ABSTRACT Dependable transfer of brain signals from motor imagery EEG must adhere to strict latency and memory constraints while maintaining accuracy in the face of noise and drift. A hybrid Echo State Network‐Long Short‐Term Memory (ESN‐LSTM) pipeline is shown here. This pipeline combines robust preprocessing with automatic time‐lag alignment between predictions and targets. To capture structure lost by linear errors alone, the evaluation combines traditional regression metrics (MSE/MAE/ R 2 ) with nonlinear dependence measures (time‐resolved distance correlation and HHG omnibus testing). A leave‐one‐subject‐out (LOSO) procedure is used to investigate cross‐subject generalization, and multi‐signal‐to‐noise ratio (SNR) stress tests are conducted to evaluate robustness. In ablation experiments, the impact of filtering, normalization, alignment, and important hyperparameters (reservoir size/spectral radius/leak; LSTM layers/hidden/dropout) is isolated. On the other hand, an efficiency snapshot reports latency and RAM usage under identical workloads for ESN‐Only, LSTM‐Only, and ESN‐LSTM modes. Across all participants, the hybrid consistently improves explained variance and dependence scores while maintaining a controlled computational cost, indicating that it is feasible for near‐real‐time use. All tables and Figures are regenerated from logged CSVs using scripts, configuration files, and fixed seeds. This ensures that reproducibility is maintained.

ABSTRACT Reliable crowd counting for IoT video analytics requires strong generalization across heterogeneous edge cameras. However, models trained on a labeled source domain often degrade on unseen cameras due to shifts in appearance, viewpoint, and density statistics. We propose MDANet, a deployment‐oriented framework for cross‐domain crowd counting that performs complementary alignment at three levels while keeping test‐time inference identical to a lightweight backbone. At the data level, Fourier Amplitude Mix reduces camera‐dependent style gaps by mixing low‐frequency amplitudes. At the feature level, global–local High‐Entropy Adversarial Regularization suppresses domain‐discriminative cues under spatial heterogeneity. At the domain level, Density‐Conditional Alignment modulates alignment strength according to predicted density to mitigate congestion‐dependent errors. Extensive experiments show that MDANet achieves competitive or state‐of‐the‐art accuracy with a favorable accuracy‐efficiency trade‐off, and additional evaluations under common stream degradations confirm its stability for edge deployment.

ABSTRACT This paper introduces a novel federated learning framework that integrates dynamic blockchain sharding and Byzantine fault tolerance mechanisms (FL‐Sharding‐BFT). The proposed framework dynamically adjusts shard configurations based on node capacity and network conditions, reducing communication overhead and enhancing model synchronization. The Byzantine fault tolerance mechanism further ensures robustness by identifying and isolating malicious nodes during model aggregation. It also ensures stronger robustness and faster convergence, highlighting its scalability and security for federated learning in 5G edge networks.

ABSTRACT In healthcare 5.0, the Internet of Medical Things (IoMT) is driving rapid developments that require energy‐efficient and secure solutions. This paper proposes a novel hybrid approach that integrates Public Key Infrastructure (PKI) with Spiking Neural Networks (SNNs) to enhance secure communication and reduce energy consumption in IoMT‐based healthcare systems. The PKI framework provides secure device authentication and encrypted data transmission, while the biologically inspired SNNs enable low‐power, real‐time anomaly detection. Furthermore, the Zebra Optimization Algorithm (ZOA) further enhances system performance. A simulation study shows that the proposed model outperforms existing protocols from a delay perspective, throughput perspective, packet delivery ratio perspective, and residual energy perspective, showing that it is suitable for real‐time, resource‐constrained healthcare.

ABSTRACT This paper presents a particle swarm optimization (PSO)‐based approach to enhance node localization accuracy in wireless sensor networks (WSNs). Traditional range‐based methods such as RSSI suffer from high positioning errors, while conventional DV‐Hop algorithms rely on assumptions that often fail under real‐world conditions. To overcome these limitations, the proposed method integrates a bounding box technique to constrain the initial search space, along with anticipatory and refinement strategies to address flip ambiguity. Additionally, inter‐node distance information, including distances between unknown nodes, is leveraged to further improve localization accuracy. Simulation results demonstrate that the proposed PSO‐based approach significantly reduces localization error, enhances accuracy and coverage, and lowers variance compared to standard DV‐Hop and Improved DV‐Hop (IDV‐Hop) algorithms. These improvements contribute to more accurate, reliable, and computationally efficient localization in WSNs.

