CIF: Small: Limited Feedback and Information Exchange for Wireless Systems

商业蜂窝网络最近经历了数据服务需求的戏剧性增长。满足这一需求将需要大幅(数量级)提高这种网络所能实现的频谱效率。与目前的系统相比，这将需要网络中节点之间更广泛的信息交流和合作。该研究基于反馈和信息交换的实用方法，研究了蜂窝网络中节点之间可能的协作方式，主要研究了在无线网络中交换和利用关于信道、干扰和服务质量的有限状态信息的技术。重点讨论了两种情况：点对点衰落链路的反馈和两个协作基站之间的信息交换。第一种方案旨在解决小区内的移动台由于衰落而施加的限制，而第二种情况则解决小区边界处的干扰所施加的限制。其中一个目标是对衰落信道的信道状态反馈和接收器状态反馈之间的权衡有基本的了解。针对两小区情况，研究了协作资源分配和协作编解码策略。重点是了解作为交换信息量的函数的合作的好处，以及跨时间、频率和空间的相关资源分配。研究人员还研究了协作编码/解码与自适应技术的组合，以进行预编码器优化，这些技术不依赖于显式交换信道测量。

Commercial cellular networks have recently experienced a dramatic increase in demand for data services. Satisfying this demand will require substantial (order-of-magnitude) increases in the spectral efficiency that can be achieved by such networks. This will require much more extensive information exchange and cooperation among nodes in the network than in current systems. This research investigates possible approaches for cooperation among nodes in a cellular network based on practical methods for feedback and information exchange.The main theme of this research is on techniques for exchanging and exploiting limited state information about channels, interference, and quality of service in a wireless network. The focus is on two scenarios: feedback for point-to-point fading links, and information exchange between two cooperative base stations. The first scenario is meant to address limits imposed by fading for mobiles well within a cell, whereas the second scenario addresses limits imposed by interference at a cell boundary. One of the goals is to gain fundamental understanding into the trade-off between channel state feedback and receiver state feedback for fading channels. For the two-cell scenario, cooperative resource allocation and cooperative coding and decoding strategies are investigated. The emphasis is on understanding the benefits of cooperation as a function of the amount of information exchanged, and on associated resource allocation across time, frequency, and space. The investigators also study the combination of cooperative coding/decoding with adaptive techniques for precoder optimization, which do not rely on the explicit exchange of channel measurements.