A Canonical Huffman Compress Algorithm Independent of Tree Structure

The Huffman tree must be built in traditional Huffman coding and the tree must be saved for decoding . Owing to the difference in the length of Huffman codes decoding is inconvenient and time-consuming against the Huffman tree. Secondly every Huffman tree must have following structure code,data,weight,parent,l-child and r-child but only code and data are needed in decoding . It causes inessential space waste. In view of the above-mentioned facts a canonical Huffman compress algorithm that is not dependent on tree structure is put forward in this paper. Experi- ments prove that the process of coding and decoding is faster and the space needed in coding and decoding is fewer with this algorithm than traditional Huffman algorithm. It is a effective fast Huffman compress algorithm.