Caching Policy for Cache-enabled D2D Communications by Learning User Preference

Prior works in designing caching policy do not distinguish content popularity with user preference. In this paper, we optimize caching policy for cache-enabled device-to-device (D2D) communications by exploiting individual user behavior in sending requests for contents. We first show the connection between content popularity and user preference. We then optimize the caching policy with the knowledge of user preference and active level to maximize the offloading probability, and develop a low-complexity algorithm to find the solution. In order to learn user preference, we model the user request behavior resorting to probabilistic latent semantic analysis, and learn the model parameters by expectation maximization algorithm. By analyzing a Movielens dataset, we find that the user preferences are less similar. The dataset also shows that the active level and topic preference of each user change slowly over time. Based on this observation, we introduce a prior knowledge based learning algorithm for user preference, which indicates the potential of a hierarchical learning strategy in accelerating convergence rate. Simulation results show remarkable performance gain of the caching policy with learned user preference over existing policy with learned content popularity, which are obtained with the synthetic data validated by the Movielens dataset.