential Equations; Wiley and Sons; 3rd
. Edition ( 1978 )
P.Hartmon - Ordinary Differential Equations; John Wiley and Sons

20
NUMBER THEORETIC FUNCTION
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
RON M JOJO
DB18CMSR16
Under the guidance of
Ms. Ajeena joseph

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Number Theoretic Function’ is a bona fide

project of RON M JOJO carried out the project work under my
supervision.

Mrs. Riya Baby Ms. Ajeena joseph

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I RON M JOJO hereby declare that the project ‘Number Theoretic Function’ is an
original record of studies and bona fide project carried out by me during the period of
2018 – 2021 under the guidance of Ms. Ajeena joseph, Department of Mathematics,
Don Bosco Arts and Science College, Angadikkadavu and has not submitted by me
elsewhere for the award of my degree, diploma, title or recognition, before.

Name
RON M JOJO

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the Almighty
who is always with me. Also, I must express my deepest gratitude to people along the
way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my project
supervisor Ms. Ajeena joseph, Department of Mathematics, Don Bosco Arts and
Science College, Angadikkadavu, for the invaluable guidance, loving encouragement
and meticulous care towards me throughout my career. I express my deep sense of
gratitude to all the faculty members of the Department of Mathematics, Don Bosco
Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and the
staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every one of
them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- ARITHMETIC FUNCTION 4-10

04 Chapter 2- EULER’S 𝜙 FUNCTION 11-20

05 Conclusion 21

06 Bibliography 22
 INTRODUCTION

A Number Theoretic Function is a complex valued function defined for all positive
integers. In Number Theory, there exist many number theoretic functions. This
includes Divisor Function, Sigma Function, Euler’s-Phi Function and Mobius
Function. All these functions play a very important role in the field of Number
Theory.

In the first chapter we will discuss about Arithmetic Function. In the second chapter
we will introduce Euler’s-Phi Function and Mobius Function.

1
 PRELIMINARY

Let n be a fixed positive integer. Two integers a and b are said to be congruent
modulo n, symbolized by
a ≡ b (mod n)
if n divides the difference a − b; that is, provided that a − b = kn for some integer k.
Example:
To fix the idea, consider n = 7. It is routine to check that
3 ≡ 24 (mod 7) − 31 ≡ 11 (mod 7) − 15 ≡ −64 (mod 7)
Because 3 − 24 = (−3)7, −31 − 11 = (−6)7 and −15 − (−64) = 77. When
n does not divide (a − b), we say that a is incongruent to b modulo n, and in this case
we write
a ≢ b (mod n). For a simple example: 25 ≢ 12 (mod 7), because 7 fails to divide

25 − 12 = 13.
It is to be noted that any two integers are congruent modulo 1, whereas two integers
are congruent modulo 2 when they are both even or both odd. In as much as
congruence modulo 1 is not particularly interesting, the usual practice is to assume
that n > 1.
Remark:
Given an integer a, let q and r be its quotient and remainder upon division by n,
so that
a = qn + r 0≤r<n
Then, by definition of congruence, a ≡ r (mod n). Because there are n choices for
r , we see that every integer is congruent modulo n to exactly one of the values
0, 1, 2, . . . , n − 1; in particular, a ≡ 0 (mod n) if and only if n | a.

2
Fundamental Theorem of Arithmetic
is Every integer 𝑛 > 1 can be represented as Product of prime factor in only one way,
apart from the order of the factors.

Residue

If a is an integer and 𝑚 is a positive integer then the residue class of a modulo 𝑚 is

denoted by 𝑎ˆ and is given by

𝑎ˆ = {𝑥: 𝑥 ≡ 𝑎(𝑚𝑜𝑑𝑚)}
= {𝑥: 𝑥 = 𝑎 + 𝑚𝑘, 𝑘 = 0, ±1, ±2, ⋯ }

3
 CHAPTER 1
ARITHMETIC FUNCTION
An arithmetic Function is a function defined on the positive integers which take
values in the real or complex numbers. i.e., A function f: N→ C is called an
arithmetic function.
An arithmetic function is called multiplicative if f(mn) = f(m)f(n) for all coprime
natural numbers m and n.
Examples
a) Sum of divisors 𝜎(n)
b) Number of divisors 𝜏(n)
c) Euler’s function 𝜙(n)
d) Mobius function 𝜇(n)

Definition 1.1

Given a positive integer n, let τ (n) denote the number of positive divisors of n and

σ(n) denote the sum of positive divisors of n.

Example

Consider n = 12. Since 12 has the positive divisors 1, 2, 3, 4, 6, 12, we find that

τ (12) = 6 and σ(12) = 1 + 2 + 3 + 4 + 6 + 12 = 28

For the first few integers,

τ (1) = 1 τ (2) = 2 τ (3) = 2 τ (4) = 3 τ (5) = 2 τ (6) = 4, . . .

σ(1) = 1, σ(2) = 3, σ(3) = 4, σ(4) = 7 , σ(5) = 6, σ(6) = 12, . . .

It is not difficult to see that τ (n) = 2 if and only if n is a prime number; also,

σ(n) = n + 1 if and only if n is a prime.

4
Theorem 1.1
𝑘 𝑘
If n = 𝑝1 1 … … … … . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

(a) τ (n) = (k1+ 1)(k2 + 1) ・ ・ ・ (kr + 1), and

𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
(b) σ(n) = ………………….
𝑝1 −1 𝑝𝑟 −1

Proof

The positive divisors of n are precisely those integers

𝑎 𝑎 𝑎
d = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟

where 0 ≤ ai ≤ ki . There are k1 + 1 choices for the exponent a1; k2 + 1 choices for a2, .

. . ; and kr + 1 choices for ar . Hence, there are

(k1 + 1)(k2 + 1) · · · (kr + 1)

possible divisors of n.

To evaluate σ(n), consider the product

𝐾 𝐾
(1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Each positive divisor of n appears once and only once as a term in the expansion of

this product, so that

𝐾 𝐾
σ(n) = (1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Applying the formula for the sum of a finite geometric series to the ith factor on the

right-hand side, we get

𝑘 +1
𝐾 𝑝𝑖 𝑖 − 1
(1 + 𝑝𝑖 + 𝑃𝑖2 + ⋯ … … … 𝑃𝑖 𝑖 ) =
𝑝𝑖 − 1

5
It follows that
𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
σ(n) = 𝑝1 −1
………………….
𝑝𝑟 −1
.

Corresponding to the ∑ notation for sums, the notation for products may be

defined using ∏ , the Greek capital letter pi. The restriction delimiting the numbers

over which the product is to be made is usually put under the ∏

sign.

Examples

With this convention, the conclusion to Theorem 1.1 takes the compact form: if
𝑘 𝑘 𝑘
n = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

and

6
Theorem 1.2

The functions τ and σ are both multiplicative functions

Proof

Let m and n be relatively prime integers. Because the result is trivially true if

either m or n is equal to 1, we may assume that m > 1 and n > 1. If

are the prime factorizations of m and n . It follows that the prime factorization of

the product mn is given by

Applying to theorem 1.1, we obtain

In a similar fashion, theorem 1.1 gives

Thus, τ and σ are multiplicative functions.

Theorem 1.3

If f is a multiplicative function and F is defined by

then F is also multiplicative.

7
Proof

Let m and n be relatively prime positive integers. Then

because every divisor d of mn can be uniquely written as a product of a divisor d1

of m and a divisor d2 of n, where gcd(d1, d2) = 1. By the definition of a

multiplicative function,

f (d1d2) = f (d1) f (d2)

It follows that

It might be helpful to take time out and run through the proof of Theorem 1.3

in a concrete case. Letting m = 8 and n = 3, we have

= f (1) + f (2) + f (3) + f (4) + f (6) + f (8) + f (12) + f (24)

= f (1 · 1) + f (2 · 1) + f (1 · 3) + f (4 · 1) + f (2 · 3)+ f (8 · 1) + f (4 · 3) + f (8 · 3)

= f (1) f (1) + f (2) f (1) + f (1) f (3) + f (4) f (1) + f (2) f (3)+ f (8) f (1)

. + f (4)f(3)+ f (8) f (3)

8
= [ f (1) + f (2) + f (4) + f (8)][ f (1) + f (3)]

= F(8)F(3)

Theorem 1.3 provides a deceptively short way of drawing the conclusion that τ

and σ are multiplicative

The Mangoldt function 𝚲(𝒏)

Definition 1.2

For every integer 𝑛 ≥ 1 we define

log 𝑝 if 𝑛 = 𝑝𝑚 for some prime 𝑝 and some 𝑚 ≥ 1 ,

Λ(𝑛) = {
0 otherwise.

Here is a short table of values of Λ(𝑛) :

𝑛: 1 2 3 4 5 6 7 8 9 10
Λ(𝑛): 0 log 2 log 3 log 2 log 5 0 log 7 log 2 log 3 0

The proof of the next theorem shows how this function arises naturally from the

fundamental theorem of arithmetic.

Theorem 1.4

If 𝑛 ≥ 1 we have

log 𝑛 = ∑ Λ(𝑑 ) … … … … … … … … … … … … (1)

𝑑∣𝑛

Proof

The theorem is true if 𝑛 = 1 since both members are 0 . Therefore, assume that 𝑛 > 1

and write

9
 𝑟

𝑛 = ∏ 𝑝𝑘 𝑎𝑘
𝑘=1

Taking logarithms we have

𝑟

log 𝑛 = ∑ 𝑎𝑘 log 𝑝𝑘
𝑘=1

Now consider the sum on the right of (1). The only nonzero terms in the sum come

from those divisors 𝑑 of the form 𝑝𝑘 𝑚 for 𝑚 = 1,2, … , 𝑎𝑘 and 𝑘 = 1,2, … , 𝑟. Hence

𝑟 𝑎𝑘 𝑟 𝑎𝑘 𝑟

∑ Λ(𝑑) = ∑ ∑ Λ(𝑝𝑘𝑚 ) = ∑ ∑ log 𝑝𝑘 = ∑ 𝑎𝑘 log 𝑝𝑘 = log 𝑛

𝑑∣𝑛 𝑘=1 𝑚=1 𝑘=1 𝑚=1 𝑘=1

which proves (1).

10
 CHAPTER 2
EULER’S 𝝓 FUNCTION
Let n be positive integer. Let Un denote the set of all positive integers less than n and
coprime to it

For example,

U6 = {1,5}

U10 = {1,3,7,9}

U18 = {1,5,7,11,13,17}

Definition 2.1
Euler’s 𝜙 function is a function 𝜙: N→N such that for any n ∈ N, 𝜙 (n) is the number
of integers less than n and coprime to it

In other words

‘Euler’s 𝜙 function counts the number of elements in Un’

For example,

𝜙(1) = 1, 𝜙(2) = 1, 𝜙(3) = 2, 𝜙(4) = 2, 𝜙(5) = 4

𝜙(6) = 2 … .

Theorem 2.1
Let p be a prime. Then 𝜙 (p) = p-1

Proof:
By definition, any natural number strictly less than p is coprime to p, hence

𝜙 (p) = p-1

Theorem 2.2

If 𝑝 is a prime and 𝑘 > 0, then

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1)

11
Proof:

Consider the successive pk natural numbers not greater than pk arranged in the
following rectangular array of p columns and pk-1 rows

1 2 . . p

p+1 p+2 . . 2p

. . . . .

. . . . .

pk-p+1 pk-p+2 . . pk

among these numbers only the ones at the rightmost sides are not coprime to pk and
there are pk-1 members in that column. So

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1).

For example, 𝜙(8) = 23 − 22 = 4 which counts the number of elements in the set
U8 = {1,3,5,7}

By the fundamental theorem of arithmetic, we can write any natural number n as

𝑘 𝑘
n= 𝑝1 1 … … … . . 𝑝𝑟 𝑟

where 𝑃𝑖 ‘s are distinct prime and k 𝑖 ≥ 1 are integers. We already know how to find
𝑘 𝑘
𝜙(𝑝𝑖 𝑖 ) we would lie to see how 𝜙(𝑛) is related to 𝜙(𝑝𝑖 𝑖 ). This follows from a very
important property of Euler’s 𝜙 Function

Multiplicativity of Euler’s 𝝓 Function

Theorem 2.3

𝜙(mn) = 𝜙(m)𝜙(n) if m and n are coprime natural numbers.

Proof:

12
Consider the array of natural numbers not greater than mn arranged in m columns and
n rows in the following manner

1 2 ⋯ 𝑟 ⋯ 𝑚
𝑚+1 𝑚+2 𝑚+𝑟 2𝑚
2𝑚 + 1 2𝑚 + 2 2𝑚 + 𝑟 3𝑚
⋮ ⋮ ⋮ ⋮
(𝑛 − 1)𝑚 + 1 (𝑛 − 1)𝑚 + 2 (𝑛 − 1)𝑚 + 𝑟 𝑛𝑚

Clearly each row of the above array has m distinct residues modulo m. Each column
has n distinct residues modulo n: for 1 ≤ 𝑖, 𝑖 ≤ 𝑛 − 1

im +j ≡ im + j (mod n)

⇒ im ≡ im (mod n)

⇒ i ≡ i (mod n) (as gcd(m,n) = 1)

⇒i≡i

Each row has 𝜙(m) residues coprime to m, and each column has 𝜙(n) residues
coprime to n. Hence in total 𝜙(m)𝜙(n) elements in the above array which are
coprime to both m and n, it follows that

𝜙(mn) = 𝜙(m)𝜙(n)

Theorem 2.4
Let n be any natural numbers, then
1 1 1
𝜙(𝑛) = 𝑛 (1 − ) (1 − ) ⋯ (1 − )
𝑝1 𝑝2 𝑝𝑟

Proof:
By fundamental theorem of arithmetic, we can write
𝑘 𝑘 𝑘
𝑛 = 𝑃1 1 𝑃2 2 … … . … 𝑃𝑟 𝑟

Where 𝑝𝑖 are the distinct prime factor of n, and 𝑘𝑖 are the non negative integers. By
previous theorem and proposition,
𝑘 𝑘
𝜙(𝑛) = 𝜙(𝑝1 1 ) ⋅ … , 𝜙(𝑝𝑟 𝑟 )
𝑘 −1 𝑘
= 𝑃1 1 (𝑃1 − 1) ⋯ 𝑃𝑟 𝑟−1 (𝑃𝑟 − 1)

13
 𝑘 1 𝑘 1
= 𝑝1 1 (1 − ) ⋯ 𝑃𝑟 𝑟 (1 − )
𝑃1 𝑃𝑟

1 1
= 𝑛 (1 − ) ⋯ ⋅ (1 − )
𝑝1 𝑝𝑟

Theorem 2.5
For n > 2, 𝜙 (n) is an even integer.

Proof:

First, assume that n is a power of 2, let us say that n = 2k , with k ≥ 2. By

theorem 2.2,
1
𝜙(𝑛) = 𝜙(2𝑘 ) = 2𝑘 (1 − ) = 2𝑘−1
2
an even integer. If 𝑛 does not happen to be a power of 2, then it is divisible by an odd
prime 𝑝; we therefore may write 𝑛 as 𝑛 = 𝑝𝑘 𝑚, where 𝑘 ≥ 1 and gcd (𝑝𝑘 , 𝑚) = 1.
Exploiting the multiplicative nature of the phi-function, we obtain

𝜙(𝑛) = 𝜙(𝑝𝑘 )𝜙(𝑚) = 𝑝𝑘−1 (𝑝 − 1)𝜙(𝑚)

which again is even because 2 | p – 1.

Theorem 2.6
For each positive integer n,

𝑛 = ∑ 𝜙(𝑑)
𝑑∣𝑛

Proof:
Let us partition the set {1,2,…….,n} into mutually disjoint subsets Sd for each d/n,
where

𝑆𝑑 = {1 ≤ 𝑚 ≤ 𝑛 ∣ gcd (𝑚, 𝑛) = 𝑑}
𝑚 𝑛 𝑚 𝑛
= {1 ≤ ≤ ∣ gcd ( , ) = 1}
𝑑 𝑑 𝑑 𝑑

Then

14
 {1,2, … … . , n} = ∑ 𝑆𝑑
𝑑∣𝑛

𝑛
⇒ 𝑛 = ∑ 𝜙( )
𝑑
𝑑∣𝑛

= ∑ 𝜙(𝑑)
𝑑∣𝑛

As for each divisor of n, n/d is also a divisor of n

MOBIUS FUNCTION

Definition 2.2
The Mobius function 𝜇: 𝑁 ⟶ {0, ±1} is defined as

1 if 𝑛 = 1
𝜇(𝑛) = {0 if 𝑝2 ⁄𝑛 for some prime 𝑝
(−1)𝑟 if 𝑛 = 𝑝1 𝑝2 ⋯ 𝑝𝑟 , where 𝑝𝑖 are distinct primes

For example,

μ(1) = 1 μ(2) = −1 μ(3) = −1

μ(4) = 0 μ(5) = −1 μ(6) = 1

If 𝑝 is a prime number, it is clear that 𝜇(𝑝) = −1; in addition, 𝜇(𝑝𝑒 ) = 0 for 𝑒 ≥ 2.

Theorem 2.7
The Mobius function is a multiplicative function i.e.

μ(mn) = μ(m)μ(n), if m and n are relatively prime

Proof:
Let m and n be coprime integers, we can consider the following to cases
2
Case 1: let μ(mn) = 0 then there is a prime p such that 𝑝 ⁄𝑚𝑛. As m and n are
𝑝2⁄ 𝑝2⁄
coprime p cannot divide both m and n hence either 𝑚 or 𝑛 . Therefore either
μ(m) = 0 or μ(n) = 0 and we have μ(mn) = μ(m)μ(n)

Case 2: suppose that μ(mn) ≠ 0 then mn is square free, hence so are m and n. let
15
𝑚 = 𝑝1 … … … 𝑝𝑟 and 𝑛 = 𝑞1 … … … 𝑞𝑠 where 𝑝𝑖 𝑎𝑛𝑑 𝑞𝑗 are all distinct primes then
mn = 𝑝1 … … … 𝑝𝑟 𝑞1 … … … 𝑞𝑠 where all the primes occurring in the factorization of
mn are distinct. Hence

𝜇(𝑚𝑛) = (−1)𝑟+𝑠

= (−1)𝑟 (−1)𝑠

= μ(m)μ(n)

Theorem 2.8

1 if 𝑛 = 1
∑ 𝜇(𝑑) = {
0 if 𝑛 > 1
𝑑∣𝑛

Where d runs through all the positive divisors of n.

Proof:

𝐿𝑒𝑡 𝐹(𝑛) = ∑ 𝜇(𝑑)

𝑑∣𝑛

As μ is multiplicative, so is F(n) by the theorem (F be a multiplicative arithmetic

function 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ) then F is also a multiplicative arthmetic function)
𝑑∣𝑛

Clearly

𝐹 (1) = ∑ 𝜇 (𝑑 )
𝑑∣𝑛

= μ(1)

=1

For integers which are prime power, i.e. of the form pk for some k ≥ 1

𝐹 (𝑝2 ) = μ(1) + μ(p) + μ(p2 ) + ⋯ … … … . . +μ(p𝑘 )

= 1 + (-1) + 0……………+ 0

=0

Now consider any integer n, and consider its prime factorization. Then
𝑘 𝑘
𝑛 = 𝑝1 1 … … … … … . 𝑝𝑟 𝑟 , 𝑘𝑖 ≥ 1

16
 𝑘
⇒ 𝐹 (𝑛) = ∏𝐹(𝑝𝑖 𝑖 )

=0

Mobius inversion formula

The following theorem is known as Mobius inversion formula

Theorem 2.9
Let F and f be two function from the set N of natural number to the field complex
number C such that

𝐹(𝑛) = ∑ 𝑓(𝑑)
𝑑∣𝑛

Then we can express f(n) as

𝑛
𝑓 (𝑛 ) = ∑ 𝜇 (𝑑 )𝐹 ( )
𝑑
𝑑∣𝑛

𝑛
= ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Proof:
First observe that if d is divisor of n so is n/d. Hence both the summation in the last
line of the theorem are same. Now

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

The crucial step in the proof is to observe that the set of S of pairs of integers (c,d)
with d|n and c|n/d is the same as the set T of pairs (c,d) with c/n and d|n/c.

