Research On Cooperative Caching Technology Based On Neighbor Discovery

In recent years rapid development of technology introduces more devices to join the wireless network which in turn causes the amount of data exchange to increase exponentially.However,limited bandwidth resource cannot meet the demand of users.In order to promote the user experience and reduce the delivery latency,the cooperative caching strategy is introduced into neighboring devices to reduce the degree of network congestion.However,the premise that devices communicate with each other and share content is that the neighboring devices are known to each other.It means that the neighbor discovery of nodes is necessary.Therefore,this thesis starts from the neighbor discovery and cooperative caching decisions between devices,and speed up the discovery of neighbors and increases the cache success rate to reduce the user's access latency.When multiple nodes send beacons at the same time in the same network,the probability of beacon collisions increases.This situation slows down the speed of the nodes' neighbor discovery.However,the existing neighbor discovery protocol does not consider the beacon collisions.In this regard,this thesis proposes a fast neighbor discovery mechanism with collision avoidance.According to this,it adopts a carrier sense mechanism,which listens to the channel state before transmitting a beacon,then selects whether to transmit a beacon or not according to the channel state.Thereby effectively reducing beacon collision.At the same time,dynamic wake-up slots increase to speed up neighbor discovery,which increases the wake-up time of nodes to increase the probability of the neighbors around the node being discovered or discovered by the neighbors such that it improves the neighbor discovery rate.Simulation results show that this mechanism can be applied to the existing neighbor discovery protocol,and it could accelerate the speed of existing protocol in both symmetric and asymmetric scenes.The cooperative caching strategy directly affects the probability that a node could obtain content through itself or neighboring nodes successfully.The existing cooperative caching strategies do not take into account that the nodes have different user preferences and cache capacity in the actual situation.To this end,a cooperative caching strategy based on heterogeneous cache capacity nodes under user preference was proposed to obtain the optimal caching deployment to maximize the cache hit ratio.A general cache hit ratio maximization problem is formulated with caching placement policies.A coordinate descent algorithm is proposed to obtain the near-optimal solution with respect to different cache capacity in different groups.Simulation results show that the proposed algorithm can achieve significant improvement on cache hit ratio.