COMPLEX PREDICTIVE ANALYTICS OF VOLCANIC ACTIVITY MASTER THESIS

Аннотация

Low-Power Wide Area Networks (LPWANs) are continuously gaining space in different IoT applications, mainly thanks to their ability to provide long-range coverage to devices and giving license-free frequency bands. Taking advantage of these tools, we can acquire different physical data from events occurring in our environment such as volcanoes. The focus of this thesis is on one of the most prominent LPWAN technologies: LoRa™. First, this thesis establishes a study about the recent techniques in Forecasting Volcanic events then we introduce the IoT technologies and their advantages, once founded the theoretical knowledge we pass through the instrumentation for the data acquisition. The construction of the data acquisition card and its sensors, then the test of a LoRa-based IoT network in a typical urban scenario Point to point to analyze and evaluate the system. Finally, the proposal of a model for a real-time 24/7 data acquisition to properly build a big dataset during several months to train a neural network and hopefully predict events or volcanic eruptions.