Delay-Constrained Throughput in Wireless Networks: Providing Reliable services over Unreliable Channels

Wireless networks are increasingly used to carry traffic of delay-constrained applications. Such applications include VoIP, video streaming, real-time monitoring, and network control. Delay-constrained applications usually generate traffic that has strict delay bound and timely-throughput requirement. Therefore, most current mechanisms, which only provide "best-effort" services, are not suitable for such applications. On the other hand, the unreliable and time-varying nature of wireless links introduces new challenges for using wireless networks to serve delay-constrained applications. In this talk, I propose a rich analytical model that jointly considers the application specifications and the unreliable nature of wireless links. Based on this model, I introduce solutions to three important problems for serving delay-constrained applications: admission control, scheduling, and utility maximization. I propose an admission control policy that characterizes when demands of clients are feasible under specified wireless limitations. I also introduce on-line scheduling policies that fulfill the demands of all feasible systems. Finally, I discuss policies for achieving utility maximization when the timely-throughput requirements are elastic.
I-Hong Hou is a Ph.D. candidate in Computer Science Department of University of Illinois, Urbana-Champaign. He has received his B.S. degree in the Department of Electrical Engineering, National Taiwan University. His research interests include wireless networks, wireless sensor networks, real-time systems, distributed systems, and vehicular ad hoc networks.