Research On Utility Based Probabilistic Routing Algorithm And Cache Management Strategy In Delay-tolerant Networks

With the continuous exploration of scientific research and the change of communication environment,a kind of special network,known as Delay Tolerant Networks(DTN),has emerged,which has the characteristics of intermittent connection,high time Delay and high bit error rate of communication links.The concept of DTN originated from interstellar networks,which are mainly used to deal with extreme network environments such as intermittent connections,dynamic changes in topology and resource scarcity.Different from the traditional Internet,there is usually no stable end-to-end transmission path between sending node and receiving node in the delay-tolerant network,so the existing TCP protocol and congestion management strategy are no longer applicable in the delay-tolerant network.In order to overcome the intermittent connection of communication links in the delay-tolerant network,the"storage-carry-forward" mode is adopted for the transmission of messages between nodes.In recent years,the research on delay-tolerant networks mainly focuses on network routing protocols and other related fields.How to select efficient routing protocols is a major research topic in the field of wireless networks.Compared with the traditional network,the transmission capacity and storage capacity of the nodes in the delay tolerant network are particularly inadequate.Therefore,it is particularly important to choose the appropriate routing algorithm to improve the message delivery rate and the appropriate cache management algorithm to reduce the message redundancy in the network.To this end.this paper mainly focuses on the routing algorithm and cache management strategy of nodes,and the specific work is as follows:(1)Considering the delay of the existing network of intermittent connection,high delay and low Delivery rate,etc,in classical probability Routing algorithm on the basis of a large number of experimental study of the prophet,and in view of the prophet algorithm under a single path of low Delivery rate and other shortcomings,this paper proposes a Probabilistic Routing Based on delivering Utility(Delivery Utility-based Probabilistic Routing,DUBPR)algorithm,which considers the node connection success rate and the effective connection time efficiency,it introduces the concept of delivering Utility,The update and decay of the delivery utility value are also considered.When two nodes meet,the delivery utility values of each node are updated and compared,and the message is delivered to the node with the higher delivery utility value first.The experimental results show that,based on the Prophet algorithm,DubPR improves the message delivery rate,reduces the load rate and reduces the consumption of resources.(2)Considering the delay in the network node Cache space is limited,when there was a Message redundancy in the network,can lead to network congestion,so this paper proposes a Cache Management strategy Based on Message discarded Utility(Message this Utility-based Cache Management,MDUBCM),namely,considering the forwarding of a Message number,remaining survival time and receiving time,put forward the Message to abandon the concept of Utility.In the event of network congestion,messages with large discarded utility values are preferred to be deleted.During forwarding,messages with smaller utility values are first forwarded and discarded,thus improving the utilization of cache resources.Experimental results show that compared with traditional cache management strategies DO and DF,MDUBCM has a better performance in improving delivery rate and reducing network overhead.