Plane array angle estimation method based on four-linear decomposition

Abstract

The invention discloses a plane array angle estimation method based on four-linear decomposition. A non-uniform noise covariance matrix is constructed according to an array signal model receiving array elements and the fourth-order tensor of the non-uniform noise covariance matrix is constructed; a receiving direction matrix, a transmitting direction matrix and a target covariance matrix are initialized, and the receiving direction matrix, the transmitting direction matrix and the target covariance matrix are initialized into a unit matrix; iterative computation of the receiving direction matrix, the transmitting direction matrix and the target covariance matrix is performed by using the alternating least square method until the convergence condition is met; the target DOD and DOA are estimated through the g least square method; the estimation method can effectively resist non-uniform white noise and can obtain great estimation effect under the condition of small snapshot and big snapshot; and the parameter estimation accuracy is higher than that of the ESPRIT algorithm and the HOSVD algorithm and the complexity is lower than that of the PARAFA C algorithm.

Description

technical field [0001] The invention relates to a radar signal processing technology, in particular to a method for estimating the angle of a plane array based on four-linear decomposition. Background technique [0002] Array signal processing is an important branch of signal processing. Its essence is to use spatially dispersed sensor arrays and multi-channel reception to obtain multi-dimensional information such as time domain and space domain of signals, so as to achieve the purpose of detecting signals and extracting their parameters. It has been widely used and developed rapidly in the fields of communication, radar, sonar, seismic exploration and radio astronomy, and has great application prospects. In recent years, multi-parameter estimation using array signal processing has received more and more attention, especially the spatial direction of arrival (DOA) estimation has become one of the hot research issues. The problem of two-dimensional parameter estimation is a ...

AI Technical Summary

Problems solved by technology

The problem of two-dimensional parameter estimation is a very practical problem. The traditional DOA estimation generally refers to estimating the azimuth angle of the signal source in a plane, which belongs to the one-dimensional parameter estimation problem and has many limitations.

At this time, the signal and noise cannot be effectively separated, which causes the performance of angle estimation to decline. Especially under the condition of low signal-to-noise ratio, the existing algorithm may be distorted or even completely invalidated, which is not suitable for engineering applications.

Images