An Adaptive Multiple Order Context Huffman Compression Algorithm Based on Markov Model

Abstract

In this paper, an adaptive multiple order context Huffman compression algorithm based on Markov chain is proposed. Firstly, the data to be compressed is traversed, and the character space of the data and the times that one character transfers to its neighboring character are figured out. According to the statistical results, we can calculate the one-step transition probability matrix and the multi-step transition probability matrix. When the conditional probability between two adjacent characters is greater than the set threshold value, the adjacent characters are merged and compressed as an independent encoding unit. Improve the compression efficiency by increasing the length of the compression characters. The experimental results show that the algorithm achieves good compression efficiency.