CS147 - Lecture 05

Kaushik Patra
([email protected])

● Standard Forms

● Karnaugh Map

[ Chapter 2 (2-3, 2-4) of Logic & Computer Design Fundamentals, 4th Edition,
M. Morris Mano, Charles R. Kime ]

© 2014 - All rights are reserved by Kaushik Patra 1

 Standard Forms ...

© 2014 - All rights are reserved by Kaushik Patra 2

 SOP & POS
● There are two types of standard forms
– Sum of Product (SOP) terms
– Product of Sum (POS) terms

● Purpose of Standard forms

– Compact forms for larger expression
– Facilitate simplification process

● Standard form contains product terms and sum terms.

● Product terms contain only AND operations. Example of product
terms will be XYZ'
● Sum terms contain only OR operations. Example of sum terms will be
X+Y'+Z.

● Sum of Product (SOP) is OR operations between product terms. An

example of SOP expression is ABC + B'C + A'B.

● Product of Sum (POS) is AND operations between sum terms. An

example of POS expression is (A+B+C)(B'+C)(A'+B).

© 2014 - All rights are reserved by Kaushik Patra 3

 Minterms & Maxterms
● Extension of two types of term in SOP, POS

– 'Minterms' are the product terms containing all the

variables of the functions appearing exactly once in their
complemented or un-complemented form.

– 'Maxterms' are the sum terms containing all the variables

of the functions appearing exactly once in their
complemented or un-complemented form.

● There are total 2n minterms or maxterms for a n-variable Boolean

functions.

● Both minterms and maxterms property is that they represent exactly one
combination of binary variable values in the truth table.

● Example of a minterm for a function F(X,Y,Z) will be X'YZ'. There will

be total 23 = 8 minterms for this function.

● Example of a minterm for a function F(X,Y,Z) will be (X' + Y + Z'). There

will be total 23 = 8 maxterms for this function.

© 2014 - All rights are reserved by Kaushik Patra 4

 SOP of minterms
X Y Z Product Term Symbol F(X,Y,Z)
0 0 0 X'Y'Z' m0 1
0 0 1 X'Y'Z m1 0
0 1 0 X'YZ' m2 1
0 1 1 X'YZ m3 0
1 0 0 XY'Z' m4 0
1 0 1 XY'Z m5 1
1 1 0 XYZ' m6 0
1 1 1 XYZ m7 1

F(X,Y,Z) = X'Y'Z' + X'YZ' + XY'Z+ XYZ = m0 + m2 + m5 + m7 = Σm(0,2,5,7)

● Each row in the truth table represents one unique minterm.

● In a minterm variable is complemented for a specific row in the truth table if its
corresponding value in the truth table for that term is 0. If the value is 1, non-
complemented form of the variable is taken for constructing the minterm. For
example for the 3rd row of the above truth table has value entry of X=0, Y=1 and
Z=0. Hence the corresponding minterm is X'YZ'.

● If a function contains high number of variables (e.g. 100 variable boolean function)
it is not very efficient to write all the variables in the minterm. That is why,
minterms are alternatively represented as mx, where x is the decimal value
corresponding the binary value represented by the variables in that row. For
example X'YZ' is represented a m2.

● The function is expressed in SOP as logical sum of all the minterms which results
the function value to be 1 in the truth table. Instead of writing explicitly, the sum
is expressed as sum of the corresponding minterm symbol. To further reduce the
expression notation classical algebraic summation symbol is used with list of
indices for the minterm. For example 'F(X,Y,Z) = Σm(0,2,5,7)' means 'F(X,Y,Z)
= X'Y'Z' + X'YZ' + XY'Z + XYZ'.

© 2014 - All rights are reserved by Kaushik Patra 5

 POS of maxterms

X Y Z Sum Term Symbol F(X,Y,Z)

0 0 0 X+Y+Z M0 1
0 0 1 X+Y+Z' M1 0
0 1 0 X+Y'+Z M2 1
0 1 1 X+Y'+Z' M3 0
1 0 0 X'+Y+Z M4 0
1 0 1 X'+Y+Z' M5 1
1 1 0 X'+Y'+Z M6 0
1 1 1 X'+Y'+Z' M7 1

F(X,Y,Z) = (X+Y+Z')(X+Y'+Z')(X'+Y+Z)(X'+Y'+Z) = M1M3M4M6 = ΠM(1,3,4,6)

● Each row in the truth table represents one unique maxterm.

