实时响应物联网中基于查询的数据转发方案  

Novel Query-Driven Real-Time Data Forwarding in Internet of Things

作　　者：李英龙[1,2,3] 陈红[1,2] 莫尚丰[1,2,3] 

机构地区：[1]教育部数据工程与知识工程重点实验室(中国人民大学),北京100872 [2]中国人民大学信息学院,北京100872 [3]湖南科技大学数学与计算科学学院,湖南湘潭411201

出　　处：《计算机学报》2012年第3期464-476,共13页Chinese Journal of Computers

基　　金：国家自然科学基金(61070056)、国家自然科学基金重点项目(61033010);中国人民大学科学研究基金(中央高校基本科研业务费专项资金资助)项目成果(12XNH175)资助~~

摘　　要：在许多基于传感器网络技术的物联网应用中,用户需要快速的查询响应,比如智能交通物联网应用中,行驶在路上的司机即时查询附近的空停车位信息.如何为此类物联网设计一种符合传感器网络特性(如能量有效等)的快速数据转发方案是一项重要的挑战性工作.已有的传感器网络实时数据转发协议大都因未解决好转发断路带来的额外开销、孤立节点处理耗时、难以适应网络拓扑动态变化等关键性问题而未取得理想的实时性效果.为此,该文提出一种新的基于查询的快速数据转发方案,利用查询消息为每个传感器节点建立最快速的数据转发路径(有向无环图),此外文中给出的综合路径代价模型可以均衡网络能量和减少网络拥塞延时,最后设计了贪婪的分布式数据转发算法及其改进算法,并用仿真实验验证了该方案的有效性和高效性.In many sensor network technologies involved Internet of Things（IoT）,users need quick query response,e.g.,the drivers driving on the road check the real-time parking information nearby in a IoTs of intelligent traffic system.Designing a fast data forwarding scheme for such kind of IoTs as well as meeting the requirements（e.g.,energy efficiency） of sensor networks is a crucial and challenging issue.Existing real-time routing protocols of sensor networks do not achieve the desired real-time goal because they suffer from the key problems of extra overhead of void forwarding paths（cannot reach the destination,but have to backtrack） and time overhead of processing isolated nodes,as well as difficult to adapt to the network topological changes.Therefore,a new query-driven data forwarding scheme is proposed in this paper,which every node gets its fastest data forwarding paths（directed acyclic graphs） based on the query message flooded by the Sink.Moreover,hybrid link cost model is given trying to achieve local energy balance and congestion avoidance,and finally the corresponding greedy distributed data forwarding algorithm and its improved one are provided.Extensive simulations show the effectiveness and efficiency of the proposed schemes and algorithms.

关 键 词：物联网 实时数据转发 转发断路 孤立节点 优先转发节点 备份转发节点 连通度 无线传感器网络 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]