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

= ∑ ( ∑ 𝜇(𝑑)𝑓(𝑐))
𝑑∣𝑛 𝑐∣(𝑛/𝑑)

17
 = ∑ 𝑓(𝑐)𝜇(𝑑)
(𝑐,𝑑)∈𝑆

= ∑ 𝑓 (𝑐 ) 𝜇 (𝑑 )
(𝑐,𝑑)∈𝑇

= ∑ (𝑓(𝑐) ∑ 𝜇(𝑑))
𝑐∣𝑛 𝑑∣(𝑛/𝑐)

= F(n)
𝐴𝑠 ∑ 𝜇(𝑑 ) = 0 𝑢𝑛𝑙𝑒𝑠𝑠 𝑛⁄𝑐 = 1, which happens when c = n
𝑑∣𝑛

Let us demonstrate this with n = 15

15
∑ 𝜇 (𝑑 )𝐹 ( ) = 𝜇 (1)[ 𝑓 (1) + 𝑓 (3) + 𝑓 (5) + 𝑓 (15)] + 𝜇(3)[ 𝑓 (1) + 𝑓 (5)]
𝑑
𝑑∣15

+𝜇(5)[ 𝑓 (1) + 𝑓 (3)] + 𝜇(15)[ 𝑓 (1)]

= f (1)[μ(1) + μ(3) + μ(5) + μ(15)] + f (3)[μ(1) + μ(5)] + f (5)[μ(1) +

μ (5)] + f(15) μ(1)

= f(1).0 + f(3).0 + f(5).0 + f(15)

= f(15)

The above theorem leads to the following interesting identities

1. we know that for any positive integer n,

∑ 𝜙 (𝑑 ) = 𝑛
𝑑∣𝑛

Where 𝜙(𝑛) is Euler’s 𝜙 function. Hence

𝑛
𝜙 (𝑛 ) = ∑ 𝜇 ( ) 𝑑
𝑑
𝑑∣𝑛

For example,

𝜙(10) = μ(1)10 + μ(2)5 + μ(5)2 + μ(10)1

18
 = 10 - 5 – 2 + 1

=4

2. similarly

𝜎(𝑛) = ∑ 𝑑
𝑑∣𝑛
𝑛
𝑛 = ∑ 𝜇 ( ) 𝜎(𝑑)
𝑑
𝑑∣𝑛

For example,

With n = 10

μ(10). 1 + μ(2)(1 + 5) + μ(5)(1 + 3) + μ(1)(1 + 3 + 5 + 10)

= 1 – 1 – 5 – 1 – 3 + 1 + 3 +5 + 10

= 10

We have seen before that if multiplicative so is 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ). But we can now

𝑑∣𝑛

Prove that converse applying the Mobius inversion formula

Theorem 2.10
If F is a multiplicative function and

𝐹 (𝑛 ) = ∑ 𝑓 (𝑑 )
𝑑∣𝑛

then f is also multiplicative.

Proof:
By the Mobius inversion formula we know that
𝑛
𝑓 (𝑛) = ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Let m and n be relatively prime positive integers. We recall that any divisor

d of mn can be uniquely written as d = d1, d2, where d1 |m, d2 | n, and

𝑚 𝑛
gcd(d1, d2) = 1 = gcd( , ).
𝑑 1 𝑑2

19
Conversely if d1/m and d2/n then d1d2/mn thus,
𝑚𝑛
𝑓(𝑚𝑛) = ∑ 𝜇(𝑑)𝐹 ( )
𝑑
𝑑∣𝑚𝑛
𝑚𝑛
= ∑ 𝜇 (𝑑1 𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇 (𝑑1 )𝜇(𝑑2 )𝐹 ( )𝐹( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇(𝑑1 )𝐹 ( ) ∑ 𝜇 (𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 ∣𝑚 𝑑2 ∣𝑛
= 𝑓(𝑚)𝑓(𝑛)

In view of the above theorem we can say that as N(n) = n is a multiplicative function
so is 𝜙(𝑛) because

∑ 𝜙(𝑑 ) = 𝑛 = N(n)
𝑑∣𝑛

20
 CONCLUSION

The purpose of this project gives a simple account of Arithmetic function, Euler’s phi
function and Mobius Function. The study of these topics given excellent introduction
to the subject called ‘NUMBER THEORETIC FUNCTION’

Number Theoretic Function demands a high standard of rigor. Thus, our presentation
necessarily has its formal aspect with care taken to present clear and detailed
argument. An understanding of the statement of the theorem, number theory proof is
the important issue. In the first chapter we discuss about function τ and σ are both
multiplicative function. If f is a multiplicative function and F is defined by

𝐹(𝑛) = ∑𝑑∣𝑛 𝑓(𝑑 ), then F is also multiplicative. In the second chapter 2 we discuss
about that if p is prime the 𝜙(𝑝) = 𝑝 − 1, 𝜙(𝑚𝑛) = 𝜙(𝑚)𝜙(𝑛). The Mobius
function is multiplicative function if f is multiplicative function and 𝐹(𝑛)=∑𝑑∣𝑛 𝑓(𝑑 ),

then F is also multiplicative.

21
 BIBLIOGRAPHY
1 DAVID M. BURTON - ELEMENTRY NUMBER THEORY
2 TOM.M. APOSTOL - INTRODUCTION TO ANALYTIC NUMBER THEORY
3 https://en.m.wikipedia.org/wiki/Euler%27s_totient_function
4 https://en.m.wikipedia.org/wiki/Arithmetic_function

22
NUMBER THEORETIC FUNCTION
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
SHAMNA THASNI T
DB18CMSR25
Under the guidance of
Ms. Ajeena joseph

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Number Theoretic Function’ is a bona fide

project of SHAMNA THASNI T carried out the project work under my
supervision.

Mrs. Riya Baby Ms. Ajeena joseph

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I SHAMNA THASNI T hereby declare that the project ‘Number Theoretic

Function’ is an original record of studies and bona fide project carried out by me
during the period of 2018 – 2021 under the guidance of Ms. Ajeena joseph,
Department of Mathematics, Don Bosco Arts and Science College, Angadikkadavu
and has not submitted by me elsewhere for the award of my degree, diploma, title or
recognition, before.

Name
SHAMNA THASNI T

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the Almighty
who is always with me. Also, I must express my deepest gratitude to people along the
way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my project
supervisor Ms. Ajeena joseph, Department of Mathematics, Don Bosco Arts and
Science College, Angadikkadavu, for the invaluable guidance, loving encouragement
and meticulous care towards me throughout my career. I express my deep sense of
gratitude to all the faculty members of the Department of Mathematics, Don Bosco
Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and the
staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every one of
them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- ARITHMETIC FUNCTION 4-10

04 Chapter 2- EULER’S 𝜙 FUNCTION 11-20

05 Conclusion 21

06 Bibliography 22
 INTRODUCTION

A Number Theoretic Function is a complex valued function defined for all positive
integers. In Number Theory, there exist many number theoretic functions. This
includes Divisor Function, Sigma Function, Euler’s-Phi Function and Mobius
Function. All these functions play a very important role in the field of Number
Theory.

In the first chapter we will discuss about Arithmetic Function. In the second chapter
we will introduce Euler’s-Phi Function and Mobius Function.

1
 PRELIMINARY

Let n be a fixed positive integer. Two integers a and b are said to be congruent
modulo n, symbolized by
a ≡ b (mod n)
if n divides the difference a − b; that is, provided that a − b = kn for some integer k.
Example:
To fix the idea, consider n = 7. It is routine to check that
3 ≡ 24 (mod 7) − 31 ≡ 11 (mod 7) − 15 ≡ −64 (mod 7)
Because 3 − 24 = (−3)7, −31 − 11 = (−6)7 and −15 − (−64) = 77. When
n does not divide (a − b), we say that a is incongruent to b modulo n, and in this case
we write
a ≢ b (mod n). For a simple example: 25 ≢ 12 (mod 7), because 7 fails to divide

25 − 12 = 13.
It is to be noted that any two integers are congruent modulo 1, whereas two integers
are congruent modulo 2 when they are both even or both odd. In as much as
congruence modulo 1 is not particularly interesting, the usual practice is to assume
that n > 1.
Remark:
Given an integer a, let q and r be its quotient and remainder upon division by n,
so that
a = qn + r 0≤r<n
Then, by definition of congruence, a ≡ r (mod n). Because there are n choices for
r , we see that every integer is congruent modulo n to exactly one of the values
0, 1, 2, . . . , n − 1; in particular, a ≡ 0 (mod n) if and only if n | a.

2
Fundamental Theorem of Arithmetic
is Every integer 𝑛 > 1 can be represented as Product of prime factor in only one way,
apart from the order of the factors.

Residue

If a is an integer and 𝑚 is a positive integer then the residue class of a modulo 𝑚 is

denoted by 𝑎ˆ and is given by

𝑎ˆ = {𝑥: 𝑥 ≡ 𝑎(𝑚𝑜𝑑𝑚)}
= {𝑥: 𝑥 = 𝑎 + 𝑚𝑘, 𝑘 = 0, ±1, ±2, ⋯ }

3
 CHAPTER 1
ARITHMETIC FUNCTION
An arithmetic Function is a function defined on the positive integers which take
values in the real or complex numbers. i.e., A function f: N→ C is called an
arithmetic function.
An arithmetic function is called multiplicative if f(mn) = f(m)f(n) for all coprime
natural numbers m and n.
Examples
a) Sum of divisors 𝜎(n)
b) Number of divisors 𝜏(n)
c) Euler’s function 𝜙(n)
d) Mobius function 𝜇(n)

Definition 1.1

Given a positive integer n, let τ (n) denote the number of positive divisors of n and

σ(n) denote the sum of positive divisors of n.

Example

Consider n = 12. Since 12 has the positive divisors 1, 2, 3, 4, 6, 12, we find that

τ (12) = 6 and σ(12) = 1 + 2 + 3 + 4 + 6 + 12 = 28

For the first few integers,

τ (1) = 1 τ (2) = 2 τ (3) = 2 τ (4) = 3 τ (5) = 2 τ (6) = 4, . . .

σ(1) = 1, σ(2) = 3, σ(3) = 4, σ(4) = 7 , σ(5) = 6, σ(6) = 12, . . .

It is not difficult to see that τ (n) = 2 if and only if n is a prime number; also,

σ(n) = n + 1 if and only if n is a prime.

4
Theorem 1.1
𝑘 𝑘
If n = 𝑝1 1 … … … … . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

(a) τ (n) = (k1+ 1)(k2 + 1) ・ ・ ・ (kr + 1), and

𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
(b) σ(n) = ………………….
𝑝1 −1 𝑝𝑟 −1

Proof

The positive divisors of n are precisely those integers

𝑎 𝑎 𝑎
d = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟

where 0 ≤ ai ≤ ki . There are k1 + 1 choices for the exponent a1; k2 + 1 choices for a2, .

. . ; and kr + 1 choices for ar . Hence, there are

(k1 + 1)(k2 + 1) · · · (kr + 1)

possible divisors of n.

To evaluate σ(n), consider the product

𝐾 𝐾
(1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Each positive divisor of n appears once and only once as a term in the expansion of

this product, so that

𝐾 𝐾
σ(n) = (1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Applying the formula for the sum of a finite geometric series to the ith factor on the

right-hand side, we get

𝑘 +1
𝐾 𝑝𝑖 𝑖 − 1
(1 + 𝑝𝑖 + 𝑃𝑖2 + ⋯ … … … 𝑃𝑖 𝑖 ) =
𝑝𝑖 − 1

5
It follows that
𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
σ(n) = 𝑝1 −1
………………….
𝑝𝑟 −1
.

Corresponding to the ∑ notation for sums, the notation for products may be

defined using ∏ , the Greek capital letter pi. The restriction delimiting the numbers

over which the product is to be made is usually put under the ∏

sign.

Examples

With this convention, the conclusion to Theorem 1.1 takes the compact form: if
𝑘 𝑘 𝑘
n = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

and

6
Theorem 1.2

The functions τ and σ are both multiplicative functions

Proof

Let m and n be relatively prime integers. Because the result is trivially true if

either m or n is equal to 1, we may assume that m > 1 and n > 1. If

are the prime factorizations of m and n . It follows that the prime factorization of

the product mn is given by

Applying to theorem 1.1, we obtain

In a similar fashion, theorem 1.1 gives

Thus, τ and σ are multiplicative functions.

Theorem 1.3

If f is a multiplicative function and F is defined by

then F is also multiplicative.

7
Proof

Let m and n be relatively prime positive integers. Then

because every divisor d of mn can be uniquely written as a product of a divisor d1

of m and a divisor d2 of n, where gcd(d1, d2) = 1. By the definition of a

multiplicative function,

f (d1d2) = f (d1) f (d2)

It follows that

It might be helpful to take time out and run through the proof of Theorem 1.3

in a concrete case. Letting m = 8 and n = 3, we have

= f (1) + f (2) + f (3) + f (4) + f (6) + f (8) + f (12) + f (24)

= f (1 · 1) + f (2 · 1) + f (1 · 3) + f (4 · 1) + f (2 · 3)+ f (8 · 1) + f (4 · 3) + f (8 · 3)

= f (1) f (1) + f (2) f (1) + f (1) f (3) + f (4) f (1) + f (2) f (3)+ f (8) f (1)

. + f (4)f(3)+ f (8) f (3)

8
= [ f (1) + f (2) + f (4) + f (8)][ f (1) + f (3)]

= F(8)F(3)

Theorem 1.3 provides a deceptively short way of drawing the conclusion that τ

and σ are multiplicative

The Mangoldt function 𝚲(𝒏)

Definition 1.2

For every integer 𝑛 ≥ 1 we define

log 𝑝 if 𝑛 = 𝑝𝑚 for some prime 𝑝 and some 𝑚 ≥ 1 ,

Λ(𝑛) = {
0 otherwise.

Here is a short table of values of Λ(𝑛) :

𝑛: 1 2 3 4 5 6 7 8 9 10
Λ(𝑛): 0 log 2 log 3 log 2 log 5 0 log 7 log 2 log 3 0

The proof of the next theorem shows how this function arises naturally from the

fundamental theorem of arithmetic.

Theorem 1.4

If 𝑛 ≥ 1 we have

log 𝑛 = ∑ Λ(𝑑 ) … … … … … … … … … … … … (1)

𝑑∣𝑛

Proof

The theorem is true if 𝑛 = 1 since both members are 0 . Therefore, assume that 𝑛 > 1

and write

9
 𝑟

𝑛 = ∏ 𝑝𝑘 𝑎𝑘
𝑘=1

Taking logarithms we have

𝑟

log 𝑛 = ∑ 𝑎𝑘 log 𝑝𝑘
𝑘=1

Now consider the sum on the right of (1). The only nonzero terms in the sum come

from those divisors 𝑑 of the form 𝑝𝑘 𝑚 for 𝑚 = 1,2, … , 𝑎𝑘 and 𝑘 = 1,2, … , 𝑟. Hence

𝑟 𝑎𝑘 𝑟 𝑎𝑘 𝑟

∑ Λ(𝑑) = ∑ ∑ Λ(𝑝𝑘𝑚 ) = ∑ ∑ log 𝑝𝑘 = ∑ 𝑎𝑘 log 𝑝𝑘 = log 𝑛

𝑑∣𝑛 𝑘=1 𝑚=1 𝑘=1 𝑚=1 𝑘=1

which proves (1).

10
 CHAPTER 2
EULER’S 𝝓 FUNCTION
Let n be positive integer. Let Un denote the set of all positive integers less than n and
coprime to it

For example,

U6 = {1,5}

U10 = {1,3,7,9}

U18 = {1,5,7,11,13,17}

Definition 2.1
Euler’s 𝜙 function is a function 𝜙: N→N such that for any n ∈ N, 𝜙 (n) is the number
of integers less than n and coprime to it

In other words

‘Euler’s 𝜙 function counts the number of elements in Un’

For example,

𝜙(1) = 1, 𝜙(2) = 1, 𝜙(3) = 2, 𝜙(4) = 2, 𝜙(5) = 4

𝜙(6) = 2 … .

Theorem 2.1
Let p be a prime. Then 𝜙 (p) = p-1

Proof:
By definition, any natural number strictly less than p is coprime to p, hence

𝜙 (p) = p-1

Theorem 2.2

If 𝑝 is a prime and 𝑘 > 0, then

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1)

11
Proof:

Consider the successive pk natural numbers not greater than pk arranged in the
following rectangular array of p columns and pk-1 rows

1 2 . . p

p+1 p+2 . . 2p

. . . . .

. . . . .

pk-p+1 pk-p+2 . . pk

among these numbers only the ones at the rightmost sides are not coprime to pk and
there are pk-1 members in that column. So

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1).

For example, 𝜙(8) = 23 − 22 = 4 which counts the number of elements in the set
U8 = {1,3,5,7}

By the fundamental theorem of arithmetic, we can write any natural number n as

𝑘 𝑘
n= 𝑝1 1 … … … . . 𝑝𝑟 𝑟

where 𝑃𝑖 ‘s are distinct prime and k 𝑖 ≥ 1 are integers. We already know how to find
𝑘 𝑘
𝜙(𝑝𝑖 𝑖 ) we would lie to see how 𝜙(𝑛) is related to 𝜙(𝑝𝑖 𝑖 ). This follows from a very
important property of Euler’s 𝜙 Function

Multiplicativity of Euler’s 𝝓 Function

Theorem 2.3

𝜙(mn) = 𝜙(m)𝜙(n) if m and n are coprime natural numbers.

Proof:

12
Consider the array of natural numbers not greater than mn arranged in m columns and
n rows in the following manner

1 2 ⋯ 𝑟 ⋯ 𝑚
𝑚+1 𝑚+2 𝑚+𝑟 2𝑚
2𝑚 + 1 2𝑚 + 2 2𝑚 + 𝑟 3𝑚
⋮ ⋮ ⋮ ⋮
(𝑛 − 1)𝑚 + 1 (𝑛 − 1)𝑚 + 2 (𝑛 − 1)𝑚 + 𝑟 𝑛𝑚

Clearly each row of the above array has m distinct residues modulo m. Each column
has n distinct residues modulo n: for 1 ≤ 𝑖, 𝑖 ≤ 𝑛 − 1

im +j ≡ im + j (mod n)

⇒ im ≡ im (mod n)

⇒ i ≡ i (mod n) (as gcd(m,n) = 1)

⇒i≡i

Each row has 𝜙(m) residues coprime to m, and each column has 𝜙(n) residues
coprime to n. Hence in total 𝜙(m)𝜙(n) elements in the above array which are
coprime to both m and n, it follows that

𝜙(mn) = 𝜙(m)𝜙(n)

Theorem 2.4
Let n be any natural numbers, then
1 1 1
𝜙(𝑛) = 𝑛 (1 − ) (1 − ) ⋯ (1 − )
𝑝1 𝑝2 𝑝𝑟

Proof:
By fundamental theorem of arithmetic, we can write
𝑘 𝑘 𝑘
𝑛 = 𝑃1 1 𝑃2 2 … … . … 𝑃𝑟 𝑟

Where 𝑝𝑖 are the distinct prime factor of n, and 𝑘𝑖 are the non negative integers. By
previous theorem and proposition,
𝑘 𝑘
𝜙(𝑛) = 𝜙(𝑝1 1 ) ⋅ … , 𝜙(𝑝𝑟 𝑟 )
𝑘 −1 𝑘
= 𝑃1 1 (𝑃1 − 1) ⋯ 𝑃𝑟 𝑟−1 (𝑃𝑟 − 1)

13
 𝑘 1 𝑘 1
= 𝑝1 1 (1 − ) ⋯ 𝑃𝑟 𝑟 (1 − )
𝑃1 𝑃𝑟

1 1
= 𝑛 (1 − ) ⋯ ⋅ (1 − )
𝑝1 𝑝𝑟

Theorem 2.5
For n > 2, 𝜙 (n) is an even integer.

Proof:

First, assume that n is a power of 2, let us say that n = 2k , with k ≥ 2. By

theorem 2.2,
1
𝜙(𝑛) = 𝜙(2𝑘 ) = 2𝑘 (1 − ) = 2𝑘−1
2
an even integer. If 𝑛 does not happen to be a power of 2, then it is divisible by an odd
prime 𝑝; we therefore may write 𝑛 as 𝑛 = 𝑝𝑘 𝑚, where 𝑘 ≥ 1 and gcd (𝑝𝑘 , 𝑚) = 1.
Exploiting the multiplicative nature of the phi-function, we obtain

𝜙(𝑛) = 𝜙(𝑝𝑘 )𝜙(𝑚) = 𝑝𝑘−1 (𝑝 − 1)𝜙(𝑚)

which again is even because 2 | p – 1.

Theorem 2.6
For each positive integer n,

𝑛 = ∑ 𝜙(𝑑)
𝑑∣𝑛

Proof:
Let us partition the set {1,2,…….,n} into mutually disjoint subsets Sd for each d/n,
where

𝑆𝑑 = {1 ≤ 𝑚 ≤ 𝑛 ∣ gcd (𝑚, 𝑛) = 𝑑}
𝑚 𝑛 𝑚 𝑛
= {1 ≤ ≤ ∣ gcd ( , ) = 1}
𝑑 𝑑 𝑑 𝑑

Then

14
 {1,2, … … . , n} = ∑ 𝑆𝑑
𝑑∣𝑛

𝑛
⇒ 𝑛 = ∑ 𝜙( )
𝑑
𝑑∣𝑛

= ∑ 𝜙(𝑑)
𝑑∣𝑛

As for each divisor of n, n/d is also a divisor of n

MOBIUS FUNCTION

Definition 2.2
The Mobius function 𝜇: 𝑁 ⟶ {0, ±1} is defined as

1 if 𝑛 = 1
𝜇(𝑛) = {0 if 𝑝2 ⁄𝑛 for some prime 𝑝
(−1)𝑟 if 𝑛 = 𝑝1 𝑝2 ⋯ 𝑝𝑟 , where 𝑝𝑖 are distinct primes

For example,

μ(1) = 1 μ(2) = −1 μ(3) = −1

μ(4) = 0 μ(5) = −1 μ(6) = 1

If 𝑝 is a prime number, it is clear that 𝜇(𝑝) = −1; in addition, 𝜇(𝑝𝑒 ) = 0 for 𝑒 ≥ 2.

