计算机科学 ›› 2014, Vol. 41 ›› Issue (12): 78-81.doi: 10.11896/j.issn.1002-137X.2014.12.017
赵礼翔,刘国庆
ZHAO Li-xiang and LIU Guo-qing
摘要： 对于时间结构信号的盲源分离(Blind Source Separation,BSS),独立成分分析(Independent Component Analysis,ICA)是十分有效的方法。在对观测信号白化处理后,ICA的关键是寻找去除高阶相关性的正交分离矩阵。鉴于任意维数正交矩阵可以表示为Givens变换矩阵的乘积,提出了一种新的时间结构信号盲源分离算法。首先,利用Givens变换矩阵参数化表示正交分离矩阵,减少了要估计参数的个数；其次,以多步时延协方差矩阵的联合近似对角化为目标函数,将盲源分离问题转化为无约束优化问题,并利用拟牛顿法中的BFGS算法对Givens变换矩阵中的参数进行估计,得到分离矩阵；最后,以实际的混合语音信号分离做仿真实验,验证了该算法对时间结构信号的盲源分离是有效的。
| [1] Koldovsky Z,Tichavsky P.Time-domain blind separation of audio sources on the basis of a complete ICA decomposition of an observation space [J].Audio,Speech,and Language Processing,IEEE Transactions on,2011,19(2):406 [2] Gao Jing-bo,Zhu Xu,Lin Hai,et al.Independent component analysis based semi-blind I/Q imbalance compensation for MIMO OFDM systems [J].IEEE Transactions on Wireless Communications,2010,9(3):914 [3] Gao Jing-bo,Zhu Xu,Nandi A K.Independent component analysis for multiple-input multiple-output wireless communication systems [J].Signal Processing,2011,91(4):607 [4] Kleinsteuber M,Shen H.Blind source separation with compressively sensed linear mixtures [J].Signal Processing Letters,IEEE,2012,19(2):107 [5] Xie Sheng-li,Yang Liu,Yang Jun-mei,et al.Time-frequency approach to underdetermined blind source separation [J].IEEE Transactions on Neural Networks and Learning Systems,2012,23(2):306 [6] Tong L,Soon V C,Huang Yih-fang,et al.AMUSE:A new blindidentification algorithm[C]∥Circuits and Systems.New Or-leans,LA:IEEE,1990:1784-1787 [7] Belouchrani A,Abed-Meraim K,Cardoso J F,et al.A blindsource separation technique using second-order statistics [J].IEEE Transactions on Signal Processing,1997,45(2):434 [8] Rutledge D N,Jouan-Rimbaud Bouveresse D.Independent Components Analysis with the JADE algorithm [J].TrAC Trends in Analytical Chemistry,2013,50:22 [9] Wang Jian-gong,Shen Yue-hong,Li Hui,et al.A Separation Algorithm for Sources with Temporal Structure Only Using Second-order Statistics [J].Radioengineering,2013,22(3):861 [10] 汤影,李建平.利用参数表示任意维数正交矩阵的ICA新算法[J].自动化学报,2008,34(1):31 [11] Watkins D S.Fundamentals of Matrix Computations(Second Edition)[M].New York:John Wiley & Sons,2002:192-194 [12] Langlois D,Chartier S,Gosselin D.An introduction to indepen-dent component analysis:InfoMax and FastICA algorithm [J].Tutorials in Quantitative Methods for Psychology,2010,6(1):31 [13] 曹军宏,韦灼彬,高屹,等.基于量子遗传算法的盲源分离时延优选 [J].计算机工程,2012,38(11):170 [14] Hyvarinen A.Independence Component Analysis[M].New York:John Wiley & Sons,2001:342-344 [15] 解可新,韩健,林友联.最优化方法(修订版)[M].天津:天津大学出版社,2004:114-123 [16] Falie D,David L.Correlation coefficient based on independentcomponent analysis[C]∥2012 9th International Conference on Communications (COMM).IEEE,2012:59-62 [17] Bach F R,Jordan M I.Kernel independent component analysis [J].The Journal of Machine Learning Research,2003,3:1 |
| No related articles found! |
|
||
首页