SVD-based angular difference estimation
Yutong Wu, Dazhuan Xu, Ying Zhou, Jingjing Pan · 2022
The paper studies the estimation of the angular difference between adjacent sources in a uniform linear array. When there is additive Gaussian noise, in order to obtain the eigenvalue expression related to the signal, we can apply singular value decomposition, getting the eigenvalue of the spatial covariance matrix. One of the feature values is related to the degree of difference between the two sources, and the expression of the angular difference estimation is derived using this difference. The larger the observed samples, the more precise the estimation. The experimental results prove our theoretical analysis. Meanwhile, the feasibility of the proposed estimation method has been proved. Importantly, the results of this paper have pragmatic guiding meaning for array signal processing.