| Version 29 (modified by , 4 years ago) ( diff ) |
|---|
Real-time machine learning-based user authentication via daily activities using wireless signals
Group Members
- Bhargav Singaraju
- Sachin Mathew
- Rishika Sakhuja
Project Objective
In this project, we plan to develop an end-to-end user authentication system that takes real-time WiFi signals as input and can recognize the user’s identity with low delay. The system first detects the presence of human activity and segment the activity which is then fed to a machine learning model. Based on the activity segment, the machine learning model determine the user identity associated with the activity segment in real-time. Tasks for this project could include wireless preprocessing-implementing real-time data segmentation, time/frequency-domain feature extraction mechanisms based on Python and scikit-learn library; machine learning-based model-developing SVM/DNN-based user identification/activity recognition models based on Python and TensorFlow library; real-time system-integrating the code for CSI preprocessing and machine learning-based model and fine-tune the system to make it capable of running in real time.
Development Tools Tutorials
- Pytorch tutorial: https://pytorch.org/tutorials/
- Tensorflow tutorial: https://www.tensorflow.org/tutorials
Week 1 Activities
Week 2 Activities
Week 3 Activities
Week 4 Activities
Week 5 Activities
Week 6 Activities
Week 7 Activities
Week 8 Activities
Final Presentation
Research Poster
Attachments (3)
- bhargav.jpg (7.7 KB ) - added by 4 years ago.
- rishika.JPG (40.0 KB ) - added by 4 years ago.
- sachin.jpg (36.9 KB ) - added by 3 years ago.
Download all attachments as: .zip
