Electromagnetic vector-sensor linear array de-coherent multiple signal classification (MUSIC) parameter estimating method

Abstract

The invention proposes an electromagnetic vector-sensor linear array de-coherent multiple signal classification (MUSIC) parameter estimating method. M electromagnetic vector-sensor linear arrays distributed randomly on the y axis receive K coherent narrow bands and the electromagnetic signals of a stable far-field; the arrays obtain synchronous sampling data for N times; according to the rotation invariant feature of the sub-matrixes, the data covariance matrixes before and after the transform and the data covariance matrixes before and after the transform are obtained; based on the data covariance matrixes before and after the transform, the data covariance matrix after the de-coherent is obtained; singular value decomposition is performed to the data covariance matrix after the de-coherent to obtain the signal subspace and noise subspace; the noise subspace is used to construct the MUSIC spatial spectrum; and through the one-dimensional MUSIC spectrum peak, the arrival angle is searched and obtained. The method of the invention is combined with the use of space rotation de-coherent and the thought of the smooth sliding of a previous and after item, which means that on the premise of no loss of the array aperture, the precision of parameter estimation is increased greatly, and the method holds true for both uniform and non-uniform linear arrays.

Description

technical field [0001] The invention belongs to the technical field of signal processing, in particular to a method for estimating the arrival angle of a coherent source of an electromagnetic vector sensor array. Background technique [0002] In practice, due to the multipath phenomenon, echo signal and human interference in the signal transmission process, the correlation signal source is very common, so the detection and estimation based on the correlation source is an important research topic of signal processing. When the incident signals are independent of each other, the subspace method has good parameter estimation performance, but if the incident signal is coherent, the subspace method represented by MUSIC will fail. When the incident signal is a coherent signal, the dimensionality of the signal subspace will be smaller than the number of signal sources, and the signal subspace spreads to the noise subspace, resulting in the signal subspace and the noise subspace not...

AI Technical Summary

Problems solved by technology

Although these methods solve the coherence problem of the signal, there are still some shortcomings. The computational complexity of the non-dimension reduction method represented by focusing processing is greatly increased.

The dimension reduction method represented by spat

Images