● In a maxterm variable is complemented for a specific row in the truth table if its
corresponding value in the truth table for that term is 1. If the value is 0, non-
complemented form of the variable is taken for constructing the maxterm. For
example for the 3rd row of the above truth table has value entry of X=0, Y=1 and
Z=0. Hence the corresponding maxterm is (X+Y'+Z).

● Maxterms are alternatively represented as Mx, where x is the decimal value

corresponding the binary value represented by the variables in that row. For
example (X+Y'+Z) is represented a M2.

● The function is expressed in POS as logical product of all the maxterms which
results the function value to be 0 in the truth table. Instead of writing explicitly,
the product is expressed as product of the corresponding maxterm symbol. To
further reduce the expression notation classical algebraic multiplication symbol
is used with list of indices for the maxterm. For example 'F(X,Y,Z) =
ΠM(1,3,4,6)' means 'F(X,Y,Z) = (X+Y+Z')(X+Y'+Z')(X'+Y+Z)(X'+Y'+Z)'.

© 2014 - All rights are reserved by Kaushik Patra 6

 Relation between SOP & POS
● Maxterm and minterm with same index holds
complement relation.
– e.g. M3 = m3'
– M3 = X + Y' + Z' = (X'YZ)' = m3'

● If F(X,Y,Z) = Σm(0,2,5,7), then F'(X,Y,Z) = Σm(1,3,4,6)

– i.e. F(X,Y,Z) = (m1 + m3 + m4 + m6)' = m1'.m3'.m4'.m6'
– Hence F(X,Y,Z) = M1.M3.M4.M6 = ΠM(1,3,4,6)
– SOP expression is equivalent to its POS expression with
exact set of disjoint indices to its SOP equivalent.
7

● To calculate SOP expression of complement of a function in SOP form, we take

those entries in the truth table which results 0. In the complement function those
rows with 0 as result in original function will become 1 and rows with 1 as the
result in the original function will become zero. The indices of the minterm set in
the original function is disjoint to the indices of the minterm set in the
complimented function.

● Fun facts
● Maxterms are seldom used directly when dealing with Boolean functions, since
we can always replace them with minterm expression.
● A function that includes all the minterms is equal to logic 1.
● Any Boolean function can be expressed as logical sum of minterms.

© 2014 - All rights are reserved by Kaushik Patra 7

 Karnaugh Map ...

© 2014 - All rights are reserved by Kaushik Patra 8

 Karnaugh Map
● It is a visual technique to reduce Boolean expression
in standard form (SOP or POS).
– It takes advantage of human pattern recognition ability.

● First introduced by Maurice Karnaugh in 1953.

– It was a refinement of Veitch diagram technique developed
in 1952 by Edward Veitch.

● It is usually called as K-map technique.

© 2014 - All rights are reserved by Kaushik Patra 9

 K-Map Table Structure

Gray Code

Y YZ
0 1 00 01 11 10
X
X
0 1 0 1 3 2
0 0

2 3 4 5 7 6
1 1

Two variable K-Map Three variable K-Map

10

● The number of rows and columns are exact power of 2. For example for a
3 variable K-Map, number of rows is 2 and columns is 4.

● Corresponding variables are written on the top-left corner of the table.

● Table row and column indices are written in Gray Code.

● Gray code encoding of number ensures that two adjacent binary
numbers differs only by one bit.
● Each column are adjacent to each other.
● The first and last columns are adjacent to each other.
● Each rows are adjacent to each other.
● The first and last row are adjacent to each other.

● Each cell / box of the table is marked with corresponding minterm

number.

© 2014 - All rights are reserved by Kaushik Patra 10

 K-Map Table Structure
Gray Code
YZ 00 01 11 10
WX
0 1 3 2
00

4 5 7 6
01
Gray Code

11 12 13 15 14

8 9 11 10
10

Four variable K-Map

11

© 2014 - All rights are reserved by Kaushik Patra 11

 K-Map Reduction Process
● For a SOP expression put 1 for each minterm in the
expression in the corresponding box in the table.

