REGULARISED DIFFERENTIATION OF MEASUREMENT DATA

Jakub Wagner, Paweł Mazurek, Roman Z. Morawski

Abstract

The paper is devoted to the numerical differentiation of measurement data. A new method, based on analytical differentiation of the least-squares approximation of data by means of shifted trigonometric functions, is proposed. It is compared with the most popular method of central difference and with a specialised version of the Kalman filter. Some conclusions, concerning metrological applicability of all three methods, are drawn.

Keywords

measurand reconstruction, numerical differentiation, regularisation

Download

IMEKO-WC-2015-TC21-406.pdf

DOI

-

IMEKO TC

TC21 - Mathematical Tools for Measurements

Back to the proceedings