WLAN-BLE Based Indoor Positioning System using Machine Learning Cloud Services

This paper presents the design and implementation of a system for micro-localization using Wireless Networking Technologies, such as WiFi and Bluetooth Low Energy (BLE) based on the Internet of Things (IoT) philosophy. The solution consists of an acquisition system of wireless signal parameters and the subsequent position system inference meanly cloud computing-based services. The proposed localization mechanism is based on a Machine Learning (ML) location algorithm stemming from the Signal-to-noise ratio (SNR) and Received Signal Strength (RSS) footprinting method, which allow us to detect the XY-position in a regression case o the reference zones in classification inside indoor environments. This paper reports a systematic description of the proposed IoT-based system and its connection to the Amazon Web Services (AWS) cloud computing services. An evaluation of different classification and regression algorithms was performed. The resulted model was deployed in the cloud, for an online and real-time inference following stage through the Internet. Real experiments were performed in order to assess the proposed system.