Eemil Lagerspetz, Software Architect on our RAIN Platform writes about the ‘Pervasive Data Science on the Edge’ published at Helsinki University Department of Computer Science.
According to the World Health Organization (WHO), in 2016 air pollution was linked to over 4.2 million deaths per year, which is 11.6 percent of all deaths. Estimates suggest that 96 percent of the world’s population currently lives in areas where air pollution exceeds safe limits and that 2 to 5 percent of GDP is spent on treating related diseases.
Earlier this year, a group of 12 researchers from the Department of Computer Science and the Institute for Atmospheric and Earth System Research at the University of Helsinki, Finland, published a research vision of real-time massive scale air quality sensing for metropolitan areas that integrates tens of thousands or even millions of air quality sensors to monitor air quality at fine spatial and temporal resolution.
Among the scientists are Dr. Eemil Lagerspetz, a Software Architect of our SaaS platform Rain, and Prof. Sasu Tarkoma, an advisor to our company regarding Rain.
Dr. Lagerspetz is a docent and an Academy of Finland postdoctoral researcher with the Department of Computer Science at the University of Helsinki, where he completed his PhD in 2014. His research interests include large-scale data analysis (big data), mobile and edge computing, and energy efficiency.
Prof. Tarkoma is the dean and a professor at said Department of Computer Science, where he completed his PhD in 2006. His research interests include mobile computing, Internet technologies, and AI.
Machine learning, edge computing, and 5G connectivity
The vision presented in the paper, ‘Towards Massive Scale Air Quality Monitoring’, holds that large numbers of sensors, ranging from expensive reference stations to low-cost sensors integrated into vehicles or carried by pedestrians, are periodically calibrated in order to provide accurate readings of air quality.
The proposed vision relies on the ability to quickly calibrate low-cost sensors with data obtained from short visits in the vicinity of reference stations, such as high quality measurement towers. For example, sensors deployed in taxis and garbage trucks can obtain reference measurements for calibration as they pass by a reference station.
Capturing local variations in air quality requires dense deployments that integrate upward of 1,000 sensors per square mile. Calibration at this scale calls for new, advanced technologies that combine machine learning, edge computing, and 5G local connectivity.
Use cases like this underscore the need for the kinds of technologies and solutions that Lempea provides. We offer world-class consulting and software development services in this area, alongside our SaaS platform, Rain, a no-hassle solution that connects all hard- and software on the edge-to-cloud continuum.
Feel free to download the research paper ‘Towards Massive Scale Air Quality Monitoring’ in PDF format here: