Binary Systems

Sheis Abolmaali

School of ECE
Semnan University
1
 Binary Numbers 1/2
• Internally, information in digital systems is of
binary form
– groups of bits (i.e. binary numbers)
– all the processing (arithmetic, logical, etc) are
performed on binary numbers.
• Example: 4392
– In decimal, 4392 = …
– Convention: write only the coefficients.
– A = a1 a0 . a-1 a-2 a-3 where aj  {0, 1, …, 9}
– How do you calculate the value of A?

2
 Binary Numbers 2/2
• Decimal system
– coefficients are from {0,1, …, 9}
– and coefficients are multiplied by powers of 10
– base-10 or radix-10 number system
• Using the analogy, binary system {0,1}
– base(radix)-2
• Example: 25.625
– 25.625 = decimal expansion
– 25.625 = binary expansion
– 25.625 =

3
 Base-r Systems
• base-r (n, m)
– A = an-1 rn-1 +… + a1 r1 + a0 r0 + a-1 r-1 + a-2 r-2 + … +a-m r-m
• Octal
– base-8 = base-23
– digits {0,1, …, 7}
– Example: (31.5)8 = octal expansion =
• Hexadecimal
– base-16
– digits {0, 1, …, 9, A, B, C, D, E, F}
– Example:
– (19.A)16 = hexadecimal expansion =
4
 Powers of 2
• 210 = 1,024 (K) -
• 220 = 1,048,576 (M) -
• 230  (G) -
• 240  (T) -
• 250  (P) -
• exa, zetta, yotta, … (exbi, zebi, yobi, ...)
• Examples:
– A byte is 8-bit, i.e. 1 B
– 16 GiB = ? B = 17,179,869,184

5
 Arithmetic with Binary Numbers

10101 21 augend 10101 21 minuend

+ 10011 19 addend - 10011 19 subtrahend
1 01000 40 sum 0 00010 2 difference

0 0 1 0 multiplicand (2)
 1 0 1 1 multiplier (11)
0 0 1 0
0 0 1 0
0 0 0 0
+ 0 0 1 0
0 0 1 0 1 1 0 product (22)
6
Multiplication with Octal Numbers

3 4 5 229 multiplicand
 6 2 1 401 multiplier
3 4 5
7 1 2
+ 2 5 3 6
2 6 3 2 6 5 91829 product

7
 Base Conversions
• From base-r to decimal is easy
– expand the number in power series and add all the
terms
• Reverse operation is somewhat more difficult
• Simple idea:
– divide the decimal number successively by r
– accumulate the remainders in the remainder
• If there is a fraction, then integer part and
fraction part are handled separately.

8
 Base Conversion Examples 1/3
• Example 1:
– 55
– (decimal to binary)

– Example 2:
– 144
– (decimal to octal)

9
 Base Conversion Examples 2/3
• Example 1: 0.6875 (decimal to binary)
– When dealing with fractions, instead of dividing by r
multiply by r until we get an integer
– …

10
 Base Conversion Examples 2/3
• We are not always this lucky
• Example 2: (144.478) to octal
– Treat the integer part and fraction part separately
– 0.4788 = 3.824 = 3 + 0.824  a-1 = 3
– 0.8248 = 6.592 = 6 + 0.592  a-2 = 6
– 0.5928 = 4.736 = 4 + 0.736  a-3 = 4
– 0.7368 = 5.888 = 5 + 0.888  a-4 = 5
– 0.8888 = 7.104 = 7 + 0.104  a-5 = 7
– 0.1048 = 0.832 = 0 + 0.832  a-6 = 0
– 0.8328 = 6.656 = 6 + 0.656  a-7 = 6
– 144.478 = (220.3645706…)8
11
 Conversions between Binary, Octal
and Hexadecimal
• r = 2 (binary), r = 8 (octal), r = 16 (hexadecimal)

