Precise detection of driver mental fatigue is critical for reducing traffic accidents and enhancing road safety. Compared with vision-based detection—which is susceptible to illumination and occlusion—multimodal physiological‑signal-based approaches integrate complementary information from diverse biosignals, delivering more faithful and objective fatigue assessments. However, adverse factors such as motion artifacts and environmental noise induce ceaseless deterioration to physiological signals, which markedly degrade the performance of existing multimodal fusion methods. To address this challenge, we propose Multimodal Uncertainty-based Self-driven Evolution, MUSE, reallocating modality contributions in real time via overall uncertainty minimization, thereby enabling efficient collaborative fusion of multi‐source predictions. Theoretically, MUSE guarantees a provably bounded cumulative error, and its generalization error approaches the Bayesian‑optimal fusion as iterations progress. Operating in a closed loop without labels or manual recalibration, MUSE presents superior suitability for real‑world driving scenarios compared to supervised algorithms. On the large‑scale driving fatigue dataset SEED‑VIG, MUSE outperforms existing models in both classification and regression tasks, substantiating its robustness and practicality as a promising driving fatigue detection solution.

This paper presents a novel binocular-encoded SSVEP (beSSVEP) method, leveraging binocular vision in virtual reality (VR) to enhance brain-computer interface (BCI) applications. We introduce the Binocular Periodically Repeated Component Analysis (bPRCA) algorithm, designed to address the unique characteristics of binocular-encoded targets, which include combinations of monocular single-frequency SSVEP units or void units, with frequency units being reused multiple times in the encoded interface. To further optimize performance, we propose the Fusion Component Analysis (FusionCA) framework, which integrates bPRCA with Task-Related Component Analysis (TRCA), effectively utilizing both steady-state periodic components and cross-trial aperiodic components. Experimental results demonstrate that ensemble-FusionCA achieves the highest information transfer rate (ITR) with an average accuracy of 71.39% and an ITR of 138.50 bits/min at 0.4 seconds, among the comparison with ensemble-bPRCA and ensemble-TRCA. Compared to traditional SSVEP approaches, beSSVEP significantly enhances frequency utilization, making VR-BCI systems more efficient and practical. This study highlights the application of physiological mechanisms of binocular vision to improve BCI systems, offering a new perspective for developing fast and scalable brain-computer interactions in VR environments.

In recent years, the diverse applications of electroencephalography (EEG) - based affective brain-computer interfaces (aBCIs) are being extensively explored. However, due to adverse factors like noise and physiological variability, the recognition capability of aBCIs can unforeseeably suffer abrupt declines. Since the timing of these aBCI failures is unknown, placing trust in aBCIs without scrutiny can lead to undesirable consequences. To alleviate this issue, we propose an algorithm for estimating the reliability of aBCI (primarily Graph Convolutional Network), synchronously delivering a probabilistic confidence score upon aBCI decision completion, thereby reflecting the aBCI's real-time recognition capabilities. Methodologically, we use the Maximum Softmax Probability (MSP) from EEG recognition networks as confidence scores and leverage the Scaling Operator to calibrate them. Then, the Projection Operator is employed to address confidence estimation biases caused by noise and subject variability. For the numerical concentration of MSP, we provide fresh insights into its causes and propose corresponding solutions. The derivation of the estimator from the Maximum Entropy Principle is also substantiated for robust theoretical underpinnings. Finally, we confirm theoretically that the estimator does not compromise BCI performance. In experiments conducted on public datasets SEED and SEED-IV, the proposed algorithm demonstrates superior performance in estimating aBCIs reliability compared to other benchmarks, and commendable adaptability to new subjects. This research has the potential to lead to more trustworthy aBCIs and advance their broader application in complex real-world scenarios.

Motor imagery-based brain-computer interfaces (MI-BCIs) hold significant potential for integration into Internet of Things (IoT) environments, enabling seamless interaction with users and external devices. In MI decoding, graph neural networks (GNNs) are increasingly employed due to their capacity to model the structures of non-grid data. However, current GNN approaches predominantly emphasize pairwise relationships, failing to account for the multi-node interactions essential for accurate decoding. To overcome these limitations, we propose NexusNet, a lightweight graph modeling network tailored for MI-BCI and IoT integration. NexusNet incorporates multi-node routing and Nexus Fusion, facilitating the modeling of intricate relationships that extend beyond pairwise connections. It effectively captures global information using a shortest path algorithm while complementing it with localized neighboring relationships. This architecture ensures efficient feature extraction and decoding while maintaining low computational overhead, making it suitable for resource-constrained scenarios. Experimental evaluations demonstrate that NexusNet achieves state-of-the-art performance, attaining 78.78% accuracy on BCIC-IV-2a, 87.21% on BCIC-IV-2b, and 94.12% on High Gamma Dataset with only 3.44K learnable parameters and 0.88M floating-point operations. These results highlight its potential for practical applications. Further analyses validate the contributions of its multi-node components, with visualizations illustrating their indispensable roles in decoding. NexusNet represents a step forward in connecting MI-BCI research with practical IoT deployment. Our code is available at https://github.com/ZikaiVan/NexusNet.

