A Survey of User Authentication Based on Channel State Information

Table 3

Gesture-based user authentication applications.

System	Users	Signal/preprocessing	Experimental scene	Purpose/classification	Performance
Wi-Sign [82]	14	Amplitude/PCA Butterworth low-pass filter	A typical office room ()	Identity recognition, stranger identification/SVM	14 people TPR: 79%; Intruder detection TNR: 86%
WiID [83]	21	Amplitude/PCA	A lab (400 ft^2), an office (150 ft^2), a living room (192 ft^2), and a bedroom (154 ft^2)	Identity recognition/SVDE	An average cross-validation accuracy of 5 users: 92.8% in four environments
Finger Pass [55]	7	Amplitude and phase/IFFT	A living room (), a bedroom (), and a kitchen ()	Identity recognition, stranger identification/SVM, LSTM-based DNN	7 people: 91.4%