Proposal of an estimator of the Hurst parameter for a self-similar process representative of the degree of randomness of the recorded traffic in IEEE 802.3-2005 networks

Abstract

An hypothesis for the existence of a process with long term memory structure, that represents the independence between the degree of randomness of the traffic generated by the sources and the pattern of traffic stream exhibited by the network is presented, discussed and developed. This methodology is offered as a new and alternative way of approaching the estimation of performance and the design of computer networks ruled by the standard IEEE 802.3-2005.

Development of a Modified Method of Network Traffic Forming

“…Self-similar traffic requires more memory resources of buffer devices than the classic M/M/1 model, which is considered the least favorable compared to others (for example, with a constant or Gaussian service time distribution). The rate of growth of the required amount of memory increases with the increase of the Hurst parameter, which is mainly due to the degree of the grouping of homogeneous packets and bursts of network load [10,11].…”

A flow control algorithm for high-speed computer networks

“…En [6] el problema de diseño del control se plantea como un problema estándar de rechazo de perturbación donde el ancho de banda disponible actúa como perturbación para el sistema. En [7] el problema se formula en términos de problema de control estocástico en el cual la perturbación se modela como proceso autoregresivo; el nodo en cuestión tiene que estimar este proceso usando mínimos cuadrados recursivos. En [8] se usa el principio de Smith para derivar un control en el caso de que un búfer FIFO se mantenga en los enlaces de salida.…”