Chapter 1 Number Systems and Codes

Chapter 1 1
 Number Systems (1)

• Positional Notation
N = (an-1an-2 ... a1a0 . a-1a-2 ... a-m)r (1.1)
where . = radix point
r = radix or base
n = number of integer digits to the left of the radix point
m = number of fractional digits to the right of the radix point
an-1 = most significant digit (MSD)
a-m = least significant digit (LSD)

• Polynomial Notation (Series Representation)

N = an-1 x rn-1 + an-2 x rn-2 + ... + a0 x r0 + a-1 x r-1 ... + a-m x r-m
n 1
=  ai r
i
(1.2)
i  m

• N = (251.41)10 = 2 x 102 + 5 x 101 + 1 x 100 + 4 x 10-1 + 1 x 10-2

Chapter 1 2
 Number Systems (2)

• Binary numbers
– Digits = {0, 1}
– (11010.11)2 = 1 x 24 + 1 x 23 + 0 x 22 + 1 x 21 + 0 x 20 + 1 x 2-1 + 1 x 2-2
= (26.75)10
– 1 K (kilo) = 210 = 1,024, 1M (mega) = 220 = 1,048,576,
1G (giga) = 230 = 1,073,741,824

• Octal numbers
– Digits = {0, 1, 2, 3, 4, 5, 6, 7}
– (127.4)8 = 1 x 82 + 2 x 81 + 7 x 80 + 4 x 8-1 = (87.5)10

• Hexadecimal numbers
– Digits = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F}
– (B65F)16 = 11 x 163 + 6 x 162 + 5 x 161 + 15 x 160 = (46,687)10

Chapter 1 3
 Number Systems (3)

• Important Number Systems (Table 1.1)

Decimal Binary Octal Hexadecimal
0 0 0 0
1 1 1 1
2 10 2 2
3 11 3 3
100 4 4
5 101 5 5
6 110 6 6
7 111 7 7
8 1000 10 8
9 1001 11 9
10 1010 12 A
11 1011 13 B
12 1100 14 C
13 1101 15 D
14 1110 16 E
15 1111 1 F
16 10000 20 10
Chapter 1 4
 Arithmetic (1)

• Binary Arithmetic
– Addition

111011 Carries
101011 Augend
+ 11001 Addend
1000100

– Subtraction

0 1 10 0 10 Borrows
1 0 0 1 0 1 Minuend
- 1 1 0 1 1 Subtrahend
1 0 1 0

Chapter 1 5
 Arithmetic (2)

– Multiplication Division

1 1 0 1 0 Multiplicand 110 Quotient

x 1 0 1 0 Multiplier Divider 1001 111101 Dividend
00000 1001
1100
11010 1001
00000 111 Remainder
11010
1 0 0 0 0 0 1 0 0 Product

Chapter 1 6
 Arithmetic (3)

• Octal Arithmetic (Use Table 1.4)

– Addition

1 1 1 Carries
5 4 7 1 Augend
+ 3 7 5 4 Addend
11445 Sum

– Subtraction

6 10 4 10 Borrows
7 4 5 1 Minuend
- 5 6 4 3 Subtrahend
1 6 0 6 Difference

Chapter 1 7
 Arithmetic (4)

– Multiplication Division

326 Multiplicand 114 Quotient

x 67 Multiplier
Divider 63 7514 Dividend
63
2732 Partial products 114
2404 63
26772 Product 364
314
50 Remainder

Chapter 1 8
 Arithmetic (5)
• Hexadecimal Arithmetic (Use Table 1.5)
– Addition

1 0 1 1 Carries
5 B A 9 Augend
+ D 0 5 8 Addend
1 2 C 0 1 Sum

– Subtraction

9 10 A 10 Borrows
A 5 B 9 Minuend
+ 5 8 0 D Subtrahend
4 D A C Difference

• Series Substitution Method

– Expanded form of polynomial representation:
N = an-1rn-1 + … + a0r0 + a-1r-1 + … + a-mr-m (1.3)
– Conversion Procedure (base A to base B)
• Represent the number in base A in the format of Eq. 1.3.
• Evaluate the series using base B arithmetic.
– Examples:
– (11010)2  ( ? )10

