A Fundamental Study on Energy Efficient Wireless Communication Networks: Modeling, Algorithms, and Applications

无线网络的快速发展和广泛部署导致了能源需求的快速增长，这迫切需要节能的网络技术。同时，无线网络正在演变为具有通信链路、无线电、信道、天线、发射功率等多维资源的复杂网络形态；对低复杂度的能效优化算法有很高的要求。本项目旨在通过建立统一的分析框架，开发高效低复杂度的算法，并将通用研究应用于第五代（5G）蜂窝系统的重要场景，对节能无线网络进行基础研究。这种跨学科的研究不仅为本科生和研究生的学习提供了丰富的训练项目，而且还将激发学生以创新、开放和跨学科的视角追求高质量的研究。本项目将通过网络的干扰模型来定义合适的传输模式，并通过调度的原则来构建一个统一的多维优化框架来进行能效优化。在这样一个统一的优化模型下，低复杂度分解技术从根本上与根据信噪比的物理干扰模型下的最大加权传输方向图（MWTP）问题相关。MWTP问题（一般是np困难问题）的近似算法和相关性能分析是需要研究的关键研究问题。解决能效优化问题的分布式算法进一步涉及多目标优化分解、物理干扰模型下创新的Lyapunov函数设计和相关稳定性分析，这些也将在本项目中进行研究。通过统一多维框架中的创新建模和算法，将在几个重要的5G蜂窝场景中实现节能解决方案，包括结合网络层流量约束和系统级基站休眠的大规模MIMO干扰缓解的联合优化，大规模MIMO干扰模型下MWTP问题的制定和算法开发。以及大规模MIMO和设备对设备通信之间相互作用的建模。在这个项目中，提出的研究无缝地整合了优化、图论、对偶分解、近似算法、无线通信和网络等领域的研究。该研究成果有望为5G蜂窝系统的发展提供重要指导。

The rapid development and wide deployment of wireless networks incur a fast escalation of energy demand, which eagerly calls for energy-efficient networking techniques. At the same time, wireless networks are evolving into complex forms with multi-dimensional resources including communication link, radio, channel, antenna, and transmit power; algorithms with low complexity for energy efficiency optimization are highly demanded. This project targets at a fundamental study on energy-efficient wireless networking through establishment of a uniformed analytical framework, development of efficient and low-complexity algorithms, and application of the generic studies into important scenarios in the fifth generation (5G) cellular systems. This interdisciplinary research will not only provide various training projects to undergraduate and graduate studies, but also inspire students to pursue high-quality research with a creative, open-minded, and cross-disciplinary perspective. This project is going to demonstrate that a uniformed multidimensional optimization framework for energy efficiency optimization can be constructed by the principle of scheduling proper transmission patterns, which are defined by the interference model of the network. With such a uniformed optimization model, low-complexity decomposition techniques are fundamentally related to a maximum weighted transmission pattern (MWTP) problem, under a physical interference model according to the signal-to-interference-plus-noise ratio. Approximation algorithms and associated performance analysis for the MWTP problem (which is NP-hard in general) are critical research issues to be studied. Distributed algorithms for solving the energy efficiency optimization problem further involves decomposition with multi-objective optimization, and innovative Lyapunov function design and associated stability analysis under the physical interference model, which will also be addressed in this project. Energy efficient solutions in a couple of important 5G cellular scenarios will be enabled through innovative modeling and algorithms in the uniformed multidimensional framework, including joint optimization that incorporates massive MIMO interference mitigation with flow constraint at network layer and base station sleeping at system level, formulation and algorithm development for a MWTP problem under the massive MIMO interference model, and modeling of the interplay between massive MIMO and device-to-device communications. In this project, the proposed research seamlessly integrates studies in the areas of optimization, graph theory, dual decomposition, approximation algorithms, and wireless communication and networking. The research outcomes are expected to provide important guidance for the development of the 5G cellular systems.