Radar signal sorting based on GRU neural network
Jigang Wu, Long Teng, Qiang Guo · 2021 IEEE 5th Information Technology,Networking,Electronic and Automation Control Conference (ITNEC) · 2021
When the features of the emitter signals are similar, the emitter signals intercepted by the reconnaissance equipment overlap seriously in the space, time, and frequency domains, and the existing technology cannot effectively separate them. This paper considers that when the antenna of the emitter is scanning, the pulse amplitude (PA) sequence intercepted by the reconnaissance receiver has certain regularity. Therefore, a signal sorting method based on the feature of pulse amplitude is proposed. Firstly, the whole pulse sequence is preprocessed, and then a neural network prediction model based on Gated Recurrent Unit (GRU) is established. The GRU neural network is used to learn the changing law of the amplitude sequence. Finally, the learned law is used to predict and sort subsequent pulses. The simulation results show that compared with the listed methods, the proposed method has a better sorting effect.