CIF: Small: From Fano to Shannon: Information Theory and Broadband Matching

Research on multiple-input, multiple-output (MIMO) communications has shown that using multiple antennas can dramatically improve the capacity of wireless channels. Since wireless devices are limited in size, using multiple antennas often requires close spacing which leads to coupling among the antennas. Coupling can profoundly impact the received power, diversity and system capacity. Moreover, this impact depends not only on the antennas and how they are arranged in space; it also depends on detailed aspects of the wireless transceiver, such as the impedance matching networks that connect the antennas to the rest of the receiver front end. Current approaches to designing these networks for coupled MIMO systems are intrinsically narrowband and exhibit poor performance under broadband conditions.

This project seeks to develop a unified theoretical framework for the design of broadband wireless transceivers. The main idea is that Fano's broadband matching theory provides a characterization of physically-realizable matching networks, while Shannon's information theory provides a way to evaluate how each network could be used to communicate in the best possible way. By combining these theories, this project considers how antennas, matching networks and communications algorithms interact to determine overall system performance, and how best to jointly optimize these components.

Three main issues are addressed: new information- and decision-theoretic bounds on the performance of broadband single-antenna systems; extensions to broadband MIMO systems; and information-theoretic design criteria to jointly optimize the antennas, matching networks and communications algorithms. This work has the potential to significantly advance science and engineering by providing a more unified view of the RF front end and by developing new communications and matching techniques that may significantly improve wireless performance.