TRAFFIC CONTROL IN ATM NETWORKS

This paper presents a framework to study the problem of traffic contro l in ATM networks. Bursty traffic sources are modelled by Interrupted Bernoulli Processes. Probability of violation and dimensioning issues are addressed for the leaky bucket mechanism and the (L, M, T) mechani sm, proposed by the authors. We present two schemes to demonstrate tha t improved statistical multiplexing can be achieved if the traffic con trol mechanisms are not used in isolation for each session, rather the y are used on groups of sessions. In the first scheme, traffic control is performed on the aggregate traffic of a group. In the second schem e, traffic control is performed on single sessions with information ab out the status of all the sessions of the group being available and us ed at each traffic control mechanism. Numerical examples that illustra te the improvement in statistical multiplexing as a result of the abov e schemes are presented.