To tackle the non-stationarity of Electroencephalogram (EEG) signals in Rapid Serial Visual Presentation (RSVP) Brain-Computer Interfaces (BCIs) for patients with Amyotrophic Lateral Sclerosis (ALS), we propose an online template updating algorithm based on the Expectation-Maximization (EM) method. This algorithm formulates EEG signals as a combination of templates and trigger matrices, and updates the templates synchronously using maximum likelihood estimation. Experimental results show that the proposed Online-Updating Hierarchical Discriminant Component Analysis (OU-HDCA) algorithm, which combines the online updating scheme and Hierarchical Discriminant Component Analysis (HDCA), outperforms the traditional static HDCA algorithm remarkably in terms of classification accuracy and information transfer rate, with both indicators maintaining a continuous upward trend. The proposed algorithm adapts to the time-varying characteristics of EEG signals, improves the long-term reliability of BCI systems, and can serve as an effective solution for clinical assistive communication.

Motor Imagery Brain-Computer Interfaces (MI-BCI) are technologies that enable information exchange between the brain and external devices by acquiring electroencephalographic (EEG) signals from the brain and performing preprocessing, feature extraction, and classification. As one of the core paradigms of BCI, motor imagery holds broad application prospects in fields such as medical rehabilitation and entertainment. This paper systematically reviews the relevant technical development in the MI-BCI field, with a particular focus on decoding algorithms based on machine learning and deep learning. Furthermore, it explores future research directions and potential applications of MI-BCI.

Brain-computer interface (BCI) technology is progressively expanding from medical rehabilitation to non-medical scenarios, forming a diverse application ecosystem covering industrial safety, sleep health management, education and training, sports enhancement, olfactory interaction, and consumer entertainment. Based on a systematic review of the technological development context, this paper analyzes the technical principles, product implementation status, and commercialization progress of six typical non-medical application directions. It further discusses the industrial landscape, technology readiness level distribution, neuroethics, and data governance issues.

Invasive brain-computer interface (BCI) neural electrodes are pivotal components in BCI systems, but they are prone to failures such as electrical performance degradation, material fatigue, and biological rejection, which necessitate efficient and reliable testing methods. This article provides an overview of electrode detection methods in intrusive BCI, analyzing failure mechanisms, summarizing evaluation indicators, and sorting out detection methods from three aspects: electrical performance, mechanical performance, and biocompatibility. Finally, it looks forward to the development of standardization, platformization, and intelligence in invasive BCI electrode testing.

As a cutting-edge interdisciplinary technology, brain-computer interface (BCI) has become a core track of global technological competition and industrial layout. Based on a systematic review of the global BCI industry landscape and China’s BCI development status, this paper conducts an in-depth analysis of prevalent global challenges, outlines a coordinated development path covering technology, industry, application scenarios and governance, and presents feasible countermeasures and recommendations. It aims to advance the collaborative innovation, inclusive accessibility, safe and responsible development of the global BCI industry, and contribute to human health and sustainable development.

Conventional brain-computer interfaces often employ rigid materials such as silicon and metals, which struggle to meet the clinical requirements for long-term implantation. Hydrogels, with their high water content, adjustable modulus, excellent biocompatibility, and efficient ion transport, offer a feasible solution for constructing flexible and highly stable neural interfaces. By summarizing the advantages, performance, and chemical structures of hydrogels, this paper elaborates on the current state of their application in neural interfaces, provides an in-depth analysis of existing challenges. Also, this paper offers prospects for the design of high-performance hydrogels, the construction of biomimetic interfaces, flexible manufacturing technologies, and clinical translation pathways. The goal is to provide a relatively comprehensive reference for the fundamental research and engineering applications of hydrogel-based brain-computer interfaces.

Wireless implantable neural recording systems serve as core components of brain-computer interface (BCI) technologies, enabling neural signal acquisition and transmission for animal neuroscience and behavioral studies. Compared with conventional wired systems, these devices integrate microelectrodes, wireless power delivery, and communication modules, allowing long-term and continuous neural recording in freely behaving conditions. This capability significantly expands their applicability in complex behavioral experiments and multi-environment scenarios. Focusing on animal-based applications, this review systematically summarizes the current development of both commercial and research-grade wireless neural recording systems. Key technologies, including neural signal acquisition, wireless power transfer, and wireless data transmission, are comprehensively analyzed. Furthermore, representative applications across terrestrial, underwater, aerial, and other extreme environments are discussed, along with future development trends. This work aims to provide valuable insights for advancing BCI-related technologies and their broader applications.

