摘要
为降低到达时差测量噪声对测向精度的干扰,避免算法结果不收敛现象,提出一种基于莱温伯格-马夸特的测向算法。用线性最小二乘(LLS)算法求出的闭式解作为该算法的初始方位估计,通过迭代运算求得辐射源的方位,从而实现对声源方位的高精度估计。实验结果表明,相对LLS算法、Taylor算法,该算法能够达到克拉姆-拉奥下界,在保证结果收敛的同时提高测向精度,且具有鲁棒性。
A direction finding algorithm based on Levenberg-Marquardt(LM)is proposed to reduce the interference of Time Difference of Arrival(TDOA)measurement noise on lateral accuracy and avoid the phenomenon that algorithm results are not convergent.The closed-form solution obtained by the Linear Least Squares(LLS)algorithm is used as the initial bearing estimation of the algorithm,and the azimuth of the radiator is obtained by iterative operation,so as to achieve a high accuracy estimation of the acoustic source azimuth.Experimental results show that,compared with LLS algorithm and Taylor algorithm,this algorithm can reach the Cramer-Rao Lower Bound(CRLB),improve the direction finding accuracy and robustness while ensuring the convergence of the results.
作者
侯东升
崔逊学
HOU Dongsheng;CUI Xunxue(Graduate Management Team;Army Artillery Air Defense Academy,Hefei 230031,China;Sixth Department,Army Artillery Air Defense Academy,Hefei 230031,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2018年第11期109-114,共6页
Computer Engineering
基金
国家自然科学基金"基于TDOA测向交叉的脉冲声源定位优化方法研究"(61672532)
关键词
到达时差
测向
迭代运算
莱温伯格-马夸特算法
克拉姆-拉奥下界
Time Difference of Arrival(TDOA)
direction finding
iterative operation
Levenberg-Marquardt(LM)algorithm
Cramer-Rao Lower Bound(CRLB)
