Parallel bit-stream neurohardware for blind separation of sources

Author

Fanghänel, Karsten ; Köllmann, Kuno ; Zeidler, H.C. ; Plessmann, Ralf ; Riemschneider, Karl-Ragmar

Author_Institution

Univ. der Bundeswehr Hamburg, Germany

Volume

5

fYear

1997

fDate

21-24 Apr 1997

Firstpage

4149

Abstract

The use of digital stochastic computing is proposed to realize a recurrent network for blind separation of undelayed linearly superposed signals into their original components. The stochastic representation of signals allows one to design very simple digital processing elements to implement all the operations necessary for the algorithm of Herault and Jutten (1986). A first hardware implementation has been designed and the experimental results are presented

Keywords

CMOS digital integrated circuits; VLSI; digital signal processing chips; learning (artificial intelligence); neural chips; neural net architecture; parallel architectures; recurrent neural nets; signal representation; stochastic processes; VLSI; blind source separation; digital CMOS; digital processing elements; digital stochastic computing; experimental results; hardware implementation; learning algorithm; parallel bit-stream neurohardware; recurrent network; stochastic signal representation; undelayed linearly superposed signals; Algorithm design and analysis; Clocks; Computer networks; Hardware; Nonlinear distortion; Sensor arrays; Signal design; Signal processing; Signal processing algorithms; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on

Conference_Location

Munich

ISSN

1520-6149

Print_ISBN

0-8186-7919-0

Type

conf

DOI

10.1109/ICASSP.1997.604860

Filename

604860