In-Network Caching and Content Placement in Cooperative Small Cell Networks

Abstract

Abstract-Anticipating multimedia file requests via caching at the small cell base stations (SBSs) has emerged as a promising technique for enhancing the quality-of-service (QoS) of cellular user equipments (UEs). Nevertheless, in traditional caching approaches, files are retrieved from evolved packet core (EPC) of the network, and without coordination among SBSs, which introduces new challenges of content duplication and unbalanced traffic load on the backhaul. In this paper, we propose a collaborative framework in which the SBSs can access files from the caches of other SBSs within the same network domain (i.e., connected to the same service gateway). We design a cost model for the content retrieval of contents across SBSs and from the EPC, and we propose a costaware decentralized algorithm, based on which the SBSs devise individual caching strategies (i.e., which files to cache, and from where). Simulation results show that the proposed cooperative caching scheme yields significant gains in terms of in-network content availability, reaching up to 21% improvement compared to a traditional approaches based on geographical distribution of the UEs.

Gibbsian On-Line Distributed Content Caching Strategy for Cellular Networks

“…The paper [13] proposes a pricing based scheme for jointly assigning content requests to cellular BSs and updating the cellular caches; but this paper focuses on certain cost minimization instead of hit rate maximization, and it is optimal only when we can represent the data by very large number of chunks which can be used in employing rateless code. The problem of collaborative but decentralized caching among small base stations for a certain cost minimization has been analyzed in [14], under the assumption that the caches have access to the contents of other caches connected to the same gateway; their formulation involves a certain cost for retrieval of a content from another cache. The authors of [15] address the problem of minimizing energy consumption under multi-cell transmission cooperation for interference reduction and content caching in heterogeneous networks.…”

Octopus: A Cooperative Hierarchical Caching Strategy for Cloud Radio Access Networks

“…Furthermore, we assume the cost of transferring a file from cache of BS k to BS r is d rk . In practice, d 0 is usually of many-fold higher than d r and d rk [12], [19], [20]. This makes it cost-effective to retrieve content from the in-network caches whenever possible rather than downloading them from the CDN.…”

A user‐centric cooperative edge caching scheme for minimizing delay in 5G content delivery networks

“…A Markov chain distribution model was proposed to assess the performance of the transition between CRAN and the CDN in terms of caching time and bandwidth consumption. Moreover, to achieve better content delivery service, independent wireless channels for content dissemination (from source server to local servers) and content delivery (among local servers) in wireless CDN were used in the work of Sung et al 28 Unlike previous schemes such as the work of Pantisano et al 26 where cached contents are dropped when the storage capacity of the SBS reaches its limit, and the work of Chen et al 27 where replicas of the popular content are stored in each SBS while saving portion of the storage capacity of SBSs for less popular files, this paper contributes to the problem of content caching and delay minimization in user-centric delivery schemes in 5G CDNs by proposing a novel cooperative caching scheme to expand the "virtual" cache capacity so that more files can be stored in the vicinity and served to users from a the shared storage capacity of all SBSs. The performance of the proposed scheme is compared against a baseline scheme and a particular benchmark clustering scheme wherein the SBSs storage capacity varies, showing that it can achieve a local serving ratio of 99.5% under realistic parameter settings.…”