基于平均传递概率的容迟网络路由算法的设计

为了提高容迟网络的传递率、降低传输延迟、对节点缓存进行更有效的管理, 结合已有的PROPHET和Spray and Wait算法, 提出了一种基于平均传递概率的容迟网络路由算法RAB-ADP。在该算法中设置了一个与时间有关的平均传递预测概率参数进行消息转发的决策, 解决了PROPHET算法容易产生路由抖动的缺点。算法综合利用了复制和知识两个属性, 采用{MOPR; FIFO}队列策略组, 通过消息传送完毕的ACK确认信息进行缓存管理和网络中冗余消息副本的删除。仿真实验表明, 该算法在节点缓存大小不同以及网络中节点数目不同的两种情况下, 传递率和路由开销比率的性能均优于其他经典路由算法。

Design of DTN routing algorithm based average delivery probability

In order to improve the delivery ratio, reduce the average latency, manage the buffer effectively in delay tolerant network, this paper proposed RAB-ADP, a routing algorithm based on the average delivery probability, by combining the existing PROPHET and Spray and Wait algorithm. This algorithm set a time-related average delivery prediction probability parameters to make the message forwarding decisions, solved the problem of route flap in PROPHET. RAB-ADP comprehensively used two properties of replication and knowledge, used {MOPR; FIFO} queuing policy group, managed the buffer as well as delete the redundant copies of the message in the network by generating ACK confirmation message when the message was delivered to the destination. Simulation results show that the algorithm has a better performance than other classic algorithms in delivery ratio and overhead ratio in the case of different buffer size and the number of nodes.