An improvement of dynamic Huffman coding with a simple repetition finder

A mixed-mode file compression scheme that incorporates interval encoding for finding duplicate occurrences of strings without storing or parsing the past sequence and a one-pass scheme for dynamic Huffman codes is presented. Results from experiments performed on various types of files are discussed. The proposed method runs in linear time, and the memory requirement depends only on the alphabet size. The code efficiency is compared experimentally to that of other schemes.<
>