Boolean Algebra

◎ Part 2: Representing Boolean Functions

1
Two important Boolean algebra problems

1. Given the values of a Boolean function, how can a Boolean

expression that represents this function be found?
○ Will be solved by showing that any Boolean function can be
represented by a Boolean sum of Boolean products of the variables
and their complements.
2. Is there a smaller set of operators that can be used to
represent all Boolean functions?
○ Will be answered by showing that all Boolean functions can be
represented using only one operator.

Both of these problems have practical importance in circuit design.

2
Representing Boolean Functions
• Sum-of-products Expansions
 Disjunctive Normal Form (DNF) orSum-of-Products Form (SOPs)
• Product-of-sums Expansions
 Conjunctive Normal Form (CNF) or Product-of-Sums Form
(POSs)
• Functional Completeness
 Minimal functionally complete sets of operators.

3
Constructing Boolean Expressions that
represent a Boolean Function
Example: Find Boolean expressions that represent the
functions F(x, y, z) and G(x, y, z), which are given in the table.

Solution:
• An expression that has the value 1 when x = z = 1 and
y = 0, and the value 0 otherwise, is needed for F. It can
be formed by taking the Boolean product of x, 𝑦,
ത and z
i.e. x𝑦z.
ത
• For G, we need an expression that equals 1 when x =
y = 1 and z = 0, or x = z = 0 and y = 1. It can be formed
by taking the Boolean sum of two different Boolean
products. The expression: xy𝑧ҧ + 𝑥yҧ 𝑧ҧ represents G.
4
Comments on the previous example

◎ The example illustrates a procedure for constructing a

Boolean expression representing a function with given values.

◎ Each combination of values of the variables for which the

function has the value 1 leads to a Boolean product of the
variables or their complements.

5
Definitions: Literals, Minterms, Maxterms

• A literal is a Boolean variable or its complement.

• A minterm of Boolean variables x1,…,xn is a Boolean product
of n literals y1…yn, where yi = xi , or yi = 𝑥ҧ𝑖 .
• A maxterm of Boolean variables x1,…,xn is a Boolean sum of
n literals y1…yn, where yi = xi , or yi = 𝑥ҧ𝑖 .

6
Examples on minterms
1. Find a minterm that equals 1 if x1 = x3 = 0 and x2 = x4 =
x5 = 1, and equals 0 otherwise.
Solution:
The minterm: 𝑥ҧ1 𝑥2 𝑥ҧ3 𝑥4 𝑥5

2. Find a Boolean product of the Boolean variables x, y,

and z, or their complements, that has the value 1 if
and only if x = y = 0, z = 1.
Solution:
The minterm: 𝑥ҧ 𝑦𝑧ത
7
Sum-Of-Products (SOP) Expansion
◎ By taking Boolean sums of distinct minterms we can build up a Boolean
expression with a specified set of values.
◎ In particular, a Boolean sum of minterms has the value 1 when exactly
one of the minterms in the sum has the value 1. It has the value 0 for all
other combinations of values of the variables.
◎ Consequently, given a Boolean function, a Boolean sum of minterms can
be formed that has the value 1 when this Boolean function has the value
1, and has the value 0 when the function has the value 0.
◎ The minterms in this Boolean sum correspond to those combinations of
values for which the function has the value 1.
◎ The sum of minterms that represents the function is called the sum-of-
products expansion or the disjunctive normal form of the Boolean
function.
8
Sum-of-Products (SOP) Expansion
• Theorem: Any Boolean function can be represented
as a sum-of-products of variables and their
complements.
– Proof: By construction from the function’s truth
table. For each row that has a function value 1,
include a product term in the sum that represents
the variables values given for that row.

9
Disjunctive Normal Form (DNF)
Sum-of-Products (SOP)

• The Disjunctive Normal Form (DNF) of Boolean function f, of a

degree-n, is the unique sum of minterms of the variables x1,…,xn that
represents f.

• A DNF is a sum-of-products representation.

10
SOP Example – Solution (1)
◎ Find the sum-of-products expansion for the function
𝐹 𝑥, 𝑦, 𝑧 = (𝑥 + 𝑦)𝑧.ҧ
Solution (1):

11
SOP Example – Solution (2)

◎ We can construct the SOP expansion by

determining the values of F for all possible
values of the variables x, y, and z, as in
table.
◎ The SOP expansion of F is the Boolean
sum of three minterms corresponding to
the three rows of this table that give the
value 1 for the function. This gives:

12
Product-of-Sums (POS) Expansion
◎ It is also possible to find a Boolean expression that represents
a Boolean function by taking a Boolean product of Boolean
sums.
◎ The resulting expansion is called the conjunctive normal form
or product-of-sums expansion of the function.
◎ The Conjunctive Normal Form (CNF) of Boolean function f, of a
degree-n, is the unique product of maxterms of the variables
x1,…,xn that represents f.
◎ A CNF is a product-of-sums representation.
◎ These expansions can be found from SOP expansions by
taking duals.