Theorem 2.7
The Mobius function is a multiplicative function i.e.

μ(mn) = μ(m)μ(n), if m and n are relatively prime

Proof:
Let m and n be coprime integers, we can consider the following to cases
2
Case 1: let μ(mn) = 0 then there is a prime p such that 𝑝 ⁄𝑚𝑛. As m and n are
𝑝2⁄ 𝑝2⁄
coprime p cannot divide both m and n hence either 𝑚 or 𝑛 . Therefore either
μ(m) = 0 or μ(n) = 0 and we have μ(mn) = μ(m)μ(n)

Case 2: suppose that μ(mn) ≠ 0 then mn is square free, hence so are m and n. let
15
𝑚 = 𝑝1 … … … 𝑝𝑟 and 𝑛 = 𝑞1 … … … 𝑞𝑠 where 𝑝𝑖 𝑎𝑛𝑑 𝑞𝑗 are all distinct primes then
mn = 𝑝1 … … … 𝑝𝑟 𝑞1 … … … 𝑞𝑠 where all the primes occurring in the factorization of
mn are distinct. Hence

𝜇(𝑚𝑛) = (−1)𝑟+𝑠

= (−1)𝑟 (−1)𝑠

= μ(m)μ(n)

Theorem 2.8

1 if 𝑛 = 1
∑ 𝜇(𝑑) = {
0 if 𝑛 > 1
𝑑∣𝑛

Where d runs through all the positive divisors of n.

Proof:

𝐿𝑒𝑡 𝐹(𝑛) = ∑ 𝜇(𝑑)

𝑑∣𝑛

As μ is multiplicative, so is F(n) by the theorem (F be a multiplicative arithmetic

function 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ) then F is also a multiplicative arthmetic function)
𝑑∣𝑛

Clearly

𝐹 (1) = ∑ 𝜇 (𝑑 )
𝑑∣𝑛

= μ(1)

=1

For integers which are prime power, i.e. of the form pk for some k ≥ 1

𝐹 (𝑝2 ) = μ(1) + μ(p) + μ(p2 ) + ⋯ … … … . . +μ(p𝑘 )

= 1 + (-1) + 0……………+ 0

=0

Now consider any integer n, and consider its prime factorization. Then
𝑘 𝑘
𝑛 = 𝑝1 1 … … … … … . 𝑝𝑟 𝑟 , 𝑘𝑖 ≥ 1

16
 𝑘
⇒ 𝐹 (𝑛) = ∏𝐹(𝑝𝑖 𝑖 )

=0

Mobius inversion formula

The following theorem is known as Mobius inversion formula

Theorem 2.9
Let F and f be two function from the set N of natural number to the field complex
number C such that

𝐹(𝑛) = ∑ 𝑓(𝑑)
𝑑∣𝑛

Then we can express f(n) as

𝑛
𝑓 (𝑛 ) = ∑ 𝜇 (𝑑 )𝐹 ( )
𝑑
𝑑∣𝑛

𝑛
= ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Proof:
First observe that if d is divisor of n so is n/d. Hence both the summation in the last
line of the theorem are same. Now

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

The crucial step in the proof is to observe that the set of S of pairs of integers (c,d)
with d|n and c|n/d is the same as the set T of pairs (c,d) with c/n and d|n/c.

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

= ∑ ( ∑ 𝜇(𝑑)𝑓(𝑐))
𝑑∣𝑛 𝑐∣(𝑛/𝑑)

17
 = ∑ 𝑓(𝑐)𝜇(𝑑)
(𝑐,𝑑)∈𝑆

= ∑ 𝑓 (𝑐 ) 𝜇 (𝑑 )
(𝑐,𝑑)∈𝑇

= ∑ (𝑓(𝑐) ∑ 𝜇(𝑑))
𝑐∣𝑛 𝑑∣(𝑛/𝑐)

= F(n)
𝐴𝑠 ∑ 𝜇(𝑑 ) = 0 𝑢𝑛𝑙𝑒𝑠𝑠 𝑛⁄𝑐 = 1, which happens when c = n
𝑑∣𝑛

Let us demonstrate this with n = 15

15
∑ 𝜇 (𝑑 )𝐹 ( ) = 𝜇 (1)[ 𝑓 (1) + 𝑓 (3) + 𝑓 (5) + 𝑓 (15)] + 𝜇(3)[ 𝑓 (1) + 𝑓 (5)]
𝑑
𝑑∣15

+𝜇(5)[ 𝑓 (1) + 𝑓 (3)] + 𝜇(15)[ 𝑓 (1)]

= f (1)[μ(1) + μ(3) + μ(5) + μ(15)] + f (3)[μ(1) + μ(5)] + f (5)[μ(1) +

μ (5)] + f(15) μ(1)

= f(1).0 + f(3).0 + f(5).0 + f(15)

= f(15)

The above theorem leads to the following interesting identities

1. we know that for any positive integer n,

∑ 𝜙 (𝑑 ) = 𝑛
𝑑∣𝑛

Where 𝜙(𝑛) is Euler’s 𝜙 function. Hence

𝑛
𝜙 (𝑛 ) = ∑ 𝜇 ( ) 𝑑
𝑑
𝑑∣𝑛

For example,

𝜙 (10) = μ(1)10 + μ(2)5 + μ(5)2 + μ(10)1

18
 = 10 - 5 – 2 + 1

=4

2. similarly

𝜎(𝑛) = ∑ 𝑑
𝑑∣𝑛
𝑛
𝑛 = ∑ 𝜇 ( ) 𝜎(𝑑)
𝑑
𝑑∣𝑛

For example,

With n = 10

μ(10). 1 + μ(2)(1 + 5) + μ(5)(1 + 3) + μ(1)(1 + 3 + 5 + 10)

= 1 – 1 – 5 – 1 – 3 + 1 + 3 +5 + 10

= 10

We have seen before that if multiplicative so is 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ). But we can now

𝑑∣𝑛

Prove that converse applying the Mobius inversion formula

Theorem 2.10
If F is a multiplicative function and

𝐹 (𝑛 ) = ∑ 𝑓 (𝑑 )
𝑑∣𝑛

then f is also multiplicative.

Proof:
By the Mobius inversion formula we know that
𝑛
𝑓 (𝑛) = ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Let m and n be relatively prime positive integers. We recall that any divisor

d of mn can be uniquely written as d = d1, d2, where d1 |m, d2 | n, and

𝑚 𝑛
gcd(d1, d2) = 1 = gcd( , ).
𝑑1 𝑑2

19
Conversely if d1/m and d2/n then d1d2/mn thus,
𝑚𝑛
𝑓(𝑚𝑛) = ∑ 𝜇(𝑑)𝐹 ( )
𝑑
𝑑∣𝑚𝑛
𝑚𝑛
= ∑ 𝜇 (𝑑1 𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇 (𝑑1 )𝜇(𝑑2 )𝐹 ( )𝐹( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇(𝑑1 )𝐹 ( ) ∑ 𝜇 (𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 ∣𝑚 𝑑2 ∣𝑛
= 𝑓(𝑚)𝑓(𝑛)

In view of the above theorem we can say that as N(n) = n is a multiplicative function
so is 𝜙(𝑛) because

∑ 𝜙(𝑑 ) = 𝑛 = N(n)
𝑑∣𝑛

20
 CONCLUSION

The purpose of this project gives a simple account of Arithmetic function, Euler’s phi
function and Mobius Function. The study of these topics given excellent introduction
to the subject called ‘NUMBER THEORETIC FUNCTION’

Number Theoretic Function demands a high standard of rigor. Thus, our presentation
necessarily has its formal aspect with care taken to present clear and detailed
argument. An understanding of the statement of the theorem, number theory proof is
the important issue. In the first chapter we discuss about function τ and σ are both
multiplicative function. If f is a multiplicative function and F is defined by

𝐹(𝑛) = ∑𝑑∣𝑛 𝑓(𝑑 ), then F is also multiplicative. In the second chapter 2 we discuss
about that if p is prime the 𝜙(𝑝) = 𝑝 − 1, 𝜙(𝑚𝑛) = 𝜙(𝑚)𝜙(𝑛). The Mobius
function is multiplicative function if f is multiplicative function and 𝐹(𝑛)=∑𝑑∣𝑛 𝑓(𝑑 ),

then F is also multiplicative.

21
 BIBLIOGRAPHY
1 DAVID M. BURTON - ELEMENTRY NUMBER THEORY
2 TOM.M. APOSTOL - INTRODUCTION TO ANALYTIC NUMBER THEORY
3 https://en.m.wikipedia.org/wiki/Euler%27s_totient_function
4 https://en.m.wikipedia.org/wiki/Arithmetic_function

22
DON BOSCO ARTS & SCIENCE COLLEGE
ANGADIKADAVU

DEPARTMENT OF MATHEMATICS
2018-2021

Project Report on

INNER PRODUCT SPACES

 DEPARTMENT OF MATHEMATICS

DON BOSCO ARTS & SCIENCE COLLEGE

ANGADIKADAVU

MARCH 2021

Project Report on

INNER PRODUCT SPACES

Dissertation submitted in the partial fulfilment

of the requirement for the award of

Bachelor of Science in Mathematics of

Kannur University

Name : SHARON JOSEPH

Roll No. : DB18CMSR29

Examiners: 1.

2.
 𝐊𝐀𝐍𝐍𝐔𝐑 𝐔𝐍𝐈𝐕𝐄𝐑𝐒𝐈𝐓𝐘

𝐁𝐎𝐍𝐀𝐅𝐈𝐃𝐄 𝐂𝐄𝐑𝐓𝐈𝐅𝐈𝐂𝐀𝐓𝐄

Certified that this project report on " INNER PRODUCT SPACES” is the bonafide
work of SHARON JOSEPH who carried out the project work under my supervision.

Mrs. Riya Baby Mr. Anil M V

Head of Department Supervisor
 𝐃𝐄𝐂𝐋𝐀𝐑𝐀𝐓𝐈𝐎𝐍

I, SHARON JOSEPH hereby declare that the project work entitled ‘INNER
PRODUCT SPACES′ has been prepared by me and submitted to Kannur University in
partial fulfilment of requirement for the award of Bachelor of Science is a record of
original work done by me under the supervision of Mr. ANIL M V, Assistant Professor,
Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I, also declare that this Project work has been submitted by me fully or partially for
the award of any Degree, Diploma, Title or recognition before any authority.

Place ∶ Angadikadavu
SHARON JOSEPH
Date ∶ DB18CMSR29
 𝐀𝐂𝐊𝐍𝐎𝐖𝐋𝐄𝐃𝐆𝐄𝐌𝐄𝐍𝐓

Introduction is the proper place to begin. But first I bow my head before the
Almighty who is always with me. Also, I must express mydeepest gratitude to
people along the way.
No words can adequately express the sense of gratitude, still I try to express my
heartfelt thanks through words. The outset, I am deeply indebted to my project
supervisor Mr. ANIL M.V, Assistant Professor, Department of Mathematics, Don
Bosco Arts and ScienceCollege, Angadikadavu, for the invaluable guidance, loving
encouragement and meticulous care towards me throughout my career.I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikadavu.
I could not name many who sincerely supported and helped for the successful
completion of this Project. It is my pleasure and duty to thank each and every one
of them who walked with me.

SHARON JOSEPH
 Contents

1. INTRODUCTION 1

2. PRELIMINARIES 2

3. CHAPTER 1 : INNER PRODUCT SPACES 6

4. CHAPTER 2 : ORTHOGONAL SETS 9

5. BIBLIOGRAPHY 16
INTODUCTION

In linear algebra, an inner product space is a vector space with an additional structure

called an inner product. This additional structure associates

each pair of vectors in the space with a scalar quantity known as the inner product of the

vectors. Inner products allow the rigorous introduction of intuitive geometrical notions

such as the length of a vector or the angle between two vectors. They also provide the

means of defining orthogonality between vectors (zero inner product). Inner product

spaces generalize Euclidean spaces (in which the inner product is the dot product, also

known as the scalar product) to vector spaces of any (possibly infinite) dimension and are

studied in functional analysis. The first usage of the concept of a vector space with an

inner product is due to Peano, in 1898.

An inner product naturally induces an associated norm, thus an inner product space is also

a normed vector space. A complete space with an inner product is called a Hilbert space.

An (incomplete) space with an inner product is called a pre-Hilbert space.

1
 PRELIMINARIES

LINEAR SPACES

Definition 1: A linear (vector) space X over a field F is a set of elements

together with a function, called addition, from X ×X into X and a function

called scalar multiplication, from F × X into X which satisfy the following

conditions for all x, y, z ∈ X and α, β ∈ F;

i. (x + y) + z = x + (y + z)

ii. x +y = y +x

iii. There is an element 0 in X such that x + 0 = x for all x ∈ X.

iv. For each x ∈ X there is an element −x ∈ X such that x + (−x) = 0.

v. (x + y) = αx + αy

vi. (α + β)x = αx + βx

vii. α(βx) = (αβ)x

viii. 1 · x = x.

Properties i to iv imply that X is an abelian group under addition and v to vi

relate the operation of scalar multiplication to addition X and to addition and

multiplication in F.

Examples:

(a) Vn(R). The vectors are n-tuples of real numbers and the scalars are real

2
 numbers with addition and scalar multiplication defined by

(1 ,···,  n ) + ( 1 ,···,  n ) = (1 + 1 ,···,  n +  n ) (1)

 (1 ,···,  n ) = ( 1 ,···,  n ) (2)

Vn(R) is a linear space over R. Similarly, the set of all n-tuples of complex

numbers with the above definition of addition and multiplication is a linear

space over C and is denoted as Vn(C).

(b) The set of all functions from a nonempty set X into a field F with addition and

scalar multiplication defined by

[f + g](t) = f (t) + g(t) and [αf ](t) = αf (t); f, g ∈ X, t ∈ T (3)

is a linear space.

Let T = N the set of all positive integers and X is the set of all sequences of

elements F with addition and scalar multiplication defined by

( n +  n ) = ( n +  n ) (4)

 ( n ) = (  n ) (5)

denoted as V∞(F), form a linear space.

3
METRIC SPACES

Remember the distance function in the Euclidean space Rn.

Let x, y, z ∈ Rn, then

(1) |x − y| ≥ 0; |x − y| = 0 if and only if x = y ;

(2) |x − y| = |y − x|;

(3) |x − y| ≤ |x − z| + z − y|.

Definition 2: A metric or distance function on a set X is a real valued function

d defined on X × X which has the following properties: for all x, y, z ∈ X.

(1) d(x, y) ≥ 0; d(x, y) = 0 if and only if x = y;

(2) d(x, y) = d(y, x);

(3) d(x, y) ≤ d(x, z) + d(z, y)

A metric space (X, d) is a nonempty set X and a metric d defined on X.

Examples: In addition to the Euclidean spaces let us have the following examples.

Here all functions are assumed to be continuous. Let Lp denotes a set of complex

p
valued functions in Rn such that f is integrable. Let us recall some results

concerning such functions.

Höder’s Inequality: If p > 1, 1/q = 1 − 1/p

 | fg |  [ | f | ] [  | g |q ]1/ q .
p 1/ p

Minkowski’s Inequality: If p ≥ 1,

[ | f + g | p ]1/ p  [ | f | p ]1/ p + [  | g | p ]1/p

4
If xk and yk for k = 1, … , m are complex numbers, let f (t) = |xk| and g(t) =

|yk| for t ∈ [k, k + 1] and f (t) = 0 = g(t) for t∈ [1, m + 1]. Then we obtain the

summation form of the above inequalities from the integral form

Hölder’s Inequality

1/ p 1/ q
m m p
 m q

 xk yk   xk    yk 
k =1  k =1   k =1 

Minkowski’s Inequality:

p 1/ p p 1/ p p 1/p
m  m  m 
  xk + yk     xk  +   yk 
 k =1   k =1   k =1 

NORMED LINEAR SPACES

Definition 3. A norm on X is a real valued function, whose value at x is denoted

by ||x||, satisfying the following conditions for all x, y ∈ X and α ∈ F;

(1) ||x|| > 0 if x ≠ 0

(2) ||αx|| = |α|||x||

(3) ||x + y|| ≤ ||x|| + ||y||.

A linear space X with a norm defined on it is called a normed linear space.

Example: l p space. On the linear space Vn(F), define

n
x = [ |  i | p ]1/ p
k =1

where p ≥ 1 is any real number and x = (1 ,···,  n ) . This defines a norm (called p-

p
norm) on Vn(F). This space is called l space .
5
 CHAPTER 1

INNER PRODUCT SPACES

INNER PRODUCTS

Let 𝐹 be the field of real numbers or the field of complex numbers, and V a vector space over

F an inner product on V is a function which assigns to each ordered’ pair of vectors 𝛼, 𝛽 in V

a scalar (𝛼|𝛽) in 𝐹 in such a way that for all 𝛼, 𝛽, γ in V and all scalars c.

(a) (𝛼 + 𝛽|𝛾) = (𝛼|𝛾) + (𝛽|𝛾) ;

(b) (c𝛼|𝛽) = 𝑐(𝛼|𝛽) ;

̅̅̅̅̅ ), the bar denoting complex conjugation

(c) (𝛽|𝛼) = (𝛼|𝛽

(d) (𝛼|𝛼) > 0 if 𝛼 ≠ 0

It should be observed that conditions (a), (b) and (c) implies that

(𝑒) = (𝛼 ∣ 𝑐𝛽 + 𝛾) = (𝑐̅(𝛼|𝛽) + (𝛼|𝛾)

One other point should be made. When 𝐹 is the field 𝑅 of real nunbers. The complex conjugates

appearing in (c) and (e) are superflom. However, in the complex case they are necessary for

the consistency of the conditions. Without these complex conjugates we would have the

contradiction

(𝛼|𝛼) > 0 and (𝑖𝛼 ∣ 𝑖𝛼) = −1(𝛼|𝛼)

Example 1:

On F 𝑛 there is an inner product which we call the standard inner product. It is defined on 𝛼 =

(𝑥1 , ⋯ 𝑥𝑛 ) and 𝛽 = (𝑦1 , … , 𝑦𝑛 ), by

6
 (𝛼|𝛽) = ∑ 𝑥𝑖 𝑦̅𝑖
𝑖

When F is R this may be also written as

(𝛼|𝛽) = ∑ 𝑥𝑖 𝑦𝑖
𝑖

In the real case, the standard inner product is often called the dot or scalar product and denoted

by 𝛼 ⋅ 𝛽.

INNER PRODUCTS SPACES

An inner product space is a real or complex vector space together with a specified inner product

on that space.

• A finite-dimensional real inner product space is often called a Euclidean spare. A

complex inner product spare often referred to as a unitary spare.

• Every inner product space is a normed linear space and every normed space is a metric

space. Hence , every inner product space is a metric space.

Theorem

If V is an inner product space, then for any vector’s 𝛼, 𝛽 in 𝑉 and any scalar c

(1) ||𝑐𝛼|| = |𝑐|||𝛼|| ;

(ii) ||𝛼|| > 0 for 𝛼 ≠ 0

(iii) |(𝛼 ∣ 𝛽)| ⩽ ||𝛼|| ||𝛽||

(iv) ∥ 𝛼 + 𝛽|| ⩽∥ 𝛼 ∥ +∥ 𝛽||

Proof:

Statements (i) and (ii) follow almost immediately form the various definitions

involved. The inequality in (iii) is clearly valid when 𝛼 = 0. if 𝛼 ≠ 0, put

(𝛽|𝛼)
𝛾=𝛽− 𝛼
∥ 𝛼 ∥2

7
 Then, (𝛾 ∣ 𝛼) = 0 and

(𝛽|𝛼) (𝛽|𝛼)
0 ⩽∥ 𝛾 ∥2 = (𝛽 − 2
𝛼 ⁄𝛽 − 𝛼)
∥𝛼∥ ∥ 𝛼 ∥2

(𝛽|𝛼)(𝛼|𝛽)
= (𝛽|𝛽) −
∥ 𝛼 ∥2

|(𝛼|𝛽)|2
=∥ 𝛽 ∥2 −
∥ 𝛼 ∥2

Hence,

|(𝛼 ∣ 𝛽)|2 ⩽∥ 𝛼 ∥2 ∥ 𝛽 ∥2

Now using (c) we find that

∥ 𝛼 + 𝛽 ∥2 = ∥ 𝛼 ∥2 + (𝛼 ∣ 𝛽) + (𝛽 ∣ 𝛼)+∥ 𝛽 ∥2

=∥ 𝛼 ∥2 + 2Re (𝛼 ∣ 𝛽)+∥ 𝛽 ∥2
⩽∥ 𝛼 ∥2 + 2 ∥ 𝛼 ∥∥ 𝛽 ∥ +∥ 𝛽 ∥2
= (∥ 𝛼 ∥ +∥ 𝛽 ∥)2

Thus,

∥ 𝛼 + 𝛽 ∥ ⩽ ∥ 𝛼 ∥ +∥ 𝛽 ∥

the inequality (iii) is called the Cauchy -Schwarz inequality. It has a wide variety of application

the proof shows that if 𝛼is non-zero then

(( 𝛼 ∣ 𝛽 )) <∥ 𝛼 ∥∥ 𝛽 ∥, unless

(𝛽|𝛼)
𝛽= 𝛼
∥ 𝛼 ∥2

Then equality occurs in (iii) if and only if 𝛼 and 𝛽 are linearly independent.

