(ICWS 2024) User Preference-Informed and Mobility-Aware Caching in a Cooperative MEC Environment

As massive data is generated by Internet of Things (IoT) devices, user-end devices, which are usually resource and storage-poor, are required to implement computation-intensive functionalities. Mobile Edge Computing (MEC) is a significant technological convergence that has the potential to fill the gap and extent the computation and storage capacities of user-end devices by decentralization of required resources and contents near users and at the edge. Cooperative caching upon edge servers is becoming popular in this direction. Since cached content may not always align with user requests, potentially impacting the perceived quality of service, a key challenge is to develop an intelligent mechanism that efficiently pre-caches content based on the preferences of mobile users to ensure a high cache hit rate. For this purpose, in this paper, we propose a User Preference-informed and Mobility-aware method for cooperative MEC caching (MCL CPC). The proposed method predicts future user preferences and destinations, clusters users accordingly, and caches content according to the clusters. Extensive experiments demonstrate that our proposed method outperforms its peers in terms of hit rate, content delivery latency, and cache utilization.