Radar repetition frequency change robust target identification method based on resampling preprocessing

Abstract

The invention provides a radar repetition frequency change robust target recognition method based on resampling preprocessing, which is used for solving the problem of radar target recognition under the condition that pulse repetition frequencies in training and testing stages are not matched, and comprises the following implementation steps of: (1) generating a training set; (2) generating a training feature matrix; (3) training an SVM classifier; (4) acquiring a test set; (5) performing conventional pretreatment on the test set; (6) carrying out resampling preprocessing on the test set; (7) generating a test feature matrix; and (8) carrying out target identification. According to the invention, through resampling preprocessing, the pulse repetition frequency of all echo signals to be classified is consistent with the pulse repetition frequency of the echo signals in the training set, and stable identification of a target can be realized in a pulse repetition frequency change scene.

Description

technical field [0001] The invention belongs to the technical field of radar, and further relates to a radar repetition frequency change robust target recognition method based on resampling preprocessing in the technical field of radar automatic target recognition RATR. This method can realize real-time recognition of different targets moving in the air and on the ground in the scene of radar repetition frequency change. Background technique [0002] Radar target recognition methods all have a premise that the pulse repetition frequency in the radar parameters needs to satisfy the Nyquist sampling theorem. However, limited by the hardware and software conditions and working modes of the radar system in practice, the radar pulse repetition frequency is often low, and spectrum aliasing will occur at this time. For pulse Doppler radar, the pulse repetition frequency determines the echo sampling rate. Generally speaking, the higher the pulse repetition frequency, the more infor...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the deficiencies in the prior art above, and propose a method for robust target recognition based on resampling preprocessing for radar repetition frequency changes, which is used to solve the problems in the prior art caused by changes in the pulse repetition frequency of test echo signals. The problem of model mismatch, which leads to a decrease in recognition rate

Images