8
 CHAPTER 2

ORTHOGONAL SETS

Definition

Let α and β be the vectors in an inner product space V. Then α is orthogonal to β if

(α | β) = 0. We simply say that and are orthogonal.

Definition

If S is a set of vectors in V, S is called an orthogonal set provided all set pairs of

distinct vectors in S are orthogonal.

Definition

An orthogonal set is an orthogonal set S with the additional property that ∥ 𝛼 ∥= 1 for

every 𝛼 in S.

• The zero vectors are orthogonal to every vector in V and is the only vector with this

property.

• It is an appropriate to think of an orthonormal set as a set of mutually perpendicular

vectors each having length l.

Example: the vector (x , y) is 𝑅 2 is orthogonal to (−y , x) with respect to the standard inner

product, for,

((x , y)|(−y , x)) = −xy + yx = 0

• The standard basis of either 𝑅 𝑛 or 𝐶 𝑛 is an orthonormal set with respect to the standard

inner product.

9
Theorem : An orthogonal set of nonzero vectors is linearly independent.

Proof:

Let S be a finite or infinite orthogonal set of nonzero vectors in a given inner product space

suppose 𝛼1, 𝛼2 , … 𝛼𝑛 are distinct vectors in S and that β=𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛

Then (β|𝛼𝑘 )=( 𝑐1 𝛼1+ + ⋯ 𝑐𝑛 𝛼𝑛 |𝛼𝑘 )

= 𝑐1 (𝛼1 |𝛼𝑘 ) + 𝑐2 ( 𝛼2 |𝛼𝑘 )+…+𝑐𝑛 (𝛼𝑛 |𝛼𝑘 )

= 𝑐𝑘 (𝛼𝑛 |𝛼𝑘 ) , since (𝛼𝑖 |𝛼𝑗 ) = 0,if i ≠ j and (𝛼𝑖 |𝛼𝑗 ) = 1,if i=j

Hence, 𝑐𝑘 = (β |𝛼𝑘 ) /(𝛼𝑘 , 𝛼𝑘 ) )

𝑐𝑘 =(β |𝛼𝑘 )/||𝛼𝑘 ||2,1≤ k ≤ m

Thus, when β=0 each 𝑐𝑘 =0; so S is a linearly independent set.

Corollary:

If {𝛼1, 𝛼2 , … 𝛼𝑚 } is an orthogonal set of nonzero vectors in a finite dimensional inner product

space V, then m ≤ dimV.

That is number of mutually orthogonal vectors in V cannot exceed the dimensional V.

Corollary:

If a vector β is linear combination of an orthogonal of nonzero vectors 𝛼1, 𝛼2 , … 𝛼𝑛 , then β is

the particular linear combination

(𝛽 |𝛼𝑘 )
β =∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘
𝑘 ||

Proof :

Since β is the linear combination of an orthogonal sequence of nonzero vectors

𝛼1, 𝛼2 , … 𝛼𝑛 , we can write β =𝑐1 𝛼1 + ⋯ 𝑐𝑛 𝛼𝑛 .

(𝛽|𝛼𝑘 )
Where 𝑐𝑘 = , 1 ≤ k ≤ m (ref. by previous theorem)
||𝛼𝑘 ||2

( 𝛽 | 𝛼𝑘 )
Hence, β= ∑𝑚
𝑘=1 ||𝛼𝑘 ||2

10
Theorem (Gram Schmidt Orthogonalization Process)

Let V be an inner product space and {𝛽1, … , 𝛽𝑛 } be any linearly independent vectors in V. Then

one may construct orthogonal vectors {𝛼1, 𝛼2 , … 𝛼𝑛 } in V, such that for each k = 1, 2, …n, the

set {𝛼1, 𝛼2 , … 𝛼𝑘 }is an orthogonal basis for the subspace of V spanned by 𝛽1, … , 𝛽𝑛 .

Proof :

The vectors are obtained by means of a construction known as the Gram Schmidt

orthogonalization process.

First let 𝛼1 =𝛽1 The other vectors are then given inductively as follows:

Suppose 𝛼1, 𝛼2 , … 𝛼𝑚 (1 ≤ m ≤ n) have been chosen so that for every k

{𝛼1, 𝛼2 , … 𝛼𝑘 } (1≤k≤m)

is an orthogonal basis for the space of v that is spanned by 𝛽1, … , 𝛽𝑛

To construct the next vector 𝛼𝑚+1, let

( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1, = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
Then  m +1  0 . For otherwise, 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘 =0, implies,
||𝛼𝑘 ||2

( 𝛽𝑚+1 |𝛼𝑘 )
𝛽𝑚+1= 𝛽𝑚+1 -∑𝑚
𝑘=1 𝛼𝑘 ,implies,  m+1 is a linear combination of 𝛼1, 𝛼2 , … 𝛼𝑚 and
||𝛼𝑘 ||2

hence a linear combination of 1 ,  2 ,...,  m , a contradiction.

Furthermore, if 1≤j≤m, then,

(𝛽𝑚+1 |𝛼𝑘 )
(𝛼𝑚+1 | 𝛼𝑗 ) = ( 𝛽𝑚+1 | 𝛼𝑗 ) -∑𝑚
𝑘=1 (𝛼𝑘 |𝛼𝑗 )
||𝛼𝑘 ||2

= ( 𝛽𝑚+1 | 𝛼𝑚 ) – ( 𝛽𝑚+1 | 𝛼𝑗 ) , using the orthonormality of {𝛼1, 𝛼2 , … 𝛼𝑚 }.

Therefore {𝛼1, 𝛼2 , …,𝛼𝑚+1 } is an orthogonal set consisting of m+1 nonzero vectors in the

subspace spanned by 𝛽1, … , 𝛽𝑚+1 . Hence by an earlier Theorem , it is a basis for this subspace

.Thus the vectors , 𝛼1, 𝛼2 , … 𝛼𝑛 may be constructed using the formula

11
 ( 𝛽𝑚+1 |𝛼𝑘 )
𝛼𝑚+1 = 𝛽𝑚+1-∑𝑚
𝑘=1 𝛼𝑘
||𝛼𝑘 ||2

In particular, when n=3 ,we have

𝛼1 =𝛽1

(𝛼2 |𝛽2 )
𝛼2 =𝛽2- α1
||𝛼𝑘 ||2

(𝛽3 |𝛼1 ) (𝛼2 |𝛽3 )

𝛼3 =𝛽3 - α1 - 𝛼
||𝛼1 ||2 ||𝛼𝑘 ||2 2

Corollary :

Every finite dimensional inner product space has an orthonormal basis.

Proof :

Let V be a finite dimensional inner product space and { 𝛽1, … , 𝛽𝑛 } a basis for V. Apply the

gram Schmidt orthogonalization process to construct an orthogonal basis , simply replace each
𝛼𝑘
vector 𝛼𝑛 by .
||𝛼𝑘 ||

Gram-Schmidt process can be used to test for linear dependence . For suppose 𝛽1, … , 𝛽𝑛 are

linearly independent vectors in an inner product space; to exclude a trivial case , assume that

β≠0. Let m be largest integers for which 𝛽1, … , 𝛽𝑚 are independent. Then 1≤m˂n.

Let 𝛼1 , 𝛼2 , … 𝛼𝑚 be the vectors obtained by applying the orthogonalization process to

(𝛽𝑚+1 |𝛼𝑘 )
𝛽1, … , 𝛽𝑚 . Then the vector 𝛼𝑚+1 given by 𝛼𝑚+1 = 𝛽𝑚+1–∑𝑚
𝑘=1 𝛼𝑘 is necessarily 0.
||𝛼𝑘 ||2

For 𝛼𝑚+1 is in the subspace spanned by 𝛼1 , 𝛼2 , … 𝛼𝑚 and orthogonal to each of the vectors ,

(β|𝛼𝑘 )
hence it is 0 as β=∑𝑚
𝑘=1 ||𝛼 2
𝛼𝑘 . Conversely, if 𝛼1 , 𝛼2 , … 𝛼𝑚 are different from 0 and 𝛼𝑚+1 =0,
𝑘 ||

then 𝛽1, … , 𝛽𝑚+1 are linearly independent .

Definition:

A best approximation to β  V by vectors in a subspace W of V is a vector α  W such that

 −    −  for every vector   W .

12
Theorem

Let W be a subspace of an inner product space V and let  V .

1. The vector  W is a best approximation to  V by vectors in W if and

only if  − is orthogonal to every vector in W .

2. If a best approximation to  V by vectors in W exists, it is unique.

3. If W is finite-dimensional and {𝛼1, 𝛼2 , … 𝛼𝑛 } is any orthonormal basis for W ,

then the vector

 =
n
(  |  k )
k
2 k
k =1

is the (unique) best approximation to  by vectors in W.

Definition:

Let V be an inner product space and S be any set of vectors in V. The orthogonal complement

of S is the set S ⊥ of all vectors in V which are orthogonal to every vector in S.

That is, S ⊥ =  V : ( |  ) = 0,    S

Definition:

Whenever the vector α in the above theorem exists it is called the orthogonal projection of

β on W. If every vector in V has an orthogonal projection on W, the mapping that assigns to

each vector in V its orthogonal projection on W is called the orthogonal projection of V on W.

Corollary :

Let V be an inner product space and W a finite dimensional subspace and E be the

orthogonal projection of V on W. Then the mapping

β →β – Eβ

is the orthogonal projection of V on W ⊥ .

13
Proof :

Let β  V . Then β – Eβ  W ⊥ , and for any γ  W ⊥ , β – γ = E β+(β – Eβ – γ)

Since Eβ  W and β – Eβ – γ  W ⊥ ,

It follows that

||𝛽 – 𝛾||2 = (Eβ+(β – Eβ – γ) ,E β+(β –Eβ – γ))

= ||𝐸𝛽||2+||𝛽 – 𝐸𝛽 – 𝛾||2

≥ ||𝛽 – (𝛽 – 𝐸𝛽)||2

with strict inequality when γ≠ β – Eβ . Therefore, β – Eβ is the best approximation to β by

vectors in W ⊥ .

Theorem

Let W be a finite dimensional subspace of an inner product space V and let E be the orthogonal

projection of V on W. Then E is an idempotent linear transformation of V onto W , W ⊥ is the

null space of E , and V= W ⨁ W ⊥ .

Proof

Let β be an arbitrary vector in V. Then Eβ is the best approximation to β that lies in W .

In particular, Eβ =β when β is in W . Therefore, E(Eβ) =Eβ for every β in V; that is, E is

idempotent : 𝐸 2 = E . To prove that E is linear transformation, let α and β be any vectors in V

and c an arbitrary scalar ,Then by theorem,

α-Eα and β-Eβ are each orthogonal to every vector in W . Hence the vector

c(α-Eα)+(β-Eβ)=(cα +β)-(cEα +Eβ)

Also belongs to W ⊥ . Since cEα+ Eβ is a vector in W , it follows from theorem that

E(cα+ β)= cEα+ Eβ.

Again let β be any vector in V. Then Eβ is the unique vector in W such that β-Eβ is in W ⊥ .

Thus Eβ=0 when β is in W ⊥ .

Conversely, β is in W ⊥ when Eβ=0. Thus W ⊥ is the null space of E.

14
The equation ,

β = E β+β – Eβ

shows that V = W + W ⊥ ; moreover W  W ⊥ = {0} ; for if α is a vector in W  W ⊥ ,then

( |  ) =0. Therefore, α=0 and V is the direct sum of W and W ⊥ .

Corollary :

Under the conditions of theorem, I − E is the orthogonal projection of V on W ⊥ .

It is an independent linear transformation of V onto W ⊥ with null space W .

Proof :

We have seen that the mapping β →β- E β is the orthogonal projection of V on W ⊥ .

Since E is a linear transformation , this projection W ⊥ is the linear transformation I − E from

its geometric properties one sees that I − E is an idempotent .Transformation of V onto W .

This also follows from the computation ( I − E )( I − E )= I − E - E +𝐸 2

=I −E

Moreover , ( I − E )β =0 If and only if β = Eβ , and this is the case if and only if β is in W .

Therefore W is the null space of I − E .

15
 BIBLIOGRAPHY

● Balmohan V .Limaye :Functional analysis

● Kenneth Hoffman,Ray Kunze:Linear Algebra,second edition

● Stephen H.Friedberg, Arnold J.Insel, Lawrence E.Spence : Linear Algebra, Fourth

edition

16
 GRAPH COLORING

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
SHARUNJITH M S
DB18CMSR30
Under the guidance of
MRS. Riya Baby

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
March 2021
Examiners 1: Examiner 2:
 CERTIFICATE

It is to certify that this project report ‘GRAPH COLORING’ is the bona fide
project of SHARUNJITH M S who carried out the project under my supervision.

Mrs. Riya Baby

Supervisor, HOD

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I, SHARUNJITH M S, hereby declare that this project report entitled

“GRAPH COLORING” is an original record of studies and bona fide project
carried out by me during the period from November 2019 to March 2020, under
the guidance of Mrs. Riya Baby, Department of Mathematics, Don Bosco Arts
and Science College, Angadikadavu and has not been submitted by me elsewhere
for the award of any degree, diploma, title or recognition, before.

SHARUNJITH M S
Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 ACKNOWLEDGEMENT

I sincerely express my deep sense of gratitude to all who have been of great
help to me during the course of my dissertation. First and foremost I thank the
almighty, for his blessing and protection during the period of this work. I express
my thanks to Dr. Fr. Francies Karakkatt, Principal, for support in the completion of
this dissertation. I express my gratitude to Mrs. Riya Baby, my project supervisor,
for the constant encouragement, valuable guidance and timely corrections, which
made this work a success.

I am also indebted to all my classmates and friends who supported me

throughout the study. I would like to express my thanks to my parents and dear
ones for their constant encouragement and support. I also thank all those who
helped me directly or indirectly to complete this project.

SHARUNJITH M S
 CONTENTS

Sl no. Title Page No.

01 Introduction 1
02 Basic Concepts 2

03 Chapter 1-Coloring 3–6

04 Chapter 2-Chromatic Number 7 – 16

05 Chapter 3-Applications Of Graph Colouring 17 – 19

06 Conclusion 20
07 Bibliography 21
 INTRODUCTION

A proper coloring of a graph is an assignment of colors to the vertices of the graph so that

no two adjacent vertices have the same color.

Usually we drop the word “proper” unless other types of coloring are also under

discussion. Of course, the “colors” don’t have to be actual colors ; may can be any distinct

labels - integers ,for examples , if a graph is not connected , each connected component can

be colored independently; except where otherwise noted , we assume graphs are

connected. We also assume graphs are simple in this section. Graph coloring has many

applications in addition to its intrinsic interest.

In the same way the most important concept of graph coloring is utilized in

resource allocation, scheduling. Also, paths, walks and circuits in graph theory are used in

tremendous applications say travelling salesman problem, database design concepts,

resource networking.

This project deals with coloring which is one of the most important topics in

graph theory. In this project there are three chapters. First chapter is coloring . The second

chapter is chromatic number. The last chapter deals with application of graph coloring.

1
 BASIC CONCEPTS

1. GRAPH

A graph is an ordered triplet. G=(V(G), E(G), I(G)); V(G) is a non empty set, E(G) is a set

disjoint from V(G) and I(G) is an incidence map that associates each element of E(G) and

unrecorded pair of element of V(G). The elements of V(G) are called vertices (or nodes or

points) of G and the elements of E(G) are Called edges or lines of G.

2. MULTIPLE EDGE / PARALLEL EDGE

A set of 2 or more edges of a graph G is called a multiple edge or parallel edge if they

have the same end vertices.

3. LOOP
An edge for which the 2 end vertices are same is called a loop.

4. SIMPLE GRAPH
A graph is simple if it has no loop and no multiple edges.

5. DEGREE

Let G be a graph and v € V the number of edge incident at V in G is called the degree or

vacancy of the vertex v in G.

2
 CHAPTER - 1

COLORING

Graph coloring is nothing but a simple way of labeling graph components such as

vertices , edges and regions under some constraints. In a graph, no two adjacent vertices,

adjacent edges , or adjacent regions are colored with minimum number of colors .This

number is called the chromatic number and the graph is called properly colored graph.

In graph theory coloring is a special case of graph labeling; it is

an assignment of labels traditionally called “colors” to elements of a graph subject to certain

constraints. In it simplest form, it is a way of coloring the vertices of a graph such that no

two adjacent vertices share the same color, it is called vertex coloring. Similarly, edge

coloring assigns a color to each edge so that no two adjacent edges share the common

color.

While graph coloring , the constraints that are set on the graph are

colors , order of coloring , the way of assigning color , etc. A coloring is given to a vertex or a

particular region . Thus, the vertices or regions having same colors form independent sets.

3
VERTEX COLORING

Vertex coloring is an assignment of colors to the vertices of a graph ‘G ’ such

that no two adjacent vertices have the same color .Simply put , no two vertices of an edge

should be of the same color.

The most common type of vertex coloring seeks to minimize the

number of colors for a given graph . Such a coloring is known as a minimum vertex coloring ,

and the minimum number of colors which with the vertices of a graph may be colored is

called the chromatic number .

CHROMATIC NUMBER:

The minimum number of colors required for vertex coloring of graph ‘ G ’

is called as the chromatic number of G , denoted by X (G) .

X(G) = 1 iff ‘ G ’ is a null graph. If ‘G ’ is not a null graph , then X(G) ≥ 2.

EXAMPLES;

1. 2.

Null Graph ( X (G) = 1 ) Not Null Graph ( X (G) = 2 )

4
EDGE COLORING

An edge coloring of a graph G is a coloring of the edges of G such that adjacent

edges ( or the edges bounding different regions ) receive different colors. An edge coloring

containing the smallest possible number of colors for a given graph is known as a minimum

edge coloring.

The edge chromatic number gives the minimum number of colours with which
graph’s edges can be colored.

CHROMATIC INDEX

The minimum number of colors required for proper edge coloring of graph is

called chromatic index.

A complete graph is the one in which each vertex is directly connected with all

other vertices with an edge. If the number of vertices of a complete graph is n, then the
chromatic

index for an odd number of vertices will be n and the chromatic index for even number of

vertices will be n-1.

5
EXAMPLES;

1.

The given graph will require 3 unique colors so that no two incident edges have the

Same color. So its chromatic index will be 3.

2.

The given graph will require 2 unique colors so that no two incident edges have

the same color. So its chromatic index will be 2.

6
 CHAPTER 2

Chromatic Number

The chromatic number of a graph is the smallest number of colors needed to color the vertices
of so that no two adjacent vertices share the same color. That is the smallest value of possible
to obtain a k-coloring.

 Graph Coloring is a process of assigning colors to the vertices of a graph.

 It ensures that no two adjacent vertices of the graph are colored with the same color.
 Chromatic Number is the minimum number of colors required to properly color any graph.

Graph Coloring Algorithm

 There exists no efficient algorithm for coloring a graph with minimum number of colors.

However, a following greedy algorithm is known for finding the chromatic number of any given
graph.

Greedy Algorithm

Step-01:

Color first vertex with the first color.

7
Step-02:

Now, consider the remaining (V-1) vertices one by one and do the following-

 Color the currently picked vertex with the lowest numbered color if it has not been used to
color any of its adjacent vertices.
 If it has been used, then choose the next least numbered color.
 If all the previously used colors have been used, then assign a new color to the currently
picked vertex.

Problems Based On Finding Chromatic Number of a Graph

Problem-01:

Find chromatic number of the following graph-

8
Solution-

Applying Greedy Algorithm, we have

Vertex a b C d e f

Color C1 C2 C1 C2 C1 C2

From here,

 Minimum numbers of colors used to color the given graph are 2.

 Therefore, Chromatic Number of the given graph = 2.

The given graph may be properly colored using 2 colors as shown below-

9
Problem-02:

Find chromatic number of the following graph-

Solution-

Applying Greedy Algorithm, we have-

Vertex a b C d e f

Color C1 C2 C2 C3 C3 C1

From here,

 Minimum numbers of colors used to color the given graph are 3.

 Therefore, Chromatic Number of the given graph = 3.

10
The given graph may be properly colored using 3 colors as shown below-

Chromatic Number of Graphs

Chromatic Number of some common types of graphs are as follows-

1. Cycle Graph-

 A simple graph of ‘n’ vertices (n>=3) and ‘n’ edges forming a cycle of length ‘n’ is called as a
cycle graph.
 In a cycle graph, all the vertices are of degree 2.

Chromatic Number
 If number of vertices in cycle graph is even, then its chromatic number = 2.
 If number of vertices in cycle graph is odd, then its chromatic number = 3.

11
Examples-

2. Planar Graphs-

A planar graph is a graph that can be embedded in the plane, that is it can be drawn on the
plane in such a way that its edges intersect only at their endpoint. In other words, it can be
drawn in such a way that no edges cross each other.

12
A Planar Graph is a graph that can be drawn in a plane such that none of its edges cross each
other.

Chromatic Number
Chromatic Number of any Planar Graph is less than or equal to 4

Examples-
+

 All the above cycle graphs are also planar graphs.

