Study On The Identification And Location Of The Vibratory Signal On Ground

With the development of Internet of things and enhancing of awareness security, the anti-intrusion technology market demand is rapidly growing. This paper focuses on the anti-intrusion sensor technology in seismic detection subsystem, which includes three blocks:sensors for detecting seismic signal and circuits for signal conditioning, identification and classification of seismic signal, feature fusion and alarm parts.In the first part of the paper, generation and propagation theory of seismic wave, which generated by intruder's moving on the ground, has been studied. Based on this theory, we determine the main exploration target is Rayleigh wave. And by the research of seismic wave modeling theory, the propagation characteristics of the seismic wave has been analysed in the different topographic and geologic conditions.A variety of card-bit methods are analyzed and compared for the detection signal. A weighted cards bit method is been proposed, which is used the characteristics of the seismic wave. In time domain, we use the number of zero to analyse the signal, and modify the threshold selection criteria of the number of zero. Kurtosis feature has been used to pick up statistical characteristics of the seismic signal, and classify the seismic signal. In frequency domain, we use the Ratio of High frequency and Low frequency(RHL) energy of seismic signal to distinguish step signal from other noises by analyzing the frequency spectrum feature. We employ Fractional Fourier Transformation (FRFT) analyze seismic signal in Time-Frequency domain,. After obtain the best order of transformation, according to the order, making the signal transform to FRFT domain, then extract time-frequency characteristics and use it for identification. The system makes ARM chips as the microprocessor, which is used for Real-time data acquisition and processing. Data is sent to the PC via serial port. Real-time data and alarm are display on software, which is written by LabVIEW. Location is based on Time-Difference and Hyperbolic principle. It can be used to estimates the location of the intruder and the invasion level. Experimental results show:a good warning effect has achieved.