Single-snapshot direction finding method under impact noise environment

Abstract

The invention relates to a single-snapshot direction finding method in an impact noise environment. The method comprises the steps of establishing a uniform linear-array single-snapshot sampling signal model; constructing an infinite norm covariance matrix based on a Gaussian kernel to obtain an infinite norm single-snapshot maximum likelihood equation based on the Gaussian kernel; initializing anAfrican buffalo herd; calculating the position adaptability of each buffalo, and recording the local optimal position of each buffalo and the global optimal position of the whole African buffalo herd; updating the positions and communication positions of the buffaloes to generate a Fibonacci weight; updating the local optimal position of each buffalo by means of a Fibonacci search strategy; calculating the adaptability of the new position of each buffalo and determining the local optimal position of each buffalo and the global optimal position of the African buffalo herd, wherein the output global optimal position of the African buffalo herd is an estimated value of an incoming wave direction. According to the method, only single-snapshot data is processed under strong impact noise and other complicated environments, so that DOA estimation operation amount is reduced, and effective estimation of the arrival direction of a received signal wave is achieved.

Description

technical field [0001] The invention relates to a single-snapshot direction finding method in an impact noise environment, in particular to a single-snapshot direction finding method based on a Fibonacci African buffalo search mechanism in an impact noise environment. The invention belongs to the field of array signal processing. Background technique [0002] Direction of Arrival (DOA) estimation, also known as direction of arrival (DOA), has always been a hot topic in array signal processing, and has been widely used in communication, radar and sonar systems. Although the traditional multiple signal classification method (MUSIC) and the rotation invariant subspace technique for signal parameter estimation (ESPRIT) have high estimation performance, these algorithms are based on the eigenvalue decomposition operation, and often In order to obtain a good estimation performance, the number of snapshots required is large, not only the real-time performance is low, but also the a...

AI Technical Summary

Problems solved by technology

[0004] According to the existing technical literature, Wei Yinsheng et al published in "System Engineering and Electronic Technology" (2013, Vol. "Sub-snapshot super-resolution algorithm" uses a dimensionality reduction method to estimate the covariance matrix in the frequency domain, which improves the resolution and resolution accuracy to a certain extent, but the second matrix decomposition of this method increases the amount of calculation and loses the array aperture. Effective direction finding in impulsive noise environments

Xie Xin et al. proposed a lossless The single-snapshot direct data domain algorithm of the array aperture has a low amount of computation, but this method is too demanding for the real receiving environment and cannot fail in the direction finding under the background of impact noise

[0005] Existing literature shows that single-snapshot DOA estimation can improve the real-time performance of the system and reduce the amount of computation, but the reduction in the number of snapshots will lead to inaccurate or even invalid estimation performance, and so far there is no effective method for single-snapshot DOA estimation under the background of impact noise. shot direction finding, so it is necessary to design a high-performance single-shot direction finding method suitable for impact noise background

Images