Advances in Wireless Networks: Topology, Signal Structure, and Feedback

点击翻译按钮获取中文摘要

The public’s engagement with applications enabled by wireless communications is transforming the way in which people around the world live their lives; from the mobile banking systems, such as M-PESA, that are transforming developing economies and the enabling of “citizen journalism”, to live coverage of sporting events, video phone calls, and ubiquitous access to information. The increasing popularity of wireless applications, and the emergence of a wide variety of wireless machine-to-machine communication applications (some of which fall under the umbrella of the envisioned Internet of Things) are driving demand for wireless access beyond the capabilities of the current generation of wireless networks. Indeed, the common consensus is that by 2020 wireless networks will be seeking to serve 10-100 times as many devices as current networks, to deliver 1000 times the data traffic, to reduce the latency by a factor of 5, and to reduce the energy consumption (in joules-per-bit) by a factor of 100. It is now widely recognized that incremental enhancements of the current (fourth) generation of networks will not be able to achieve these targets and that the time has come to envision the architectures of a fifth and future generations. To address the extent of the aspirations for the capabilities of the next generation, networks are becoming more dense. This reduces the propagation losses in the links and increases the probability of line-of-sight communication, but it also reduces the propagation losses of interfering signals sent by neighbouring nodes. The loss in spectral efficiency incurred when traditional techniques for avoiding this interference are employed, and advances in the processing capabilities of nodes in the network, are driving a re-examination of the pragmatic architectural decisions that simplified the design and operation of previous generation networks. The goal of the proposed research program is to combine insights from network information theory, communication theory, signal processing, optimization and stochastic geometry to develop signaling strategies that will enable networks to dynamically manage interference. The proposed strategies will exploit the structure of interference in conventional networks and will seek to synthesize structure in networks that are more heterogeneous. This structure will reduce the signaling overhead, leaving more of the channel’s resources available for communication. The main goal of the proposed research will be the development of broadly applicable design guidelines and signaling strategies. However, it is hoped that some of the associated insight will help to shape the development of practical techniques, akin to the influence that some earlier work had on the development of the Evolved Multimedia Broadcast/Multicast Service in the 4G-LTE standard