Source-linked AI summary
A Wireless Sensor Network Air Pollution Monitoring System
Kavi K. Khedo, Rajiv Perseedoss, Avinash Mungur, University of Mauritius, Mauritius
TL;DR
Mauritius’s growing industrialization, urbanization, and traffic contribute to an air-pollution monitoring problem that existing bulky and resource-limited systems address inefficiently. The paper proposes WAPMS, combining dense wireless sensing, AQI-based reporting, hierarchical routing, sleeping motes, and RCQ aggregation. WAPMS is reported to provide timely monitoring, alerts, flexible visualization, and reduced transmission data through quartile-based summarization.
Problem
Mauritius faces increasing air pollution, while existing monitoring is resource-limited and uses bulky instruments; an AQI is also not presently available.
Method
WAPMS uses distributed wireless sensor nodes, AQI categorization, DSR-based multihop transmission, duplicate elimination, and Recursive Converging Quartiles data aggregation.
Results
WAPMS provides real-time regional pollution information, AQI-colored displays, alerts for drastic air-quality changes, and both summarized and fine-grained results.
Takeaways & Limitations
The system is presented as a flexible monitoring approach for high-risk regions, supporting timely information and responses by authorities.
Abstract
from arXiv · showhide
Sensor networks are currently an active research area mainly due to the potential of their applications. In this paper we investigate the use of Wireless Sensor Networks (WSN) for air pollution monitoring in Mauritius. With the fast growing industrial activities on the island, the problem of air pollution is becoming a major concern for the health of the population. We proposed an innovative system named Wireless Sensor Network Air Pollution Monitoring System (WAPMS) to monitor air pollution in Mauritius through the use of wireless sensors deployed in huge numbers around the island. The proposed system makes use of an Air Quality Index (AQI) which is presently not available in Mauritius. In order to improve the efficiency of WAPMS, we have designed and implemented a new data aggregation algorithm named Recursive Converging Quartiles (RCQ). The algorithm is used to merge data to eliminate duplicates, filter out invalid readings and summarise them into a simpler form which significantly reduce the amount of data to be transmitted to the sink and thus saving energy. For better power management we used a hierarchical routing protocol in WAPMS and caused the motes to sleep during idle time.
1. INTRODUCTION
Mauritius faces worsening air pollution amid industrial growth, urbanization, and increased traffic, while existing monitoring is costly, bulky, and slow. The paper proposes WAPMS to provide denser, more timely monitoring, AQI-based interpretation, and operational reporting.
- Mauritius’s air pollution has increased alongside industrial activity, infrastructure development, urbanization, and rising vehicle numbers.Industrial combustion, stone crushing, and vehicle emissions are identified as contributors to deteriorating air quality.
- Existing monitoring in Mauritius relies on bulky instruments and lacks resources, limiting flexibility and control.Traditional data logging is described as time-consuming and expensive.
- WAPMS targets coordinated collection from thousands of nodes while minimizing duplicate and invalid readings and transmission power.The system includes architecture, aggregation, visualization, reporting, and real-time notifications for serious pollution states.
- Monitoring supplies pollutant measurements that can be analyzed to assess daily severity, spatial differences, and changes over time.The paper presents monitoring as a way to address insufficient understanding of air quality across the country.
2. RELATED WORKS
Related work shows that WSNs support distributed environmental monitoring, control, and surveillance across difficult or large-scale settings. Examples include fire and flood detection, environmental control, precision agriculture, habitat monitoring, and wildlife tracking.
- WSNs have expanded from military origins into civilian applications including environmental monitoring, healthcare, home automation, fire detection, and traffic control.Their small, versatile sensors support monitoring, tracking, and controlling applications.
- Environmental WSN deployments monitor forests, weather, biodiversity, air, soil, and water, often connecting sensors to processing centers.These systems can also adjust environmental conditions through actuators.
- 2.1. Fire and Flood Detection: The FFSS and ALERT systems use sensor measurements for real-time forest-fire alarms and flood detection or prevention.FFSS monitors temperature, humidity, and smoke, while ALERT uses rainfall, water-level, and weather sensors.
- 2.2. Biocomplexity Mapping and Precision Agriculture: Precision agriculture applies WSNs to monitor pesticides, soil erosion, air pollution, soil, crops, and climate across large fields.Such applications can generate very large amounts of sensor data.
- 2.3. Habitat Monitoring: Great Duck Island used low-power motes to unobtrusively monitor habitat microclimates and relay real-time environmental data over the Internet.Measured variables included infrared levels, humidity, rainfall, and temperature.
- 2.3. Habitat Monitoring: DeerNet uses a WSN-based system to track deer behavior and support wildlife and disease-related studies.The stated goal is long-lived, unobtrusive, real-time video monitoring.
