Frequency-Hopping Signal Sorting Based on Deep Leaming

In order to solve the problem that the sorting threshold of traditional frequency-hopping signal needs to be manually adjusted, Faster-RCNN and clustering algorithm is proposed. In this paper, the Faster-RCNN is firstly used to identify and locate all frequency-hopping points in the time- frequency spectrum diagram, and then AlexNet is used to obtain the number of frequency-hopping signal. Experimental results show that the Faster-RCNN can be effectively used for automatic signal sorting when the number of frequency- hopping signal is small.