Lempel-Ziv-Welch

LZW

The most common form of compression is the LZW scheme. This is a lossless form of data compression and is commonly found in off-line compression/archiving utilities.

 

The LZ compression algorithm is a substitutional method ie a dictionary based encoding algorithm (also called a translation table). Basically the operation of the LZ compression is:

·        Patterns of data (substrings) are identified in the data file

·        These substrings are matched with dictionary entries

·        Then the dictionary item is written to the compressed output file

·        If the substring is not present in the dictionary a new dictionary entry is created and stored which can be reused at later re-occurrences.

 

Decompression (decoding) is the reverse of compression (encoding).

Pkzip varies on LZW compression by using a hash table to enhance performance. It gives it quicker access to the dictionary.

LZW is used in GIF, TIFF, PKZIP and Stuffit(used by Macintosh).

 

Determining the compression ratio

Compression software uses a range of different algorithms. The smaller the compressed file the longer it takes to compress.

 

Original Size - Compressed Size  x100%

                 Original Size

Some software programs quote their compression as a ratio. A ratio of 2:1 would mean 50% and 4:1 would mean 25%.