120 Solutions Manual of Elements of Discrete Mathematics

CHAPTER

ELEVEN
BOOLEAN ALGEBRAS

11.1 Suppose a Ù b = b. Then b Ú a = (a Ù b) Ú a = a

Suppose b Ú a = a. Then a Ù b = (b Ú a) Ù b = b
11.2 a Ú (b Ù c) £ (a Ú b) Ù (a Ú c) £ b Ù (a Ú c)
11.3 a Ú (b Ù c) £ a Ú b
a Ú (b Ù c) £ a Ú c
Thus, a Ú (b Ù c) £ (a Ú b) Ù (a Ú c)
The second inequality follows from duality.
11.4 If b £ a, then a Ù b = b which contradicts the condition a Ù b < b.
If a £ b, then a Ù b = a which contradicts the condition a Ù b < a.
Thus, if a Ù b < a and a Ù b < b, then a and b are incomparable.
On the other hand, suppose a and b are incomparable.
Since a Ù b £ a, if a Ù b = a, then a £ b, which is a contradiction.
Since a Ù b £ b, if a Ù b = b, then b £ a, which is a contradiction.
11.5 Since a Ú (a Ù x) = a, we have a Ù a = a Ù (a Ú (a Ù x)). However,
according to the absorption law a Ù (a Ú (a Ù x)) = a. Thus, a Ù a = a.
11.6 (a) Transitivity: a £ b, b £ c Þ a Ù b = a, b Ù c = b
Þ a Ù c = (a Ù b) Ù c = a Ù (b Ù c)
= a Ù b = a.
Antisymmerty: a £ b, b £ a Þ a Ù b = a, a Ù b = b Þ a = b.
Reflexivity: Let a Ú a = b. Then a Ù b = a Ù (a Ú a) = a
so a Ú a = a Ú (a Ù b) = a by absorption.
Thus a Ù a = a Ù (a Ú a) = a and hence a £ a.
(b) a £ a Ú b since a Ù (a Ú b) = a by absorption. Similarly, b £ a Ú b.
Thus, a Ú b is an upper bound of a and b.
 Boolean Algebras 121

Also, (a Ú b) Ù (a Ú b Ú c) = a Ú b by absorption.
If a £ c and b £ c, then c = c Ú (c Ù a) = c Ú a. Similarly, b Ú c = c.
Hence, (a Ú b) Ù (a Ú b Ú c) = (a Ú b) Ù (a Ú c) = (a Ú b) Ù c.
Thus, a Ú b = (a Ú b) Ù c and a Ú b £ c.
Similarly, a Ù b £ a and a Ù b £ b since a Ù (a Ù b) = a Ù b and b Ù
(b Ù a) = b Ù a.
If c £ a and c £ b, then a Ù c = c and b Ù c = c.
Thus, c Ù (a Ù b) = c Ù b = c, so c £ a Ù b.
11.7 Clearly, (a Ú b) Ù c £ a Ú (b Ù c) £ (a Ù c) Ú (b Ù c). For the reverse
inclusion we note that the lattice is modular since if a £ c, then a Ú (b Ù c)
³ (a Ú b) Ù c by assumption. While a £ a Ú b, a £ c, and b Ù c £ a Ú b, b
Ù c £ c, so a Ú (b Ù c) £ (a Ú b) Ù c. Thus, since a Ù c £ c, (a Ù c) Ú (b Ù
c) = ((a Ù c) Ú b) Ù c ³ ((a Ú b) Ù c) = (a Ú b) Ù c.
11.8 We have x Ú (a Ù x) = (x Ú a) Ù (x Ú x) = x
However, x Ú (a Ù x) = x Ú (a Ù y)
= (x Ú a) Ù (x Ú y)
= (y Ú a) Ù (x Ú y)
= (a Ù x) Ú y
= (a Ù y) Ú y
=y
11.9 If (A, £) is distributive, then
(a Ú b) Ù (b Ú c) Ù (c Ú a) = [(a Ú b) Ù (b Ú c) Ù c] Ú [(a Ú b) Ù (b Ú c)
Ù a]
= [(a Ú b) Ù c] Ú [(b Ú c) Ù a]
= [(a Ù c) Ú (b Ù c)] Ú [(b Ù a) Ú (c Ù a)]
= (a Ù b) Ú (b Ù c) Ú (c Ù a).
Conversely, if this condition holds, then
a Ù (b Ú c) = a Ù ((a Ú b) Ù (a Ú c) Ù (b Ú c)),
(since (a Ú b) Ù a = a = (a Ú c) Ù a)
= a Ù [(b Ù c) Ú (c Ù a) Ú (a Ù b)],
(by the assumed condition)
= a Ù [(b Ù c) Ú a] Ù [(b Ù c) Ú (c Ù a) Ú (a Ù b)]
(since a Ù [(b Ù c) Ú a] = a)
= [a Ú [(c Ù a) Ú (a Ù b)]] Ú [(b Ù c) Ù a] Ú [(b Ù c) Ù [(c Ù a)
Ú (a Ù b)]]
(since a Ú (c Ù a) Ú (a Ù b) = a)
= [a Ù [(c Ù a) Ú (a Ù b)]] Ú [(b Ù c) Ù a] Ú [(b Ù c) Ù [(c Ù a)
Ú (a Ù b)]]
(by the assumed condition)
122 Solutions Manual of Elements of Discrete Mathematics

