A PCA-BP fast estimation method for broadband two-dimensional DOA of high subsonic flight targets based on the acoustic vector sensor array

Abstract

Aiming at the problem of fast estimation of the two-dimensional Direction-Of-Arrival (DOA) of low-altitude and high-subsonic flight targets, a PCA-BP estimation method based on a uniform linear array of acoustic vector sensors arranged in a limited space was presented. For the fast direction finding of low-altitude flight targets, the traditional Principal Component Analysis (PCA) algorithm was extended to a uniform linear array of two-dimensional acoustic vector sensors, and the corresponding PCA-BP fast estimation algorithm using a neural network implementation structure was proposed. While maintaining the advantages of multiple signal classification algorithms, this method expands the application range of PCA DOA estimation methods and effectively improves the real-time performance of acoustic two-dimensional DOA estimations.