时间同步是无线传感器网络的一项关键技术.针对目前时间同步算法能耗较大等问题,通过单向同步技术建立全网同步数学模型,提出一种基于最大互邻集合的同步算法.在层次发现阶段生成同层节点的最大互邻集合,利用有限的消息交互分布式地保留尽量少的广播节点,并加入子节点注册、低层节点监听和时序控制等策略提高算法效率.在NS2软件平台进行了仿真,并与相关文献算法对比,结果表明所提出的算法在达到相同同步精度前提下,可显著降低同步阶段的消息开销,提高成功同步节点比例.
Time Synchronization is a critical issue in Wireless Sensor Networks. The energy consumption problem that exists in current synchronization algorithms is studied. A network-wide mathematical model is established with one-way synchronization technique, and a synchronization algorithm based on max ad- jacent set is proposed. During the level discovery phase, the max adjacent set of the same level nodes is generated. Nodes communicate with each other through few messages to reserve as less broadcast nodes as possible. Furthermore, some strategies such as child node registering, lower nodes overhearing and time sequence control are used to improve the efficiency of the algorithm. The simulation on NS2 platform and comparison to related work confirm that the algorithm can significantly reduce the message consumption during synchronization phase, and improve the ratio of nodes which achieve synchronization successfully.