= [(c Ù a) Ú (a Ù b)] Ú [(b Ù c) Ù a] Ú [(b Ù c) Ù [(c Ù a) Ú (a

Ù b)]]
(since a ³ (c Ù a) Ú (a Ù b))
= [(c Ù a) Ú (a Ù b)] Ú [(b Ù c) Ù a]
(since [(b Ù c) Ù [(c Ù a) Ú (a Ù b)]]
£ [(c Ù a) Ú (a Ù b)])
= (c Ù a) Ú (a Ù b)
(since (a Ù b Ù c) £ (a Ù b))
11.10 If (A, £) is modular, then a £ a Ú c, so a Ú (b Ù (a Ú c)) = (a Ú b) Ù (a Ú
c).
If the condition holds, let a £ c. Then c = a Ú c and
a Ú (b Ù c) = a Ú [b Ù (a Ú c)]
= (a Ú b) Ù (a Ú a Ú c)
= (a Ú b) Ù (a Ú c)
= (a Ú b) Ù c.
11.11 Clearly (c Ú b) Ù a ³ b Ù a and (c Ú b) Ù a £ a.
To show (c Ú b) Ù a £ b we observe that
(c Ú b) Ù a £ (c Ú b) Ù (a Ú b) £ b Ú [c Ù (a Ú b)]
by the modular law. But by assumption, c Ù (a Ú b) = b Ù c.
Thus, (c Ú b) Ù a £ b Ú (b Ù c) = b. And (c Ú b) Ù a £ a Ù b.
11.12 (a) (2) Þ (2 ): If x £ a, then a Ù x = x.
¢

Therefore, a Ù x ÎI Þ x ÎI
(2 ) Þ (2): For any a Î I and any x Î A, a Ù x £ a. Thus, a Ù x is in I.
¢

(b) Let I be an ideal. Clearly, (1²) is satisfied. According to condition (2)

if a Ú b is in I, a = (a Ú b) Ù a is also in I, and b = (a Ú b) Ù b is also
in I. Thus condition (2² ) is satisfied.
Now, suppose I is a set satisfying conditions (1² ) and (2 ² ).
Consider any a in I and x £ a. Note that a Ú x = a. According to
condition (2²), a Ú x in I implies a Î I and x Î I. Thus condition (2¢)
is satisfied.
11.13 (i) For any c and d in I(a, b)
aÙc =bÙc
aÙd =bÙd
a Ù (c Ú d) = (a Ù c) Ú (a Ù d)
= (b Ú c) Ú (b Ù d)
= b Ù (c Ú d)
Thus, c Ú d is in I(a, b).
 Boolean Algebras 123

(ii) For any c in I(a, b) and any y in A

a Ú (c Ù y) = (a Ù c) Ù y
= (b Ù c) Ù y
= b Ù (c Ù y)
Thus, c Ù y is in I(a, b).
11.14 Let x be a complement of 0. By definition, 0 Ú x = 1, 0 Ù x = 0. Also 0 £ x,
and 0 Ú x = x. Thus, x = 1.
11.15 a Ú ( a Ù b) = (a Ú a ) Ù (a Ú b) = a Ú b
a Ù (a Ú b) = (a Ù a ) Ú (a Ù b) = a Ù b
11.16 (A, Å) is certainly closed. Commutativity is clear. Furthermore, 0 is the
identity since a Å 0 = a, and a is its own inverse since a Å a = 0. As to
associativity, we note that
(a Å b) Å c = (((a Ù b ) Ú ( a Ù b)) Ù c ) Ú ( ((a ˜ b ) ™ (a ˜ b)) Ù c)
= (a Ù b Ù c ) Ú ( a Ù b Ù c ) Ú ( ( a ˜ b ) Ù ( a ˜ b) Ù c)
= (a Ù b Ù c ) Ú ( a Ù b Ù c ) Ú (( a Ú b) Ù (a Ú b ) Ù c)
= (a Ù b Ù c ) Ú ( a Ù b Ù c ) Ú (((a Ù b) Ú ( a Ù b )) Ù c)
= (a Ù b Ù c ) Ú ( a Ù b Ù c ) Ú (a Ù b Ù c) Ú ( a Ù b Ù c)
A similar expansion for a Å (b Å c) yields the same expression.
11.17 (a) (i) a H a = e implies for all a, a £ a. (reflexivity)
(ii) a £ b and b £ a mean a H b = e and b H a = e.
By Problem 12.13 (e), a = b. (antisymmetry)
(iii) a £ b and b £ c mean a H b = b H c = e.
a H c = a H (e H c) = a H [(b H c) H c]
= a H [(c H b) H b] = (c H b) H (a H b)
= (c H b) H e = e. Thus, a £ c (transitivity)
Hence, £ is a partial enduring relation.
(b) If b = x H a, then a H b = a H (x H a) = x H (a H a) = x H e = e
If a H b = e, then b = e H b = (a H b) H b = (b H a) H a = x H a
where x = b H a.
(c) a H [(a H b) H b] = e Þ
a £ (a H b) H b.
Similarly, b £ (b H a) H a = (a H b) H b.
Thus, (a H b) H b is an upper bound for a and b.
Now suppose a £ c, b £ c. According to (b), c = x H a for some x.
Also, b H c = e.
Hence [(a H b) H b] H c = [(a H b) H b] H (x H a)
= x H [((a H b) H b) H a]
= x H (b H a)
= b H (x H a) = b H c = e.
124 Solutions Manual of Elements of Discrete Mathematics

