SSP'05 -- Proceedings

	IEEE/SP 13th workshop on Statistical Signal Processing July, 17-20, 2005 - Bordeaux - France

Information regarding the paper

Title

Statistical Analysis of Local Features in Network Traffic Processes

Author(s)

Giada Giorgi	Department of Information Engineering - Università di Padova
Claudio Narduzzi	Department of Information Engineering - Università di Padova
Paolo Attilio Pegoraro	Department of Information Engineering - Università di Padova

Get the paper in PDF format

To obtain Acrobat Reader (version 5 minimum required) necessary to his read.

Abstract

This work introduces an approach to the analysis of network traffic traces and their scaling characteristics, based on the measurement of maximal rate envelopes over short time intervals. The method can be employed both to estimate traffic parameters at large time scales and to detect, on a reasonably fine-grained scale, local features that may affect the outcome of scaling analysis. As such, the approach complements wavelet algorithms and should prove useful particularly in the understanding of short time scale phenomena in network traffic.