辅助定位信标节点的移动路径规划算法研究
Study on Movement Path Planning Algorithm of Auxiliary Locating Beacon Node
作者：陈友荣(浙江树人大学 信息科技学院, 浙江 杭州 310015;常州大学 信息科学与工程学院, 江苏 常州 213164)；万锦昊(常州大学 信息科学与工程学院, 江苏 常州 213164)；苏子漪(浙江树人大学 信息科技学院, 浙江 杭州 310015)；张瑞(浙江树人大学 信息科技学院, 浙江 杭州 310015)
Author：CHEN Yourong(College of Info. Sci. and Technol., Zhejiang Shuren Univ., Hangzhou 310015, China;School of Info. Sci. & Eng., Changzhou Univ., Changzhou 213164, China)；WAN Jinhao(School of Info. Sci. & Eng., Changzhou Univ., Changzhou 213164, China)；SU Ziyi(College of Info. Sci. and Technol., Zhejiang Shuren Univ., Hangzhou 310015, China)；Zhang Rui(College of Info. Sci. and Technol., Zhejiang Shuren Univ., Hangzhou 310015, China)
收稿日期：2016-10-27 年卷（期）页码：2017,49(2):160-168
期刊名称：工程科学与技术
Journal Name：Advanced Engineering Sciences
关键字：无线传感网;传感节点;位置;路径规划
Key words：wireless sensor networks;sensor nodes;localization;path planning
基金项目：国家自然科学基金资助项目（61501403）；浙江省自然科学基金资助项目（LY15F030004）；浙江省公益性技术应用研究计划项目资助（2016C33038）
中文摘要
为快速实现监控区域内所有传感节点的定位，利用辅助定位信标节点的移动，提出无线传感网中辅助定位信标节点的移动路径规划算法（MPPA）。在MPPA算法中，考虑由多个六边形网格组成的监控区域，分析sink节点的移动特点，考虑其移动路径中停留位置只是六边形网格的顶点和中心，不在同一位置停留，相邻3个停留位置不共线以及每一个网格至少被3个以上不同停留位置覆盖等约束条件，提出信标节点的移动路径约束和传感节点定位约束，并建立其移动路径规划模型。根据邻居停留位置的信息素浓度决定下一个停留位置，根据蚂蚁选择的路径释放和挥发信息素。经过蚁群算法的多次迭代，可获知能覆盖所有网格的信标节点最优移动路径。信标节点沿着该路径移动时，传感节点可获知信标节点的不同位置信息，收集通信时的RSSI值，采用Kalman滤波算法降低通信噪声，采用最大似然估计算法计算自身位置坐标。仿真结果表明：MPPA算法可根据网格中心和顶点的位置，收敛于移动距离最短且能实现监控区域任何位置上传感节点定位的最优移动路径。MPPA算法降低了信标节点的移动路径长度和停留位置个数，降低了网络启动后所有传感节点获知自身位置所需要的时间，并将传感节点平均定位误差保持在较低的水平。在一定的条件下，MPPA算法比SCAN、DOUBLE_SCAN、HILBERT、CIRCLES和ZSCAN算法更优。
英文摘要
To quickly realize the localization of all sensor nodes in the monitoring area,a movement path planning algorithm of auxiliary locating beacon node in wireless sensor networks (MPPA) was proposed by using movement of auxiliary locating beacon node.In MPPA algorithm,the monitoring area consisting of multiple hexagonal grids was considered.Analyzing mobile characteristic of sink node,four constraint conditions were assumed:1) Sojourn locations of sink node only were centers and vertexes of hexagonal grids;2) The sojourn location was not stayed twice by sink node;3) Three adjacent sojourn locations were non-collinear;4) Each grid was covered by at least three different sojourn locations.Movement path constraints of beacon node and localization constraint of sensor nodes were then proposed.Movement path planning model was flowingly established.Next sojourn location was judged according to the pheromone concentration of neighbor sojourn locations and the pheromone was released and volatilized according to the selected path of ant.After several iterations of ant colony algorithm,the optimal path of beacon node covering all grids was obtained.When beacon node moved along the path,in each sensor node different sojourn location information of beacon node was obtained and RSSI value of communication was collected.Kalman filter algorithm was used to reduce communication noise and maximum likelihood estimation algorithm was used to calculate its own location coordinates.Simulation results showed that according to the location information of centers and vertexes of grids,MPPA algorithm converged to optimal movement path with shortest movement distance and realized the localization of sensor nodes at any location in the monitoring area.MPPA algorithm reduced the movement path length of beacon node and the number of sojourn locations.Meanwhile,it reduced the time that sensor nodes need to know their own locations, and maintained localization error at a low level.Under certain conditions,MPPA algorithm outperforms SCAN,DOUBLE_SCAN,HILBERT,CIRCLES and ZSCAN algorithms.
【关闭】
