Research On Opportunity Network Routing Algorithms And Cache Space Management Strategy

With the rapid development of wireless network technology,opportunity network has become a research hotspot in this field.Opportunity network communication is different from traditional network.Opportunity network communication uses storage-carry-forward mode to transfer data packets.In this mode,communication path selection is very important,reasonable use of buffer space to ensure efficient transmission of information is equally important.Therefore,routing mechanism and cache management strategy have always been the focus of opportunity network research.After studying the sociality of opportunity network,this paper puts forward the routing algorithm of opportunity networks based on interest partitioning and the cache management strategy based on community routing algorithm,mainly from the following aspects to study:1.Opportunity network routing algorithm based on community of interest partition.In view of the problem that the current opportunity network routing algorithm can not accurately estimate the interest of nodes under the condition of fewer data packets,which leads to unreasonable community partition and invalid transmission of data packets among nodes,thus increasing the communication overhead,this paper proposes a method to quantify the interest of various messages by combining the historical situation of receiving messages and the similarity between historical messages and various messages.According to this degree of interest,the interest level community routing algorithm ILCR(interest level community route)is divided.The specific forwarding strategy of ILCR is to select the nodes with high probability of arriving at the destination node in the target community or-with high activity and reliability as the relay.Through the simulation of ILCR on ONE platform and comparing with Epidemic and Prophet,the results show that ILCR has a 13%higher delivery rate than Prophet,a 113%higher delivery rate than Epidemic,a 94.4%lower network overhead than Prophet,and an 81%lower cost than Epidemic,which ensures the possibility of successful communication under frequent network discontinuity and lack of network resources.2.Research on cache management strategy based on community routing algorithm.Du(Drop Unimportant)strategy is proposed to solve the problem of low network performance caused by message replica size and lifecycle in cache management.Du strategy quantifies the current node's social relationship strength mainly according to the social connection degree of the node in which the message is located in the community and the regional connection strength in the community.The importance of each message is calculated according to the different weights of the social relationship of the message,the number of copies of the message and the remaining survival time to the importance of the message.When the node cache congestion occurs,priority is given to deleting messages of relatively low importance.Through the simulation of DU strategy on ONE platform and comparing with classical caching strategies such as FIFO and DO,the results show that the delivery rate of DU strategy is about 5%to 17%higher than that of FIFO strategy,the average cost is about 9%-14%lower,and the delay is also slightly lower.DU-based ILCR routing is 10%higher than ILCR,16%higher than Epidemic and 55%higher than Direct Delivery.Network overhead is only about 0.08 times higher than Epidemic network overhead.The average delay is about 4%lower than ILCR and about 20%lower than Direct Delivery.