The assessment of Low Cost Air Quality Multisensor Systems (LCAQMS) performance is a crucial issue in the Air Quality (AQ) monitoring framework. The devices calibration model is one of the most important drivers of the overall performances. As we know, on field calibration is increasingly considered as the best performing approach for air quality monitor devices. Field recorded sensor data together with co-located reference data allow to build suitable datasets that are more representative of the complexity of real world conditions. In this work a co-location experiment is presented, in which four multisensor devices are co-located with a mobile ARPAC (Campania Regional Agency for Environmental Protection) reference analyzer station. Two types of calibration models, linear and nonlinear have been tested on the recorded datasets, in order to determine the best calibration strategy to use to optimize the calibration procedure time in the real world operative phase. The results show that for pervasive AQ scenario, a reasonable choice is provided by a multilinear approach during one-week short co-location period.
Optimal Field Calibration of Multiple IoT Low Cost Air Quality Monitors: Setup and Results
Esposito E.;D'Elia G.;Ferlito S.;Del Giudice A.;Fattoruso G.;De Vito S.;Di Francia G.
2020-01-01
Abstract
The assessment of Low Cost Air Quality Multisensor Systems (LCAQMS) performance is a crucial issue in the Air Quality (AQ) monitoring framework. The devices calibration model is one of the most important drivers of the overall performances. As we know, on field calibration is increasingly considered as the best performing approach for air quality monitor devices. Field recorded sensor data together with co-located reference data allow to build suitable datasets that are more representative of the complexity of real world conditions. In this work a co-location experiment is presented, in which four multisensor devices are co-located with a mobile ARPAC (Campania Regional Agency for Environmental Protection) reference analyzer station. Two types of calibration models, linear and nonlinear have been tested on the recorded datasets, in order to determine the best calibration strategy to use to optimize the calibration procedure time in the real world operative phase. The results show that for pervasive AQ scenario, a reasonable choice is provided by a multilinear approach during one-week short co-location period.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
