Radar Target Detection using Convolutional Neutral Network in Clutter

In our paper, we present CNN detector which is a new approach to achieve radar target detection. As a binary hypothesis testing, radar target detection can be viewed as a binary classification. We utilize convolutional neural network (CNN) to perform detection owing to its extraordinary classification performance. CNN detector searches targets and determines locations of potential targets so it can be directly used in radar system. Based on actual radar data, we test the performance of CNN detector and compare it with CFAR detector. The results of evaluation illustrate that CNN detector is pretty better than traditional method. In particular, it detects much more true targets than CFAR detector under the condition of the same number of false alarms in the presence of clutter.