On Energy-Efficient Resource Allocation In Wireless Networks

The rapid development of wireless networks incurs enormous energy cost consumed by the information and communication industry and meanwhile leads to a series of problems such as the environmental pollution and the operation of wireless networks. For these rea-sons, green communications achieve the worldwide recognition upon proposition and are becoming an inevitable trend for the future wireless network design. Aiming at construct-ing resource-saving and eco-friendly networks, green communications focus on minimizing the power consumption or maximizing the network energy efficiency (EE) subject to users’ quality-of-service or quality-of-experience. Therefore, it is well worth studying and break-ing through the key theories and technologies for energy-efficient transmission in wireless networks.To this end, this thesis devotes to investigating the energy-efficient transmission problem-s through adaptive resource allocation in spectrum-sharing wireless networks, Orthogonal Frequency Division Multiple Access (OFDMA) networks, and Sparse Code Multiple Ac-cess (SCMA) networks. Specifically, the main contents and contributions are summarized as follows.1. We investigate the fundamental energy-aware power allocation problem in spectrum-sharing wireless networks. We introduce the revenue-cost (RC) index to characterize how efficiently the energy is consumed and formulate an optimization problem to maximize the RC subject to the power budget constraints. We first show that the RC maximization problem is intimately connected with the extensively investigated energy efficiency and spectral efficiency maximization problems. Due to the nonconvexity and NP-hardness of the formulation, we focus on computation-efficient and easy-implementation algorithm design. Leveraging the high signal-to-interference-plus-noise ratio approximation, we develop an extremely simple algorithm with fast convergence, fully distributed frame-work, and tuning-free properties. Simulation results exhibit the effectiveness of the pro-posed algorithm.2. We joint subcarrier assignment and power allocation to explore the green transmission problem in downlink OFDMA systems. We formulate it as a mixed-integer nonlinear programming, which maximizes the system energy efficiency subject to rate requirements of users, transmit power budget, and subcarrier assignment constraints. Due to the mixed combinatorial features, it is prohibitive to find its globally optimal solution in terms of computational cost. On account of this, we propose an effectively iterative algorithm with good performance and low complexity leveraging the nonlinear fractional programming and the Lagrangian dual decomposition. Simulation results show the fast convergence and effectiveness of the proposed algorithm.3. We investigate the energy-efficient transmission problem by jointing codebook assign-ment and power allocation in SCMA networks. We formulate it as an optimization problem to maximize the network energy efficiency subject to quality-of-service require-ments, codebook assignment, power allocation, and subcarrier reuse constraints. Due to its mixed combinatory, we separate codebook assignment and power allocation to devise suboptimal but cost-efficient algorithms. Simulation results exhibit the superior of the proposed algorithms against the existing classical schemes and of SCMA over OFDMA in terms of the network EE.