Future wireless systems, providing high data-rate services with limited spectrum, will increase capacity by exploiting additional degrees of freedom provided by the spatial dimension. Examples of such systems are multiple-input multiple-output (MIMO) systems, where multiple antennas are employed at each end to scale the number of degrees of freedom. Further degrees of freedom can be obtained through employing various electric and magnetic field components of the electromagnetic wave. Assuming that the same configuration is employed at both the transmitter and receiver, information theoretic results for independent fading demonstrate that the system capacity scales linearly with the product of the number of antennas and the number of field components per antenna employed. This has led system designers to consider systems with a vast number of antennas where possible. However, as these systems push rapidly towards implementation, it is important to be able to analyze and design such systems for realistic propagation environments, where the amount of independence is inherently limited. Thus, under this project, the PIs will: (1) develop stochastic propagation models that include depolarization and multiple scattering of waves (2) study capacity limitations arising from element correlation and mutual coupling that results from restricting antennas to a fixed volume and realistic environment, and (3) design and analyze robust single-user and multi-user communication techniques that exploit various forms of channel state information in these environments.
The technical merit of this project is significant both in fundamental advances in the individual areas of the PIs and also in the close connection of such. The modeling techniques for MIMO systems will represent significant advances in electromagnetic modeling for this important new environment. The capacity analyses will introduce new techniques and mathematical devices that will prove of general use to the community. The techniques of design and analysis of communication systems based on the derived electromagnetic models represent a new approach to robust communication system design. Just as importantly, however, is how the connection of the two areas of the PIs furthers the goals of the project; in particular, the communications research helps define the parameters of the electromagnetic model that will have a significant impact on system design and analysis, and the electromagnetic models drive the techniques and tools required to design and analyze MIMO systems. This promising technical merit is substantiated in this proposal by the description of the proposed and recent work of the PIs - both individually and in collaboration.
The broader impact of this project will be significant in a number of ways. First, the models and techniques developed in this research will be used widely by researchers and designers of MIMO systems. In particular, the development of such models and techniques by this multidisciplinary team will guarantee both the accuracy of such and that the results are in a form that can serve the research community in the analysis and design of algorithms. Second, this project will continue the team's successful program of undergraduate research, which has sent a number of University of Massachusetts undergraduates to top graduate programs around the country, and supports two PIs with a sincere dedication to undergraduate teaching. Finally, this project will further the strong collaboration of the PIs across the boundary that separates electromagnetics from communication theory, and it will further the understanding that integrated academic research is not only needed across layers in the typical networking model but also within the physical layer itself.
