Distributed Optimal Green Power Allocation for D2D Based Cellular Networks with Long-Term QoS Constraint
- Additional Document Info
- View All
© 2016 IEEE. We investigate the energy efficiency (EE) maximization problem under the scenario of Device-to-Device (D2D) communications underlaying cellular networks. By jointly considering time-varying channels and stochastic arrivals, we propose a distributed power allocation algorithm for both D2D user groups (DUGs) and cellular user terminals (CUTs) with long- term quality of service (QoS) provision. To effectively deal with the dynamics of D2D based cellular networks, stochastic optimization problems are formulated, so as to optimize the long-term EE of both DUGs and CUTs under the constraints of queue stability, battery capacity and average transmission rate. With weighted sum method and Lyapunov optimization approach, we can sufficiently handle the non-convexity of the original problem by deriving an equivalent one. Through optimal solutions of the problems, each DUG and CUT can manage their power resources more efficiently, which improves their long-term EE performance individually. The simulation results further validate the theoretical analyses and evaluate the effectiveness of our proposed algorithm.
author list (cited authors)