Desenvolvimento de um sistema de localização indoor utilizando machine learning e fingerprinting de Wi-Fi na UFAM: aplicações e desafios

Abstract

This work proposes the development of an indoor localization system based on machine learning and Wi-Fi fingerprinting at the Federal University of Amazonas (UFAM). The project consists of six blocks, divided into two parts: offline and online. In the offline phase, processes are carried out to capture Wi-Fi signals using ESP32 devices, create a database with the collected samples, choose and train a classification model using Support Vector Machine (SVM). In the online phase, the trained model is used to recognize patterns in real-time and determine the expected indoor location. The system aims to overcome the limitations of GPS in indoor environments, such as signal blockages caused by UFAM's infrastructure, providing a precise and efficient solution for locating objects or devices within the university campus.