An Efficient Memory Construction Scheme for an Arbitrary Side Growing Huffman Table

Abstract

By grouping the common prefix of a Huffman tree, in stead of the commonly used single-side rowing Huffman tree (SGHtree), we construct a memory efficient Huffman table on the basis of an arbitrary-side growing Huffman tree (AGH-tree) to speed up the Hufman decoding. Simulation results show that, in Huffman decoding, an AGH-tree based Huffman table is 2.35 times faster that of the Hashemian's method (an SGHtree based one) and needs only one-fifth the corresponding memory size. In summary, a novel Huffman table construction scheme is proposed in this paper which provides better performance than existing construction schemes in both decoding speed and memory usage.

Memory Efficient Hierarchical Lookup Tables for Mass Arbitrary-Side Growing Huffman Trees Decoding

“…1(d)] is only heuristics-based and, more importantly, the total memory access or memory size usage does not guarantee reaching to the optimal solution. A compact data structure to represent HASGT and the corresponding discussions can be found in [13].…”

“…Following the discussions given in Section II-C, we define the cost of as (13) The penalty, , and resource constraint, , of are decided by the current choice of op and the cost of its associated subtrees, thus the penalty of partitioned by op would be (14) in which is the probability associated with , and op partitions and induces . The suffix is decided by .…”

“…With the proposed metric, any optimal solution algorithm for finding a Lagrangian multiplier can be adopted. As to the second issue, [13] provided two hierarchical approaches for partitioning any arbitrary-side growing HT effectively in memory usage. Since optimizing a hierarchical partition plays a crucial role in partitioning an arbitrary-side-growing Huffman tree (AGH-tree), in Section III, we develop a Viterbi-like algorithm for searching optimal hierarchical structures in AGH-trees.…”