Joint Edge Content Cache Placement and Recommendation: Bayesian Approach

Content caching is a key technology driving beyond 5G mobile edge computing and hence, an efficient mechanism is needed to satisfy ultra-reliable low-latency communication. One such mechanism to reduce latency is to use recommendation influenced caching algorithm. To enable a more efficient wireless caching, in this paper, joint optimization of both caching and recommendation is formulated and the influence of the recommendation on the popularity is modelled through a probability transition matrix. To maximize the cache hits, an algorithm is presented to find the optimal joint caching and recommendation actions. Two estimation algorithms namely Point estimation and Bayesian estimation methods are presented. Further, theoretical guarantees are provided on the performance of the algorithm. Finally, simulation results are provided to demonstrate that the proposed algorithm significantly outperforms the existing algorithms in terms of average cache hit rate.