一种基于LMS自适应滤波的互相关时延估计优化算法

doi:10.16180/j.cnki.issn1007-7820.2020.06.006

摘要/Abstract

摘要：

在多站时差定位系统中使用基于LMS自适应滤波的互相关法进行时延估计时,若采用固定步长因子则会在收敛速度和稳态失调之间存在较大矛盾,从而影响时延估计精度。针对这一问题,文中提出了一种基于分段变步长LMS自适应滤波和希尔伯特差值的互相关时延估计优化算法。该方法首先采用分段变步长LMS自适应滤波对信号进行滤波处理,然后将滤波后的信号作互相关运算,最后通过希尔伯特差值法锐化相关函数的峰值,进一步提高时延估计精度。在相同条件下,文中模拟分析了不同算法的时延估计精度。实验结果表明,新的优化算法时延估计精度更高。在不同信噪比下,新方法相较传统时延估计方法精度提高了2.2%以上,具有良好的抗噪声性能。

关键词: 时差定位, 自适应滤波, 互相关, 变步长LMS, 时延估计, 希尔伯特差值

Abstract:

In multi-station time difference positioning system, when using the cross-correlation method based on LMS adaptive filtering for time delay estimation, the fixed step factor has a large contradiction between convergence speed and steady-state offset, which affects the accuracy of delay estimation. Aiming at this problem, a cross-correlation time delay estimation optimization algorithm based on piecewise variable step size LMS adaptive filtering and Hilbert difference was proposed in this study. Firstly, the piecewise variable step size LMS adaptive filtering was used to filter the signal, then the filtered signal was cross-correlated, and finally the peak of the correlation function was sharpened by the Hilbert difference method to obtain higher delay estimation accuracy. Under the same conditions, the time delay estimation accuracy of different algorithms was simulated and analyzed. The results showed that the proposed optimization algorithm had higher delay estimation accuracy, and the prediction accuracy could be improved by 2.2% compared with the traditional method under different SNR, indicating proposed algorithm presented good anti-noise performance.

Key words: TDOA location, adaptive filtering, CC, VSS-LMS, time delay estimation, Hilbert difference

中图分类号:

TN95

魏文亮,茅玉龙. 一种基于LMS自适应滤波的互相关时延估计优化算法[J]. 电子科技, 2020, 33(6): 29-34.

WEI Wenliang,MAO Yulong. Cross-correlation Time Delay Estimation Optimization Algorithm Based on LMS Adaptive Filtering[J]. Electronic Science and Technology, 2020, 33(6): 29-34.

