A Blind Source Separation Algorithm Based on Dynamic Niching Particle Swarm Optimization

Abstract

Abstract. In this paper, the dynamic niching particle swarm optimization (DNPSO) is proposed to solve linear blind source separation problem. The key point is to use the DNPSO rather than particle swarm optimization (PSO) and fast-ICA as the optimization algorithm in Independent Component Analysis (ICA). By using DNPSO, which has global superiority, the performance of ICA will be improved in accuracy and convergence rate. The idea of sub-population in DNPSO leads to the greater efficiency compared with other methods when solving high dimensional cost functions in ICA.

Bigradient neural network-based quantum particle swarm optimization for blind source separation

“…Where  represents the contraction-expansion factor, is the control parameter of the algorithm convergent [17], [18].…”

Improved Swarm Intelligent Blind Source Separation Based on Signal Cross-Correlation

“…The separated signal is . The mathematical model of blind source separation [ 1 , 2 , 3 , 5 , 16 , 17 , 19 , 20 , 21 , 23 , 41 , 42 , 43 , 44 , 45 , 46 ] is a linear mixture model, as shown in Equation (1): where A is the linear mixing matrix of , and n represents the number of source signals. According to the de-correlation and non-Gaussian criteria, the separation matrix W is solved, and then the separated signal is extracted from the , which can be expressed as Equation (2): where is the separation matrix of .…”

“…Blind Source Separation (BSS) refers to the process of obtaining source signals from mixed signals when the theoretical model and source signals cannot be accurately known [ 1 , 2 , 3 ]. It is a powerful signal processing method, which has been widely used in sonar and radar signal processing [ 4 , 5 ], wireless communication [ 6 ], geophysical exploration [ 7 , 8 ], biomedical signal processing [ 9 , 10 ], speech and image processing [ 11 , 12 ], and machine fault diagnosis [ 13 , 14 ].…”

“…Independent Component Analysis (ICA) is the traditional method of linear BSS to obtain the best estimate of the source signals [ 15 ]. Since the traditional optimization algorithms are easily trapped into local optimum and have a slow convergence rate, researchers have introduced swarm intelligence optimization algorithms to ICA for blind source separation [ 3 ]. Swarm intelligence (SI) algorithm is a new branch of artificial intelligence with low cost, fast speed, and good robustness [ 16 ].…”

“…In the end, the final separated signals are identified by the cross-correlations values, which means that the lower the values of the cross-correlations of the separated signals are, the more mutual independence they have. In order to verify the effectiveness of our algorithm, six swarm intelligence algorithms (PSO [ 3 , 24 ], GA [ 25 , 26 ], DE [ 27 , 28 , 29 ], SCA [ 30 , 31 , 32 ], BOA [ 33 , 34 ], and CSA [ 35 , 36 ]) are applied. The results of separation were compared in terms of separation performance, separation success rate, and applicability according to the cross-correlation and three kinds of separation evaluation indexes (scale invariant signal-to-distortion ratio (SI-SDR), perceptual evaluation of speech quality (PESQ), and short-time objective intelligibility (STOI) [ 37 , 38 , 39 , 40 ]).…”