– (627)8  ( ? )10

– (11110.101)2  ( ? )8

– (2AD.42)16  ( ? )10

Chapter 1 9
 Arithmetic (6)

– Multiplication Division

B9A5 Multiplicand 79B Quotient

x D50 Multiplier
Divider B9 57F6D Dividend
50F
3A0390 Partial products 706
96D61 681
9A76490 Product 85D
7F3
6A Remainder

Chapter 1 10
 Base Conversion (1)

• Series Substitution Method

– Expanded form of polynomial representation:
N = an-1rn-1 + … + a0r0 + a-1r-1 + … + a-mr-m (1.3)
– Conversation Procedure (base A to base B)
• Represent the number in base A in the format of Eq. 1.3.
• Evaluate the series using base B arithmetic.
– Examples:
• (11010)2 ( ? )10
N = 10
= (16)10 + (8)10 + 0 + (2)10 + 0
= (26)10
• (627)8  ( ? )10
N = 6
= (384)10 + (16)10 + (7)10
= (407)10

Chapter 1 11
 Base Conversion (2)

• Radix Divide Method

– Used to convert the integer in base A to the equivalent base B integer.
– Underlying theory:
• (NI)A = bn-1Bn-1 + … + b0B0 (1.4)
Here, bi’s represents the digits of (NI)B in base A.
• NI bn-1Bn-1 + … + b1B1 + b0B0 ) / B
= (Quotient Q1: bn-1Bn-2 + … + b1B0 ) + (Remainder R0: b0)
• In general, (bi)A is the remainder Ri when Qi is divided by (B)A.

– Conversion Procedure
1. Divide (NI)B by (B)A, producing Q1 and R0. R0 is the least significant
digit, d0, of the result.
2. Compute di, for i = 1 … n - 1, by dividing Qi by (B)A, producing Qi+1
and Ri, which represents di.
3. Stop when Qi+1 = 0.
Chapter 1 12
 Base Conversion (3)

– Examples
• (315)10 = (473)8

8 315 3 LSD
8 39 7
8 4 4 MSD
0

• (315)10 = (13B)16

16 315 B LSD
16 19 3
16 1 1 MSD
0

Chapter 1 13
 Base Conversion (4)

• Radix Multiply Method

– Used to convert fractions.
– Underlying theory:
• (NF)A = b-1B-1 + b-2B-2 + … + b-mB-m (1.5)
Here, (NF)A is a fraction in base A and bi’s are the digits of (NF)B in base
A.
• B NF = B (b-1B-1 + b-2B-2 + … + b-mB-m )
= (Integer I-1: b-1) + (Fraction F-2: b-2B-1 + … + b-mB-(m-1))
• In general, (bi)A is the integer part I-i, of the product of F-(i+1) (BA).

– Conversion Procedure
1. Let F-1 = (NF)A.
2. Compute digits (b-i)A, for i = 1 … m, by multiplying Fi by (B)A,
producing integer I-i, which represents (b-i)A, and fraction F-(i+1).
3. Convert each digits (b-i)A to base B.
Chapter 1 14
 Base Conversion (5)

– Examples
• (0.479)10 = (0.3651…)8
MSD 3.832 0.479  8
6.656 0.832  8
5.248 0.656  8
LSD 1.984 0.248  8
…

• (0.479)10 = (0.0111…)2
MSD 0.9580 0.479  2
1.9160 0.9580  2
1.8320 0.9160  2
LSD 1.6640 0.8320  2
…

Chapter 1 15
 Base Conversion (6)

• General Conversion Algorithm

• Algorithm 1.1
To convert a number N from base A to base B, use
(a) the series substitution method with base B arithmetic, or
(b) the radix divide or multiply method with base A arithmetic.

• Algorithm 1.2
To convert a number N from base A to base B, use
(a) the series substitution method with base 10 arithmetic to convert N
from base A to base 10, and
(b) the radix divide or multiply method with decimal arithmetic to convert
N from base 10 to base B.

• Algorithm 1.2 is longer, but easier and less error prone.

