Blind source separation method based on membrane computing and PSO algorithm

被引：0
|
作者：
Sun Y. [1 ]
Yang F. [1 ]
Zheng J. [2 ]
Xu M. [1 ]
Pei S. [2 ]
机构：
[1] School of Mechanical Electronic & Information Engineering, China University of Mining & Technology (Beijing), Beijing
[2] State Key Lab of Coal Resources and Safe Mining, China University of Mining & Technology (Beijing), Beijing
来源：
Yang, Feng | 2018年 / Chinese Vibration Engineering Society卷 / 37期
关键词：
Blind source separation (BSS); Inertia weight; Membrane computing (MC); Particle swarm optimization (PSO);
D O I：
10.13465/j.cnki.jvs.2018.17.009
中图分类号：
学科分类号：
摘要：
In order to solve problems of slower convergence and lower separating performance of the existing blind source separation methods, a new method based on membrane computing (MC) and particle swarm optimization (PSO) was proposed. The separated signals' neg-entropy was taken as the fitness function of PSO. Particles were uniformly distributed into each elementary membrane. The velocity and position of population particles were updated with a particle own optimal position and the population global optimal position. The optimal solution to PSO was used to adjust step function of blind source separation and then separate signals. The proposed algorithm simplified inertia weight's choosing to ensure the accuracy of PSO local search and satisfy the variety of global search. The simulation and actual application showed that mixing signals can be well separated with this new method and the premature convergence problem of PSO can be avoided; this new method has a faster convergence speed and a more excellent separating performance. © 2018, Editorial Office of Journal of Vibration and Shock. All right reserved.
引用
收藏
页码：63 / 71
页数：8
相关论文
共 16 条
  • [1] Xu X., Duan C., Liu L., Et al., Research on a new structural traits based on complex-valued space joint diagonalization algorithm for blind source separation and its applications, Acta Electronica Sinica, 45, 1, pp. 29-36, (2017)
  • [2] Wang J., Zhang S., EMD denoising method based on frequency domain constrained independent component analysis, Journal of China Coal Society, 42, 3, pp. 621-629, (2017)
  • [3] Li L., Ren J., A universal particle swarm-optimized independent component analysis algorithm, Journal of Vibration and Shock, 34, 8, pp. 7-11, (2015)
  • [4] Wang R., Zhan Y., Zhou H., A class of sequential blind source separation method in order using swarm optimization algorithm, Circuits Systems and Signal Processing, 35, 9, pp. 3220-3243, (2016)
  • [5] Bohm M., Stadlthanner K., Gruber P., Et al., On the use of simulated annealing to automatically assign decorrelated components in second-order blind source separation, IEEE Transactions on Biomedical Engineering, 53, 5, pp. 810-820, (2006)
  • [6] Wen G., Zhang C., Lin Z., Et al., Independent component analysis based on genetic algorithmsInternational Conference on Natural Computation, pp. 214-218, (2014)
  • [7] Yang J., Yu W., Tian H., Et al., Underdetermined blind source separation method based on independent component analysis, Journal of Vibration and Shock, 32, 7, pp. 30-33, (2013)
  • [8] Commom P., Et al., Independent component analysis, a new concept?, Signal Processing, 36, 3, pp. 287-314, (1994)
  • [9] Zhang C., Zhang J., Sun X., Blind source separation based on adaptive particle swarm optimization, Systems Engineering and Electronics, 31, 6, pp. 1275-1278, (2009)
  • [10] Zhu X., Zhang X., Overdetermined blind source separation based on singular value decomposition, Journal of Electronics & Information Technology, 26, 3, pp. 337-343, (2004)
12