Harmonic radar target recognition method based on machine learning

At present, the research of harmonic radar is mainly concentrated in the field of target detection. The difficulty is that the second harmonic component is obviously higher than the third harmonic component in the echo information of some third harmonic objects, which leads to the result of detection errors. Through the exploration of machine learning methods, this paper finds out a set of feasible recognition process. It aims to make up for the shortcomings in the field of detection through classification theory. And the objects that are easily confused in daily life are predicted and recognized. Experiments show that in the recognition method based on machine learning, the size of data set and waveform characteristics have important contributions to the classification results. And it has a certain accuracy for easily confused objects.