基于UKF区域交叉定位的无线传感器网络sink节点动态跟踪算法  

Dynamic wireless sensor network sink node tracking algorithm based on UKF and regional cross location

作　　者：殷荣网[1] 李赵鑫 邵安贤[1] 庞京玉[1] 

机构地区：[1]合肥学院基础教学与实验中心,合肥230601 [2]北京大学信息科学技术学院,北京100871

出　　处：《计算机应用研究》2015年第9期2729-2732,2736,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(61170263;60972036)

摘　　要：在传感器网络中sink点(或网关)是收集信息的节点,但其生存周期受能量限制,为了提高传感器网络的使用寿命,对传感器网络的目标跟踪方式进行研究,提出基于无迹Kalman滤波(UKF)的传感器网络sink节点动态跟踪算法,以实现高效节能的资源管理和利用方式。该算法主要包括位置预测和目标定位两个步骤:首先利用UKF算法对目标节点的下一位置进行预测,通过选择各节点的开启/睡眠状态使sink节点靠近预测位置;然后通过四圆区域定位交叉定位算法对sink节点的位置区域进行局部准确定位。实验结果表明,这种动态的sink节点预测定位算法能够有效缩短数据发射传感器与sink点之间的距离,减少跳数,从而实现负载均衡降低能耗的效果。Sink（ or gateway） in the sensor network point is node collects informations but its lifetime is constrainted by ener- gy, in order to improve the sensor network lifetime, research on object tracking in sensor networks, and proposed an unscented Kalmau filter （UKF） based sensor network sink node dynamic tracking algorithm in order to realize resource management effi- ciency. The algorithm mainly included the location prediction and target of two steps, firstly, it used the UKF algorithm of sink node a position prediction, by selecting nodes on/sleep makes sink nodes gradually close to the predicted position, and then used the four round of regional positioning algorithm of cross location location area of sink node locally accurate positioning. The experimental resuhs show that the dynamic prediction of sink node localization algorithm can effectively shorten the data transmission between sensor and the sink distance of a point to reduce the number of hops, achieving load balancing energy consumption reducing effect.

关 键 词：无迹卡尔曼滤波 区域交叉定位 SINK节点 动态跟踪 生存周期 

分 类 号：TP393.07[自动化与计算机技术—计算机应用技术]