Chapter 1 16
 Base Conversion (7)

• Example
(18.6)9 = ( ? )11

(a) Convert to base 10 using series substitution method:

N10 = 1  91 + 8  90 + 6  9-1
= 9 + 8 + 0.666…
= (17.666…)10
(b) Convert from base 10 to base 11 using radix divide and multiply
method:
7.326 0.666  1111 17
.
6
3.586 0.326  11 11 1 1
6.446 0.586  11 0

N11 = (16.736 …)11

Chapter 1 17
 Base Conversion (8)

• When B = Ak
• Algorithm 1.3
(a) To convert a number N from base A to base B when B = Ak and k is a
positive integer, group the digits of N in groups of k digits in both directions
from the radix point and then replace each group with the equivalent digit in
base B
(b) To convert a number N from base B to base A when B = Ak and k is a
positive integer, replace each base B digit in N with the equivalent k digits in
base A.

• Examples
– (001 010 111. 100)2 = (127.4)8 (group bits by 3)
– (1011 0110 0101 1111)2 = (B65F)16 (group bits by 4)

Chapter 1 18
 Signed Number Representation

• Signed Magnitude Method

– N =  (an-1 ... a0.a-1 ... a-m)r is represented as
N = (san-1 ... a0.a-1 ... a-m)rsm, (1.6)
where s = 0 if N is positive and s = r -1 otherwise.
– N = -(15)10
– In binary: N = -(15)10 = -(1111)2 = (1, 1111)2sm
– In decimal: N = -(15)10 = (9, 15)10sm

• Complementary Number Systems

– Radix complements (r's complements)
[N]r = rn - (N)r (1.7)
where n is the number of digits in (N)r.
– Positive full scale: rn-1 - 1
– Negative full scale: -rn - 1
– Diminished radix complements (r-1’s complements)
[N]r-1 = rn - (N)r - 1

Chapter 1 19
 Radix Complement Number Systems (1)

• Two's complement of (N)2 = (101001)2

[N]2 = 26 - (101001)2 = (1000000)2 - (101001)2 = (010111)2

• (N)2 + [N]2 = (101001)2 + (010111)2 = (1000000)2

If we discard the carry, (N)2 + [N]2 = 0.
Hence, [N]2 can be used to represent -(N)2.
• [ [N]2 ]2 = [(010111)2]2 = (1000000)2 - (010111)2 = (101001)2 = (N)2.

• Two's complement of (N)2 = (1010)2 for n = 6

[N]2 = (1000000)2 - (1010)2 = (110110)2.

• Ten's complement of (N)10 = (72092)10

[N]10 = (100000)10 - (72092)10 = (27908)10.

Chapter 1 20
 Radix Complement Number Systems (2)

• Algorithm 1.4 Find [N]r given (N)r .

– Copy the digits of N, beginning with the LSD and proceeding toward the
MSD until the first nonzero digit, ai, has been reached
– Replace ai with r - ai .
– Replace each remaining digit aj , of N by (r - 1) - aj until the MSD has been
replaced.

• Example: 10's complement of (56700)10 is (43300)10

• Example: 2's complement of (10100)2 is (01100)2.
• Example: 2’s complement of N = (10110)2 for n = 8.
– Put three zeros in the MSB position and apply algorithm 1.4
– N = 00010110
– [N]2 = (11101010)2

• The same rule applies to the case when N contains a radix point.
Chapter 1 21
 Radix Complement Number Systems (3)

• Algorithm 1.5 Find [N]r given (N)r .

– First replace each digit, ak , of (N)r by (r - 1) - ak and then add 1 to the
resultant.
• For binary numbers (r = 2), complement each digit and add 1 to the result.

• Example: Find 2’s complement of N = (01100101)2 .

N = 01100101
10011010 Complement the bits
+1 Add 1
[N]2 = (10011011)10
• Example: Find 10’s complement of N = (40960)10
N = 40960
59039 Complement the bits
+1 Add 1
[N]2 = (59040)10
Chapter 1 22
 Radix Complement Number Systems (4)

• Two's complement number system (See Table 1.6):

– Positive number :
• N = +(an-2, ..., a0)2 = (0, an-2, ..., a0)2cns,
n 1
where 0  N  2  1.
– Negative number:
• N = (an-1, an-2, ..., a0)2
• -N = [an-1, an-2, ..., a0]2 (two's complement of N),
n 1
where  1  N  2 .

