CS2100 Computer Organisation

Number Systems and Codes

NUMBER SYSTEMS & CODES

Information Representations
Number Systems
Base Conversion
Negative Numbers
Excess Representation
Floating-Point Numbers
Decimal codes: BCD, Excess-3, 2421, 84-2-1
Gray Code
Alphanumeric Code

2011 Sem 1

Number Systems

INFORMATION REPRESENTATION (1/3)

Numbers are important to computers

Represent information precisely
Can be processed

Examples
Represent yes or no: use 0 and 1
Represent the 4 seasons: 0, 1, 2 and 3

Sometimes, other characters are used

Matriculation number: 8 alphanumeric characters (eg:
U071234X)

2011 Sem 1

Number Systems

INFORMATION REPRESENTATION (2/3)

Bit (Binary digit)

0 and 1
Represent false and true in logic
Represent the low and high states in electronic devices

Other units
Byte: 8 bits
Nibble: 4 bits (seldom used)
Word: Multiples of byte (eg: 1 byte, 2 bytes, 4 bytes, 8
bytes, etc.), depending on the architecture of the computer
system

2011 Sem 1

Number Systems

INFORMATION REPRESENTATION (3/3)

N bits can represent up to 2N values.

Examples:
2 bits represent up to 4 values (00, 01, 10, 11)
3 bits rep. up to 8 values (000, 001, 010, , 110, 111)
4 bits rep. up to 16 values (0000, 0001, 0010, ., 1111)

To represent M values, log2M bits are required.

Examples:

2011 Sem 1

32 values requires 5 bits

64 values requires 6 bits
1024 values requires 10 bits
40 values how many bits?
100 values how many bits?

Number Systems

log 2 M
5

DECIMAL (BASE 10) SYSTEM (1/2)

A weighted-positional number system

Base or radix is 10 (the base or radix of a number system is the
total number of symbols/digits allowed in the system)
Symbols/digits = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }
Position is important, as the value of each symbol/digit is
dependent on its type and its position in the number
Example, the 9 in the two numbers below has different values:
(7594)10 = (7 103) + (5 102) + (9 101) + (4 100)

(912)10 = (9 102) + (1 101) + (2 100)

In general,

2011 Sem 1

(anan-1 a0 . f1f2 fm)10 =

(an x 10n) + (an-1x10n-1) + + (a0 x 100) +
(f1 x 10-1) + (f2 x 10-2) + + (fm x 10-m)
Number Systems

The weighted-positional system

For any radix r:

an an1 L

a1a0 .b1b2 L

k 0

k 1

r ak r k bk r k

Note: The Roman numeral system is not

weighted-positional: I, II, III, IV, V,

2011 Sem 1

Number Systems

DECIMAL (BASE 10) SYSTEM (2/2)

Weighing factors (or weights) are in powers of 10:

103 102 101 100 . 10-1 10-2 10-3

To evaluate the decimal number 593.68, the digit in each

position is multiplied by the corresponding weight:
5102 + 9101 + 3100 + 610-1 + 810-2
= (593.68)10

2011 Sem 1

Number Systems

A bit of history
the Hindu-Arab Numeral System
From Wikipedia:
The symbols for 1 to 9 in the Hindu-Arabic numeral system evolved from the Brahmi
numerals. Buddhist inscriptions from around 300 BC use the symbols which became 1, 4
and 6. One century later, their use of the symbols which became 2, 7 and 9 was
recorded.
The first universally accepted inscription containing the use of the 0 glyph is first recorded
in the 9th century, in an inscription at Gwalior dated to 870. Indian documents on copper
plates, with the same symbol for zero in them, dated back as far as the 6th century AD,
abound.
The numeral system came to be known to both the Persian mathematician Al-Khwarizmi,
whose book On the Calculation with Hindu Numerals written about 825, and the Arab
mathematician Al-Kindi, who wrote four volumes, On the Use of the Indian Numerals
about 830, are principally responsible for the diffusion of the Indian system of numeration
in the Middle East and the West. In the 10th century, Middle-Eastern mathematicians
extended the decimal numeral system to include fractions, as recorded in a treatise by
Syrian mathematician Abu'l-Hasan al-Uqlidisi in 95253.
2011 Sem 1

Number Systems

OTHER NUMBER SYSTEMS (1/2)

Binary (base 2)
Weights in powers of 2
Binary digits (bits): 0, 1

Octal (base 8)
Weights in powers of 8
Octal digits: 0, 1, 2, 3, 4, 5, 6, 7.

