Comparative Analysis of Lossless Text Compression Techniques

Abstract

Data compression is an effective means for saving storage space and channel bandwidth. There are two main types of compression lossy and lossless. This paper will deal with lossless compression techniques named Huffman, Arithmetic, LZ-78 and Golomb coding. The paper attempts to do comparative analysis in terms of their compression efficiency and speed. The test files used for this include English text files, Log files, Sorted word list and geometrically distributed data text file. The implementation results of these compression algorithms suggest the efficient algorithm to be used for a certain type of file to be compressed taking into consideration both the compression ratio and speed of operation. In terms of compression ratios, Golomb is best suited for very low frequency Text files, arithmetic for moderate and high frequency. Implementation is done using MATLAB software. General TermsText Compression.

Performance evaluation of arithmetic coding data compression for internet of things applications

“…According to the researchers, the Arithmetic algorithm produced a high compression ratio and used less memory space. However, the weakness of this algorithm is similar to previous project [7], which had slow compression and decompression processes.…”

“…A stream of input symbols are replaced with a single floating point number as the output [6]. According to Jacob, Somvanshi, and Tornekar [7], the Arithmetic algorithm is a very effective mechanism in eliminating redundancy in the encoding of data. Based on the explanation of the Arithmetic Coding in paper [6], the main objective of the Arithmetic Coding is to allocate an interval to each potential symbol.…”

“…Paper [7] studied the comparative analysis in terms of the compression efficiency. The authors dealt with lossless compression approaches such as the Huffman, Arithmetic, LZ-78, and Golomb Coding.…”

“…Based on paper [8], Porwal, Chaudhary, Joshi, and Jain focused on the lossless data compression methodologies and comparison of their performances. The authors used a traditional Arithmetic algorithm in their project, which was similar with projects [6] and [7]. However, they only used single-data compression during the comparative performance, such as the Huffman method.…”

“…One of the similarities of project [6] with the other projects is in the implementation of the Arithmetic algorithm method in the project. Project [6] used the traditional Arithmetic approach, which was also similar with projects [7] and [8]. Each of these projects used the text file as the dataset.…”