基于极端梯度提升法的漏缆周界入侵定位技术研究

Research on Leaky Coxial Cable Perimeter Intrusion Locating Detection Based on XGBoost Algorithm

基于泄漏电缆的周界入侵检测系统具有安全隐蔽、可随形敷设、全方位警戒、全天候工作等优势,并针对现如今市面上泄漏电缆入侵探测定位系统探测精度低,误报率高等问题,提出了一种基于机器学习的泄露电缆入侵检测定位技术,首先利用泄漏电缆对入侵数据进行采集并处理为多维输入特征量数据,采用极端梯度提升算法(XGBoost)模型对数据进行监督学习得到入侵检测定位模型。并在仿真环境下检测系统的实际性能测试实验。测试结果表明,基于机器学习的入侵检测系统的探测精度为1米,误报率低于1.5%,不需要前期矫正就可以自适应于不同背景环境噪声,具有良好的应用前景。

The perimeter intrusion detection system based on leakage cable has the advantages of security concealment,can be laid along with the shape,omni-directional vigilance,all-weather work and so on,in view of the low detection accuracy and high false alarm rate of leakage cable intrusion detection and positioning system in the market today.This paper proposes a type leakage cable intrusion detection based on machine learning on positioning technology.Firstly,the leakage cable is used to collect and process the intrusion data into multi-dimensional input feature data,and the XGBoost model is used to monitor and learn the data to obtain the intrusion detection and location model.And the actual performance of the system is tested under the simulation environment.The test results show that the detection accuracy of the intrusion detection system based on machine learning is 1 meter,the false alarm rate is less than 1.5%,and it can adapt to the noise of different background environment without prior correction,so it has a good application prospect.