基于猫群算法的震动感知周界安防系统  

Seismic sensing perimeter security system based on cat swarm algorithm

作　　者：周求湛[1] 冀泽宇 王聪 胡继康[1] 李明明[2] 陈禹竺 周险峰 刘萍萍[5] ZHOU Qiu-zhan;JI Ze-yu;WANG Cong;HU Ji-kang;LI Ming-ming;CHEN Yu-zhu;ZHOU Xian-feng;LIU Ping-ping(College of Communication Engineering,Jilin University,Changchun 130012,China;Big Data&Management Center,Jilin University,Changchun 130012,China;Key Laboratory of Geoexploration Instrumentation of Ministry of Education,Jilin University,Changchun 130061,China;College of Instrumentation&Electrical Engineering,Jilin University,Changchun 130061,China;College of Computer Science and Technology,Jilin University,Changchun 130012,China)

机构地区：[1]吉林大学通信工程学院,长春130012 [2]吉林大学大数据与管理中心,长春130012 [3]吉林大学地球信息探测仪器教育部重点实验室,长春130061 [4]吉林大学仪器科学与电气工程学院,长春130061 [5]吉林大学计算机科学与技术学院,长春130012

出　　处：《吉林大学学报（工学版）》2023年第4期1187-1199,共13页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(62071199);吉林省自然科学基金项目(20200201283JC)。

摘　　要：针对传统周界安防系统入侵目标定位方法定位精度较差且常用的基于信号到达时间差(TDOA)定位方法在面对多组传感器情况难以确定最优解的问题,本文设计了基于猫群算法的震动感知智能周界安防系统。首先利用动圈式振动传感器采集入侵目标在浅地表产生的振动信号,并通过振动信号对入侵目标进行TDOA定位。然后通过猫群算法对TDOA初步定位结果进行优化。最后针对系统的不同应用场景,本文设计实现LoRa无线通信和北斗短报文无线通信两种通信模式,解决了系统在偏远地区部署的通信问题。为验证本文算法的有效性,设计了不同算法参数环境下的验证实验以及在实际环境下的测试实验。结果表明:本文算法优化的目标定位方法在浅地表震动波传播速度误差小于自身10%的情况下,平均定位误差小于1 m。算法定位误差相较于TDOA初步定位结果减小约46.9%。Traditional perimeter-security systems either suffer from low positioning accuracy of the target or cannot determine the optimal result generated by multiple groups of sensors in common TDOA.In response to the above issues,this paper designs an intelligent perimeter-security system,by means of seismic sensors and cat-swarm algorithm.First,moving coil seismic sensors are used for collecting the seismic signals generated by targets on shallow surfaces.Then,the collected signals are processed to obtain initial TDOA localization results of the target.Later,these initial results are further optimized via catswarm algorithm.Aiming at different application scenarios of the system,this paper designs two communication modes(i.e.,LoRa and Beidou short message)to ensure system communication in remote regions.In order to verify the effectiveness of the optimization algorithm,several experiments are conducted in real world under different parameter settings.According to the experimental results,when the error in propagation velocity of seismic wave is less 10%,the average localization error of the proposed system(i.e.,optimized by cat-swarm algorithm)is less 1 meter.In particular,compared with the initial TDOA localization results,the localization error is reduced by 46.9%.

关 键 词：通信与信息系统 周界安防 TDOA定位 群体智能 猫群算法 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]