Bluetooth Low Energy
Bluetooth Low Energy (BLE) is a low-power wireless technology increasingly used for various applications, particularly in the Internet of Things (IoT). Current research focuses on improving BLE's accuracy for location tracking and proximity detection, often combining it with other sensing modalities (e.g., ultrasound, inertial measurement units, WiFi) and employing machine learning techniques like neural networks (including attentional models and autoencoders) and Long Short-Term Memory networks for data processing and prediction. These advancements are driving improvements in areas such as smart factory worker monitoring, contact tracing, and real-time device tracking, while also addressing challenges related to energy consumption prediction and RF fingerprinting for enhanced security.
Papers
Large Scale Passenger Detection with Smartphone/Bus Implicit Interaction and Multisensory Unsupervised Cause-effect Learning
Valentino Servizi, Dan R. Persson, Francisco C. Pereira, Hannah Villadsen, Per Bækgaard, Jeppe Rich, Otto A. Nielsen
"Is not the truth the truth?": Analyzing the Impact of User Validations for Bus In/Out Detection in Smartphone-based Surveys
Valentino Servizi., Dan R. Persson, Francisco C. Pereira, Hannah Villadsen, Per Bækgaard, Inon Peled, Otto A. Nielsen