Hexadecimal (base 16)

Weights in powers of 16
Hexadecimal digits: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F.

Base/radix R:
Weights in powers of R

2011 Sem 1

Number Systems

10

OTHER NUMBER SYSTEMS (2/2)

In

some programming languages/software,

special notations are used to represent
numbers in certain bases
In programming language C
Prefix 0 for octal. Eg: 032 represents the octal number (32)8
Prefix 0x for hexadecimal. Eg: 0x32 represents the
hexadecimal number (32)16

In PCSpim (a MIPS simulator)

Prefix 0x for hexadecimal. Eg: 0x100 represents the
hexadecimal number (100)16
2011 Sem 1

Number Systems

11

BASE-R TO DECIMAL CONVERSION

Easy!
1101.1012 =
572.68

2A.816

341.245 =

2011 Sem 1

Number Systems

12

QUICK REVIEW QUESTIONS (1)

DLD page 37
Questions 2-1 to 2-4.

2011 Sem 1

Number Systems

13

DECIMAL TO BINARY CONVERSION

Method 1
Sum-of-Weights Method

Method 2
Repeated Division-by-2 Method (for whole numbers)
Repeated Multiplication-by-2 Method (for fractions)

2011 Sem 1

Number Systems

14

SUM-OF-WEIGHTS METHOD

Determine the set of binary weights whose sum is

equal to the decimal number
(9)10 = 8 + 1 = 23 + 20 = (1001)2
(18)10 = 16 + 2 = 24 + 21 = (10010)2
(58)10 = 32 + 16 + 8 + 2 = 25 + 24 + 23 + 21 = (111010)2
(0.625)10 = 0.5 + 0.125 = 2-1 + 2-3 = (0.101)2

2011 Sem 1

Number Systems

15

REPEATED DIVISION-BY-2
To convert a whole number to binary, use successive division by 2
until the quotient is 0. The remainders form the answer, with the
first remainder as the least significant bit (LSB) and the last as the
most significant bit (MSB).
(43)10 = (101011)2

2011 Sem 1

Number Systems

16

REPEATED MULTIPLICATION-BY-2
To convert decimal fractions to binary, repeated multiplication
by 2 is used, until the fractional product is 0 (or until the
desired number of decimal places). The carried digits, or
carries, produce the answer, with the first carry as the MSB,
and the last as the LSB.
Carry
(0.3125)10 = (.0101)2
MSB
0.31252=0.625 0
1
0.6252=1.25
0
0.252=0.50
1
LSB
0.52=1.00

2011 Sem 1

Number Systems

17

CONVERSION BETWEEN DECIMAL AND

OTHER BASES
Base-R to decimal: multiply digits with their corresponding
weights.

Decimal to binary (base 2)

Whole numbers repeated division-by-2
Fractions: repeated multiplication-by-2

Decimal to base-R
Whole numbers: repeated division-by-R
Fractions: repeated multiplication-by-R

2011 Sem 1

Number Systems

18

QUICK REVIEW QUESTIONS (2)

DLD

page 37
Questions 2-5 to 2-8.

2011 Sem 1

Number Systems

19

CONVERSION BETWEEN BASES

In general, conversion between bases can be done

via decimal:
Base-2
Base-3
Base-4

Base-R

Decimal

Base-2
Base-3
Base-4
.
Base-R

Shortcuts for conversion between bases 2, 4, 8, 16

(see next slide)

2011 Sem 1

Number Systems

20

BINARY TO OCTAL/HEXADECIMAL CONVERSION

Binary Octal: partition in groups of 3

(10 111 011 001 . 101 110)2 =

Octal Binary: reverse

(2731.56)8 =

Binary Hexadecimal: partition in groups of 4

(101 1101 1001 . 1011 1000)2 =

Hexadecimal Binary: reverse

(5D9.B8)16 =

2011 Sem 1

Number Systems

21

QUICK REVIEW QUESTIONS (3)

DLD

page 37
Questions 2-9 to 2-10.

