Huffman Algorithm

7/24/2019 Huffman Algorithm

http://slidepdf.com/reader/full/huffman-algorithm 1/44

Huffman Coding:

CS 102

Trees and Priority Queues



Encoding and

Compression of Data

• Fax Machines

• ASCII

• Variations on ASCII

CS 102

– min number of bits needed – cost of savings

– patterns

– modifications



Purpose of Huffman

Coding

•Proposed by Dr. David A.Huffman in 1952

– “A Method for the Construction

CS 102

of Minimum Redundancy Codes”

• Applicable to many forms of

data transmission

– Our example: text files



The Basic Algorithm

• Huffman coding is a form of

statistical coding

• Not all characters occur with the

CS 102

same requency• Yet all characters are allocated

the same amount of space

– 1 char = 1 byte, be it e or x



The Basic Algorithm

•

Any savings in tailoring codesto frequency of character?

• Code word lengths are no longer

CS 102

fixed like ASCII.

• Code word lengths vary and will

be shorter for the morefrequently used characters.



The (Real) Basic

Algorithm 1. Scan text to be compressed and tally

occurrence of all characters.

2. Sort or prioritize characters based on

number of occurrences in text.

CS 102

3. Build Huffman code tree based on prioritized list.

4. Perform a traversal of tree to determine

all code words.

5. Scan text again and create new file

using the Huffman codes.



Building a Tree

Scan the original text

•Consider the following shorttext:

CS 102

Eerie eyes seen near lake.

• Count up the occurrences of all

characters in the text



Building a Tree


Eerie eyes seen near lake.• What characters are present?

CS 102

E e r i space

y s n a r l k .



Building a Tree



• What is the frequency of each

character in the text?

CS 102

Char Freq. Char Freq. Char Freq.E 1 y 1 k 1e 8 s 2 . 1r 2 n 2i 1 a 2space 4 l 1



Building a Tree

Prioritize characters

• Create binary tree nodes with

character and frequency of

each character

CS 102

•Place nodes in a priorityqueue

– The lower the occurrence, the

higher the priority in the queue



Building a Tree

Prioritize characters

• Uses binary tree nodes

public class HuffNode{

ublic char m Char

CS 102

public int myFrequency; public HuffNode myLeft, myRight;}

priorityQueue myQueue;



Building a Tree

• The queue after inserting all nodes

CS 102

• Null Pointers are not shown

E

1

i

1

y

1

l

1

k

1

.

1

r

2

s

2

n

2

a

2

sp

4

e

8



Building a Tree

• While priority queue contains two

or more nodes– Create new node

–

CS 102

– Dequeue next node and make it right

subtree

– Frequency of new node equals sum of

frequency of left and right children

– Enqueue new node back into queue



Building a Tree

E i y l k . r s n a sp e

CS 102



Building a Tree

y

1

l

1

k

1

.

1

r

2

s

2

n

2

a

2

sp

4

e

8

CS 102

E1

i1

2



Building a Tree

y

1

l

1

k

1

.

1

r

2

s

2

n

2

a

2

sp

4

e

82

CS 102

E1 i1



Building a Tree

k

1

.

1

r

2

s

2

n

2

a

2

sp

4

e

82

CS 102

E1 i1

y1

l1

2



Building a Tree

k

1

.

1

r

2

s

2

n

2

a

2

sp

4

e

8

2

y l

2

CS 102

E1 i1



Building a Tree

r

2

s

2

n

2

a

2

sp

4

e

8

2 2

CS 102

E1 i1 1 1

k1

.1

2



Building a Tree

r

2

s

2

n

2

a

2

sp

4

e

8

22 2

CS 102

E1 i1y1

l1

k

1

.

1



Building a Tree

n

2

a

2

sp

4

e

8

22 2

CS 102

E1 i1 y1 l1 k1

.1

r2

s2

4



Building a Tree

n

2

a

2

sp

4

e

8

22

2 4

CS 102

E1 i1 y1

l1

k1

.1

r2

s2



Building a Tree

sp

4

e

8

2 2 2 4

CS 102

E1

i1

y

1 1 1

.

1 2 2

n2

a2

4



Building a Tree

sp

4

e

8

2 2 2 4 4

CS 102

E1

i1

y

1 1 1

.

12 2

2 2



Building a Tree

sp

4

e

8

2 4 4

CS 102

E1

i1

2

y1

l1

2

1

.

1 2 2 2 2

4



Building a Tree

sp

4

e

82 2

2 4 4 4

CS 102

E1

i1

y1

l1

1

.

