Electrical computers: arithmetic processing and calculating – Electrical digital calculating computer – Particular function performed
Patent
1998-02-05
1999-12-07
Mai, Tan V.
Electrical computers: arithmetic processing and calculating
Electrical digital calculating computer
Particular function performed
708300, G06F 1710
Patent
active
05999956&
ABSTRACT:
A source separation system (10) processes input signals formed by mixtures of primary signals originating from sources and estimates the primary signals so that the estimates do not differ even in the presence of non-stationary input signals. The system includes a first separation sub-assembly (12) which produces first estimates of the primary signals, a second sub-assembly (14) which adaptively determines the separation coefficients, and a third sub-assembly (15) which standardizes the first estimates and produces first standardized estimates used for the calculation of the separation coefficients. An output module (17) produces estimates which have between them the same proportionality ratio as that existing between the primary signals. A selection module (19) avoids an estimate being duplicated on various outputs in the case where certain primary signals are absent or very weak.
REFERENCES:
patent: 3696203 (1972-10-01), Leonard
patent: 5293405 (1994-03-01), Gersbach et al.
patent: 5652541 (1997-07-01), Yang et al.
Multi-layer Neural Networks WIth A Local Adaptive Learning Rule Ror Blind Separation of Source Signals, by A. Cichocki, W. Kasprzak, S. Amari, International Symposium on Nonlinear Theory And Its Applications (NOLTA '95) Las Vegas, Dec. 10-14, 1995, 1Ce-5, pp. 61 to 65.
Goodman Edward W.
Mai Tan V.
U.S. Philips Corporation
LandOfFree
Separation system for non-stationary sources does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Separation system for non-stationary sources, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Separation system for non-stationary sources will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-836130