Title :
Length requirements of recorded data of network traffic
Author :
Li, Ming ; Jia, Weijia ; Au, Pui On ; Leung, Cheng
Author_Institution :
Dept. of Comput. Sci., City Univ. of Hong Kong, Kowloon, China
Abstract :
Measuring and processing real-traffic data in the real-time analysis of network performances are related to the practical issue of the required length of the traffic data with a desired accuracy so as to make the processing results reliable. This paper presents a reference relationship between the block size of the measured traffic data, the accuracy of the estimated autocorrelations and the Hurst parameter. This relationship represents reference requirements of the data length in measuring and processing traffic data
Keywords :
correlation methods; data communication; telecommunication traffic; Gaussian distribution; Hurst parameter; TCP traces; block size; estimated autocorrelation accuracy; length requirements; measured traffic data; network traffic; real-time analysis; real-traffic data measurement; real-traffic data processing; recorded data; traffic data length; Autocorrelation; Computer science; Gold; Length measurement; Performance analysis; Performance evaluation; Random processes; Real time systems; Telecommunication traffic; Traffic control;
Conference_Titel :
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on
Conference_Location :
Beijing
Print_ISBN :
0-7803-7010-4
DOI :
10.1109/ICII.2001.983551