Real-Time Air Pollution Monitoring Dataset - IoT Environmental Sensors Bangladesh

Dataset Overview

This real-time air pollution dataset from Mendeley Data (March 2024) provides extensive air quality monitoring data collected using IoT sensor networks deployed in Dhaka, Bangladesh - one of the world's most polluted cities. With 155,406 timestamped records, it enables comprehensive environmental research and smart city air quality management.

Pollutant Monitoring - 6 Key Parameters

Gaseous Pollutants

• Carbon Monoxide (CO - ppm): Toxic gas from vehicle emissions and incomplete combustion
• Nitrogen Dioxide (NO2 - ppb): Traffic-related pollutant causing respiratory issues
• Sulfur Dioxide (SO2 - ppb): Industrial emissions indicator
• Ozone (O3 - ppb): Secondary pollutant formed from photochemical reactions

Particulate Matter

• PM2.5 (μg/m³): Fine particles penetrating deep into lungs, major health hazard
• PM10 (μg/m³): Coarse particles from dust, construction, and industrial activities

IoT Hardware Infrastructure

Sensor Array

Professional-grade air quality sensors:

• Electrochemical Sensors: For CO, NO2, SO2 detection with ppb/ppm sensitivity
• Optical Particle Counters: Laser-based PM2.5 and PM10 measurement
• Ozone Sensors: UV absorption or electrochemical O3 detection

Data Acquisition System

• Microcontroller: Arduino platform for sensor integration and data processing
• Real-Time Logging: Continuous data collection with precise timestamping
• Data Transmission: Excel Data Streamer enabling live data flow to PC/cloud
• Storage: CSV format for easy analysis and archival

Temporal Coverage and Resolution

• Total Records: 155,406 measurements
• Sampling Frequency: High-resolution temporal data enabling hourly and daily pattern analysis
• Duration: Extended monitoring period capturing seasonal variations
• Continuity: Minimal data gaps ensuring reliable time-series analysis

Urban Air Quality Research Applications

Pollution Forecasting

• Train ML models predicting pollutant concentrations hours or days ahead
• Enable proactive public health warnings and traffic management
• Support early warning systems for vulnerable populations

Source Apportionment

• Identify contribution of different pollution sources (traffic, industry, cooking, construction)
• Correlate temporal patterns with human activities
• Guide targeted pollution control policies

Health Impact Studies

• Link air quality measurements to respiratory disease incidence
• Assess population exposure levels in different city zones
• Evaluate effectiveness of pollution reduction interventions

Smart City Integration

• Develop real-time air quality displays for citizens
• Integrate with traffic management systems for dynamic routing
• Support green infrastructure planning (urban forests, green corridors)

Machine Learning Applications

• Time-Series Forecasting: LSTM/GRU networks predicting next-hour PM2.5 or NO2 levels
• Anomaly Detection: Identify pollution episodes or sensor malfunctions
• Correlation Analysis: Study relationships between different pollutants
• Spatial Interpolation: Estimate air quality in unmeasured locations using neighboring sensors
• Classification: Categorize air quality into health advisory levels (good/moderate/unhealthy/hazardous)

Environmental and Public Health Value

Dhaka's severe air pollution makes this dataset particularly valuable for studying extreme urban air quality challenges. The high record count (155,406) provides statistical power for robust model training. Real-time IoT collection methodology is replicable for other cities seeking to establish air quality monitoring networks.

Data Format

Excel/CSV format with columns for timestamp and six pollutant concentrations. Clean, preprocessed, and ready for analysis. Documentation includes sensor specifications, calibration procedures, and deployment locations.