Radio Resource Allocation for Multi User Spectrum Sharing with Parameter Uncertainty

频谱共享（SpS）技术自2002年被美国联邦通信委员会（FCC）引入作为替代频谱管理策略以来一直在持续发展。无线电资源分配（RRA）是SpS技术中的关键功能，其将可用的无线电频谱分配给次级用户，并且以有效的方式调整传输参数（诸如次级用户发射功率和速率），使得系统性能最大化。RRM在次级网络中实现，并且需要解决以下重大技术挑战。RRA算法的输入参数包括频谱可用性和信道增益。频谱可用性评估使用频谱感测机制具有一定的精度水平。所需的信道增益也使用具有给定水平的估计误差的信号处理技术来评估。主要挑战之一是，在大多数情况下，在主网络和辅网络之间没有信令或信令非常有限。这使得参数估计成为具有有限精度水平的非常复杂的任务，特别是对于需要在主网络中测量/估计的那些参数。不确定的参数可能会导致效率降低，或不可行的RRA算法，设计假设完美的参数。还需要解决多个次级用户之间的高效和自适应无线电资源分配。其他挑战包括分布式实现以及开发RRA算法性能评估的分析方法。RRA功能的性能分析揭示了基本的性能界限，如最大可实现的频谱利用率，缩放属性的次要用户的数量以及其他系统的权衡。申请人的建议旨在理论研究参数不确定性对多用户辅助服务频谱共享系统中的无线电资源分配效率的影响。该建议还包括基于不同于许多常规研究的理论研究的新型RRA技术的开发，其中还考虑了实际问题，包括对不确定输入参数的鲁棒性、最小化系统间和系统内信令开销的水平以及多个次级服务用户的影响。本研究提案的总体目标是：i）开发一种分析框架，其将物理和更高层功能结合到多次级用户环境中的系统模型中，并且能够对参数中的不确定性对网络性能的影响进行建模，ii）基于该分析框架开发实用的RRA算法以管理参数不确定性的影响，以及iii）获得具有不确定参数的多个次要服务的基本性能极限。申请人的最终目标是向公众提供SpS技术。

Spectrum sharing (SpS) technology has continued to develop since its introduction as an alternative spectrum management policy by the US's Federal Communication Commission (FCC) in 2002. Radio resource allocation (RRA) is a crucial function in SpS technology which allocates the available radio spectrum to the secondary users and adjusts the transmission parameters such as secondary user transmit power and rate in an efficient way so that the system performance is maximised. RRM is implemented in the secondary network and needs to address the following significant technical challenges. Input parameters to RRA algorithms include spectrum availability and channel gains. Spectrum availability is evaluated using a spectrum sensing mechanism with a certain level of accuracy. The required channel gains are also evaluated using signal processing techniques with a given level of estimation errors. One of the main challenges is that in most cases there is either no or very limited signalling between the primary and secondary networks. This makes parameter estimation a very complicated task with a limited level of accuracy, particularly for those parameters which needed to be measured/estimated in the primary network. Uncertain parameters may result in either decreasing the efficiency, or infeasibility of an RRA algorithm that is designed assuming perfect parameters. It is also required to address efficient and adaptive radio resource allocation among multiple secondary users. Other challenges include distributed implementations as well as developing analytical methods for performance evaluations of RRA algorithms. Performance analysis of RRA function sheds light on the fundamental performance bounds such as maximum achievable spectrum utilization, scaling property regarding the number of secondary users as well as other system trade-offs. Addressing these challenges is required for efficient adoption of SpS technology in full scale commercial wireless communication systems.The applicant's proposal aims at the theoretical investigation of the impact of parameter uncertainty on the efficiency of the radio resource allocation in multi user secondary service spectrum sharing systems. This proposal also includes development of novel RRA techniques based on the theoretical investigations that differs from many conventional studies, in that practical concerns including robustness to uncertain input parameters, minimizing the level of inter and intra system signalling overhead and impact of multiple secondary service users are also taken into consideration. The overall aims of this research proposal are: i) to develop an analytical framework which incorporates physical and higher layer functionalities into the system model in a multi secondary user environment and is able to model the impact of uncertainty in the parameters on the network performance, ii) to develop practical RRA algorithms based on this analytical framework to manage the impact of parameter uncertainty, and iii) to obtain fundamental performance limits for the multiple secondary service with uncertain parameters. The applicant's ultimate goal is to make SpS technology available to general public.