– Example: Two's complement number system representation of  (N)2

when (N)2 = (1011001)2 for n = 8:
• +(N)2 = (0, 1011001)2cns
• -(N)2 = [+(N)2]2 = [0, 1011001]2 = (1, 0100111)2cns

Chapter 1 23
 Radix Complement Number Systems (5)

• Example: Two's complement number system representation of -(18)10 , n = 8:

– +(18)10 = (0, 0010010)2cns
– -(18)10 = [0, 0010010]2 = (1, 1101110)2cns

• Example: Decimal representation of N = (1, 1101000)2cns

– N = (1, 1101000)2cns = -[1, 1101000]2 = -(0, 0011000)2cns = -(24)2 .

Chapter 1 24
 Radix Complement Arithmetic (1)

• Radix complement number systems are used to convert subtraction to addition,

which reduces hardware requirements (only adders are needed).
• A - B = A + (-B) (add r’s complement of B to A)

• Range of numbers in two’s complement number system:

 2 n 1  N  2 n 1  1 , where n is the number of bits.
• 2n-1 -1 = (0, 11 ... 1)2cns and -2n-1 = (1, 00 ... 0)2cns
• If the result of an operation falls outside the range, an overflow condition is
said to occur and the result is not valid.

• Consider three cases:

– A = B + C,
– A = B - C,
– A = - B - C,
(where B  and C .)
Chapter 1 25
 Radix Complement Arithmetic (2)

• Case 1: A = B + C
– (A)2 = (B)2 + (C)2
– If A > 2n-1 -1 (overflow), it is detected by the nth bit, which is set to 1.

– Example: (7)10 + (4)10 = ? using 5-bit two’s complement arithmetic.

• + (7)10 = +(0111)2 = (0, 0111)2cns
• + (4)10 = +(0100)2 = (0, 0100)2cns
• (0, 0111)2cns + (0, 0100)2cns = (0, 1011)2cns = +(1011)2 = +(11)10
• No overflow.
– Example: (9)10 + (8)10 = ?
• + (9)10 = +(1001)2 = (0, 1001)2cns
• + (8)10 = +(1000)2 = (0, 1000)2cns
• (0, 1001)2cns + (0, 1000)2cns = (1, 0001)2cns (overflow)

Chapter 1 26
 Radix Complement Arithmetic (3)

• Case 2: A = B - C
– A = (B)2 + (-(C)2) = (B)2 + [C]2 = (B)2 + 2n - (C)2 = 2n + (B - C)2
– If B  C, then A  2n and the carry is discarded.
– So, (A)2 = (B)2 + [C]|carry discarded
– If B < C, then A = 2n - (C - B)2 = [C - B]2 or A = -(C - B)2 (no carry in this
case).
– No overflow for Case 2.

– Example: (14)10 - (9)10 = ?

• Perform (14)10 + (-(9)10)
• (14)10 = +(1110)2 = (0, 1110)2cns
• -(9)10 = -(1001)2 = (1, 0111)2cns
• (14)10 - (9)10 = (0, 1110)2cns + (1, 0111)2cns = (0, 0101)2cns + carry
= +(0101)2 = +(5)10

Chapter 1 27
 Radix Complement Arithmetic (4)

– Example: (9)10 - (14)10 = ?

• Perform (9)10 + (-(14)10)
• (9)10 = +(1001)2 = (0, 1001)2cns
• -(14)10 = -(1110)2 = (1, 0010)2cns
• (9)10 - (14)10 = (0, 1001)2cns + (1, 0010)2cns = (1, 1011)2cns
= -(0101)2 = -(5)10

– Example: (0, 0100)2cns - (1, 0110)2cns = ?

