Title :
Parallel processing for biomedical signal processing
Author :
Manolakos, Elias S. ; Stellakis, Haris M. ; Brooks, Dana H.
Author_Institution :
Dept. of Electr. & Comput. Eng., Northeastern Univ., Boston, MA, USA
fDate :
3/1/1991 12:00:00 AM
Abstract :
The authors describe how systematic mapping methodologies can be used to derive special-purpose processor arrays for the estimation of the bispectrum via the third-order moments. A novel design that is optimal in terms of total execution time for multiple pipelined data blocks is proposed, and it is shown how formal verification of the design can be achieved analytically. The discussion focuses on mapping regular algorithms to parallel architectures, bispectrum estimation from the third-order moments, and systolic array design for the third-order moments.<>
Keywords :
biomedical electronics; computerised signal processing; digital signal processing chips; parallel architectures; systolic arrays; biomedical signal processing; bispectrum estimation; multiple pipelined data blocks; parallel architectures; parallel processing; regular algorithms; special-purpose processor arrays; systolic array design; third-order moments; Biomedical computing; Biomedical imaging; Biomedical signal processing; Digital signal processing; High performance computing; Parallel processing; Signal analysis; Signal processing; Signal processing algorithms; Spectral analysis;