2011 Sem 1

Number Systems

22

READING ASSIGNMENT
Binary

arithmetic operations

Read up DLD section 2.6, pg 20 21.

2011 Sem 1

Number Systems

23

NEGATIVE NUMBERS
Unsigned numbers: only non-negative values.
Signed numbers: include all values (positive and
negative)
There are 3 common representations for signed
binary numbers:

Sign-and-Magnitude
1s Complement
2s Complement

2011 Sem 1

Number Systems

24

DEFINITIONS
Given a n-bit binary string: an 1an 2 L a1a0

Sign-and-magnitude interpretation:

n2

ai 2i
i 0

1s complement interpretation:

an1

n 1

n2

1 an 1 ai 2i
i 0

2s complement interpretation:

2
2011 Sem 1

n 1

n2

an 1 ai 2i
i 0

Number Systems

25

SIGN-AND-MAGNITUDE (1/3)

The sign is represented by a sign bit

0 for +
1 for -

Eg: a 1-bit sign and 7-bit magnitude format.

magnitude

sign

00110100 +1101002 = +5210

10010011 -100112 = -1910
2011 Sem 1

Number Systems

26

SIGN-AND-MAGNITUDE (2/3)

Largest value:

01111111 = +12710

Smallest value: 11111111 = -12710

Zeros:

Range: -12710 to +12710

Question:

00000000 = +010
10000000 = -010

For an n-bit sign-and-magnitude representation, what is

the range of values that can be represented?

2011 Sem 1

Number Systems

27

SIGN-AND-MAGNITUDE (3/3)
To negate a number, just invert the sign bit.
Examples:

How to negate 00100001sm (decimal 33)?

Answer: 10100001sm (decimal -33)
How to negate 10000101sm (decimal -5)?
Answer: 00000101sm (decimal +5)

2011 Sem 1

Number Systems

28

1s COMPLEMENT (1/3)
Given a number x which can be expressed as an n-bit binary
number, its negated value can be obtained in 1s-complement
representation using:
-x = 2n x 1
Example: With an 8-bit number 00001100 (or 1210), its
negated value expressed in 1s-complement is:
-000011002
= 28 12 1 (calculation in decimal)
= 243
= 111100111s

(This means that -1210 is written as 11110011 in 1scomplement representation.)

2011 Sem 1

Number Systems

29

1s COMPLEMENT (2/3)
Essential technique to negate a value: invert all the bits.
Largest value:
01111111 = +12710

Smallest value:

Zeros:

Range: -12710 to +12710

The most significant (left-most) bit still represents the sign: 0

for positive; 1 for negative.

2011 Sem 1

10000000 = -12710
00000000 = +010
11111111 = -010

Number Systems

30

1s COMPLEMENT (3/3)

Examples (assuming 8-bit numbers):

(14)10 = (00001110)2 = (00001110)1s
-(14)10 = -(00001110)2 = (11110001)1s
-(80)10 = -( ? )2 = ( ? )1s

2011 Sem 1

Number Systems

31

2s COMPLEMENT (1/3)
Given a number x which can be expressed as an n-bit binary
number, its negated value can be obtained in 2s-complement
representation using:
-x = 2n x
Example: With an 8-bit number 00001100 (or 1210), its
negated value expressed in 1s-complement is:
-000011002
= 28 12 (calculation in decimal)
= 244
= 111101002s

(This means that -1210 is written as 11110100 in 2scomplement representation.)

2011 Sem 1

Number Systems

32

2s COMPLEMENT (2/3)
Essential technique to negate a value: invert all the
bits, then add 1.
Largest value:
01111111 = +12710

Smallest value: 10000000 = -12810

Zero:

Range: -12810 to +12710

The most significant (left-most) bit still represents

the sign: 0 for positive; 1 for negative.

