Radar Behavior Recognition Based on Radar Word Coding and Intelligent Dual Mode Recognition

In order to solve the problem of low recognition accuracy of radar working mode caused by the reuse of the same radar phrase by multiple working modes, this paper proposes a method based on mapping coding and intelligent dual-mode recognition. First, the physical entity of radar waveform is mapped to a digital code. Thirdly, the long and short term memory (LSTM) neural network embedded with attention mechanism is used to extract the digital code sequence to realize the automatic recognition of the working mode of radar emitter. Finally, an auxiliary recognition method based on the knowledge map is proposed, which inputs the double time sequence code into the central network through the knowledge map to assist the decision, so as to improve the recognition accuracy. The simulation test shows that the recognition method proposed in this paper can effectively improve the recognition accuracy of radar working mode compared with traditional methods.