CIF Small: Massive MIMO in the MM-Wave Range: The Theory of Making it Practical

University Of Southern California, Los Angeles CA

Investigators

Abstract

Fifth generation cellular communications will critically depend on making available large swaths of bandwidth in the mm-wave range. To enhance spectral efficiency, and compensate for the large pathloss of the mm-wave bands, large antenna arrays (massive MIMO) will be used at both the transmitter and receiver. This project aims to develop new techniques to reduce both the cost and energy consumption normally associated with such large arrays. Solving these problems will be an important step forward for 5G, which in turn will be a cornerstone of the presidential vision of a wireless broadband revolution in the coming years. It will allow consumers to benefit from higher data rates at lower costs, and network operators and equipment manufacturers to better leverage their investment in mm-wave technology. The focus of this work is on hybrid analog/digital transceiver structures that combine analogue beam forming based on second-order channel statistics (CSI) with digital beam forming based on instantaneous CSI, a structure that has both practical and fundamental advantages. The key point of the project is to tackle research questions that, while inspired by practical requirements, are novel and deep challenges in communication theory and signal processing. This project considers setups involving hybrid transceivers at both the base station and the user equipment, thus requiring optimization over matrices that change on different timescales, with different dimensions, and are linked over channels that do not follow the convenient, but unrealistic, simplifications commonly used. Another important practically motivated challenge is the impact of the properties of the propagation channel, which will be investigated through novel measurement and parameter extraction methods.