利用复旋转矩阵的非酉联合对角化算法

doi:10.3969/j.issn.1001-2400.2016.05.002

Abstract

Abstract:

The proposed algorithm is based on the combination of complex Givens and complex Shear rotations, which decomposes the de-mixing matrix into a product of complex rotation matrices. By ingenious utilization of the structure of the complex Givens and complex Shear rotations, together with the adequate expression of the concerned variables, the minimization of the F-norm criterion function can be transformed into a sequence of problems of calculating the generalized eigenvector of two 3×3 real symmetric matrices. Numerical simulations illustrate the good performance and convergence property of the proposed algorithm.

Key words: blind source separation, non-orthogonal joint diagonalization, complex Givens rotation, complex Shear rotation

LIU Wenjuan;FENG Dazheng. Non-unitary joint diagonalization by complex Givens and complex Shear rotation[J].Journal of Xidian University, 2016, 43(5): 6-11.

References

［1］ SOULOUMIAC A. Nonorthogonal Joint Diagonalization by Combining Givens and Hyperbolic Rotations ［J］. IEEE Transactions on Signal Processing, 2009, 57(6): 2222-2231.
［2］ CHABRIEL G, KLEINSTEUBER M, MOREAU E, et al. Joint Matrices Decompositions and Blind Source Separation: a Survey of Methods, Identification, and Applications ［J］. IEEE Signal Processing Magazine, 2014, 31(3): 34-43.
［3］ ZIEHE A, LASKOV P, NOLTE G, et al. A Fast Algorithm for Joint Diagonalization with Non-orthogonal Transformations and Its Application to Blind Source Separation ［J］. Journal of Machine Learning Research, 2004, 5(3): 777-800.
［4］ XU X F, FENG D Z, ZHENG W X. A Fast Algorithm for Nonunitary Joint Diagonalization and Its Application to Blind Source Separation ［J］. IEEE Transactions on Signal Processing, 2011, 59(7): 3457-3463.
［5］ MESLOUB A, ABED-MERAIM K, BELOUCHRANI A. A New Algorithm for Complex Non-orthogonal Joint Diagonalization Based on Shear and Givens Rotations ［J］. IEEE Transactions on Signal Processing, 2014, 62(8): 1913-1925.
［6］ GONG X F, WANG K, LIN Q H. Complex Non-orthogonal Joint Diagonalization with Successive Givens and Hyperbolic Rotations ［C］//Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway: IEEE, 2012: 1889-1892.
［7］ MAURANDI V, MOREAU E. A Decoupled Jacobi-like Algorithm for Non-unitary Joint Diagonalization of Complex-valued Matrices ［J］. IEEE Signal Processing Letters, 2014, 21(12): 1453-1456.
［8］ GONG X F, WANG X L, LIN Q H. Generalized Non-orthogonal Joint Diagonalization with LU Decomposition and Successive Rotations ［J］. IEEE Transactions on Signal Processing, 2015, 63(5): 1322-1334.
［9］ YEREDOR A. Non-orthogonal Joint Diagonalization in the Least-squares Sense with Application in Blind Source Separation ［J］. IEEE Transactions on Signal Processing, 2002, 50(7): 1545-1553.
［10］ FENG D Z, ZHANG H, ZHENG W X. Bi-iterative Algorithm for Extracting Independent Components from Array Signals ［J］. IEEE Transactions on Signal Processing, 2011, 59(8): 3636-3646.
［11］ TICHAVSKY P, YEREDOR A. Fast Approximate Joint Diagonalization Incorporating Weight Matrices ［J］. IEEE Transactions on Signal Processing, 2009, 57(3): 878-891.
［12］ ZENG T J, FENG Q Y. A Parallel Dual Matrix Method for Blind Signal Separation ［J］. Neural Computation, 2014, 26(3): 592-610.
［13］ JOHO M. Newton Method for Joint Approximate Diagonalization of Positive Definite Hermitian Matrices ［J］. SIAM Journal on Matrix Analysis ＆ Applications, 2008, 30(3): 1205-1218.
［14］ CARDOSO J F, SOULOUMIAC A. Blind Beamforming for Non-Gaussian Signals ［J］. IEE Proceedings F: Radar and Signal Processing, 1993, 140(6): 362-370.

[1]	ZHAO Wei,CHOU Shengnan,YANG Shuo,LI Xiongfei. Anti-jamming algorithm for spread spectrum communication using blind source separation [J]. Journal of Xidian University, 2020, 47(1): 73-79.
[2]	WANG Chuanchuan,ZENG Yonghu,FU Weihong,WANG Liandong. Estimation algorithm for an underdetermined mixing matrix based on maximum density point searching [J]. Journal of Xidian University, 2019, 46(1): 106-111.
[3]	FU Weihong;WANG Lu;MA Lifen. Improved laplace mixed model potential function algorithm for UBSS [J]. J4, 2014, 41(6): 1-5+88.
[4]	JI Jian;LI Xiao. Method for sparse component analysis in the shearlet domain [J]. J4, 2014, 41(1): 45-52+146.
[5]	YIN Hai-qing;LIU Hong-wei. New blind source separation algorithm based on L₁ sparse regularization and nonnegative matrix factorization [J]. J4, 2010, 37(5): 835-841.
[6]	ZHANG Wei-tao;LOU Shun-tian;ZHANG Yan-liang. SVD based robust approach for blind source separation [J]. J4, 2009, 36(6): 985-989.
[7]	ZHANG Wei-tao;LIU Ning;LOU Shun-tian. Two efficient algorithms for joint diagonalization with exception of the singular solution [J]. J4, 2009, 36(5): 857-861.
[8]	Zhang Yanliang1,LOU Shuntian1,ZHANG Weitao2. Blind source separation method applicable to dependent sources [J]. J4, 2009, 36(3): 401-432.
[9]	ZHANG Wei-tao1;LOU Shun-tian1;ZHANG Yan-liang1;2. Adaptive Least-Squares method for blind source separation with an equivariant property [J]. J4, 2008, 35(6): 999-1002.
[10]	WANG Fa-song1;3;LI Hong-wei1;LI Rui2. Efficient non-parametric GKNN independent component analysis algorithm [J]. J4, 2008, 35(4): 764-768.
[11]	LIU Jian-qiang;FENG Da-zheng. Block recursive blind source separation method [J]. J4, 2008, 35(2): 233-236.
[12]	ZHANG Hua;FENG Da-zheng;NIE Wei-ke;XU Xian-feng. Non-orthogonal joint diagonalization for blind source separation [J]. J4, 2008, 35(1): 27-317.
[13]	ZHAO Zhi-jin1;2;XIE Ting-ting2;LI Xiao-ping3;ZHAO Zhi-dong2. Blind source separation method in the wavelet domain [J]. J4, 2007, 34(3): 423-427.
[14]	YE Ji-min(1);JIN Hai-hong(2);LOU Shun-tian(1);ZHANG Xian-da(1). Projected natural gradient-based BSS algorithm with an unknown source number [J]. J4, 2006, 33(2): 190-194.
[15]	ZHANG Jun-ying1;2;LIU Li-ping1. Blind signal separation based on the partially independent component analysis [J]. J4, 2004, 31(3): 334-337.

Non-unitary joint diagonalization by complex Givens and complex Shear rotation

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Metrics

Comments

Recommended 10