Device-free healthcare approaches for human activity monitoring and positioning by using Wi-Fi signals

As the world aging process quickened, the need for healthcare solutions to support seniors living on their own is recognized as a serious medical and social problem. Though an extensive amount of research has been carried out to investigate human activity based on a range of device-oriented (e.g., wearable) and device-free (e.g., vision based) sensing technologies. Monitoring activities of clinical relevance for senior well-being (e.g., eating, sleeping and falls) is still very challenging. Since the elderly people may not like to carry/wear sensors and they also have privacy concerns about non-wearable devices (e.g., in-home cameras). Moving along this direction, this project aims to provide a non-intrusive device-free approach for elderly activity monitoring and positioning by using the already deployed commodity Wi-Fi infrastructures. We focus on analyzing the specific motion of the elderly people (e.g., slowed movements, unstable transfers) to identify different events by leveraging the Wi-Fi Channel State Information (CSI) measurements, which is less discussed in the existing work. Moreover, both classical machine learning algorithms (e.g., SVM and KNN) and deep models (e.g., CNN, LSTM and attention model) will be deployed to enhance the system performance (…)

Faculty Supervisor:

Xue (Steve) Liu

Student:

Landu Jiang

Partner:

Aerial Technologies Inc.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Program: