Networking

Minimum Expected Length of Fixed-to-Variable Lossless Compression Without Prefix Constraints

Download Now Free registration required

Executive Summary

Lossless symbol-by-symbol compressors are required to satisfy the condition of "Unique decodability" whereby different input strings are assigned different compressed versions. Uniquely decodable non-prefix codes do not offer any advantages over prefix codes since any uniquely decodable code must assign lengths to the various symbols that satisfy Kraft's inequality, while a prefix code is guaranteed to exist with those symbol lengths. Achieved by the Huffman code, an exact expression for the minimum average length of a prefix symbol-by-symbol binary code is unknown.

  • Format: PDF
  • Size: 221.63 KB