Resource allocation of fog radio access network based on deep reinforcement learning

With the development of energy harvesting technologies and smart grid, future trend radio access networks will present a multi-source power supply. In this article, joint renewable cooperation resource allocation scheme fog (F-RANs) with hybrid supplies (including both conventional grid sources) is studied. our objective to maximize average throughput F-RAN architecture sources while satisfying constraints signal noise ratio (SNR), available bandwidth, harvesting. To solve problem, dynamic in network studied by using Q-learning Deep Q Network respectively. Simulation results show that proposed two algorithms have low complexity can improve whole compared other traditional algorithms.