Inherent Flexibility of Mathematical Splines for Pulse Shape Reconstruction

Abstract

Effective retrieval of information from digitised pulse shape data depends on a good data fitting procedure. Polynomial approximation, although easy to compute, results in curve oscillation and interpolation when high order polynomials are used. In addition, the analyticity of polynomials introduces a major drawback of global dependence on local properties in the interval of approximation. The use of mathematical splines helps to over-come these problems. A quartic spline of order 5 with 6 knots has been used to reconstruct digitised Cerenkov light pulse shapes. The pulse shape parameters measured are consistent with measurements from other experiments.