• Perform (0, 0100)2cns + (- (1, 0110)2cns)
• - (1, 0110)2cns = two’s complement of (1,0110)2cns
= (0, 1010)2cns
• (0, 0100)2cns - (1, 0110)2cns = (0, 0100)2cns + (0, 1010)2cns
= (0, 1110)2cns = +(1110)2 = +(14)10
• +(4)10 - (-(10)10) = +(14)10
Chapter 1 28
 Radix Complement Arithmetic (5)

• Case 3: A = -B - C
– A = [B]2 + [C]2 = 2n - (B)2 + 2n - (C)2 = 2n + 2n - (B + C)2 = 2n + [B + C]2
– The carry bit (2n) is discarded.
– An overflow can occur, in which case the sign bit is 0.

– Example: -(7)10 - (8)10 = ?

• Perform (-(7)10) + (-(8)10)
• -(7)10 = -(0111)2 = (1, 1001)2cns , -(8)10 = -(1000)2 = (1, 1000)2cns
• -(7)10 - (8)10 = (1, 1001)2cns + (1, 1000)2cns = (1, 0001)2cns + carry
= -(1111)2 = -(15)10
– Example: -(12)10 - (5)10 = ?
• Perform (-(12)10) + (-(5)10)
• -(12)10 = -(1100)2 = (1, 0100)2cns , -(5)10 = -(0101)2 = (1, 1011)2cns
• -(7)10 - (8)10 = (1, 0100)2cns + (1, 1011)2cns = (0, 1111)2cns + carry
• Overflow, because the sign bit is 0.

Chapter 1 29
 Radix Complement Arithmetic (6)

• Example: A = (25)10 and B = -(46)10

– A = +(25)10 = (0, 0011001)2cns , -A = (1, 1100111)2cns
– B = -(46)10 = -(0, 0101110)2 = (1, 1010010)2cns , -B = (0, 0101110)2cns

– A + B = (0, 0011001)2cns + (1, 1010010)2cns = (1, 1101011)2cns = -(21)10

– A - B = A + (-B) = (0, 0011001)2cns + (0, 0101110)2cns
= (0, 1000111)2cns = +(71)10
– B - A = B + (-A) = (1, 1010010)2cns + (1, 1100111)2cns
= (1, 0111001)2cns + carry = -(0, 1000111)2cns = -(71)10
– -A - B = (-A) + (-B) = (1, 1100111)2cns + (0, 0101110)2cns
= (0, 0010101)2cns + carry = +(21)10
– Note: Carry bit is discarded.

Chapter 1 30
 Radix Complement Arithmetic (7)

• Summary

Case Carry Sign Bit Condition Overflow ?

B+C 0 0 B + C  2n-1 - 1 No
0 1 B + C > 2n-1 - 1 Yes
B-C 1 0 B  C No
0 1 B>C No
-B - C 1 1 -(B + C) -2n-1 No
1 0 -(B + C) < -2n-1 Yes

• When numbers are represented using two’s complement number system:

– Addition: Add two numbers.
– Subtraction: Add two’s complement of the subtrahend to the minuend.
– Carry bit is discarded, and overflow is detected as shown above.

– Radix complement arithmetic can be used for any radix.

Chapter 1 31
 Diminished Radix Complement Number systems (1)

• Diminished radix complement [N]r-1 of a number (N)r is:

[N]r-1 = rn - (N)r - 1 (1.10)

• One’s complement (r = 2):

[N]2-1 = 2n - (N)2 - 1 (1.11)

• Example: One’s complement of (01100101)2

[N]2-1 = 28 - (01100101)2 - 1
= (100000000)2 - (01100101)2 - (00000001)2
= (10011011)2 - (00000001)2
= (10011010)2

Chapter 1 32
 Diminished Radix Complement Number systems (2)

• Example: Nine’s complement of (40960)

[N]2-1 = 105 - (40960)10 - 1
= (100000)10 - (40960)10 - (00001)10
= (59040)10 - (00001)10
= (59039)10

• Algorithm 1.6 Find [N]r-1 given (N)r .

Replace each digit ai of (N)r by r - 1 - a. Note that when r = 2, this simplifies
to complementing each individual bit of (N)r .

• Radix complement and diminished radix complement of a number (N):

[N]r = [N]r-1 + 1 (1.12)

Chapter 1 33
 Diminished Radix Complement Arithmetic (1)

• Operands are represented using diminished radix complement number system.

• The carry, if any, is added to the result (end-around carry).

• Example: Add +(1001)2 and -(0100)2 .

