Computationally efficient subspace-based method for direction-of-arrival estimation without eigendecomposition
Computationally efficient subspace-based method for direction-of-arrival estimation without eigendecomposition
复制标题
DOI：
10.1109/tsp.2004.823469
复制
发表时间：
2004-04
影响因子：
5.4
通讯作者：
J. Xin;A. Sano
中科院分区：
文献类型：
--
作者：
J. Xin;A. Sano
一种计算简单且具有良好统计性能的波达方向（DOA）估计方法在阵列处理的许多实际应用中颇具吸引力。在本文中，我们通过利用阵列几何结构及其平移不变性，针对均匀线性阵列（ULA）上的相干窄带信号提出了一种新的计算高效的无需特征分解的基于子空间的方法（SUMWE）。入射信号的相干性通过子阵列平均去相关，零空间通过由一些传感器数据之间的互相关形成的矩阵的线性运算获得，其中加性噪声的影响被消除。因此，无需进行特征分解就可以估计波达方向，也无需计算阵列数据的所有相关性。此外，SUMWE也适用于部分相干或非相干信号的情况，并且通过选择适当的子阵列可以扩展到空间相关噪声的情况。对SUMWE进行了统计分析，并推导了估计误差的渐近均方误差（MSE）表达式。展示了SUMWE的性能，并通过数值例子验证了理论分析。结果表明，SUMWE在少量快拍和低信噪比（SNR）情况下分辨紧密间隔的相干信号方面具有优势，并且对于不相关和相关的入射信号都提供了良好的估计性能。
A computationally simple direction-of-arrival (DOA) estimation method with good statistical performance is attractive in many practical applications of array processing. In this paper, we propose a new computationally efficient subspace-based method without eigendecomposition (SUMWE) for the coherent narrowband signals impinging on a uniform linear array (ULA) by exploiting the array geometry and its shift invariance property. The coherency of incident signals is decorrelated through subarray averaging, and the null space is obtained through a linear operation of a matrix formed from the cross-correlations between some sensor data, where the effect of additive noise is eliminated. Consequently, the DOAs can be estimated without performing eigendecomposition, and there is no need to evaluate all correlations of the array data. Furthermore, the SUMWE is also suitable for the case of partly coherent or incoherent signals, and it can be extended to the spatially correlated noise by choosing appropriate subarrays. The statistical analysis of the SUMWE is studied, and the asymptotic mean-squared-error (MSE) expression of the estimation error is derived. The performance of the SUMWE is demonstrated, and the theoretical analysis is substantiated through numerical examples. It is shown that the SUMWE is superior in resolving closely spaced coherent signals with a small number of snapshots and at low signal-to-noise ratio (SNR) and offers good estimation performance for both uncorrelated and correlated incident signals.