Abstract:
Most Wi-Fi-based gait recognition systems prioritize gait cycle detection. Dynamic measurements from commercial Wi-Fi devices are noisy, making gait cycle detection difficult. We propose a cycle-independent human gait recognition and walking direction estimation system, AGait, in Wi-Fi networks using the attention-based Recurrent Neural Network (RNN) encoder-decoder.
Two receivers and one transmitter in different spatial layouts capture more walking dynamics. An integrated walking profile is created by assembling and refining Channel State Information (CSI) from multiple receivers. The RNN encoder converts the walking profile into primary feature vectors. The decoder computes an attention vector for a gait or direction sensing task and predicts the target.
The attention scheme encourages AGait to adapt to different critical clips of CSI data for different tasks. AGait can achieve average F1 scores of 97.32 to 89.77 percent for gait recognition from a group of 4 to 10 subjects and 97.41 percent for direction estimation from 8 walking directions on commercial Wi-Fi devices in three indoor environments.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
