Title :
Code Compression using Huffman and Dictionary-based Pattern Blocks
Author :
Azevedo Dias, Wanderson Roger ; Moreno, Edward David
Author_Institution :
Coordenadoria de Inf., Inst. Fed. de Sergipe, Itabaiana, Brazil
fDate :
7/1/2015 12:00:00 AM
Abstract :
This paper presents a new code compression method where we merge the Huffman algorithm with the formation of patterns of blocks, which are found into the dictionary. We seek two consecutive blocks in two levels, one of them into compressed code and the other related to uncompressed code. We have used simulation with SimpleScalar and programs from MiBench. Our proposal has achieved a compression rate of 34.5% and 7% improvement in code compression when compared to the Huffman traditional method.
Keywords :
Huffman codes; data compression; Huffman algorithm; MiBench program; SimpleScalar simulation; code compression; consecutive blocks; dictionary-based pattern blocks; pattern block formation; Benchmark testing; Computational modeling; Computers; Dictionaries; Hardware; Random access memory; Reduced instruction set computing; Code Compression; Dictionary-based Pattern Blocks; Embedded Systems; Huffman; MiBench;
Journal_Title :
Latin America Transactions, IEEE (Revista IEEE America Latina)
DOI :
10.1109/TLA.2015.7273793