In motor imagery (MI) decoding, it still remains challenging to excavate enough contextual information of MI in different brain regions and to bridge the cross-session variance in feature distributions. In light of these issues, our study presents an innovative Bi-Stream Adaptation Network (BSAN) to bolster network efficacy, aiming to improve MI-based brain-computer interface (BCI) robustness across sessions. Our framework consists of the Bi-attention module, feature extractor, classifier, and Bi-discriminator. Precisely, we devise the Bi-attention module to reveal granular context information of MI with performing multi-scale convolutions asymptotically. Then, after features extraction, Bi-discriminator is involved to align the features from different MI sessions such that a uniform and accurate representation of neural patterns is achieved. By such a workflow, the proposed BSAN allows for the effective fusion of context coherence and session-invariance within the network architecture, therefore diminishing the reliance of redundant MI trials for MI-BCI re-calibration. To empirically substantiate BSAN, comprehensive experiments are conducted based on two public MI datasets. With average accuracies of 78.97% and 83.79% on two public datasets, and an inference time of 2.99 ms on CPU-only devices, it is believed that our approach has the potential to accelerate the practical deployment of MI-BCI.

Brain-computer interfaces (BCIs) have been widely focused and extensively studied in recent years for their huge prospect of medical rehabilitation and commercial applications. Transfer learning exploits the information in the source domain and applies in another different but related domain (target domain), and is therefore introduced into the BCIs to figure out the inter-subject variances of electroencephalography (EEG) signals. In this article, a novel transfer learning method is proposed to preserve the Riemannian locality of data structure in both the source and target domains and simultaneously realize the joint distribution adaptation of both domains to enhance the effectiveness of transfer learning. Specifically, a Riemannian graph is first defined and constructed based on the Riemannian distance to represent the Riemannian geometry information. To simultaneously align the marginal and conditional distribution of source and target domains and preserve the Riemannian locality of data structure in both domains, the Riemannian graph is embedded in the joint distribution adaptation (JDA) framework and forms the proposed Riemannian locality preserving-based transfer learning (RLPTL). To validate the effect of the proposed method, it is compared with several existing methods on two open motor imagery datasets, and both multi-source domains (MSD) and single-source domains (SSD) experiments are considered. Experimental results show that the proposed method achieves the highest accuracies in MSD and SSD experiments on three datasets and outperforms eight baseline methods, which demonstrates that the proposed method creates a feasible and efficient way to realize transfer learning.

Recently, reconfigurable intelligent surface (RIS) is emerging as an enabling technology to further enhance the performance of wireless communication systems. RIS is actually a metasurface which is consist of a large number of reflection elements. Each of the elements is capable of re-scatter the incident wave by an independent reflection coefficient. Signals can be redirected to the receiver by delicately assigning reflection coefficients to the elements. Thus a better receiving condition can be realized, and system performance can be enhanced. Existing studies already proved the superiority of general RIS-aided systems, e.g., single-user and multi-user, unicast and multicast, orthogonal and non-orthogonal multiple access. However, most of them considered ideal reflection elements. They assumed that full reflection can be realized at any values of phase-shift, which however is unrealistic due to energy dissipation in reflection elements. To circumvent this problem, in this paper we consider a practical phase shift model where reflection amplitude and phase shift are coupled. Specifically, under this practical model we try to maximize the achievable rate of an RIS-aided MIMO system. An alternating optimization based algorithm is proposed to solve the corresponding optimization problem. Reflection coefficients at the RIS and transmit covariance matrix at the transmitter are optimized alternately with the others being fixed. Simulation results are presented to manifest the validity of the proposed scheme.

To further enhance the performance of traditional wireless multicasting, in this paper a novel multicast system is proposed based on non-orthogonal multiple access (NOMA) and device-to-device relay (D2D-relay). In the new system, all destination users are sliced into multiple tiers according to their different levels of channel states, and the multicast data are also divided correspondingly and then disseminated through power domain NOMA transmission. After executing NOMA decoding independently and locally, the destination users can share multicast data through D2D links. The joint tier slicing and power control problem is considered in this paper to minimize the overall system power consumption in the new system while satisfying that all users can successfully receive all multicast data. Extensive simulations are conducted and the superiority of the proposed new multicast system over traditional multicast system and another benchmark multicast system based on orthogonal multiple access (OMA) and D2D-relay is verified.