Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes

doi:10.19682/j.cnki.1005-8885.2020.0047

中国邮电高校学报(英文) ›› 2020, Vol. 27 ›› Issue (6): 73-86.doi: 10.19682/j.cnki.1005-8885.2020.0047

• Image Processing • 上一篇下一篇

Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes

王川川,姜林,曾勇虎,汪连栋

CEMEE实验室

收稿日期:2020-05-08 修回日期:2020-11-24 出版日期:2020-12-31 发布日期:2020-12-31
通讯作者: 姜林 E-mail:328458366@qq.com
基金资助:
国家自然科学基金;国家973项目

Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes

Wang Chuanchuan, Jiang Lin, Zeng Yonghu, Wang Liandong

Received:2020-05-08 Revised:2020-11-24 Online:2020-12-31 Published:2020-12-31

摘要/Abstract

摘要：

To solve the problem of mixing matrix estimation for underdetermined blind source separation (UBSS) when the number of sources is unknown, this paper proposed a novel mixing matrix estimation method based on average information entropy and cluster validity index (CVI). Firstly, the initial cluster center is selected by using fuzzy C-means (FCM) algorithm and the corresponding membership matrix is obtained, and then the number of clusters is obtained by using the joint decision of CVI and average information entropy index of membership matrix, then multiple cluster number estimation results can be obtained by using multiple CVIs. Then, according to the results of the number of multiple clusters estimation, the number of radiation sources is determined according to the principle of the subordination of the minority to the majority. The cluster center vectors obtained from the clustering operation of the estimated number of radiation sources are fused, that is the mixing matrix is estimated based on the degree of similarity of the cluster center vectors. When the source signal is not sufficiently sparse, the time-frequency single source detection processing can be combined with the proposed method to estimate the mixing matrix. The effectiveness of the proposed method is validated by experiments.

关键词: UBSS, estimation of the number of radiation sources, mixing matrix, FCM clustering, CVI

Abstract: To solve the problem of mixing matrix estimation for underdetermined blind source separation (UBSS) when the number of sources is unknown, this paper proposed a novel mixing matrix estimation method based on average information entropy and cluster validity index (CVI). Firstly, the initial cluster center is selected by using fuzzy C-means (FCM) algorithm and the corresponding membership matrix is obtained, and then the number of clusters is obtained by using the joint decision of CVI and average information entropy index of membership matrix, then multiple cluster number estimation results can be obtained by using multiple CVIs. Then, according to the results of the number of multiple clusters estimation, the number of radiation sources is determined according to the principle of the subordination of the minority to the majority. The cluster center vectors obtained from the clustering operation of the estimated number of radiation sources are fused, that is the mixing matrix is estimated based on the degree of similarity of the cluster center vectors. When the source signal is not sufficiently sparse, the time-frequency single source detection processing can be combined with the proposed method to estimate the mixing matrix. The effectiveness of the proposed method is validated by experiments.

Key words: UBSS, estimation of the number of radiation sources, mixing matrix, FCM clustering, CVI

中图分类号:

TN911

Wang Chuanchuan, Jiang Lin, Zeng Yonghu, Wang Liandong. Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes[J]. The Journal of China Universities of Posts and Telecommunications, 2020, 27(6): 73-86.

参考文献

1. Wang C C, Zeng Y H. Research status and prospects of underdetermined blind source separation algorithm. Journal of Beijing University of Posts and Telecommunications, 2018, 41(6): 113-119 (in Chinese).

2. Hao Y S, Song L Y, Wang M Y, et al. Underdetermined source separation of bearing faults based on optimized intrinsic characteristic-scale decomposition and local non-negative matrix factorization. IEEE Access, 2019, 7: 11427-11435.

3. Rashi D, Anuradha B, Prasad V R C, et al. Discrete wavelet transform based unsupervised underdetermined blind source separation methodology for radar pulse deinterleaving using antenna scan pattern. IET Radar, Sonar & Navigation, 2019, 13(8): 1350-1358.

4. Wang C C, Zeng Y H, Fu W H, et al. Estimation method for an underdetermined mixing matrix based on maximum density point searching. Journal of Xidian University, 2019, 46(1): 106-111 (in Chinese).

5. Wang C C, Jia R. A source signal recovery method for underdetermined blind source separation based on shortest path. Applied Computational Electromagnetics Society Journal, 2020, 35(4): 406-414.