ABSTRACT The increasing complexity and interdependence of global supply chains necessitate innovative solutions to enhance collaboration while safeguarding data privacy. This paper presents a lightweight multi‐blockchain framework designed to address the challenges of privacy‐preserving collaboration in Internet of Things (IoT)‐driven supply chains. By integrating multiple blockchain networks, the proposed framework ensures data confidentiality, integrity, and traceability across various stakeholders. The system employs advanced cryptographic techniques, including zero‐knowledge proofs and differential privacy, to protect sensitive information during data exchange and processing. Additionally, the framework incorporates lightweight consensus mechanisms to accommodate the resource constraints of IoT devices. Experimental evaluations demonstrate the effectiveness of the proposed framework in improving data privacy and system scalability compared to existing solutions.

ABSTRACT Single backhaul networks face significant limitations in simultaneously satisfying the user coverage and stringent service requirements of future Industrial Internet of Things (IIoT) systems envisioned for Industry 5.0. To address this challenge, this study proposes a joint optimization method for dual‐mode backhaul link selection and power allocation based on satellite‐terrestrial base station cooperation, tailored for resilient IIoT connectivity. The approach constructs a Space‐Terrestrial Integrated Network (STIN) backhaul architecture to ensure ubiquitous coverage and reliable communication for distributed industrial assets. It formulates an optimization problem with the objective of maximizing delay tolerance resilience for critical IIoT applications, such as real‐time monitoring and autonomous control. A decomposition optimization strategy is employed for solution—applying the Hungarian algorithm for the link allocation subproblem and the Lagrangian dual method for the power allocation subproblem. Simulation results demonstrate that compared to existing algorithms, the proposed method reduces average latency by 18% for uRLLC packets and 13% for eMBB packets, significantly enhancing system robustness while effectively reducing user service latency. This advancement supports the ultra‐reliable, low‐latency communication essential for human‐centric, intelligent, and sustainable operations in Industry 5.0 environments.

ABSTRACT In today's world, most individuals are suffering from health complications due to sedentary lifestyles, food habits, and aging. This might have distracted them from their daily activities, and many individuals are not seeking medical help with their busy work schedules. To overcome these challenges, Wireless Body Area Network (WBAN) provides challenging solutions with continuous and remote monitoring systems. Tiny sensors are either embedded in the body or worn by the patients for tracking physiological factors, including heart pulse rate (bilirubin), pressures (alkphos), and levels of hemoglobin (albumin). The inserted or implanted sensors collect the data and transmit it to the central sink node for the aggregation of the readings, and reports are comprehensively generated and transmitted to the medical professionals for analysis. While transmitting the sensitive data, a security breach might occur with the wireless channels. To enhance security, Enhanced Juels and Sudan (EJS) encryption algorithm is deployed to safeguard both the data and sensor systems. This prevents unauthorized access to data. Whale Optimization Algorithm is deployed for improving optimized network performance and security. Simulations are conducted in the Cooja simulator and demonstrate enhanced convergence efficacy, optimal weight computation, and mitigated error rates, including False Rejection Rate (FRR) and False Acceptance Rate (FAR) for a robust health monitoring system.

ABSTRACT Cloud‐assisted physical education teaching is an important direction for the development of smart education. However, the data processing and transmission of videos severely restrict the real‐time performance of action analysis. To this end, this paper proposes an efficient edge cloud‐assisted student movement recognition framework based on the graph convolutional network and human skeleton data. On the edge server, we use the YOLO‐pose algorithm to generate robust human skeleton sequences and design an improved spatial–temporal dual stream graph convolutional neural network with an early fusion structure, which introduces the node weight module and the dynamic graph module to exploit long‐distance dependency relationships of nodes. In the cloud server, we use a federated learning framework based on the density clustering mechanism to collaboratively train and aggregate parameters of models scattered across edge nodes. The experimental results show that our proposed model achieves excellent recognition accuracy on the self‐built sports action dataset, providing an effective solution for intelligent and real‐time feedback in outdoor sports teaching.

ABSTRACT RNA base pairing probability is a critical determinant of RNA structure, influencing its stability and functionality. This property is pivotal for mRNA vaccine design, disease diagnosis, and drug discovery. We introduce an innovative Internet of Medical Things (IoMT)‐driven Conv‐Transformer model to predict RNA base pairing probabilities. Leveraging IoMT's edge intelligence, our model efficiently collects and processes real‐time RNA sequence data from diverse medical devices, thereby enhancing our model's feature extraction and prediction accuracy. Given the data‐driven advantages from IoMT, the proposed Conv‐Transformer model utilizes word embedding and position coding to encode primary and secondary RNA structures. It then employs a multi‐headed self‐attention mechanism to capture global base interactions and one‐dimensional convolution to extract local features. Our experiments demonstrate superior performance compared to five classical machine learning models, reducing RMSE and MAE by at least 16% and 20%, respectively. This work highlights the potential of IoMT and advanced deep learning techniques in enhancing RNA structure prediction.