10110001101001.101100010111

10 110 001 101 001.101 100 010 111

10 1100 0110 1001.1011 0001 0111

• Octal and hexadecimal representations are more
compact.
• Therefore, we use them in order to communicate
with computers directly using their internal
representation
12
 Complement
• Complementing is an operation on base-r numbers
• Goal: To simplify subtraction operation
– Rather turn the subtraction operation into an
addition operation
• Two types
1. Radix complement (a.k.a. r’s complement)
2. Diminished complement (a.k.a. (r-1)’s complement)
• When r = 2
1. 2’s complement
2. 1’s complement

13
 How to Complement?
• A number N in base-r (n-digit)
1. rn – N r’s complement
2. (rn-1) – N (r-1)’s complement
– where n is the number of digits we use
• Example: r = 2, n = 4, N = 7
– rn = 24 = 16, rn -1 = 15.
– 2’s complement of 7  ?
– 1’s complement of 7  ?
• Easier way to compute 1’s and 2’s complements
– Use binary expansions
– 1’s complement: negate
– 2’s complement: negate + increment
14
 Subtraction with Complements 1/4
• Conventional subtraction
– Borrow concept
– If the minuend digit is smaller than the subtrahend
digit, you borrow “1” from a digit in higher significant
position
• With complements
– M-N = ?
– rn – N r’s complement of N
– M + (rn – N) =

15
 Subtraction with Complements 2/4
• M + (rn – N) = M – N + rn
1. if M  N,
– the sum will produce a carry, that can be discarded
2. Otherwise,
– the sum will not produce a carry, and will be equal to
rn – (N-M), which is the r’s complement of N-M

16
 Subtraction with Complements 3/4
• Example:
– X = 1010100 (84) and Y = 1000011 (67)
– X-Y = ? and Y-X = ?
X 1010100
2’s complement of Y + 0111101
10010001

Y 1000011
2’s complement of X + 0101100
1101111

-0010001

17
 Subtraction with Complements 4/4
• Example: Previous example using 1’s complement
X 1010100
1’s complement of Y + 0111100

Y 1000011
1’s complement of X + 0101011

18
 Signed Binary Numbers
• Pencil-and-paper
– Use symbols “+” and “-”
• We need to represent these symbols using bits
– Convention:
• 0 positive
1 negative
• The leftmost bit position is used as a sign bit
– In signed representation, bits to the right of sign bit
is the number
– In unsigned representation, the leftmost bit is a part
of the number (i.e. the most significant bit (MSB))

19
 Signed Binary Numbers
• Example: 5-bit numbers
– 01011  ? (unsigned binary)
–  ? (signed binary)
– 11011  ? (unsigned binary)
–  ? (signed binary)
– This method is called “signed-magnitude” and is rarely
used in digital systems (if at all)
• In computers, a negative number is represented
by the complement of its absolute value.
• Signed-complement system
– positive numbers have always “0” in the MSB position
– negative numbers have always “1” in the MSB position
20
 Signed-Complement System
• Example:
– 11 = (01011)
– How to represent –11 in 1’s and 2’s complements
1. 1’s complement –11 = ?
2. 2’s complement -11 = ?
– If we use eight bit precision:
– 11 = 00001011
– 1’s complement -11 = ?
– 2’s complement -11 = ?

21
 Signed Number Representation
Signed magnitude One’s complement Two’s complement
000 +0 000 +0 000 0
001 +1 001 +1 001 +1
010 +2 010 +2 010 +2
011 +3 011 +3 011 +3
100 -0 111 -0 111 -1
101 -1 110 -1 110 -2
110 -2 101 -2 101 -3
111 -3 100 -3 100 -4

• Issues: balance, number of zeros, ease of operations

• Which one is best? Why?
22
 Which One?
• Signed magnitude:
– Where to put the sign bit?
– Adders may need an additional step to set the sign
– There are two representations for 0.
• Try to subtract a large number from a smaller one.
2 = 0 0 1 0
5 = 0 1 0 1
= 1 1 0 1
• 2’s complement provides a natural way to represent signed
numbers (every computer today uses two’s complement)
• Think that there is an infinite number of 1’s in a signed
number
-3 = 1101  …11111101
23
• What is 11111100?
 Arithmetic Addition
• Examples:

+11 00001011 -11 11110101

+9 + 00001001 +9 + 00001001

+11 00001011 -11 11110101

-9 + 11110111 -9 + 11110111

• No special treatment for sign bits

24
 Arithmetic Overflow 1/2
• In hardware, we have limited resources to
accommodate numbers
– Computers use 8-bit, 16-bit, 32-bit, and 64-bit
registers for the operands in arithmetic operations.
– Sometimes the result of an arithmetic operation get
too large to fit in a register.

