Airplane target classification method based on time domain correlation characteristics

Abstract

The invention discloses an airplane target classification method based on time domain correlation characteristics, and relates to the technical field of radar. The method includes the steps that firstly, a training sample peak value function is calculated; secondly, the variance, the entropy, the number of peaks with the values larger than a first training peak value threshold and a time domain point corresponding to the first peak with the value larger than a second training peak value threshold of the training sample peak value function are calculated; thirdly, the amplitude variance and the amplitude entropy of training samples are calculated; fourthly, training sample characteristic vectors are normalized, and a classifier is trained; fifthly, a test sample peak value function is calculated; sixthly, the variance, the entropy, the number of peaks with the values larger than a first test peak value threshold and a time domain point corresponding to the first peak with the value larger than a second test peak value threshold of the test sample peak value function are calculated; seventhly, the amplitude variance and the amplitude entropy of test samples are calculated; eighthly, test sample characteristic vectors are normalized and input into the classifier for class judgment. The method has good advantages at low repetition frequency and can be used for classification of three classes of airplane targets.

Description

technical field [0001] The invention belongs to the technical field of radar and relates to a radar signal classification method, in particular to an aircraft target classification method based on time-domain correlation features. Background technique [0002] In modern warfare, helicopters undertake important tasks such as artillery fire correction, reconnaissance, airborne landing behind enemy lines, and maneuver transfer. The main mission of propeller aircraft is to seize control of low and ultra-low altitudes. Performance, can fight at an extremely fast speed in the air. The three types of aircraft play their respective important roles on the battlefield, so it is of great significance to realize the classification of the three types of aircraft. [0003] So far, the process of feature extraction in relevant literature is mainly carried out in the Doppler domain and the characteristic spectral domain; the features extracted in the Doppler domain can mainly reflect the d...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the shortcomings of the feature extraction methods in the prior art, an aircraft target classification method based on time-domain correlation features is proposed. In the case of spectral aliasing and Doppler domain feature separability decline, the purpose of better classification effect is still achieved, and compared with the prior art four-dimensional feature extraction method based on feature spectrum distribution features, the present invention also has the advantages of The advantage of fast calculation speed

Images