基于线谱特征保持的单枚浮标多目标信号分离算法

doi:10.12305/j.issn.1001-506X.2024.06.07

摘要/Abstract

摘要：

针对单枚被动全向声纳浮标多目标信号的分离问题, 提出联合非负矩阵分解(non-negative matrix factorization, NMF)和快速独立成分分析(fast independent component analysis, FastICA)的多目标信号盲分离算法。首先, 基于空间与谱间相关性优化NMF算法, 以增强NMF算法对水声信号调制线谱特征的适应性, 提高对线谱的保持优势; 然后, 以NMF基矩阵优势结合FastICA算法实现水声多目标信号的盲分离。仿真信号实验结果表明, 所提算法取得了较高的信号分离精度, 可以较好地保持信号的调制特征, 同时对分离信号进行了一定的降噪增强, 更好地保证了后续目标识别的特征支撑。

关键词: 被动全向声纳浮标, 多目标信号盲分离, 非负矩阵分解, 快速独立成分分析, 线谱特征保持

Abstract:

To solve the problem of multi-target signal separation of a single passive omnidirectional sonar buoy, a multi-target blind signal separation algorithm combining non-negative matrix factorization (NMF) and fast independent component analysis (FastICA) is proposed. Firstly, the proposed algorithm optimizes the NMF algorithm based on the correlation between space and spectrum to enhance the adaptability of the NMF algorithm to the characteristics of underwater acoustic signal modulation line spectrum and improve the advantage of maintaining line spectrum. Then, the blind separation of underwater acoustic multi-target signals is realized by the FastICA algorithm with NMF basis matrix advantage. Experimental results of simulated signals show that the proposed algorithm achieves high signal separation accuracy, can better maintain the modulation characteristics of the signal, and enhance the noise of the separated signal to a certain extent at the same time, which better ensures the feature support of subsequent target recognition.

Key words: passive omnidirectional sonar buoy, blind separation of multi-target signals, non-negative matrix factorization (NMF), fast independent component analysis (FastICA), line spectrum feature preservation

中图分类号:

TN911.7

李大卫, 吴明辉, 单志超, 宋广明, 蔡召鹏. 基于线谱特征保持的单枚浮标多目标信号分离算法[J]. 系统工程与电子技术, 2024, 46(6): 1892-1898.

Dawei LI, Minghui WU, Zhichao SHAN, Guangming SONG, Zhaopeng CAI. Multi-target signal separation algorithm for single buoy based on line spectrum feature preservation[J]. Systems Engineering and Electronics, 2024, 46(6): 1892-1898.

