I-Corps: Remote Health Monitoring from Wearable Antenna Measurements

点击翻译按钮获取中文摘要

The broader impact/commercial potential of this I-Corps project is significant as the proposed antenna measurement system and deep learning classification algorithms can be integrated with 5G mobile devices and provide real-time, continuous tracking and analysis of a wide range of potential activities. The technology will potentially benefit a diverse population including, but not limited to, tracking elderly people at home or in assisted living facilities, soldiers in the battlefield, and athletes in training or competition. The number of connected wearable devices worldwide is expected to surpass 1.1 billion by the year of 2022 and become a $27 billion market. The proposed approach is adaptive and scalable as more and more wearable devices are being deployed in the era of internet of things (IoT).This I-Corps project aims to enable dual-use of antennas embedded inside wearable devices for both wireless communication and motion sensing purposes. Recent research findings demonstrate that the input reflection coefficient from an on-body antenna exhibits unique patterns for different motions and can be used for high fidelity human activity classification as an integrated measurement system. It does not require integration of specialized motion sensors into wearable devices, which can reduce the complexity of the system design, the overall power consumption, and the cost. It can monitor both large-scale activities such as walking and small-scale activities such as finger movements, and can operate in both indoor and outdoor environments. The measurement data can be trained and tested by a customized deep convolutional neural network application to provide high-fidelity human activity classification results, realizing the goal of continuously, remote, user-dependent health monitoring.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.