基于势博弈的分布式目标跟踪传感器分配算法

发布时间：2018-01-05 19:06

本文关键词：基于势博弈的分布式目标跟踪传感器分配算法　出处：《电子与信息学报》2017年11期 　论文类型：期刊论文

【摘要】：针对无线传感器网络多目标跟踪传感器分配问题,考虑传感器能量、带宽、观测距离和通信距离受限,该文提出一种基于势博弈的分布式目标跟踪传感器分配算法。算法基于目标预测位置,以几何精度衰减因子作为优化准则,建立了基于邻居合作促进整体最优化的传感器分配局部信息博弈模型,并证明了该模型是一个精确势博弈模型,存在纯策略纳什均衡。在此基础上,提出了一种并行最佳响应动态的传感器分配方法,从理论上证明了所提算法只需要传感器与一跳邻居进行信息交互,就能收敛到纳什均衡并且拥有比传统最佳响应动态更快的收敛速度。同时借鉴通信网中的载波侦听多路访问机制,设计了一种完全分布式的决策节点推选机制,更加符合无线传感器网络自组织的特性。仿真结果表明所提算法在收敛速度、跟踪精度和能量有效性方面的优势。
[Abstract]:For multi target tracking sensor allocation problem in wireless sensor networks, sensor energy, bandwidth, observation distance and communication distance is limited, this paper proposes a distributed target tracking sensor allocation algorithm based on game theory. The potential target prediction algorithm based on location, with the geometric accuracy of the attenuation factor as the optimization criterion, establish cooperation to promote the local neighbor information game the overall optimization model of sensor based distribution, it is proved that the model is an exact potential game model, has a pure strategy Nash equilibrium. On this basis, put forward a method for sensor allocation and optimal dynamic response, theoretically proves that the proposed algorithm requires only the sensor and the one hop neighbor information exchange can convergence to the Nash equilibrium and has faster convergence speed than traditional optimal dynamic response. And use the communication network in the multi carrier sense Based on the access mechanism, a fully distributed decision node selection mechanism is designed, which is more consistent with the self-organizing characteristics of wireless sensor networks. The simulation results show the advantages of the proposed algorithm in terms of convergence speed, tracking accuracy and energy efficiency.

【作者单位】：解放军信息工程大学信息系统工程学院;
【基金】：国家科技重大专项(2014ZX03006003)~~
【分类号】：TN929.5;TP212.9
【正文快照】： 随着微机电系统(Micro-Electro-MechanicalSystems,MEMS)技术的发展,低能耗、廉价、多功能的传感器节点的出现,无线传感器网络(WirelessSensor Networks,WSN)已经在民用和军用领域发挥着重要作用[1,2]。目标跟踪是无线传感器网络最重要的应用之一,为保证观测区域的充分覆盖、

【相似文献】