针对事件驱动型传感器网络的多路径编码路由协议

Multipath Coded Routing Protocol for Event-Driven Wireless Sensor Networks

针对事件驱动型传感器网络Burst-Bulk的流量特性,提出了一个基于网络编码和机会转发的多路径单播路由协议——CodePaths.定义了一个考虑时延和拥塞的路由度量,以该度量为基础建立了所有节点到Sink的梯度场.当网络中有事件发生时,节点对数据包进行Intra-flow随机线性编码和再编码,并依据路由梯度和链路质量为多个下游节点分配"转发机会",从而形成从源节点到Sink的无状态交叉多径Mesh,编码包最终汇聚于Sink,并解码得到原始数据.实验结果表明,CodePaths能够适应事件的时空随机性,具有较高的吞吐率和良好的实时性,包传输成功率在95%以上,并在保证可靠性条件下最小化数据传输量,提高能量效率.

An intra-flow network coding based opportunistic multipath routing protocol called CodePaths is proposed referring to the Burst-Bulk characteristic of traffic in event-driven wireless sensor networks.In CodePaths,a new routing metric,taking delay and congestion factors into account,is defined and implemented for establishing routing gradient field from nodes to sink.When an event occurs in the network area,nodes employ random linear network coding and recoding to mix packets,and assign ＇forwarding opportunities＇ to downstream nodes based on the routing gradients and link qualities.Consequently,an interleaving-paths mesh is constructed,along which the encoded packets are routed and finally decoded by the sink node.The extensive simulation results indicate that CodePaths can well adapt to the randomness of events.Compared with other single and multi-path routing,CodePaths with higher throughput and lower latency minimizes the data transmission and energy consumption while keeps the packet delivery rate over 95%.