Highly secure access control requires Swiss-cheese-type multi-layer security protocols. The use of electroencephalogram (EEG) to provide cognitive indicators for human workload and fatigue has created environments where the EEG data are well-integrated into systems, making it readily available for more forms of innovative uses including biometrics. However, most of the existing studies on EEG biometrics rely on resting state signals or require specific and repetitive sensory stimulation, limiting their uses in naturalistic settings. Moreover, the limited discriminatory power of uni-variate measures denies an opportunity to use dependences information inherent in brain regions to design more robust biometric identifiers. In this paper, we proposed a novel model for ongoing EEG biometric identification using EEG collected during a diverse set of tasks. The novelty lies in representing EEG signals as a graph based on within-frequency and cross-frequency functional connectivity estimates, and the use of graph convolutional neural network (GCNN) to automatically capture deep intrinsic structural representations from the EEG graphs for person identification. An extensive investigation was carried out to assess the robustness of the method against diverse human states, including resting states under eye-open and eye-closed conditions and active states drawn during the performance of four different tasks. We compared our method with the state-of-the-art EEG features, classifiers, and models of EEG biometrics. Results show that the representation drawn from EEG functional connectivity graphs demonstrates more robust biometric traits than direct use of uni-variate features. Moreover, the GCNN can effectively and efficiently capture discriminative traits, thus generalizing better over diverse human states.

Electroencephalography (EEG) is one of the most promising methods in the field of Brain-Computer Interfaces (BCIs) due to its rich time-domain resolution and the availability of advanced and portable sensor technology. One of the major challenges for EEG signal analysis is the small size of its datasets as it is usually demanding for human subjects to perform lengthy experiments. Consequently, this challenge can limit the performance of EEG signal classification models. In this paper, we propose a novel generative adversarial network (GAN) model that can learn the statistical characteristics of the EEG signal and augment its datasets size to enhance the performance of classification models. Results show that the proposed model significantly outperforms other generative models on the utilized EEG dataset. Furthermore, it significantly enhances the performance of classification models working on small size EEG datasets after augmenting them with generated samples.

Deep learning has become a predominant method for solving data analysis problems in virtually all fields of science and engineering. The increasing complexity and the large volume of data collected by diverse sensor systems have spurred the development of deep learning methods and have fundamentally transformed the way the data are acquired, processed, analyzed, and interpreted. With the rapid development of deep learning technology and its ever-increasing range of successful applications across diverse sensor systems, there is an urgent need to provide a comprehensive investigation of deep learning in this domain from a holistic view. This survey paper aims to contribute to this by systematically investigating deep learning models/methods and their applications across diverse sensor systems. It also provides a comprehensive summary of deep learning implementation tips and links to tutorials, open-source codes, and pretrained models, which can serve as an excellent self-contained reference for deep learning practitioners and those seeking to innovate deep learning in this space. In addition, this paper provides insights into research topics in diverse sensor systems where deep learning has not yet been well-developed, and highlights challenges and future opportunities. This survey serves as a catalyst to accelerate the application and transformation of deep learning in diverse sensor systems.

Electroencephalography (EEG) signals are complex dynamic phenomena that exhibit nonlinear and nonstationary behaviors. These characteristics tend to undermine the reliability of existing hand-crafted EEG features that ignore time-varying information and impair the performances of classification models. In this paper, we propose a novel method that can automatically capture the nonstationary dynamics of EEG signals for diverse classification tasks. It consists of two components. The first component uses an autoregressive-deep variational autoencoder model for automatic feature extraction, and the second component uses a Gaussian mixture-hidden Markov model for EEG classification with the extracted features. We compare the performance of our proposed method and the state-of-the-art methods in two EEG classification tasks, subject, and event classification. Results show that our approach outperforms the others by averages of 15% ± 6.3 (p-value <; 0.05) and 22% ± 5.7 (p-value <; 0.05) for subject and event classifications, respectively.

As a promising candidate to complement traditional biometric modalities, brain biometrics using electroencephalography (EEG) data has received a widespread attention in recent years. However, compared with existing biometrics such as fingerprints and face recognition, research on EEG biometrics is still in its infant stage. Most of the studies focus on either designing signal elicitation protocols from the perspective of neuroscience or developing feature extraction and classification algorithms from the viewpoint of machine learning. These studies have laid the ground for the feasibility of using EEG as a biometric verification modality, but they have also raised security and privacy concerns as EEG data contains sensitive information. Existing research has used hash functions and cryptographic schemes to protect EEG data, but they do not provide functions for revoking compromised templates as in cancellable template design. This paper proposes the first cancellable EEG template design for privacy-preserving EEG-based verification systems, which can protect raw EEG signals containing sensitive privacy information (e.g., identity, health and cognitive status). A novel cancellable EEG template is developed based on EEG features extracted by a deep learning model and a non-invertible transform. The proposed transformation provides cancellable templates, while taking advantage of EEG elicitation protocol fusion to enhance biometric performance. The proposed verification system offers superior performance than the state-of-the-art, while protecting raw EEG data. Furthermore, we analyze the system’s capacity for resisting multiple attacks, and discuss some overlooked but critical issues and possible pitfalls involving hill-climbing attacks, second attacks, and classification-based verification systems.