Lecture Notes in Discrete Mathematics

Marcel B. Finan
Arkansas Tech University
All
c Rights Reserved
22 RECURSION 141

Review Problems
Problem 22.1
Find the first four terms of the following recursively defined sequence:

v1 = 1, v2 = 2
vn = vn−1 + vn−2 + 1, n ≥ 3
Problem 22.2
Prove each of the following for the Fibonacci sequence:
a. Fk2 − Fk−1
2
= Fk Fk+1 − Fk+1 Fk−1 , k ≥ 1.
2
b. Fk+1 − Fk2 − Fk−1
2
= 2Fk Fk−1 , k ≥ 1.
2 2
c. Fk+1 − Fk = Fk−1 Fk+2 , k ≤ 1.
2
d. Fn+2 Fn − Fn+1 = (−1)n for all n ≥ 0.

Problem 22.3
Find limn→∞ FFn+1
n
where F0 , F1 , F2 , · · · is the Fibonacci sequence. (Assume
that the limit exists.)

Problem 22.4
Define x0 , x1 , x2 , · · · as follows:
p
xn = 2 + xn−1 , x0 = 0.
Find limn→∞ xn .

Problem 22.5
a. Make a list of all bit strings of lengths zero, one, two, three, and four that
do not contain the pattern 111.
b. For each n ≥ 0 let dn = the number of bit strings of length n that do not
contain the bit pattern 111. Find d0 , d1 , d2 , d3 , and d4 .
c. Find a recurrence relation for d0 , d1 , d2 , · · ·
d. Use the results of (b) of (c) to find the number of bit strings of length five
that do not contain the pattern 111.

Problem 22.6
Find a formula for each of the following sums:

a. 1 + 2 + · · · + (n − 1), n ≥ 2.
b. 3 + 2 + 4 + 6 + 8 + · · · + 2n, n ≥ 1.
c. 3 · 1 + 3 · 2 + 3 · 3 + · · · 3 · n, n ≥ 1.
142 RELATIONS AND FUNCTIONS

Problem 22.7
Find a formula for each of the following sums:

a. 1 + 2 + 22 + · · · + 2n−1 , n ≥ 1.
b. 3n−1 + 3n−2 + · · · + 32 + 3 + 1, n ≥ 1.
c. 2n + 3 · 2n−2 + 3 · 2n−3 + · · · + 3 · 22 + 3 · 2 + 3, n ≥ 1.
d. 2n − 2n−1 + 2n−2 − 2n−3 + · · · + (−1)n−1 · 2 + (−1)n , n ≥ 1.
Problem 22.8
Use iteration to guess a formula for the following recusively defined sequence
and then use mathematical induction to prove the validity of your formula:
c1 = 1, cn = 3cn−1 + 1, for all n ≥ 2.
Problem 22.9
Use iteration to guess a formula for the following recusively defined sequence
and then use mathematical induction to prove the validity of your formula:
w0 = 1, wn = 2n − wn−1 , for all n ≥ 2.
Problem 22.10
Determine whether the recursively defined sequence: a1 = 0 and an = 2an−1 +
n − 1 satisfies the explicit formula an = (n − 1)2 , n ≥ 1.
Problem 22.11
Which of the following are second-order homogeneous recurrence relations
with constant coefficients?

a. an = 2an−1 − 5an−2 .
b. bn = nbn−1 + bn−2 .
c. cn = 3cn−1 · c2n−2 .
d. dn = 3dn−1 + dn−2 .
e. rn = rn−1 − rn−2 − 2.
f. sn = 10sn−2 .
Problem 22.12
Let a0 , a1 , a2 , · · · be the sequence defined by the explicit formula
an = C · 2n + D, n ≥ 0
where C and D are real numbers.
a. Find C and D so that a0 = 1 and a1 = 3. What is a2 in this case?
b. Find C and D so that a0 = 0 and a1 = 2. What is a2 in this case?
22 RECURSION 143

Problem 22.13

Let a0 , a1 , a2 , · · · be the sequence defined by the explicit formula

an = C · 2n + D, n ≥ 0

where C and D are real numbers. Show that for any choice of C and D,

an = 3an−1 − 2an−2 , n ≥ 2.

