Reasearch On Processing Algorithms Of Weak Life Target Echos

The periodic motion generated by breathing and heartbeat of organisms, such ashumans and animals, can cause small fluctuation to their outside body (e.g., the chest).Accordingly, it will bring about Micro-Doppler modulation to the electromagneticsignals. By extracting, analyzing and processing Micro-Doppler characteristic ofliving organisms, we can not only estimate the frequency of their breathing andheartbeat, but also implement detection and location of the organism targets.Therefore, such research has significant application values in rescue areas such asearthquake, fire and debris flow. Also, it can help to ensure the safety of hostages inthe war on terror. Based on the micro-motion model of life, this paper makes study onaspects of the traditional processing algorithms, the time-frequency analysisalgorithms, and the algorithms based on spectrum estimation. The specific content ofeach part is as follows:1. The double micro-motion model of breathing and heartbeat is built for thephysical characteristics of weak life targets. Also, traditional processing algorithmswhich are based on Fourier and Multi-layer Autocorrelation are analyzed. Theeffectiveness of the two algorithms are validated and compared through differentsimulations.2. For the built echo model of weak life targets, three different time-frequencyanalysis algorithms which are respectively based on Gabor Transform, SPWVD andWavelet Transform are studied. Furthermore, an improved processing method basedon Wavelet Transform is proposed, which can improve the signal to noise ratio (SNR)of echo signals, by modifying the threshold function. Finally, the performance ofthese three methods is validated by simulations, which shows that such methods basedon the time-frequency analysis are superior to the traditional algorithms.3. Spectrum estimation algorithms which are based on the MUSIC, the MVDRand the AR model are researched as the third part in this paper and simulation resultsgive analysis on their performance. As It turns out, such methods based on spectrumestimation can achieve high SNR for respiratory signals which are strong. However,for the relatively weak signals created by life’s heartbeat, the SNR is rather low.4. Ultimately, the echo model of the Stepped-Frequency continuous wave is built forthe locating problem of weak life targets. Signal processing methods based on Fourier Transform, Wavelet Transform and MUSIC algorithm are studied respectively. Also,simulation and verification of measured data have been done about all these proposedmethods. The results show that the radar which is based on the Stepped-Frequencycontinuous wave system not only can detect the weak life, but also can performaccurate locating.