This is a repository copy of MEC-assisted immersive VR video streaming over terahertz wireless networks: A deep reinforcement learning approach.

In order to improve the edge caching efficiency of the fog radio access network (F-RAN), this paper put forward a distributed deep Q-learning-based content caching scheme based on user preference prediction and content popularity prediction. Given that the constraint that the storage capacity of each device is limited, and the optimization problem is formulated so as to maximize the caching hit rate. Specifically, by taking users' selfishness into consideration, user preference is predicted in an offline manner by applying popular topic models. Then, the online predicted content popularity is achieved by combining the network topology relationship together with the obtained user preference. Finally, with the predicted user preference and content popularity, the deep Q-learning network (DQN)-based content caching algorithm is proposed to achieve the optimal content caching strategy. Moreover, we further present a content update policy with user preference and content popularity prediction, so that the proposed algorithm can handle the variations of contents popularity in a timely manner. Simulation results demonstrate that the proposed scheme achieves better caching hit rate compared with existing algorithms.

No abstract

Most mobile edge computing (MEC) works assume that the mobile devices (MDs) can offload their tasks to the MEC-severs at anytime, which may not be a practical assumption due to the tension between a large number of MDs and the scarce spectrum resources. In this paper, a framework for wireless powered cognitive radio (CR)-based MEC-enabled networks is proposed, which integrates three technologies: MEC, CR, and wireless power transfer (WPT). CR technology with imperfect spectrum sensing is adopted by the MD to find the spectrum access opportunities. An optimization problem is formulated to maximize the average calculated number of bits (CNoB) of the MD, which is non-convex and hard to solve. A two-loop procedure using a one-dimensional line search method is proposed. The time for spectrum sensing, WPT, energy harvesting (EH) and offloading, the central processing unit (CPU) frequency, and the transmit power of the MD are jointly optimized. Some semi-closed form solutions are obtained through Lagrangian dual decomposition and successive pseudo-convex approximation (SPCA) methods. Simulation results are presented to show the effectiveness of the proposed CR-based MEC scheme with different parameters.INDEX TERMS Mobile edge computing, cognitive radio, imperfect spectrum sensing, wireless power transfer, semi-closed form.

No abstract