Problem 22.14
Let a0 , a1 , a2 , · · · be the sequence defined by the explicit formula

a0 = 1, a1 = 2
an = 2an−1 + 3an−2 , n ≥ 2

Find an explicit formula for the sequence.

Problem 22.15
Let a0 , a1 , a2 , · · · be the sequence defined by the explicit formula

a0 = 1, a1 = 4
an = 2an−1 − an−2 , n ≥ 2

Find an explicit formula for the sequence.

Problem 22.16
The triangle inequality for absolute value states that for all real numbers a
and b, |a+b| ≤ |a|+|b|. Use the recursive definition of summation, the triangle
inequality, the definition of absolute value, and mathematical induction to
prove that for all positive integers n, if a1 , a2 , · · · , an are real numbers then
n
X n
X
| ak | ≤ |ak |.
k=1 k=1

Problem 22.17
Use the recursive definition of union and intersection to prove the following
general distributive law: For all positive integers n, if A and B1 , B2 , · · · , Bn
are sets then
A ∩ (∪nk=1 Bk ) = ∪nk=1 (A ∩ Bk ).
144 RELATIONS AND FUNCTIONS

Problem 22.18
Use mathematical induction to prove the following generalized De Morgan’s
law.
(∩ni=1 Ai )c = ∪ni=1 Aci
Exercise 367
Show that the relation F : IN → Z given by the rule

 1 if n = 1.
n
F (n) = F(2) if n is even
1 − F (5n − 9) if n is odd and n > 1


does not define a function.

23 PROJECT VII: APPLICATIONS TO RELATIONS 145

23 Project VII: Applications to Relations

Part I: Relational Database

The ”bi” in binary relation R refers to the fact that R is a subset of the
cartesian product of two sets. Let A1 , A2 , · · · , An be given sets. If R is a
subset of A1 × A2 × · · · × An then we call R an n-ary relation. An n-ary
relation can be represented by a table or a set of ordered n-tuples.

Example 23.1

ID# Name Position Age

22012 Johnsonbaugh c 22
93831 Glover of 24
58199 Battey p 18
84341 Cage c 30
01180 Homer lb 37
26710 Score p 22
61049 Johnsonbaugh of 30
39826 Singleton 2b 31

A database is a collection of records that are manipulated by a computer.

Database management systems are programs that help users access the
information in databases. The relational database model is based on the
concept of an n-ary relation.
When an n-ary relation is represented by a table then the columns in this ta-
ble are called attributes. In the above table, the attributes are ID Number,
Name, Position, and Age. A single attribute or a combination of attributes
for a relation is called a key if the values of the attributes uniquely define
an n-tuple. For example, in the above table, we can take the attribute ID
Number as a key since every person has a unique identification number. The
attribute Name is not a key because different persons can have the same
name. For the same reason, we cannot take the attribute Age as a key. A
database management system responds to queries. A query is a request for
information from the database. For example, ”Find all persons that are 22
years old”.
146 RELATIONS AND FUNCTIONS

Problem 23.1
a. Express the above 4-ary as a set of 4-tuples.
b. Answer the query: PLAYER[Name]
c. Answer the query: PLAYER[Name, Position]
Part II: Representing a Relation by a Matrix.

Let A be a set with n elements and R be a binary relation on A. Define

the n × n matrix M (R) = (mij ) as follows:

1 if (ai , bj ) ∈ R
mij =
0 if (ai , bj ) 6∈ R

If the numbers on the main diagonal of M (R) are all equal to 1 then R is
reflexive. If M (R)T = M (R), where M (R)T is the transpose of M (R), then
the relation R is symmetric. If mij = 0 or mji = 0 for i 6= j then R is
antisymmetric.
Problem 23.2
Let A = {1, 2, 3} and R = {(1, 1), (1, 2), (2, 2), (2, 3), (3, 1), (3, 2), (3, 3)}.
Find M (R) and use it to determine if the relation R is reflexive, symmetric
or antisymmetric.
Part III: Cryptology
An important application to congruences is cryptology, which is the study of
secret messages.