6. Tichy O, Bodiova L, Smidl V. Bayesian non-negative matrix factorization with adaptive sparsity and smoothness prior. IEEE Signal Processing Letters, 2019, 26(3): 510-514.

7. Kuang L D. Blind source separation methods for group analysis of multi-subject complex-valued fMRI data. Ph D Thesis. Dalian, China: Dalian University of Technology, 2018 (in Chinese).

8. Ge S N, Han J, Han M. Nonnegative mixture for underdetermined blind source separation based on a tensor algorithm. Circuits, System and Signal Processing, 2015, 34(9): 2935-2950.

9. Xie Y, Xie K, Xie S L. Underdetermined convolutive blind separation of sources integrating tensor factorization and expectation maximization. Digital Signal Processing, 2019, 87: 145-154.

10. Hao R J, An X J, Shi Y L. Single-channel blind source separation based on EMD and CICA and its application to gearbox multi-fault diagnosis. Journal of Vibration and Shock, 2019, 38(8): 225-231.

11. Wang C C, Zeng Y H, Wang L D. Mixing matrix estimation method for UBSS based on observed data segmentation. Proceedings of the 8th International Conference on Communication and Network Security (ICCNS'18), 2018, Nov 2-4, Qingdao, China. New York, NY, USA: ACM, 2018: 149-154.

12. Smith S, Pischella M, Terre M. A moment-based estimation strategy for underdetermined single-sensor blind source separation. IEEE Signal Processing Letters, 2019, 26(6): 788-792.

13. Wang C C, Xu J Q, Zeng Y H. Research on influence of source number estimation on application of blind source separation algorithms. Procedia Computer Science, 2017, 107: 379-384.

14. Li C J, Zhu L D, Luo Z Q. Underdetermined blind source separation of adjacent satellite interference based on sparseness. China Communications, 2017, 14(4): 140-149.

15. Feng F, Kowalski M. Underdetermined reverberant blind source separation: Sparse approaches for multiplicative and convolutive narrowband approximation. ACM Trans on Audio Speech and Language Processing, 2019, 27(2): 442-456.

16. Xie Y, Xie K, Xie S L. Underdetermined blind source separation for heart sound using higher-order statistics and sparse representation. IEEE Access, 2019, 7: 87606-87616.

17. Piao S Z, Chaomurilige, Yu J. Cluster validity indexes for FCM clustering algorithm. Pattern Recognition and Artificial Intelligence, 2015, 28(5): 452-461.

18. Wang X, Xu Y S. An improved index for clustering validation based on Silhouette index and Calinski-Harabasz index. IOP Conference Series: Materials Science and Engineering, 2019, 569(5) (Proceeding of the 239th ECS Meeting ,with 18th International Conference on Chemical Sensors (IMCS'18), 2021, May 30-Jun 3, Chicago, IL, USA) : 052024/1-6.

19. Zhu E Z, Wang X, Liu F. A new cluster validity index for overlapping datasets. Journal of Physics: Conference Series, 2019, 1168(3) (Proceeding of the 2018 International Conference on Computer Information Science and Application Technology (CISAT'18), 2018, Dec 7-9, Daqing, China) : 032070/1-7.

20. Zhang Y J, Wang W N, Zhang X N, et al. A cluster validity index for fuzzy clustering. Information Sciences, 2008, 178(4): 1205-1218.

21. Zhang X B, Li J. A new validity index of Fuzzy c-means clustering. Proceeding of the 2009 International Conference on Intelligent Human-Machine Systems and Cybernetics: Vol 2, 2009, Aug 26-27, Hangzhou, China. Piscataway, NJ, USA: IEEE, 2009: 218-221.

22. Chen J S, Pi D C. A cluster validity index for fuzzy clustering based on non-distance. Proceeding of the 5th International Conference on Computational and Information Sciences, Shenyang, China, Piscataway, NJ, USA: IEEE, 2013: 880-883.

23. Fu W H, Wang L, Ma L F. Improved laplace mixed model potential function algorithm for UBSS. Journal of Xidian University, 2014, 41(6): 1-5 (in Chinese).

24. Liu C. The research on undetermined blind source separation algorithm based on sparse properties. Master Thesis. Harbin, China: Harbin Engineering University, 2018.