图/表 10

图1

图2

表1

图3

图4

图5

图6

表2

图7

图8

参考文献 30

1	NIKNAZAR H , NASRABADI A , SHAMSOLLAHI M . A new blind source separation approach based on dynamical similarity and its application on epileptic seizure prediction[J]. Signal Processing, 2021, 183 (1): 108045.
2	赵孟晨, 姚秀娟, 王静, 等. 基于Stacked-TCN的空间混叠信号单通道盲源分离方法[J]. 系统工程与电子技术, 2021, 43 (9): 2628- 2636.
	ZHAO M C , YAO X J , WANG J , et al. Single-channel blind souce separation method of spatial aliasing signal based on Stacked-TCN[J]. Systems Engineering and Electronics, 2021, 43 (9): 2628- 2636.
3	HUANG R , LI X , ZHAO L . Spectral-spatial robust nonnegative matrix factorization for hyperspectral unmixing[J]. IEEE Trans. on Geoscience and Remote Sensing, 2019, 57 (10): 5235- 8254.
4	XU P , JIA Y , JIANG M . Blind audio source separation based on a new system model and the Savitzky-Golay filter[J]. Journal of Electrical Engineering, 2021, 72 (3): 208- 212. doi: 10.2478/jee-2021-0029
5	MA B Z , ZHANG T Q . Underdetermined blind source separation based on source number estimation and improved sparse component analysis[J]. Circuits, System, and Signal Processing: CSSP, 2021, 40 (7): 3417- 3436. doi: 10.1007/s00034-020-01629-x
6	LU X Q , WU H , YUAN Y , et al. Manifold regularized sparse NMF for hyperspectral unmixing[J]. IEEE Trans. on Geoscience and Remote Sensing, 2013, 51 (5): 2815- 2826. doi: 10.1109/TGRS.2012.2213825
7	LIANG L , DING X Y , WEN H B , et al. Impulsive components separation using minimum-determinant KL-divergence NMF of bi-variable map for bearing diagnosis[J]. Mechanical Systems & Signal Processing, 2022, (175): 109129- 109150.
8	WANG W H , QIAN Y T , LIU H F . Multiple clustering guided nonnegative matrix factorization for hyperspectral unmixing[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13 (6): 5162- 5179.
9	LU X Q , DONG L , YUAN Y . Subspace clustering constrained sparse NMF for hyperspectral unmixing[J]. IEEE Trans. on Geoscience and Remote Sensing, 2020, 58 (5): 3007- 3019. doi: 10.1109/TGRS.2019.2946751
10	韦娟, 杨皇卫, 宁方立. 基于NMF与CNN联合优化的声学场景分类[J]. 系统工程与电子技术, 2022, 44 (5): 1433- 1438.
	WEI J , YANG H W , NING F L . Acoustic scene classification based on joint optimization of NMF and CNN[J]. Systems Engineering and Electronics, 2022, 44 (5): 1433- 1438.
11	TEPPER M , SAPIRO G . Compressed nonnegative matrix factorization is fast and accurate[J]. IEEE Trans. on Signal Processing, 2016, 64 (9): 2269- 2283. doi: 10.1109/TSP.2016.2516971
12	YANG H D , XU N , HAO Y P . Integrating pure pixel identification into nonnegative matrix factorization for endmember extraction[J]. IAENG International Journal of Computer Science, 2019, 46 (2): 291- 299.
13	MA B Z , ZHANG T Q , AN Z L , et al. A blind source separation method for time-delayed mixtures in underdetermined case and its application in modal identification[J]. Digital Signal Processing, 2021, 112 (8): 103007- 103022.
14	崔建峰, 邓泽平, 申飞, 等. 基于非负矩阵分解和长短期记忆网络的单通道语言分离[J]. 科学技术与工程, 2019, 19 (12): 1671- 1815.
	CUI J F , DENG Z P , SHEN F , et al. Single channel speech separation based on non-negative matrix factorization and long short-term memory network[J]. Science Technology and Engineering, 2019, 19 (12): 1671- 1815.
15	HUANG R S , LI X R , ZHAO L Y . Spectral-spatial robust nonnegative matrix factorization for hyperspectral unmixing[J]. IEEE Trans. on Geoscience and Remote Sensing, 2019, 57 (10): 8235- 8254. doi: 10.1109/TGRS.2019.2919166
16	刘贤忠, 吴明辉, 郑晓庆, 等. 改进NMF优化的水声目标信号增强算法[J]. 电光与控制, 2021, 28 (9): 6-9, 53. doi: 10.3969/j.issn.1671-637X.2021.09.002
	LIU X Z , WU M H , ZHENG X Q , et al. Underwater acoustic target signal enhancement algorithm optimized by improved NMF[J]. Electronics Optics & Control, 2021, 28 (9): 6-9, 53. doi: 10.3969/j.issn.1671-637X.2021.09.002
17	HOU L S , CHU D L , LIAO L Z . Convergence of a fast hierarchical alternating least squares algorithm for nonnegative matrix factorization[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36 (1): 77- 89. doi: 10.1109/TKDE.2023.3279369
18	李海鹏, 孙大军, 郑翠娥. 强干扰环境下水声时延估计技术研究[J]. 电子与信息学报, 2021, 43 (3): 873- 880.
	LI H P , SUN D J , ZHENG C E . Time of arrival estimation in presence of strong interference[J]. Journal of Electronics & Information Technology, 2021, 43 (3): 873- 880.
19	李大卫, 杨日杰, 韩建辉. 舰艇噪声环境下改进语音信号增强算法[J]. 西安电子科技大学学报, 2016, 43 (5): 133- 138. doi: 10.3969/j.issn.1001-2400.2016.05.024
	LI D W , YANG R J , HAN J H . Study of speech enhancement in the background of ship-radiated noise[J]. Journal of Xidian University, 2016, 43 (5): 133- 138. doi: 10.3969/j.issn.1001-2400.2016.05.024
20	袁博. 基于混合像元空间与谱间相关性模型的NMF线性盲解混[J]. 测绘学报, 2019, 48 (9): 1151- 1160.
	YUAN B . NMF linear blind unmixing method based on mixed pixel's spatial and spectral correlation mode[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48 (9): 1151- 1160.
21	QIU Y , SUN C , TANG J . Seismic attribute fusion approach using optimized FastICA-based blind source separation algorithm[J]. Geophysical Prospecting for Petroleum, 2022, 57 (5): 733- 743.
22	E J W , YE J M , HE L L , et al. Performance analysis for complex-valued FastICA and its improvement based on the Tukey M-estimator[J]. Digital Signal Processing, 2021, 115 (3): 103077.
23	XIE Y , XIE K , YANG Q Y , et al. Reverberant blind separation of heart and lung sounds using nonnegative matrix factorization and auxiliary function technique[J]. Biomedical Signal Processing and Control, 2021, 69, 102899- 102899. doi: 10.1016/j.bspc.2021.102899
24	DO H D , TRAN S T , CHAU D T . Speech separation in the frequency domain with autoencoder[J]. Journal of Communication, 2020, 15 (11): 841- 848.
25	PASEDDULA C , GANGASHETTY S V . Late fusion framework for acoustic scene classification using LPCC, SCMC, and Log-Mel band energies with deep neural networks[J]. Applied Acoustics, 2021, 172, 107568. doi: 10.1016/j.apacoust.2020.107568
26	LEE S , PANG H S . Feature extraction based on the non-negative matrix factorization of convolutional neural networks for monitoring domestic activity with acoustic signals[J]. IEEE Access, 2020, 8, 122384- 122395. doi: 10.1109/ACCESS.2020.3007199
27	LI Y B , WANG Y F , DONG Q H . A novel mixing matrix estimation algorithm in instantaneous under determined blind source separation[J]. Signal Image and Video Processing, 2020, 14 (5): 1- 8.
28	YANG Z J , LING W K , BINGHAM C . Fault detection and signal reconstruction for increasing operational availability of industrial gas turbines[J]. Measurement Journal of the International Measurement Confederation, 2013, 46 (6): 1938- 1946. doi: 10.1016/j.measurement.2013.02.016
29	DAM H , LOW S , NORDHOLM S . Two-level optimization approach with accelerated proximal gradient for objective measures in sparse speech reconstruction[J]. Journal of Industrial and Management Optimization, 2022, 18 (5): 3701- 3717. doi: 10.3934/jimo.2021131
30	HONG F , LIU C , GUO L , et al. Underwater acoustic target recognition with a residual network and the optimized feature extraction method[J]. Applied Sciences, 2021, 11 (4): 1442- 1454. doi: 10.3390/app11041442