One’s complement of +(1001) = 01001
One’s complement of -(0100) = 11011
01001 + 11011 = 100100 (carry)
Add the carry to the result: correct result is 00101.

• Example: Add +(1001)2 and -(1111)2 .

One’s complement of +(1001) = 01001
One’s complement of -(1111) = 10000
01001 + 10000 = 11001 (no carry, so this is the correct result).

Chapter 1 34
 Diminished Radix Complement Arithmetic (2)

• Example: Add -(1001)2 and -(0011)2 .

One’s complement of the operands are: 10110 and 11100
10110 + 11100 = 110010 (carry)
Correct result is 10010 + 1 = 10011.

• Example: Add +(75)10 and -(21)10 .

Nine’s complements of the operands are: 075 and 978
075 + 978 = 1053 (carry)
Correct result is 053 + 1 = 054

• Example: Add +(21)10 and -(75)10 .

Nine’s complements of the operands are: 021 and 924
021 + 924 = 945 (no carry, so this is the correct result).

Chapter 1 35
 Computer Codes (1)

• Code is a systematic use of a given set of symbols for representing

information.
• Example: Traffic light (Red: stop, Yellow: caution, Blue: go).

• Numeric Codes
– To represent numbers.
– Fixed-point and floating-point number.

• Fixed-point Numbers
– Used for signed integers or integer fractions.
– Sign magnitude, two’s complement, or one’s complement systems are
used.
– Integer: (Sign bit) + (Magnitude) + (Implied radix point)
– Fraction: (Sign bit) + (Implied radix point) + (Magnitude)

Chapter 1 36
 Computer Codes (2)

• Excess or Biased Representation

– An excess-K representation of a code C: Add K to each code word C.
– Frequently used for the exponents of floating-point numbers.
– Excess-8 representation of 4-bit two’s complement code: Table 1.8

Chapter 1 37
 Floating Point Numbers (1)

• N = M  rE, where (1.13)

– M (mantissa or significand) is a significant digits of N
– E (exponent or characteristic) is an integer exponent.

• In general, N =  (an-1 ... a0 .a-1 ... a-m)r is represented by

– N =  (.an-1 ... a-m)r  rn

• M is usually represented in sign magnitude:

– M = (SM.an-1 ... a-m)rsm , where (1.14)
– (.an-1 ... a-m)r represents the magnitude
– SM = (1)  (. a ... a ) (0: positive, 1: negative)
SM
n 1 m r
(1.15)

Chapter 1 38
 Floating Point Numbers (2)

• E is usually coded in excess-K two’s complement.

• K is called a bias and usually selected to be 2e-1 (e is the number of bits).
• So, biased E is:
– -2e-1 
 

• Excess-K form of E is written as: E = (be-1, be-2 ... b0)excess-K (1.16)

where be-1 is the sign bit.

• Combining Eqs. (1.14) and (1.16), we have

N = (SMbe-1be-2 ... b0an-1 ... a-m)r (1.17)
e 1
( 1) S M  (. an 1 ... a  m ) r  r (be1be2 ...b0 )  2
representing N = (1.18)

• The number 0 is represented by an all-zero word.

Chapter 1 39
 Floating Point Numbers (3)

• Multiple representations of a given number:

N = M rE (1.19)
= (M r)  rE+1 (1.20)
= (M  r)  rE-1 (1.21)

• Example: M = +(1101.0101)2
M = +(1101.0101)2
= (0.11010101)2 24 (1.22)
= (0.011010101)2 25 (1.23)
= (0.0011010101)2 26 (1.24)
…
• Normalization is used for a unique representation: mantissa has a nonzero
value in its MSD position.
• Eq. 1.22 gives the normalization representation of M.

Chapter 1 40
 Floating Point Numbers (4)

• Floating-point Number Formats

– Typical single-precision format

S M Exponent E Mantissa M

Sign of mantissa

– Typical extended-precision format

S M Exponent E Mantissa M (most significant part)

Mantissa M (least significant part)

Chapter 1 41
 Floating Point Numbers (5)

• Example: N = (101101.101)2, where n + m = 10 and e = 5. Assume that a

normalized sign magnitude fraction is used for M and that Excess-16 two’s
complement is used for E.