 Chromatic number of each graph is less than or equal to 4.

13
3. Complete Graphs-

 A complete graph is a graph in which every two distinct vertices are joined by exactly one
edge.
 In a complete graph, each vertex is connected with every other vertex.
 So to properly it, as many different colors are needed as there are number of vertices in the
given graph.

Chromatic Number
Chromatic Number of any Complete Graph

= Number of vertices in that Complete Graph

Examples-

14
4. Bipartite Graphs-

A bipartite graph is a graph whose vertices can be divided into two disjoint and independent
sets U and V such that every edge connects a vertex in U to one in V. Vertex sets U and V are
usually called the parts of the graph.

 A Bipartite Graph consists of two sets of vertices X and Y.

 The edges only join vertices in X to vertices in Y, not vertices within a set.

Chromatic Number
Chromatic Number of any Bipartite Graph

=2

Example-

15
5. Trees-

A tree is an undirected graph in which any two vertices are connected by exactly one path, or
equivalently a connected acyclic undirected graph.

 A Tree is a special type of connected graph in which there are no circuits.

 Every tree is a bipartite graph.
 So, chromatic number of a tree with any number of vertices = 2.

Chromatic Number
Chromatic Number of any tree

=2

Examples-

16
 CHAPTER-3
APPLICATIONS OF GRAPH COLORING

1) Making Schedule or Time Table:

Suppose we want to make an exam schedule for a university. We have list

different subjects and students enrolled in every subject. Many subjects would have common
students (of same batch, some backlog students, etc). How do we schedule the exam so that no
two exams with a common student are scheduled at same time? How many minimum time
slots are needed to schedule all exams? This problem can be represented as a graph where
every vertex is a subject and an edge between two vertices mean there is a common student.
So this is a graph coloring problem where minimum number of time slots is equal to the
chromatic number of the graph.

2) Mobile Radio Frequency Assignment:

When frequencies are assigned to towers, frequencies assigned to all towers at the
same location must be different. How to assign frequencies with this constraint? What is the
minimum number of frequencies needed? This problem is also an instance of graph coloring
problem where every tower represents a vertex and an edge between two towers represents
that they are in range of each other.

3) Register Allocation:

In compiler optimization, register allocation is the process of assigning a large number

of target program variables onto a small number of CPU registers. This problem is also a graph
coloring problem.

4) Sudoku:

Sudoku is also a variation of Graph coloring problem where every cell represents a
vertex. There is an edge between two vertices if they are in same row or same column or same
block.

17
 5) Map Coloring:

Geographical maps of countries or states where no two adjacent cities cannot be

assigned same color. Four colors are sufficient to color any map.

6) Bipartite Graphs:

We can check if a graph is bipartite or not by coloring the graph using two colors. If a
given graph is 2-colorable, then it is Bipartite, otherwise not. See this for more details.

Explanation;

Algorithm:

A bipartite graph is possible if it is possible to assign a color to each vertex such that no
two neighbour vertices are assigned the same color. Only two colors can be used in this
process.

Steps:
1. Assign a color (say red) to the source vertex.

2. Assign all the neighbours of the above vertex another color (say blue).

3. Taking one neighbour at a time, assign all the neighbour's neighbours the color red.

4. Continue in this manner till all the vertices have been assigned a color.

5. If at any stage, we find a neighbour which has been assigned the same color as that of the
current vertex, stop the process. The graph cannot be colored using two colors. Thus the graph
is not bipartite.

18
Example:

19
 CONCLUSION

This project aims to provide a solid background in the basic topics of graph coloring. Graph
coloring problem is to assign colors to certain elements of a graph subject to certain
constraints. The nature of coloring problem depends on the number of colors but not on what
they are.

The study of this topic gives excellent introduction to the subject called “Graph
Coloring”.

This project includes two important topics such as vertex coloring and edge coloring and came
to know about different ways and importance of coloring.

Graph coloring enjoys many practical applications as well as theoretical challenges.

Besides the applications, different limitations can also be set on the graph or on the away a color
is assigned or even on the color itself. It has been reached popularity with the general public in
the form of the popular number puzzle Sudoku and it is also use in the making of time
management which is an important application of coloring. So graph coloring is still a very
active field of research.

20
 BIBLIOGRAPHY

1) V.K Balakrishnan, outlines, graph theory.

2) J.A Bondy and U.S.R Murty, graph theory with applications.

3) S.Arumugam, S.Ramachandran, invitation to graph theory.

21
NUMBER THEORETIC FUNCTION
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
SHILJI KURIAN
DB18CMSR08
Under the guidance of
Ms. Ajeena joseph

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE

Certified that this project ‘Number Theoretic Function’ is a bona fide

project of SHILJI KURIAN carried out the project work under my
supervision.

Mrs. Riya Baby Ms. Ajeena joseph

Head of Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 DECLARATION

I SHILJI KURIAN hereby declare that the project ‘Number Theoretic Function’ is
an original record of studies and bona fide project carried out by me during the period
of 2018 – 2021 under the guidance of Ms. Ajeena joseph, Department of Mathematics,
Don Bosco Arts and Science College, Angadikkadavu and has not submitted by me
elsewhere for the award of my degree, diploma, title or recognition, before.

Name
SHILJI KURIAN

Department of Mathematics
Don Bosco Arts and Science College,
Angadikkadavu
 ACKNOWLEDGEMENT
Introduction is the proper place to begin. But first I bow my head before the Almighty
who is always with me. Also, I must express my deepest gratitude to people along the
way.

No words can adequately express the sense of gratitude; still, I try to express my
heartfelt thanks through words. At the outset, I am deeply indebted to my project
supervisor Ms. Ajeena joseph, Department of Mathematics, Don Bosco Arts and
Science College, Angadikkadavu, for the invaluable guidance, loving encouragement
and meticulous care towards me throughout my career. I express my deep sense of
gratitude to all the faculty members of the Department of Mathematics, Don Bosco
Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the Principal and the
staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the successful
completion of this project. It is my pleasure and duty to thank each and every one of
them who walked with me.

My greatest debt is always, to God Almighty.

 CONTENTS

SI no. Title Page No.

01 Introduction 1

02 Preliminary 2-3

03 Chapter 1- ARITHMETIC FUNCTION 4-10

04 Chapter 2- EULER’S 𝜙 FUNCTION 11-20

05 Conclusion 21

06 Bibliography 22
 INTRODUCTION

A Number Theoretic Function is a complex valued function defined for all positive
integers. In Number Theory, there exist many number theoretic functions. This
includes Divisor Function, Sigma Function, Euler’s-Phi Function and Mobius
Function. All these functions play a very important role in the field of Number
Theory.

In the first chapter we will discuss about Arithmetic Function. In the second chapter
we will introduce Euler’s-Phi Function and Mobius Function.

1
 PRELIMINARY

Let n be a fixed positive integer. Two integers a and b are said to be congruent
modulo n, symbolized by
a ≡ b (mod n)
if n divides the difference a − b; that is, provided that a − b = kn for some integer k.
Example:
To fix the idea, consider n = 7. It is routine to check that
3 ≡ 24 (mod 7) − 31 ≡ 11 (mod 7) − 15 ≡ −64 (mod 7)
Because 3 − 24 = (−3)7, −31 − 11 = (−6)7 and −15 − (−64) = 77. When
n does not divide (a − b), we say that a is incongruent to b modulo n, and in this case
we write
a ≢ b (mod n). For a simple example: 25 ≢ 12 (mod 7), because 7 fails to divide

25 − 12 = 13.
It is to be noted that any two integers are congruent modulo 1, whereas two integers
are congruent modulo 2 when they are both even or both odd. In as much as
congruence modulo 1 is not particularly interesting, the usual practice is to assume
that n > 1.
Remark:
Given an integer a, let q and r be its quotient and remainder upon division by n,
so that
a = qn + r 0≤r<n
Then, by definition of congruence, a ≡ r (mod n). Because there are n choices for
r , we see that every integer is congruent modulo n to exactly one of the values
0, 1, 2, . . . , n − 1; in particular, a ≡ 0 (mod n) if and only if n | a.

2
Fundamental Theorem of Arithmetic
is Every integer 𝑛 > 1 can be represented as Product of prime factor in only one way,
apart from the order of the factors.

Residue

If a is an integer and 𝑚 is a positive integer then the residue class of a modulo 𝑚 is

denoted by 𝑎ˆ and is given by

𝑎ˆ = {𝑥: 𝑥 ≡ 𝑎(𝑚𝑜𝑑𝑚)}
= {𝑥: 𝑥 = 𝑎 + 𝑚𝑘, 𝑘 = 0, ±1, ±2, ⋯ }

3
 CHAPTER 1
ARITHMETIC FUNCTION
An arithmetic Function is a function defined on the positive integers which take
values in the real or complex numbers. i.e., A function f: N→ C is called an
arithmetic function.
An arithmetic function is called multiplicative if f(mn) = f(m)f(n) for all coprime
natural numbers m and n.
Examples
a) Sum of divisors 𝜎(n)
b) Number of divisors 𝜏(n)
c) Euler’s function 𝜙(n)
d) Mobius function 𝜇(n)

Definition 1.1

Given a positive integer n, let τ (n) denote the number of positive divisors of n and

σ(n) denote the sum of positive divisors of n.

Example

Consider n = 12. Since 12 has the positive divisors 1, 2, 3, 4, 6, 12, we find that

τ (12) = 6 and σ(12) = 1 + 2 + 3 + 4 + 6 + 12 = 28

For the first few integers,

τ (1) = 1 τ (2) = 2 τ (3) = 2 τ (4) = 3 τ (5) = 2 τ (6) = 4, . . .

σ(1) = 1, σ(2) = 3, σ(3) = 4, σ(4) = 7 , σ(5) = 6, σ(6) = 12, . . .

It is not difficult to see that τ (n) = 2 if and only if n is a prime number; also,

σ(n) = n + 1 if and only if n is a prime.

4
Theorem 1.1
𝑘 𝑘
If n = 𝑝1 1 … … … … . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

(a) τ (n) = (k1+ 1)(k2 + 1) ・ ・ ・ (kr + 1), and

𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
(b) σ(n) = ………………….
𝑝1 −1 𝑝𝑟 −1

Proof

The positive divisors of n are precisely those integers

𝑎 𝑎 𝑎
d = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟

where 0 ≤ ai ≤ ki . There are k1 + 1 choices for the exponent a1; k2 + 1 choices for a2, .

. . ; and kr + 1 choices for ar . Hence, there are

(k1 + 1)(k2 + 1) · · · (kr + 1)

possible divisors of n.

To evaluate σ(n), consider the product

𝐾 𝐾
(1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Each positive divisor of n appears once and only once as a term in the expansion of

this product, so that

𝐾 𝐾
σ(n) = (1 + 𝑝1 + 𝑃12 + ⋯ … … … 𝑃1 1 ) (1 + 𝑝2 + 𝑃22 + ⋯ … … … 𝑃2 2 )………………..

𝐾
………….(1 + 𝑝𝑟 + 𝑃𝑟2 + ⋯ … … … 𝑃𝑟 𝑟 )

Applying the formula for the sum of a finite geometric series to the ith factor on the

right-hand side, we get

𝑘 +1
𝐾 𝑝𝑖 𝑖 − 1
(1 + 𝑝𝑖 + 𝑃𝑖2 + ⋯ … … … 𝑃𝑖 𝑖 ) =
𝑝𝑖 − 1

5
It follows that
𝑘 +1 𝑘 +1
𝑝1 1 −1 𝑝𝑟 𝑟 −1
σ(n) = 𝑝1 −1
………………….
𝑝𝑟 −1
.

Corresponding to the ∑ notation for sums, the notation for products may be

defined using ∏ , the Greek capital letter pi. The restriction delimiting the numbers

over which the product is to be made is usually put under the ∏

sign.

Examples

With this convention, the conclusion to Theorem 1.1 takes the compact form: if
𝑘 𝑘 𝑘
n = 𝑝1 1 𝑝2 2 … … . . 𝑝𝑟 𝑟 is the prime factorization of n > 1, then

and

6
Theorem 1.2

The functions τ and σ are both multiplicative functions

Proof

Let m and n be relatively prime integers. Because the result is trivially true if

either m or n is equal to 1, we may assume that m > 1 and n > 1. If

are the prime factorizations of m and n . It follows that the prime factorization of

the product mn is given by

Applying to theorem 1.1, we obtain

In a similar fashion, theorem 1.1 gives

Thus, τ and σ are multiplicative functions.

Theorem 1.3

If f is a multiplicative function and F is defined by

then F is also multiplicative.

7
Proof

Let m and n be relatively prime positive integers. Then

because every divisor d of mn can be uniquely written as a product of a divisor d1

of m and a divisor d2 of n, where gcd(d1, d2) = 1. By the definition of a

multiplicative function,

f (d1d2) = f (d1) f (d2)

It follows that

It might be helpful to take time out and run through the proof of Theorem 1.3

in a concrete case. Letting m = 8 and n = 3, we have

= f (1) + f (2) + f (3) + f (4) + f (6) + f (8) + f (12) + f (24)

= f (1 · 1) + f (2 · 1) + f (1 · 3) + f (4 · 1) + f (2 · 3)+ f (8 · 1) + f (4 · 3) + f (8 · 3)

= f (1) f (1) + f (2) f (1) + f (1) f (3) + f (4) f (1) + f (2) f (3)+ f (8) f (1)

. + f (4)f(3)+ f (8) f (3)

8
= [ f (1) + f (2) + f (4) + f (8)][ f (1) + f (3)]

= F(8)F(3)

Theorem 1.3 provides a deceptively short way of drawing the conclusion that τ

and σ are multiplicative

The Mangoldt function 𝚲(𝒏)

Definition 1.2

For every integer 𝑛 ≥ 1 we define

log 𝑝 if 𝑛 = 𝑝𝑚 for some prime 𝑝 and some 𝑚 ≥ 1 ,

Λ(𝑛) = {
0 otherwise.

Here is a short table of values of Λ(𝑛) :

𝑛: 1 2 3 4 5 6 7 8 9 10
Λ(𝑛): 0 log 2 log 3 log 2 log 5 0 log 7 log 2 log 3 0

The proof of the next theorem shows how this function arises naturally from the

fundamental theorem of arithmetic.

Theorem 1.4

If 𝑛 ≥ 1 we have

log 𝑛 = ∑ Λ(𝑑 ) … … … … … … … … … … … … (1)

𝑑∣𝑛

Proof

The theorem is true if 𝑛 = 1 since both members are 0 . Therefore, assume that 𝑛 > 1

and write

9
 𝑟

𝑛 = ∏ 𝑝𝑘 𝑎𝑘
𝑘=1

Taking logarithms we have

𝑟

log 𝑛 = ∑ 𝑎𝑘 log 𝑝𝑘
𝑘=1

Now consider the sum on the right of (1). The only nonzero terms in the sum come

from those divisors 𝑑 of the form 𝑝𝑘 𝑚 for 𝑚 = 1,2, … , 𝑎𝑘 and 𝑘 = 1,2, … , 𝑟. Hence

𝑟 𝑎𝑘 𝑟 𝑎𝑘 𝑟

∑ Λ(𝑑) = ∑ ∑ Λ(𝑝𝑘𝑚 ) = ∑ ∑ log 𝑝𝑘 = ∑ 𝑎𝑘 log 𝑝𝑘 = log 𝑛

𝑑∣𝑛 𝑘=1 𝑚=1 𝑘=1 𝑚=1 𝑘=1

which proves (1).

10
 CHAPTER 2
EULER’S 𝝓 FUNCTION
Let n be positive integer. Let Un denote the set of all positive integers less than n and
coprime to it

For example,

U6 = {1,5}

U10 = {1,3,7,9}

U18 = {1,5,7,11,13,17}

Definition 2.1
Euler’s 𝜙 function is a function 𝜙: N→N such that for any n ∈ N, 𝜙 (n) is the number
of integers less than n and coprime to it

In other words

‘Euler’s 𝜙 function counts the number of elements in Un’

For example,

𝜙(1) = 1, 𝜙(2) = 1, 𝜙(3) = 2, 𝜙(4) = 2, 𝜙(5) = 4

𝜙(6) = 2 … .

Theorem 2.1
Let p be a prime. Then 𝜙 (p) = p-1

Proof:
By definition, any natural number strictly less than p is coprime to p, hence

𝜙 (p) = p-1

Theorem 2.2

If 𝑝 is a prime and 𝑘 > 0, then

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1)

11
Proof:

Consider the successive pk natural numbers not greater than pk arranged in the
following rectangular array of p columns and pk-1 rows

1 2 . . p

p+1 p+2 . . 2p

. . . . .

. . . . .

pk-p+1 pk-p+2 . . pk

among these numbers only the ones at the rightmost sides are not coprime to pk and
there are pk-1 members in that column. So

𝜙(𝑝𝑘 ) = 𝑝𝑘 − 𝑝𝑘−1 = 𝑝𝑘−1 (𝑝 − 1).

For example, 𝜙(8) = 23 − 22 = 4 which counts the number of elements in the set
U8 = {1,3,5,7}

By the fundamental theorem of arithmetic, we can write any natural number n as

𝑘 𝑘
n= 𝑝1 1 … … … . . 𝑝𝑟 𝑟

where 𝑃𝑖 ‘s are distinct prime and k 𝑖 ≥ 1 are integers. We already know how to find
𝑘 𝑘
𝜙(𝑝𝑖 𝑖 ) we would lie to see how 𝜙(𝑛) is related to 𝜙(𝑝𝑖 𝑖 ). This follows from a very
important property of Euler’s 𝜙 Function

Multiplicativity of Euler’s 𝝓 Function

Theorem 2.3

𝜙(mn) = 𝜙(m)𝜙(n) if m and n are coprime natural numbers.

Proof:

12
Consider the array of natural numbers not greater than mn arranged in m columns and
n rows in the following manner

1 2 ⋯ 𝑟 ⋯ 𝑚
𝑚+1 𝑚+2 𝑚+𝑟 2𝑚
2𝑚 + 1 2𝑚 + 2 2𝑚 + 𝑟 3𝑚
⋮ ⋮ ⋮ ⋮
(𝑛 − 1)𝑚 + 1 (𝑛 − 1)𝑚 + 2 (𝑛 − 1)𝑚 + 𝑟 𝑛𝑚

Clearly each row of the above array has m distinct residues modulo m. Each column
has n distinct residues modulo n: for 1 ≤ 𝑖, 𝑖 ≤ 𝑛 − 1

im +j ≡ im + j (mod n)

⇒ im ≡ im (mod n)

⇒ i ≡ i (mod n) (as gcd(m,n) = 1)

⇒i≡i

Each row has 𝜙(m) residues coprime to m, and each column has 𝜙(n) residues
coprime to n. Hence in total 𝜙(m)𝜙(n) elements in the above array which are
coprime to both m and n, it follows that

𝜙(mn) = 𝜙(m)𝜙(n)

Theorem 2.4
Let n be any natural numbers, then
1 1 1
𝜙(𝑛) = 𝑛 (1 − ) (1 − ) ⋯ (1 − )
𝑝1 𝑝2 𝑝𝑟

Proof:
By fundamental theorem of arithmetic, we can write
𝑘 𝑘 𝑘
𝑛 = 𝑃1 1 𝑃2 2 … … . … 𝑃𝑟 𝑟

Where 𝑝𝑖 are the distinct prime factor of n, and 𝑘𝑖 are the non negative integers. By
previous theorem and proposition,
𝑘 𝑘
𝜙(𝑛) = 𝜙(𝑝1 1 ) ⋅ … , 𝜙(𝑝𝑟 𝑟 )
𝑘 −1 𝑘
= 𝑃1 1 (𝑃1 − 1) ⋯ 𝑃𝑟 𝑟−1 (𝑃𝑟 − 1)

13
 𝑘 1 𝑘 1
= 𝑝1 1 (1 − ) ⋯ 𝑃𝑟 𝑟 (1 − )
𝑃1 𝑃𝑟

1 1
= 𝑛 (1 − ) ⋯ ⋅ (1 − )
𝑝1 𝑝𝑟

Theorem 2.5
For n > 2, 𝜙 (n) is an even integer.

Proof:

First, assume that n is a power of 2, let us say that n = 2k , with k ≥ 2. By

theorem 2.2,
1
𝜙(𝑛) = 𝜙(2𝑘 ) = 2𝑘 (1 − ) = 2𝑘−1
2
an even integer. If 𝑛 does not happen to be a power of 2, then it is divisible by an odd
prime 𝑝; we therefore may write 𝑛 as 𝑛 = 𝑝𝑘 𝑚, where 𝑘 ≥ 1 and gcd (𝑝𝑘 , 𝑚) = 1.
Exploiting the multiplicative nature of the phi-function, we obtain

𝜙(𝑛) = 𝜙(𝑝𝑘 )𝜙(𝑚) = 𝑝𝑘−1 (𝑝 − 1)𝜙(𝑚)

which again is even because 2 | p – 1.

