PIR传感器网络中行为轨迹重构的几何代数方法

发布时间：2018-05-14 06:05

  本文选题：行为轨迹 + 传感器网络　； 参考：《电子学报》2017年05期

【摘要】：基于高密度、低成本的PIR(Passive Infrared Detector)被动式红外探测器传感器进行人群行为轨迹的定位与重建是人群行为分析的重要方向.PIR传感器通过0-1响应确定附近是否有行人通过,并通过不同传感器响应序列的时空序列关系实现对人群行为轨迹的识别与跟踪.已有PIR传感器网络分析方法多从群体行为视角揭示人群运动的统计特征,在行为估算精度、不确定性等方面仍存在诸多问题.本文针对PIR传感器网络中个人行为轨迹的重构问题,引入几何代数工具,实现PIR传感器网络拓扑、时空响应数据序列和行为轨迹的代数化统一表达,进而以传感器响应的时空关系为基础进行网络连通性的判定.基于几何代数算子构建轨迹路径的延拓与扩展规则,并集成传感器对轨迹的时空响应关系构建基于PIR传感器网络响应序列数据的行为轨迹重构算法.最后以MERL实验室连续观测数据为案例,论证了方法的合理性与有效性.
[Abstract]:Location and reconstruction of crowd behavior trajectory based on high-density, low-cost PIR(Passive Infrared detector passive infrared sensor is an important direction of crowd behavior analysis. PIR sensor determines whether there is a pedestrian passing by 0-1 response. The identification and tracking of human behavior locus are realized by the temporal and spatial sequence relationship of different sensor response sequences. The existing PIR sensor network analysis methods often reveal the statistical characteristics of crowd movement from the perspective of group behavior. There are still many problems in the estimation accuracy and uncertainty of behavior. In this paper, aiming at the reconstruction of personal behavior trajectory in PIR sensor networks, a geometric algebraic tool is introduced to realize the algebraic unified representation of topology, spatiotemporal response data series and behavior trajectories of PIR sensor networks. Furthermore, the network connectivity is determined based on the temporal and spatial relationship of sensor response. The extension and extension rules of trajectory path are constructed based on geometric algebra operator, and the behavior trajectory reconstruction algorithm based on response sequence data of PIR sensor network is constructed by integrating the space-time response of sensor to trajectory. Finally, the rationality and validity of the method are demonstrated by taking the continuous observation data of MERL laboratory as an example.
【作者单位】： 南京师范大学虚拟地理环境教育部重点实验室;重庆知行地理信息咨询服务有限公司;江苏省地理环境演化国家重点实验室培育建设点;江苏省地理信息资源开发与利用协同创新中心;
【基金】：国家自然科学基金(No.41571379,No.41231173)
【分类号】：TP212
，

本文编号：1886644