25
 Arithmetic Overflow 2/2
• Example:

+2 0010 -3 1101
+4 + 0100 -5 + 1011

+7 0111 -3 1101
+6 + 0110 -6 + 1010

• Rule: If the MSB and the bits to the left of it

differ, then there is an overflow
26
 Subtraction with Signed Numbers
• Rule: is the same
• We take the 2’s complement of the subtrahend
– It does not matter if the subtrahend is a negative
number.
– (A) - (-B) = A+B

-6 11111010 -6 11111010
-13 - 11110011 +

• Signed-complement numbers are added and subtracted in

the same way as unsigned numbers
• With the same circuit, we can do both signed and unsigned
arithmetic 27
 Alphanumeric Codes
• Besides numbers, we have to represent other
types of information
– letters of alphabet, mathematical symbols.
• For English, alphanumeric character set includes
– 10 decimal digits
– 26 letters of the English alphabet (both lowercase and
uppercase)
– several special characters
• We need an alphanumeric code
– ASCII
– American Standard Code for Information Exchange
– Uses 7 bits to encode 128 characters
28
 ASCII Code
• 7 bits of ASCII Code
– (b6 b5 b4 b3 b2 b1 b0)2
• Examples:
– A  65 = (1000001),…, Z  90 = (1011010)
– a  97 = (1100001), …, z  122 = (1111010)
– 0 48 = (0110000), …, 9  57 = (0111001)
• 128 different characters
– 26 + 26 + 10 = 62 (letters and decimal digits)
– 32 special printable characters %, *, $
– 34 special control characters (non-printable): BS, CR,
etc. 29
 Representing ASCII Code
• 7-bit
• Most computers manipulate 8-bit quantity as a
single unit (byte)
– One ASCII character is stored using a byte
– One unused bit can be used for other purposes such as
representing Greek alphabet, italic type font, etc.
• The eighth bit can be used for error-detection
– parity of seven bits of ASCII code is prefixed as a bit
to the ASCII code.
– A  (0 1000001) even parity
– A  (1 1000001) odd parity
– Detects one, three, and any odd number of bit errors
30
 Binary Logic
• Binary logic is equivalent to what it is called “two-
valued Boolean algebra”
– Or we can say it is an implementation of Boolean
algebra
• Deals with variables that take on “two discrete
values” and operations that assume logical
meaning
• Two discrete values:
– {true, false}
– {yes, no}
– {1, 0}
31
 Binary Variables and Operations
• We use A, B, C, x, y, z, etc. to denote binary
variables
– each can take on {0, 1}
• Logical operations
1. AND  x · y = z or xy = z
2. OR x+y=z
3. NOT  x = z or x’ = z
– For each combination of the values of x and y, there
is a value of specified by the definition of the logical
operation.
– This definition may be listed in a compact form called
truth table.
32
 Truth Table

x y AND OR NOT
x·y x + y x’
0 0

0 1

1 0

1 1

33
 Logic Gates
• Binary values are represented as electrical
signals
– Voltage, current
• They take on either of two recognizable values
– For instance, voltage-operated circuits
– 0V  0
– 4V  1
• Electronic circuits that operate on one or more
input signals to produce output signals
– AND gate, OR gate, NOT gate
34
 Range of Electrical Signals
• What really matters is the range of the signal
value

35
Logic Gate Symbols

36
 Gates Operating on Signals
V
x 0 1 1 0 0
t

y 0 0 1 1 0
t

AND: xy 0 0 1 0 0
t

OR: xy 0 1 1 1 0
t

NOT: x’ 1 0 0 1 1
t
Input-Output Signals for gates
Gates with More Than Two Inputs

38