Joint User Association, Content Caching and Recommendations in Wireless Edge Networks

Abstract

In this paper, we investigate the performance gains that are achievable when jointly controlling (i) in which Small-cell Base Stations (SBSs) mobile users are associated to, (ii) which content items are stored at SBS co-located caches and (iii) which content items are recommended to the mobile users who are associated to different SBSs. We first establish a framework for the joint user association, content caching and recommendations problem, by specifying a set of necessary conditions for all three component functions of the system. Then, we provide a concrete formulation of the joint problem when the objective is to maximize the total hit ratio over all caches. We analyze the problems that emerge as special cases of the joint problem, when one of the three functions is carried out independently, and use them to characterize its complexity. Finally, we propose a heuristic that tackles the joint problem. Proof-of-concept simulations demonstrate that even this simple heuristic outperforms an optimal algorithm that takes only caching and recommendation decisions into account and provide evidence of the achievable performance gains when decisions over all three functions are jointly optimized.

Fairness in Network-Friendly Recommendations

“…The paradigm of network-friendly (or, network-aware) recommendations has been recently proposed and studied under different network setups and content services [1]- [5], [7]- [17], [19]- [21]. The proposed NFR schemes aim to increase the network gains (and/or improve the quality of content delivery) by selecting recommendations [4], [5], [12], [20] or by jointly designing the recommendation and network policy (e.g., caching) [2], [3], [7]- [11], [13]- [17]. The majority of related works considers cache-friendly recommendations in mobile networks [1]- [4], [7]- [9].…”

“…The majority of related works considers cache-friendly recommendations in mobile networks [1]- [4], [7]- [9]. However, the same principles apply to generic network setups [12], such as coded caching [7], broadcast communications [15]- [17], user association to base stations [13], or swarming systems [21]. While some of the proposed schemes take into account the user perspective by accounting the QoR, none of them has considered the fairness in recommendations from the perspective of the content provider.…”

Network-aware Recommendations in the Wild: Methodology, Realistic Evaluations, Experiments

“…Previous works generalize the above example [4]- [13], [16]- [22], by considering more general recommendation techniques and parameters (e.g., number and order of recommendations [19]), more general content delivery schemes (e.g., multiple caches [6], [20], broadcasting [16], [17], [22]), and more general user demand models (e.g., acceptance of recommendations [18], sequential requests [7]).…”

Socio-aware content allocation in complex networks via efficient monitoring of information dissemination

“…Measurements over the YouTube service show that CABaRet increases the cache hit ratio significantly. The trade-off between caching efficiency and quality of recommendations is studied in [140], [141] and [142]. The objective in [140] is to maximize the cache hit rate by forming user demands towards cached content via recommendations.…”

“…This is achieved by a heuristic algorithm which first places content in caches driven by user preferences and then makes recommendations that promote the cached content. This is further examined by taking into account the associations of mobile users to a caching network of SBSs with limited service capacity in [141], where a methodology is developed to maximize the cache hit rate while guaranteeing a minimum quality of service and quality of recommendations for the users of the Content Provider (CP) platform.…”