噪声向量模值最小的水声网络TDOA目标定位方法

TDOA-based target localization method by minimizing module of noise vector in underwater acoustic networks

由于水下环境的复杂性导致水声网络节点通常存在一定的漂移,从而引起网络节点自定位的不准确;又因为水下测距不准确导致TDOA测距中也存在一定的误差。以上两类前期噪声误差均会降低网络对目标定位时的精度。针对以上问题,本文提出一种基于噪声向量模值最小的高精度水声网络TDOA目标定位方法。该方法利用LS(least-squares)算法得到目标定位的初值,通过考虑节点自定位误差和TDOA测距误差对算法精度的影响,经过一系列转换得到目标函数,使得上述两种前期噪声误差对定位精度的影响达到最小;根据初值及目标函数,采用模拟退火智能优化算法得到目标位置。仿真结果表明:与WLS(weighted least-squares)算法、CTLS(constrained total least-squares)算法相比较,本文算法定位精度高且前期误差对算法性能影响小,鲁棒性强。

Given the complexity of underwater environments,network nodes are usually unstable,which leads to inaccurate self-localization. Consequently,time difference of arrival（ TDOA） measurements are not accurate,which decreases the precision of the resulting location information. To solve the aforementioned problems,we propose a TDOA-based target localization method that employs the minimizing the module of noise vector in underwater acoustic networks. This algorithm uses the least-squares method to calculate the initial target position.Then,by considering the TDOA measurement error and the self-positioning error,an objective function is obtained through a series of transformations which minimizes the influence of the above errors on location accuracy. According to the initial value and the objective function,a simulated anneal algorithm is used to obtain the exact position of the target. Simulation results demonstrate that MMNV is superior to the weighted least-squares（ WLS） and the constrained total least-squares（ CTLS） algorithms in terms of positioning accuracy,robustness,and effect of errors on the result.