DocumentCode :
3116739
Title :
Design of a processor optimized for syntax parsing in video decoders
Author :
Siret, Nicolas ; Nezan, Jean-François ; Rhatay, Aimad
Author_Institution :
Eur. Univ. of Brittany, Brest, France
fYear :
2011
fDate :
2-4 Nov. 2011
Firstpage :
1
Lastpage :
8
Abstract :
Heterogeneous platforms aim to offer both performance and flexibility by providing designers processors and programmable logical units on a single platform. Processors implemented on these platforms are usually soft-cores (e.g. Altera NIOS) or ASIC (e.g. ARM Cortex-A8). However, these processors still face limitations in terms of performance compared to full hardware designs in particular for real-time video decoding applications. We present in this paper an innovative approach to improve performance using both a processor optimized for the syntax parsing (an Application-Specific Instruction-set Processor) and a FPGA. The case study has been synthesized on a Xilinx FPGA at a frequency of 100 MHz and we estimate the performance that could be obtained with an ASIC.
Keywords :
application specific integrated circuits; decoding; field programmable gate arrays; microprocessor chips; video coding; ARM Cortex-A8; ASIC; Altera NIOS; Xilinx FPGA; application specific instruction set processor; frequency 100 MHz; heterogeneous platforms; processor design; programmable logical units; real-time video decoding; soft-cores; syntax parsing; video decoders; Computer architecture; Decoding; Hardware; Registers; Software; Streaming media; Syntactics; Application-Specific Instruction-set Processor (ASIP); Instruction Set Architecture (ISA); bit manipulation operations; circuit design; embedded processor; multimedia; processor architecture; video decoding;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Design and Architectures for Signal and Image Processing (DASIP), 2011 Conference on
Conference_Location :
Tampere
Print_ISBN :
978-1-4577-0620-2
Electronic_ISBN :
978-1-4577-0619-6
Type :
conf
DOI :
10.1109/DASIP.2011.6136850
Filename :
6136850
Link To Document :
بازگشت