A Study on Radar Target Detection Based on Deep Neural Networks

Target detection is one of the most important radar applications widely used in practice. Target detection can be regarded as a kind of classification, which distinguishes whether the signal undertested consists of an echo from a target (target present) or just corresponds to the noise (target absent). The deep neural network (DNN) is a popular topic for classification and has successfully been applied in different fields of science. Recently, many researchers have proposed DNNs for radar applications. In this article, we analyze a possible application of DNN- for target detection in radar, DNN-based detectors are designed, and the performance of the detector is demonstrated by comparison with traditional target detectors.