无线传感器网络时钟参数追踪与一致性同步  

Clock parameters tracking and consensus synchronization in wireless sensor network

作　　者：万羊所 蔡春扬 曾培[1] Wan Yangsuo;Cai Chunyang;Zeng Pei(School of Automation,Chongqing University of Posts&Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆邮电大学自动化学院,重庆400065

出　　处：《计算机应用研究》2018年第5期1479-1483,1487,共6页Application Research of Computers

基　　金：国家自然科学基金资助项目(51605065);重庆市教委科学技术研究项目(KJ1600402)

摘　　要：针对无线传感器网络分布式时钟同步问题,在不可靠网络以及随机性干扰的情况下,提出了基于卡尔曼滤波的最大一致性时钟同步算法。在获得硬件时钟参数后,通过设置预定偏斜目标,各节点可不通过网络交换来调整时钟偏斜。为了进一步使节点间时钟偏移达到同步,设计了最大一致性控制方案来补偿节点,并基于图论给出算法收敛性证明。仿真结果表明算法能够快速跟踪硬件时钟参数,较加权最大一致性时钟同步算法收敛速度更快,全局平均同步误差下降了一个数量级。Aiming at the problem of distributed clock synchronization in wireless sensor network(WSN),in the case of unreliable network and random interference,this paper proposed a maximum clock synchronization algorithm based on Kalman filter(KF-MTS)after establishing state space model of clock synchronization.The algorithm set a predetermined skew target after obtaining the hardware clock parameters.Each node could adjust the clock skew without network exchanging.In order to further synchronize the clock offset between the nodes,it designed the maximum consensus control scheme to compensate the nodes,and given the proof of convergence of the algorithm based on graph theory.Simulation results show that the proposed algorithm can quickly track hardware.Compared with the weighted maximum time synchronization(WMTS)protocol,the clock synchronization algorithm converges faster and the global average synchronization error was lower one order than WMTS.

关 键 词：无线传感器网络 分布式时钟同步 卡尔曼滤波 一致性 

分 类 号：TP393.04[自动化与计算机技术—计算机应用技术] TP301.6[自动化与计算机技术—计算机科学与技术]