Title :
Random access in Huffman-coded files
Author_Institution :
AT&T Bell Labs., Murray Hill, NJ, USA
Abstract :
Presents a technique for building an index into a Huffman-coded file that permits efficient random access to the encoded data. The technique provides the ability to find the starting position of the jth symbol of the uncompressed file in an n-bit compressed file in O(log n) bit-examinations of the compressed file plus its index. Furthermore, the size of the index is o(n) bits. In other words, the ratio of the space occupied by the index to the space occupied by the data approaches zero as the length of the data file increases without bound.<>
Keywords :
Huffman codes; data compression; file organisation; Huffman coded files; compressed file; data file; index size; random access; starting position; uncompressed file; Arithmetic; Costs; Database systems; Decoding; Encoding; Jacobian matrices;
Conference_Titel :
Data Compression Conference, 1992. DCC '92.
Conference_Location :
Snowbird, UT, USA
Print_ISBN :
0-8186-2717-4
DOI :
10.1109/DCC.1992.227444