Local spline approximation of time-varying channel model

Methods of local spline approximation of random functions described by Clar ke's model are investigated. Such a model is known to describe time-varying amplitudes of multipath components in mobile fading channels. Splines of z ero, first and second orders with different spline coefficients are compare d. The comparison shows that local linear and parabolic splines with quasi- optimal spline coefficients provide a low approximation error. This makes t hem attractive for estimating and modelling fast fading channels.