The recent convergence of blockchain with federated learning (BCFL) offers enhanced data privacy, traceability, and ownership protection for decentralized data sharing. However, current research lacks consensus on accurate and fair data valuation methods for decentralized environments-a critical yet unresolved challenge. Existing Shapley value (SV)-based data valuation methods, while promising, rely on centralized frameworks ill-suited for decentralized environments and incur high computational costs. To address this, we propose a scalable dual-chain architecture: the main chain handles primary BCFL data-sharing operations, while a dedicated side chain offloads data valuation tasks. For the side chain, we introduce the Proof-of-Lightweight-Shapley (PoLS) protocol, enabling decentralized SV computation with theoretical convergence guarantees. PoLS leverages a proof-of-work-inspised mechanism to facilitate parallel and lightweight computation of shapley values using Monte Carlo sampling. Furthermore, we develop a SV-based data aggregation protocol to validate PoLS effectiveness and enhance resilience against low-quality data attacks. Extensive experiments confirm that PoLS significantly reduces computational overhead while maintaining an average deviation as low as 2.3% from the true shapley value. Moreover, in adversarial scenarios, the PoLS-driven aggregation strategy improves model accuracy by 25.7% over baselines, thereby strengthening the robustness and fairness of the decentralized valuable data sharing.

Unmanned surface vehicle (USV) formation navigation in constrained harbor environments presents challenges for maritime Internet of Things (IoT) applications, including surveillance, inspection, and rapid response missions. This paper addresses the trajectory planning problem for high-speed penetration, specifically focusing on minimizing oscillations in the reference trajectories. We propose S-Flocking, an integrated framework that reduces abrupt transitions in behavioral vector synthesis through the sigmoid-based transformation. For global path planning, we propose the Bisection-based Rewiring RRT* (B-RRT*) algorithm, which combines bidirectional search with dual backtracking optimization and bisection-based node insertion strategy to enhance computational efficiency and path quality. For reactive local planning, we introduce a dual-layer avoidance strategy (DAS) that integrates COLREGs-compliant maneuvers with 1-vs-1 emergency mechanism, resolving conflicts between formation-keeping and avoidance in dynamic encounters. Comprehensive simulations demonstrate that S-Flocking generates smoother reference trajectories with effective formation cohesion, thereby providing a robust solution for autonomous cooperative navigation in partially unknown dynamic maritime environments.

Autonomous on-ramp merging requires a safe and efficient decision and planning framework to navigate dynamic and complex traffic scenarios. Reinforcement Learning (RL) offers adaptability in such environments but struggles to ensure safety in unseen situations, whereas Model Predictive Control (MPC) can enforce safety constraints but relies on accurate models. To leverage the strengths of both, we propose a hierarchical Safety-Augmented RL and MPC integration approach, i.e., SARMI, for decision and planning in the on-ramp merging scenario. The high-level decision-making layer employs a discrete Soft Actor-Critic (SAC-Discrete) algorithm enhanced with the Augmented Lagrangian method to generate actions. MPC generates reference trajectories based on RL actions and feeds predicted states back to RL for reward and cost design, ensuring consistency between RL and MPC. Our key innovations include: i) incorporating a Gaussian-based risk field model into the cost design of constrained RL, which quantifies collision risk based on MPC-predicted states, enabling agents to make proactive decisions; ii) an Augmented Lagrangian SAC-Discrete method with barrier-like quadratic penalties to promote compliance with safety constraints and alleviate oscillations in dual gradient descent; iii) theoretical analysis proving the equivalence between the optimal solutions of the primal and dual problems in Augmented Lagrangian SAC-Discrete and iv) a dual safety mechanism combining action masking (to filter invalid actions) and action shielding (to replace unsafe actions in RL), enhancing safety during the exploration and execution stages, respectively. Experiments demonstrate the superiority of our method over the baseline SAC-Discrete, showing improved safety and efficiency.

Phase-modulated continuous-wave (PMCW) radar is a promising technology for internet of things (IoT) applications due to its high resilience against interference and compatibility with communication systems. To address the computational burden in PMCW radar, conventional approaches accumulate sequences in the time domain to reduce the number of samples used for Doppler processing in velocity estimation. However, the direct summation of sequences without phase compensation leads to severe performance degradation caused by the Doppler effect, including velocity ambiguity and signal-to-noise ratio loss. Therefore, this paper proposes a low-complexity sequence accumulation method for detecting fast-moving targets. By adjusting the desired velocity search region, the proposed method simultaneously resolves the velocity ambiguity and reduces computational complexity. Simulation results show that the proposed method achieves a processing time approximately 3.5 times faster than the non-accumulation method. In addition, the proposed method compensates for the Doppler phase difference along the slow-time to enable coherent accumulation over sequences, thereby preventing performance degradation and achieving higher detection probability compared to conventional accumulation methods.