– N = (101101.101)2 = (0.101101101)2  26
– M = +(0.1011011010)2 = (0.1011011010)2sm

– E = +(6)10 = +(0110)2 = (00110)2cns

– Add the bias 16 = (10000)2 to E
E = 00110 + 10000 = 10110
So, E = (1, 0110)excess-16

– Combining M and E, we have

N = (0, 1, 0110, 1011011010)fp

Chapter 1 42
 Characters and Other Codes (1)

• To represent information as strings of alpha-numeric characters.

• Binary Coded Decimal (BCD)

– Used to represent the decimal digits 0 - 9.
– 4 bits are used.
– Each bit position has a weight associated with it (weighted code).
– Weights are: 8, 4, 2, and 1 from MSB to LSB (called 8-4-2-1 code).
– BCD Codes:
0: 0000 1: 0001 2: 0010 3: 0011 4: 0100
5: 0101 6: 0110 7: 0111 8: 1000 9: 1001
– Used to encode numbers for output to numerical displays
– Used in processors that perform decimal arithmetic.

– Example: (9750)10 = (1001011101010000)BCD

Chapter 1 43
 Characters and Other Codes (2)

• ASCII (American Standard Code for Information Interchange)

– Most widely used character code.
– See Table 1.11 for 7-bit ASCII code.
– The eighth bit is often used for error detection (parity bit)
– Example: ASCII code representation of the word Digital

Character Binary Code Hexadecimal Code

D 1000100 44
i 1101001 69
g 1100111 67
i 1101001 69
t 1110100 74
a 1100001 61
l 1101100 6C

Chapter 1 44
 Characters and Other Codes (3)

• Gray Code
– Cyclic code: A circular shifting of a code word produces another code
word.
– Gray code: A cyclic code with the property that two consecutive code
words differ in only 1 bit (the distance between the two code words is 1).
– Gray code for decimal numbers 0 - 15: See Table 1.12

Chapter 1 45
 Error Detection Codes and Correction Codes(1)

• An error: An incorrect value in one or more bits.

• Single error: An incorrect value in only one bit.
• Multiple error: One or more bits are incorrect.
• Errors are introduced by hardware failures, external interference (noise), or
other unwanted events.

• Error detection/correction code: Information is encoded in such a way that a

particular class of errors can be detected and/or corrected.

• Let I and J be n-bit binary information words

– w(I): the number of 1’s in I (weight)
– d(I, J): the number of bit positions in which I and J differ (distance)
• Example: I = (01101100) and J = (11000100)
– w(I) = 4 and w(J) = 3
– d(I, J) = 3.
Chapter 1 46
 Error Detection Codes and Correction Codes(2)

• General Properties
– Minimum distance, dmin, of a code C: for any two code words I and J in C,
d(I, J) dmin
– A code provides t error correction plus detection of s additional errors if
and only if the following inequality is satisfied.
2t + s + 1  dmin (1.25)
– Example:
• Single-error detection (SED): s = 1, t = 0, dmin = 2.
• Single-error correction (SEC): s = 0, t = 1, dmin = 3.
• Single-error correction and double-error detection (SEC and DED):
s = t = 1, dmin = 4.

Chapter 1 47
 Error Detection Codes and Correction Codes(3)

• Relationship between the minimum distance between code words and

the ability to detect and correct errors:

Error word

Valid code word

d min = 2 d min = 3
(a) SED (b) SEC or DED

d min = 4 d min = 5
(c) (SEC and DED) or TED (d) DEC, (SEC and 3ED), or 4ED

Chapter 1 48
 Error Detection Codes and Correction Codes(4)

• Simple Parity Code

– Concatenate (|) a parity bit, P, to each code word of C.

– Odd-parity code: w(P|C) is odd.

P Information bits
– Even-parity code: w(P|C) is even.
Parity bit

– Parity coding on magnetic tape:

01011000
0
1
0
1 Information
1 tracks
0
0
0
1 Parity track

Chapter 1 49
 Error Detection Codes and Correction Codes(5)

– Example: Odd-parity code for ASCII code characters:

Character ASCII Code Odd-parity Code

0 0110000 10110000
X 1011000 01011000
= 0111100 1111100
BEL 0000111 00000111

– Error detection: Check whether a code word has the correct parity.
– Single-error detection code (dmin = 2).

