Feature Extraction Of Laser Micro-doppler Effect

Now, searching for an efficient method for analyzing weak non-stationary signal to extract the Micro-Doppler effect induced by weak vibrating signal is one of the top interests in weak signal research of laser radar. This paper includes studying many feature extraction methods such as time-frequency method, multi-resolution analysis and correlation analysis to analyze the Micro-Doppler signal. The main contents are as follows:1. As a powerful tool to analyze the non-stationary signals, the time-frequency analysis shows us clearly frequency varies with time. The experiment data are analyzed with several time-frequency analysis methods such as STFT, WVD, CWD to extract the micro-Doppler effect of the target, and then the experiment results are talked about and analyzed.2. With high resolution in time and frequency area, the wavelet analysis can extract the local information of signal effectively. Doing pretreatment to micro-Doppler signal with multi-resolution analysis can eliminate noise, the results show that the multi-resolution analysis is suitable for micro-Doppler features extraction.3. Doing self or inter-correlation analysis to micro-Doppler signal with correlation analysis technology can extract the micro-Doppler feature. Using the multi-resolution analysis as pretreatment method association with the correlation analysis can improve the effect of data analysis.