Summary: PREDICTIVE MODELING OF VIDEO PACKET DELAY IN IP NETWORKS
Ali C. Begen
, Mehmet A. Begen
and Yucel Altunbasak

School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332­0250 USA

Department of Operations and Logistics, University of British Columbia, Vancouver, BC V6T1Z2 Canada
acbegen@ece.gatech.edu, mehmet.begen@sauder.ubc.ca, yucel@ece.gatech.edu
ABSTRACT
This paper studies linear prediction algorithms for packet-delay
modeling. A detailed examination of the delay traces collected
from video streams encoded at different bitrates, suggests that
autoregressive (AR) models can exploit the correlation among
the delay samples and produce the best estimates in terms of
the mean-squared error criterion. Simulation results show that
AR prediction can reduce the average prediction-error power
significantly as compared to the exponentially-weighted moving
average prediction as well as the recursive weighted median filtering.
This is a promising result since many layers in the multimedia

Source: Altunbasak, Yucel - School of Electrical and Computer Engineering, Georgia Institute of Technology

Collections: Engineering