The modeling of an access multiplexer in an ATM network and the characterization of the cell loss behaviour have received a large attention recently. In fact this is a fmt necessary step to understand what are the relevant parameters for the traffic control in an ATM environment. In particular, we considex the superposition of heterogeneous ON-OFF sources, both because they are significant representatives of real-life bursty emission processes and for lhcir amenability to some analytical investigation. For the last purpose. we adopt the stochastic fluid flow approach. The focus of the paper is to discuss the extent to which the long-term average cell loss probability is a significant performance parameter and/or what else should be checked for. Both simulations and analytical results are examined in order to study the cell loss behaviour of the multiplexer, in a long-term average perspective as well as looking at the temporal behaviour.
On the significant parameters for the characterization of the cell loss behaviour in ATM multiplexing / Baiocchi, Andrea; N., BLÉFARI MELAZZI; F., Salvatore. - 2:(1992), pp. 682-688. (Intervento presentato al convegno IEEE ICC'92 tenutosi a Chicago, IL, USA nel 14-17 giugno) [10.1109/ICC.1992.268196].
On the significant parameters for the characterization of the cell loss behaviour in ATM multiplexing
BAIOCCHI, Andrea;
1992
Abstract
The modeling of an access multiplexer in an ATM network and the characterization of the cell loss behaviour have received a large attention recently. In fact this is a fmt necessary step to understand what are the relevant parameters for the traffic control in an ATM environment. In particular, we considex the superposition of heterogeneous ON-OFF sources, both because they are significant representatives of real-life bursty emission processes and for lhcir amenability to some analytical investigation. For the last purpose. we adopt the stochastic fluid flow approach. The focus of the paper is to discuss the extent to which the long-term average cell loss probability is a significant performance parameter and/or what else should be checked for. Both simulations and analytical results are examined in order to study the cell loss behaviour of the multiplexer, in a long-term average perspective as well as looking at the temporal behaviour.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.