Multisensor Data Fusion Calibration in IoT Air Pollution Platforms

Abstract

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.

Establishing A Sustainable Low-Cost Air Quality Monitoring Setup: A Survey of the State-of-the-Art

“…In order to tackle the error sources mentioned above, researchers have suggested a calibration process that transforms the raw output to the corresponding reference-grade instrument values. Numerous studies have reported significant improvement in the accuracy with the calibration process [ 23 , 75 , 84 , 85 , 93 , 94 , 95 , 96 , 97 , 98 ]; single variable regression, multiple variable linear regression, polynomial regression, random forests, k-nearest neighbours, artificial neural networks are some of the models already used for LCS calibration [ 20 , 99 , 100 , 101 ]. Calibration needs to be done both before deployment (pre-deployment) and after deployment (post-deployment)…”

“…We conclude from the existing studies that evaluation should be made in three scenarios to ensure the sensors data accuracy and reliability. (1) Evaluation of the sensors against reference station for accuracy (sensor vs. reference) [ 20 , 22 , 23 , 24 , 25 , 27 , 79 , 84 , 93 , 94 , 100 , 120 , 123 , 125 , 135 , 153 ]. (2) Evaluation of sensor against the same type of sensor for precision (sensor vs. sensor) [ 49 , 93 , 95 , 100 , 133 , 154 ].…”

“…(1) Evaluation of the sensors against reference station for accuracy (sensor vs. reference) [ 20 , 22 , 23 , 24 , 25 , 27 , 79 , 84 , 93 , 94 , 100 , 120 , 123 , 125 , 135 , 153 ]. (2) Evaluation of sensor against the same type of sensor for precision (sensor vs. sensor) [ 49 , 93 , 95 , 100 , 133 , 154 ]. (3) Comparison of different models performance to finalize better one (model vs. model) [ 20 , 21 , 122 ].…”

Research on Software Design of Intelligent Sensor Robot System Based on Multidata Fusion

“…The performance of each sensor is complementary, and the information collected is not related to each other. The entire system receives complete information that cannot be received by any sensor [10]. Therefore, when one of the sensors fails, there will always be another sensor collecting environmental data as supplementary information, which will make the system less sensitive to interference caused by changes in the external environment, thereby improving the stability of the entire system [11].…”

Embedded LUKS (E-LUKS): A Hardware Solution to IoT Security

“…Many facts indicate that this growth is an upward trend [1], with IoT data traffic expected to reach around 2000 petabytes of information by 2024 [2]. The use of IoT devices has reached many fields such as industrial production [3], health care [4], or quality-of-life-related devices [5]. Its implementations in our homes and personal environments has had a great impact on many daily life processes [6,7].…”