Energy-Efficient Power Control: A Look at 5G Wireless Technologies
Energy-Efficient Power Control: A Look at 5G Wireless Technologies
复制标题
DOI：
10.1109/tsp.2015.2500200
复制
发表时间：
2015-03
影响因子：
5.4
通讯作者：
Alessio Zappone;L. Sanguinetti;G. Bacci;Eduard Axel Jorswieck;M. Debbah
中科院分区：
文献类型：
--
作者：
Alessio Zappone;L. Sanguinetti;G. Bacci;Eduard Axel Jorswieck;M. Debbah
本文开发了无线网络中能量效率（EE）最大化（以比特/焦耳为单位）的功率控制算法。与以前的相关工作不同，施加了最小速率约束，并且信号与干扰加噪声比采用了更一般的表达式，这使得人们可以包含一些最有前途的5G候选技术。以网络为中心和以用户为中心的EE最大化都被考虑。在以网络为中心的场景中，执行全局EE的最大化和网络的最小EE。与以前的贡献，我们开发的集中式算法，保证收敛，负担得起的计算复杂性，所考虑的非凸优化问题的Karush-Kuhn-Tucker点。此外，封闭形式的可行性条件。在以用户为中心的情况下，博弈论被用来研究网络的均衡，并推导出收敛的功率控制算法，该算法可以以完全分散的方式实现。在单个或多个资源块被用于数据传输的假设下研究上述两种场景。数值结果评估所提出的解决方案的性能，分析最小速率约束的影响，并比较以网络为中心和以用户为中心的方法。
This paper develops power control algorithms for energy efficiency (EE) maximization (measured in bit/Joule) in wireless networks. Unlike previous related works, minimum-rate constraints are imposed and the signal-to-interference-plus-noise ratio takes a more general expression, which allows one to encompass some of the most promising 5G candidate technologies. Both network-centric and user-centric EE maximizations are considered. In the network-centric scenario, the maximization of the global EE and the minimum EE of the network is performed. Unlike previous contributions, we develop centralized algorithms that are guaranteed to converge, with affordable computational complexity, to a Karush-Kuhn-Tucker point of the considered non-convex optimization problems. Moreover, closed-form feasibility conditions are derived. In the user-centric scenario, game theory is used to study the equilibria of the network and to derive convergent power control algorithms, which can be implemented in a fully decentralized fashion. Both scenarios above are studied under the assumption that single or multiple resource blocks are employed for data transmission. Numerical results assess the performance of the proposed solutions, analyzing the impact of minimum-rate constraints, and comparing the network-centric and user-centric approaches.