Title :
Robust source separation using ranks
Author :
Xiang, Lan ; Zhang, Yinglu ; Kassam, Saleem A.
Author_Institution :
Dept. of Electr. Eng., Pennsylvania Univ., Philadelphia, PA, USA
Abstract :
Robustness against deviations from nominal source PDF assumptions is very desirable in blind source separation (BSS) algorithms. A new approach for robust BSS is proposed. We modify the EASI (equivariant adaptive separation by independence) algorithms to use ranks of observed signals. Two different methods for evaluation of ranks are introduced. Our modified algorithm can be applied to both real-valued data and complex-valued data. Design guidelines are discussed for the nonlinear rank weighting functions in the modified algorithm. Simulation results and some examples are given, showing very good performance
Keywords :
adaptive estimation; adaptive signal processing; identification; probability; EASI algorithms; PDF; blind source separation; complex-valued data; equivariant adaptive separation by independence; nonlinear rank weighting functions; observed signals; performance; rank evaluation; real-valued data; robust source separation; simulation results; Algorithm design and analysis; Approximation algorithms; Blind source separation; Guidelines; Independent component analysis; Probability density function; Robustness; Signal processing; Source separation;
Conference_Titel :
Statistical Signal and Array Processing, 2000. Proceedings of the Tenth IEEE Workshop on
Conference_Location :
Pocono Manor, PA
Print_ISBN :
0-7803-5988-7
DOI :
10.1109/SSAP.2000.870137