2011 Sem 1

00000000 = +010

Number Systems

33

2s COMPLEMENT (3/3)

Examples (assuming 8-bit numbers):

(14)10 = (00001110)2 = (00001110)2s
-(14)10 = -(00001110)2 = (11110010)2s
-(80)10 = -( ? )2 = ( ? )2s
Compare with slide 30.

2011 Sem 1

Number Systems

34

COMPARISONS
4-bit system
Positive values

Negative values

Value

Sign-andMagnitude

1s
Comp.

2s
Comp.

Value

Sign-andMagnitude

1s
Comp.

2s
Comp.

+7
+6
+5
+4
+3
+2
+1
+0

0111
0110
0101
0100
0011
0010
0001
0000

0111
0110
0101
0100
0011
0010
0001
0000

0111
0110
0101
0100
0011
0010
0001
0000

-0
-1
-2
-3
-4
-5
-6
-7
-8

1000
1001
1010
1011
1100
1101
1110
1111
-

1111
1110
1101
1100
1011
1010
1001
1000
-

1111
1110
1101
1100
1011
1010
1001
1000

2011 Sem 1

Number Systems

35

COMPLEMENT ON FRACTIONS
We can extend the idea of complement on fractions.
Examples:

Negate 0101.01 in 1s-complement

Answer: 1010.10
Negate 111000.101 in 1s-complement
Answer: 000111.010
Negate 0101.01 in 2s-complement
Answer: 1010.11

2011 Sem 1

Number Systems

36

2s COMPLEMENT
ADDITION/SUBTRACTION (1/3)

Algorithm for addition, A + B:

1. Perform binary addition on the two numbers.
2. Ignore the carry out of the MSB.
3. Check for overflow. Overflow occurs if the carry in and carry
out of the MSB are different, or if result is opposite sign of A
and B.

Algorithm for subtraction, A B:

A B = A + (-B)
1. Take 2s-complement of B.
2. Add the 2s-complement of B to A.

2011 Sem 1

Number Systems

37

2s COMPLEMENT
ADDITION/SUBTRACTION (2/3)

Examples: 4-bit system

+3
+ +4
---+7
----

0011
+ 0100
------0111
-------

-2
+ -6
----8
----

1110
+ 1010
------11000
-------

+6
+ -3
---+3
----

0110
+ 1101
------10011
-------

+4
+ -7
----3
----

0100
+ 1001
------1101
-------

Which of the above is/are overflow(s)?

2011 Sem 1

Number Systems

38

2s COMPLEMENT
ADDITION/SUBTRACTION (3/3)

Examples: 4-bit system

-3
+ -6
----9
----

1101
+ 1010
------10111
-------

+5
+ +6
---+11
----

0101
+ 0110
------1011
-------

Which of the above is/are overflow(s)?

2011 Sem 1

Number Systems

39

1s COMPLEMENT
ADDITION/SUBTRACTION (1/2)

Algorithm for addition, A + B:

1. Perform binary addition on the two numbers.
2. If there is a carry out of the MSB, add 1 to the result.
3. Check for overflow. Overflow occurs if result is opposite sign of
A and B.

Algorithm for subtraction, A B:

A B = A + (-B)
1. Take 1s-complement of B.
2. Add the 1s-complement of B to A.

2011 Sem 1

Number Systems

40

1s COMPLEMENT
ADDITION/SUBTRACTION (2/2)

Examples: 4-bit system

+3
+ +4
---+7
----2
+ -5
----7
----

2011 Sem 1

0011
+ 0100
------0111
------1101
+ 1010
-----10111
+
1
-----1000
Number Systems

+5
+ -5
----0
----3
+ -7
----10
----

0101
+ 1010
------1111
------1100
+ 1000
------10100
+
1
------0101
41

OVERFLOW

Signed numbers are of a fixed range.