Hence (a H b) H b £ c
i.e., (a H b) H b is a least upper bound.

11.18 (i) (a H b) H a = ( a ™ b) Ú a = (a Ù b ) Ú a = a
(ii) (a H b) H b = ( a ™ b) Ú b = (a Ù b ) Ú b = a Ú b
(b H a) H a = (b ™ a ) Ú a = (b Ù a ) Ú a = a Ú b
(iii) a H (b H c) = a Ú ( b Ú c)
b H (a H c) = b Ú ( a Ú c)
Hence, (A, H) is an implication algebra.

11.19 E(x1, x2, x3) = (x1 Ù x2 Ù x 3) Ú (x1 Ù x2 Ù x3) Ú ( x1 Ù x 2 Ù x3)

Ú ( x1 Ù x 2 Ù x3)
E(x1, x2, x3) = (x1 Ú x2 Ú x3) Ù (x1 Ú x 2 Ú x3) Ù (x1 Ú x 2 Ú x 3)
Ù ( x1 Ú x2 Ú x3)
11.20 E(x1, x2, x3) = (x1 Ù x2 Ù x3) Ú (x1 Ù x2 Ù x 3) Ú (x1 Ù x2 Ù x3) Ú (x1 Ù x 2
Ù x 3) Ú ( x 1
Ù x2 Ù x 3)
= (x1 Ú x2 Ú x3) Ù (x1 Ú x2 Ú x 3) Ù (x1 Ú x 2 Ú x 3)

11.21 E(x1, x2, x3, x4) = (x1 Ù x2 Ù x 3 Ù x4) Ú (x1 Ù x2 Ù x 3 Ù x 4 ) Ú (x1 Ù x 2 Ù x3

Ù x4) Ú (x1 Ù x 2 Ù x 3 Ù x4) Ú ( x1 Ù x2 Ù x 3 Ù x 4 )
= (x1 Ú x2 Ú x3 Ú x4) Ù (x1 Ú x2 Ú x3 Ú x 4 ) Ù (x1 Ú x2 Ú x 3
Ú x4) Ù (x1 Ú x2 Ú x 3 Ú x 4 )
Ù (x1 Ú x 2 Ú x3 Ú x 4 ) Ù (x1 Ú x2 Ú x 3 Ú x4) Ù (x1 Ú x 2
Ú x 3 Ú x 4 ) Ù ( x1 Ú x 2 Ú x 3 Ú x 4 )
Ù ( x1 Ú x 2 Ú x 3 Ú x4) Ù ( x1 Ú x2 Ú x3 Ú x4) Ù ( x1 Ú x2 Ú
x 3 Ú x4)

11.22 (a) ( x1 Ù x 2 Ù x 3) Ú ( x1 Ù x2 Ù x 3) Ú (x1 Ù x 2 Ù x3) Ú (x1 Ù x2 Ù x3)

(b) (x1 Ú x2 Ú x 3) Ù (x1 Ú x 2 Ú x 3) Ù ( x1 Ú x2 Ú x3) Ù ( x1 Ú x 2 Ú x3)

11.23 (a) b Ù (a Ú c)
(b) (a Ù b) Ú (a Ù c) Ú (b Ù c)
(c) c
(d) (a Ú b ) Ù c
11.24 (a) The basis of induction is obvious. Let E(x1, x2,…, xn) be an
expression of length k + 1. We consider the following cases:
 Boolean Algebras 125

