Local area networks supporting speech traffic

Abstract

This paper presents a new analytical model for estimating the average queue length and the average queueing delay of data packets in a SENET-like integrated voice and data system. The analytical model also computes the mean queue lengths and average queueing delays of data packets for every number of calls allowed in the system. The analysis uses the fluid flow approximation of data queues. The equations used to represent the integrated system provide further insight into the system behavior and can be useful to similar systems. The method presented here is easy to apply, avoids the numerical difficulties of other methods in the literature, and yields estimates in very good agreement with simulation results. Also presented is a new flow control procedure that significantly improves the delay performance and considerably reduces the mean data queue length.