Parameter estimation of cascades model for network traffic
Yi Zhang, Lemin Li, Bo Li · 2004
The cascades model was proposed recently to represent network traffic comprehensively. A parameter estimation procedure for the cascades model is proposed. Examples of the estimation procedure using artificially generated and real network collected traffic are given. It is pointed out that the stochastic process, which is in the cascades model, should not be the sequence of packet counts in unit time or the sequence of inter-arrival time between packets; instead, their cumulative sequences should be used.