Adam Cooman

Distortion Contribution Analysis
with the Best Linear Approximation

Adam Cooman, Piet Bronders, Dries Peumans,
Gerd Vandersteen and Yves Rolain

A Distortion Contribution Analysis (DCA) obtains the distortion at the output of an analog electronic circuit as a sum of distortion contributions of its sub-circuits. Similar to a noise analysis, a DCA helps a designer to pinpoint the actual source of the distortion. Classically, the DCA uses the Volterra theory to model the circuit and its sub-circuits. This DCA has been proven useful for small circuits or heavily simplified examples. In more complex circuits however, the amount of contributions increases quickly, making the interpretation of the results difficult. In this paper, the Best Linear Approximation (BLA) is used to perform the DCA instead. The BLA represents the behaviour of a sub-circuit as a linear circuit with the unmodelled distortion represented by a noise source. Combining the BLA with a classical noise analysis yields a DCA which is simple to understand, yet capable to handle complex excitation signals and complex strongly non-linear circuits.

This page contains the full text of the accepted version of the paper published in IEEE Transactions on Circuits and Systems I: Regular Papers. The content was modified slightly to better fit the format of this website

This work was sponsored in part by the Fund for Scientific Research (FWO-Vlaanderen), in part by the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen), in part by the Flemish Government (Methusalem), in part by the Belgian Federal Government (IUAP VII), and in part by the Strategic Research Program of the VUB (SRP-19). The authors are with the Department of fundamental Electricity and Instrumentation (ELEC), Vrije Universiteit Brussel, 1050 Brussels, Belgium.

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17