基于非凸分式规划的多用户无线传输能效优化研究

In recent years, the energy consumption has been a bottle-neck for mobile terminals due to limited battery capacity and high rates service in wireless communications. To solve this problem, Green Radio (GR) has been presented, which emphasizes energy efficiency (EE). The EE optimization in wireless transmission schemes plays an important role in Green Radio, which can effectively improve the EE and reduce the energy consumption. The academic and the industrial have paid much effort on the EE optimization in multi-user transmission systems. In this project, we will conduct research in three levels: the unified mathematical model of energy efficient communications, the transmission strategies of base stations and mobile terminals, and the trade-off between EE and spectral efficiency (SE). We will firstly get the mathematical model of the EE optimization in multi-user systems and give solutions to such optimization problem. Next, we will discuss how to design the transmission strategies of base station in downlink and mobile terminals in uplink, which aims to maximize the system EE. Finally, the trade-off between EE and SE will be investigated and a novel metric, called EST (EE and SE Trade-off), will be proposed. The main contributions are as follows: based on the theoretical analysis model in multi-user systems, we give solutions to the EE fractional optimization problem with dual parameterization transformation and low-complexity iterative methods; the joint optimization with adaptive mode selection and power allocation for multi-user transmission scheme is proposed to improve system EE; we design the novel EST metric with weighted sum approach (WSA)，Pareto filtering (PF), and Multiattribute Utility Analysis (MAUA) for multi-user systems.

未来移动通信网络中能量有效无线传输方案研究，对提高网络能量效率，降低系统能耗，实现绿色通信不断的发展和完善具有重要意义，是目前国内外学术界和工业界共同关注的前沿科学问题。本项目将在基于能效优化的多用户传输系统非线性非凸分式规划问题建模，基于分式优化理论的能效传输方案研究，以及能量效率和频谱效率联合优化设计三个层次上开展研究，并对能效优化模型的求解方法、基于能效优化的上下行传输策略、能量效率和频谱效率折中新准则设计等关键科学问题进行重点攻关，为未来绿色通信提供有力的解决方案。主要创新包括：针对建立的能效传输模型非线性非凸分式优化问题，提出含参对偶变换和低复杂度迭代等求解方法；针对多用户传输方案，提出将用户自适应模式选择机制与功率分配问题联合优化以提高系统可达能量效率；针对能量效率和频谱效率的折中问题，提出利用加权求和、帕累托过滤、多属性效用分析等方法来设计能量效率-频谱效率折中新准则。

未来移动通信网络中能量有效无线传输方案研究，对提高网络能量效率，降低系统能耗，实现绿色通信不断的发展和完善具有重要意义，是目前国内外学术界和工业界共同关注的前沿科学问题。在基于能效优化的多用户传输系统非线性非凸分式优化方法研究方面，我们针对非理想信道估计场景的多天线系统的功率分配问题，利用分式优化理论成果提出了一种鲁棒的功率分配机制，能够获得近似最优解，且复杂度更低，仿真结果证实了所提功率分配方案的有效性和鲁棒性。在能效传输方案研究方面，我们针对广播和通信融合网络场景的能效性能分析问题，提出了一种新的系统模型及能效分析方法。从用户请求业务角度出发，将请求的业务分类为热门与非热门业务。并基于随机空间网络模型，将融合网络中的蜂窝基站与用户建模为2个独立的空间泊松点过程（PPP）随机分布，而广播大塔位于服务区域中心。通过考虑广播网的覆盖范围及蜂窝网的中断概率推导出广播与通信融合网络下系统的能量效率（EE）闭合表达式，进一步讨论热门业务概率对融合网络能效的影响。仿真结果表明，基于讨论的融合系统模型，并考虑了通用场景下的基站和用户密度，当热门业务超过总业务的5%后，融合网络能效相比于单蜂窝网络有显著提高。这种新的系统模型及能效分析方法为后期分析融合网络总体性能奠定了基础。在能量效率和频谱效率联合优化设计方面，针对下行MU-MIMO系统效用最大化问题，建立了能够表达能量效率与频谱效率折中关系的系统效用函数，并提出了一种迭代用户功率分配算法，给出了能够实现系统效用最大的传输模式选择方法。仿真结果表明，所提算法能够使得系统效用取得最大值。同时，在不同的效用偏好参数下，所提方案均可以获得最优的系统效用。