● Group adjacent 1s in maximum way in power of two –

1, 2, 4, 8, etc.

● Write one product term for each such group of one.

12

● For a Boolean expression F(X,Y,Z) = Σm(2,3,5,6), 1 will be placed in

the box with minterm indices 2,3,5, and 6.

● Different group of one can overlap each other. However, they must
contain some unique 1s, i.e. one group can not be a subset of the union
of the other group.

● Product terms will contain only the variable (or its complementary form)
if and only if the variable does not change its value in the group.

● All the identified product terms are then summed to have a reduced SOP
form of the expression.

© 2014 - All rights are reserved by Kaushik Patra 12

 K-Map Reduction Examples

Y Y
0 1 0 1
X X
0 1 0 1
1 1 0 1
0

2 3 2 3
1 1
1 1

Two variable K-Map Two variable K-Map

F(X,Y) = Σm(1,2) = XY' + X'Y

F(X,Y) = Σm(0,1,3) = X' + Y

13

● For Boolean expression F(X,Y) = Σm(0,1,3) there are two groups chosen.
● The boxes in the table for the minterm 0 and 1 are horizontally
adjacent hence they are grouped together. For this group, only
variable X does not vary and stays at value 0. This gives the term X'
in the reduced expression.
● The boxes in the table for the minterm 1 and 3 are vertically adjacent,
hence they also group together. For this group variable Y does not
vary, but X does. The variable Y stays at 1. This gives the term Y in
the reduced expression.
● Combining two groups, we get X'+Y as the reduced expression for the
given Boolean expression.

● For the Boolean expression F(X,Y) = Σm(1,2), there are also two groups
can be chosen. However these two groups are singular group (meaning
that it contains only one 1 in the group).
● The box for minterm 2 gives the term XY' in the reduced expression
sine both X and Y remains constant at 1 and 0 respectively.
● The box for minterm 1 gives the term X'Y in the reduced expression
sine both X and Y remains constant at 0 and 1 respectively.
● Combining two groups we get the reduced (nothing reduced here in
this case) expression as X'Y + XY'

© 2014 - All rights are reserved by Kaushik Patra 13

 K-Map Reduction Examples
YZ YZ 11
00 01 11 10 00 01 10
X X
0 1 3 2 0 1 3 2
0 1 1 1 1 0 1 1

4 5 7 6 4 5 7 6
1 1
1 1 1 1 1

F(X,Y,Z) = Σm(0,1,2,3,4,5) = X' + Y' F(X,Y,Z) = Σm(0,2,4,5,6) = XY' + Z'

YZ 11
00 01 10
This red group is X
0 1 3 2
not taken as it
0 1 1
does not include
any unique 1s
4 5 7 6
compare to union 1
of other groups 1 1 1

14
F(X,Y,Z) = Σm(1,3,4,5,6) = XY' + X'Z + XZ'

● For the expression F(X,Y,Z) = Σm(0,1,2,3,4,5) the following groups are chosen.
● All the boxes for minterms 0,1,3,2 are adjacent and X remains constant at 0.
Hence the term X'.
● All the boxes for minterms 0,1,4,5 are adjacent and Y remain constant at 0.
Hence the term Y'.
● Combining both the term, we get reduced expression as X' + Y'

● For the expression F(X,Y,Z) = Σm(0,2,4,5,6) the following groups are chosen.
● All the boxes for minterms 0,2,4,6 are adjacent (outer two columns are adjacent
to each other) and Z remains constant at 0. Hence the term Z'.
● All the boxes for minterms 4,5 are adjacent and both X and Y remain constant at
1 and 0 respectively. Hence the term XY'.
● Combining both the term, we get reduced expression as XY' + Z'

