2021
DOI: 10.1016/j.dib.2021.107127
|View full text |Cite
|
Sign up to set email alerts
|

H2020 project CAPTOR dataset: Raw data collected by low-cost MOX ozone sensors in a real air pollution monitoring network

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
11
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
3

Relationship

4
4

Authors

Journals

citations
Cited by 16 publications
(11 citation statements)
references
References 6 publications
0
11
0
Order By: Relevance
“… [7] . These nodes were deployed in testbeds in Spain and Italy, and allowed the development of sensor calibration studies, in addition to publishing the data openly for other researchers to perform their studies [8] . As a continuation of the nodes that measured with MOX sensor technology, we have developed two prototype nodes in which we could connect the electrochemical (EQ) sensors in various conditions that would allow us to decide the working parameters of a node in the final deployment.…”
Section: Experimental Design Materials and Methodsmentioning
confidence: 99%
“… [7] . These nodes were deployed in testbeds in Spain and Italy, and allowed the development of sensor calibration studies, in addition to publishing the data openly for other researchers to perform their studies [8] . As a continuation of the nodes that measured with MOX sensor technology, we have developed two prototype nodes in which we could connect the electrochemical (EQ) sensors in various conditions that would allow us to decide the working parameters of a node in the final deployment.…”
Section: Experimental Design Materials and Methodsmentioning
confidence: 99%
“…2) Spanish O 3 reference station network for Catalonia: this data set is formed by forty-three nodes in the area of Catalonia capturing hourly tropospheric ozone data between the months of January and February of 2021, with a total of 1076 samples. 3) H2020 Captor network [35]: this data set is formed by eight nodes, five low-cost sensors and three reference stations, deployed in the area of Catalonia (Spain) during the summer of 2017 to capture half-hourly tropospheric ozone concentration levels. This data set has a total of 2368 samples.…”
Section: Outlier Detection: Residual-basedmentioning
confidence: 99%
“…We call this algorithm Volterra graph-based outlier detection (VGOD). We perform several experiments on networks of reference stations in Spain measuring tropospheric ozone (O 3 ), as well as a sensor drift detection experiment using a small heterogeneous sensor network involving high-cost instrumentation and lowcost sensors, deployed in the Captor H2020 project [35]. Specifically, in this article we:…”
Section: Introductionmentioning
confidence: 99%
“…Their applicability and effectiveness in solving specific monitoring and local decision support tasks are astonishing [ 55 ] compared to conventional methods. Most of these systems rely on simple sensory signal filtering methods [ 10 ], while others use AI-based solutions [ 21 , 56 ], deep learning [ 57 ], and other machine learning methods [ 58 ] to filter noises in the raw data [ 59 ], classifying events [ 60 ], detecting patterns [ 61 ], eliminating cyber threats in communication networks [ 60 , 62 , 63 ], compressing the data [ 64 ], etc.…”
Section: Review Of Recent Advancementsmentioning
confidence: 99%