Huffman Coding( a Lossless data

compression technique)
Encoding messages
 Encode a message composed of a string of
characters
 Codes used by computer systems
 ASCII
• uses 8 bits per character
• can encode 256 characters
 Unicode
• 16 bits per character
• can encode 65536 characters
• includes all characters encoded by ASCII
 ASCII and Unicode are fixed-length codes
 all characters represented by same number of bits
Problems
 Suppose that we want to encode a message
constructed from the symbols A, B, C, D, and E
using a fixed-length code
 How many bits are required to encode each
symbol?
 at least 3 bits are required
 2 bits are not enough (can only encode four

symbols)
 How many bits are required to encode the

message DEAACAAAAABA?
 there are twelve symbols, each requires 3 bits
 12*3 = 36 bits are required
Drawbacks of fixed-length codes
 Wasted space
 Unicode uses twice as much space as ASCII
• inefficient for plain-text messages containing
only ASCII characters
 Same number of bits used to represent all characters
 ‘a’ and ‘e’ occur more frequently than ‘q’ and ‘z’

 Potential solution: use variable-length codes

 variable number of bits to represent characters
when frequency of occurrence is known
 short codes for characters that occur frequently
Advantages of variable-length codes
 The advantage of variable-length codes over fixed-
length is short codes can be given to characters that
occur frequently
 on average, the length of the encoded message is
less than fixed-length encoding
 Potential problem: how do we know where one
character ends and another begins?
• not a problem if number of bits is fixed!

A = 00 0010110111001111111111
B = 01
C = 10 ACDBADDDDD
D = 11
Prefix property
 A code has the prefix property if no character code
is the prefix (start of the code) for another character
 Example:

Symbol Code
P 000
01001101100010
Q 11
R 01 RSTQPT
S 001
T 10
 000 is not a prefix of 11, 01, 001, or 10
 11 is not a prefix of 000, 01, 001, or 10 …
Code without prefix property
 The following code does not have prefix property

Symbol Code
P 0
Q 1
R 01
S 10
T 11
 The pattern 1110 can be decoded as QQQP, QTP,
QQS, or TS
Problem
 Design a variable-length prefix-free code such that
the message DEAACAAAAABA can be encoded
using 22 bits
 Possible solution:
 A occurs eight times while B, C, D, and E each
occur once
 represent A with a one bit code, say 0
• remaining codes cannot start with 0
 represent B with the two bit code 10
• remaining codes cannot start with 0 or 10
 represent C with 110
 represent D with 1110
 represent E with 11110
Encoded message

DEAACAAAAABA

Symbol Code
A 0
B 10
C 110
D 1110
E 11110

1110111100011000000100 22 bits
Another possible code

DEAACAAAAABA

Symbol Code
A 0
B 100
C 101
D 1101
E 1111

1101111100101000001000 22 bits
Better code

DEAACAAAAABA

Symbol Code
A 0
B 100
C 101
D 110
E 111

11011100101000001000 20 bits
What code to use?
 Question: Is there a variable-length code that makes
the most efficient use of space?

Answer: Yes!
Huffman coding tree
 Binary tree
 each leaf contains symbol (character)
 label edge from node to left child with 0
 label edge from node to right child with 1
 Code for any symbol obtained by following path from
root to the leaf containing symbol
 Code has prefix property
 leaf node cannot appear on path to another leaf
 note: fixed-length codes are represented by a
complete Huffman tree and clearly have the prefix
property
Building a Huffman tree
 Find frequencies of each symbol occurring in
message
 Begin with a forest of single node trees
 each contain symbol and its frequency
 Do recursively
 select two trees with smallest frequency at the root
 produce a new binary tree with the selected trees
as children and store the sum of their frequencies
in the root
 Recursion ends when there is one tree
 this is the Huffman coding tree
Example
 Build the Huffman coding tree for the message
This is his message
 Character frequencies

A G M T E H _ I S

1 1 1 1 2 2 3 3 5

 Begin with forest of single trees

1 1 1 1 2 2 3 3 5
A G M T E H _ I S
Step 1

1 1 1 1 2 2 3 3 5
A G M T E H _ I S
Step 2

2 2

1 1 1 1 2 2 3 3 5
A G M T E H _ I S
Step 3

2 2 4

1 1 1 1 2 2 3 3 5
A G M T E H _ I S
Step 4

2 2 4

1 1 1 1 2 2 3 3 5
A G M T E H _ I S
Step 5

2 2 4 6

1 1 1 1 2 2 3 3 5
A G M T E H _ I S
Step 6

4 4

2 2 2 2 6
E H

1 1 1 1 3 3 5
A G M T _ I S
Step 7

8 11

4 4 5 6
S

2 2 2 2 3 3
E H _ I

1 1 1 1
A G M T
Step 8
19

8 11

4 4 5 6
S

2 2 2 2 3 3
E H _ I

1 1 1 1
A G M T
Label edges
19
0 1

8 11
0 1
0 1

4 4 5 6
0 1 0 1 S 0 1

2 2 2 2 3 3
0 1 0 1 E H _ I

1 1 1 1
A G M T
Huffman code & encoded message
This is his message

S 10
E 010
H 011
_ 110
I 111
A 0000
G 0001
M 0010
T 0011

00110111011110010111100011101111000010010111100000001010