Cooperation and Cognition in Wireless NetworksDate: 2008-04-24
Location: Holmes 389
Speaker: Speaker: Dr. Ivana Maric, Stanford University
The key to high-performance wireless network design is managing the interferencebetween users. Many decades of research on the fundamental capacity limits of these networks has failed to uncover optimal strategies for mitigating or exploiting this interference. In this talk we propose novel cooperative and cognitive methods between network nodes that exploit interference to increase capacity. We first consider a multiuser network with an intermediate relay. We investigate generalized relay strategies whereby the relay performs joint encoding of the messages it receives. We also propose "interference-forwarding", where the relay magnifies interference at a given receiver to assist in interference cancellation. We show that both of these strategies provide significant performance improvement in terms of the network capacity region. We also consider scenarios in which one ``cognitive'' user overhears the other user's message. We propose a new communication paradigm to exploit this information that is capacity-achieving under certain channel conditions. Our results offer innovative ideas to dramatically improve performance in wireless networks as well as guidelines for new wireless networking paradigms that improve efficiency of spectrum usage in both licensed and unlicensed frequency bands.
Ivana Maric received her B.S. degree from the University of Novi Sad, Serbia. She finished her M.S and Ph.D. in the Wireless Network Information Laboratory (WINLAB), Rutgers University in 2000 and 2006, respectively. She was a summer intern at ATT Research Labs in 1998. She is currently a postdoctoral scholar at Stanford University. Her research focuses on network information theory and wireless communications.