Linear measurements from nonlinear sensors: identifying distortion with incidental noise

Lachlan J Gunn, Andrew Allison, Derek Abbott

Research output: Contribution to journalArticleScientificpeer-review


Nonlinearity in many systems is heavily dependent on component variation and environmental factors such as temperature. This is often overcome by keeping signals close enough to the device’s operating point that it appears approximately linear. But as the signal being measured becomes larger, the deviation from linearity increases, and the device’s nonlinearity specification will be exceeded. This limits the range over which the device will produce directly useful measurements, often to far less than the device’s safe range of operation.

Although these nonlinear measurements may exceed the device’s nonlinearity specification, they nevertheless contain useful information. We have shown that the input noise of a system can provide the necessary information to determine the function relating the input signal to the output signal, allowing us to invert it numerically and recover the original signal, even if the device has been driven quite heavily into nonlinearity.

In this article, we will provide an overview the technique and results from [1]–[3] and show how it can be used to provide real-time compensation of distortion, providing a significant advantage over the state-of-the-art techniques standardized in [4] when dealing with grossly-distorting systems. We begin by showing how noise can be used to measure the nonlinearity of a system; then we discuss different models of nonlinearity that are suitable for use in real-time compensation tools and show how this can be applied in practice.
Original languageEnglish
Article number8979524
Pages (from-to)50-55
Number of pages6
Issue number1
Publication statusPublished - 3 Feb 2020
MoE publication typeA1 Journal article-refereed


Dive into the research topics of 'Linear measurements from nonlinear sensors: identifying distortion with incidental noise'. Together they form a unique fingerprint.

Cite this