Target Detection in Mainlobe Jamming Using Convolutional Neural Network

Abstract

The traditional array adaptive beamforming methods yield serious target loss in the presence of mainlobe jamming. Super-resolution methods based on sparse recovery can separate the mainlobe jamming from the target in the space domain, but it cannot reconstruct the target information when the angle between the jamming and the target is too small or the signal-to-noise ratio is too low. In order to solve above problems, we propose a novel super-resolution method based on the convolutional neural network. The proposed method can effectively separate the mainlobe jamming and the weak target even within the half-power mainlobe width, thus can achieve the target detection in the mainlobe jamming scenario. Simulation experiments verify the effectiveness of the proposed method.