If the result of addition/subtraction goes beyond this range, an
overflow occurs.
Overflow can be easily detected:
positive add positive negative
negative add negative positive

Example: 4-bit 2s-complement system

Range of value: -810 to 710
01012s + 01102s = 10112s
510 + 610 = -510 ?! (overflow!)
10012s + 11012s = 101102s (discard end-carry) = 01102s
-710 + -310 = 610 ?! (overflow!)

2011 Sem 1

Number Systems

42

Why it works?
Comes

from algebras ring theory.

The

2n possible values of n bits actually

form a ring of equivalence classes,
namely the integers modulo 2n.

2011 Sem 1

Number Systems

43

QUICK REVIEW QUESTIONS (4)

DLD

page 37
Questions 2-13 to 2-18.

2011 Sem 1

Number Systems

44

EXCESS REPRESENTATION (1/2)

Besides sign-and-magnitude and

complement schemes, the excess
representation is another scheme.
It allows the range of values to be
distributed evenly between the
positive and negative values, by a
simple translation
(addition/subtraction).
Example: Excess-4 representation
on 3-bit numbers. See table on the
right.
Questions: What if we use Excess-2
on 3-bit numbers? Excess-7?

2011 Sem 1

Number Systems

Excess-4
Representation

Value

000

-4

001

-3

010

-2

011

-1

100

101

110

111

45

EXCESS REPRESENTATION (2/2)

Example: For 4-bit numbers, we may use excess-7 or

excess-8. Excess-8 is shown below. Fill in the values.
Excess-8
Representation

2011 Sem 1

Value

Excess-8
Representation

0000

1000

0001

1001

0010

1010

0011

1011

0100

1100

0101

1101

0110

1110

0111

1111
Number Systems

Value

46

FIXED POINT NUMBERS (1/2)

In fixed point representation, the binary point is

assumed to be at a fixed location.
For example, if the binary point is at the end of an 8-bit
representation as shown below, it can represent integers
from -128 to +127.

binary point

2011 Sem 1

Number Systems

47

FIXED POINT NUMBERS (2/2)

In general, the binary point may be assumed to be

at any pre-fixed location.
Example: Two fractional bits are assumed as shown below.

integer part

fraction part
assumed binary point

If 2s complement is used, we can represent values like:

011010.112s = 26.7510
111110.112s = -000001.012 = -1.2510
2011 Sem 1

Number Systems

48

FLOATING POINT NUMEBRS (1/4)

Fixed point numbers have limited range.
Floating point numbers allow us to represent very
large or very small numbers.
Examples:

0.23 1023 (very large positive number)

0.5 10-37 (very small positive number)
-0.2397 10-18 (very small negative number)

2011 Sem 1

Number Systems

49

FLOATING POINT NUMEBRS (2/4)

3 parts: sign, mantissa and exponent
The base (radix) is assumed to be 2.
Sign bit: 0 for positive, 1 for negative.

sign

mantissa

exponent

Mantissa is usually in normalised form (the integer part is

zero and the fraction part must not begin with zero)
0.01101 24 normalised
101011.0110 2-4 normalised

Trade-off:
More bits in mantissa better precision
More bits in exponent larger range of values

2011 Sem 1

Number Systems

50

FLOATING POINT NUMEBRS (3/4)

Exponent is usually expressed in complement or excess

format.

Example: Express -6.510 in base-2 normalised form

-6.510 = -110.12 = -0.11012 23

Assuming that the floating-point representation contains 1-bit,

5-bit normalised mantissa, and 4-bit exponent. The above
example will be stored as if the exponent is in 1s or 2s
complement.

2011 Sem 1

11010

0011

Number Systems

51

FLOATING POINT NUMBERS (4/4)

Example: Express 0.187510 in base-2 normalised form

0.187510 = 0.00112 = 0.11 2-2

Assume this floating-point representation:1-bit sign, 5-bit

normalised mantissa, and 4-bit exponent.

The above example will be represented as

2011 Sem 1

11000

1101

If exponent is in 1s complement.

11000

1110

If exponent is in 2s complement.

11000

0110

If exponent is in excess-8.

