Robust WiFi-based Indoor Presence Detection and Localization

In this project, we are interested in device-free methods that passively sense, monitor, and track peoples indoor presence, location, and movement using off-the-shelf Wi-Fi-enabled devices. We use information extracted from the physical layer of wireless links to detect and interpret human presence, location, and physical activities. The current design and implementation of Wi-Fi-based systems exhibit some temporal inconsistencies and limitations due to the complexity of the wireless signal propagation in indoor environment and the challenging nature of human’s behavior itself. This project focus on feature extraction techniques to reduce data inconsistencies and improving the performance of classical machine learning algorithms and deep learning models, for building robust smart-home applications such as presence detection and indoor localization.

Faculty Supervisor:

Xue (Steve) Liu

Student:

Qianyu Liu

Partner:

Aerial Technologies Inc.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Program: