The Research On A Lossless Data Compression Algorithm

With the coming of the informationization and the advent of the knowledge-based economy era, more and more data is needed for processing and transmission on the daily life, which inevitably brings redundant data. Redundant data is not only a waste of memory, but also on the retrieval of information brought unnecessary time overhead. Therefore, we urgently need to have good compression algorithms to eliminate unnecessary data redundancy.A compression algorithm based on arithmetic of common sense is presented. Judging by its realization it still belongs to the compression algorithms based on the dictionary model. The innovation of current algorithm is that from a broader perspective to understand the concept of redundancy of data. It is no longer simply to find and eliminate duplicate bit string, but for some of the longer string of data, we find some of the more streamlined method derived data and one more brief representation of the source data.The specific implementation plan based on the compression algorithm of arithmetic common sense is presented. The concept of optimal coding and optimal sequence and the properties of linear transformations based on the optimal coding and optimal sequence which should be satisfied and also have been proved are introduced. Wherein, based on knowledge of arithmetic compression algorithm and the encoding optimization plan combination, may require large amounts of memory to store the dictionary, compressed slow, but its nature has changed, no longer depends on the characteristics of the local data redundancy, but on the size of the dictionary. The decompression algorithm is just a arithmetic computation. Experiments show that, the compression algorithm of arithmetic common sense has a better compression effect for encoding unpredictable random data compared with LZ77 compression algorithm. And there is further room for optimization.