A Comparative Study on the Performance of Huffman Coding and Arithmetic Coding

This paper conducts a systematic and comprehensive comparison of the performance differences between Huffman coding and arithmetic coding in terms of coding efficiency, time/space complexity, and practical application scenarios through in - depth theoretical derivations and rigorous experimental verifications. Three typical datasets (English text, binary data, and DNA sequences) are used in the experiments, which are carried out in a unified experimental environment. The results show that the average compression ratio of arithmetic coding is increased by 12.3% compared with that of Huffman coding, but the encoding time is increased by 2.8 times; the decoding speed of Huffman coding reaches 4.1 times that of arithmetic coding. The research further proposes an efficient implementation scheme based on Java, and elaborates on the key code implementation process, verifying the advantages of Huffman coding in real - time systems and the applicability of arithmetic coding in scenarios with high compression requirements. This paper provides a quantitative decision - making basis for the selection of coding algorithms in engineering practice, and also explores the optimization directions of the two coding algorithms and future research trends.