Micro moving target feature extracting method based on micro Doppler effect

Abstract

The invention provides a micro moving target feature extracting method based on micro Doppler effect, which comprises the steps that HHT (Hilbert-Huang transform) is introduced into micro moving target feature extracting, an HHT algorithm based on the down sampling EMD (empirical mode decomposition) is provided by aiming at the problem of the modal mixing of the feature extracting of the HHT, and procedures of resolving, summing and averaging on the noisy EMD are performed by the multiple groups of data obtained by performing the down sampling on original signals, thus effectively solving the mixing problem of spectrogram modes in the vibration target feature extracting of the HHT, inhibiting the noise of the original signals, improving the signal to noise ratio, reducing the EMD operating complexity of the multiple groups of data, greatly reducing the operating amount, improving the operating speed and achieving a better micro Doppler feature extracting effect. The micro Doppler feature extracting model based on the improved HHT is provided by integrating the advantages of the traditional time frequency analysis method and the improved HHT algorithm, a spectrogram peak value estimation method is added into the model, the resolution in the traditional time frequency spectrogram is improved to be used as an assisting method for the HHT feature extracting, and the requirements on accuracy and practicability of the extracted vibration target feature are achieved.

Description

technical field [0001] The invention relates to a micro-Doppler effect micro-moving target feature extraction method, which is suitable for the field of non-contact target detection and recognition. Background technique [0002] When there is relative motion between the detector and the measured target, a Doppler frequency shift will occur, and this phenomenon is called the Doppler effect. In addition to relative motion, the target itself has other motions such as the rotation of the helicopter propeller, the rotating radar antenna on the ship and other small vibrations or rotations. The phenomenon of broadening the signal frequency caused by its own additional motion, that is, the micro-Doppler effect. It is manifested in the frequency spectrum that there are spectral sidelobes or broadening. This sidelobe or broadening characterizes the unique parameter data of the micro-moving target itself, such as reflecting the electromagnetic characteristics, geometric structure and ...

AI Technical Summary

Problems solved by technology

In the current micro-Doppler feature extraction, most of the traditional time-frequency analysis methods used, such as short-time Fourier transform, WVD, SPWVD, etc., these time-frequency analysis methods have resolution problems in the time domain and frequency domain Or the choice of kernel function, there are great defects in the feature extraction of micro-Doppler, so a better method of micro-Doppler feature extraction is needed, which can extract the micro-Doppler feature parameters well

Images