Caching of popular content on wireless nodes is recently proposed as a meansto reduce congestion in the backbone of cellular networks and to improveQuality of Service. From a network point of view, the goal is to offload asmany users as possible from the backbone network to the wireless caches whileat the same time offering good service to cache-unrelated users. Aggressiveoffloading can lead to an unbalanced user association. Some wireless nodes canbe overloaded by cache-related traffic while the resources of others remainunderused. Given a fixed content placement, this work proposes an efficientdistributed algorithm to control and balance the association of cache-relatedtraffic among cellular cache memories. The algorithm allows the network toachieve the globally optimal solution and can be executed on base stationsusing a limited amount of information exchange between them. It is based on anovel algorithm we call Bucket-filling. The solution limits the cache-users pernode by balancing the total load among the nodes in a fair way. The improvementcompared to common user assignment policies is highlighted for single- as wellas for multi-tier random networks.
展开▼