Signal and noise subspace decomposition for a linear antenna array using SVD

摘要

In the vast majority of signal processing applications, the signal is intended to be analyzed to obtain a variety of data from the signal. It is decomposed into signal subspaces to perform this analysis. The noise in the signal during analysis is undesirable. If the signal incoming to an antenna array contains noise, it should be filtered before processing signal. In this study, we have tried to divide the uniform linear antenna array into subspaces of incoming signals from multiple signal sources with various frequencies, noise and arrival angles. The Singular Value Decomposition (SVD) method is used to separate the signal subspace and noise subspace. According to the singular values obtained, the information about the frequencies of the signals and whether the signal contains noise. Also, the number of sources can be estimated by means of singular values.