Recent research on multiple-input multiple-output (MIMO) communications has shown that deploying arrays at the transmitter and receiver can dramatically improve the capacity of wireless multipath channels. Since the physical size of a transceiver is often limited, increasing the number of array elements often requires closer inter-element spacing and leads to signal correlation and mutual coupling. Coupling can profoundly impact the received power, diversity and system capacity. Moreover, this impact depends essentially on aspects of the transceiver design, such as antenna matching and the dominant sources of noise.
Intellectual Merit: This project seeks to develop a systems-level perspective on the design of compact array transceivers for wireless communications. The aim is to understand how antennas, matching networks, amplifiers and communications algorithms interact to determine overall performance, and to jointly optimize the design of these interacting subsystems. Three issues are addressed: (1) channel models which incorporate diverse noise sources, transceiver design and interference from other users for both narrowband and broadband channels; (2) the impact of different noise sources and propagation environments on the fundamental performance limits of coupled MIMO systems, as well as on performance of specific diversity and multiplexing techniques; (3) information-theoretic design criteria to jointly optimize the array, matching, amplifiers and communications algorithms.
Broader Impacts: This multi-disciplinary project combines theoretical studies with experiments using an antenna testbed. The mix of theory and hardware demonstrations will provide opportunities for student participation at all levels. 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 models, communications algorithms and matching techniques which may significantly improve wireless performance.
The goal of this project was to develop a systems-level approach to the analysis and design of compact multiple-antenna transceivers for wireless communications. This approach seeks to understand how antennas, matching networks, amplifiers, termination and communications algorithms interact to determine overall system performance, and to determine how best to jointly optimize the design of these interacting subsystems. Three main issues were addressed: (1) channel models that incorporate diverse noise sources, transmitter design, the propagation environment and interference from other users for both narrow and broadband channels; (2) a study of the impact of different noise sources and propagation environments on the fundamental performance limits of coupled multi-antenna systems as well as on performance of specific diversity, spatial multiplexing and space-time coding techniques; (3) information-theoretic design criteria to jointly optimize the array, matching networks, amplifiers and communications algorithms. All of these issues were addressed from a unified systems-level perspective that reflects the combined impact of the communications algorithms, antennas and propagation, and antenna matching techniques. This project has contributed to the theory and practice of wireless communications in several important ways. First, a new multi-antenna wireless transceiver model was introduced that identifies the dominant physical noise sources and relates their spatial correlation to the properties of the antennas, front-end amplifiers and matching networks. Second, this model was used to analyze the performance of communications and matching techniques in the presence of mutual coupling, and to develop new techniques that are optimized for these conditions. This work included a study of the impact of correlation, mutual coupling and matching on the performance of the multi-input, multi-output beamforming techniques adopted in recent wireless standards, such as WiMAX (IEEE 802.16e) and Wi-Fi (IEEE 802.11n). Third, new low-noise design principles for multi-antenna RF front ends were developed based on information theory. This work revealed that one optimal matching network exists that simultaneously optimizes a wide range of information-theoretic/communications metrics. It further shows that the damaging effects of antenna coupling and correlation can often be mitigated by redesigning the RF front end and using optimum receiver processing. All of these results contribute to an improved understanding of the aspects of RF front end design that affect communications performance, as well as providing communications engineers with the tools needed to strike a better tradeoff among power, bandwidth and data-rate for a given transceiver size.
