Development of IoT-based Landslide Monitoring System Using Wireless Sensor Networks

Thesis Abstract

Landslides remain a significant natural hazard causing extensive damage to infrastructure, loss of life, and economic setbacks in vulnerable regions worldwide. The unpredictability and complexity of landslide phenomena necessitate the development of real-time, accurate, and cost-effective monitoring systems to facilitate early warning and disaster preparedness. This study aims to develop an Internet of Things (IoT)-based landslide monitoring system leveraging wireless sensor networks (WSNs) to enhance early detection capabilities and improve decision-making processes for disaster risk reduction. The specific objectives are to design and implement a sensor deployment framework tailored to landslide-prone terrains, develop a data acquisition and transmission protocol optimized for real-time monitoring, and evaluate the system's performance through field testing in a selected geological hazard zone. The research adopts a mixed-methods approach grounded in the positivist paradigm, combining quantitative experimental design with qualitative system evaluation. The study population comprises 150 sensor nodes deployed across a 2-square-kilometer landslide-prone area in hilly terrain. A stratified random sampling technique selects 50 sensor nodes for in-depth performance testing, covering parameters such as soil moisture, slope movement, and rainfall. Data collection instruments include high-precision wireless sensors with embedded accelerometers and moisture probes, a central data aggregation server, and user interfaces for real-time monitoring. The validity and reliability of the sensors are established through calibration against laboratory-grade instruments prior to deployment, and data integrity is maintained through encryption and synchronization protocols. Data analysis involves descriptive statistical techniques to summarize sensor readings, followed by inferential analyses such as multiple regression to identify relationships between environmental factors and landslide initiation. Temporal and spatial data patterns are analyzed using Geographic Information Systems (GIS) and anomaly detection algorithms to assess system responsiveness to precursor events. The analytical framework incorporates the Theory of Sensors Data-Driven Decision Making to underpin the system's design, while the Diffusion of Innovations theory guides the adoption and user interaction aspects of the monitoring platform. Key expected findings include demonstration of the system's capability to provide accurate, real-time alerts with minimal false positives, improved detection of soil instability indicators, and enhanced spatial resolution of hazard mapping. It is anticipated that the deployment will reveal critical thresholds for environmental parameters that precede landslide occurrences, contributing empirical data to landslide risk models. The integration of IoT and WSN technologies is expected to prove effective in reducing man-hours for ground monitoring, lowering operational costs, and increasing community awareness through timely information dissemination. This research makes a significant contribution to the body of knowledge by providing a replicable, scalable framework for IoT-based landslide monitoring that combines technological innovation with geohazard analysis. It advances understanding of real-time hazard detection in landslide-prone environments and offers a viable solution adaptable to diverse geological settings. The main conclusion highlights that IoT-driven monitoring platforms can substantially improve early warning systems, thereby enhancing resilience among vulnerable communities. Based on the findings, recommendations include scaling the system for broader regional implementation, integrating machine learning algorithms for predictive analytics, and establishing standardized data-sharing protocols among stakeholders. Future research directions suggest exploring autonomous drone deployment for sensor maintenance and expanding the system to monitor other geo-hazards such as earthquakes and floodwaters.