A Comparative Study of Calibration Methods for Low-Cost Ozone Sensors in IoT Platforms

Ferrer-Cid, Pau; Barceló-Ordinas, José M.; García-Vidal, Jorge; Ripoll, Anna; Viana, Mar

doi:10.1109/jiot.2019.2929594

Cited by 38 publications

(57 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…al. [26] show that identical ozone MOX sensors behave with large variability given the same calibration model. Thus, we will first investigate the fusion of multi-sensor data between sensors of the same family.…”

Section: A Multi-sensor Data Fusion Calibration With Mox Sensorsmentioning

confidence: 89%

“…The concept is that in this context of calibration, a number of sensors (called array of sensors) participate in the calibration of the target sensor to reduce calibration errors. The sensor array principle is widely used [4], [11], [13], [17], [18], [19], [26] in the calibration of air pollution sensors, where the sensor calibration consists of measuring all cross sensitivities of the sensor array to compensate for all interfering contaminants and environmental conditions. Different calibration methods covering linear and non-linear models have been used with arrays of sensors depending on the type of contaminant and cross sensitivities.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multisensor Data Fusion Calibration in IoT Air Pollution Platforms

Ferrer-Cid

Barceló-Ordinas

García-Vidal

et al. 2020

IEEE Internet Things J.

Self Cite

View full text Add to dashboard Cite

This paper investigates the calibration of low-cost sensors for air pollution. The sensors were deployed on three IoT (Internet of Things) platforms in Spain, Austria, and Italy during the summers of 2017, 2018, and 2019. One of the biggest challenges in the operation of an IoT platform, which has a great impact on the quality of the reported pollution values, is the calibration of the sensors in an uncontrolled environment. This calibration is performed using arrays of sensors that measure cross sensitivities and therefore compensate for both interfering contaminants and environmental conditions. The paper investigates how the fusion of data taken by sensor arrays can improve the calibration process. In particular, calibration with sensor arrays, multi-sensor data fusion calibration with weighted averages, and multi-sensor data fusion calibration with machine learning models are compared. Calibration is evaluated by combining data from various sensors with linear and nonlinear regression models.

show abstract

Section: A Multi-sensor Data Fusion Calibration With Mox Sensorsmentioning

confidence: 89%

Section: Related Workmentioning

confidence: 99%

Multisensor Data Fusion Calibration in IoT Air Pollution Platforms

Ferrer-Cid

Barceló-Ordinas

García-Vidal

et al. 2020

IEEE Internet Things J.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, in some works additional measurements are used to generate specific models for certain environmental considerations. For instance, [33] evaluates and compares metal-oxide sensors and electro-chemical sensors with linear and nonlinear regression techniques for ozone estimation; however, the seasonal appearance of ozone in summer and meteorological variations may require different calibration models for other seasons and pollutants, reducing the accuracy in results from extrapolated values (i.e. lower temperatures from winter).…”

Section: B Related Workmentioning

confidence: 99%

“…Machine learning techniques such as Multilayer Perceptron (MLP) in Artificial Neural Networks (ANN), can produce better approximations than multiple linear regression (MLR) [26], [27]; however, for some pollutants, hybrid methods combining linear and non-linear techniques can produce more accurate results [28], [29]. In addition to hybrid methods, other variants in machine learning showed to ameliorate the classic ANN, by including correction factors (temperature (T), relative humidity (RH)) or temporal information (previous samples) of the environment with Dynamic Neural Networks (DNN) [30], [31], utilizing Random Forest (RF) [32], [33], k-Nearest Neighbors (k-NN) [33], Support Vector Regression (SVR) [33], [34], or adding Fuzzy Logic for qualitative calibration [35] and prediction [36].…”

Section: Introductionmentioning

confidence: 99%

Design and Evaluation of a Reliable Low-Cost Atmospheric Pollution Station in Urban Environment

2020

View full text Add to dashboard Cite

The pollution of the air constitutes an environmental risk to health, crops, animals, forests and water. There are several policies for reducing air pollution regarding industry, energy, transportation, and agriculture. Unfortunately, there is limited monitoring of the air quality in cities and rural areas for supervising the accomplishment of these policies. Reliable monitoring of air pollutants is, typically, based on expensive fixed stations, which constitutes a barrier to tackle. This research presents the design, implementation and evaluation of a small, low-cost, station for monitoring atmospheric pollution. The prototype registers ozone (O 3) and carbon monoxide (CO) using inexpensive sensors. To assure high reliability of the measurements obtained by the sensors installed in this station, it is proposed a calibration procedure based on the selection of the best performance analysis of the following machine learning techniques: multiple linear regression, artificial neural networks, and random forest. Additionally, a decision rule is implemented to select an optimal combination of sensors for the estimation models, while the sample timestamp is considered as a temporal heuristic at the input of the system, assuming similarities in the daily environmental dynamics. In order to test the station in a realistic scenario, the calibration and evaluation sets were taken in two different time frames of one and two months, respectively. The overall process was implemented with reference data coming from a certified air quality fixed station in the city of Cuenca-Ecuador. Experimental results showed that the real-time reports of ozone provided by the prototype are quite similar to the fixed station during the evaluation period, with a resulting correlation of up to r = 0.92 and r = 0.91 in the calibration and evaluation set, respectively. However, signal drift and aging in CO x sensors diminished the accuracy of carbon monoxide calibration models, resulting in lower correlation (r ≤ 0.76) with the evaluation set.

show abstract

“…In order to produce well-calibrated data, both multiple linear regression and nonlinear techniques were used in [ 59 ]. Additionally, a recalibration was done to mitigate the bias presented by the sensors and improve the variance.…”

Section: Introductionmentioning

confidence: 99%

Robust Inferential Techniques Applied to the Analysis of the Tropospheric Ozone Concentration in an Urban Area

Hernández

Méndez

González-Posadas

et al. 2021

Sensors

View full text Add to dashboard Cite

This paper analyzes 12 years of tropospheric ozone (O3) concentration measurements using robust techniques. The measurements were taken at an air quality monitoring station called Belisario, which is in Quito, Ecuador; the data collection time period was 1 January 2008 to 31 December 2019, and the measurements were carried out using photometric O3 analyzers. Here, the measurement results were used to build variables that represented hours, days, months, and years, and were then classified and categorized. The index of air quality (IAQ) of the city was used to make the classifications, and robust and nonrobust confidence intervals were used to make the categorizations. Furthermore, robust analysis methods were compared with classical methods, nonparametric methods, and bootstrap-based methods. The results showed that the analysis using robust methods is better than the analysis using nonrobust methods, which are not immune to the influence of extreme observations. Using all of the aforementioned methods, confidence intervals were used to both establish and quantify differences between categories of the groups of variables under study. In addition, the central tendency and variability of the O3 concentration at Belisario station were exhaustively analyzed, concluding that said concentration was stable for years, highly variable for months and hours, and slightly changing between the days of the week. Additionally, according to the criteria established by the IAQ, it was shown that in Quito, the O3 concentration levels during the study period were not harmful to human health.

show abstract

A Comparative Study of Calibration Methods for Low-Cost Ozone Sensors in IoT Platforms

Cited by 38 publications

References 24 publications

Multisensor Data Fusion Calibration in IoT Air Pollution Platforms

Multisensor Data Fusion Calibration in IoT Air Pollution Platforms

Design and Evaluation of a Reliable Low-Cost Atmospheric Pollution Station in Urban Environment

Robust Inferential Techniques Applied to the Analysis of the Tropospheric Ozone Concentration in an Urban Area

Contact Info

Product

Resources

About