Research On Early Warning Method Of Perimeter Invasion Based On Vibration Sensing

With the continuous development and progress of society,people's safety precautions are constantly being strengthened.In the current perimeter warning system,there are more and more high-tech applications,mainly using infrared,laser,microwave,video,vibration,etc.the way.These current warning systems all have their own characteristics and scope of use.At the same time,they also have their own limitations,such as poor equipment concealment,low early warning efficiency,and harsh equipment installation conditions.For the problem of the above perimeter warning system,this paper studies the use of a geophone as a signal detection unit.The camera serves as a real-time picture feedback unit of the monitoring system and builds a perimeter intrusion prevention system based on vibration signals.The system has the advantages of low false alarm rate,strong concealment of the detection unit,strong resistance to external interference,limited deployment area,and simple installation,realizing real-time monitoring of intrusion sources without being noticed.The system is simple and convenient to operate,and has great practical value in military and civilian use.This paper first introduced the mechanism of the generation and propagation of seismic waves,and explained in detail the propagation characteristics of the Rayleigh wave,the main carrier of energy in seismic waves and the working principle of the geophone was briefly explained and a large number of sample signals were collected at the same time which mainly includes background noise signals,single-person walking,single-person running,two-person walking,two-person walking and low-velocity motion signals.The characteristics of each sample signal are studied.Then,based on the analysis of the collected sample signals,a theoretical analysis method for the target signal is proposed.For the target signal,zero-crossing analysis and kurtosis analysis are used in the time domain to eliminate the irrelevant signals quickly,and the frequency characteristics of the signal are quickly obtained through power spectrum analysis in the frequency domain.Then,the principles of wavelet theory and support vector machine are introduced in detail.The process of wavelet packet decomposition,wavelet packet and how to extract the feature vector of the sample signal and its normalization,and how to construct the optimal classification plane and selected the nuclear function etc.At last,in conjunction with video linkage,the vibration signal is associated with real-time video surveillance,which greatly reduces false alarms in the early warning process and improves the system's practicality.Finally,a perimeter intrusion warning system based on vibration signal is designed,which consists of software system and acquisition unit.The host computer software system mainly uses the LabVIEW and MATLAB joint programming methods to complete the storage of the collected data,the application of the algorithm,and the data processing functions and the data acquisition unit consists of a capture card,a geophone,and a camera.