一种应用幅值信息的一单元定点复数ICA-R算法

李镜; 林秋华

doi:10.3724/SP.J.1146.2007.00608

一种应用幅值信息的一单元定点复数ICA-R算法

doi: 10.3724/SP.J.1146.2007.00608

李镜,
林秋华

基金项目:

国家自然科学基金(60402013)资助课题

计量
- 文章访问数: 3749
- HTML全文浏览量: 83
- PDF下载量: 1140
- 被引次数: 0
出版历程
- 收稿日期: 2007-04-20
- 修回日期: 2007-10-31
- 刊出日期: 2008-11-19

One-Unit Fixed-Point Complex-valued ICA-R Algorithm Using Magnitude Information

摘要

摘要: 参考独立分量分析(Independent Component Analysis with Reference, ICA-R)通过引入参考信号而实现期望实值源信号的抽取。然而，目前尚无复数域ICA-R算法。该文在约束ICA框架下，利用期望源信号的幅值信息提出了一种定点复数ICA-R算法，用于抽取某个期望的复数源信号。首先，采用复数fastICA算法的差异函数和关于复数信号幅值信息的不等式约束建立了复数ICA-R模型，然后采用增广朗格朗日函数和K-T条件推导了复数ICA-R定点算法。计算机仿真和性能分析结果表明，由于利用了幅值信息，复数ICA-R的估计性能优于传统的复数fastICA算法。
- 参考独立分量分析 /
- 独立分量分析 /
- 半盲分离 /
- 复数信号 /
- 参考信号
Abstract: Independent Component Analysis with Reference (ICA-R) extracts only desired signals by incorporating prior information as reference signals. It can provide output signals with definite order and improved performance. However, no ICA-R algorithm in complex domain has been reported till now. Motivated by the fact that the magnitude information of a complex-valued signal is readily obtained, this paper proposes a fixed-point complex-valued ICA-R algorithm to extract a desired signal by utilizing its magnitude information in the framework of constrained ICA. Specifically, the complex ICA-R is formulated as maximizing the contrast function of a blind complex fastICA algorithm under an inequality constraint corresponding to the magnitude information, the augmented Lagrangian function and Kuhn-Tucker conditions are then used to derive the fixed-point algorithm. The results of computer simulations and performance analysis demonstrate that the complex-valued ICA-R algorithm outperforms the blind complex fastICA algorithm by virtue of incorporation of magnitude information.
- I /
- n /
- d /
- e

HTML全文

参考文献(1)

[1] Comon P. Independent component analysis, a new concept?Signal Processing, 1994, 36(3): 287-314. [2] Hyvinen A, Karhunen J, and Oja E. Independentcomponent analysis. New York: John Wiley, 2001, chapter1,chapter8. [3] Cichocki A and Amari S. Adaptive Blind Signal and ImageProcessing: Learning Algorithms and Applications.Chichester: John Wiley, 2003, chapter1, chapter6. [4] Lin Q H, Zheng Y R, Yin F L, Liang H L, and Calhoun V D.A fast algorithm for one-unit ICA-R[J].Information Sciences.2007, 177(5):1265-1275 [5] Lu Wei and Rajapakse J C. ICA with Reference. Proc. ThirdInt. Conf. on Independent Component Analysis and BlindSource Separation (ICA2001), San Diego, California, 2001:120-125. [6] Lu Wei and Rajapakse J C. Approach and applications ofconstrained ICA[J].IEEE Trans. on Neural Networks.2005, 16(1):203-212 [7] Amari S, Douglas S C, Cichocki A, and Yang H H.Multichannel blind deconvolution and equalization using thenatural gradient. First IEEE Signal Processing Workshop onSignal Processing Advances in Wireless Communications.Paris, France, 1997: 101-104. [8] Calhoun V D, Adali T, Pearlson G D, VanZijl P C M, andPekar J J. Independent component analysis of fMRI data inthe complex domain[J].Magnetic Resonance in Medicine.2002,48(1):180-192 [9] Bingham E and Hyvinen A. A fast fixed-point algorithm forindependent component analysis of complex-valued signals.International Journal of Neural Systems, 2000, 10(1): 1-8. [10] Eriksson J and Koivunen V. Complex random vectors andICA models: identifiability, uniqueness, and separability[J].IEEE Trans. on Information Theory.2006, 52(3):1017-1029 [11] Cardoso J F and Adali T. The maximum likelihood approachto complex ICA. IEEE International Conference on Acoustics,Speech and Signal Processing. Toulouse, France, 2006, (5):673-676. [12] Hualiang Li and Adali T. Gradient and fixed-point complexICA algorithms based on kurtosis maximization. Proceedingsof the 16th IEEE Signal Processing Society Workshop onMachine Learning for Signal Processing. Maynooth, Ireland,2006: 85-90. [13] Luenberger D G. Optimization by Vector Space Methods.New York: John Wiley, 1969: 213-236. [14] Hyvinen A, Karhunen J, and Oja E. A fast fixed-pointalgorithm for independent component analysis. NeuralComputation, 1997, 9(7): 1483-1492. [15] 陈宝林. 最优化理论与算法. 第二版. 北京: 清华大学出版社,2005: 405-408.

施引文献

资源附件(0)

访问统计