Robust Classification Method of Aircraft Target Noise Based on Generalized Matched Filter

Abstract

The invention belongs to the technical field of radar target classification and recognition, and particularly relates to an airplane target noise stable classification method based on generalized matched filtering. The airplane target noise stable classification method based on the generalized matched filtering comprises the following steps that S1, measured data are received through a radar, and the S2, S3, S4 and S5 are executed on the received measured data; S2, the autocorrelation matrix of an airplane target echo sample is obtained; S3, the whitenization matrix of the generalized matched filtering is obtained; S4, the autocorrelation matrix of the airplane target echo sample is obtained after clutter and noise are filtered out; S5, the three-dimensional characteristic spectrum distribution characteristics of the measured data are obtained; S6, a training data characteristic matrix is obtained, and a supporting vector machine classifier is trained through the training data characteristic matrix; S7, the three-dimensional characteristic spectrum distribution characteristics of the measured data are classified through the trained supporting vector machine classifier.

Description

technical field [0001] The invention belongs to the technical field of radar target classification and recognition, in particular to a generalized matching filter-based aircraft target noise robust classification method, which can classify and recognize air aircraft targets. Background technique [0002] When an object moves, in addition to its own translation, some parts on it usually have micro-motions such as rotation and vibration, such as the rotation of the wheel when the vehicle is running, and the rotation of the blades of the helicopter when it is flying. Researchers at the U.S. Naval Laboratory named the Doppler modulation phenomenon in radar echoes produced by such micro-motions as the micro-Doppler effect. The JEM (Jet Engine Modulation) phenomenon is the manifestation of the micro-Doppler effect in the field of aircraft targets, that is, when the electromagnetic data emitted by the radar irradiates the rotating blades on the moving aircraft target, the echo will...

AI Technical Summary

Problems solved by technology

[0004] In the field of radar data processing, the existing clutter suppression methods mainly include moving target indication (MTI), CLEAN and generalized matched filter (GMF). The three clutter suppression methods have their own characteristics: the MTI method is simple to implement, The amount of calculation is small, and it can be used to remove zero-frequency ground clutter, but the nonlinear modulation characteristics of MTI will affect the target echo; the CLEAN method can well retain the target while removing the clutter component in a specific frequency range information, but this method requires prior knowledge of clutter bandwidth; someone proposed a clutter suppression method based on generalized matched filtering, which is suitable for removing stationary ground clutter and does not require prior information of clutter bandwidth, and Little impact on target information

However, the computational complexity and amount of matrix operations involved in the generalized matched filtering process have high requirements on the system, and it is difficult to implement

At the same time, the above method is only to achieve clutter suppression, and does not consider the noise component contained in the radar echo

Images