An SGH-tree based memory efficient and constant decoding algorithm for Huffman codes

摘要

We proposed an SGH-based data structure to obtain a memory efficient and constant Huffman decoding algorithm. In cooperation with the Aggarwal and Narayan's algorithm, the proposed data structure spends a constant time on decoding each symbol and needs less memory size. In addition, we derive some properties to reduce the proposed codeword searching time. We believe that the proposed approach is useful for various multimedia coding applications, especially when the applications are conducted on memory constrained devices.