April 2, 2007
EE Conference Room, 1312 MUDD
Speaker: Akbar M. Sayeed, University of Wisconsin-Madison
Multipath signal propagation, a salient feature of wireless channels, results in detrimental signal fading and yet it is also a key source of diversity to combat the effects of fading and to increase the capacity and reliability of wireless communication systems. Technological advances in wideband, frequency-agile RF front-ends and reconfigurable antenna arrays are enabling new modes for sensing and adapting to multipath in time,frequency and space. However, our current understanding of the design and fundamental performance limits of such cognitive wireless transceivers is still in its infancy. One emerging insight is that physical channels exhibit a sparse multipath structure as the dimension of the spatio-temporal signal space increases. I will present a model for sparse multipath channels and discuss two examples of exploiting multipath sparsity with reconfigurable transceivers. First, I will discuss optimum signal spreading in time and frequency in the context of channel learning, capacity and reliability in the wideband regime. Second, I will discuss optimum antenna array configurations for maximizing capacity at any operating SNR. These results reveal new tradeoffs in sparse multipath channels that can be exploited by reconfigurable transceivers for significant performance gains over fixed transceivers.