Research On Seismic Sensing Perimeter Security System Based On Swarm Intelligence

As an effective protection method for key areas,the perimeter security system is widely used in people’s production and life,whether it is an unattended perimeter security system deployed on the border or the security of key civilian facilities such as warehouses and prisons.The systems all play an important role in ensuring social stability and people’s safety.In recent years,artificial intelligence technology has continued to develop,and the intelligentization of perimeter security systems has become the mainstream direction of development.Traditional perimeter security systems have a single function,usually only with intrusion detection functions,and lack intrusion positioning,trajectory prediction and big data analysis functions.The intelligent perimeter security system will be able to achieve richer functions,especially the positioning function for intrusion targets,which can greatly facilitate the management and control of key areas of perimeter security by duty personnel.This paper uses the idea of swarm intelligence algorithm and integrates the swarm intelligence cat swarm algorithm to design and implement a group-based vibrationsensing intelligent perimeter security system,which intends to solve the shortcomings of various existing security systems for intrusion target tracking and positioning.This article mainly analyzes the requirements of the perimeter security system in terms of deployment flexibility,intrusion target detection,identification,positioning,and later maintenance and upgrades.From the system hardware design,system communication mode design,system software function design,and group intelligence optimization positioning method four aspects describe the realization of the overall group intelligent perimeter security system.The specific content is as follows:1.System hardware design: select a moving coil vibration sensor to collect the vibration signal generated by the intrusion target on the shallow surface;design and realize the signal conditioning circuit to amplify and filter the vibration signal generated by the intrusion target on the shallow surface to increase the signal-to-noise ratio,Improve the detection distance of the system;in order to achieve the intrusion target positioning function and ensure the signal time reference is consistent,the synchronous acquisition analog-to-digital conversion chip AD7779 is used to collect the intrusion target vibration signal;the STM32L496RGT3 single-chip microcomputer is selected to form the main control circuit to realize the functional control of the overall system.2.System communication mode: Aiming at different application scenarios of the perimeter security system,this paper designs and implements two modes of system Lo Ra wireless communication and Beidou short message wireless communication to meet the needs of different application environments.3.System software function: complete the detection of intrusion behavior through the detection and recognition model carried by the MCU,and identify the type of intrusion target(person,car);use the TDOA positioning method based on the signal arrival time difference to realize the initial positioning of the intrusion target,and from the perspective of productization,the design implements auxiliary functions such as system factory configuration,model parameter change query,etc.,which is convenient for system upgrade and maintenance in the later period.4.The positioning method of swarm intelligence optimization: In order to improve the overall positioning accuracy of the system,this paper combines the idea of swarm intelligence algorithm and adopts the cat swarm algorithm to optimize the preliminary positioning results of TDOA and improve the positioning accuracy.Aiming at the target location method optimized by the cat swarm algorithm,design experiments to verify the function of the algorithm.When the seismic wave propagation velocity error on the shallow surface is less than 10% of its own,the average location error is less than 1m.According to the actual collected vibration data,the positioning error of the verification algorithm is 0.5m higher than the preliminary positioning result of TDOA.