Application of information theory in wireless networks relying on a cross layer approach

未来的无线系统有望以更低的成本支持更高的比特率，这对频谱的使用施加了巨大的压力。多输入多输出（MIMO）天线系统被广泛认为是实现未来无线网络频谱效率提高的关键解决方案。近年来，学界对MIMO无线网络的信息论方面的研究引起了极大的关注。网络信息论将香农点对点（两台终端）通信推广到两台以上终端的系统。这个通用框架允许考虑多个源和/或多个信道的传输。多年来，这一主题的理论研究已经显示出实现比传统点对点通信技术更高收益的潜力。然而，需要付出巨大的努力来提高我们对这些网络的理解，以弥合这一新兴领域中目前存在的理论与实践之间的巨大差距。除了巨大的计算复杂性（必须在资源有限的移动设备上部分实现）之外，还对底层信道和（快/慢衰落）以及发射器/接收器上信道状态信息的可用性做出了许多相当不切实际的假设，这些假设应该更好地理解并根据实践进行调整。此外，传统的通信网络具有许多抽象级别，其中协议层被设计为单独的模块。最近，通信网络中的跨层优化被广泛认为是提高网络性能的一种有吸引力的方法。提出的研究将解决与上述概念在未来一代无线网络设计中的实际应用相关的几个问题，依赖于跨层优化。这涉及到多路复用、合作传输、研究这些网络中的基本权衡、多用户分集、分散网络和干扰管理/校准等问题。

Future wireless systems are expected to support much higher bit rates at a lower cost, which imposes tremendous pressure on the usage of spectrum. Multiple Input-Multiple Output (MIMO) antenna systems are widely recognized as the key solution to realize the increase in the spectral efficiency required in future wireless networks. In the recent years, research community has paid significant attention to study the Information Theoretic aspects of MIMO wireless networks. Network information theory generalizes Shannon point-to-point (two terminals) communication to systems with more than two terminals. This general framework allows considering transmission of more than one source, and/or over more than one channel. For many years, theoretical studies in this subject have shown potential for realizing high gains over conventional point-to-point communication techniques. However significant effort is needed to improve our understanding of these networks towards bridging the huge gap that currently exists between theory and practice in this emerging field. In addition to the huge computational complexity (which has to be implemented in part at the resource limited mobile unit), many rather unrealistic assumptions are made about the underlying channel and (fast/slow fading) and availability of the channel state information at the transmitters/receivers which should be better understood and tailored to practice. In addition, traditional communication networks have many levels of abstraction where protocol layers are designed as separate modules. More recently, cross layer optimization in communication networks has been widely considered as an attractive approach to improve network performance. The proposed research will address several problems related to the practical application of the above concepts to the design of the future generation of wireless networks, relying on a cross layer optimization. This involves problems in Multiplexing, Co-operative Transmission, studying the fundamental tradeoffs in these networks, Multi-user Diversity, Decentralized networking, and interference management/alignment.