Radar Adaptive Waveform Selection Behavior Recognition Based on Neural Network

Cognitive radar has the ability to perceive the environment and target information, and complete the work state optimization through analysis and decision. Radar's behavior is defined as all the characteristics of a radar's performance and the process of changing its state according to the law of change and the external influences. In this paper, the adaptive selection behavior principle of radar emission waveform is firstly described. Then, for the waveform adaptive selection behavior, a black box model is constructed by using neural network. The environment, target characteristics and waveform parameters are used as input training data, and the waveform selection result is taken as output. By learning the training waveform selection process of the radar, we will obtain a neural network model with the same capabilities as the radar adaptive waveform selection system, thus completing the identification of the radar adaptive waveform selection behavior. Finally, through simulation analysis, the method has good results.