ВДОСКОНАЛЕННЯ МЕТОДУ КОМПРЕСІЇ ДАНИХ НА ОСНОВІ КОДУ ФІБОНАЧЧІ

Abstract

This article discusses the effectiveness of data compression based on Fibonacci code. The prob-lem of using variable-length code for data compression is studied on the example of a test message encoding. The general principle of data compression, its variants are considered and the connection between the distribution of the frequency of the use of certain letters and the possibility of applying compression on the basis of codes of dynamic size is analyzed. Also, the general principle of constructing the coding tables for this method is considered, and practical calculations of the efficiency of the method for cases of symbols distribution according to statistics and at a uniform distribution of symbols are carried out. Based on calculations of compression efficiency in different cases, the main disadvantages of the method are outlined for further improvement of the algorithm. Taking into account certain deficiencies, the compression method on the basis of Fibonacci codes has been improved, which ensures the increase of the efficiency of real data compression.A method for increasing the compression efficiency by means of Fibonacci code with a large number of unique symbols is offered and the analysis of the efficiency of the proposed method compared with the standard method has been conducted. It has been determined that the improved excep-tion method has a higher percentage of compression over the standard one, but the results are significantly dependent on the frequency division of the symbols. The improved method is directed on the compression of texts using multiple languages, special symbols, as well as on the compression of files of any format where the standard method has a small or even negative compression ratio.