基于冲激脉冲体制超宽带雷达的生命信号分析

发布时间：2019-06-24 18:45

【摘要】：面对自然灾害,可能我们无能为力,尤其是作为一个已经经历过四川多次地震的当地人。在地震的那一刻,高楼大厦顷刻间倒塌,山崩地裂,有无数的生命被掩埋在了地下。虽然现在科学技术已经取得了一定的进步,但是大家还是可以看到:每每遇到这些灾难,都会带来无法估量的损失,无论是生命安全还是物质财产。根据地理学家的分析,我们国家地处太平洋与亚欧两大地震带之间,十分容易受到周围陆地板块与海洋板块的相互挤压,所以地震频频发生,从数次灾后的救援过程中发现:灾后救援黄金时间是72小时,如果能在这段时间内更加快速、准确的找到受害者,那么受害者得救的机会就越大。正因如此,在某些紧急的情况下能够更短的时间内找到受害者,寻找一种更加先进的快速的生命探测方法就尤其重要。对被困生命体的探测就是使用各式各样的方法去采集能够代表生命特征的呼吸和心跳等相关信息进行分析,进而确定是否有存活的生命体。研究人员通过多次试验发现:以前的基于光波或声波等探测仪器极易受到环境的干扰;而电磁波则相对较稳定,在探测过程中受到周围环境影响较小,因此就有越来越多的研究者们开始将电磁波探测用于生命探测的试验中。文中根据对三种常用的超宽带进行分析对比,选择了脉冲式的超宽带对被掩埋生命体进行分析。建立一个受害人员的模型,通过脉冲式超宽带发射一个可以将生命特征信号带回的窄脉冲。根据所选超宽带的冲击响应,及生命体的振动幅度与发射天线的距离仿真接收端接收到的信息。然后对接收到的信息做初步处理,去掉十分明显的直达波与杂波等,这里主要选择的是曲波变换去除噪声的方法,该方法不仅能够去除干扰的杂波还可以提高信噪比。预处理后得到的波形信号中还是有许多种频率的信息,为了能够从中找出是否有生命特征信号的存在,选取了可根据信息本身自适应分解的方法——聚类经验模态分解;将预处理后的波形分解为若干个固有模态分量和余项序列,对这些若干个分量进行频域分析可知是否存在生命体,再用这些分量重新去构造原来的有用信息。最后为了更加确定所探测的准确性,对重新构造的生命体特征信号做频域分析和高阶频域分析,对比两种方法得到的频谱图像,从而也引进高阶累积法求超宽带探测生命的频谱分析。本文通过引入超宽带雷达的发展及其到目前为止的应用领域,接着详细的介绍了几种常见的用于地震探测生命体的超宽带雷达和它们工作的原理,并选取脉冲式的超宽带作为模拟探测的仪器,建立地震生命探测的模型。通过对雷达接收端接收信息的分析,对接收到的波形中的直达波和杂波有了更深的理解;而选择曲波变换去除噪声的过程中对波形的边缘也有了进一步认识;重点介绍了根据波形自身特点自适应分解的聚类模态分解法;最后讨论了重新构造的生命体特征信号的频域分析和高阶累积量的频域分析图,从而得出:高阶累积的频域分析的在探测中更加合适。
[Abstract]:In the face of natural disasters, there may be nothing we can do, especially as a local population that has experienced multiple earthquakes in Sichuan. At the moment of the earthquake, the building of the high-rise building collapsed, the mountain broke, and countless lives were buried under the ground. Although science and technology have made some progress, it can be seen that every one of these disasters will bring incalculable losses, whether life-safety or material property. According to the analysis by the geographer, our country is located between the Pacific and the Asia-Europe seismic belt, which is very easy to be pressed by the surrounding land plate and the ocean plate, so the earthquake happens frequently, and it is found that the golden time of the post-disaster rescue is 72 hours, The more quickly and accurately the victim can be found during this period, the greater the chance of the victim to be saved. That is why, in some emergency situations, it is particularly important to find the victim in a shorter time and to find a more advanced and rapid life detection approach. The detection of the trapped living body is to use a variety of methods to collect relevant information, such as breathing and heartbeat, which can represent the characteristics of life, so as to determine whether there is a living body. The researchers have found that previous sounding instruments based on light waves or acoustic waves are highly susceptible to environmental interference, while the electromagnetic waves are relatively stable and are less affected by the surrounding environment during the detection process. As a result, more and more researchers have begun to probe electromagnetic waves into the test for life detection. In this paper, based on the analysis and comparison of three commonly used ultra-wideband, the pulse-type ultra-wideband is selected to analyze the buried living body. A model of an injured person is established, and a narrow pulse which can bring the life characteristic signal back through the pulse-type ultra-wideband is transmitted. According to the impact response of the selected ultra-wideband and the vibration amplitude of the living body and the distance of the transmitting antenna. Then, the received information is subjected to preliminary processing to remove the very obvious direct wave and the clutter, and the method mainly selects the method for removing the noise by the wave-wave transformation, and the method not only can remove the interference clutter, but also can improve the signal-to-noise ratio. There are many kinds of frequency information in the waveform signal obtained after the pre-processing, in order to be able to find out whether there is the existence of the life characteristic signal, the method _ cluster empirical mode decomposition can be selected according to the self-adaptive decomposition of the information itself; The pre-processed waveform is decomposed into a plurality of natural mode components and a residual term sequence, and the plurality of components are subjected to frequency domain analysis to find out whether a living body exists, and then the original useful information is re-constructed by using the components. Finally, in order to determine the accuracy of the detection, the frequency-domain analysis and the high-order frequency-domain analysis of the reconstructed living body characteristic signal are performed, and the spectrum images obtained by the two methods are compared, so that the spectrum analysis of the ultra-wideband detection life is also obtained by the high-order accumulation method. In this paper, by introducing the development of ultra-wideband radar and the application field so far, several common ultra-wideband radar for earthquake detection living body and the principle of their work are introduced in detail, and the pulse-type ultra-wideband is used as an instrument for analog detection. And the model of the seismic life detection is established. by analyzing the receiving information of the radar receiving end, a deeper understanding of the direct wave and the clutter in the received waveform is provided; and the edge of the waveform is further recognized in the process of selecting the wave-wave transformation to remove the noise; In this paper, the method of cluster mode decomposition based on the self-adaptive decomposition of the waveform is introduced, and the frequency-domain analysis of the reconstructed living body characteristic signal and the frequency-domain analysis of the higher-order cumulant are discussed. The results show that the higher-order accumulated frequency-domain analysis is more suitable in the detection.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN957.51

【参考文献】