DocumentCode
3698729
Title
Features of information flows in the backbone Internet-channel: the analysis of the statistical characteristics of the relationship between the number of packets and the time
Author
S.V. Porshnev;A.S. Koposov;D.A. Bozhalkin
Author_Institution
RTF, «
fYear
2015
Firstpage
437
Lastpage
440
Abstract
The flows of traffic dumps of high-speed Internet backbone channel were analyzed. Streams were classified into three groups according to the amount of transmitted information. Density function was calculated for the number of packets transmitted by different classes of flows (time series) according to the method of image sources and the Rosenblatt-Parzen approximation. The obtained results show non-stationarity of considered time series.
Keywords
"Approximation methods","Internet","Mice","Time series analysis","Probability density function","Data mining"
Publisher
ieee
Conference_Titel
Application of Information and Communication Technologies (AICT), 2015 9th International Conference on
Print_ISBN
978-1-4673-6855-1
Type
conf
DOI
10.1109/ICAICT.2015.7338596
Filename
7338596
Link To Document