Research On Joint Content Recommendation And Delivery In Mobile Communication Networks

Personalized content retrieval(PCR)service has become a major information service that consumes a large portion of mobile Internet traffic.Joint content recommendation and delivery is a promising design philosophy that could effectively improve the overall user experience with personalized content retrieval services.Existing research on joint content recommendation and delivery mostly focused on a push-type design paradigm called proactive caching,which,however,has multiple inherent drawbacks such as high device cost and low energy efficiency.To overcome the drawbacks of proactive caching systems,this paper proposes a novel pull-type joint content recommendation and delivery system as an alternative.In addition,compared with conventional pull-type protocols,our protocol is more advanced in that it jointly considers the physical layer constraint(i.e.,channel information)and application layer semantics(e.g.,user interests)to optimize the overall system performance.Through such cross-layer design,our protocol can not only give users better experience in content retrieval,but also optimize user experience on the premise of avoiding network congestion and achieve better coexistence with other applications.The main contributions of this article include two parts.First,we investigate the performance of the scheme in a simplified system model that assumed synchronized user behavior,inflexible delay constraint,and identical quality-of-service(QoS)requirement among users.Based on the simplified model,we propose the optimal algorithm and heuristic algorithms for joint content recommendation and delivery problem,establish an analysis framework for system performance,and compare the performance of various algorithms through simulations.Second,we study the performance of joint content recommendation and delivery scheme in a generalized and realistic system model with asynchronous user behavior,flexible outage delay constraint,and customized user QoS.We propose three heuristic algorithms with polynomial complexity based on the generalized model and show the numerical results that characterize the trade-off relationship of system performance.The simulation results show that the proposed joint content recommendation and delivery system can effectively optimize the user QoE of the PCR service in the changing communication scenarios.Our research has shown that by exploiting the user interest semantics as new degree of freedom for system design,the proposed joint content recommendation and delivery system offers a novel and effective approach to better manage the overall QoE of personalized content retrieval services.It represents a promising design paradigm with broad application prospects.