TITLE: Rethinking wireless networks with network coding: Overhead and stability challenges
ABSTRACT: Network coding changes the traditional networking view and emerges as a unifying tool for networking and information theory. In this talk, we review the challenges of network coding for adaptation to wireless communications and address the new research problems that arise when network coding is cast in a resource allocation framework. In particular, we address the two basic challenges of protocol overhead and traffic dynamics for wireless network coding. First, we evaluate the protocol information in terms of overhead that is necessary to operate network coding in distributed wireless networks. We present an adaptive approach of joint training and channel coding for dynamic network topologies and show how the rates achievable by network coding may be limited by the protocol overhead. Then, we present the benefits of network coding in terms of queue stability for wireless communications. We focus on the broadcast erasure channels with feedback and evaluate the capacity and stability regions through the application of queue-based dynamic network coding. These results aim at extending our understanding of the fundamental communication limits and illustrate new design opportunities and trade-offs for wireless networks.
BIO: Yalin Sagduyu is a senior research scientist at Intelligent Automation Inc. (IAI) and a visiting research scientist at the University of Maryland. He received his Ph.D. from the University of Maryland and then worked as a postdoctoral fellow at Northwestern University. His research interests are in the areas of design and optimization of wireless networks, multi-user communications, network coding, network security and game theory.