多站时差定位中的病态性分析及其改善方法

在多站时差定位过程中，传统定位算法在传感器布站不合理的情况下通常会表现出明显的病态性，其导致定位结果不稳定、精度低。针对此问题，首先分析了时差定位过程中病态性产生的物理场景，并通过奇异值分解揭示了病态性的本质；其次根据时差定位观测矩阵的特征值分布特点，提出了一种改进的多站时差定位改善方法，该方法基于LS估计信息矩阵的条件指标进行奇异值修正，兼顾了目标定位的分辨率与方差，能显著提高多站时差目标定位的精度和稳定性，且没有增加系统和算法复杂度；最后将该算法与传统算法进行了仿真比较，结果证明了该算法优越性。

Ill-Conditions Analyzed and Modified Estimation for Multi-Sensor TDOA Location

In connection with the problem of unstable Location and low precision using TDOA owing to diseased info matrix, firstly the reason of ill-condition is finded, and eigenvalue of observation matrix using TDOA is analyzed,a modified location algorithm using TDOA was presented,which decompose singularly value based on LS condition index and includes distinguishability and error.The proposed algorithm can enhance precision and stability of TDOA location,it not only bring system complexity,but also algorithm complexity. The simulations have shown the method was effective.