Increasing the Runs in Compressed Data Using a Dedicated Move to Front Algorithm

Abstract

Compressed data is a random data with high entropy and low redundancy, so they are hard for another round of compression. Reducing the randomness of compressed data is a step towards dealing with this problem. We detail a modification to the MTF (Move-to-front) algorithm that is able to reduce the randomness of compressed data by increasing the number of runs and its corresponding quality. The experiments confirmed that our modification results in better value and increased runs among randomly selected sequences. Two formats were tested; binary and alpha-betical representation. Both reported increasing runs and qualities. A common feature was found in the alphabetical representation. The ratios of the number of each simple symbol to its file size is similar in the case of the selected compression format.