Programmable Architectures for Realtime Music Decompression

Authored by

Martin Botteck, Holger Blume, Jörg Von Livonius, Martin Neuenhahn, Tobias G. Noll

Abstract

Field Programmable Gate Array Architectures (FPGA) are known to facilitate efficient implementations of signal processing algorithms. Their computational efficiency for arithmetic datapaths in terms of power and performance is typically ranked better than that of general purpose processors. This paper analyses the efficiency gain for MP3 decoding on an FPGA. The results of an exemplary FPGA implementation of a complete MP3 decoder featuring arithmetic datapaths as well as control overhead are compared concerning performance and power consumption to further implementation alternatives. Furthermore, the results are compared to requirements for a deployment in portable environments.

Details

External Organisation(s): Nokia Corporation
RWTH Aachen University
Type: Conference contribution
Pages: 777-784
No. of pages: 8
Publication date: 2008
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: General Computer Science