KOLMOGOROV-WIENER FILTER FOR CONTINUOUS TRAFFIC PREDICTION IN THE GFSD MODEL

Context. We investigate the Kolmogorov-Wiener filter weight function for prediction of continuous stationary telecommunication traffic in GFSD (Gaussian fractional sum-difference) model.
 Objective. The aim work is to obtain an approximate solution corresponding and illustrate convergence truncated polynomial expansion method used this paper.
 Method. obtaining KolmogorovWiener under consideration. In paper we on basis Chebyshev polynomials first kind orthogonal time interval which input data are given. It expected that results based other sets will be similar obtained Results. investigated approximations up eighteen-polynomial one. shown rather large numbers lead a good coincidence left-hand side right-hand Wiener-Hopf integral equation. quality illustrated by calculation MAPE errors.
 Conclusions. devoted theoretical construction model. correlation framework model positively defined one, guarantees method. errors misalignment equation may applied practical systems with packet transfer.