The design and performance evaluation of multimedia systems require th
e availability of adequate models to mimic the statistical properties
of the traffic generated by a multimedia source. In this paper the aut
hors propose a very simple method for the generation of discrete-time
and discrete-state autocorrelated random variables which can be used t
o model a traffic source by simulation. The probability distribution a
nd autocorrelation sequence of the variables generated exactly match t
he corresponding experimental histograms of the source to be modelled
when the experimental autocorrelation sequence is decreasing with down
ward convexity. In this paper an analytical demonstration of the metho
d proposed is given and its use is illustrated by three telecommunicat
ions examples.