Research And Design Of Opportunistic Network Cache Strategy Based On Multi-dimensional Attributes

In order to ensure that the data has a higher delivery rate in this special network environment,the opportunistic network generally adopts a multi-copy routing algorithm and a transmission mode that stores and then forwards.Due to the limited network resources,this special transmission mode and a large number of replicas' competition for node resources will degrade the network performance.The role of the cache management strategy is to effectively handle these redundant copies,thereby maximizing the utilization of node resources.In the existing caching strategy,some mechanisms are based on the single attribute of the message(such as the size of the message or a copy of the message)as the standard directly;some are based on the degree of association between the message and the node or The node's own mobility characteristics are used as a reference;some are based on the specific route as the premise of the proposed management strategy.These strategies have a certain effect on the improvement of network performance,but they have their own limitations,besides,they are not universal in application scope.This paper studies these caching strategies and analyzes the existing problems to design new caching management strategies.The specific research results are as follows:(1)In order to reduce the number of invalid forwarding times of nodes and thus improve the transmission performance of the network,this paper proposes a priority scheduling strategy based on message multidimensional attributes.Firstly,the strategy calculates the probability of a message encountering the next hop as the destination node,the average transmission delay of the message between nodes,and the copy overhead of the message through the multidimensional attribute model of the message.Then based on the calculated results,it estimates the consumption of network resources in the next scheduling period.Finally,a priority queue is established with the resource consumption value of the message as a parameter.When data is exchanged,the message with the highest priority is selected to be preferentially forwarded.(2)In order to manage the redundant copies in the cache effectively,this paper designs a new cache dropping strategy.The strategy uses the message's forwarding frequency,buffer share and delivery probability to calculate the message's dependence on the node.When the buffer is insufficient,messages with the least dependencies are discarded first.At the same time,messages that cannot be forwarded again with a high probability in the cache and messages that have been successfully delivered to the destinationnode are periodically deleted,thereby releasing the cache resources in time.In summary,this paper has done a lot of analysis and research on the cache management of opportunistic networks,and proposed a new solution to the shortage of existing caching strategies.The simulation results show that the related strategies proposed in this paper can significantly improve the transmission performance of networks in terms of message delivery rate,average transmission delay,and network overhead,etc.