Methods for Estimating the Required Volume of Resource for Multiservice Access Nodes

We construct and study a mathematical model for the distribution of information transmission resource in a multiservice access node. The model considers an arbitrary number of multimedia traffic streams, which differ in the intensity of claims arrival, the amount of resource used to service one claim, and the resource occupation time. The time intervals between the arrival of claims have an exponential distribution with a parameter depending on the number of claims being serviced in the considered flow. We construct a recursive algorithm for evaluating the characteristics. Relations are established between the integral and flow characteristics of the quality of service for the claims. We construct an efficient algorithm for estimating the amount of resource required to service given traffic flows with required quality. The efficiency of the computational procedure is achieved as a result of organizing recursion over the volume of resource and the use of normalized values of state probabilities. We consider a solution to the problem of estimating the required amount of resource for a multiservice node model, which allows to use resource reservation and dynamic distribution mechanisms when servicing elastic traffic. We show numerical examples that illustrate the features of the implementation of the constructed computational procedures.