● For the expression F(X,Y,Z) = Σm(1,3,4,5,6) the following groups are chosen.
● All the boxes for minterms 1,3 are adjacent. In this group X and Z remains
constant at 0 and 1 respectively. Hence the term X'Z.
● All the boxes for minterms 4,5 are adjacent. In this group X and Y remain
constant at 1 and 0 respectively. Hence the term XY'.
● All the boxes for minterms 4,6 are adjacent (outer two columns are adjacent to
each other. In this group X and Z are constant at 1 and 0. Hence the term XZ'
● Combining both the term, we get reduced expression as XY'+X'Z+XZ'
● The minterms 1,5 also could be grouped, but they are not including any unique
minterms that is not included in other group. Hence this group was not
considered. If it had been considered, then we could drop the group 4.5 and
the reduced expression would be X'Z + XY' + Y'Z.
© 2014 - All rights are reserved by Kaushik Patra 14
 K-Map Reduction Examples
YZ 11
00 01 10
WX
0 1 3 2
00 1 1 1

4 5 7 6
01 1 1 1

12 13 15 14
11
1 1

8 9 11 10
10 1 1 1

F(W,X,Y,Z) = Σm(0,1,2,4,5,6,8,9,10,12,13) = Y' + W'Z' + X'Z' 15

● In the Boolean expression F(W,X,Y,Z) = Σm(0,1,2,4,5,6,8,9,10,12,13)

the groups are chosen as following.
● The minterms 0,1,4,5,12,13,8,9 are adjacent to each other. For this
group only Y remains constant at 0 giving the term Y'.
● The minterms 0,2,4,6 are adjacent (since the boundary columns are
adjacent). For this group both W and Z are constant at 0 giving the
term W'Z'.
● The minterms 0,2,8,10 are adjacent since the boundary rows and
columns are adjacent to each other. For this group both X and Z are
constant at 0 giving the term X'Z'.
● Combining three individual product terms we get he SOP expression
as Y'+W'Z'+X'Z'.

© 2014 - All rights are reserved by Kaushik Patra 15

 Review of K-Map Grouping
YZ 11
00 01 10
WX
0 1 3 2
00 1 1 1

4 5 7 6
01 1 1 1

12 13 15 14
11
1 1

8 9 11 10
10 1 1 1

F(W,X,Y,Z) = Σm(0,1,2,4,5,6,8,9,10,12,13) = Y' + W'Z' + X'Z' 16

● We group adjacent to reduce the product term. Grouping two adjacent

term means that they have only one variable changing and hence can
be reduced to a new product term by eliminating the variable literal.
For example grouping 0 and 2 minterms means (W'X'Y'Z' + W'X'YZ')
= W'X'Z'(Y'+Y) = W'X'Z'.1 = W'X'Z'. Similarly we can reduce 4 and 6
as W'XZ'. Now since 0 and 4 are adjacent and 2,6 are adjacent, then
we can group (0,4) and (2.6) into (0,4,2,6) which will have exactly one
variable literal. In this case it is X. So we can express the term as
(W'X'Z' + W'XZ') = W'Z'(X+X') = W'Z'.1 = W'Z'.

© 2014 - All rights are reserved by Kaushik Patra 16

 Review of K-Map Grouping
YZ 11
00 01 10
WX
0 1 3 2
00 1 1 1

4 5 7 6
01 1 1 1

12 13 15 14
11
1 1

8 9 11 10
10 1 1 1

F(W,X,Y,Z) = Σm(0,1,2,4,5,6,8,9,10,12,13) = Y' + W'Z' + X'Z' 17

● We can group two mutually exclusive adjacent groups with same number of
minterms. This means, we group two adjacent groups with one minterm, which
will give a group of 2 minterms. Then we can group two adjacent groups with 2
minterms each to for a group of 4 minterms and so on. For example, initial group
can be (0,1) (4,5) (12,13) (8.9). Then we can group ((0,1) (4,5)) and ((12, 13)
(8,9)) because
● Boundary element in the individual group has exact one adjacent to another
boundary element in other group.
● For example of ((0,1) (4,5)), 0 is adjacent to 4 and 1 is adjacent to 5 hence
these two groups can be merged into one group of (0,1,4,5).
● Each group has exact same number of minterms.
● Both groups are mutually exclusive.

● Now we can also merge group (0,1,4,5) and (12,13,8,9) because boundary elements
(4,5) is adjacent to boundary elements (12,13) in other group. Also each group
has exactly same number of elements and they are mutually exclusive. Hence
they can be merged into (0,1,4,5,12,13,8,9).

● This is why we group minterms into power of 2, like 1, 2, 4, 8, etc. number of

minterms in a group.

