传感器网络邻居发现协议综述

Survey on Neighbor Discovery Protocol for Sensor Networks

邻居发现协议是传感器网络协议族的重要组成部分,用于快速高效地发现通信范围内的邻居节点,为其他协议和应用提供邻居信息.文中根据近年来的相关研究成果,介绍传感器网络邻居发现的问题定义,提出了形式化的问题模型以及简化的时隙模型,并指出了协议设计面临的挑战:(1)在电池供电的前提下保证能量有效性;(2)保证实时性以满足应用的延迟约束;(3)适应网络拓扑的动态性,尤其是近年来基于位置移动应用的出现对邻居发现协议的性能提出了更高的要求.文中根据各类邻居发现协议的对称性、确定性、协作性、是否使用多信道和按需发现等特点分类,选定了发现延迟、能耗和最优近似比作为协议的性能评价标准,并针对各类具有代表性邻居发现协议分析研究其主要设计思想,剖析了相似协议间存在的关联性和演进过程,论述各自的优缺点及适用范围,归纳协议设计方法并通过理论分析比较协议性能.最后,通过真实环境中的实验比较了各类协议的实际性能,并总结和展望了邻居发现技术的未来发展趋势.

Neighbor discovery protocol is an important component in wireless sensor network protocols,which focuses on finding out neighbors within node’s communication range fast and efficiently,and provides neighbor information for other protocols and applications.With surveying neighbor discovery protocols developed in recent years,we introduce the definition of neighbor discovery in sensor networks and present the formal model of the problem as well as a simplified slot model.The design of neighbor discovery protocol is facing three challenges：（1 ）ensuring energy-efficiency in battery-powered nodes;（2）satisfying the low-latency requirement in diverse applications;（3 ）adapting network topology dynamics,especially in emerging location-based mobile applications which make higher demand on the neighbor discovery protocol performance. Neighbor discovery protocols are divided into several categories by symmetry,determinacy, collaboration,whether multi-channel is involved and whether the discovery request is on-demand. Moreover,we propose discovery latency,energy consumption and approximation to optimal as the metrics of protocol performance.We research the primary design principal of representational discovery protocols and analyze their relationship with those in the same protocol family.With discussing their advantages and drawbacks,we list the application scenarios of different protocols. Meanwhile,we summarize the various design principals and compare the protocol performance by theoretical analysis.Finally,we compare their performances with real-world experiments and try to predict the future research trends of the neighbor discovery technology.