To address the problems of insufficient multimodal feature mining and limited generalization in emotion recognition across different cultures, this paper proposes a Hierarchical Multi-scale Branch Residual Transformation-Canonical Correlation Attention Fusion Network, which consists of multi-scale feature extraction, canonical correlation analysis-based enhancement, and attention-weighted fusion modules. Based on the Chinese, German, and French subsets of the SJTU emotion electroencephalography dataset (SEED), namely SEED-CHN, SEED-GER, and SEED-FRA, intra-cultural subject-dependent, intra-cultural subject-independent, and cross-cultural subject-independent experiments were conducted. The experimental results show that, in the intra-cultural subject-dependent experiments, the proposed method outperforms several baseline methods overall, indicating that it has good emotion recognition performance and stability in different cultural scenarios.

The rich connotations of data assets give rise to inherent complexity in the process of data assetization. Currently, three major consensuses on data assets have been reached both domestically and internationally, whereas two prominent divergences remain with respect to the implementation paths of data assetization. These divergences further result in discrepancies in data assetization practices at home and abroad. This paper conducts a comprehensive assessment of the three mainstream paths of data assetization at present, and explores each path in detail from five aspects: objectives, applicable entities, practical effects, existing problems, and exploratory suggestions. It intends to offer path recommendations for various entities engaged in data assetization.

Based on an analysis of policy backgrounds, technological commonalities, industrial chain complementarities, and integration pathways, this paper proposes specific strategies to promote the synergistic development of the low-altitude economy and vehicle-road-cloud integration. These include establishing a national air-ground collaborative policy framework, formulating a unified technical standard system, and innovating business models for hybrid transportation services. The aim is to provide theoretical support and practical guidance for the development of related industries.

From the perspective of information science ontology, this paper reviews research on agent technologies addressing contemporary challenges in data flow, namely the dynamic ambiguity of trust boundaries, the multidimensional interweaving of security attributes, the real-time evolution of threat landscapes, and the inherent conflict between privacy protection and value utilization. First, it elucidates that ontology constitutes the foundational underpinning for achieving cross-domain semantic consensus in security. Second, it delineates governance pathways for low-altitude data security across four dimensions: dynamic fine-grained access control, trusted auditing across the entire data lifecycle, collaborative threat perception and response, and automated security compliance. Finally, it proposes four future research directions: embodied security agent interaction, low-altitude security digital twin testbeds, quantum-semantic integrated security communication, and generative adversarial security evolution. By establishing an understandable, trustworthy, evolvable, and immune urban low-altitude data security collaborative governance system, this study provides a systematic academic reference and a technical development roadmap for the field.

Currently, the energy consumption of data centers has become increasingly prominent. The HVAC system accounts for a significant proportion of total energy use, with the cooling system being the key to energy saving. Traditional cooling system control methods are simplistic and cannot fully utilize external natural cooling sources. This paper proposes an intelligent control method for cooling towers based on the composite control of wet-bulb temperature and water vapor partial pressure. By dynamically adjusting fan speed, the cooling system can adaptively match external climatic conditions. Experiments and applications show that this method can significantly improve cooling system efficiency, reduce PUE values, and provide technical support for green and low-carbon operation of data centers.

The rapid advancement of Artificial Intelligence (AI) technology has introduced novel requirements for data engineering, catalyzing a paradigm shift in data value morphology from data resources to high-quality datasets. This paper systematically analyzes the underlying mechanisms driving the transformation from traditional DataOps to AI-ready data engineering paradigms, and constructs an intelligence-oriented data engineering architecture. The study first delineates the connotation of AI-ready DataOps, conceptualizing it as an engineering methodology that integrates DataOps principles to efficiently, securely, and reliably supply data for AI scenarios. Building upon this foundation, an AI-ready DataOps capability reference framework is proposed, encompassing four critical phases: research and development, delivery, technical operations, and data governance, thereby establishing a clearly defined, step-by-step dataset pipeline. Furthermore, the study elaborates a five-step practical implementation methodology—inventory, construction, processing, governance, and operations—offering actionable guidelines for enterprise-level high-quality dataset construction. This study provides theoretical frameworks and methodological guidance for national high-quality dataset infrastructure development.