• Two-out-of-Five Code
– Each code word has exactly two 1’s and three 0’s.
– Detects single errors and multiple errors in adjacent bits.

Chapter 1 50
 Hamming Codes (1)

• Multiple check bits are employed.

• Each check bit is defined over (or covers) a subset of the information bits.
• Subsets overlap so that each information bit is in at least two subsets.
• dmin is equal to the weight of the minimum-weight nonzero code word.

• Hamming Code 1 (Table 1.14)

– dmin = 3, single error correction code.
– Let the set of all code words: C
an error word with single error: ce
the correct code word for the error word: c
then, d(ce,c) = 1 and d(ce, w) > 1 for all other w  C (see Table 1.15)
– So, a single error can be detected and corrected by finding out the code
word which differs in 1 bit position from the error word.

Chapter 1 51
 Hamming Codes (2)

– A code word consists of 4 information bits and 3 check bits:

c = (i3 i2 i1 i0 c2 c1 c0)
– Each check bit covers:
c 2 : i 3 , i 2 , i1 c1: i3, i2, i0 c0: i3, i1, i0
– This relationship is specified by the generating matrix, G:

1000111  1000 p11 p12 p13 

0100110 0100 p p p 
G  21 22 23  (1.26)
0010101 0010 p31 p32 p33 
   
 0001011  0001 p p p
41 42 43 

– Encoding of an information word i to produce a code word, c:

c = iG (1.27)

Chapter 1 52
 Hamming Codes (3)

– Decoding can be done using the parity-check matrix, H:

(1.28)  p11 p21 p31 p41100  1110100

   
H   p12 p22 p32 p42 010  1101010
 p13 p23 p33 p43 001  1011001

• H matrix is can be derived from G matrix.

– An n-tuple c is a code word generated by G if and only if
HcT = 0 (1.29)
– Let d be a data word corresponding to a code word c, which has been
corrupted by an error pattern e. Then
d = c + e (1.30)
– Decoding:
• Compute the syndrome, s, of d using H matrix.
• s tells the position of the erroneous bit.
Chapter 1 53
 Hamming Codes (4)

– Computation of the syndrome:

s = HdT (1.31)
= H(c + e)T
= HcT + HeT
= 0 + HeT
= HeT (1.32)
• Note: All computations are performed using modulo-2 arithmetic.

– See Table 1.16 for the syndromes and error patterns.

Chapter 1 54
 Hamming Codes (5)

• Hamming Code 2 (Table 1.14)

– dmin = 4, single error correction and double-error detection.
– The generator and parity-check matrices are:

10000111  01111000
01001110 11100100 
G  (1.33) H 
(1.34)
00101101 11010010 
   
00011011 10110001 
 

Odd-weight-column code:
• H matrix has an odd number of ones in each column.
• Example: Hamming Code 2.
• Has many properties; single-error correction, double-error detection,
multiple-error detection, low cost encoding and decoding, etc.
Chapter 1 55
 Hamming Codes (6)

• Hamming codes are most easily designed by specifying the H matrix.

• For any positive integer m  3, there exists an (n, k) SEC Hamming code with
the following properties:
– Code length: n = 2m - 1
– Number of information bits: k = 2m - m - 1
– Number of check bits: n - k = m
– Minimum distance: dmin = 3
• The H matrix is an n m matrix with all nonzero m-tuples as its column.
• A possible H matrix for a (15, 11) Hamming code, when m = 4:

111101110001000
111011001100100
H  (1.35)
110110100110010
 
101110011010001

Chapter 1 56
 Hamming Codes (7)
• Example: A Hamming code for encoding five (k = 5) information bits.
– Four check bits are required (m = 4). So, n = 9.
– A (9, 5) code can be obtained by deleting six columns from the (15,11)
code shown above.
– The H and G matrices are:

100001111 
111101000  
111010100 010001110 
H 
(1.36) G  001001101 (1.37)
110110010  
  000101011 
101110001  000010111
 

Chapter 1 57