Measurement and Estimation of Quality of Service, Usage, and Path Sharing in Data Networks

Mr. Dogu Arifler

Ph.D. Student
Embedded Signal Processing Laboratory (ESPL)
Electrical and Computer Engineering
The University of Texas at Austin
Austin, TX 78712, USA

Monday, November 25th, 12:00 PM, ENS 637

arifler@ece.utexas.edu

This presentation is the open-to-public part of Mr. Arifler's Ph.D. Qualifying Examination

Abstract

Developing effective methods to measure network statistics has become critical as a result of an increasing need for monitoring and charging for data services. In this proposal, I present a number of challenging research problems that may significantly impact the ways content and service providers design and operate their networks. The proposed reseach aims to make the following contributions:
(1) Inferring which flows share paths or bottlenecks by applying multivariate statistical methods to local flow measurements collected at a monitoring agent.
(2) Estimating the average number of users and the QoS perceived by users of streamed content services on large-scale multicast networks.
(3) Developing better usage estimates for best-effort flows in wireless networks.

Biography

Dogu Arifler received his B.S.E.E. and M.S. degrees from The University of Texas at Austin in 1997 and 1999, respectively. He worked as a full-time software engineer at National Instruments from 1999 until 2000. He is currently pursuing his Ph.D. degree in the Department of Electrical and Computer Engineering, The University of Texas at Austin. His primary research interests are network measurements and statistical signal processing applications in networking. He is a member of the Institute of Electrical and Electronics Engineers (IEEE) and Sigma Xi (The Scientific Research Society).

A list of Wireless Networking and Communications Seminars is available at from the ECE department Web pages under "Seminars". The Web address for the Wireless Networking and Communications Seminars is http://signal.ece.utexas.edu/seminars