1. Science
  2. Publications
  3. Information Processing Systems
  4. 4(150)'2017
  5. Entropy recursive group coding for double-byte alphabets

Entropy recursive group coding for double-byte alphabets

N. Kozhemiakina, N. Ponomarenko
Annotations languages:

In the paper a modification of entropy recursive group coding (ERGC) that by usage of dynamic frequency modeling and fixed sizes of group allows in the first iteration to code double-byte symbols is proposed. A model of test text with double-byte symbols allowing to approve effectiveness of the modification of ERGC is described. It is shown that the proposed modification provides larger compression ratios not only than arithmetical coding and Huffman coding but also than such high-level compression methods as WinRar and PAQ8.
Keywords: recursive group coding, entropy coding, arithmetical coding, Huffman coding, dynamic frequency modeling