A Localization and Monitoring Framework for Indoor Environments Leveraging IoT Signal Processing and ML Algorithms with Feedback Process

Driven by the latest advancements in wireless technology location-based services have attracted the interest of computing and telecommunication industries, as well as academia, to launch fast and accurate localization systems. The aim of this work is to propose a closed-loop localization framework for large-scale deployments facilitating both the modeling and continuous monitoring of Activities of Daily Living (ADLs). The design of these localization systems is very challenging, time consuming and their adaptation in environmental changes is hard. The proposed methodology takes advantage of limited RSSI measurements at different distances, enriches the data and accurately models the attenuation of the propagated signal. These measurements are then used as input in the data-enrichment process, where the proposed framework generates datasets at different distances. Therefore, all created datasets (gathered and generated) are exploited to train the proposed ML-based chain. The primary purpose of the ML-chain is to determine the distance between the mobile nodes and each installed beacon. The position is then calculated using trilateration methods. Finally, the collected RSSI along with the estimated position will be stored and used for increasing position accuracy, allowing our proposed framework to continuously and automatically optimize its processes and accuracy. Furthermore, to be useful and practical, once reliable position estimation is achieved, the proposed framework can detect predefined Activities of Daily Living (ADLs) based on location patterns and movement behaviors. This capability opens new opportunities for context-aware services and smart environment applications. Each module of the framework was individually tested and evaluated, demonstrating strong performance both in isolation and as part of the integrated system.