Graph Signal Reconstruction Techniques for IoT Air Pollution Monitoring Platforms
Authors:
Pau Ferrer-Cid,
Jose M. Barcelo-Ordinas,
Jorge Garcia-Vidal
Abstract:
Air pollution monitoring platforms play a very important role in preventing and mitigating the effects of pollution. Recent advances in the field of graph signal processing have made it possible to describe and analyze air pollution monitoring networks using graphs. One of the main applications is the reconstruction of the measured signal in a graph using a subset of sensors. Reconstructing the si…
▽ More
Air pollution monitoring platforms play a very important role in preventing and mitigating the effects of pollution. Recent advances in the field of graph signal processing have made it possible to describe and analyze air pollution monitoring networks using graphs. One of the main applications is the reconstruction of the measured signal in a graph using a subset of sensors. Reconstructing the signal using information from sensor neighbors can help improve the quality of network data, examples are filling in missing data with correlated neighboring nodes, or correcting a drifting sensor with neighboring sensors that are more accurate. This paper compares the use of various types of graph signal reconstruction methods applied to real data sets of Spanish air pollution reference stations. The methods considered are Laplacian interpolation, graph signal processing low-pass based graph signal reconstruction, and kernel-based graph signal reconstruction, and are compared on actual air pollution data sets measuring O3, NO2, and PM10. The ability of the methods to reconstruct the signal of a pollutant is shown, as well as the computational cost of this reconstruction. The results indicate the superiority of methods based on kernel-based graph signal reconstruction, as well as the difficulties of the methods to scale in an air pollution monitoring network with a large number of low-cost sensors. However, we show that scalability can be overcome with simple methods, such as partitioning the network using a clustering algorithm.
△ Less
Submitted 23 August, 2022; v1 submitted 2 January, 2022;
originally announced January 2022.
Sampling Trade-Offs in Duty-Cycled Systems for Air Quality Low-Cost Sensors
Authors:
Pau Ferrer-Cid,
Julio Garcia-Calvete,
Aina Main-Nadal,
Zhe Ye,
Jose M. Barcelo-Ordinas,
Jorge Garcia-Vidal
Abstract:
The use of low-cost sensors in conjunction with high-precision instrumentation for air pollution monitoring has shown promising results in recent years. One of the main challenges for these sensors has been the quality of their data, which is why the main efforts have focused on calibrating the sensors using machine learning techniques to improve the data quality. However, there is one aspect that…
▽ More
The use of low-cost sensors in conjunction with high-precision instrumentation for air pollution monitoring has shown promising results in recent years. One of the main challenges for these sensors has been the quality of their data, which is why the main efforts have focused on calibrating the sensors using machine learning techniques to improve the data quality. However, there is one aspect that has been overlooked, that is, these sensors are mounted on nodes that may have energy consumption restrictions if they are battery-powered. In this paper, we show the usual sensor data gathering process and we study the existing trade-offs between the sampling of such sensors, the quality of the sensor calibration, and the power consumption involved. To this end, we conduct experiments on prototype nodes measuring tropospheric ozone, nitrogen dioxide, and nitrogen monoxide at high frequency. The results show that the sensor sampling strategy directly affects the quality of the air pollution estimation and that each type of sensor may require different sampling strategies. In addition, duty cycles of 0.1 can be achieved when the sensors have response times in the order of two minutes, and duty cycles between 0.01 and 0.02 can be achieved when the sensor response times are negligible, calibrating with hourly reference values and maintaining a quality of calibrated data similar to when the node is connected to an uninterruptible power supply.
△ Less
Submitted 2 June, 2022; v1 submitted 14 December, 2021;
originally announced December 2021.