Number Systems

52

QUICK REVIEW QUESTIONS (5)

DLD

page 38
Questions 2-19 to 2-20.

2011 Sem 1

Number Systems

53

READING ASSIGNMENT

Arithmetic operations on floating point numbers

DLD page 29

IEEE floating point representation

DLD page 30
IEEE standard 754 floating point numbers:
http://steve.hollasch.net/cgindex/coding/ieeefloat.html

2011 Sem 1

Number Systems

54

DECIMAL CODES
Decimal numbers are favoured by humans. Binary numbers
are natural to computers. Hence, conversion is required.
If little calculation is required, we can use some coding
schemes to store decimal numbers, for data transmission
purposes.
Examples: BCD (or 8421), Excess-3, 84-2-1, 2421, etc.
Each decimal digit is represented as a 4-bit code.

2011 Sem 1

Number Systems

55

BINARY CODE DECIMAL (BCD) (1/2)

Decimal
digit

BCD

0000

0001

0010

Easy to convert, but arithmetic

operations are more
complicated.

0011

0100

0101

Suitable for interfaces such as

keypad inputs.

0110

0111

1000

1001

Some codes are unused, like

1010BCD, 1011BCD, 1111BCD.
These codes are considered as
errors.

2011 Sem 1

Number Systems

56

BINARY CODE DECIMAL (BCD) (2/2)

Examples of conversion between BCD values and decimal

values:
(234)10 = (0010 0011 0100)BCD
(7093)10 = (0111 0000 1001 0011)BCD
(1000 0110)BCD = (86)10
(1001 0100 0111 0010)BCD = (9472)10

Note that BCD is not equivalent to binary

Example: (234)10 = (11101010)2

2011 Sem 1

Number Systems

57

OTHER DECIMAL CODES

Decimal Digit
0
1
2
3
4
5
6
7
8
9

BCD
8421
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001

Excess-3

84-2-1

2*421

0011
0100
0101
0110
0111
1000
1001
1010
1011
1100

0000
0111
0110
0101
0100
1011
1010
1001
1000
1111

0000
0001
0010
0011
0100
1011
1100
1101
1110
1111

Biquinary
5043210
0100001
0100010
0100100
0101000
0110000
1000001
1000010
1000100
1001000
1010000

Self-complementing code: codes for complementary digits

are also complementary to each other.

Error-detecting code: biquinary code (bi=two, quinary=five).

2011 Sem 1

Number Systems

58

SELF-COMPLEMENTING CODES
The codes representing the pair of complementary digits are
also complementary to each other.
0: 0011
Example: Excess-3 code

241: 0101 0111 0100

758: 1010 1000 1011

1:
2:
3:
4:
5:
6:
7:
8:
9:

0100
0101
0110
0111
1000
1001
1010
1011
1100

Question: What are the other self-complementing codes?

2011 Sem 1

Number Systems

59

GRAY CODE (1/3)

Unweighted (not an arithmetic code)

Only a single bit change from one code value to the next.
Not restricted to decimal digits: n bits 2n values.
Good for error detection.
Example: 4-bit standard Gray code

2011 Sem 1

Number Systems

60

GRAY CODE (2/3)

Generating a 4-bit standard Gray code sequence.

0000
0001
0011
0001
0010
0000
0010
0110
0111
0011
0101
0001
0100
0000

1100
0100
0101
1101
0111
1111
0110
1110
0010
1010
0011
1011
0001
1001
0000
1000

Questions: How to generate 5-bit standard Gray code

sequence? 6-bit standard Gray code sequence?

2011 Sem 1

Number Systems

61

0
0

GRAY CODE (3/3)

0
1

1
0

0
0

mis-aligned
sensors

mis-aligned
sensors

1
010 11
111
1 1
111 1 1
1
0

0
11 00
110 1
0
00 1 0
11 0
1

1
11 11
00 0
0
1
000 00 1
0
0

0
1 01
000 10
000 00 1
0 1
0

0
0 01
11 01
0
1
000 11 1
1
0

Binary coded: 111 110 000

2011 Sem 1

sensors

0
00 00
001 0
111 00 0
0
0
1

1
1

0
11 01
111 0
111 0 1
00 1
0

0
010 10
010
1 1
111 1 1
1
0

Number Systems

Gray coded: 111 101

62

READING ASSIGNMENT

Conversion between standard Gray code and binary

DLD pages 35 36.

