Automatic WiFi

Automatic WiFi sensing leverages readily available WiFi signals to passively infer information about the environment and its occupants, aiming to replace or augment traditional sensor systems. Current research focuses on developing robust machine learning models, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and transformer architectures, for tasks such as human activity recognition, localization, and pose estimation, often employing self-supervised or few-shot learning techniques to address data scarcity. This field is significant due to its potential for low-cost, privacy-preserving solutions in diverse applications ranging from robotics and healthcare monitoring to smart home automation and security systems.