Title :
A dual-linear predictor approach to blind source extraction for noisy mixtures
Author :
Liu, Wei ; Mandic, Danilo P. ; Cichocki, Andrzej
Author_Institution :
Dept. of Electron. & Electr. Eng., Sheffield Univ., Sheffield
Abstract :
A second-order statistics based dual-linear predictor structure is proposed for blind source extraction from noisy instantaneous mixtures. The noise component is assumed to be spatially and temporally white, but the variance information of noise is not required. A detailed proof of the proposed approach is provided and an adaptive algorithm is developed. Simulation results show that it can extract the source signals successfully.
Keywords :
adaptive estimation; blind source separation; adaptive algorithm; blind source extraction; blind source separation; dual-linear predictor approach; noisy instantaneous mixtures; second-order statistics; Adaptive algorithm; Algorithm design and analysis; Data mining; Educational institutions; Higher order statistics; Signal processing; Signal processing algorithms; Source separation; Vectors; Working environment noise; Blind Source Extraction; Linear Predictor; Noisy Mixtures; Second-order Statistics;
Conference_Titel :
Sensor Array and Multichannel Signal Processing Workshop, 2008. SAM 2008. 5th IEEE
Conference_Location :
Darmstadt
Print_ISBN :
978-1-4244-2240-1
Electronic_ISBN :
978-1-4244-2241-8
DOI :
10.1109/SAM.2008.4606924
