A Python Tool to Predict Wireless Network Signals in Indoor Environments using Neural Networks

Código em python para previsão propagação de sinais em ambientes indoor utilizando RNA's (Redes Neurais artificiais)

Resumen

Abstract — The use of neural networks proved to be effective in creating more accurate predictive models compared to traditional approaches. The Python tool developed made it possible to train and adjust these models based on the information collected, taking into account factors such as the physical structure of the site, obstacles present and building materials. The results obtained during the research indicated significant improvements in prediction accuracy compared to conventional methods. This suggests great potential for the practical use of the tool in real-world scenarios, such as the planning and optimisation of indoor wireless networks, contributing to more stable and reliable connectivity indoors. The aim of this work was to create a Python-based tool that uses neural networks to predict wireless network signals in indoor environments. The innovative approach, which combines mapping and field measurements, demonstrated an increase in the accuracy of predictions, promoting advances in the efficiency and reliability of wireless networks in indoor spaces.