GPS异常监测数据的关联负选择分步识别算法

Multi-step identification algorithm based on relational negative selection for GPS abnormal monitoring data

为了有效地判别GPS异常监测数据,建立了GPS监测序列的关联模型;针对关联分析中无法准确判定异常数据发生范围的问题,提出了基于关联负选择的异常监测数据分步识别算法;对于固定数目和固定半径的检测器,通过设定空间覆盖率来自动生成合适数目的可变半径检测器,并采用蒙特卡洛的方法解决了可变半径检测器的空间重叠问题;对于负选择算法中自体空间动态变化的问题,提出了自适应半径的自体表示方法,使得自体半径能够随着自体数据的聚集特征自适应调整,满足了动态监测的需求;采用仿真数据检验了所提方法的可行性和有效性,分析结果表明自适应半径的负选择算法能够以较少的检测器覆盖较大的异体空间,提高了GPS数据的异常检测率,且该算法能够准确给出异常监测数据的发生范围,具有较高的实用价值。

In order to effectively identify the abnormal monitoring data of GPS,the relational model of GPS monitoring time series is established.Then,the multi-step identification algorithm of abnormal GPS monitoring data based on the relational negative selection is proposed to overcome the drawback that the relational analysis cannot accurately determine the extent of the abnormal data.For the fixed-number and constant-sized detectors,an appropriate number of variable-sized detectors can be automatically generated by setting the space coverage rate,and the problem of space overlap is solved by Monte Carlo method.In addition,the self-representing method with self-adaptive size is proposed to solve the dynamic change of self space in negative selection algorithm,allowing the self radius adaptively adjust with the clustering features of self data,which satisfies the requirement of GPS based dynamic monitoring.Finally,the simulation data are used to verify the feasibility and efficiency of the method respectively.The analysis results show that the negative selection algorithm with self-adaptive radius can cover larger non-self space by fewer detectors,which improves the abnormal detection rate of GPS data.Moreover,the proposed algorithm can accurately determine the extent of abnormal data,which has high practical values in civil engineering.