Blind Source Separation Algorithm Based on Restructuring a New Separation Performance Index

Abstract:

In view of the contradiction between convergence speed and steady state error in blind source separation algorithm, an adaptive algorithm based on improved separation performance index is proposed. The proposed algorithm restructures a new performance index by adding a parallel system. According to the corresponding relationship between the performance index and the step-size, then a new step-size equation with the Rayleigh distribution function is defined. Simulation results show that the convergence speed and steady state error of the proposed algorithm is better than the classic blind source separation algorithm and the sigmoid function variable step-size blind source separation algorithm in different environments.

• Periodical: Advanced Materials Research (Volumes 1030-1032)
• Pages: 1676-1679
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.1030-1032.1676
• Online since: September 2014
• Authors: Ye Cai Guo*, Zheng Zhang
• Keywords: Blind Source Separation, Parallel System, Performance Index, Variable Step-Size

References

[1] Y.B. Quan, J.F. Wang and Y. Yue: Communications Technology, Vol. 44 (2011), p.13.

[2] J.Q. Liu, D.Z. Feng and W.W. Zhang: Neural computation, Vol. 21(2009), p.872.

[3] A. Bell. and T. Sejnowski: Neural Computation, Vol. 7(1995), p.1129.

[4] Y. Chen and X.L. Wen: Advanced Science Letters, Vol. 7 (2012), p.440.

[5] S.F. OU, Y. Gao and G. Jin: Natural Computation, Vol. 3 (2009), p.649.

[6] H. Yang and S. Amari: Neural computation, Vol. 9 (1997), p.1457.

[7] S.Y. Sun, J.L. Zheng, M. Zhao and Q. Zhang: Journal on Communications, Vol. 25 (2004), p.132.

[8] W.T. Zhang, S.T. Lou and Y.L. Zhang: Journal of Xidian University, Vol. 35 (2008), p.999.

[9] S.T. Hsieh, T.Y. Sun and C.L. Lin: Evolutionary Computation, Vol. 12 (2008), p.242.