© 2014 - All rights are reserved by Kaushik Patra 17

 Review of K-Map Grouping
YZ 11
00 01 10
WX
0 1 3 2
00 1 1 1

4 5 7 6
01 1 1 1

12 13 15 14
Can not 11
1 1
group
with the
other 8 9 11 10
group of 10 1 1 1
(0,1,4,5)

F(W,X,Y,Z) = Σm(0,1,2,4,5,6,8,9,10,12,13) = Y' + W'Z' + X'Z' 18

● We need to merge mutually exclusive group because otherwise it will not

result in any further reduction. For example, (0,1,4,5) can not be
merged with (4.5.12.13) and give a group (0,1,4,5,12,13). First of all
this will result in a group cardinal of 6 which is not a power of 2.
● Group (0.1.4.5) gives a reduced term as W'Z' and (4,5,12,13) gives
XZ'. Grouping them together will give a term W'Z' + XZ' which can
not be reduced further. This happened because we did not group the
minterms which are differing from each other with respect to
change in only one variable. In this case, Z' remains same, but W'
suddenly disappeared in other term and a new literal of X came in.
With respect to row indices, grouping (0,1,4,5) and (4,5,12,13)
means both W and X are varying in this grouping, which contradicts
the grouping rule where only one variable can change its value.

© 2014 - All rights are reserved by Kaushik Patra 18

 Map Manipulation
● How do we choose the groups for reduction?

● Let's define the following terminology.

– An implicant is any product term of a SOP form for
expression.
– A prime implicant is an product term which can not be
covered by any other more general (reduced term – terms
with fewer literals).
– An essential prime implicant is an prime implicant that
covers an output of the function in such a way the no other
combination of prime implicant can cover.

19

● For a boolean function F, if a product term P evaluated to be 1 will also

make F as 1 them P is an implicant of F. In this case, we say P implies
F.
● For example, F(X,Y,Z,W) = XY + YZ + W has may implicant, such as
XY, YZ, XYZ. XYZW, and many mores.

● If we removed any term from a prime implicant, it can no longer stays as

prime implicant. For the above example, ZY, YZ and W are the prime
implicants. XYZ and XYZW are not prime implicant since can be
further generalized to XY, YZ, and W.

● If a prime implicant contains a minterm that no other prime implicant

has, the prime implicant is called essential prime implicant. In this
example XY and YZ and W are the essential prime implicants.

● An essential prime implicant contains a minterm that no other prime

implicant has.

● The sum of all prime implicant is called complete sum or minimal

covering sum.

© 2014 - All rights are reserved by Kaushik Patra 19

 Map Manipulation
● Identifying essential and non-essential prime implicants in the
K-map.

YZ 00 01 11 10
WX
0 1 3 2
00 1 1
Essential prime
Non essential 4 5 7 6
01 1 1 1 1 implicants
prime implicant
12 13 15 14
11
1 1
8 9 11 10
10

F(W,X,Y,Z) = Σm(1,3,4,5,6,7,12,14) = W'Z + XZ' + W'X = W'Z + XZ'

20

● The expression F(W,X,Y,Z) = Σm(1,3,4,5,6,7,12,14) contains prime

implicants of W'Z, XZ' and W'X. Among these W'Z is one essential
prime implicant since it contains minterms 1,3 which are not included in
any other prime implicants. Similarly XZ' is also an essential prime
implicant since it contains the minterms 12, 14 which are not included in
any other prime implicants. However the term W'X is not an essential
prime implicant since all the minterms it covers (4,5,6,7) are included
into other prime implicants.

● The essential prime implicant should contain at least one minterm which
is not included into any other prime implicants. If it is not the case, then
we call the prime implicant as non-essential prime implicant.

● This example also demonstrate the K-map based explanation of the

Consensus theorem. In the consensus theorem, the non-essential prime
implicant gets dropped from the expression.

© 2014 - All rights are reserved by Kaushik Patra 20

 Map Manipulation
● There can be a situation where selecting one essential prime
implicant will make another prime implicant non-essential.

