Low Cost Sensor Networks: How Do We Know the Data Are Reliable?

Area of Science:

• Environmental Science
• Sensor Technology
• Data Science

Background:

• Informal networks of low-cost devices are increasingly used for air quality monitoring.
• The plausibility of data from these networks is contentious due to the need for regular on-site calibration.
• On-site calibration is a specialist task, and its cost can outweigh the initial installation expense.

Purpose of the Study:

• To address the contentious issue of data plausibility in low-cost sensor networks.
• To propose alternative methods for ensuring data believability without relying solely on on-site calibration.
• To illustrate potential solutions using examples from air quality monitoring networks.

Main Methods:

• Exploring the use of independent information to validate sensor data.
• Analyzing approaches to data validation that circumvent the need for frequent, costly on-site calibration.
• Focusing on air quality monitoring as a case study for these methods.

Main Results:

• Independent information, when appropriately used, can resolve the contention surrounding data plausibility.
• This approach offers a cost-effective alternative to traditional calibration methods for low-cost sensor networks.
• The findings are particularly relevant for the growing field of informal air quality monitoring.

Conclusions:

• The use of independent information is a promising strategy for enhancing the credibility of low-cost sensor network data.
• This method addresses the economic and logistical challenges associated with on-site calibration.
• The study provides a framework for improving the reliability of environmental monitoring data from informal networks.