Theorem 2.6
For each positive integer n,

𝑛 = ∑ 𝜙(𝑑)
𝑑∣𝑛

Proof:
Let us partition the set {1,2,…….,n} into mutually disjoint subsets Sd for each d/n,
where

𝑆𝑑 = {1 ≤ 𝑚 ≤ 𝑛 ∣ gcd (𝑚, 𝑛) = 𝑑}
𝑚 𝑛 𝑚 𝑛
= {1 ≤ ≤ ∣ gcd ( , ) = 1}
𝑑 𝑑 𝑑 𝑑

Then

14
 {1,2, … … . , n} = ∑ 𝑆𝑑
𝑑∣𝑛

𝑛
⇒ 𝑛 = ∑ 𝜙( )
𝑑
𝑑∣𝑛

= ∑ 𝜙(𝑑)
𝑑∣𝑛

As for each divisor of n, n/d is also a divisor of n

MOBIUS FUNCTION

Definition 2.2
The Mobius function 𝜇: 𝑁 ⟶ {0, ±1} is defined as

1 if 𝑛 = 1
𝜇(𝑛) = {0 if 𝑝2 ⁄𝑛 for some prime 𝑝
(−1)𝑟 if 𝑛 = 𝑝1 𝑝2 ⋯ 𝑝𝑟 , where 𝑝𝑖 are distinct primes

For example,

μ(1) = 1 μ(2) = −1 μ(3) = −1

μ(4) = 0 μ(5) = −1 μ(6) = 1

If 𝑝 is a prime number, it is clear that 𝜇(𝑝) = −1; in addition, 𝜇(𝑝𝑒 ) = 0 for 𝑒 ≥ 2.

Theorem 2.7
The Mobius function is a multiplicative function i.e.

μ(mn) = μ(m)μ(n), if m and n are relatively prime

Proof:
Let m and n be coprime integers, we can consider the following to cases
2
Case 1: let μ(mn) = 0 then there is a prime p such that 𝑝 ⁄𝑚𝑛. As m and n are
𝑝2⁄ 𝑝2⁄
coprime p cannot divide both m and n hence either 𝑚 or 𝑛 . Therefore either
μ(m) = 0 or μ(n) = 0 and we have μ(mn) = μ(m)μ(n)

Case 2: suppose that μ(mn) ≠ 0 then mn is square free, hence so are m and n. let
15
𝑚 = 𝑝1 … … … 𝑝𝑟 and 𝑛 = 𝑞1 … … … 𝑞𝑠 where 𝑝𝑖 𝑎𝑛𝑑 𝑞𝑗 are all distinct primes then
mn = 𝑝1 … … … 𝑝𝑟 𝑞1 … … … 𝑞𝑠 where all the primes occurring in the factorization of
mn are distinct. Hence

𝜇(𝑚𝑛) = (−1)𝑟+𝑠

= (−1)𝑟 (−1)𝑠

= μ(m)μ(n)

Theorem 2.8

1 if 𝑛 = 1
∑ 𝜇(𝑑) = {
0 if 𝑛 > 1
𝑑∣𝑛

Where d runs through all the positive divisors of n.

Proof:

𝐿𝑒𝑡 𝐹(𝑛) = ∑ 𝜇(𝑑)

𝑑∣𝑛

As μ is multiplicative, so is F(n) by the theorem (F be a multiplicative arithmetic

function 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ) then F is also a multiplicative arthmetic function)
𝑑∣𝑛

Clearly

𝐹 (1) = ∑ 𝜇 (𝑑 )
𝑑∣𝑛

= μ(1)

=1

For integers which are prime power, i.e. of the form pk for some k ≥ 1

𝐹 (𝑝2 ) = μ(1) + μ(p) + μ(p2 ) + ⋯ … … … . . +μ(p𝑘 )

= 1 + (-1) + 0……………+ 0

=0

Now consider any integer n, and consider its prime factorization. Then
𝑘 𝑘
𝑛 = 𝑝1 1 … … … … … . 𝑝𝑟 𝑟 , 𝑘𝑖 ≥ 1

16
 𝑘
⇒ 𝐹 (𝑛) = ∏𝐹(𝑝𝑖 𝑖 )

=0

Mobius inversion formula

The following theorem is known as Mobius inversion formula

Theorem 2.9
Let F and f be two function from the set N of natural number to the field complex
number C such that

𝐹(𝑛) = ∑ 𝑓(𝑑)
𝑑∣𝑛

Then we can express f(n) as

𝑛
𝑓 (𝑛 ) = ∑ 𝜇 (𝑑 )𝐹 ( )
𝑑
𝑑∣𝑛

𝑛
= ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Proof:
First observe that if d is divisor of n so is n/d. Hence both the summation in the last
line of the theorem are same. Now

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

The crucial step in the proof is to observe that the set of S of pairs of integers (c,d)
with d|n and c|n/d is the same as the set T of pairs (c,d) with c/n and d|n/c.

𝑛
∑ 𝜇(𝑑)𝐹 ( ) = ∑ (𝜇(𝑑) ∑ 𝑓(𝑐))
𝑑
𝑑∣𝑛 𝑑∣𝑛 𝑐∣(𝑛/𝑑)

= ∑ ( ∑ 𝜇(𝑑)𝑓(𝑐))
𝑑∣𝑛 𝑐∣(𝑛/𝑑)

17
 = ∑ 𝑓(𝑐)𝜇(𝑑)
(𝑐,𝑑)∈𝑆

= ∑ 𝑓 (𝑐 ) 𝜇 (𝑑 )
(𝑐,𝑑)∈𝑇

= ∑ (𝑓(𝑐) ∑ 𝜇(𝑑))
𝑐∣𝑛 𝑑∣(𝑛/𝑐)

= F(n)
𝐴𝑠 ∑ 𝜇(𝑑 ) = 0 𝑢𝑛𝑙𝑒𝑠𝑠 𝑛⁄𝑐 = 1, which happens when c = n
𝑑∣𝑛

Let us demonstrate this with n = 15

15
∑ 𝜇 (𝑑 )𝐹 ( ) = 𝜇 (1)[ 𝑓 (1) + 𝑓 (3) + 𝑓 (5) + 𝑓 (15)] + 𝜇(3)[ 𝑓 (1) + 𝑓 (5)]
𝑑
𝑑∣15

+𝜇(5)[ 𝑓 (1) + 𝑓 (3)] + 𝜇(15)[ 𝑓 (1)]

= f (1)[μ(1) + μ(3) + μ(5) + μ(15)] + f (3)[μ(1) + μ(5)] + f (5)[μ(1) +

μ (5)] + f(15) μ(1)

= f(1).0 + f(3).0 + f(5).0 + f(15)

= f(15)

The above theorem leads to the following interesting identities

1. we know that for any positive integer n,

∑ 𝜙 (𝑑 ) = 𝑛
𝑑∣𝑛

Where 𝜙(𝑛) is Euler’s 𝜙 function. Hence

𝑛
𝜙 (𝑛 ) = ∑ 𝜇 ( ) 𝑑
𝑑
𝑑∣𝑛

For example,

𝜙 (10) = μ(1)10 + μ(2)5 + μ(5)2 + μ(10)1

18
 = 10 - 5 – 2 + 1

=4

2. similarly

𝜎(𝑛) = ∑ 𝑑
𝑑∣𝑛
𝑛
𝑛 = ∑ 𝜇 ( ) 𝜎(𝑑)
𝑑
𝑑∣𝑛

For example,

With n = 10

μ(10). 1 + μ(2)(1 + 5) + μ(5)(1 + 3) + μ(1)(1 + 3 + 5 + 10)

= 1 – 1 – 5 – 1 – 3 + 1 + 3 +5 + 10

= 10

We have seen before that if multiplicative so is 𝐹 (𝑛) = ∑ 𝑓 (𝑑 ). But we can now

𝑑∣𝑛

Prove that converse applying the Mobius inversion formula

Theorem 2.10
If F is a multiplicative function and

𝐹 (𝑛 ) = ∑ 𝑓 (𝑑 )
𝑑∣𝑛

then f is also multiplicative.

Proof:
By the Mobius inversion formula we know that
𝑛
𝑓 (𝑛) = ∑ 𝜇 ( ) 𝐹(𝑑)
𝑑
𝑑∣𝑛

Let m and n be relatively prime positive integers. We recall that any divisor

d of mn can be uniquely written as d = d1, d2, where d1 |m, d2 | n, and

𝑚 𝑛
gcd(d1, d2) = 1 = gcd( , ).
𝑑1 𝑑2

19
Conversely if d1/m and d2/n then d1d2/mn thus,
𝑚𝑛
𝑓(𝑚𝑛) = ∑ 𝜇(𝑑)𝐹 ( )
𝑑
𝑑∣𝑚𝑛
𝑚𝑛
= ∑ 𝜇 (𝑑1 𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇 (𝑑1 )𝜇(𝑑2 )𝐹 ( )𝐹( )
𝑑1 𝑑2
𝑑1 |𝑚
𝑑2 |𝑛
𝑚 𝑛
= ∑ 𝜇(𝑑1 )𝐹 ( ) ∑ 𝜇 (𝑑2 )𝐹 ( )
𝑑1 𝑑2
𝑑1 ∣𝑚 𝑑2 ∣𝑛
= 𝑓(𝑚)𝑓(𝑛)

In view of the above theorem we can say that as N(n) = n is a multiplicative function
so is 𝜙(𝑛) because

∑ 𝜙(𝑑 ) = 𝑛 = N(n)
𝑑∣𝑛

20
 CONCLUSION

The purpose of this project gives a simple account of Arithmetic function, Euler’s phi
function and Mobius Function. The study of these topics given excellent introduction
to the subject called ‘NUMBER THEORETIC FUNCTION’

Number Theoretic Function demands a high standard of rigor. Thus, our presentation
necessarily has its formal aspect with care taken to present clear and detailed
argument. An understanding of the statement of the theorem, number theory proof is
the important issue. In the first chapter we discuss about function τ and σ are both
multiplicative function. If f is a multiplicative function and F is defined by

𝐹(𝑛) = ∑𝑑∣𝑛 𝑓(𝑑 ), then F is also multiplicative. In the second chapter 2 we discuss
about that if p is prime the 𝜙(𝑝) = 𝑝 − 1, 𝜙(𝑚𝑛) = 𝜙(𝑚)𝜙(𝑛). The Mobius
function is multiplicative function if f is multiplicative function and 𝐹(𝑛)=∑𝑑∣𝑛 𝑓(𝑑 ),

then F is also multiplicative.

21
 BIBLIOGRAPHY
1 DAVID M. BURTON - ELEMENTRY NUMBER THEORY
2 TOM.M. APOSTOL - INTRODUCTION TO ANALYTIC NUMBER THEORY
3 https://en.m.wikipedia.org/wiki/Euler%27s_totient_function
4 https://en.m.wikipedia.org/wiki/Arithmetic_function

22
NORMED LINEAR SPACES
Project report submitted to
The Kannur University
for the award of the degree
of
Bachelor of Science
by
SOORYA DEVI K S
DB18CMSR09
Under the guidance of
Ms. Athulya P

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu
 CERTIFICATE
It is to certify that this project report ‘Normed Linear Spaces’ is the bonafide project of

Soorya Devi K S carried out the project work under my supervision.

Mrs. Riya Baby Ms. Athulya P

Head Of Department Supervisor

Department Of Mathematics
Don Bosco Arts And Science College
Angadikadavu

2
 DECLARATION
I Soorya Devi K S hereby declare that the project ‘Normed Linear Space’ is an original
record of studies and bona fide project carried out by me during the period of 2018 – 2021 under
the guidance of Ms. Athulya P , Department of Mathematics, Don Bosco Arts and Science
College, Angadikadavu and has not submitted by me elsewhere for the award of my degree,
diploma, title or recognition, before.

Soorya Devi K S

DB18CMSR09

Department of Mathematics
Don Bosco Arts and Science College
Angadikadavu

3
 ACKNOWLEDGEMENT

First and foremost, praises and thanks to God, the Almighty, for His showers of blessings
throughout my work to complete the project successfully. I had finally managed to finish up
this project with great enthusiasm and determination. All the time spent to search and discuss
ideas as well as justifying theoretical clues to drive the output were worth my effort and time.

Therefore, I would like to express my sincere gratitude to my supervisor Ms. Athulya P,

Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu, for providing
invaluable guidance, comments and suggestions throughout the course of the project. I also
acknowledge with a deep sense of reverence, my gratitude towards all the faculty members of
the Department of Mathematics, Don Bosco Arts and Science College, Angadikadavu.

I owe and respectfully offer my thanks to the principal and staff of Don Bosco Arts and
Science College, Angadikadavu for their constant moral support and mellifluous affection
provided to me.

I express my gratitude to all who directly or indirectly helped me to complete this project
successfully. Their guidance and support was very helpful in bringing this work to
conclusion.

4
 CONTENTS

SI no Title Page no

1 Introduction 6

2 Preliminaries 7-9

3 Chapter 1 10 - 18

4 Chapter 2 19 - 25

5 Conclusion 26

6 Bibliography 27

5
 INTRODUCTION

This chapter gives an introduction to the theory of normed linear spaces. A skeptical reader may
wonder why this topic in pure mathematics is useful in applied mathematics. The reason is quite
simple: Many problems of applied mathematics can be formulated as a search for a certain
function, such as the function that solves a given differential equation. Usually the function
sought must belong to a definite family of acceptable functions that share some useful properties.
For example, perhaps it must possess two continuous derivatives. The families that arise
naturally in formulating problems are often linear spaces. This means that any linear combination
of functions in the family will be another member of the family. It is common, in addition, that
there is an appropriate means of measuring the “distance” between two functions in the family.
This concept comes into play when the exact solution to a problem is inaccessible, while
approximate solutions can be computed. We often measure how far apart the exact and
approximate solutions are by using a norm. In this process we are led to a normed linear space,
presumably one appropriate to the problem at hand. Some normed linear spaces occur over and
over again in applied mathematics, and these, at least, should be familiar to the practitioner.
Examples are the space of continuous functions on a given domain and the space of functions
whose squares have a finite integral on a given domain.

6
 PRELIMINARIES

1) LINEAR SPACES
We introduce an algebraic structure on a set 𝑋and study functions on X which are well behaved
with respect to this structure. From now onwards , K will denote either R , the set of all real
numbers or C, the set of all complex numbers. For k ∈ C , Re k and Im k will denote the real and
imaginary part of k.

A linear space(or a vector space) over K is a non-empty set 𝑋 along with a function
+ : 𝑋 × 𝑋 → 𝑋, called addition and a function ·: K× 𝑋 → 𝑋 called scalar multiplication, such
that for all 𝑥 , 𝑦 , 𝑧 ∈ 𝑋and 𝑘 , 𝑙 ∈ K , we have

𝑥+ 𝑦 = 𝑦 + 𝑥

𝑥 + (𝑦 + 𝑧) = (𝑥 + 𝑦) + 𝑧

∃0∈𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + 0 = 𝑥,

∃ − 𝑥 ∈ 𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡 𝑥 + (− 𝑥) = 0 ,

𝑘 · (𝑥 + 𝑦) = 𝑘 · 𝑥 + 𝑘 · 𝑦,

(𝑘 + 𝑙)⋅𝑥 = 𝑘 · 𝑥 + 𝑙 · 𝑥,

(𝑘𝑙)⋅𝑥 = 𝑘 · (𝑙 · 𝑥),

1⋅𝑥 = 𝑥.

We shall write 𝑘𝑥 in place of 𝑘 · 𝑥. We shall also adopt the following notations. For
𝑥, 𝑦 ∈ 𝑋, 𝑘 ∈ K and subsets 𝐸, 𝐹of 𝑋 ,

𝑥 + 𝐹 = {𝑥 + 𝑦: 𝑦 ∈ 𝐹},

𝐸 + 𝐹 = {𝑥 + 𝑦: 𝑥 ∈ 𝐸, 𝑦 ∈ 𝐹},`

𝑘𝐸 = {𝑘𝑥: 𝑥 ∈ 𝐸}.

2) BASIS

A nonempty subset 𝐸 of 𝑋 is said to be a subspace of 𝑋 if 𝑘𝑥 + 𝑙𝑦 ∈ 𝐸 whenever 𝑥, 𝑦 ∈ 𝐸 and

𝑘, 𝑙 ∈ K . If ∅≠𝐸 ⊂ 𝑋, then the smallest subspace of 𝑋 containing 𝐸 is

7
 {
𝑠𝑝𝑎𝑛⁡𝐸 = 𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛: 𝑥1, …, 𝑥𝑛 ∈ 𝐸 , 𝑘1, …, 𝑘𝑛 ∈ 𝐾 }
It is called the span of 𝐸. If span 𝐸 = 𝑋, we say that 𝐸 spans 𝑋. A subset 𝐸 of 𝑋 is said to be
linearly independent if for all 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈K , the equation
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0implies that 𝑘1 = ⋯ = 𝑘𝑛 = 0. It is called linearly dependent if it is not
linearly independent, that is, if there exist 𝑥1, …, 𝑥𝑛 ∈ 𝐸 and 𝑘1, …, 𝑘𝑛 ∈ K such that
𝑘1𝑥1 + ⋯ + 𝑘𝑛𝑥𝑛 = 0, where at least one 𝑘𝑗is nonzero.

A subset E of X is called a Hamel basis or simply basis for X if span of E = X and E is

linearly independent .

3) DIMENSION

If a linear space 𝑋 has a basis consisting of a finite number of elements , then X is called finite
dimensional and the number of elements in a basis for 𝑋 is called the dimension of 𝑋, denoted as
dimX . Every basis for a finite dimensional linear space has the same (finite) number of elements
and hence the dimension is well-defined. The space {0} is said to have zero dimension. Note that
it has no basis !

If a linear space contains an infinite linearly independent subset, then it is said to be infinite
dimensional.

4)METRIC SPACE

We introduce a distance structure on a set 𝑋 and study functions on 𝑋 which are well-behaved
with respect to this structure.

A metric 𝑑 on a nonempty set 𝑋 is a function 𝑑: 𝑋 × 𝑋 →R

such that for all 𝑥, 𝑦, 𝑧 ∈ 𝑋

d(x, y) ≥0 and d(x , y) = 0 iff x=y

d(y , x) = d(x , y)

d(x , y) ≤ d(x , z) + d(z , y) .

The last condition is known as the triangle inequality. A metric space is a nonempty set 𝑋 along
with a metric on it.

8
5)CONTINUOUS FUNCTIONS

Roughly speaking, a function from a metric space to a metric space is continuous if it sends
‘nearby’ points to ‘nearby’ points. If 𝑋 and 𝑌 are metric spaces with metrics 𝑑 and 𝑒 respectively,
then a function 𝐹: 𝑋 → 𝑌 is said to be continuous at 𝑥0 ∈ 𝑋 if for every ϵ⟩ 0 , there is some
(
δ > 0 (possibly depending on ϵ and 𝑥0 ) such that 𝑒 𝐹(𝑥), 𝐹 𝑥0 ( )) < ϵ for all 𝑥 ∈ 𝑋 satisfying
( )
𝑑 𝑥, 𝑥0 < δ. Further, 𝐹 is said to be continuous on 𝑋 if it is continuous at every point of 𝑋. It is
easy to see that 𝐹 is continuous on 𝑋 if and only if the set F -1(E) is open in X whenever the set E
is open inY. Also , this happens iff F(xn) →F(x) in Y whenever xn →x in X.

6) UNIFORM CONTINUITY

We note that a continuous function 𝐹: 𝑇 → 𝑆 is, in fact, uniformly

continuous, that is, for every ϵ > 0, there exists some δ > 0
such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < ϵ whenever 𝑑 𝑡, 𝑢( )< δ. This can be seen as follows. Let 𝑡 ∈ 𝑇. By
ε
the continuity of 𝐹 at 𝑡 ∈ 𝑇, there is some δ𝑡, such that 𝑒(𝐹(𝑡), 𝐹(𝑢)) < 2 whenever
𝑑(𝑡, 𝑢) < δ𝑡.

7) FIELD

A ring is a set R together with two binary operations + and ·( which we call addition and
multiplication ) such that the following axioms are satisfied .

➢ R is an abelian group with respect to addition

➢ Multiplication is associative
➢ ∀𝑎 , 𝑏, 𝑐 ∈ 𝑅the left distributive law a(b + c) = (a·b) + (a·c) and the right distributive
law (a + b)c = (a·c) + (b·c) , hold .

A field is a commutative division ring

9
 CHAPTER 1

NORMED LINEAR SPACE

Let X be a linear space over K . A norm on X is the function || || from 𝑋 to R such that ∀
x,y ∈ X and k ∈ K ,

||𝑥||≥0 and ||𝑥|| = 0 if and only if x = 0 ,

||x + y|| ≤||x|| + ||y|| ,

||kx|| =|k| ||x|| .

A norm is the formalization and generalization to real vector spaces of the intuitive
notion of “ length” in the real world .

A normed space is a linear space with norm on it .

For x and y in X , let

d(x,y) = ||x - y||

Then d is a metric on X so that (X,d) is a metric space , thus every normed space is a metric
space

➢ Every normed linear space is a metric space . But converse may not be true .