25. Starck J L, Murtagh F, Fadili J. Sparse image and signal processing: Wavelets, curvelets, morphological diversity. London, UK: Cambridge University Press, 2010.

26. Zhao G Q. Principle of radar countermeasure. 2nd ed. Xi'an, China: Xidian University Press, 2012 (in Chinese).

27. Fu W H, Zhou X B, Nong B, et al. Blind estimation of underdetermined mixing matrix based on similarity measurement. Journal of Beijing University of Posts and Telecommunications, 2017, 40 (6): 24-29 (in Chinese).

28. Kim S G, Yoo C D. Underdetermined blind source separation based on subspace representation. IEEE Trans on Signal Processing, 2009, 57(7): 2604-2614.

[1]	束丰张玲华丁寅. Iterative subspace matching pursuit for joint sparse recovery[J]. 中国邮电高校学报(英文版), 2023, 30(2): 26-35.
[2]	Wang Chuanchuan, Zeng Yonghu, Wang Liandong, Li Zhipeng. Source enumeration method by combining diagonal loading and constructing second-order statistics[J]. 中国邮电高校学报(英文版), 2023, 30(1): 47-55.
[3]	Guan Sihai, Cheng Qing, Zhao Yong, Liu Fangyao. Variable step-size adaptive filtering algorithm based on an exponent sin function[J]. 中国邮电高校学报(英文版), 2023, 30(1): 56-65.
[4]	Jiang Fan, Chen Jiajun, Gao Youjun, Sun Changyin. Research on ECG classification based on transfer learning [J]. 中国邮电高校学报(英文版), 2022, 29(6): 83-96.
[5]	张承畅徐余杨建鹏李晓梦. Automatic modulation classification based on AlexNet with data augmentation[J]. 中国邮电高校学报(英文版), 2022, 29(5): 51-61.
[6]	Yao Shengshi, Wang Sixian, Dai Jincheng, Niu Kai, Xu Wenjun, Zhang Ping. Semantic information processing in industrial networks[J]. 中国邮电高校学报(英文版), 2022, 29(1): 41-49.
[7]	Li Hui, Li Shanshan, Zou Borong, Chen Yannan. Modulation classification based on the collaboration of dual-channel CNN-LSTM and residual network[J]. 中国邮电高校学报(英文版), 2022, 29(1): 113-124.
[8]	Li Xiaohui, Liu Shuaishuai, Fan Tao, Fang Kun. Dual-CRC polar codes and dual-SCFlip decoding algorithm for cell search in 5G system [J]. 中国邮电高校学报(英文版), 2021, 28(6): 82-90.
[9]	Ma Shexiang, Mei Xiaobing. Side-lobe constrained beamforming under virtual expansion of L-shaped array[J]. 中国邮电高校学报(英文版), 2021, 28(2): 79-88.
[10]	周游段瑞枫蒋伯峰. Reduced-state noncoherent MLSD method for CPM in aeronautical telemetry [J]. 中国邮电高校学报(英文版), 2020, 27(4): 43-53.
[11]	Wang Chuanchuan, Zeng Yonghu, Wang Liandong, Fu Weihong. Improved statistical sparse decomposition principle method for underdetermined blind source signal recovery[J]. 中国邮电高校学报(英文版), 2019, 26(6): 94-102.
[12]	Gao Miao, Yu Zhiguo, Li Qingqing, Wei Jinghe, Gu Xiaofeng. Improved ECG signal compression method for E-healthcare[J]. 中国邮电高校学报(英文版), 2019, 26(4): 62-69.
[13]	张凯尹长川 Xu Guixian 刘丹谱. Robust hybrid beamforming design for millimeter-wave massive MIMO systems[J]. 中国邮电高校学报(英文版), 2019, 26(3): 44-49.
[14]	Yang Jinsheng, Xiang Yang, Chen Weigang, Dong Yangyang. Computationally efficient 2-D DOA estimation for non-uniform two-L-shaped array[J]. 中国邮电高校学报(英文版), 2018, 25(6): 81-89.
[15]	Yuan Fang, Tian Bin. Modified min-sum SCAN decoding algorithm for polar codes[J]. 中国邮电高校学报(英文版), 2018, 25(6): 90-96.

Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes

Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价