(a) The process of making a message secret is called encryption. This

process consists of assigning the alphabet A, B, C, · · · , Z by the integers
0, 1, 2, · · · , 25. Then the encrypted version of the message, the letter rep-
resented by p is replaced with the letter represented by the remainder of the
division of (p + 3) by 26.
Problem 23.3
What is the encrypted message produced from the message ”MEET YOU
IN THE PARK”?
(b) Decryption is the process of determining the original message. In this
case the letter represented by p is replaced by the letter represented by the
remainder of the division of (p − 3) by 26.
23 PROJECT VII: APPLICATIONS TO RELATIONS 147

Problem 23.4
What is the message produced from the encrypted message ”PHHW BRX
LQ WKH SDUN”?
148 RELATIONS AND FUNCTIONS

24 Project VIII: Well-Ordered Sets and Lat-

tices
Let [A, ≤] be a poset. Let B ⊆ A. An element b ∈ B is called a least
element of B if and only if b ≤ x for all x ∈ B. If x ≥ b for all x ∈ B then
we call b the greatest element of B.
Problem 24.1
Consider the set IN with the inequality relation ≤ . Let B = {2, 4, 5, 6, 7, 8, 9}.
What is the least element of B? What is the greatest element of B?
A poset [A, ≤] is said to be well-ordered if and only if ≤ is a total order
and every subset of A has a least element.
Problem 24.2
a. Show that (IN, ≤) is well-ordered.
b. Show that (ZZ, ≤) is not well-ordered.
An element a ∈ A is called a lower bound of B if a ≤ x for all x ∈ B. We
call a ∈ A an upper bound of B if x ≤ a for all x ∈ B. Note that a lower
bound or an upper bound is not unique.
Problem 24.3
Consider the poset [IN, ≤]. Let B = {2, 4, 8, 10}. Find a lower bound of B as
well as an upper bound.
The greatest element of the set of lower bounds of B is called the greatest
lower bound, in symbol g.l.b(B). The least upper bound of the set of upper
bounds of B is called the least upper bound, in symbol l.u.b(B).
Problem 24.4
Consider the poset [IR, ≤] and B = (−1, 1). Find g.l.b(B) and l.u.b(B).
A lattice is a poset [A, ≤] such that every pair of elements in A have a l.u.b
and g.l.b in A.
Problem 24.5
Show that [IR, ≤] is a lattice.
Problem 24.6
Let A = {2, 3, 4, 9, 12, 18} and R be the binary relation ”divides” on A. Show
that [A, R] is not a lattice.
25 PROJECT IX: THE PIGEONHOLE PRINCIPLE 149

25 Project IX: The Pigeonhole Principle

The Pigeonhole principle asserts that if n pigeons fly into k holes with
n > k then some of the pigeonholes contain at least two pigeons. The
reason this statement is true can be seen by arguing by contradiction. If the
conclusion is false, each pigeonhole contains at most one pigeon and, in this
case, we can account for at most k pigeons. Since there are more pigeons
than holes, we have a contradiction.

Problem 25.1
Ten persons have first names George, William, and Laura and last names
Bush, Perry, and Gramm. Show that at least two persons have the same
first and last names.

A mathematical way to formulate the pigeonhole principle is given by the

following exercise

Problem 25.2
Let S be a finite set and {A1 , A2 , · · · , An } be a partition of S. Use the method
of contradiction to show that there is an index 1 ≤ i ≤ n such that |Ai | ≥ |S| n
.

One can uses the previous exercise to solve the following exercise.

Problem 25.3
Let S and T be two finite sets such that |S| > k|T | where k is a positive
integer. Show that for any function f : S → T there is a t ∈ T such that the
set {s ∈ S : f (s) = t} has more than k elements.
Hint: Show that the family At = {s ∈ S : f (s) = t}, where t ∈ T, partitions
S into n sets with n ≤ |T |. Then apply the previous exercise.

As a consequence of the above exercise we have

Problem 25.4
If S and T are finite sets such that |S| > |T | then any function f : S → T is
not one-to-one.
150 RELATIONS AND FUNCTIONS

26 Project X: Countable Sets

We say that two sets have the same cardinality if and only if there is a
bijective function between them. A set A is called countably infinite if
and only if A has the same cardinality as the set IN∗ of positive integers. A
set A is called countable if it is either finite or countably infinite. A set
that is not countable is said to be uncountable. Examples of uncountable
sets are IR and the intervals in IR..
The purpose of this project is to look at some countably infinite sets.

Problem 26.1
Show that the function f : IN∗ → IN given by f (n) = n − 1 is a bijective
function. Thus, IN is countably infinite.

