Title :
Modeling VBR traffic with autoregressive Gaussian processes
Author_Institution :
Dept. of Electr. Eng., Nat. Cheng Kung Univ., Tainan, Taiwan
Abstract :
Previous studies about network traffic measurement show that today´s network traffic exhibits long-range dependence (LRD). The computation effort of generating LRD traffic is directly proportional to the length of the traces. This paper presents a traces-generating framework based on TES (transform-expand-samples) and synthetic autoregressive Gaussian processes. The proposed scheme can fit both the probability density function and the autocorrelation of the empirical traces. Besides, the computation effort of this scheme is independent of the length of the LRD traces
Keywords :
Gaussian processes; autoregressive processes; correlation methods; probability; queueing theory; signal sampling; telecommunication traffic; LRD traffic; VBR traffic modeling; autocorrelation; autoregressive Gaussian processes; long-range dependence; network traffic measurement; probability density function; queueing performance; synthetic AR Gaussian processes; traces-generating framework; transform-expand-samples; Autocorrelation; Computational modeling; Electric variables measurement; Gaussian processes; Length measurement; Probability density function; Quantum computing; Tail; Telecommunication traffic; Traffic control;
Conference_Titel :
Networks, 2000. (ICON 2000). Proceedings. IEEE International Conference on
Print_ISBN :
0-7695-0777-8
DOI :
10.1109/ICON.2000.875835
