Title :
Adaptive Quasi-Newton Algorithm for Source Extraction via CCA Approach
Author :
Wei-Tao Zhang ; Shun-Tian Lou ; Da-Zheng Feng
Author_Institution :
Sch. of Electron. Eng., Xidian Univ., Xi´an, China
Abstract :
This paper addresses the problem of adaptive source extraction via the canonical correlation analysis (CCA) approach. Based on Liu´s analysis of CCA approach, we propose a new criterion for source extraction, which is proved to be equivalent to the CCA criterion. Then, a fast and efficient online algorithm using quasi-Newton iteration is developed. The stability of the algorithm is also analyzed using Lyapunov´s method, which shows that the proposed algorithm asymptotically converges to the global minimum of the criterion. Simulation results are presented to prove our theoretical analysis and demonstrate the merits of the proposed algorithm in terms of convergence speed and successful rate for source extraction.
Keywords :
Newton method; blind source separation; CCA approach; Lyapunov method; adaptive Quasi-Newton algorithm; adaptive source extraction; canonical correlation analysis approach; quasi-Newton iteration; Algorithm design and analysis; Convergence; Correlation; Cost function; Prediction algorithms; Standards; Vectors; Blind source extraction (BSE); Lyapunov method; Newton iteration; canonical correlation analysis (CCA); stability;
Journal_Title :
Neural Networks and Learning Systems, IEEE Transactions on
DOI :
10.1109/TNNLS.2013.2280285