【作者】 陈绪君；

【导师】 刘守印；

【作者基本信息】 华中师范大学 ， 无线电物理， 2011， 博士

【摘要】 随着无线通信技术的发展,对无线应用的需求迅速增长,然而无线信道的多径衰落严重影响了通信的质量。研究表明,分集技术是克服多径衰落影响的有效手段之一。其中空间分集由于不占用额外带宽和时隙资源被引起广泛关注。但是具备空间分集的多天线系统由于成本和尺寸的原因,在实际应用中存在诸多限制。Sendonaris和Laneman等人提出利用源节点和周围一个或多个相邻节点组成的分布式虚拟多天线系统来获得空间分集,这种利用相邻节点获得分集的方式称为协作分集。目前,关于协作分集的研究主要存在如下问题：(1)大多数的研究都是假定协作通信的收发端均已知理想的信道状态信息,对于信道估计误差对协作通信的影响研究甚少；(2)“Best"选择协作通信研究大多假定是“全天候协作”,即无论中继链路质量的好坏都保持协作,很少考虑分集和带宽效率的折中问题；(3)当“Best"中继已经被其它用户调度,当前用户必须选择次优或更次优中继,该条件下的协作系统性能分析有待深入研究；(4)多用户协作可以同时获得多用户分集和协作分集,该系统性能的理论分析仍有待深入研究。针对以上问题,论文研究了相应的协作系统及算法,主要包括：(1)研究了存在信道估计误差条件下协作分集性能分析的数学模型,分析了信道估计误差对协作系统的影响。①对于DF协作系统,得到了Rayleigh信道存在估计误差条件下的平均码元错误概率闭式解及其上界表达式。进一步考虑总功率约束条件,分析了优化功率分配后的平均码元错误概率性能。对于Nakagami-m信道,得到了存在信道估计误差条件下的平均码元错误概率表达式,分析了不同Nakagami分布参数m对协作系统的影响。②对于AF协作系统,通过非直接链路的端到端信噪比的近似表达式得到了接收信号信噪比的概率密度函数PDF、累积分布函数CDF和矩生成函数MGF,并推导了该系统的平均码元错误概率和中断概率表达式。(2)研究了中继选择协作系统和中继选择算法：①对于中继选择系统,考虑分集性能和带宽效率的折中问题,提出了基于S-D链路信噪比门限的中继选择算法和基于S-R-D/S-D信噪比比值门限的中继选择算法,推导了上述中继选择算法的平均码元错误概率和带宽效率表达式,分析了不同的门限值对平均码元错误概率和带宽效率的影响。理论分析和仿真结果表明,对于S-D准则,可通过减小协作门限值提高分集效果,可通过增大协作门限值提高带宽效率；对于S-R-D准则,当链路处于低信噪比时,可通过减小协作门限值获得更好的分集效果。当链路处于高信噪比时,可通过增大协作门限值提高系统的带宽效率。做到分集和带宽效率的折中。不仅解决了“与谁协作”的问题,而且解决了“何时协作”的问题。②对于集中调度的多源-多中继协作系统,利用Order Statistic理论研究了公平调度下选择次优或第k优中继的协作模型,推导了该模型下的端到端信噪比的PDF、CDF和MGF,并推导了非相同分布信道参数条件下的平均码元错误概率和平均信噪比增益表达式。(3)研究了Rayleigh信道独立同分布条件下的联合多用户分集和协作分集系统,得到了该协作系统平均码元错误概率和平均信道容量闭式表达式,分析了不同用户数目对协作系统的影响。理论分析和仿真结果表明,联合多用户分集和协作分集的协作系统分集阶数不再是2而是2K,即多用户协作分集阶数随着用户数K的增多而成倍提高。更多还原

【Abstract】 With the development of wireless communication, the demand for wireless applications increases rapidly, however QOS of wireless communications is badly degradated by multi-path fading channel. Research shows that diverstiy is a valid technology to overcome degradation. Space diversity technology attracts broad attention owing to no extra bandwith and tme slots occupied. However the multi-antenna system with space diversity is restricted by the cost and the size. Sendonaris and Laneman studied a system that source node can resort to one or more neighboring nodes who can help source retransmit the received signal from the source to the destination node, called cooperative diversity.The present research on cooperative diversity exists following questions:(1) most literatures assume that both the transmitter and the receiver know ideal CSI for cooperative diversity. Few studied the performance of cooperative communication with channel estimation errors. (2) Research on "best selection" most assumes that cooperation is’all-participate-in’cooperative scenes, seldom studied trade-off between diversity and bandwith efficiency. (3) While the "best" relay is completely scheduled by other users, the source must resort to sub-optimal or kth optimal-relay, this model remains intensive study for future. (4) Multi-user cooperation can achieve not only multi-user diversity but also cooperative diversity, the exact theoretical expressions awaits deep research.In order to answer these questions, serveral cooperative models and algorithms are studied. The contibution of the thesis is listed as follows:(1) We study the mathematical model for performance analysis of cooperative diversity with channel estimation errors. The impact of channel estimation errors on cooperation is analyzed.①As for DF cooperation, we obtain the average symbol error probability(SEP) and its upper bound expressions for Rayleigh fading channel with estimation errors. For the case of cooperative diversity with total power constraint, we analyse the performance of the average SEP with optimal power allocation. As for Nakagami-m fading channel, we also obtain the average SEP expressions and analyse different m parameters how to affect the performance of cooperative diversity.②As for AF cooperation, we analyse the performance of AF cooperative diversity with channel estimation error and derived PDF,CDF and MGF expression of End-to-End signal-to-noise ratio based on non-direct link with approximation bound. Then we derive the average SEP and outage probability expressions.(2) We study relay-selection system and relay-selection algorithms:①in order to implement the trade-off between diversity gain and bandwidth efficiency for AF relay selection system, we propose S-D SNR threshold-ruler relay-selection algorithm and S-R-D /S-D SNR threshold-ruler relay-selection algorithm, and derive the average SEP and bandwidth efficiency expressions respectively. We also analyse different threshold value how to impact on the performance of SEP and bandwidth efficiency. Theoretics and simulation show that,as for S-D SNR ruler, raising diversity gains can be resolved by decrease threshold value, at the same time enhancing bandwidth efficiency can be resolved by increasing threshold value. As for S-R-D/S-D ruler, we can achieve the better diversity gains by decreasing threshold value when channel is under low SNR condition, and we can also enhance bandwidth efficiency by increasing threshold value when channel is under high SNR condition. So we can implement trade-off between diversity gain and bandwidth efficiency. Not only can we solve the problem of "Whom to cooperation"but also solve the problem of "When to cooperation".②As for multi-source multi-relay cooperation with centralized scheduling, we study sub-optimal or Kth optimal cooperative models with Order Statistic, and derive PDF,CDF and MGF expressions of End-to-End SNR, then apply these expressions to derive the Average SEP and the Average SNR gains for non-idential distributed parameters.(3) We investigate the model of joint multi-user diversity and cooperative diversity for i.i.d Rayleigh fading channel, and derive the average SEP and the average channel capacity expressions. Then we analyse the number of users K how to impact on cooperative system. Theoretics and simulation show that,as for single-relay multi-user AF cooperation,, the diversity order is no longer 2 but 2K for joint multi-user diversity and cooperative diversity system. That is to say that diversity-order double increases along with the number of users K.更多还原