3. RECURSIVE CONVERGING QUARTILES (RCQ) DATA AGGREGATION ALGORITHM
RCQ is a data aggregation algorithm for WAPMS that removes duplicate packets and summarizes sensor readings through recursive quartile computation. It reduces transmitted data while retaining a representation intended to resist extreme or invalid values.
- RCQ addresses high WSN data volumes by combining duplicate elimination with data fusion before transmission to the sink.The algorithm explicitly targets reduced overhead while balancing transmitted-data volume and reliability.
- 3.1. Duplicate Elimination Technique: Duplicate elimination groups packets by node identifier and retains only one packet instance for each repeated identifier.Each packet contains a sensor reading and a unique node identifier.
- 3.2. Proposed Data Fusion Technique: Quartile fusion uses lower, median, and upper quartiles because quartiles are unaffected by extreme values that may be invalid.The method reduces a list to three values while preserving a representation of the original readings.
- 3.2. Proposed Data Fusion Technique: RCQ partitions a list into groups, computes three quartiles per subgroup, merges those quartiles, and repeats until one group remains.Group selection maximizes the number of groups while keeping subgroup size above a threshold.
- 3.2. Proposed Data Fusion Technique: 33 values are aggregated to 3 values using RCQ.The paper identifies this reduction as the algorithm’s demonstrated operation in Figure 3.
4. WAPMS: THE PROPOSED AIR POLLUTION MONITORING SYSTEM
WAPMS is a hierarchical wireless sensor system for collecting, aggregating, transmitting, and interpreting air-quality data across Mauritius. It uses AQI-based reporting and a partitioned deployment with cluster heads, sinks, and a gateway.
- System architecture: WAPMS gathers sensor readings autonomously and forwards them through base stations to a server for processing.The system can also send commands to nodes to fetch data, while nodes may transmit data autonomously.
- Deployment strategy: The deployment partitions the monitored region into smaller areas, with cluster heads collecting and aggregating readings from sensor nodes.Nodes communicate with their area cluster head through multihop routing.
- Deployment strategy: Multiple sinks collect aggregated cluster-head data and transmit it to a gateway, which relays results to the database and application.For the Port Louis prototype, a single sink is used and the gateway also performs the sink role.
- Prototype setting: The Port Louis prototype divides the site into 6 smaller areas because it is an urban region more exposed to air pollution than rural areas.The system is simulated over this small region before planned extension to the whole island.
- Air Quality Index: AQI reports air quality using pollutant measurements and six colour-coded health-concern categories.The system considers ozone, fine particulate matter, nitrogen dioxide, carbon monoxide, sulphur dioxide, and total reduced sulphur compounds.
5. SIMULATIONS AND RESULTS
WAPMS was simulated with JiST/SWANS using Dynamic Source Routing and evaluated under increasing numbers of areas and nodes. The system produced rapid analysis and visualized AQI results for selected locations.
- Simulation setup: WAPMS was simulated using the JiST/SWANS wireless-network simulation platform.SWANS supports scalable wireless and sensor-network configurations and efficient signal-propagation computation.
- Simulation setup: Dynamic Source Routing provided self-organizing, self-configuring multihop data transmission in WAPMS.DSR uses request-driven route discovery and route maintenance.
- Monitoring outputs: The system displays nodes using AQI colours, provides AQI readings and health concerns for an area and date, and generates line graphs for selected areas.These outputs support visual inspection of collected air-quality data.
- Performance evaluation: The evaluation varied simulated areas from 1 to 6 and nodes per area from 50 to 200, recording execution time.Results were reported in Table 2 and Figure 13.
- Performance evaluation: Less than 20 minutes was the maximum simulator running time for 6 areas with 200 nodes in each area.The authors report that this enables timely monitoring and near-immediate detection of abnormal situations compared with the existing monitoring unit.
6. CONCLUSION
WAPMS combines real-time air-pollution monitoring, AQI-based health communication, RCQ data aggregation, and flexible visualization for Mauritius. Its design aims to improve response readiness while reducing transmitted data and energy use.
- WAPMS provides real-time air-pollution information and alerts for drastic changes, supporting prompt authority responses such as evacuation or emergency deployment.The system targets high-risk regions and is intended to provide actionable monitoring information.
- AQI categories and intuitive colours communicate pollution levels and evaluate health concern for specific areas.The index is intended to support an air-pollution categorization system in Mauritius.
- RCQ summarizes readings into three quartile values, reducing data transmission and required transmission energy while representing original values accurately.The aggregation technique is designed for power-consumption minimization in WSNs.
- WAPMS supports both area-level summaries and fine-grained sensor readings through tables, line graphs, AQI displays, and sensor-location maps.Users can inspect individual readings, trends across areas, and mapped sensor locations in Port Louis.