13 January 2020 UST, YEMEN 1

• Arithmetic coding is a more modern coding method that

usually outperforms Huffman coding in practice.

• Arithmetic coding yields a single codeword for each encoded

string of characters

• The first step is to divide the numeric range from 0 to 1 into a

number of different characters present in the message to be
sent

• The size of each segment by the probability of the related

character
13 January 2020 UST, YEMEN 2
• A single code is given for each string of characters

13 January 2020 UST, YEMEN 3

13 January 2020 UST, YEMEN 4
At the termination character ‘ . ’ is encoded, the segment range from 0.81602 to 0.81620
and hence the codeword for the complete string is any number within the range:
0.81602 ≤ codeword<0.81620
For example, the codeword=(0.81604004)10=(0.1101000011101)2 . As compared with
ASCII code:
13 𝑏𝑖𝑡𝑠
𝐶𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑖𝑜𝑛 𝑟𝑎𝑡𝑖𝑜 𝑐𝑟 = ∗ 100% = 37.14%
7 𝑏𝑖𝑡𝑠/𝑠𝑦𝑚𝑏𝑜𝑙 ∗ 5 𝑠𝑦𝑚𝑏𝑜𝑙
13 January 2020 UST, YEMEN 5
13 January 2020 UST, YEMEN 6
13 January 2020 UST, YEMEN 7
• The Lempel-Ziv-Welch (LZW) algorithm
employs an adaptive, dictionary-based
compression technique.

• Unlike variable-length coding, in which the

lengths of the codewords are different, LZW uses
fixed-length codewords to represent variable
length strings of symbols/characters that
commonly occur together, such as words in
English text.
13 January 2020 UST, YEMEN 8
• The LZW encoder and decoder builds up the
same dictionary dynamically while receiving
the data—the encoder and the decoder both
develop the same dictionary.
• LZW proceeds by placing longer and longer
repeated entries into a dictionary, then emitting
the code for an element rather than the string
itself, if the element has already been placed in
the dictionary.

13 January 2020 UST, YEMEN 9

 LZW Encoding Algorithm

13 January 2020 UST, YEMEN 10

 Example …

The following test string is to be compressed

with the help of the LZW compression
algorithm:
ABRAKADABRAABRAKADABRA
a) Show the compression process in a table.
b) How much the compression ratio as
compared with ASCII code (8 bit/character).

13 January 2020 UST, YEMEN 11

 While the original character string needs 176 bits (22 characters of each 8 bits with ASCII code),
so 𝐵0 = 22 × 8 = 176 𝑏𝑖𝑡𝑠. After LZW compression, there only 13 characters. There each
character must be represented by 12 bit, so 𝐵1 = 13 × 12 = 156 𝑏𝑖𝑡𝑠 (Note: you can consider
only 9 bits per character, but the computer uses the hexadecimal digits).
𝐵1 156
𝑐𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑖𝑜𝑛 𝑟𝑎𝑡𝑖𝑜% = × 100% = ∗ 100% = 88.64%
𝐵0 176
13 January 2020 UST, YEMEN 12
 LZW Decoding Algorithm

13 January 2020 UST, YEMEN 13

When a LZW compressed file is decoded, the dictionary can gradually be reconstructed . This is because the
output of the LZW algorithm always contains only dictionary entries that were already in the dictionary. In
compression, each dictionary entry begins with the last character of the previously added dictionary entry.
Conversely, the last character of a new dictionary entry, added to the dictionary in decompression, is at the same
time the first output character in decoding. The following table illustrates the sequence of LWZ decompression.
The first column contains the sequence of each of the code characters to be decoded . The second column shows
the output of decompression and the third column contains the current dictionary entry added consisting of a
character string and its associated code.

13 January 2020 UST, YEMEN 14

 Assignment

Given an initial dictionary:

Index Entry
1 a
2 b
3 h
4 i
5 s
6 t
and output of an LZW encoder:
6 3 4 5 1 3 1 6 2 9 11 16
decode the above sequence (which is not intended to represent
meaningful English).
13 January 2020 UST, YEMEN 15
 Just For Thinking

13 January 2020 UST, YEMEN 16