【作者】 李红岩；

【导师】 吴伟陵；

【作者基本信息】 北京邮电大学 ， 信号与信息处理， 2009， 博士

【摘要】 随着无线技术的飞速发展,人们对宽带无线业务的需求不断增加,频谱这种不可再生的宝贵资源正逐渐变得短缺。一方面,频谱供需的矛盾日益突出;另一方面,在当前的频谱管理策略下,已分配频谱的利用率又极其不足。在这一背景下,认知无线电技术应运而生。认知无线电是一种具备动态频谱接入功能的智能无线通信系统,能在不干扰授权用户的前提下伺机利用频谱机会,提高已分配频谱的利用率。目前,认知无线电技术已成为无线通信的热点研究领域。国内外学者和研究机构都开展了相关研究,但是这些研究尚处于初期阶段,还很不成熟。鉴于此,本论文以提高认知无线电的性能和效率为目的,对认知无线电的功率优化、MAC层频谱检测效率、802.22系统中基站间的频谱共享等关键技术方面做了重点研究,所得结果对未来认知无线电系统的最终实现具有一定的参考价值和实际意义。本文所作的几点工作如下:一、本论文通过讨论当前认知无线电的研究现状,分析了其特点和面临的挑战性技术问题,并主要采用博弈论探讨了认知无线电的若干问题。认知无线电技术的主要目标是实现与授权系统的频谱共存以及与其他认知无线电系统的频谱共享,因而,协调多个相互竞争实体的频谱接入行为是其面临的关键问题。这类问题适合采用微观经济学中的博弈论来解决。博弈论是一种处理多个相互影响的实体之间竞争和协作的数学公具,它提供了丰富的博弈模型用于实现不同目标的优化。二、认知无线电应用的前提是能够与授权系统共存,不侵犯到授权用户的正常通信。干扰温度模型是一种认知无线电系统与授权系统共存的模型,基于该模型,共存问题转变为认知无线电网络的功率优化问题。本论文的第3章针对认知无线电ad hoc网络和授权系统的空域共存范畴,研究了多测量点场景下认知无线电网络的功率控制问题。基于干扰温度模型,从最大化认知无线电网络总效用的角度出发,利用博弈理论,提出一种全局最优的分布式功率控制(TPCG)算法。该算法不需要测量点的管理和控制,且适用于存在多个测量点的频谱共享场景。文中通过理论证明和计算机仿真验证了TPCG算法的性能。三、本论文的第4章研究了认知无线电的MAC层频谱检测周期优化问题。由于认知无线电终端硬件条件的限制,在仅配置一副天线的情况下,为了能够在单位时间内检测到并利用较多的频谱,需要设计高效的MAC层检测算法以提高频谱检测效率。本论文在对授权用户的信道使用情况进行建模的基础上,提出了一种新型的MAC层检测周期优化算法。该算法通过自适应的调整每条信道的检测周期,提高了单位时间内的信道利用率,仿真结果验证了本算法的自适应周期检测相比于固定周期检测的性能优势。四、本论文的第5章研究了IEEE 802.22 WRAN网络的基站间频谱共享问题。当多个WRAN基站的覆盖范围相互重叠时,WRAN基站间存在频谱机会的接入竞争问题。考虑到拍卖理论在平衡资源需求和激励资源共享方面的独特优势,为了鼓励基站间平衡信道分配,避免频谱接入冲突,提高系统的频谱利用率,本论文提出一种基于拍卖理论的WRAN基站间频谱租借方案。该方案将WRAN小区间的频谱共享情况抽象为一个拍卖基站和多个竟拍基站的拍卖模型,在满足基站间的干扰约束下,以提升拍卖收益为优化目标,给出一种频谱分配的近似算法,并通过计算机仿真度量了分配效率。最后,总结全文工作的同时,给出了下一步的研究方向。更多还原

【Abstract】 With the rapid development of wireless communication technologies and the increasing broadband communication demands, the shortage of spectrum is becoming an outstanding problem. On one hand, the conflicion between spectrum supply and demand is increasing. On the other hand, according to the survey of related organizations the utilization of licensed spectrums is very low under the present static spectrum management mode. Under the above background, the cognitve radio (CR) technology is prposed, which aims at enhancing the utilization of licensed spectrum with intelligent using the unused spectrum of licensed systems.CR becomes a hot topic in wireless communication research. Related research are undergoing by scholars from universities all over the world. But the CR technology is also in the beginning stage and there are many drowbacks need to be improved. In this paper, our researches focus on CR power control scheme, MAC layer sensing efficiency and spectrum sharing in IEEE 802.22 systems. The game theory and auction theory in microeconomics are introduced to inspire new ideas. Several new algorithms are proposed and the performances are verified through computer simulations. The researches have meaningful value for the realization of CR systems.The work in this dissertation is concluded as following points.Firstly, the research progress of CR is discussed detailedly, and the characteristic and technology challeges are analyzed.Secondaly, in chaper 3, we considered decentralized cognitive radio networks, and we especially focusd on the multiple measurement points scenario. Interference temperature model is proposed for spectrum sharing. Based on interference temperature model, the problem of spectrum sharing can be formulated as a power optimization problem at physical layer. Game theory is used to investigate the distributed power control for providing the maximum throughput in cognitive radio networks. A tax-based power control game algorithm is introduced to implement power allocation optimization in a distributed mode, and guarantees the convergence to globally optimal power allocations.Thirdly, chapter 4 discussed the MAC layer spectrum sensing in cognitive radio network. Efficient MAC layer spectrum sensing scheme can improve sensing performances and find more available spectrum opportunities. Focusing on solving MAC layer the sensing period optimization problem, we proposed an adaptive sensing period adjustment algorithm. The proposed algorithm realized the automation adjusting of sensing period based on the channel model of primary users, and found more opportunities than fixed period sensing.Fourthly, in chapter 5 we studied the spectrum sharing problem among different IEEE 802.22 WRAN cells. Aiming at efficiently solving the spectrum sharing problem, a dynamic spectrum renting scheme is proposed. Based on auction theory, the proposed spectrum renting algorithm obtains higher spectrum utilization and higher auction benefit, and guarantees no interference between competitive WRAN base stations.Finally, a conclusion is drawn for the dissertation, and valuable research directions in the future are presented.更多还原