Problem 26.2
Show that the function f : IN∗ → Z defined by

n

2
if n is even
f (n) = 1−n
2
if n is odd

is bijective. Hence, Z is countably infinite.

Problem 26.3
Show that the function f : Z → 2Z Z defined by f (n) = 2n is a bijective
function. Hence, the set of even integers is countably infinite.

Problem 26.4
Show that the set of rational numbers Q
I is countably infinite. (Hint:
26 PROJECT X: COUNTABLE SETS 151

)
152 RELATIONS AND FUNCTIONS

27 Project XI: Finite-State Automaton

A finite-state machine can be looked at as a mathematical model that can
accept input, store and process information and produce output. Examples
include digital computers, compilers, vending machines, coin changers, tele-
phones, and elevators.
This model has an input/output unit, and, consequently, has a way of com-
municating with the world using a set of symbols. Let I be the set of input
symbols and O, the set of output symbols. In the case of an elevator I might
be up, down, and floor selection, while O might be stops on particular floors.
Besides input and output symbols, there is a set of states S for our model. A
state is like a snapshot of what is happening in the machine at a particular
instant. An elevator might be in a state of going down to the first floor to
pick up a passenger or in a state of stopping on the third floor on the way
up to the fifth floor. There are always an intial state of our model, denoted
by s0 , and final or accepting state(s).
Also, our model has a function, called the next state function. This function
returns the next state based on the present state and input. For instance, if
the elevator is in the state of moving up to the fifth floor and has an input
of someone pressing the down button on the third floor, it goes to a state
that says,” Remember, when coming back down from stopping and picking
someone on the third floor.”
The above discussion is formalized as follows:
A finite-state automaton A consists of five objects:
1. a set I, called the input alphabet, of input symbols;
2. a set O, called the output alphabet, of output symbols;
3. a set S of states the machine can be in;
4. a subset of S whose elements are called accepting states;
5. a next-state function or transition function N : S × I → S. If s ∈ S
and m ∈ I then N (s, m) is the state to which A goes if m is input to A when
A is in state s. The initial state of the machine is s0 .

The operation of a finite-state machine is commomly described by a dia-

gram called a transition diagram. The edges labeled with inputs and
nodes with states. A double circle stands for the final or accepting state(s).

Problem 27.1
Consider the finite-state automaton defined by the transition diagram
27 PROJECT XI: FINITE-STATE AUTOMATON 153

a. What are the elements of S?

b. What are the input symbols?
c. What is the initial state?
d. What are the accepting states?
e. Find N (s3 , 1) and N (s3 , 0).

Let A be a finite-state automaton with input alphabet I. Let I ∗ be the set

of all words with letters from I. A word w ∈ I ∗ is said to be accepted by
A if, and only if, A goes to an accepting state when the symbols of w are
input to A in sequence starting when A is in its initial state. The language
accepted by A, denoted by L(A), is the set of all words that are accepted
by A.

Problem 27.2
Consider the finite-state automaton defined by the following transition dia-
gram
154 RELATIONS AND FUNCTIONS

a. To what states does A go if the symbols of the following words are input
to A in sequence starting from the initial state?
(i) 1101 (ii) 0011 (iii) 0101010.
b. Which of the words in part (a) send A to an accepting state?
c. Show that L(A) = {0(10)n : n ≥ 0} where (10)n = 1010 · · · with n copies
of 10 juxtaposed into one word.

Let A be a finite-state automaton with input alphabet I and states S. Let

N ∗ : S × I ∗ → S be the function defined as follows: N ∗ (s, w) is the state to
which A goes if the symbols of w are input to A in sequence starting when
A is in state s. We call N ∗ the eventual -state function.

Problem 27.3
A finite-state automaton A, given by the transition diagram below, has tran-
sition function N and eventual-state functioin N ∗ .
27 PROJECT XI: FINITE-STATE AUTOMATON 155

a. Find N (s2 , 0) and N (s1 , 0).

b. Find N ∗ (s2 , 11010) and N ∗ (s0 , 01000).

Problem 27.4
Design a finite-state machine that recognizes words of the form 01, 011, 0111, 01111, · · · .
156 RELATIONS AND FUNCTIONS
Introduction to the Analysis of
Algorithms

Informally, an algorithm is any well-defined computational procedure that

takes a set of values as input and produces a set of values as output.
The subject of the analysis of algorithms consists of the study of efficiency of
algorithms. Two aspects of the algorithm efficiency are: the amount of time
required to execute the algorithm and the memory space it consumes. In this
chapter we introduce the basic techniques for calculating time efficiency.