图/表 12

图1

图3

表1

图4

图5

图6

图7

表2

表3

表4

表5

参考文献 16

[1]	刘磊, 姚剑, 谢学锋 , 等. TDOA无源定位系统的原理与影响因素分析[J]. 无线通信技术, 2017,26(4):42-46.
	Liu Lei, Yao Jian, Xie Xuefeng , et al. Principle and influencing factors of TDOA passive location system[J]. Wireless Communication Technology, 2017,26(4):42-46.
[2]	Knapp C, Carter G . The generalized correlation method for estimation of time delay[J]. IEEE Transactions on Acoustics Speech & Signal Processing, 2003,24(4):320-327.
[3]	Carter G C . Coherence and time delay estimation[J]. Proceedings of the IEEE, 1987,75(2):236-255.
[4]	翟晓光, 武传华 . 一种稳健的循环平稳信号时延估计算法[J]. 火力与指挥控制, 2017,42(1):71-74.
	Zhai Xiaoguang, Wu Chuanhua . Arobust cyclostationary signal delay estimation algorithm[J]. Fire Control & Command Control, 2017,42(1):71-74.
[5]	童浪 . 基于卫星联合定位的TDOA参数估计方法的研究[D]. 西安:beoplay体育提现, 2015.
	Tong Lang . Research on TDOA parameter estimation method based on satellite joint positioning[D]. Xi’an:Xidian University, 2015.
[6]	乔振岳 . 最小均方误差自适应时延估计算法研究[D]. 西安:beoplay体育提现, 2014.
	Qiao Zhenyue . Research on adaptive delay estimation algorithm of minimum mean square error[D]. Xi’an:Xidian University, 2014.
[7]	Paulo S R Diniz. 自适应滤波算法与实现[M]. 2版.北京: 电子工业出版社, 2004.
	Paulo S R Diniz . Adaptive filtering algorithm and implementation[M]. 2 ^ed Edition.Beijing:Publishing House of Electronics Industry , 2004.
[8]	陈霄, 徐慨, 杨海亮 . 基于LMS自适应滤波和希尔伯特差值的二次相关时延估计算法[J]. 指挥控制与仿真, 2015,37(3):69-74.
	Chen Xiao, Xu Kai, Yang Hailiang . Quadratic correlation time delay estimation algorithm based on Hilbert interpolation and LMS adaptive filtering method[J]. Command Control & Simulation, 2015,37(3):69-74.
[9]	汪潮, 单家方 . 一种新的变步长LMS自适应滤波算法仿真及性能分析[J]. 科技通报, 2015,31(1):87-89.
	Wang Chao, Shan Jiafang . Simulation of new variable step size LMS adaptive filtering algorithm[J]. Bulletin of Science and Technology, 2015,31(1):87-89.
[10]	张宇, 严天峰, 许富馨 , 等. 一种基于幅值加权的互相关时延估计新方法[J]. 电子科技, 2019,32(8):1-5.
	Zhang Yu, Yan Tianfeng, Xu Fuxin , et al. A new method based on amplitude weighting for cross-correlation delay estimation[J]. Electronic Science and Technology, 2019,32(8):1-5.
[11]	李红姝 . 时差估计及三星定位算法研究[D]. 哈尔滨:哈尔滨工业大学, 2010.
	Li Hongshu . Estimation of TDOA and research on three-satellite location algorithms[D]. Harbin:Harbin Institute of Technology, 2010.
[12]	Simon H. 自适应滤波原理[M]. 郑宝玉, 译.4版.北京:电子工业出版社, 2010.
	Simon H. Principles of adaptive filters[M]. Zheng Baoyu, Translated.4th Edition.Beijing:Electronic Industry Press, 2010.
[13]	彭宏, 陈泓宇 . 一种新的变步长LMS算法及分析[J]. 浙江工业大学学报, 2018,46(1):45-50,82.
	Peng Hong, Chen Hongyu . Anew variable step size LMS algorithm and analysis[J]. Journal of Zhejiang University of Technology, 2018,46(1):45-50,82.
[14]	王锋, 刘鹏远, 李兵 . 相关Hilbert差值时延估计方法误差分析[J]. 中国测试, 2015,41(12):14-16,24.
	Wang Feng, Liu Pengyuan, Li Bing . Error analysis of related Hilbert difference time delay estimation methods[J]. China Test, 2015,41(12):14-16,24.
[15]	孙洋, 徐慨, 杨海亮 . 基于广义互相关时延估计算法的性能分析[J]. 计算机与数字工程, 2013,41(1):33-34,144.
	Sun Yang, Xu Kai, Yang Hailiang . Performance analysis based on generalized cross-correlation time delay estimation algorithm[J]. Computer and Digital Engineering, 2013,41(1):33-34,144.
[16]	邓华 . MATLAB通信仿真及应用实例讲解[M]. 北京: 人民邮电出版社, 2003.
	Deng Hua. MATLAB simulation of communication and the explanation of application examples[M]. Beijing: The People’s Posts and Telecommunications Press, 2003.

SNR	算法1	算法2	算法3
20 dB	1.035×10^-6 s	1.030×10^-6 s	1.015×10^-6 s
10 dB	1.040×10^-6 s	1.025×10^-6 s	1.020×10^-6 s
0 dB	1.060×10^-6 s	1.030×10^-6 s	1.025×10^-6 s
-5 dB	-	1.085×10^-6 s	1.040×10^-6 s

时延估计方法	真实时延	估计均值	估计均值标准差	估计误差
算法1	1	1.034	0.034	3.4%
算法2	1	1.030	0.030	3.0%
算法3	1	1.012	0.012	1.2%

时延估计方法	真实时延	估计均值	估计均值标准差	估计误差
算法1	1	1.037 75	0.037 75	3.775%
算法2	1	1.031 00	0.031 00	3.100%
算法3	1	1.013 25	0.013 25	1.325%

时延估计方法	真实时延	估计均值	估计均值标准差	估计误差
算法1	1	1.053 00	0.053 00	5.300%
算法2	1	1.033 25	0.033 25	3.325%
算法3	1	1.020 75	0.020 75	2.075%

时延估计方法	真实时延	估计均值	估计均值标准差	估计误差
算法1	1	-	-	-
算法2	1	1.061 25	0.061 25	6.125%
算法3	1	1.031 00	0.031 00	3.100%