Arrival Rate-Based Average Energy-Efficient Resource Allocation for 5G Heterogeneous Cloud RAN

Heterogeneous cloud radio access network (H-CRAN) needs more elegant design to achieve higher energy efficiency and spectral efficiency than traditional cloud radio access networks. In this paper, we propose an energy-efficient resource allocation algorithm by taking into account the impact of arrival rates of various user traffic. Firstly, based on the power consumption model of the H-CRAN, the average energy-efficiency of the whole network is adopted as the optimization objective with multiple constraints of maximum transmit power, average power, and minimum data rate of each users, etc. In order to solve the non-convex and non-deterministic polynomial time- hardness (NP-hard) problem, we transform the objective function into analyzable multiple sub-problems by using fractional programming and norm approximation. Secondly, by the Lyapunov optimization method, we turn the original problem into a problem of system stability. Thirdly, we derive the closed expression of the optimal power allocation matrix and the optimal user association matrix with the Lagrangian dual decomposition. We propose a two-layer iterative algorithm to balance the power consumption and energy efficiency with a designed control factor. Both theoretical bound of average energy efficiency and length of data queuing are derived. Finally, the comprehensive numerical results demonstrate of convergence of the proposed algorithm and verify the performance gain by proposed energy-efficient resource allocation scheme.