2011 Sem 1

Number Systems

63

QUICK REVIEW QUESTIONS (6)

DLD

page 38
Questions 2-21 to 2-24.

2011 Sem 1

Number Systems

64

ALPHANUMERIC CODES (1/3)

Computers also handle textual data.
Character set frequently used:

alphabets:
A Z, a z
digits:
0 9
special symbols: $, ., @, *, etc.
non-printable: NULL, BELL, CR, etc.

Examples
ASCII (8 bits), Unicode

2011 Sem 1

Number Systems

65

ALPHANUMERIC CODES (2/3)

ASCII
American Standard Code
for Information
Interchange
7 bits, plus a parity bit for
error detection
Odd or even parity

2011 Sem 1

Number Systems

Character
0
1
...
9
:
A
B
...
Z
[
\

ASCII Code
0110000
0110001
...
0111001
0111010
1000001
1000010
...
1011010
1011011
1011100

66

ALPHANUMERIC CODES (3/3)

ASCII

LSBs
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
1100
1101
1110
1111
2011 Sem 1

A: 1000001

table
000
NUL
SOH
STX
ETX
EOT
ENQ
ACK
BEL
BS
HT
LF
VT
FF
CR
O
SI

001
DLE
DC1
DC2
DC3
DC4
NAK
SYN
ETB
CAN
EM
SUB
ESC
FS
GS
RS
US

010
SP
!

#
$
%
&

(
)
*
+
,
.
/

MSBs
011 100
0
@
1
A
2
B
3
C
4
D
5
E
6
F
7
G
8
H
9
I
:
J
;
K
<
L
=
M
>
N
?
O

Number Systems

101
P
Q
R
S
T
U
V
W
X
Y
Z
[
\
]
^
_

110
`
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o

111
p
q
r
s
t
u
v
w
x
y
z
{
|
}
~
DEL
67

Unicode
International

standard for representing

multiple languages
UTF-8 an 8-bit variable-width encoding
which maximizes compatibility with ASCII.
UTF-16 a 16-bit, variable-width
encoding
UTF-32 a 32-bit, fixed-width encoding
2011 Sem 1

Number Systems

68

ERROR DETECTION (1/4)

Errors can occur during data transmission. They
should be detected, so that re-transmission can be
requested.
With binary numbers, usually single-bit errors occur.

Example: 0010 erroneously transmitted as 0011 or 0000 or

0110 or 1010.

Biquinary code uses 3 additional bits for errordetection.

2011 Sem 1

Number Systems

69

ERROR DETECTION (2/4)

Parity bit
Even parity: additional bit
added to make total
number of 1s even.
Odd parity: additional bit
added to make total
number of 1s odd.

Example of odd parity on

ASCII values.

2011 Sem 1

Character
0
1
...
9
:
A
B
...
Z
[
\

Number Systems

ASCII Code
0110000 1
0110001 0
...
0111001 1
0111010 1
1000001 1
1000010 1
...
1011010 1
1011011 0
1011100 1

Parity bits

70

ERROR DETECTION (3/4)

Parity bit can detect odd

number of errors but not even
number of errors.
Example: Assume odd parity,
10011 10001 (detected)
10011 10101 (not detected)

Parity bits can also be applied

to a block of data.

0110 1
0001 0
1011 0
1111 1
1001 1
0101 0

Column-wise parity

Row-wise parity
2011 Sem 1

Number Systems

71

ERROR DETECTION (4/4)

Sometimes, it is not enough to do error detection.
We may want to correct the errors.
Error correction is expensive. In practice, we may
use only single-bit error correction.
Popular technique: Hamming code single error
correction and double error detection (SECDED)
(not covered).

2011 Sem 1

Number Systems

72

END

2011 Sem 1

Number Systems

73