基于经典多跳通信子网系统的协作频谱共享

Radio spectrum is a scarce and precious resource. With the exponential growth in the number of users and the immense demand for more capable wireless systems and services, the contradictory between the scarcity in the radio spectrum and the inherent inefficiency of current spectrum allocation policies has spurred much research in the last decade, giving birth to the notion of cognitive radios.In traditional interweave techniques, the cognitive radio keeps sensing the spectrum and fit in if spectrum holes are available. However, when there is a mis-detection, collision occurs between licensed and cognitive users. In the alternative underlay techniques, although the cognitive radio is allowed to operate in the same spectrum simultaneously with the licensed user by controlling its power, performance degradation is inevitable for the licensed user. To achieve more flexible spectrum sharing and meanwhile better protect the licensed user, we focus on an overlay model in this proposal. By exploiting the awareness of the radio environment, e.g. knowledge of the channel state information and licensed user's codebooks and messages, the cognitive radio is able to access the spectrum simultaneously with the licensed user by serving as a cooperative relay, i.e. cooperative spectrum sharing. Thus even in urban areas where rare spectrum holes limit the potential of the interweave model, cognitive spectrum access is still possible through this overlay technique, meanwhile the performance of the licensed user is well protected due to the cooperation from the cognitive radio. Thus a "win-win" situation is created for both licensed and cognitive users. However, most current works about cooperative spectrum sharing focus on one-way/two-way relay channels. To achieve the full potential of this technique, we investigate how to implement cooperative spectrum sharing on more complex network topologies in this project, e.g. interference channels with a cognitive relay, two-path relay channels, multiple-access relay channels and etc. Studies on these canonical multi-hop networks will provide valuable insights into the applications of cognitive radio in more complex network scenarios. To demonstrate the feasibility of the proposed spectrum sharing protocols, the respective performance of cognitive and licensed users will be extensively evaluated. Meanwhile, in order to mitigate the mutual interference between cognitive and licensed users, coding strategies like dirty-paper coding will be adopted. The joint design of cooperative spectrum sharing and interference management will provide significant benefits in both spectral efficiency and power efficiency of the whole system.

本项目着重研究一类通过协作技术实现不同用户之间频谱共享的认知无线电模型。与传统的侦听-躲避及干扰温度模型中认知用户总是不可避免的对授权用户造成有害干扰相比，协作频谱共享技术能够通过协作补偿使认知用户和授权用户在同一频带内操作，同时提高授权用户的性能。但是现有的工作大都仍停留在传统的单/双向中继信道上。为了充分发挥协作的潜力，在本项目中我们将重点研究如何在一些更为复杂的网络拓扑结构中实现用户之间的频谱共享，如有中继的干扰信道、双路径中继信道等。通过研究这些经典的通信子网，宏观上我们能够得到构建协作频谱共享系统的通用准则。微观上，我们将对不同用户的性能进行精确量化以充分证明所提出方案的可行性。除此之外，对于用户之间在建立协作的同时所带来的干扰，我们将采用干扰消除、脏纸编码等技术进行抑制。在本项目中,通过在协作频谱共享的框架内引入干扰管理机制，我们尝试在提高频谱效率的同时进一步提高系统的能效。

为了缓解有限的无线电频谱资源与海量的移动设备及数据业务之间的矛盾，该项目重点研究了一类通过协作中继转发技术实现不同用户之间共享无线电频谱资源的认知无线电模型。通过建立授权用户与认知用户之间的协作传输，如认知用户帮助转发授权用户信号的同时传输自己的信号，并采用叠加编码、功率分配、分集、干扰消除等先进的信号处理技术，能够在保证授权用户达到既定系统性能的情况下使授权用户与认知用户同时接入同一频带。具体的，我们研究了有中继的干扰信道、双路径中继信道、多路径中继信道等多用户无线通信系统中的典型协作通信系统模型，基于这些模型建立了不同用户之间的协作频谱共享，并对不同用户及整个系统的性能进行了分析, 如中断概率，吞吐量等。实验结果表明，在合适的系统配置及参数条件下，我们提出的协作频谱共享认知无线电模型能够在充分保护授权用户性能的基础上优化认知用户的性能，从而显著提高授权频带的频谱效率，实现“双赢”。除此之外，利用自适应功率控制、干扰消除等技术，可以进一步降低能耗，提升系统的能量效率。该项目及其成果为下一代无线通信系统提供了理论依据和解决思路。