无线传感器网络层次参考时间同步算法的研究.pdfVIP

Computer Engineering and Applications计算机工程与应用2013，49（21） 1 引言 时间同步技术是无线传感器网络（Wireless Sensor Net- work，WSN）的一项关键支撑技术，无线传感器网络作为分 布式系统，在完成MAC链路调度、协同休眠、定位、数据融 合等任务时，都需要网络中的节点时钟同步，节点间准确保 持时间同步是WSN实现各种技术的基础和重要保证[1]。由 于无线传感器网络的特殊性，一些传统的时间同步技术， 如NTP[2]，GPS[3]和无线测距技术等均不再适用于现在的无 线传感器网络。目前WSN中经典的时间同步算法主要可 分为三类：一是基于发送者的同步机制，如 DMTS[4]；二是 基于发送者-接收者交互的同步机制，如 TPSN[5]；三是基于 接收者-接收者的时间同步机制，如 RBS[6]。DMTS算法简 单，通信开销小，但精度不高；TPSN算法采用双向通信，可 以把部分延迟的影响降低一半，但没有对时钟频差进行估 计，需频繁同步，开销较大；RBS算法大大降低了传播延迟 对精度的影响，并对时钟偏移做了补偿，精度比DMTS算 法高，但RBS算法报文交互频繁，能量消耗大，且采用线性 拟合，复杂度高。HRTS（Hierarchy Referencing Time Syn- 无线传感器网络层次参考时间同步算法的研究 王玉秀，黄 剑，石 欣 WANG Yuxiu, HUANG Jian, SHI Xin 重庆大学 自动化学院，重庆 400030 School of Automation, Chongqing University, Chongqing 400030, China WANG Yuxiu, HUANG Jian, SHI Xin. Research on hierarchy referencing time synchronization for Wireless Sensor Net- works. Computer Engineering and Applications, 2013, 49（21）：70-74. Abstract：The applications of Wireless Sensor Network need clock of sensors nodes to be synchronized. Due to the impact of node clock skew and the delay in information transmission, the receipt time of commonly used classic time synchronization algo- rithm is inaccurate, which will reduce the time synchronization accuracy to a certain extent. In order to improve time synchroni- zation precision and balance node energy consumption of the network, this paper proposes an improved hierarchy referencing time synchronization algorithm, which is based on the time characteristics of each node in the hierarchy having the unique physical location. This paper uses Bayesian estimation methods to compute the receipt time of each node, narrows the error scope of time deviation, and finally obtains more precise synchronization time offset. It adopts wireless channel broadcast nature and bi-directional synchronization mechanisms, minimizes communication load and balances node energy consumption. The simulation results show that the use of Bayesian estimation method can significantly improve t