Research On Multi-Strategy Routing Algorithm Based On Community In Opportunistic Networks

Opportunistic network originated from the early delay tolerant network,which is different from the traditional TCP/IP based network.It does not require the complete link between the start node and the destination node,which is more suitable for the reality than the traditional network.Most of the time,a complete end-to-end path cannot be established between the source node and the target node due to the reasons of node movement,signal attenuation,and imperfect infrastructure.Therefore,opportunity network has a wide range of application prospects and use value in various fields through the "store carry forward" communication mode.The mobility of nodes in opportunistic networks makes the communication link between nodes unstable,so data forwarding is an important research content in opportunistic net-works.This thesis aims at the application scenario of portable intelligent devices carried by people,and studies the impact of human social attributes on routing transmission.According to the research results of sociology and psychology,we can see that people have clustering,which can make people form a community and meet more easily in the same community node.This characteristic indicates that the movement path of the node is not irregular' so the reasonable use of community has a positive effect on the research of opportunity network.The research content of this thesis is as follows:1.The social attributes of the nodes make the movement of the nodes follow certain rules,and reasonable use of the social attributes between nodes can make more effec-tive routing decisions.Therefore,this thesis divides nodes into different communities based on their social attributes,and uses routing transmission strategies according to different situations.2.When nodes are in the same community,they have a higher deliver ratio in transmitting messages to each other.Based on the above principle,if the node carrying the message is in the same community as the destination node,the message will be transmitted within the community,and the intermediate node will be selected through the social prediction value to reduce the geographical scope of message forwarding.3.The active frequency of nodes is defined based on social utility.If the node carrying the message is in the different community as the destination node,the node with higher social degree can be selected as the intermediate node to deliver the message to the node that may meet the destination node as much as possible.4.By using ONE simulation platform the MSRC algorithm proposed in this thesis is simulated and compared with the classical routing algorithm,Epidemic algorithm and Prophet algorithm.From the experimental results,we can see that the overall perfor-mance of MSRC algorithm is better in delivery rate,transmission delay and network overhead.