Research On Probabilistic Routing Algorithm And Message Management Strategy Based On Node Similarity In Opportunistic Networks

An opportunity network is one that does not require a fixed end-to-end path between the source and destination nodes.The traditional network cannot be connected most of the time due to the movement of nodes and the weak signal of the network in remote areas,so the opportunity network is developed.In the opportunistic network,communication is carried out in the mode of "store-carry-forward",which can avoid the shortcoming that messages cannot be delivered in time due to the constant change of routing caused by node movement.Opportunity network has a very broad market prospect and application value in wildlife tracking,vehicle-mounted network and interplanetary network.Due to the flexibility of nodes to move and the limitations of energy and storage space,it is extremely necessary to design efficient routing algorithm.On the basis of analyzing the research achievements of researchers at home and abroad on opportunistic network routing algorithms in recent years,the main research content of this paper is given.Since the communication time between nodes is not stable,the traditional Probabilistic routing Protocol using history of Encounters&transitivity is studied.Probabilistic routing algorithm based on node similarity rate in opportunistic network was proposed.S-Prophet).S-Prophet algorithm,which makes node routing more efficient,higher delivery rate,lower network overhead.The main work is as follows:(1)Based on the node encounter duration,the attenuation formula of the delivery prediction value in the traditional Prophet algorithm is improved,and the node encounter duration is used as an indicator to measure the attenuation of the delivery prediction value.Nodes based on historical information,introduce the concept of the similar ratio,through node similar rate,Prophet of traditional algorithm of computing nodes delivery forecast formula to make improvement,but the message delivery strategy remains the same,still keep the message delivery forecast large relay passed to the node,and then forward the relay node to destination node,An ACK deletion mechanism is added to the algorithm to remove copies of messages that have been successfully delivered.(2)The simulation experiment of the algorithm in this paper was carried out through ONE simulation platform,and S-Prophet algorithm,BA-Prophet algorithm,Epidemic algorithm and PROPHET algorithm were compared from four aspects:node cache,message life cycle,message generation interval and simulation time.Use Python script to read experimental data and draw at one time to achieve the visual effect.The experimental results show that the overall performance of S-PROPHET algorithm is the best in delivery rate,average delay,network overhead and average hop number.In view of the message forwarding strategy,the message forwarding strategy stage and message scheduling stage of S-Prophet algorithm are improved by following Prophet's forwarding rules.Probabilistic routing algorithm message management strategy based on node similarity rate is proposed.MS-Prophet).In the message forwarding stage,the MS-Prophet algorithm adds the average encounter duration of nodes as the judgment basis,and in the message deletion stage,it uses the utility value as the judgment basis,and achieves the purpose of message management by increasing the delivery opportunities and deleting the messages with small utility value.Then,ONE platform was used to compare MS-Prophet algorithm,S-Prophet algorithm,Ba-Prophet algorithm and DARP-DO algorithm from four aspects:node cache,message life cycle,message generation interval and simulation time.Use Python script to read experimental data and draw at one time to achieve the visual effect.The experimental results show that the MS-PROPHET algorithm has the best performance in terms of delivery rate and average delay,and can deliver messages quickly and accurately.