基于融合的无线传感器网络k-集覆盖的分布式算法  

Distributed Set k-Cover Algorithms for Fusion-Based Coverage in Wireless Sensor Networks

作　　者：李劲[1,2] 岳昆[3] 刘惟一[3] 

机构地区：[1]云南大学软件学院,云南昆明650500 [2]云南省软件工程重点实验室,云南昆明650500 [3]云南大学信息学院,云南昆明650091

出　　处：《电子学报》2013年第4期659-665,共7页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61063009;No.61163003);教育部博士点基金新教师类课题(No.20105301120001);教育部科学技术研究重点项目(No.211172);云南省应用基础研究项目(No.2011FB020);云南省教育厅科学研究项目(No.2010Y251;No.2011Z015);云南大学软件学院学科建设基金项目(No.2012SE303)

摘　　要：当节点采用概率感知模型且融合多个节点的数据进行联合感知的情况下,提出了一个新的无线传感器网络的覆盖优化问题:基于融合的k-集覆盖优化问题.首先,将优化问题建模为融合覆盖博弈,证明该博弈是势博弈,且势函数与优化目标函数一致,因此,最优解是一个纯策略Nash均衡解.其次,给出了节点间融合覆盖效用独立的判定条件,进而分别提出同步、异步控制的、基于局部信息的、分布式的覆盖优化算法,证明了算法收敛到纯策略Nash均衡.最后,仿真实验结果表明,当算法收敛时,网络能达到高的覆盖率且具有好的覆盖稳定性.A new coverage optimization problem named disjoint set k-cover for fusion-based coverage of WSN is investigat- ed in this paper where sensor nodes are assumed using a fusion-based collective probabilistic sensor model. First, the problem is for- mulated as a fusion-based coverage game and then the game is proved as a potential game. So that the optimal solution is a pure Nash equilibrium. Second, we present the conditions that determine the independence of coverage utility among sensor nodes. Fur- thermore, two distributed algorithms only based on local information are proposed and proven to be convergent to pure Nash equlib- ria. Finally, experimental results show that Nash equilibria can provide a near-optimal and well-balanced solution to the problem.

关 键 词：无线传感器网络 覆盖优化 融合感知 博弈论 分布式计算 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]