基于粒子群优化的工频干扰消除算法
|
摘要
提出了一种基于粒子群优化的消除微弱信号采集过程中工频干扰的算法。通过人工构造观测信号,使系统模型符合盲源分离的数学模型要求。使用信号的四阶累积量作为信号独立性的判据,利用粒子群优化算法寻找使判据最大化的分离矩阵,进而消除被采集信号中的工频干扰。在粒子群优化算法的求解过程中,采用将对分离矩阵的直接辨识转换成对一系列Givens矩阵的辨识方法,从而减少了算法中对未知元素辨识的数量,避免反复白化过程,有效降低了算法的计算量,克服了粒子群优化过程中容易早熟收敛的问题。仿真结果表明,本算法在保护有用信号的前提下,能够有效地消除微弱信号中的工频干扰。
This paper proposed a power interference removal algorithm in weak signal collection based on particle swarm optimization. It gave the system model consistent with BSS mathematical model by constructing observation signal artificially,and uesd fourth-order cumulant for estimating independence of the signal,found the separation matrix maximizing the criterion using particle swarm optimization and then the power interference could be removed. In the process of particle swarm optimization,transformed direct identification of the separation matrix into identification of a series of Givens matrices and reduced the number of unknown elements,avoided the whiting process and reduced computational complexity. Synchronously,overcame premature convergence problem in process of particle swarm optimization. The simulation results show that the algorithm is useful in eliminating power interference in weak signal and the useful signal can be protected efficiently.
This paper proposed a power interference removal algorithm in weak signal collection based on particle swarm optimization. It gave the system model consistent with BSS mathematical model by constructing observation signal artificially,and uesd fourth-order cumulant for estimating independence of the signal,found the separation matrix maximizing the criterion using particle swarm optimization and then the power interference could be removed. In the process of particle swarm optimization,transformed direct identification of the separation matrix into identification of a series of Givens matrices and reduced the number of unknown elements,avoided the whiting process and reduced computational complexity. Synchronously,overcame premature convergence problem in process of particle swarm optimization. The simulation results show that the algorithm is useful in eliminating power interference in weak signal and the useful signal can be protected efficiently.
引文
[1]沈凤麟.生物医学随机信号处理[M].合肥:中国科技大学出版社,1999:441-442.
[2]KENNEDY J,EBERHART R C.Particle swarm optimization[C]//Proc of IEEE International Conference on Neutral Networks.Perth:[s.n.],1995:1942-1948.
[3]SHI Yu-hui,EBERHART R C.A modified particle swarm optimizer[C]//Proc of IEEE World Congress on Computational Intelligence.Anchorage,Alaska:[s.n.],1998:69-73.
[4]TICHAVSKY P,KOLDOVSKY Z,YEREDOR A,et al.A hybrid technique for blind separation of non-Gaussian and time-correlated sources using a multicomponent approach[J].IEEE Trans on Neu-ral Networks,2008,19(3):421-430.
[5]SUN T Y,LIU Chan-cheng,HSIEH S T,et al.Blind separation with unknown number of sources based on auto-trimmed neural network[J].Neurocomputing,2008,71(10):2271-2280.
[6]LI Shu-jun,LI Cheng-qing,LO K T,et al.Cryptanalyzing an en-cryption scheme based on blind source separation[J].IEEE Trans on Circuits and Systems I:Regular Papers,2008,55(4):1055-1063.
[7]ZHANG Hong-juan,SHI Zhen-wei,GUO Chong-hui.Blind source extraction based on generalized autocorrelations and complexity pursuit[J].Neurocomputing,2009,72(10-12):2556-2562.
[8]BELL A J,SEJNOWSKI T J.An information-maximization approach to blind separation and blind deconvolution[J].Neuralcomputa-tion,1995,7(6):1129-1159.
[9]HYVARINEN A.Fast and robust fixed-point algorithms for indepen-dent component analysis[J].IEEE Trans on Neural Networks,1999,10(3):626-634.
[10]AMARI S,CICHOCKI A,YANG H H.A new learning algorithm for blind signal separation[C]//Proc of Advances in Neural Information Processing Systems.Cambridge,MA:MIT Press,1996:757-763.
[11]凌云,高军,张汝杰,等.随时间推移地震勘探处理方法研究[J].石油地球物理勘探,2001,36(2):173-179.
[12]刘洋.强工频干扰波的提取与消除方法[J].石油物探,2003,42(2):154-159.
[13]周静.心电信号中工频干扰的消除[J].生物医学工程研究,2003,22(4):61-64.
[14]吴小培,詹长安,周荷琴,等.采用独立分量分析方法消除信号中的工频干扰[J].中国科技大学学报,2000,30(6):671-676.
[15]刘俊豪.基于粒子群算法和鱼群算法的盲源分离的研究[D].太原:太原理工大学,2006.
[16]覃和仁,谢胜利.基于QR分解与罚函数方法的盲分离算法[J].计算机工程,2003,29(17):55-57.
[2]KENNEDY J,EBERHART R C.Particle swarm optimization[C]//Proc of IEEE International Conference on Neutral Networks.Perth:[s.n.],1995:1942-1948.
[3]SHI Yu-hui,EBERHART R C.A modified particle swarm optimizer[C]//Proc of IEEE World Congress on Computational Intelligence.Anchorage,Alaska:[s.n.],1998:69-73.
[4]TICHAVSKY P,KOLDOVSKY Z,YEREDOR A,et al.A hybrid technique for blind separation of non-Gaussian and time-correlated sources using a multicomponent approach[J].IEEE Trans on Neu-ral Networks,2008,19(3):421-430.
[5]SUN T Y,LIU Chan-cheng,HSIEH S T,et al.Blind separation with unknown number of sources based on auto-trimmed neural network[J].Neurocomputing,2008,71(10):2271-2280.
[6]LI Shu-jun,LI Cheng-qing,LO K T,et al.Cryptanalyzing an en-cryption scheme based on blind source separation[J].IEEE Trans on Circuits and Systems I:Regular Papers,2008,55(4):1055-1063.
[7]ZHANG Hong-juan,SHI Zhen-wei,GUO Chong-hui.Blind source extraction based on generalized autocorrelations and complexity pursuit[J].Neurocomputing,2009,72(10-12):2556-2562.
[8]BELL A J,SEJNOWSKI T J.An information-maximization approach to blind separation and blind deconvolution[J].Neuralcomputa-tion,1995,7(6):1129-1159.
[9]HYVARINEN A.Fast and robust fixed-point algorithms for indepen-dent component analysis[J].IEEE Trans on Neural Networks,1999,10(3):626-634.
[10]AMARI S,CICHOCKI A,YANG H H.A new learning algorithm for blind signal separation[C]//Proc of Advances in Neural Information Processing Systems.Cambridge,MA:MIT Press,1996:757-763.
[11]凌云,高军,张汝杰,等.随时间推移地震勘探处理方法研究[J].石油地球物理勘探,2001,36(2):173-179.
[12]刘洋.强工频干扰波的提取与消除方法[J].石油物探,2003,42(2):154-159.
[13]周静.心电信号中工频干扰的消除[J].生物医学工程研究,2003,22(4):61-64.
[14]吴小培,詹长安,周荷琴,等.采用独立分量分析方法消除信号中的工频干扰[J].中国科技大学学报,2000,30(6):671-676.
[15]刘俊豪.基于粒子群算法和鱼群算法的盲源分离的研究[D].太原:太原理工大学,2006.
[16]覃和仁,谢胜利.基于QR分解与罚函数方法的盲分离算法[J].计算机工程,2003,29(17):55-57.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心