基于能量收集的认知无线电系统中感知接入理论研究

Cognitive radio based on energy harvesting is a new wireless communication technology, which has high efficiency of spectrum utilization and is self-sustainable in energy. This technique is an effective solution in front of scarce spectrum resources and urgent restriction of energy consumption and carbon emission. In a cognitive radio system based on energy harvesting, how to design the strategy of sensing and access, realize the balance between energy harvesting and consumption, fully utilize spectrum opportunities is the key issue. This project will investigate the key issue. In this project, the process of spectrum sensing and access by the secondary user and the process of energy arrival will be modelled at first. Then the mechanism between the spectrum sensing and access parameters, energy scheduling parameters, and system performance will be disclosed. At last the optimization of spectrum sensing and access parameters given the energy scheduling parameters and the design of energy scheduling mechanism will be carried out with the aid of convex optimization, multiple-level optimization, monotonic programming, semi-infinite programming, optimal stopping, Markov decision process To this end, the design of sensing and access strategy has been accomplished, and the objective of maximizing the average throughput while limiting the interference to primary users and the occurrence of the energy supplying outage event. The research results of this project will promote the development of the research on “cognitive radio based on energy harvesting” and will provide theoretical guidance and technical support for the application of cognitive radio based on energy harvesting.

基于能量收集的认知无线电是一项频谱利用率高、能量自我永续的新型无线通信技术，是在频谱资源日益紧缺、节能减排迫在眉睫形势下的一种有效解决方案。在基于能量收集的认知无线电系统中，如何设计感知接入策略，确保能量收支平衡、充分利用频谱机会是核心问题。本项目将对上述核心问题开展研究。通过对次级用户感知接入过程和能量到达过程进行理论建模，探究频谱感知接入参数、能量调度参数与系统性能指标间的作用机理，采用凸优化、分层优化、单调优化、半无限优化、最优停时、马尔可夫判决过程等理论工具，分别开展给定能量调度参数条件下频谱感知接入参数的优化和能量调度机制设计，最终完成感知接入策略的整体设计，实现在限制对主用户的干扰和能量供应中断事件发生的条件下最大化次级用户平均速率的研究目标。本课题的研究成果将可以推动“基于能量收集的认知无线电”这一研究领域的发展，为其实际应用提供理论指导和技术支撑。

为提高频谱利用效率、实现能量自我永续，基于能量收集的认知无线电系统是实现上述愿景的有效技术手段。随着物联网等应用的兴起，这种技术的大规模应用更加值得期待。本研究围绕基于能量收集的认知无线电系统，着力解决该系统中的关键问题：频谱资源的高效使用，以及收集能量的充分利用。为解决上述问题，研究从多个角度依次展开。在频谱资源的高效使用方面，考虑多用户对频谱租赁和衬垫接入等模式都进行了研究，利用博弈论、凸优化、统计学等理论与方法，实现了多主用户频谱拍卖的纳什均衡、有限干扰信道样本数据下干扰可控接入。在能量收集系统方面，分别针对单跳信能同传、两跳多中继信能同传、多跳多中继信能同传型网络进行了资源分配的优化，利用凸优化等方法将其对应的一系列非凸问题转化为凸问题或者单调规划问题，获得全局最优解，还结合能量的随机游走过程研究了最优能量调度问题，给出了低复杂度的能量调度策略。