参数	取值
f₀₁	20
f₀₂	35
f₀₃	45
f_c1	2 500
f_c2	2 000
f_c3	1 500

分离信号	源信号
分离信号	1	2	3	混合信号
1	0.925	0.106	0.197	0.515
2	0.109	0.095	0.893	0.612
3	0.210	0.913	0.115	0.609

[1]	赵庆媛, 赵志强, 叶春茂, 鲁耀兵. 气动目标多频点调制谱融合增强识别方法[J]. 系统工程与电子技术, 2023, 45(7): 2043-2050.
[2]	韦娟, 杨皇卫, 宁方立. 基于NMF与CNN联合优化的声学场景分类[J]. 系统工程与电子技术, 2022, 44(5): 1433-1438.
[3]	同鸣，张伟，吴扬成. 正交指数约束的平滑非负矩阵分解方法及应用[J]. 系统工程与电子技术, 2013, 35(10): 2221-2228.
[4]	郭睿, 包敏, 李军, 臧博, 邢孟道. 极化SAR图像中的弱小舰船目标检测[J]. Journal of Systems Engineering and Electronics, 2011, 33(6): 1253-1256.
[5]	刘昶，周激流，郎方年. 广义判别正交非负矩阵分解及其应用[J]. Journal of Systems Engineering and Electronics, 2011, 33(10): 2327-2330.