The Department of Computer Science at the University of Cyprus cordially invites you to the Seminar entitled:
On the Characterization of Network Traffic Dynamics
Speaker: Angelos Marnerides
Traffic characterization poses a significant challenge due to the unstable and fluctuated behavior exhibited by network traffic dynamics. Modeling techniques developed for volume-based traffic profiling (i.e., based on packet and/or byte counts) rely on the statistical assumptions of stationarity, Gaussianity and linearity, whose applicability has not been thoroughly investigated throughout past and recent work. We argue that these modeling assumptions should be rigorously validated since they determine the accuracy of any model applied to describe the traffic process. In this work we introduce and illustrate the suitability of Time- Frequency (TF) representations and the Hinich algorithms for the validation of modeling assumptions on captured backbone and edge link network traces. Through the employment of these algorithms, we statistically show that link traffic is extremely dynamic and model characteristics change in small timescales. In addition, we propose the bispectrum as a good candidate for volume-based traffic profiling since it can accurately capture traffic dynamics. Furthermore, we highlight the benefits of a transport-layer traffic decomposition approach where protocols are modeled independently and protocol-specific characteristics are revealed, as opposed to analysis based solely on volume aggregates.
Angelos Marnerides received the BSc degree in Computer Science from Middlesex University, in London, in 2006 and the MSc degree in Critical Software Engineering from the University of Lancaster, UK, in 2007. He is currently a PhD candidate, funded by the UK Engineering and Physical Sciences Research Council (EPSRC) at the University of Lancaster, UK and is expected to earn his doctorate in computer science, in summer 2011. His research is mainly dealing with the characterization of traffic dynamics and anomaly detection on backbone and edge networks using statistical, information-theoretic and signal processing techniques. During his PhD, Angelos has been involved with 2 EU Projects (ANA, Resumenet) as well as with the India-UK Centre of Excellence in Next Generation Networks (IU-ATC).
|Other Presentations Web: https://www.cs.ucy.ac.cy/colloquium/presentations.php|
|Colloquia Web: https://www.cs.ucy.ac.cy/colloquium/|