ESTIMATING TRAFFIC PARAMETERS IN QUEUING-SYSTEMS WITH LOCAL INFORMATION

In ATM networks a new connection is accepted if the CAC (call acceptan ce controller) at the entering node decides that the network has the r equisite resources for the incoming traffic. For this, the CAC will ne ed to monitor the network congestion based on the local information av ailable to it. In this paper we propose several schemes that the CAC c an use to estimate the traffic of the users on the various routes in t he network by sending a 'probing' stream. First we consider a single n ode where the probing stream could be arbitrary while the total interf ering traffic is a Poisson process. Knowing the mean service time of t he interfering traffic and the delays experienced by the probing strea m we provide algorithms to estimate the arrival rate of the interferin g traffic. We model the node as a continuous time as well as a discret e time queue. Then we generalize the assumptions on the interfering tr affic to make the overall traffic at the node a general stationary erg odic stream. For the Poisson interfering traffic we also consider a ta ndem of queues. Finally, for a product form Kelly network we provide e stimates of the arrival rates of interfering traffic and the service r ates of customers on the route of the probing stream of the CAC. These estimates are also used to provide the optimal 'reshaping' of the tra ffic stream of a user based on the delays experienced by the user in t he network. (C) 1998 Elsevier Science B.V.