PIR传感器网络中行为轨迹重构的几何代数方法

doi:10.3969/j.issn.0372-2112.2017.05.011

PDF(2974 KB)

电子学报 ›› 2017, Vol. 45 ›› Issue (5) : 1096-1102. DOI: 10.3969/j.issn.0372-2112.2017.05.011

学术论文

PIR传感器网络中行为轨迹重构的几何代数方法

王健健¹, 冯琳耀^1,2, 朱帅¹, 罗文¹, 俞肇元^1,3,4, 袁林旺^1,3,4

作者信息 +

Geometric Algebraic Method for Behavior Trajectory Reconstruction in PIR Sensor Networks

WANG Jian-jian¹, FENG Lin-yao^1,2, ZHU Shuai¹, LUO Wen¹, YU Zhao-yuan^1,3,4, YUAN Lin-wang^1,3,4

Author information +

文章历史 +

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

Localization and reconstruction of the human motion trajectory based on high density and low cost PIR sensor is an important direction for the behavior analysis of people.The walking of persons can be detected through the 0-1 response of the PIR sensors.The identification and tracking of the behavior of the crowd can be achieved through the different sensor response sequence of spatio-temporal sequence relationship.Most of the existing analysis methods for PIR sensor network have revealed the statistical characteristics of crowd movement from the perspective of group behavior,but there still exists many problems in the estimation accuracy and uncertainty of behavior.Arming at the reconstruction of personal behavior trajectory in PIR sensor network,the geometric algebra is introduced to achieve the united expression of topological network of PIR sensor,spatial and temporal response of data sequence and the behavior of the person,thus judging the network connectivity according to the response of spatial and temporal relations of the sensors.Based on the geometric algebra operators,the extension and expansion rules of the trajectory are constructed,and behavioral trajectory reconstruction algorithm is constructed with the PIR sensor network response sequence data.Finally,the rationality and effectiveness of the method are demonstrated by using the continuous observation data of MERL laboratory.

导出引用

王健健, 冯琳耀, 朱帅, 罗文, 俞肇元, 袁林旺. PIR传感器网络中行为轨迹重构的几何代数方法[J]. 电子学报, 2017, 45(5): 1096-1102. https://doi.org/10.3969/j.issn.0372-2112.2017.05.011

WANG Jian-jian, FENG Lin-yao, ZHU Shuai, LUO Wen, YU Zhao-yuan, YUAN Lin-wang. Geometric Algebraic Method for Behavior Trajectory Reconstruction in PIR Sensor Networks[J]. Acta Electronica Sinica, 2017, 45(5): 1096-1102. https://doi.org/10.3969/j.issn.0372-2112.2017.05.011

中图分类号： TP208

参考文献

[1] 王凡,周怀北.基于信号衰减的蜂窝移动定位技术[J].计算机工程与应用,2006,42(9):129-131. Wang Fan,Zhou Huaibei.Mobile location technique based on signal attenuation[J].Computer Engineering and Applications,2006,42(9):129-131.(in Chinese)
[2] 赵军,李鸿斌,王智.无线网络室内定位系统研究[J].信息与控制,2008,37(4):465-470. Zhao Jun,Li Hong-bin,Wang Zhi.Based on single neuron adaptive PI controller[J].Information and Control,2008,37(4):465-470.(in Chinese)
[3] Ladd A M,Bekris K E,Rudys A,et al.Robotics-based location sensing using wireless ethernet[J].Wireless Networks,2002,11(1-2):189-204.
[4] Bahl P,Padmanabhan V N.RADAR:An in-building RF-based user location and tracking system[A].Proceedings of IEEE Infocom[C].Tel Aviv:2000.2:775-784.
[5] 俞波,陈一民.基于立体视觉的三维视频轨迹跟踪[J].计算机应用,2003,23(4):72-74. Yu Bo,Chen Yi-min.Stereo vision-based 3D visual tracking[J].Journal of Computer Applications,2003,23(4):72-74.(in Chinese)
[6] 常发亮,马丽,乔谊正.视频序列中面向人的多目标跟踪算法[J].控制与决策,2007,22(4):418-422. Chang Fa-liang,Ma Li,Qiao Yi-zheng.Human oriented multi-target tracking algorithm in video sequence[J].Journal of Control and Decision,2007,22(4):418-422.(in Chinese)
[7] 程光,罗予频,王宏宝.联合轮廓法在低分辨率视频下的多目标追踪[J].计算机工程与应用,2007,43(2):64-66. Cheng Guang,Luo Yu-pin,Wang Hong-bao.Multiple objects tracking in low resolution based on joint contour method[J].Computer Engineering and Applications,2007,43(2):64-66.(in Chinese)
[8] Niebles J C,Wang H,Li F F.Unsupervised learning of human action categories using spatial-temporal words[J].International Journal of Computer Vision,2008,79(3):299-318.
[9] Kasteren T L M,Englebienne G,Krøse B J A.Transferring knowledge of activity recognition across sensor networks[A].Proceedings of the 8th international conference on Pervasive Computing[C].Berlin:Springer-Verlag,2010.283-300.
[10] 袁林旺,闾国年,罗文,等.GIS多维统一计算的几何代数方法[J].科学通报,2012,57(4):282-290. Yuan L W,Lü G N,Luo W,et al.Geometric algebra method for multidimensionally-unified GIS computation[J].Chin Sci Bull,2012,57(4):282-290.(in Chinese)
[11] 袁林旺,俞肇元,罗文,等.基于共形几何代数的GIS三维空间数据模型[J].中国科学:地球科学,2010(12):1740-1751. Yuan L W,Yu Z Y,Luo W,et al.A 3D GIS spatial data model based on conformal geometric algebra[J].Sci China:Earth Sci,2010(12):1740-1751.(in Chinese)
[12] Yuan L,Yu Z,Luo W,et al.Clifford algebra method for network expression,computation,and algorithm construction[J].Mathematical Methods in the Applied Sciences,2014,37(10):1428-1435.