YZ 00 01 11 10
WX
0 1 3 2
00 1
4 5 7 6
01 1
12 13 15 14
11
1 1 1
8 9 11 10
10 1 1

F(W,X,Y,Z) = Σm(0,5,10, 11, 12,13,15) = W'X'Y'Z' + XY'Z + WX'Y + WXZ

OR
W'X'Y'Z' + XY'Z + WX'Y + WYZ 21

● In this example of F(W,X,Y,Z) = Σm(0,5,12,13,15,11,10) we have three

distinct essential prime implicants – W'X'Y'Z' , XY'Z , WX'Y. Now to
include the minterm 15 exclusively it can be grouped into either WXZ or
WYZ. In this case, if we use WXZ in the expression then WYZ does not
remain as an essential prime implicant. Similarly if we use WYZ in the
expression then WXZ is no longer an essential prime implicant.

© 2014 - All rights are reserved by Kaushik Patra 21

 Map Manipulation
● Selection rule states that select non-overlapping essential
prime implicants as much as possible.
YZ 00 01 11 10
WX
0 1 3 2
00 1 1 1
4 5 7 6
01 1 1
12 13 15 14
11
1 1
8 9 11 10
10 1 1

F(W,X,Y,Z) = Σm(0,1,2,4,5,10,11,13,15) = W'Y' + WXZ + WX'Y + W'X'Z'

or
W'Y' + XY'Z + WYZ + X'YZ'

but not
22
W'Y' + XY'Z + WYZ + WX'Y + W'X'Z'

● In this example of F(W,X,Y,Z) = Σm(0,1,2,4,5,10,11,13,15) \

● We can group the minterms as (0,1,4,5),(13,16),(10,11),(0,2) with one
overlap of minterm 0 (marked in bold in group). This gives 4
product terms in the SOP form of this expression.
● We also can group minterms as (0,1,4,5),(5,13),(11,15),(2,12) with one
overlap of minterm 5 (marked in bold in group). This also gives 4
product terms in the SOP form of this expression.
● We can also group (0,1,4,5),(5,12),(15,11),(10,11),(0,2). Even though
all of them are essential prime implicants, it gives 3 overlaps and
produce 5 product terms in SOP form of this expression which is
clearly not most reduced form that can be obtained for this
expression.

© 2014 - All rights are reserved by Kaushik Patra 22

 Use of don't care term
● Treat don't care term (x) as 1 if required to minimize the
expression better.
YZ 00 01 11 10
WX
0 1 3 2
00 x 1 1 x
4 5 7 6
01 x 1
12 13 15 14
11
1
8 9 11 10
10 1

F(W,X,Y,Z) = Σm(1,3,7,11,15) + d(0,2,5) = YZ + W'X'Z

or
YZ + W'X'
or
YZ + W'Z 23

● Don't care terms are the minterms which does not dictate overall function value.
Even if the minterm becomes 1, it does not make the function 1. Even if the
minterm becomes 0, it does not make the function 0.

● Don't care terms in an expression are denoted as d(list of minterms). For example,
d(0,2,5) means the minterms 0,2, and 5 are don't care terms for the expression
F(W,X,Y,Z) = Σm(1,3,7,11,15) + d(0,2,5). In this case, we can choose not to involve
don't care terms and have a reduced from of YZ+W'X'Z).

● Don't care terms can be involved in K-map based minimization because we can
include them in the reduced form since they do not affect the final value of the
function. In the above example, if we include minterms 0 and 2 then the reduced
form will be (YZ+W'X'). If we include minterm 5, the the reduced form will be
(YZ+W'Z).

● Inclusion of don't care term reduce the product terms. In the above example, before
inclusion of the don't care terms, the minimization process gave two product term
with 2 and 3 variable involved. By including the don't care terms, we were able to
express the function as two product terms with 2 variables each.

● Though represented by 'x', “don't care” and “unknown” terminology are different
from each other. They convey totally different ideas. Don't care term is applicable in
context of minterm, where as the unknown value is applicable to denote a value (not
known) for a Boolean variable.

© 2014 - All rights are reserved by Kaushik Patra 23

 CS147 - Lecture 05

Kaushik Patra
([email protected])

24

● Standard Forms

● Karnaugh Map

[ Chapter 2 (2-3, 2-4) of Logic & Computer Design Fundamentals, 4th Edition,
M. Morris Mano, Charles R. Kime ]

© 2014 - All rights are reserved by Kaushik Patra 24