28 Time Complexity and O-Notation

The primary efficiency criterion for analyzing the efficiency of an algorithm
is the running time of the algorithm as a function of the number of values
it processes. The running time of an algorithm is not measured by counting
the minutes and seconds for the algorithm written in a particular language
and running on a particular machine. Rather it is defined to be an estimate
of the number of operations performed by the algorithm given a particular
number of input values.
Generally, given an algorithm that performs a task, we will be interested in
estimating the running time as a function of the problem size. For example,
let us consider the sequential search of an item X from a list of n items.
Here, we say that the problem size is n. Let T (n) be a measure of the time
required to execute an algorithm of problem size n. We call T (n) the time
complexity function of the algorithm. If n is sufficiently small then the
algorithm will not have a long running time. Thus, the interesting question
is:”How fast T (n) increases as n increases?” This is called the asymptotic
behavior of the time complexity function.
In our time analysis we will restrict ourselves to the worst case behavior of

157
158 INTRODUCTION TO THE ANALYSIS OF ALGORITHMS

an algorithm; that is, the longest running time for any input of size n.
Since we are considering asymptotic efficiency of algorithms, basically we will
be focusing on the leading term of T (n). For example, if T (n) = 4n3 − 2n2 +
n + 5 then T (n) = n3 (4 − n2 + n1 + n5 ) and for large n we have T (n) ∼ n3 . We
say that T (n) has a growth of order n3 .
We say that one algorithm is more efficient then another if its worse case
running time has a lower order of growth.

Example 28.1
Estimate the time complexity of the following algorithm:

i := 1
p := 1
for i := 1 to n
p := p · i
i := i + 1
next i

Solution.
Prior to entering the loop, it takes two assignment statements to initialize
the variables i and p. The loop is executed n times, and each time it executes
the two assignment statements in the body of the loop with a total of two
arithmetic operations. Thus, the time complexity of the algorithm is given
by
T (n) = 4n + 2
so the growth is of order n.

Example 28.2
What is the run-time complexity based on n for the following program sege-
ment:

for i := 1 To n
for j := 1 To n
A(i,j) := x
next j
next i

Solution.
The inner loop is executed n times and the outer loop also is executed n
28 TIME COMPLEXITY AND O-NOTATION 159

times. Hence, T (n) = n2 so that the growth is of order n2 .

In the above two problems we found a precise expression for the time com-
plexity of the algorithm. What usually interests us is the order of growth.
We next introduce some of the concepts of growth orders. Let g : IN → IR.
We define the set

O(g(n)) = {f (n) : there exists positive constants n0 and C such that

|f (n)| ≤ C|g(n)|, f or n ≥ n0 }.

We say that a function f is order at most g or f big-oh of g if and

only if f (n) ∈ O(g(n)). Sometimes We write f (n) = O(g(n)). Graphically,
this means that for n ≥ n0 the graph of f is below the graph of g.
Example 28.3
Show that the time complexity found in Exercise 390 is O(n).
Solution.
To show that T (n) = O(n) we must produce constants C and n0 such that
T (n) ≤ Cn for n ≥ n0 . Indeed, T (n) ≤ 5n for n ≥ 2 so that n0 = 2 and
C = 5.
Example 28.4
Show that the run-time complexity based on n for the following program
segement is O(n2 ).

s := 0
for i := 1 To n
for j := 1 To i
s := s + j · (i − j + 1)
next j
next i
Solution.
Prior to entering the loop there is one assignment statement. Now, there are
two additions, one subtraction, one multiplication and one assignment for
each iteration of the inner loop. The total number of time the inner loop is
executed is
n(n + 1)
1 + 2 + 3 + ··· + n =
2
220 ELEMENTS OF GRAPH THEORY

and then the subtrees T1 , T2 , · · · , Tn are listed, from left to right, in order
of their roots. The preorder traversal begins by visiting r. It continues by
traversing T1 in preorder, thenT2 in preorder, and so on, until Tn is traversed
in preorder. In which order does a preorder traversal visit the vertices in the
following rooted tree?