基于时间序列启发式信息的室内轨迹跟踪算法

Indoor Trajectory Tracking Algorithm Based on Time Series Heuristic Information

现有的无线传感器网络室内轨迹跟踪算法是通过定位形成轨迹的,没有利用一定空间范围内相邻信标节点RSSI定位信息在一段时间内的启发式信息.提出了基于RSSI时间序列启发式信息的轨迹跟踪算法,该算法构建基于定位信息时空关联特性的轨迹跟踪模型,对定位信息进行一维重构边界时间序列、二维重构区域统计量、移动最小二乘法检测分别得到动态时间窗口及与之匹配的区域信息及边界信息,在此基础上完成受启发式信息约束的动态时间弯曲轨迹跟踪,并对时空关联模型轨迹跟踪算法中定位信息融合处理的原理进行了严谨的数学论证.通过现场实验与仿真实验表明:该算法轨迹光滑、误差不累积、环境适应性好,相比现有方法基于启发式信息有效克服噪声的影响、减小搜索范围,提高轨迹跟踪的准确性.

Existing indoor trajectory tracking algorithms on wireless sensor network are based on continuous localization and can not make use of the heuristic information of RSSI time series within a certain temporal and spatial range.The heuristic information of RSSI time series is a key factor of trajectory tracking procedure.This paper proposes a new trajectory tracking algorithm on spatiotemporal correlation model based on heuristic information.According to the heuristic information related to moving trajectory,the new method contains the following essential phases.Firstly,we model the trajectory tracking model reflecting spatiotemporal correlation and statistical characteristics.Secondly,we detect spanning boundary event and judge which subarea the unknown node was in by means of information fusion of RSSI time series and moving least square method.Finally,the moving trajectory of unknown node is formed by means of dynamic time warping fingerprinting matching algorithm with heuristic information constraints.The principles of information fusion are strictly proved in mathematics.The field experiments and the simulation experiments show that the algorithm has good environment adaptability,smooth trajectory and the error does not accumulate among the subareas.Compared with the existing methods,the accuracy of trajectory tracked is improved.