{
(i) E(x1, x2,…, xn) = E1 ( x1 , x2 , , xn ) , where E1 (x1, x2,…, xn)
is an expression of length k. According to the induction
hypothesis:
E(x1, x2,…, xn) = ( x i ˜ E1 ( xi = 0)) ™ ( xi ˜ E1 ( xi = 1))

= (xi Ú E1 ( xi = 0) ) Ù ( x i Ú E1 ( xi = 1) )

= (xi Ù E1 ( xi = 1) ) Ú ( x i Ù E1 ( xi = 0) )
= (xi Ù E(xi = 1)) Ú ( x i Ù E(xi = 0))
(ii) E(x1, x2,…, xn) = E1(x1, x2,…, xn) Ù E2(x1, x2,…, xn).
According to the induction hypothesis:
E(x1, x2,…, xn) = [( x i Ù E1(xi = 0)) Ú (xi Ù E1(xi = 1))] Ù
[( x i Ù Ee2(xi = 0)) Ú (xi Ù E2(xi = 1))]
= [ x i Ù E1(xi = 0) Ù E2(xi = 0)] Ú
[xi Ù E1(xi = 1) Ù E2 (xi = 1)]
= ( x i Ù E(xi = 0)) Ú (xi Ù E(xi = 1))
(iii) E(x1, x2,…, xn) = E1(x1, x2,… xn) Ú E2(x1, x2,…, xn).
(Similar to (ii).)
(b) Repeated applications of the result in (a) yield
cd1d 2…d n = E(x1 = d1, x2 = d2,…, xn = dn)
where di = 0 if ~x i = x i and di = 1 if ~
x i = xi.
(c) (2 Ù x1 Ù x 2 ) Ú (2 Ù x1 Ù x2).
(d) Determine the disjunctive normal form from the 2 n values of f (d1,
d2,…, dn) where di = 0 or 1.
(e) According to f (0, 0) = 1, f (0, 1) = 0, f (1, 0) = 1, f (1, 1) = 1 from
Figure 11.8, we should have
f (x1, x2) = ( x1 Ù x 2 ) Ú (x1 Ù x 2 ) Ú (x1 Ù x2)
However, if that is the case

f (0, 2) = ( 0 Ù 2 ) Ú (0 Ù 2 ) Ú (0 Ù 2) = 3
which is not consistent with the value of f (0, 2) in Figure 11.8.
(f) Similar to (a).
(g) Similar to (b).

11.25 ((a Ú b) Ù d) Ú ((a Ú b) Ù (c Ù d) Ù e ) Ú [((a Ú b) Ú (c Ú d)) Ù f ] Ú g

= ((a Ú b) Ù d) Ú ((a Ú b) Ù c Ù e ) Ú ((a Ú b Ú c Ú d) Ù f ) Ú g
Condition (2) can be simplified as:
126 Solutions Manual of Elements of Discrete Mathematics

(2) He got a B or better in the mid-term examination and a B in the final

examination and did not miss any homework assignment.
11.26 (a Ú c) Ù (a Ú b Ú d) Ù (b Ú c Ú e) Ù (b Ú e) Ù d
= (a Ú c) Ù (b Ú e) Ù d
= (a Ù b Ù d) Ú (a Ù e Ù d) Ú (c Ù b Ù d) Ú (c Ù e Ù d)
11.27 Corresponding to the five conditions, we have
(1) (a Ù b ) Ú ( a Ù b)
(2) c Ú e
(3) d Ú b
(4) (a Ù c) Ú ( a Ù c)
(5) e Ú (c Ù d).
Simplifying the conjunction of these five expressions, we obtain a Ù b Ùc
ÙdÙe as the only way of selection.
11.28

b c d e

f h

(a Ú b Ú c Ú d Ú e) Ù (b Ú a Ú c Ú g) Ù (c Ú a Ú b Ú d Ú f ) Ù (d Ú a Ú c Ú
e Ú f ) Ù (e Ú a Ú d Ú h) Ù (f Ú c Ú d Ú g Ú h) Ù (g Ú b Ú f Ú h) Ù (h Ú e Ú
f Ú g)
= (a Ù f ) Ú (a Ù g) Ú (a Ù h) Ú (b Ù e Ù f ) څ
11.29

a
c

b
d

x
11.30 (a)
x⁄y
11.30 (a)
y
 Boolean Algebras 127

xŸy
(b)
y

–
(c) x NOR x

–
x
– NOR xŸy
y

(d) x– x
NAND x⁄y NOR NOR x⁄y
y– y

(d)
x
NAND NAND xŸy x NAND x–
y

11.31 (a) The three simplified expressions are

(a) ( a Ù (b Ú c )) Ú (a Ù b ) Ú (d Ù ( a Ú c ))
(b) ( a Ù b ) Ú (a Ù c) Ú d
(c) ( a Ù b) Ú ( b Ù c)

(b) a

b
output
c

d
(a)

output
b

c
d

(b)

a
output
b

c
(c)