DocumentCode :
2937486
Title :
Traffic analysis for high speed networks
Author :
Bragg, A.W. ; Chou, Wushow
Author_Institution :
North Carolinia State Univ., Raleigh, NC, USA
fYear :
1991
fDate :
2-5 Dec 1991
Firstpage :
1421
Abstract :
A description is presented of the use of time series analysis techniques to characterize aspects of traffic patterns in high-speed wide-area communication networks. One aspect relates to the possible periodic behavior of the traffic; another aspect relates to the bursts or spikes characteristic of traffic approaching the maximum load. It is found that a time-of-day periodicity exists at several levels of network traffic and should be considered regardless of the analytical technique used to estimate traffic load. The authors have confirmed that a composite or average day model is not precise enough for modeling network traffic in their environment
Keywords :
statistical analysis; telecommunication traffic; time series; wide area networks; WAN; bursts; high-speed wide-area communication networks; maximum load; periodic behavior; spikes; statistical analysis; time series analysis techniques; time-of-day periodicity; traffic load estimation; traffic patterns; Bandwidth; Communication networks; High-speed networks; Intelligent networks; Pattern analysis; Predictive models; Telecommunication traffic; Time series analysis; Traffic control; Wide area networks;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Global Telecommunications Conference, 1991. GLOBECOM '91. 'Countdown to the New Millennium. Featuring a Mini-Theme on: Personal Communications Services
Conference_Location :
Phoenix, AZ
Print_ISBN :
0-87942-697-7
Type :
conf
DOI :
10.1109/GLOCOM.1991.188602
Filename :
188602
Link To Document :
بازگشت