An Improved ICA Algorithm Based on the Negative Entropy and Simulated Annealing Algorithm

Abstract:

The Independent Component Analysis (ICA) is a classical algorithm for exploring statistically independent non-Gaussian signals from multi-dimensional data, which has a wide range of applications in engineering, for instance, the blind source separation. The classical ICA measures the Gaussian characteristic by kurtosis, which has the following two disadvantages. Firstly, the kurtosis relies on the value of samples, and is not robust to outliers. Secondly, the algorithm often falls into local optima. To address these drawbacks, we replace the kurtosis by negative entropy, utilize the simulated annealing algorithm for optimization, and finally propose an improved ICA algorithm. Experimental results demonstrate that the proposed algorithm outperforms the classical ICA in its robustness to outliers and convergent rate.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 1125-1128
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.1125
• Online since: September 2013
• Authors: Hong Yi Li, Meng Ye, Di Zhao
• Keywords: Blind Source Separation, ICA, Negative Entropy, Simulated Annealing Algorithm

References

[1] Jia Yin-Jie, Xu Peng-Fei. Blind separation of mixed audio signal based on FastICA [J]. Information and Electronic Engineering, 2009, 7(4): 321-325 (in Chinese).

[2] Dai Huahe, Jiang Shouda. Atrial fibrillation wave extraction algorithm based on ICA and wavelet transform[J]. Chinese Journal of Scientific Instrument, 2011, 32(8): 1716-1723.

[3] YU Ren, CHEN Zhi, ZHANG Zi, LIU Liang-hui. A WAV-ICA Based On-Line Fault Diagnoisis Method for Redundant Sensors in Nuclear Power Plants[J]. Nuclear Power Engineering, 2012, 33(3): 109-120.

[4] Chen Xiao-Jun, Cheng Hao, Tang Bin. Underdetermined Blind Rador Signal Seperation Based on ICA[J]. Journal of Electronic & Information Technology, 2010, 32(4): 919-924. (in Chinese).

DOI: 10.3724/sp.j.1146.2009.00291

[5] Zhang Deng-Ke. Approaches for Checking Determinnancy in ICA-Based BSS. ACTA ELECTRONICA SINICA, 2012, 40(11): 2303-2306. (in Chinese).

[6] WAN Jun, ZHANG Xiao-hui, RAO Jiong-hui. Detecting of Wake Flow Backscattering Optical Signal Based on PCA and ICA[J]. Fire Control & Command Control, 2012, 37(7): 121-127.

[7] Shi Jiyong, Zou Xiaobo, Zhao Jiewen, Yin Xiaoping. Measurement of chlorophyll content in cucumber leaves based on GA-ICA and hyper-spectral imaging technique[J]. JOURNAL OF JIANGSU UNIVERSITY, 2011, 32(2): 134-139.

[8] Hyvarinen A., Karhunen J., Oja E. Independent Component Analysis [M]. New York: John Wiley & Sons, Inc., (2001).

[9] Hyvarinen A. Survey on Independent Component Analysis [J]. Neural Computing Surveys, 1999(2): 94-128.

[10] Ji Zhi-Hua, Cai Kun-Bao, Ye Yan-Fang. Blind Separation of Image Based on ICA [J]. Journal of Chongqing Institute of Technology, 2008, 22(4): 104-107 (in Chinese).

[11] MENG Ling-ling, SUN Chang-dong, HAN Bao-ru. Algorithm for inter-harmonic detection based on least square method and ICA[J]. Power System Protection and Control. 2012, 40(11): 77-81.

[12] Liang Liang, Yang Ming-Hua, Li Ying-Fang. Hyperspectral Remote Sensing Image Classification Based on ICA and SVM Algorithm[J]. Spectroscopy and Spectral Analysis, 2010, 30(10): 2724-2728. (in Chinese).

[13] LU Zhao, WU Xiao-pei, LI Mi. Speech feature enhancement based on frequency-domain ICA[J]. JOURNAL OF VIBRATION AND SHOCK, 2011, 30(2): 238-257.

[14] Yang Fu-Sheng, Hong Bo. The Principle and Application of Independent Component Analysis[M]. Beijing: Tsing Hua University Press, 2006. (in Chinese).

[15] Balram S., Nazish H., Shweta J. Orthogonal Simulated Annealing for multi-objective optimization [J]. Computers & Chemical Engineering, 2010, 34(10), 1618-1631.

[16] Balram S. Study of simulated annealing based algorithms for multi-objective optimization of a constrained problem [J]. Computers & Chemical Engineering, 2004, 28(9), 1849-1871.

DOI: 10.1016/j.compchemeng.2004.02.037

[17] Zhang Xu-Xiu, Qiu Tian-Shuang. Random Variable Analogy Based ICA Method[J]. Journal of Electronic & Information Techonology, 2009, 31(9): 2104-2108. (in Chinese).

[18] ZHANG He-fa, LI Li-ping. Expectation-Maximization Algorithm for Noisy Independent Component Analysis[J]. Journal of University of Electronic Science and Technology of China, 2012, 41(4): 527-531.

[19] QIU Tian-shuang, LI Bing, ZHA Dai-hua. Elimination of pulse-noise from mixed-noise based on fraction lower order ICA[J]. Journal on Communication, 2011, 32(9): 78-81.