High capacity future cellular data networks

点击翻译按钮获取中文摘要

Demand for broadband communication services has increased over the years as newer services such a video streaming are introduced. The usage habits of Internet access users have changed as these new services have been introduced. Typically users of broadband services accessed over a fixed network (DSL or CATV) "consume" say 50 Gbytes/month. On the other hand a typical data plan for mobile data services is 1 GByte/month. These data amounts for the fixed and mobile network have been constantly increasing, but the ratio is approximately 50 and has been large for some time. With this limitation the user experience on a mobile data network will be very different from the experience in a fixed network. With advances in technology we expect the absolute data rates to increase but with the current approaches to cellular networks the ratio will remain high and the capability of wireless networks to meet user expectations will remain low. As a result the mobile data service will be viewed as a more limited service, for use as a last resort, and the great potential for mobile access will not be realized. The goal of this proposal is to narrow the above gap between transmission capabilities for fixed and mobile Internet access by devising new architectures/techniques for the design of cellular networks. There are two main ways to increase the capacity of these networks: 1) increase the spectral efficiency per base station, and 2) deploy more (denser) base stations, or small cells, deploy different architectures such as relays and manage more efficiently spectrum blocks among different technologies. This proposal concentrates on the 2nd approach. However in order to undertake such deployments successfully we need methods to automatically install and configure access points (APs) and relays. These methods entail novel approaches to the study and management of interference as a result of random positioning of APs, differing power levels, differing antenna capabilities, variations in traffic demand, or hot spot traffic characteristics. These deployments should be driven both by operators and users in a manner that the growth in wireless infrastructure is organic. However the spectrum is owned by the network operator and the operator runs the associated algorithms for a unified spectrum management, in a combination of large and small cells, after a self-configuration procedure.