Data Acquisition Filtering Focused on Optimizing Transmission in a LoRaWAN Network Applied to the WSN Forest...

Area of Science:

• Environmental Science
• Computer Science
• Electrical Engineering

Background:

• Forest fire detection is crucial for environmental protection.
• Wireless Sensor Networks (WSNs) are increasingly used for remote monitoring.
• Power consumption and bandwidth efficiency are key challenges for WSNs in forest environments.

Purpose of the Study:

• To develop and evaluate an energy-efficient data acquisition system for early forest fire detection using WSNs.
• To reduce the power consumption of individual sensor modules in a LoRaWAN network.
• To improve the timeliness and accuracy of anomaly detection for fire ignition.

Main Methods:

• Implementation of a Wireless Sensor Network (WSN) for forest data acquisition.
• Testing and calibration of four algorithms: Exponential Smoothing, two Moving Averages, and Least Mean Square.
• Development of a combined algorithm for pre-filtering data before LoRaWAN transmission.
• Validation using Wildfire Simulation Events (WSE).

Main Results:

• The developed pre-filtering system achieved an accuracy rate of 0.73 with 0.5 possible false alerts.
• Sensor module battery life was improved by nearly 53% due to reduced data transmission.
• Abrupt changes indicative of fire ignition were recognized within 60 seconds.

Conclusions:

• The proposed pre-filtering algorithms effectively reduce energy consumption in WSN modules for forest monitoring.
• The system demonstrates potential for early fire detection, though further improvements with cloud-based algorithms are suggested.
• Optimized data transmission and anomaly detection enhance the operational efficiency of forest fire monitoring systems.