American Institute of Mathematics, Palo Alto, California
Robert Calderbank, George Papanicolaou, and Arogyaswami Paulraj
The growing trend in the design of communications systems is the accomodation of high data rates. This can be achieved with the use of multiple antenna technology, which involves increased transmitter/receiver device cost. The alternative is the use of large transmission bandwidths. The channels under consideration in this case are referred to as high delay spread bandwidth channels (HDSBWC), which open several exciting opportunities but pose several challenges. HDSBWC benefit from spatial focusing, high delay resolution and a great degree of frequency diversity, whereas channel estimation and equalization become more challenging tasks.
A simple transmission scheme that can exploit all the benefits and efficiently address the inherent challenges is Time reversal (TR). In TR, signals transmitted with a time reversed filter back propagate and focus tightly in space-time near the intended receiver. The space-time focusing is even more tight and robust when the propagation medium is randomly inhomogeneous. Recent experimental and theoretical studies have firmly established this remarkable effect and have motivated applications to detection, imaging and communications. More advanced prefiltering techniques can be used to optimize system performance based on a specific metric.
The space-time focusing properties of HDSBWC's can lead to a new class of wireless communications systems with significant advantages over current approaches. In the case of dedicated spectrum allocation, the spatial focusing potential of HDSBWC is an appealing quality, since service providers assume the high cost of buying the spectrum, and would therefore like to be able to accomodate as many users as possible within it. In the case of shared spectrum, interference issues become important. In both situations, the constraint is the simplification of receiver design. Potential applications include commercial wireless as well as specialized systems, such as communication systems that demand low probability of intercept.
In order to fully exploit the benefits of HDSBWC's and overcome the challenges, there is a need for accurate channel modeling and parameterization. There is also a need for equalization techniques that have reasonable complexity. Existing equalization techniques may be computationally too expensive for channels with a few hundred channel taps.
These tasks can be accomplished most effectively through the collaboration between the mathematicians, who have been addressing the issues of channel modelling in inhomogeneous channels, and the communications engineers, who deal with the limitations of hardware and algorithm design. The aim of this workshop is to explore these ideas in depth, and to bring the two disciplines together.
The workshop schedule.
A report on the workshop activities.
The workshop had several discussion sessions which raised many interesting questions.