A combined Kalman filter and natural gradient algorithm approach for blind separation of binary distributed sources in time-varying channels

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 5, Issue , 2001, Pages 2769-2772

  Jafari, M G ^a,b   Seah, H W ^a,b   Chambers, J A ^a,b  

^a IMPERIAL COLLEGE LONDON   (United Kingdom)

^b UNIVERSITY OF BATH   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

ALGORITHMS; COMMUNICATION CHANNELS (INFORMATION THEORY); ERROR ANALYSIS; GRADIENT METHODS; KALMAN FILTERING; LEARNING SYSTEMS; MATRIX ALGEBRA; TIME VARYING SYSTEMS; VECTORS;

ADAPTIVE LEARNING RATE; BINARY DISTRIBUTED SIGNALS; MINIMUM MEAN SQUARE ERROR; NATURAL GRADIENT ALGORITHM;

BLIND SOURCE SEPARATION;

EID: 0034852707     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2001.940220     Document Type: Article

References (8)

1. Adaptive blind signal processing - Neural network approaches
2. Isolating epileptiform discharges in the unaveraged EEG using independent component analysis
3. Blind source separation of real world signals
4. A new learning algorithm for blind signal separation
5. Survey on independent component analysis
6. On-line adaptive algorithms in non-stationary environments using a modified conjugate gradient approach