What is it about?
The project is designed to address the conflict between society's demand for security monitoring and individuals' insistence on preserving personal privacy. For it to work, an optical fiber that contains a specific engraved optical structure is positioned on the ground. As individuals pass through this area, variations in pressure give rise to alterations in the optical structure. Since different people have different physical conditions, like weight and stride length, a trained neural network is used to distinguish changes in optical structural characteristics so it can recognize walking patterns. In addition, to ensure real-time monitoring capabilities, the project integrates HLS (High-Level Synthesis) acceleration technology into the FPGA data flow design to effectively manage the massive data load generated by the fiber optic sensing system. This technology serves as a remarkably effective complementary tool for safeguarding critical infrastructure.
Featured Image
Photo by Compare Fibre on Unsplash
Why is it important?
Technology like this can be used in places like nursing homes, where comprehensive resident security monitoring is required. It not only ensures safe monitoring of occupants in private spaces, but it circumvents potential privacy invasions associated with camera surveillance.
Perspectives
I am aware that the project still has numerous challenges to overcome before it can truly address real-world issues, but it's a good start and a fun one.
Yuqi Li
University of Pittsburgh
Read the Original
This page is a summary of: FiberFlex: Real-time FPGA-based Intelligent & Distributed Fiber Sensor System for Pedestrian Recognition, ACM Transactions on Reconfigurable Technology and Systems, August 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3690389.
You can read the full text:
Contributors
The following have contributed to this page