1 2 2 2 2



Building a Tree

e

82 2r s

4

n a

4 4

CS 102

E1

i1

sp4

y1

l1

k1

.1

2

2 2 2 2

6



Building a Tree

sp

e

82

4 4 4 6

CS 102

E1

i1

4

y1

l1

k1

.1

r

2

s

2

n

2

a

2

What is happening to the characters

with a low number of occurrences?



Building a Tree

s

e

82 2 2

46

CS 102

E1

i1

4

y1

l1

k1

.1

r2

s2

4

n2

a2

4

8



Building a Tree

s

e

82 2 2

46 8

CS 102

E1

i1

4

y1

l1

k1

.1

r2

s2

n2

a2



Building a Tree

e

84 4

8

CS 102

E1

i1

sp4

2

y1

l1

2

k1

.1

2

r2

s2

n2

a2 4

6

10



Building a Tree

e

84 4 4

8 10

CS 102

E1

i1

sp4

2

y1

l1

2

k1

.1

2r2

s2

n2

a2

6



Building a Tree

4

10

16

CS 102

E1

i1

sp4

e8

2

y1

l1

2

k1

.1

2

r2

s2

4

n2

a2

4

8



Building a Tree

4

1016

CS 102

E1

i1

sp4

e82

y1

l1

2

k1

.1

2

r2 s2

4

n2 a2

4

8



Building a Tree

1016

26

CS 102

E

1

i

1

sp4

e8

2

y

1

l

1

2

k

1

.

1

2

r2 s2

4

n2 a2

4

4

68



Building a Tree

26

• After

enqueueingthis node

there is only

CS 102

E1

i1

sp

4

e8

2

y1

l1

2

k1

.1

2

r2

s2

4

n2

a2

4

46

8

10 one node left

in priority

queue.



Building a Tree

Dequeue the single node

left in the queue.

This tree contains the 1016

26

CS 102

character.

Frequency of root node

should equal number ofcharacters in text.

E1

i1

sp4

e8

2

y1

l1

2

k1

.1

2

r

2

s

2

4

n

2

a

2

4

46 8

Eerie eyes seen near lake. 26 characters

E di th Fil



Encoding the File

Traverse Tree for Codes

• Perform a traversal

of the tree to

obtain new code

words

26

CS 102

• Going left is a 0going right is a 1

• code word is only

completed when a

leaf node is

reached

E1

i1

sp

4

e8

2

y1

l1

2

k1

.1

2

r2

s2

4

n2

a2

4

46

8

10

E di th Fil



Encoding the File

Traverse Tree for Codes

Char CodeE 0000i 0001y 0010l 0011

26

CS 102

k 0100. 0101space 011e 10

r 1100s 1101n 1110a 1111

E1

i1

sp

4

e8

2

y1

l1

2

k1

.1

2

r2

s2

4

n2

a2

4

46

8

10



Encoding the File

• Rescan text and

encode file using

new code words


Char Code

E 0000i 0001y 0010l 0011

CS 102

k 0100. 0101space 011e 10

r 1100s 1101n 1110a 1111

1000101011011010011

1110101111110001100

1111110100100101

• Why is there no need

for a separator

character?

.

Encoding the File



Encoding the File

Results

• Have we made

things any better?

• 73 bits to encode

0000101100000110011

1000101011011010011

1110101111110001100

CS 102

the text• ASCII would take

8 * 26 = 208 bits

hIf modified code used 4 bits percharacter are needed. Total bits

4 * 26 = 104. Savings not as great.



Decoding the File

• How does receiver know what the codes are?

• Tree constructed for each text file.

– Considers frequency for each file

–

CS 102

,

files

• Tree predetermined

– based on statistical analysis of text files or

file types

• Data transmission is bit based versus byte

based



Decoding the File

• Once receiver has

tree it scans

incoming bit stream

• 0 ⇒⇒⇒⇒ go left10

16

26

CS 102

• 1 ⇒⇒⇒⇒ go right

E1

i1

sp4

e

82

y1

l1

2

k1

.1

2

r2

s2

4

n2

a2

4

6

101000110111101111

01111110000110101



Summary

• Huffman coding is a technique used

to compress files for transmission

• Uses statistical coding

CS 102

–

more frequently used symbols haveshorter code words

• Works well for text and fax

transmissions

• An application that uses several

data structures

Date post:	21-Feb-2018
Category:	Documents
Upload:	ines-sassi
View:	230 times
Download:	0 times

Huffman Algorithm

Documents