A Modular IoT Platform for Real-Time Indoor Air Quality Monitoring

Mohieddine Benammar, Abderrazak Abdaoui, Sabbir H. M. Ahmad, Farid Touati, Abdullah KadriSensors 2018 · DOI

TL;DR

This paper introduces a modular, end-to-end Internet of Things (IoT) platform for real-time indoor air quality monitoring (IAQM). The system addresses the critical need for continuous, accessible air quality data by integrating wireless sensor nodes, intelligent gateways, and a cloud-based IoT server, with a strong emphasis on data reliability and seamless integration.

What Problem Does This Paper Address?

Indoor air quality (IAQ) is a significant concern for human health and comfort, as people spend up to 90% of their time in enclosed environments. Poor IAQ can lead to various health issues, and current monitoring systems often lack real-time data, ubiquity, or robust data handling, especially in the event of network disruptions. Furthermore, building managers often resort to over-ventilation to ensure safety, which is energy-intensive and costly. This paper tackles the challenge of creating a reliable, scalable, and accessible IAQM system that can monitor multiple pollutants in real-time and provide data to users anywhere, anytime, while ensuring data integrity even with intermittent internet connectivity.

Methods

The core of the proposed solution is a modular IAQM architecture. This includes:

  • Wireless Sensor Nodes: These nodes are built around the Libelium Waspmote platform, featuring an ATmega1281 microcontroller and equipped with sensors for CO2, CO, SO2, NO2, O3, Cl2, ambient temperature, and relative humidity. Communication with the gateway is handled by XBee PRO S2 radio modules. A key feature is the inclusion of a real-time clock (RTC) and a rechargeable battery for maintaining time and providing backup power.
  • Intelligent Gateways: The gateway, based on a Raspberry Pi 2, acts as a central hub. It receives data from sensor nodes via XBee modules and processes it before transmitting it to an IoT server. A crucial aspect of the gateway's design is its ability to detect transmission errors and re-submit lost packets, ensuring data integrity even during internet outages. It utilizes a local stack for temporary data storage.
  • IoT Server: The system leverages Emoncms, an open-source web application, for data logging, visualization, and user access. The gateway communicates with Emoncms using its APIs, allowing for real-time monitoring and historical data analysis.

The system's modularity allows for easy integration of various sensing technologies and WSNs. The firmware for both sensor nodes and gateways is custom-developed to manage data acquisition, communication, and the data backup/resubmission mechanisms.

Datasets

The paper does not explicitly mention the use of pre-existing datasets for training. Instead, it presents results from data collected using the developed IAQM system in two distinct locations at Qatar University:

  1. Office Space: Data collected in a large open-space office in the research building of the college of engineering.
  2. Chemical Engineering Lab: Data collected in a chemical engineering research lab.

These real-world data collections serve as the evaluation benchmarks for the system's performance and the insights derived from the monitored air quality.

Key Insights and Ideas

  • Reliability through Redundancy: The paper emphasizes the importance of data reliability. The inclusion of data backup at both the sensor node (battery for RTC and temporary power) and gateway (local stack for un-transmitted packets) levels is a critical design choice to combat data loss due to communication failures.
  • Modular and Scalable Architecture: The adoption of a modular design allows for flexibility in integrating different sensors and communication technologies, making the system adaptable to various IAQM needs and future expansions.
  • Intelligent Gateway for Data Integrity: The gateway's role extends beyond simple data forwarding. Its ability to detect and recover lost packets, coupled with local data storage, significantly enhances the robustness of the IAQM system, especially in environments with unreliable internet connectivity.
  • Real-time Visualization and Accessibility: The integration with Emoncms provides users with immediate access to IAQ data through graphical interfaces, enabling informed decision-making and a better understanding of indoor environmental conditions.
  • Understanding Indoor Pollutant Dynamics: The collected data from different environments (office vs. lab) reveal distinct patterns and correlations between pollutants, highlighting the influence of human activity, ventilation systems, and external factors like traffic.

Key Results

  • Packet Loss Evaluation: The study evaluated packet loss rates in an indoor environment (Qatar University library) using XBee PRO S2 modules. Results showed that unicast transmission generally outperformed broadcast transmission, with packet loss being significantly affected by obstacles (walls, doors). For instance, unicast transmission had at least 50% lower packet loss than broadcast transmission through double and triple concrete obstacles.
  • IAQM Data Collection: The system successfully collected and displayed real-time IAQM data for CO2, O3, CO, Cl2, NO2, SO2, temperature, and humidity in both an office space and a chemical engineering lab.
  • Pollutant Behavior Insights:
    • In the office, CO2 levels peaked around 1000 ppm during weekdays, aligning with recommended limits, and were low on weekends. O3 levels were significantly higher than WHO recommendations.
  • In the lab, CO2 concentration was strongly dependent on student presence and experimental activities. CO and NO2 levels showed a high correlation, likely due to nearby traffic and lab processes.
  • Data Storage Capacity: The gateway's 6 GB of usable storage on an SD card, combined with a data packet size of 114 bytes and a 15-minute sampling interval, is estimated to be sufficient for storing IAQM data for over 300 years, with an additional 0.01% of storage reserved for the backup stack, capable of storing data for approximately 11 days of internet disconnection.

Limitations and Future Work

  • Sensor Accuracy: The paper notes limitations with the accuracy of certain sensors, particularly for CO and NO2, which may not be sufficient for precise compliance checks against EPA limits. The O3 sensor's accuracy also limits its ability to measure levels within the recommended range.
  • Limited Scope of Initial Deployment: The current deployment is focused on specific locations within Qatar University.
  • Future Work: The authors suggest that the developed platform can be easily modified to assess other gases (e.g., VOCs) and that the IAQ parameters monitored can be adapted based on the specific environment and pollution sources. The platform is intended for broader deployment across various locations in Doha and for extended periods to cover different seasons. Further research could focus on developing more accurate sensors or implementing advanced calibration techniques.

Summary updated

BibTeX
@article{benammar2018modular,
  author = {Mohieddine Benammar and Abderrazak Abdaoui and Sabbir H. M. Ahmad and Farid Touati and Abdullah Kadri},
  title = {{A Modular IoT Platform for Real-Time Indoor Air Quality Monitoring}},
  journal = {Sensors},
  year = {2018},
  doi = {10.3390/s18020581},
}
Figure