Private authentication keys based on wearable device EEG recordings

Abstract

Abstract-In this paper, we study an Electroencephalography (EEG) based biometric authentication system with privacy protection. We use motor imagery EEG, recorded using a wearable wireless device, as our biometric modality. To obtain EEG-based authentication keys we employ the fuzzy-commitment like scheme with soft-information at the decoder, see Ignatenko and Willems [2014]. In this work we study the effect of multi-level quantization together with binary encoding of EEG biometric at the encoder on the system performance, when EEG feature vectors have limited length. We demonstrate our findings on an experimental EEG dataset of ten healthy subjects.

Biometric Identification Systems with Noisy Enrollment for Gaussian Sources and Channels

“…Another extension is to construct practical codes that can achieve the capacity regions. In the BIS with a single user, convolutional and turbo codes that control the privacy-leakage were investigated in [31] and applied to real-life application, Electroencephalograph, in [32]. In these studies, it was shown that by applying vector quantization at the encoder and softdecision at the decoder for Gaussian sources, a lower privacy-leakage rate was realizable.…”

Review on EEG‐Based Authentication Technology

“…e performance of the biometric cryptosystem is a true-positive rate (TPR) of 0.9974. Yang et al [83] used the two MI activity EEG signals obtained by the C channel from 10 subjects, filtered them with CAR to remove the average potential of all electrodes, and further normalized them to obtain the original EEG data. Feature vectors are obtained with AR.…”

“…Methods Key length (bits) Number of subjects Singandhupe et al [80] Fuzzy extractor 128 Damaševičius et al [82] Fuzzy commitment 400 42 Yang et al [83] Fuzzy commitment 21 10 Tuiri et al [84] Quantization 230 8 Bajwa and Dantu et al [76] Quantization 230 120 Nguyen et al [30] Quantization 256 3 Ravi et al [85] Quantization 62 10 improving the reliability of the system and the security of the biometric template. Albermany and Baqer [90] choose the C3, CZ, and C4 three channels of the left-hand and right-hand MI data from BCI Competition 2008-Graz data set A as their EEG data set.…”