Example :
|𝑥 − 𝑦|
d(x,y) = , ∀ x , y ∈X
1 + |𝑥 − 𝑦|

|𝑥 − 𝑦|
⇒ ||x - y|| =
1 + |𝑥 − 𝑦|

|𝑧|
⇒ ||z|| = , z = x - y ∈X
1 + |𝑧|

10
 |α𝑧|
||αz|| =
1 +|α𝑧|

|α| |𝑧|
=
1 + |α| |𝑧|

= |α| ( |𝑧|
1 + |α| |𝑧| )
≠ |α| ||z|| .

➢ Result

Let X be a normed linear space . Then ,

| ||x|| - ||y|| | ≤ ||x - y|| , ∀x , y ∈ X

Proof :

||x|| = || ( x - y ) + y|| ≤ ||x - y|| + ||y||

⇒||x|| - ||y|| ≤||x - y|| → (1)

x↔y

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||→(2)

From (1) and (2)

|||x|| - ||y||| ≤ ||x - y||

➢ Norm is a continuous function

Let xn →x , as n → ∞

11
 ⇒ xn - x →0 as n →∞

⇒ ||xn - x||→0 as n→∞

| ||xn|| - ||x|| | ≤ ||xn - x|| →0 as n→∞

⇒||xn|| - ||x|| →0 , as n →∞

⇒||x|| is continuous

➢ Norm is a uniformly continuous function

We have , || || :X→R . Let x,y ∈ X and ε> 0

Then ||x|| = ||x - y + y ||

≤||x - y|| + ||y||

⇒ ||x|| - ||y|| ≤ ||x - y||→ (1)

Interchanging x and y ,

||y|| - ||x|| ≤ ||y - x||

⇒ - ( ||x|| - ||y|| ) ≤ ||x - y||

⇒ ||x|| - ||y|| ≥ - ||x - y|| → (2)

Combining (1) and (2)

- ||x - y|| ≤ ||x|| - ||y|| ≤ ||x - y||

That is ,

| ||x ||- ||y|| | ≤ ||x - y||

Take δ = ε , then whenever ||x - y|| < δ, | ||x|| -|| y|| |< ε

12
 Therefore || || is a uniformly continuous function .

➢ Continuity of addition and scalar multiplication

To show that + : X×X →X and ·: K ×X →X are continuous functions.

Let (x,y) ∈ X×X . To show that + is continuous at (x ,y) , that is , to show

that for each (x,y) ∈ X×X if xn→x and yn→y in X , then

+(xn , yn) →+(x , y) ;

That is ,
xn + yn →x + y .

Consider
||( xn + yn) - (x + y )|| = ||xn - x + yn - y||

≤||xn - x|| + ||yn - y||

Given xn→x and yn→y , for each ϵ > 0, ∃ N1 ∋

ε
||xn - x|| < 2
∀ n ≥N1 , and ∃ N2 ∋

ε
||yn - y|| < 2
∀ n ≥N2

\
Take N = max { N1, N2}

ε ε
Then ||xn - x|| < 2
and ||yn - y|| < 2
∀n≥N

ε ε
Therefore ||(xn + yn) - (x + y)|| ≤ 2
+ 2
= ε ∀n ≥N

That is , xn + yn →x + y

Now to show that ·: K ×X→X is continuous

Let (k , x) ∈ K×X

13
 To show that if kn →k and xn →x , then knxn →kx

Since kn →k , ∀ ε > 0 ∃ N1 ∋ |kn - k| < ε ∀ n ≥ N1

2

ε
Since xn →x , ∀ ε > 0 ∃ N2 ∋ ||xn - x|| < ∀ n ≥ N2
2

Consider ||knxn - kx|| = ||knxn - kx + xnk - xnk ||

= ||xn (kn - k) + k(xn - x)||

≤ ||xn(kn - k)|| + ||k(xn - x)||

= ||xn|| |kn- k| + |k| ||xn - x||

ε ε
≤ ||xn|| 2 + |k| 2

∴ knxn→kx

➢ Examples of normed space

1) Spaces Kn (K = R or C)

For n = 1 , the absolute value of function | | is a norm on K , since ∀k ∈K

We have ,

||k|| = ||k · 1|| = |k| ||1|| , by definition .

But ||1|| is a positive scalar .

∴ ||k|| is a positive scalar multiple of the absolute value function .

∴ any norm on K is a positive scalar multiple of the absolute value

function

For n > 1 , let p≥ 1 be a real number

14
 Kn = { ( x(1) , x(2) , . . . , x(n) ) : x(i) ∈K , i = 1 , 2 , . . . , n }

For x ∈Kn , that is , x = ( x(1) , x(2) , . . . , x(n) ) , define

𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (|𝑥(1)| +. . . + |𝑥(𝑛)| )

Then || ||p is a norm on Kn

When p = 1 ,

Then , ||x||1 = |x(1)| + |x(2)| + . . . + |x(n)|

Since |x(i)| ≥0 ∀ i = 1 , 2 , . . . , n , ||x||1 ≥0

And ||x||1 = 0 ⇔ |x(1)| + . . . +|x(n)| = 0

⇔ |x(i)| = 0 ∀ i

⇔ x(i) = 0 ∀ i

⇔ x = ( x(1) , . . . , x(n) ) = 0

Now ||kx||1 = |kx(1)| + |kx(2)| + . . . + |kx(n)|

= |k| |x(1)| + . . . + |k| |x(n)|

= |k| ( |x(1)| + . . . + |x(n)| )

= |k| ||x||1

||x + y||1 = |(x + y)(1)| + . . . + |(x + y)(n)|

= |x(1) + y(1)| + . . . + |x(n) + y(n)|

≤|x(1)| + |y(1)| + . . . + |x(n)| + |y(n)|

= |x(1)| + . . . + |x(n)| + |y(1)| + . . . + |y(n)|

= ||x||1 + ||y||1

15
 Consider 1<p<∞

𝑝 𝑝 1/𝑝
Now , ||𝑥||𝑝 = ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

𝑝
Since |x(i)| ≥0 ∀i , we have ||x||p ≥0

𝑝 𝑝 1/𝑝
And ||𝑥||𝑝 = 0⇔( |𝑥(1)| +. . . + |𝑥(𝑛)| ) = 0

𝑝
⇔ |𝑥(𝑖)| = 0 ∀i

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 ∀i

⇔ x = ( x(1) , . . . , x(n) ) = 0 .
Now
𝑝 𝑝 1/𝑝
||𝑘𝑥||𝑝 = ( |𝑘𝑥(1)| +. . . + |𝑘𝑥(𝑛)| )

𝑝 𝑝 𝑝 𝑝 1/𝑝
= ( |𝑘| |𝑥(1)| +. . . + |𝑘| |𝑥(𝑛)| )

𝑝 𝑝 1/𝑝
= |𝑘| ( |𝑥(1)| +. . . + |𝑥(𝑛)| )

= |𝑘| ||𝑥||𝑝 .

𝑝 𝑝 1/𝑝
||𝑥 + 𝑦||𝑝 = ( |𝑥(1) + 𝑦(1)| +. . . + |𝑥(𝑛) + 𝑦(𝑛)| )

We have by Minkowski’s inequality ,

1/𝑝 1/𝑝 1/𝑝

( ) ( ) ( )
𝑛 𝑛 𝑛
𝑝 𝑝 𝑝
∑ |𝑥(𝑖) + 𝑦(𝑖)| ≤ ∑ |𝑥(𝑖)| + ∑ |𝑦(𝑖)|
𝑖=1 𝑖=1 𝑖=1
Then

16
 𝑝 𝑝 1/𝑝 𝑝 𝑝 1/𝑝
(
||𝑥 + 𝑦||𝑝 ≤ |𝑥(1)| +. . . + |𝑥(𝑛)| ) (
+ |𝑦(1)| +. . . + |𝑦(𝑛)| )
= ||𝑥||𝑝 + ||𝑦||𝑝

Then , for 1≤p<∞ ,|| ||𝑝 is a norm on Kn

When p = ∞, define ||𝑥||∞ = 𝑚𝑎𝑥 { |𝑥(1)| , |𝑥(2)| ,. . . , |𝑥(𝑛)| }

Then it is a norm on Kn

||𝑥||𝑝 ≥ 0 since each values |x(i)|≥0

So that

max {|x(i)| , i=1, . . . , n} ≥ 0

||𝑥||∞ = 0 ⇔𝑚𝑎𝑥 { |𝑥(𝑖)| : 𝑖 = 1,. . . , 𝑛 }= 0

⇔|x(i)| = 0 ∀i

⇔x(i) = 0 , ∀i

⇔x = 0

||𝑘𝑥||∞ = 𝑚𝑎𝑥 { |𝑘𝑥(1)| ,. . . , |𝑘𝑥(𝑛)| }

= max { |k| |x(1)| , . . . , |k| |x(n)|}

= |k| max {|x(1)| , . . . , |x(n)|}

= |k| ||x||∞

||x + y||∞ = max { |x(1) + y(1)| , . . . , |x(n) + y(n)| }

≤max { |x(1)| + |y(1)| , . . . , |x(n)| + |y(n)| }

17
 ≤max { |x(1)| , . . . , |x(n)| } + max { |y(1)| , . . . , |y(n)| }

= ||x||∞ + ||y||∞

2) Sequence space

∞
𝑝 𝑝 𝑝
Let 1≤p < ∞, 𝑙 = { x = ( x(1) , x(2) , . . . ) ; x(i) ∈K and ∑ |x(j)| <∞} , that is , 𝑙 is the
𝑗=1
𝑝
space of p-summable scalar sequences in K . For x = (x(1) , x(2) , . . . ) ∈ 𝑙 ,

let ||x||p = ( |x(1)|p + |x(2)|p + . . . )1/p . Then it is a norm on lp.

That is , || ||p is a function from lp to R .

If p = 1 , then l1 is a linear space and ||x||1 = ( |x(1)| + |x(2)| + . . . ) is a norm on l1

∞
Let p = ∞. Then 𝑙 is the linear space of all bounded scalar sequences . And ,

||x||∞ = sup { |x(j)| : j = 1, 2, 3, . . . }

∞
Then || ||∞ is a norm on 𝑙

18
 CHAPTER 2

THEOREMS ON NORMED SPACES

a) Let Y be a subspace of a normed space X , then Y and its closure 𝑌 are normed spaces with
the induced norm.
b) Let Y be a closed subspace of a normed space X , for x +Y in the quotient space X/Y, let
|||x +Y||| = inf { ||x+y|| : y ∈Y} . Then ||| ||| is a norm on X/Y , called the quotient norm.

A sequence (xn + Y) converges to x + Y in X/Y iff there is a sequence (yn) in Y , (xn+ yn)
converges to x in X.

c) Let || ||pbe a norm on the linear space Xp , j = 1,2,…. . Fix p such that 1 ≤p ≤∞

For x = (x(1) , x(2) , … , x(m)) that is the product space X =X1× X2 × …× Xm ,

1/𝑝
Let ||𝑥||𝑝 = ( ||𝑥(1)|| 𝑝
1
𝑝
+ ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚
𝑝
) , if 1≤ p <∞

||x||p = max { ||x(1)||1 , … , ||x(m)||m } , if p = ∞.

Then || ||p is a norm on X.

A sequence (xn) converges to x in X ⇔ (xn(j)) converges to x(j) in Xj ∀ j=1,2,…,m.

Proof:
a) Since X is a normed space, there is a norm on X to Y . Since Y is a subspace of X,

|| ||y: Y → R is a function. To show that || ||y is a norm on Y.

For y∈ Y , || y||Y = ||y|| , then

||y||Y ≥ 0 ( ∵||y||≥ 0 ) and ||y||Y = 0 ⇔ y = 0

||ky||Y = ||ky|| = |k| ||y|| = |k| ||y||y .

Let y1 , y2 ∈Y. Then ,

||𝑦1 + 𝑦2||𝑦 = ||𝑦1 + 𝑦2|| ≤ ||𝑦1|| + ||𝑦2|| = ||𝑦1||𝑦 + ||𝑦2||𝑦

Now the continuity of addition and scalar multiplication shows that 𝑌is a subspace of X, since if
xn →x and yn → y , xn , yn ∈ 𝑌 , then

xn + yn →x + y (by continuity of addition) and

19
 kxn →kx (by continuity of scalar Xn) .

Since 𝑌is closed , x + y ∈ 𝑌 and kx ∈ 𝑌 . Therefore 𝑌 ≤ X.

∴ norm on X induces a norm on Y and 𝑌

b) X/Y , the quotient space equals X/Y={ x + Y : x ∈X }.

|||x + y||| = inf { ||x + y|| : y ∈Y }

Claim: ||| ||| is a norm on X/Y , called quotient norm

• Let x ∈X ,

|||x + Y||| = inf { ||x + y|| : y ∈Y } ≥ 0.

∴ |||x + Y||| ≥0 .

If |||x + y||| = 0 ( 0 in X/Y is Y) , then there is a sequence (yn) in Y ∋

||x + yn || →0

⇒ x + yn →0

⇒ yn →-x

Since yn ∈Y and Y is closed

-x ∈Y ⇔ x ∈Y (∵ Y is a subspace)

⇔x + Y = Y , zero in X/Y.

• For k ∈K ,

|||k(x + Y)||| = |||kx + Y|||

= inf { ||k(x + y)|| : y ∈Y}

= inf { |k| ||x + y|| : y ∈Y}

= |k| inf { ||x + y|| : y ∈Y}

= |k| |||x + Y||| .

• Let x1 , x2 ∈X . Then

|||x1 + Y||| = inf { ||x1 + y|| : y ∈Y } . Then ∃ y1 ∈ Y ∋

ε
|||x1 + Y||| + > ||x1 + y1|| , and
2

20
 |||x2 + Y||| = inf { ||x2 + y|| : y ∈Y} , Then ∃y2 ∈Y ∋
ε
|||x2 + Y||| + 2
> ||x2 + y2|| .

||x1 + y1 + x2 + y2 || ≤ ||x1 + y1|| + ||x2 + y2||

ε ε
≤ |||x1 + Y||| + 2 + |||x2 + Y||| + 2

Let y = y1 + y2 ∈Y . Then ,

||(x1+x2) + y|| ≤ |||x1 + Y||| + |||x2 + Y||| + ℇ —(1)

Now , |||(x1 + Y) + (x2 + Y)||| = |||x1 + x2 + Y|||

=inf { ||x1 + x2 + y|| : y ∈Y }

< ||x1 + x2 + y||

≤ |||x1 + Y||| + |||x2 + Y||| + ℇ (by (1) )

since ℇ is arbitrary , we have

|||(x1 + Y) + (x2 + Y)||| ≤|||x1 + Y||| + |||x2 + Y|||

∴ ||| ||| is a norm on X/Y.

Let (xn + Y) be a sequence in X/Y . Assume that (yn) is a sequence in Y ∋ (xn + yn) converges
to x in X.
That is , (xn - x + yn) converges to 0 . —(1)
Claim: (xn + Y) converges to x + Y.
Consider
|||xn + Y - (x+Y)||| = |||(xn - x) + Y|||

= inf { ||xn - x + yn|| : y ∈Y }

≤ ||xn - x + yn|| ∀yn ∈Y .

Then by (1) , xn + Y converges to x + Y in X/Y.

Conversely assume that the sequence (xn + Y) → x + Y in X/Y.

Consider |||xn + Y - (x + Y)||| = |||xn - x + Y|||

= inf { ||xn -x + y|| : y ∈Y }

Then we can choose yn ∈Y ∋

21
 1
||xn - x + yn|| < |||(xn - x) + Y||| + 𝑛 , n=1,2,3,….

Since xn+Y →x+Y , we get

(xn - x + yn) converges to zero as n →∞

That is , (xn + yn) converges to x in X as n →∞

c) Consider 1≤ p < ∞

Given that
𝑝 𝑝 𝑝 1/𝑝
||𝑥||𝑝 = (||𝑥(1)||1 + ||𝑥(2)||2 +. . . + ||𝑥(𝑚)||𝑚)

Clearly , ||x||p ≥0 .
𝑝
Since each ||𝑥(𝑖)||𝑖 ≥ 0 .

𝑝
||x||p = 0 ⇔ |𝑥(𝑗)|𝑗 = 0 ∀ j = 1, . . . ,m

⇔ x(j) = 0 ∀ j.

⇔ x = (x(1), . . . ,x(m)) = 0
1/𝑝
( 𝑝
||kx||p = ||𝑘𝑥(1)||1 +. . . + ||𝑘𝑥(𝑚)||𝑚
𝑝
)
1/𝑝
= ( |𝑘| ||𝑥(1)|| )
𝑝 𝑝 𝑝 𝑝
1
+. . . + |𝑘| ||𝑥(𝑚)||𝑚

1/𝑝
( 𝑝
= |𝑘| ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= |𝑘| ||𝑥||𝑝 , k∈K and x∈X

1/𝑝
(
Now, ||𝑥 + 𝑦||𝑝 = ||𝑥(1) + 𝑦(1)||1 +. . . + ||𝑥(𝑚) + 𝑦(𝑚)||𝑚
𝑝 𝑝
)
(by Minkowski’s inequality)

≤
1/𝑝
( (||𝑥(1)||1
+ ||𝑦(1)||1 )𝑝
(
+. . . + ||𝑥(𝑚)||𝑚 + ||𝑦(𝑚)||𝑚 ))
𝑝

1/𝑝 1/𝑝

( ) ( )
𝑚 𝑚
𝑝 𝑝
≤ ∑ ||𝑥(𝑗)||𝑗 + ∑ ||𝑦(𝑗)||𝑗 (Minkowski’s inequality )
𝑗=1 𝑗=1

22
 1/𝑝
( 𝑝
= ||𝑥(1)||1 +. . . + ||𝑥(𝑚)||𝑚
𝑝
)
= ||x||p + ||y||p

Now suppose p =∞

||x||∞ = max { ||x(1)||1 , . . . , ||x(m)|| m }

||x||∞ ≥0 Since ||x(j)|| ≥0 , ∀j

||x||∞ = 0 ⇔ ||x(m)|| = 0 ∀m

⇔ x(m) = 0 ∀m

⇔x=0

||kx||∞ = max { ||kx(1)||1 , . . . , ||kx(m)||m }

= |k| max { ||x(1)||1 , . . . , ||x(m)||m }

= |k| ||x||∞

||x + y||∞ = max { ||x(1) + y(1)||1, . . . , ||x(m) + y(m)||m }

≤max { ||x(1)||1 + ||y(1)||1 , . . . , ||x(m)||m + ||y(m)||m }

= max { ||x(1)||1 , . . . , ||x(m)||m } + max { ||y(1)||1 , . . . , ||y(m)||m }

= ||x||∞ + ||y||∞

We now consider ,

𝑝 𝑝 1/𝑝
( )
||𝑥𝑛 − 𝑥(1)||𝑝 = ||𝑥𝑛(1) − 𝑥(1 ||1 +. . . + ||𝑥𝑛(𝑚) − 𝑥(𝑚)||𝑚)

Then

xn →x in X ⇔ ||xn - x ||p → 0
𝑝
⇔ ||xn(j) - x(j)||𝑗 → 0

⇔ xn(j) - x(j) → 0

⇔ xn(j) → x(j) in X ∀j .

23
 RIESZ LEMMA

Let 𝑋 be a normed space . 𝑌be a closed subspace of 𝑋and 𝑋 ≠ 𝑌 . Let 𝑟be a real number
such that 0 < 𝑟 < 1 . Then there exist some xr ∈X such that ||xr|| = 1 and

r<dist ( xr , Y )≤1

Proof :
We have ,

dist (x , Y) = inf { d(x , y) : y∈Y}

= inf { ||x - y|| : y ∈Y}

Since Y ≠X , consider x ∈X ∋ x∉Y.

If dist(x , Y) = 0 , then ||x - y|| = 0 ⇒ x∈𝑌= Y ( ∵ Y is closed )

Therefore ,

dist (x , Y) ≠0

That is ,
dist (x , Y) > 0
1
Since 0 < r < 1 , >1
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
⇒ > dist (x , Y)
𝑟

𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)
That is , is not a lower bound of { ||x - y|| : y ∈Y }
𝑟

𝑑𝑖𝑠𝑡(𝑥 , 𝑌)
Then ∃ y0 ∈Y ∋ ||x - y0|| <
𝑟
→(1)
𝑥 − 𝑦0
Let xr = . Then xr ∈X
||𝑥 − 𝑦0||

( ∵y0 ∈Y , x∉Y ⇒x - y0 ∈X and ||x - y0|| ≠0 )

24
 𝑥 − 𝑦0 ||𝑥 − 𝑦0||
Then ||xr|| = || || = =1
||𝑥 − 𝑦 ||0
||𝑥 − 𝑦0||

Now to prove r < dist(𝑥𝑟, 𝑌)≤1

We have dist(xr , Y) = inf { ||xr - y|| : y∈Y }

≤||xr - y|| ∀y∈Y

In particular, 0∈ 𝑌, so that dist(xr , Y) ≤||xr - 0|| = 1

That is ,

dist (xr , Y) ≤1

Now ,
𝑥 − 𝑦0
dist (xr , Y) = dist ( ,Y )
||𝑥−𝑦0||

1
= dist ( x - y0 , Y)
||𝑥−𝑦0||

1
= inf { ||x - y0 - y|| : y∈ 𝑌}
||𝑥−𝑦0||

1
= inf {||x - (y0+ y)|| : y0 + y∈ 𝑌}
||𝑥−𝑦0||

1
= dist (x , Y)
||𝑥−𝑦0||

𝑟
> dist (x , Y) by (1)
𝑑𝑖𝑠𝑡 (𝑥 , 𝑌)

⇒ dist (xr , Y) > r

That is ,

r < dist (xr , Y) ≤ 1

25
 CONCLUSION

This project discusses the concept of normed linear space that is fundamental to
functional analysis . A normed linear space is a vector space over a real or complex
numbers ,on which the norm is defined . A norm is a formalization and generalization to
real vector spaces of the intuitive notion of “length” in real world
In this project , the concept of a norm on a linear space is introduced and thus
illustrated . It mostly includes the properties of normed linear spaces and different proofs
related to the topic.

