Kok, G; van Dijk, M; Harris, P; Vedurmudi, A (2025) Modelling and determining correlations in sensor networks. Measurement: Sensors, 38. 101793 ISSN 26659174

Full text not available from this repository.

Abstract

Sensor networks are nowadays widely used to measure and monitor various measurands like the concentrations of pollutants in ambient air. When a mathematical model is fitted to the measurement data, it is often assumed that the measurement errors of the sensors are independent and identically distributed (i.i.d.) and with a Gaussian distribution. In this paper we will take a closer look at this assumption using a specific dataset with measurements made by air quality sensors. We propose a space-time dependent correlation model for the sensor measurement errors and show that calculated uncertainties of aggregated and interpolated measurements can be significantly different using this model compared to the standard Gaussian i.i.d. sensor noise model.

Item Type:	Article
Keywords:	Sensor networks, Correlation analysis, Uncertainty evaluation, Gaussian processes
Subjects:	Mathematics and Scientific Computing > Measurement Uncertainties
Divisions:	Data Science
Identification number/DOI:	10.1016/j.measen.2024.101793
Last Modified:	11 May 2026 08:22
URI:	https://eprintspublications.npl.co.uk/id/eprint/10379