TDOA-Based Source Localization With Distance-Dependent Noises
TDOA-Based Source Localization With Distance-Dependent Noises
复制标题
基于 TDOA 的具有距离相关噪声的源定位
DOI：
10.1109/twc.2014.2351798
复制
发表时间：
2015-01-01
影响因子：
10.4
通讯作者：
Yang, Zai
中科院分区：
文献类型：
--
作者：
Huang, Baoqi;Xie, Lihua;Yang, Zai
本文主要研究二维和三维空间中利用到达时间差（TDOA）测量的声源定位问题。与现有的研究不同，TDOA测量噪声的方差被假设为独立的相关的源到传感器的距离，我们认为更现实的模型，方差是一个函数的源到传感器的距离，这大大复杂TDOA为基础的源定位。在建立了距离相关的噪声模型后，证明了利用函数方差中的源位置信息可以提高基于TDOA的源定位的估计精度，但在噪声水平足够小的情况下，其贡献很小。进一步，我们从理论上分析了传感器的最优配置问题，并推导出在不同情况下优化定位性能的充分必要条件。然后，定位方案的迭代加权广义最小二乘（IRGLS）方法的基础上，提出了有效地利用额外的源位置信息。最后，仿真分析证实了我们的理论研究，并表明，所提出的定位方案的性能是相媲美的Cramer-Rao下限（CRLB）给予温和的TDOA测量噪声。
This paper focuses on the problem of source localization using time-difference-of-arrival (TDOA) measurements in both 2-D and 3-D spaces. Different from existing studies where the variance of TDOA measurement noises is assumed to be independent of the associated source-to-sensor distances, we consider the more realistic model where the variance is a function of the source-to-sensor distances, which dramatically complicates TDOA-based source localization. After formulating the distance-dependent noise model, we prove that using the extra information about the source location in the functional variance improves the estimation accuracy of TDOA-based source localization, but contributes little under a sufficiently small noise level. Further, we theoretically analyze the problem of optimal sensor placement, and derive the necessary and sufficient conditions for optimizing localization performance under different circumstances. Then, a localization scheme based on the iteratively reweighted generalized least squares (IRGLS) method is proposed to efficiently exploit the extra source location information. Finally, a simulation analysis confirms our theoretical studies, and shows that the performance of the proposed localization scheme is comparable to the Cramer-Rao lower bound (CRLB) given moderate TDOA measurement noises.