非均匀噪声环境下宽带近场源定位的两种改进方法

Two Novel Methods for Wideband Source Localization in the Presence of Unknown Nonuniform Noise

研究了宽带近场信号源基于最大似然方法和相关信号子空间方法在非均匀噪声下的被动定位算法 ,并进行了比较。这两种算法均可在传感器任意分布的情况下有效地进行信号源定位。最大似然法采用了迭代的方法来估计噪声的协方差矩。而信号子空间法给出了聚焦阵构造的新方法。

In this paper, we derive the maximum-likelihood (ML) and coherent signal-subspace location estimator for wideband sources in the near field of the sensor array on the background of nonuniform white noise. By comparison of the two methods, the proposed algorithms can deal with source localization efficiently for randomly distributed sensors array. The estimation of noise covariance matrix is based on iterative procedure in ML. And a new focussing matrix is proposed in coherent signal-subspace method. Extensive simulations demonstrate that the algorithms have good performance.