13
POS Example
Find the POS expansion of the Boolean function:
𝐹 𝑥, 𝑦, 𝑧 = 𝑥 𝑦.
ത
Solution: 𝒙 𝒚 𝒛 ഥ 𝒙ഥ
𝒚 𝒚

• First we construct the table of values for 𝐹 𝑥, 𝑦, 𝑧 : 0 0 0 1 0

0 0 1 1 0

• We then find the six maxterms that have value of 0 0 1 0 0 0

0 1 1 0 0
and construct their product: 1 0 0 1 1
1 0 1 1 1
𝐹 𝑥, 𝑦, 𝑧 = 𝑥 + 𝑦 + 𝑧 𝑥 + 𝑦 + 𝑧ҧ 𝑥 + 𝑦ത + 𝑧 1 1 0 0 0
(𝑥 + 𝑦ത + 𝑧)(
ҧ 𝑥ҧ + 𝑦ത + 𝑧)(𝑥ҧ + 𝑦ത + 𝑧)ҧ 1 1 1 0 0

14
Convert DNF (SOP) to CNF (POS) representation
To get the CNF representation for f,
• take the DNF (SOP) representation for complement 𝑓,ҧ 𝒙 𝒚 𝒛 ഥ
𝒙 𝑥ҧ + 𝑦
• and then complement both sides (apply De Morgan’s 0 0 0 1 1
laws) to get the CNF (POS) representation of f. 0 0 1 1 1
Example: Using the above procedure find the POS expansion 0 1 0 1 1
of the Boolean function: 0 1 1 1 1
1 0 0 0 0
𝐹 𝑥, 𝑦, 𝑧 = 𝑥𝑦. ത
Solution: 1 0 1 0 0

• 𝐹(𝑥, 𝑦, 𝑧) = 𝑥ҧ + 𝑦 (then construct table) 1 1 0 0 1

• 𝐹(𝑥, 𝑦, 𝑧) = 𝑥ҧ 𝑦ത𝑧ҧ + 𝑥ҧ 𝑦𝑧
ത + 𝑥𝑦
ҧ 𝑧ҧ + 𝑥𝑦𝑧
ҧ + 𝑥𝑦𝑧ҧ + 𝑥𝑦𝑧 1 1 1 0 1

• Complement both sides of this SOP representation and apply

De Morgan laws to get:
𝐹 𝑥, 𝑦, 𝑧 = 𝑥 + 𝑦 + 𝑧 𝑥 + 𝑦 + 𝑧ҧ 𝑥 + 𝑦ത + 𝑧
(𝑥 + 𝑦ത + 𝑧)(
ҧ 𝑥ҧ + 𝑦ത + 𝑧)(𝑥ҧ + 𝑦ത + 𝑧)ҧ
15
Row 𝒙𝒚𝒛 𝒎𝒊𝒏𝒕𝒆𝒓𝒎𝒔 𝑴𝒂𝒙𝒕𝒆𝒓𝒎𝒔
No.
0 000 𝑥ҧ 𝑦ത𝑧ҧ = 𝑚0 𝑥 + 𝑦 + 𝑧 = 𝑀0
1 001 𝑥ҧ 𝑦𝑧
ത = 𝑚1 𝑥 + 𝑦 + 𝑧ҧ = 𝑀1
2 010 𝑥𝑦
ҧ 𝑧 ҧ = 𝑚2 𝑥 + 𝑦ത + 𝑧 = 𝑀2
3 011 𝑥𝑦𝑧
ҧ = 𝑚3 𝑥 + 𝑦ത + 𝑧ҧ = 𝑀3
4 100 𝑥 𝑦ത 𝑧ҧ = 𝑚4 𝑥ҧ + 𝑦 + 𝑧 = 𝑀4
5 101 𝑥 𝑦𝑧
ത = 𝑚5 𝑥ҧ + 𝑦 + 𝑧ҧ = 𝑀5
6 110 𝑥𝑦𝑧ҧ = 𝑚6 𝑥ҧ + 𝑦ത + 𝑧 = 𝑀6
7 111 𝑥𝑦𝑧 = 𝑚7 𝑥ҧ + 𝑦ത + 𝑧ҧ = 𝑀7
ഥ0 = 𝑚0 and 𝑚
Note that: 𝑀 ഥ 7 = 𝑀7
16
Functional Completeness
• Since every Boolean function can be expressed in terms of ·,+,¯,
we say that the set of operators {·,+,¯} is functionally complete
• There are smaller sets of operators that are also functionally
complete.
– We can eliminate either · or + using De Morgan’s law.
– So, {+,¯} and {·,¯} are functionally complete.
• So, we have sets of 3 operators and sets of 2 operators that are
functionally complete. Are there sets of a single operator that are
functionally complete? See next slide.

17
Functional Completeness x y x|y x y xy

NAND, | , and NOR, ↓ operators, defined in the 00 1 00 1

tables on the right, are functionally complete, 01 1 01 0

each by itself (as a singleton set, {|} and {}). 10 1 10 0

11 0 11 0

NAND (|) NOR ()

• To see that {|} is functionally complete: Because {. ,¯} is functionally

complete, all that we have to do is show that both of the operators ・
and ¯ can be expressed using just the | operator
• Similar treatment for the case of {} can be performed. Both results are
given in the table below:
NAND NOR
𝑥ҧ = 𝑥 | 𝑥 𝑥ҧ = 𝑥 ↓ 𝑥
𝑥. 𝑦 = 𝑥 𝑦)| 𝑥 𝑦) 𝑥 + 𝑦 = (𝑥 ↓ 𝑦) ↓ (𝑥 ↓ 𝑦)
18