Research On Edge Cache Resource Allocation Algorithm In Cellular Networks

With the increasing demand of mobile users for multimedia content,mobile data traffic is also dramatically increasing,which will not only bring heavy traffic transmission pressure to mobile network operators(MNOs),but also increase the delay of users to obtain the required video content during the peak period of user request videos,so as to reduce the experience of users watching video.The introduction of edge cache can effectively alleviate the above problems.On the one hand,edge cache can cache the popular video content to the edge node(EN)in advance,reducing the delay for users to obtain the required content;on the other hand,edge cache can effectively reduce the redundant transmission of the backhaul link in cellular network,greatly easing the traffic transmission pressure of network operators.Based on this,this paper mainly studies the design of edge caching scheme in cellular network,aiming to design an edge caching incentive mechanism to improve the efficiency of caching.Specifically,this paper mainly studies from the following aspects:Firstly,in the scenario which consists of multiple MNOs and multiple content providers(CPs),each MNO sets the unit price for storage space of its ENs and sells these storages to CPs.In order to improve the experience(for example,reduce the delay of users' access to video)of subscribers when watching video,CPs will purchase cache space from MNOs,and cache popular video content at ENs in advance.The goal of MNO is to maximize the revenue from selling cache space.Similarly,the goal of each CP is to maximize their own utilities.The cache resource allocation problem is modeled as a multiple leaders and multiple followers Stackelberg game,in which MNOs,leaders in this game,move first to make actions(set the unit price of the cache space);CPs,followers in this game,make their own decisions(how much cache space to buy from the network operator)after receiving MNOs' strategies.In this game,the followers compete with each other and there is a non-cooperative game,and the exsistence of Nash equilibrium among the followers' non-cooperative game is proved.The distributed iterative algorithm is used to solve the game problem.Simulation results show that the proposed incentive caching mechanism ensures that both MNOs and CPs can obtain the optimal utilities.Secondly,considering that users may have different choices for video definition due to network conditions and personal preferences when watching video,some layers of video files can be cached without caching the whole video file by introducing scalable video coding(SVC)technology.In the scenario which consists of a single EN and multiple CPs,the EN sells storage space to CPs.By predicting user's possible demand for video clarity,CPs will purchase cache space from EN to cache some video layer files of SVC encoded video in advance.The cache resource allocation problem is modeled as a Stackelberg game in which EN is the leader and the CPs are the followers.Due to the limited storage space,CPs need to compete with each other,so there is a non cooperative game.The existence and uniqueness of NE in noncooperative games are proved.The game is solved by backward induction and distributed iterative algorithm.Simulation results show that the proposed layered video cache resource allocation scheme can improve the cache hit rate and effectively reduce the transmission delay of the required video.