Land-sea clutter classification method based on time domain and frequency domain multi-features

Abstract

The invention discloses a land-sea clutter classification method based on time domain and frequency domain multi-features. The method comprises the following steps: 1, calculating normalized power variance, Doppler bandwidth, Doppler spectrum entropy and Doppler spectrum peak power ratio of echoes of a single radar spatial resolution unit; 2, generating feature data and category data of ground clutter and sea clutter based on the step 1; 3, constructing a BP neural network; 4, calculating normalized power variance, Doppler bandwidth, Doppler spectrum entropy and Doppler spectrum peak power ratio of each spatial resolution unit in a radar detection scene; and 5, performing judgment according to an output result. According to the land-sea clutter classification method based on time domain and frequency domain multi-features, land and sea clutter classification of different motion platform radars can be realized.

Description

technical field [0001] The invention belongs to the field of research on radar clutter characteristics, and in particular relates to a classification method for ground-sea clutter based on time-domain and frequency-domain multi-features in the field, which can be used for the classification of ground-sea clutter in radar detection scenes. Background technique [0002] Maritime surveillance radar detection scenarios are complex, including not only ocean areas but also land areas and sea islands. In order to improve the performance of radar target detection, prior clutter knowledge is needed. The radar sea surveillance scene is divided into sea clutter area and ground clutter area, and the radar detection performance can be improved by processing them separately. The study of clutter characteristics shows that there are certain differences in amplitude statistical characteristics, spectral characteristics, and scattering characteristics between ground and sea clutter, which pr...

AI Technical Summary

Problems solved by technology

There are several limitations in this method: one is that the phase information of the radar echo is not considered; the other is that the amplitude distribution model of the clutter is limited by the K distribution shape parameter of the radar echo amplitude; the third is that for each radar spatial resolution unit The estimation of statistical properties such as amplitude distribution and time correlation requires a large number of data samples. In the literature, the echo data of multiple scanning cycles is used to solve the problem of insufficient data samples of a single radar spatial resolution unit, which makes this method only applicable to work. Radar on a stationary platform; Fourth, there are differences in the physical parameters of different radar spatial resolution units in the same detection scene, and there is an aliasing area in the dynamic range of ground clutter and sea clutter scattering intensities, making the amplitude mean feature important in ground-sea clutter classification. less effective in

Images