Title :
Maximizing the information transfer for adaptive unsupervised source separation
Author :
Castedo, Luis ; Macchi, Odile
Author_Institution :
Dept. de Electronica y Sistemas, Univ. de La Coruna, La Coruna, Spain
Abstract :
The problem of adapting linear multi-input-multi-output systems for unsupervised separation of linear mixtures of sources arises in a number of applications in multiuser wireless communications, such as mobile telephony. In this paper we propose a new statistical criterion to adapt the separating system. It involves the well-known Godard criterion as part of it and is interpreted by information theory as the maximization of information transfer in a single layer nonlinear neural network. The proposed criterion is free from undesirable stationary points provided that the signals to be separated have negative kurtosises, which is the case in communications.
Keywords :
MIMO systems; adaptive signal processing; neural nets; optimisation; statistical analysis; Godard criterion; adaptive unsupervised source separation; information theory; information transfer maximization; linear mixtures of sources; linear multi-input-multi-output systems; mobile telephony; multiuser wireless communications; negative kurtosises; single layer nonlinear neural network; statistical criterion; Information theory; MIMO; Mobile communication; Multiaccess communication; Neural networks; Sensor arrays; Sensor systems; Source separation; Telephony; Vectors;
Conference_Titel :
Signal Processing Advances in Wireless Communications, First IEEE Signal Processing Workshop on
Conference_Location :
Paris, France
Print_ISBN :
0-7803-3944-4
DOI :
10.1109/SPAWC.1997.630067