Spline fitting in presense of uniform noise

The method of estimation using splines is basically a curve fitting technique[1] for smoothing a collection of random data. Spline fitting involves estimation of coefficients of a polynomial which fits a set of data points between ´knots´. A technique is presented in this paper for spline fitting when the data scattering is most accurately described by a non-Gaussian density function. The polynomial coefficients can be optimally estimated in the sense that the mean square error is minimized. If there are a sufficient number of data points between a pair of knots, the estimate error approaches zero. The technique can be used for real time processing of data[2]. Nonlinear fitting using Bayesian estimation is used to estimate the spline coefficients.