Local Weather Monitoring using WSN and IoT as an Early Warning for Extreme Weather

Abstract

This research develops a local weather-monitoring system based on the Internet of Things (IoT) and a Wireless Sensor Network (WSN), employing four nodes to collect real-time data on temperature, humidity, air pressure, wind speed, and rainfall. Each node transmits its data to Firebase, where it is displayed on a web dashboard and used to trigger early-warning notifications via Telegram. Testing results show that the anemometer recorded an average deviation of 0.33 km/h, while the BME280 demonstrated high accuracy across three parameters: a 0.21°C (0.74%) deviation for temperature, 0.83% (1.31%) for humidity, and 0.28 hPa (0.02%) for air pressure. The system also exhibited stable data synchronization and rapid alert response times. The testing results demonstrate the potential of a multi-node approach to capture local microclimate variability and indicate its suitability for further development in machine learning–based predictive models.