S函数变步长LMS算法的一种L_2范数修正
|
摘要
为了提高S函数变步长LMS算法的综合性能和抗干扰能力,引入L_2范数控制步长更新.在L_2范数中引入输入信号,对输入信号实施动态地跟踪,以提高算法的抗干扰能力;利用归一化输入信号代替原始信号,便于处理信号且可以降低输入信号动态范围对算法的影响.在高和低信噪比条件下的仿真结果表明:相较于其他的变步长LMS算法,所提出算法收敛速度更快和稳态误差更小.
In order to improve the comprehensive performance and anti-interference ability of S function variable step LMS algorithm,the paper used the L_2 norm to control step update.The input signal is introduced into the L_2 norm to track the input signal dynamically,so as to improve the anti-interference ability of the algorithm.In this study,original signals are replaced by normalized input signals,so that the signal could be processed conveniently and the influence of the dynamic range of the input signal on the algorithm could be reduced as well.Simulation results under high and low SNR conditions show that the proposed algorithm has the advantages of faster convergence rate and smaller steady-state error over other variable step size LMS algorithms.
In order to improve the comprehensive performance and anti-interference ability of S function variable step LMS algorithm,the paper used the L_2 norm to control step update.The input signal is introduced into the L_2 norm to track the input signal dynamically,so as to improve the anti-interference ability of the algorithm.In this study,original signals are replaced by normalized input signals,so that the signal could be processed conveniently and the influence of the dynamic range of the input signal on the algorithm could be reduced as well.Simulation results under high and low SNR conditions show that the proposed algorithm has the advantages of faster convergence rate and smaller steady-state error over other variable step size LMS algorithms.
引文
[1] HU M,TANG W,CAI C.A new variable step-size LMS adaptive filtering algorithm for beamforming technology[C]//9th IEEE International Symposium on the Antennas Propagation and EM Theory.Guangzhou:IEEE Press,2010:101-104.
[2] WU F Y,TONG F.Non-uniform norm constraint LMS algorithm for sparse system identification[J].IEEE Communications Letters,2013,17(2):385-388.
[3] HUI Y T,LI B B,TONG Z,et al.A short-delay multipath mitigation algorithm based on variable step-size LMS adaptive filtering[C]//Applied Mechanics and Materials.Switzerland:Trans.Tech.Publications,2014,571:368-375.
[4] LI Y P,LEE T S,WU B F.A variable step-size sign algorithm for channel estimation[J].Signal Processing,2014,102:304-312.
[5] 覃景繁,欧阳景正.一种新的变步长LMS自应用滤波算法[J].数据采集与处理,1997,12(3):171-174.
[6] 罗小东,贾振红,王强.一种新的变步长LMS自适应滤波算法[J].电子学报,2006,34(6):1123-1126.
[7] 高鹰,谢胜利.一种变步长LMS自适应滤波算法及分析[J].电子学报,2001(8):1094-1097.
[8] 韩彦博,高丽.基于Q函数的变步长 LMS 算法[J].兰州交通大学学报,2017,36(6):42-46.
[9] 彭继慎,刘爽,安丽.低信噪比下基于新型变步长LMS的自适应滤波算法[J].传感技术学报,2013,26(8):1116-1120.
[10] 还秋云,邱晓晖,刘晓飞.引用范数的双曲正切函数变步长LMS算法[J].信号处理,2014,30(1):93-99.
[11] SUDHA K L.Performance analysis of New Time Varying LMS (NTVLMS) adaptive filtering algorithm in noise cancellation system[C]//2015 International Conference on Communication,Information & Computing Technology .Maharashtra:IEEE Press,2015:1-6.
[12] LAU Y S,HUSSAIN Z M,HARRIS R.A time-varying convergence parameter for the LMS algorithm in the presence of white Gaussian noise[C]//Australian Telecommunications,Networks and Applications Conference.Melbourne:Telstra Communication Press,2003:1151-1156.
[13] 刘建成,赵宏志,全厚德,等.迭代变步长LMS算法及性能分析[J].电子与信息学报,2015,37(7):1674-1680.
[14] 张红梅,韩万刚.一种新的变步长LMS自适应滤波算法研究及其应用[J].仪器仪表学报,2015,36(8):1822-1830.
[15] SIMON H.自适应滤波器原理[M].4版.郑宝玉,译.北京:电子工业出版社,2010:206-212.
[16] 何英杰.一种改进自适应谐波检测算法研究[J].电力自动化设备,2006,26(9):23-26.
[17] 吕振肃,熊景松.一种改进的变步长LMS自适应算法[J].信号处理,2008,24(1):144-146.
[2] WU F Y,TONG F.Non-uniform norm constraint LMS algorithm for sparse system identification[J].IEEE Communications Letters,2013,17(2):385-388.
[3] HUI Y T,LI B B,TONG Z,et al.A short-delay multipath mitigation algorithm based on variable step-size LMS adaptive filtering[C]//Applied Mechanics and Materials.Switzerland:Trans.Tech.Publications,2014,571:368-375.
[4] LI Y P,LEE T S,WU B F.A variable step-size sign algorithm for channel estimation[J].Signal Processing,2014,102:304-312.
[5] 覃景繁,欧阳景正.一种新的变步长LMS自应用滤波算法[J].数据采集与处理,1997,12(3):171-174.
[6] 罗小东,贾振红,王强.一种新的变步长LMS自适应滤波算法[J].电子学报,2006,34(6):1123-1126.
[7] 高鹰,谢胜利.一种变步长LMS自适应滤波算法及分析[J].电子学报,2001(8):1094-1097.
[8] 韩彦博,高丽.基于Q函数的变步长 LMS 算法[J].兰州交通大学学报,2017,36(6):42-46.
[9] 彭继慎,刘爽,安丽.低信噪比下基于新型变步长LMS的自适应滤波算法[J].传感技术学报,2013,26(8):1116-1120.
[10] 还秋云,邱晓晖,刘晓飞.引用范数的双曲正切函数变步长LMS算法[J].信号处理,2014,30(1):93-99.
[11] SUDHA K L.Performance analysis of New Time Varying LMS (NTVLMS) adaptive filtering algorithm in noise cancellation system[C]//2015 International Conference on Communication,Information & Computing Technology .Maharashtra:IEEE Press,2015:1-6.
[12] LAU Y S,HUSSAIN Z M,HARRIS R.A time-varying convergence parameter for the LMS algorithm in the presence of white Gaussian noise[C]//Australian Telecommunications,Networks and Applications Conference.Melbourne:Telstra Communication Press,2003:1151-1156.
[13] 刘建成,赵宏志,全厚德,等.迭代变步长LMS算法及性能分析[J].电子与信息学报,2015,37(7):1674-1680.
[14] 张红梅,韩万刚.一种新的变步长LMS自适应滤波算法研究及其应用[J].仪器仪表学报,2015,36(8):1822-1830.
[15] SIMON H.自适应滤波器原理[M].4版.郑宝玉,译.北京:电子工业出版社,2010:206-212.
[16] 何英杰.一种改进自适应谐波检测算法研究[J].电力自动化设备,2006,26(9):23-26.
[17] 吕振肃,熊景松.一种改进的变步长LMS自适应算法[J].信号处理,2008,24(1):144-146.