A General Framework for Spectral Efficiency, Energy Efficiency and Delay Tradeoff in D2D Networks

In this paper, we propose a novel framework to investigate the three-factor tradeoff among spectral efficiency (SE), energy efficiency (EE) and delay in D2D communications underlaying cellular networks. Taking both channel state information and queue state information into account, a stochastic multi objective optimization problem is formulated aiming at maximizing SE and EE simultaneously subject to the queue stability constraint, where subchannel allocation and power optimization are jointly considered. The single-objective optimization problem is then derived with the help of weighted sum method and Lyapunov techniques, where an effective algorithm is designed to achieve the three-factor tradeoff among SE, EE and delay. Afterwards, the optimal subchannel and power allocation is obtained via the Lagrangian dual decomposition method, which only requires polynomial computational complexity. Both in-depth theoretical analysis and simulation results verify that the proposed algorithm enables the system controller to comprehensively balance the three-factor tradeoff among SE, EE and delay by flexibly adjusting two introduced weighting parameters.