Radar high resolution range image target recognition method based on mmfa model

Abstract

The invention discloses a radar high-resolution range profile target recognition method based on an MMFA model in order to mainly solve the problems of high solving complexity, and poor classification and rejection performances in the prior art. The realization steps of the method include: firstly, extracting radar HRRP data features and obtaining a power spectrum feature set; secondly, building the MMFA model, and obtaining a probability density function of power spectrum features and the combined condition posterior distribution of various parameters; thirdly, deducing the condition posterior distribution of various parameters; fourthly, sampling the various parameters for I<0> times in a circulating manner; fifthly, saving sampling results of the required parameters in the test stage for T<0> times; sixthly, obtaining a hidden variable of the power spectrum features of the tested radar HRRP via an FA model, calculating the probability density value, and determining whether the tested radar HRRP is an outside-base sample; performing rejection if yes; and brining the hidden variable into a classifier for determining target classification and outputting classification labels if not. According to the method, the complexity is low, the recognition and rejection performances are high, and the method can be used for the recognition of radar targets.

Description

Technical field [0001] The present invention belongs to the technical field of radar, and relates to a radar target recognition method, in particular to a radar high-resolution range image target recognition method based on a maximum boundary factor analysis MMFA model, which is used to recognize targets such as airplanes and vehicles. Background technique [0002] Radar target recognition is to use the radar echo signal of the target to realize the judgment of the target type. Broadband radar usually works in the optical zone, at this time the target can be seen as composed of a large number of scattering points with different intensities. The high-resolution range profile HRRP is the vector sum of the echoes of various scattering points on the target body obtained by using broadband radar signals. It reflects the distribution of scattering points on the target along the line of sight of the radar, contains important structural features of the target, and is widely used in the ...

AI Technical Summary

Problems solved by technology

However, these methods are unsupervised models, and the features extracted by these methods are not necessarily suitable for back-end classification tasks.

Linear discriminant analysis (LDA) is a commonly used supervised dimensionality reduction method. However, LDA requires all kinds of data to be Gaussian distributed and have the same covariance matrix, which is difficult to meet in practical applications.

Images