26
 BIBLIOGRAPHY

➢ Balmohan V Limaye , Functional Analysis

➢ S.H Friedberg , Arnold J.Insel and Lawrence E.Spence , Linear Algebra , 2 nd Edition, PH
Inc .
➢ S. Narayanan and Mittai , A Text Book of Matrices , Revised Edition, S. Chand

27
PLANARITY IN GRAPH THEORY

Project report submitted to

The Kannur University
for the award of the degree
of
Bachelor of Science
by
SREESHMA SHIBU
DB18CMSR10
Under the guidance of
Mrs. PRIJA V

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu
June 2021
 CERTIFICATE

Certified that this project ‘Planar Graph’ is a bona fide project of Sreeshma
Shibu carried out the project work under my supervision.

Mrs.
Mrs. Riya
Riya Baby
Baby Mrs. Prija V
Head
Head of
of Department
Department Supervisor

Department of Mathematics
Don Bosco Arts and Science College
Angadikkadavu

I
 DECLARATION

I Sreeshma Shibu hereby declare that the project ‘Planar Graph’ is an original
record of studies and bona fide project carried out by me during the period of
2018 – 2021 under the guidance of Mrs. Prija V, Department of Mathematics,
Don Bosco Arts and Science College, Angadikkadavu and has not submitted by
me elsewhere for the award of my degree, diploma, title, or recognition, before.

Sreeshma Shibu

DBI8CMSR10

Department of Mathematics

Don Bosco Arts and Science College,

Angadikkadavu

II
 ACKNOWLEDGEMENT

Introduction is the proper place to begin. But first I bow my head before
the Almighty who is always with me. Also, I must express my deepest gratitude
to people along the way.

No words can adequately express the sense of gratitude; still, I try to

express my heartfelt thanks through words. At the outset, I am deeply indebted to
my project supervisor Mrs. Prija V, Department of Mathematics, Don Bosco Arts
and Science College, Angadikkadavu, for the invaluable guidance, loving
encouragement, and meticulous care towards me throughout my career. I express
my deep sense of gratitude to all the faculty members of the Department of
Mathematics, Don Bosco Arts and Science College, Angadikkadavu.

I can never forget the support and encouragement rendered by the principal
and the staff of Don Bosco Arts and Science College, Angadikkadavu.

I could not name many who sincerely supported and helped for the
successful completion of this project. It is my pleasure and duty to thank each and
every one of them who walked with me.

My greatest debt is always, to God Almighty.

III
 CONTENTS

Sl No Title Page No

1 Introduction 1

2 Chapter 1 - Basic Concepts 2-9

3 Chapter 2 – Planar Graph 10 - 17

4 Chapter 3 – Euler’s Formula 18 – 19

5 Chapter 4 – Dual of a Plane Graph 20

6 Conclusion 21

7 Bibliography 22

IV
 INTRODUCTION

In recent years, Graph Theory has established itself as an important

mathematical tool in a wide variety of subjects, ranging from Operational
Research and Chemistry to Genetics and Linguistics, and from Electrical
Engineering and Geography to Sociology and Architecture. At the same time, it
has also emerged as a worthwhile mathematical discipline in its own right.

A great mathematician, Euler become the Father of Graph Theory, when

in 1736, he solved a famous unsolved problem of his days, called Konigsberg
Bridge Problem. This is today, called as the First Problem of the Graph theory.
This problem leads to the concept of the planar graph as well as Eulerian Graphs,
while planar graphs were introduced for practical reasons, they pose many
remarkable mathematical properties. In 1936, the psychologist Lewin used planar
graphs to represent the life space of an individual.

1
Chapter 1

BASIC CONCEPTS

Graph

A graph is an ordered triple 𝐺 = {𝑉(𝐺), 𝐸(𝐺), 𝐼𝐺 } where 𝑉(𝐺) is a non-

empty set, 𝐸(𝐺) is a set disjoint from 𝑉(𝐺) and 𝐼(𝐺) is an incidence map that
associates each element of 𝐸(𝐺) and unordered pair of elements of 𝑉(𝐺). The
elements of 𝑉(𝐺) are called vertices (or nodes or points) of 𝐺 and the element
of 𝐸(𝐺) are called edges or lines of 𝐺.

Example:

Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 }

𝐸(𝐺) = {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 }
𝐼𝐺 (𝑒1 ) = {𝑣1 , 𝑣2 } or {𝑣2 , 𝑣1 }
I𝐺 (𝑒2 ) = {𝑣2 , 𝑣3 } or {𝑣3 , 𝑣2 }
𝐼𝐺 (𝑒3 ) = {𝑣3 , 𝑣4 } or {𝑣4 , 𝑣3 }
𝐼𝐺 (𝑒4 ) = {𝑣4 , 𝑣1 } or {𝑣1 , 𝑣4 }

Multiple edges

A set of two or more edges of a graph 𝐺 is called multiple edges or parallel

edges if they have the same end vertices.

2
Loop

An edge for which the two end vertices are same is called a loop.

Here {𝑒1 , 𝑒2 , 𝑒3 , 𝑒4 } form the parallel edges.

𝑒7 is the Loop.
Simple Graph

A graph is simple if it has no loops and no multiple edges.

Finite & Infinite Graphs

A graph is called finite if both 𝑉(𝐺) & 𝐸(𝐺) are finite. A graph that is not
finite is called infinite graph.

Adjacent Vertices

Two vertices 𝑢 and 𝑣 are said to be adjacent vertices if and only if there is
an edge with 𝑢 and 𝑣 as its end vertices.

3
Adjacent Edges

Two distinct edges are said to be adjacent edges if and only if they have a
continuous end vertex.

Complete Graph

A simple graph 𝐺 is said to be complete if every pair of distinct vertices of

𝐺 are adjacent in 𝐺. A complete graph with n vertices is denoted by 𝐾𝑛 .

Bipartite Graph

A graph is bipartite if its vertex set can be partitioned into two non-empty
subsets 𝑋 and 𝑌 such that each edge of 𝐺 has one end in 𝑋 and the other in 𝑌. The
pair (𝑋, 𝑌) is called a bipartition of the bipartite graph 𝐺. The bipartite graph 𝐺
with bipartition (𝑋, 𝑌) denoted by 𝐺 (𝑋, 𝑌).

4
Here 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , 𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }
The Bipartition is
𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 , 𝑣6 , 𝑣7 }

Complete Bipartite Graph

A simple bipartite graph 𝐺 (𝑋, 𝑌) is complete if each vertex 𝑋 is adjacent

to all the vertices of 𝑌.

Here 𝑋 = {𝑣1 , 𝑣2 , 𝑣3 }
𝑌 = {𝑣4 , 𝑣5 }

Subgraph

A graph 𝐻 is called subgraph of 𝐺 if 𝑉(𝐻) ⊆ 𝑉(𝐺), 𝐸(𝐻) ⊆ 𝐸(𝐺) and IH

is the restriction of 𝐼𝐺 to 𝐸(𝐻) [ie, 𝐼𝐻 (𝑒) = 𝐼𝐺 (𝑒) whenever 𝑒 ∈ 𝐸(𝐻).

5
Degrees of Vertices

The number of edges incident with vertex 𝑉 is called degree of a vertex

or valency of a vertex and it is denoted by 𝑑(𝑣).

Isomorphism of Graph

A graph isomorphism from a graph 𝐺 to a graph 𝐻 is a pair (𝜙, 𝜃), where

𝜙 ∶ 𝑉(𝐺) → 𝑉(𝐻) and 𝜃 ∶ 𝐸(𝐺) → 𝐸(𝐻) are bijection with a property that
𝐼𝐺 (𝑒) = {𝑢, 𝑣} and 𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}.

Walk

A walk in a graph 𝐺 is an alternative sequence 𝑊=

𝑣0 𝑣1 𝑒1 𝑣2 𝑒2 … 𝑣𝑛 𝑒𝑛 vertices and edges, beginning and ending with vertices where
𝑣0 is the origin and 𝑣𝑛 is the terminus of 𝑊.

𝑊 = 𝑣6 𝑒8 𝑣1 𝑒1 𝑣2 𝑒2 𝑣3 𝑒3 𝑣2 𝑒1 𝑣1

6
Closed Walk

A walk to begin and ends at the same vertex is called a closed walk. That
is, the walk 𝑊 is closed if 𝑣0 = 𝑣𝑛 .

Open Walk

If the origin of the walk and terminus of the walk are different vertices,
then it is called an open walk.

Trail

A walk is called a trail if all the edges in the walk are distinct.

Path

A walk is called a path if all the vertices are distinct.

Example:

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒6 𝑣1 → A trail

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 → A path

𝑣0 𝑒1 𝑣1 𝑒2 𝑣2 𝑒3 𝑣3 𝑒5 𝑣1 → A trail, but not a path

Euler’s Theorem

The sum of the degrees of the vertices of a graph is equal to the twice the
number of edges.

ie: ∑𝑛𝑖=1 𝑑(𝑣𝑖 ) = 2𝑚

7
Isomorphic Graph

𝐼𝐻 (𝜃(𝑒)) = {𝜙(𝑢), 𝜙(𝑣)}

A graph 𝐺1 = (𝑉1 , 𝐸1 ) is said to be isomorphic to graph 𝐺2 = (𝑉2 , 𝐸2 ) if

there is a one-to-one correspondence between the edge sets 𝐸1 and 𝐸2 in such a
way that if 𝑒1 is an edge with end vertices 𝑢1 and 𝑣1 in 𝐺1 then the corresponding
edge 𝑒2 in 𝐺2 has its end vertices 𝑢2 and 𝑣2 in 𝐺2 . This correspondence is called
a graph isomorphism.

Example:

𝐺=

𝐻=

ie: G and H are isomorphic.

Components

A connected component of a graph is a maximal connected subgraph. The

term is also used for maximal subgraph or subset of a graph 's vertices that have
some higher order of connectivity, including bi-connected components, tri-
connected components and strongly connected components.

Tree

A connected graph without cycles is called a tree.

Vertex Cut

Let 𝐺 be a connected graph. The set 𝑉՛ subset of 𝑉(𝐺) is called a Vertex

cut of 𝐺, if 𝐺 − 𝑉՛ is a disconnected graph.

8
Cut Vertex

If 𝑉՛ = {𝑣} is a Vertex cut of the connected Graph 𝐺, then the vertex 𝒗 is

called a Cut vertex.

Edge Cut

Let 𝐺 be a non-trivial connected graph with vertex set 𝑉 and let 𝑆 be a non-
empty subset of 𝑉 and 𝑆̅ = 𝑉 − 𝑆. Let 𝐸՛ = [𝑆, 𝑆̅] denote the set of all edges of 𝐺
that have one end vertex is 𝑆 and the other is 𝑆̅. Then 𝐺 − 𝐸՛ is a disconnected
graph and 𝐸՛ = [𝑆, 𝑆̅] is called an edge cut of 𝐺.

Cut Edge

If 𝐸՛ = {𝑒} is an edge cut of 𝐺 then 𝑒 is called a cut edge of 𝐺.

Block

A block is a Connected graph without any cut vertices.

Eg:

Graph 𝐺 Blocks of 𝐺

9
Chapter 2

PLANAR GRAPHS

Plane Graph

A plane graph is a graph drawn in the plane, such a way that any pair of
edges meet only at their end vertices.

Example:

Planar Graph

A planar graph is a graph which is isomorphic to a plane graph, ie: it can

be drawn as a plane graph.

A plane graph is a graph that can be drawn in the plane without any edge crossing.

10
Example of Planar graph:

Planar Representation

The pictorial representation of a planar graph as a plane graph is called a

planar representation.

Eg: Is Q3 shown below, planar?

The graph Q3

Planar representation of Q3 is:

11
Jordan Curve

A Jordan Curve in the plane is a continuous non-self-intersecting curve

where Origin and Terminals coincide.

Example:

Jordan Curves

Non-Jordan Curves

Remark

If J is a Jordan Curve in the plane, then the part of the plane enclosed by J
is called interior of J and is denoted by ‘int J’. We exclude from ‘int J’ the points
actually lying on J. Similarly, the part of the plane lying outside J is called the
exterior of J and is denoted by ‘ext J’.

Example:

Arc connecting point 𝑥 in int J with point 𝑦 in ext J.

12
Theorem

Let J be a Jordan Curve, if 𝑥 is a point in int J and 𝑦 is a point in ext J then

any line joining 𝑥 to 𝑦 must meet J at some point, ie: must cross J. this is called
Jordan Curve Theorem.

Boundary

The set of edges that bound a region is called its boundary.

Definition

A graph which is not planar is known as non-planar graph or a graph that

cannot be drawn in the plane without any edge crossing is known as non-planar
graph.

Theorem

K5 is nonplanar:

Every drawing of the complex graph K5 in the plane (or sphere) contains
at least one edge crossing.

Proof:

Label the vertices 0, 1, 2, 3, 4. By the Jordan Curve theorem any drawing of the
cycle (1, 2, 3, 4, 1) separates the plane into two regions. Consider the region with

13
vertex 0 in its interior as the ‘inside’ of the circle. By the Jordan Curve theorem,
the edges joining vertex 0 to each of its vertices 1, 2, 3 and 4 must also lie entirely
inside the cycle, as illustrated below.

Drawing most of the K5 in the plane

Moreover, each of the 3-cycles {0, 1, 2, 0}, {0, 2, 3, 0}, {0, 3, 4, 0} and {0, 4, 1, 0}
also separates the plane and hence the edges (2, 4) must also lie to the exterior of
the cycle {1, 2, 3, 4} as shown. It follows that the cycle formed by edges (2, 4),
(4, 0) and (0, 2) separates the vertices 1 and 3, again by Jordan Curve theorem.
Thus, it is impossible to draw edge (1, 3) without crossing an edge of that cycle.
So, it is proven that the drawing of the K5 in the plane contains at least one edge-
crossing.

Theorem

K33 is nonplanar:

Every drawing of the complete bipartite graph K33 in the plane (or sphere)
contains at least one edge crossing.

Proof:

Label the vertices of one partite set 0, 2, 4 and of the order 1, 3, 5. By the
Jordan Curve theorem, cycle {2, 3, 4, 5, 2} separates the plane into two regions,

14
and as in the previous proof (K5), we regard the region containing the vertex 0 as
the ‘inside’ of the cycle. By the Jordan Curve theorem, the edges joining vertex
0 to each of the vertices 3 and 5 lie entirely inside that cycle, and each of the cycle
{0, 3, 2, 5, 0} and {0, 3. 4, 5, 0} separates the plane, as illustrated below.

Drawing most of the K33 in the plane

Thus, there are 3 regions: the exterior of cycles {2, 3, 4, 5, 2} and the inside
of each of the other two cycles. It follows that no matter which region contains
vertex 1, there must be some even numbered vertex that is not in that region, and
hence the edge from vertex 1 to that even-numbered vertex would have to cross
some cycle edge.

Corollary

Subgraph of a planar graph is planar.

Definition

A plane graph partitions the plane into number of regions called faces.

Let G be plane graph. If x is a point on the plane which is not in G, ie: 𝑥 is not a
vertex of G or a point on any edge of G, then we define the faces of G containing
𝑥 to be the set of all points on the plane which can be reached from 𝑥 by a line
which does not cross any edge of G or go through any vertex of G.

15
The number of faces of a plane graph G denoted by 𝑓(𝑎) or simply 𝑓.

Each plane graph has exactly one unbounded face called the exterior face.

Here 𝑓(𝐺) = 4

Degree of faces

The degree 𝑑(𝑓) of a face 𝑓 is the number of edges with which it is

incident, that is the number of edges in the boundary of a face.

Cut edge being counted twice.

Eg:

𝑑(𝑓2 ) = 3
𝑑(𝑓1 ) = 4
𝑑(𝑓3 ) = 3

Theorem

A graph is planar if and only if each of its blocks is planar.

Proof:

If G is planar, then each of its blocks is planar since a subgraph of planar

graph is planar.

Conversely, suppose that each block of G is planar. We now use induction

on the number of blocks of G to prove the result. Without loss of generality, we
16
assume that G is connected. If G has only one block, then G itself is a block, and
hence G is planar.

Now suppose G has k planar blocks and that the result has been proved for
all connected graph having (k-1) planar blocks. Choose any end block B0 of G
and delete from G all the vertices of B0 except the unique cut vertex, say 𝑣0 of G
in B0. The resulting connected graph G` of G contains (k-1) planar blocks. Hence,
by the induction hypothesis G` is planar. Let G~` be plane embedded of G` such
that 𝑣0 belongs to the boundary of unbounded face, say 𝑓 `. Let B0~ be a plane
embedding of B0 in 𝑓 `, so that 𝑣0 is in the exterior face of B0~. Then G~` and B0~
is a plane embedding of G.

17
Chapter 3

EULER’S FORMULA

Theorems

Euler Formula:

For a connected plain graph 𝐺, 𝑛 − 𝑚 + 𝑓 = 2 where 𝑛, 𝑚, and 𝑓 denote

the number of vertices, edges and faces of 𝐺 respectively.

Proof:

We apply the induction on 𝑓.

If 𝑓 = 1 the 𝐺 is a tree and 𝑚 = 𝑛 − 1.

Hence 𝑛 − 𝑚 + 𝑓 = 2 and suppose that 𝐺 has 𝑓 faces.

Since 𝑓 ≥ 2, 𝐺 is not a tree and hence contains a cycle 𝐶. Let 𝑒 be an edge of 𝐶.

Then 𝑒 belongs to exactly 2 faces, say 𝑓1 and 𝑓2 and the deletion of 𝑒 from 𝐺
results in the formation of a single face from 𝑓1 and 𝑓2 . Also, since 𝑒 is not a cut
edge of 𝐺. 𝐺 − 𝑒 is connected.

Further the number of faces of 𝐺 − 𝑒 is 𝑓 − 1, number of edges in 𝐺 − 𝑒 is 𝑚 −

1 and number of vertices in 𝐺 − 𝑒 is 𝑛. So, applying induction to 𝐺 − 𝑒, we get
𝑛 − (𝑚 − 1) + (𝑓 − 1) = 2 and this implies that 𝑛 − 𝑚 + 𝑓 = 2. This
completes the proof of theorem.

Corollary 1

All plane embedding of a planar graph have the same number of faces.

18
Proof:

Since 𝑓 = 𝑚 − 𝑛 + 2 the number of faces depends only on 𝑛 and 𝑚 and not on

the particular embedding.

Corollary 2

If 𝐺 is a simple planar graph with at least 3 vertices, then 𝑚 ≤ 3𝑛 − 6.

Proof:

Without the generality we can assume that 𝐺 is a simple connected plane graph.
Since 𝐺 is simple and 𝑛 ≥ 3, each face of 𝐺 has degree at least 3. Hence if 𝑓
denote the set of faces of 𝐺 ∑𝑓𝜖𝐹 𝑑(𝑓) ≥ 3𝑓. But ∑𝑓𝜖𝐹 𝑑(𝑓) = 2𝑚.

2𝑚
Consequently 2𝑚 ≥ 3𝑓 so that 𝑓 ≤ .
3

2𝑚 2m
By the Euler formula 𝑚 = 𝑛 + 𝑓 − 2 now 𝑓 ≤ implies m ≤ n + ( ) − 2.
3 3

This gives. 𝑚 ≤ 3𝑛 − 6.

19
Chapter 4

DUAL OF A PLANE GRAPH

Definition

Let G be a plane graph. One can form out of G a new graph H in the
following way corresponding to each face f(g), take the vertex f* and
corresponding to each edge e(g), take an edge e*. Then edge e* joins vertices f*
and g* in H iff edge e is common to the boundaries of faces f and g in G. The
graph H is then called dual of G.

Example:

Plane graph and its Dual

20
 CONCLUSION

In this project we discussed the topic planar graph in graph theory.

We discussed about Euler formula and verified that some graphs are planar, and
some are non-planar. A related important property of planar graphs, maps and
triangulations is that they can be enumerated very nicely.

We also discussed about duality of a graph.in mathematical discipline of graph

theory, the dual graph of a plane graph G is a graph that has a vertex of each face
of G .it has many applications in mathematical and computational study.

In fact, graph theory is being used in our so many routine activities. For eg; using
GPS or google maps to determine a route based on used settings.

21
 BIBLIOGRAPHY

1) R. Balakrishnan and K. Ranganathan, A textbook of Graph Theory,

University Text

2) J. A. Bondy and U. S. R. Murty, Graph Theory with Great Britain: The

Macmillian Press Ltd. 1976

22