Research Article
Unobtrusive User Verification using Piezoelectric Energy Harvesting
@INPROCEEDINGS{10.4108/eai.7-11-2017.2273542, author={Dong Ma and Guohao Lan and Weitao Xu and Mahbub Hassan and Wen Hu}, title={Unobtrusive User Verification using Piezoelectric Energy Harvesting}, proceedings={14th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services}, publisher={ACM}, proceedings_a={MOBIQUITOUS}, year={2018}, month={4}, keywords={energy-efficiency gait verification sparse representation wearable device}, doi={10.4108/eai.7-11-2017.2273542} }- Dong Ma
Guohao Lan
Weitao Xu
Mahbub Hassan
Wen Hu
Year: 2018
Unobtrusive User Verification using Piezoelectric Energy Harvesting
MOBIQUITOUS
ACM
DOI: 10.4108/eai.7-11-2017.2273542
Abstract
With the capability to harvest energy from low frequency motions or vibrations, piezoelectric energy harvesting has become a promising solution to achieve self-powered wearable system. Apart from generating energy to power the wearable devices, the output electricity signal of the PEH can also be used as an information source as it reflects the activity or motion patterns of the user. In this paper, we have designed and built an insole-based user authentication system by leveraging the AC voltage generated by the PEH during human walking. Meanwhile, the generated power is also collected and stored, which could be later used as the power source of the mobile system. By using a dataset of 20 subjects, we have demonstrated that our system can achieve 89.76% of human recognition accuracy when using only one gait cycle signal, and the accuracy can be further increased to 95.86% when two gait cycles are utilized.
