Title :
VLSI architectures for recursive and multiple-window order statistic filtering
Author :
Hakami, M. ; Warter, P. ; Boncelet, C. ; Nassimi, D.
Author_Institution :
Dept. of Electr. Eng., Delaware Univ., Newark, DE, USA
Abstract :
Based on a recently developed class of sorting networks, new VLSI architectures suitable for order statistic filtering are developed. The major advantage of these architectures is minimal response-time regardless of the number of stages in the pipeline; an effective characteristic for implementing recursive order statistic filters. The devised word-parallel architecture is the only one introduced to date that is capable of operating in both recursive and standard modes with optimal throughput. The proposed architectures are also suitable for implementing order statistic filters with multiple overlapping windows
Keywords :
VLSI; digital filters; parallel architectures; signal processing; statistical analysis; VLSI architectures; minimal response-time; multiple-window order statistic filtering; recursive order statistic filters; sorting networks; word-parallel architecture; Computer architecture; Digital filters; Digital signal processing; Filtering; Pipeline processing; Random variables; Sorting; Statistics; Throughput; Very large scale integration;
Conference_Titel :
Parallel Processing Symposium, 1992. Proceedings., Sixth International
Conference_Location :
Beverly Hills, CA
Print_ISBN :
0-8186-2672-0
DOI :
10.1109/IPPS.1992.223030