Self-supervised monocular depth estimation leverages photometric consistency across views as supervisory signals, yet inherently suffers from compromised geometric fidelity at object boundaries, surface smoothness, and cross-view consistency due to 3D structural information loss during 2D projection. This paper introduces Diff3D-Net, a novel framework integrating multi-level 3D geometric constraints to overcome these limitations. Key innovations include: a fixed-parameter RPM-Net point cloud registration module generating geometrically grounded supervision; a pose consistency constraint that directly optimizes the pose network using RPM-Net-refined transformations to suppress error accumulation; a dual-path photometric constraint adaptively fusing reconstructions from original and optimized poses to enhance low-texture robustness; and a multi-view 3D structural consistency constraint enforcing global point alignment via minimum residual selection. Evaluations on KITTI and DDAD show Diff3D-Net's state-of-the-art performance, achieving 0.094/4.220 Abs. Rel/RMSE on KITTI. The model also exhibits strong cross-domain generalization in zero-shot settings, outperforming competitors on Cityscapes, NYU Depth V2 and DrivingStereo datasets without fine-tuning, highlighting its IoT application potential. Despite higher training costs, its inference achieves an optimal speed-accuracy trade-off, with a lightweight variant reaching 51.73 FPS -sufficient to enable real-time applications such as UAV navigation on resource-limited IoT edge devices. Ablations confirm each constraint's contribution, and qualitative results demonstrate robustness in dynamic scenes and complex geometries.

Radar-based, contactless in-home vital sign monitoring can significantly improve the healthcare of the rapidly growing aging population by enabling early detection of clinical events and continuous tracking of physiological states, while maintaining privacy, comfort, and user compliance. We introduce a unique experimental environment, collected for over a year – consisting of populated test homes, equipped with arrays of continuous-wave radar sensors placed on ceilings and walls, with occupants engaged in everyday activities within the homes. We applied and compared multiple analytical methods to extract heart rate from the radar data, using combinations of normalized cross-correlation (NXC), machine learning with individual sensors, and sensor fusion. Our results show that sensor fusion leads to the lowest mean absolute errors, reduced by over 30% from single-sensor machine learning models and over 80% from NXC. We additionally observed that a dense sensor array design enhances signal quality and reduces noise by more than 15%, providing more accurate heart rate monitoring even during dynamic activities, such as jumping and walking. These results highlight the effectiveness of sensor fusion and dense array designs in real-world settings. Our novel experimental platform and analytical results represent a meaningful step toward scalable application of radar technology in real residential settings for non-contact health monitoring.

Unmanned maritime surveillance systems (UMSS), consisting of Unmanned Surface Vessels (USVs) and smart buoys, as a typical application of the maritime Communication-Computing Converged Network (CCCN), play a crucial role in combating illegal fishing, smuggling, and piracy. However, the 2D images on which UMSS relies often lack spatial depth, and although 3D point cloud data can compensate for this limitation, complex and harsh weather degrades their quality and increases data volume. This increases the communication and computational burden, making efficient resource allocation far more complex, impairing UMSS’s responsiveness in dynamic maritime conditions. To address these issues, this paper proposes an edge intelligence-based real-time resource allocation framework for UMSS. Unlike existing works that only consider resource allocation, our framework also accounts for the impact of 3D point cloud data quality on system performance in harsh weather conditions. Specifically, it first leverages 3D point cloud data with a dehazing algorithm to mitigate haze-induced data expansion. Then we formulate an optimization problem to dynamically trade off throughput, latency, and Energy Consumption (EC) under varying maritime conditions. Finally, we utilize Deep Deterministic Policy Gradient (DDPG) methods to solve this problem. Additionally, we design task allocation strategies tailored to different maritime services. Simulation results show that our approach reduces average latency by 13%, lowers EC by 20%, and increases throughput by 10%, to guarantee the timely fulfillment of all tasks and improve the efficiency of UMSS under harsh weather conditions.

Defect inversion is a critical quantitative assessment technique to evaluate the integrity of energy transportation systems. In practice, non-stationary states and frequent situation switches in real-world industrial processes make it difficult to acquire ideal sensor signals with precise annotations, there by compromising accurate defect inversion. Simulation systems is a viable solution. However, the dramatic inter-domain disparity between simulated and real system poses challenges to existing studies. Moreover, intra-domain divergence within the real energy transportation systems is also unexplored. To address the above issues, an intra- and inter-domain interactive learning method is proposed for simulation-to-real (sim2real) defect inversion. Specifically, reality-aware feature stylization is proposed which diversifies simulated signals with real style variants, so as to alleviate the domain gap initially. Then, bi-level inter-domain alignment is proposed to not only penalize hard-to-align samples in abstract feature space, but also innovatively exploits intrinsic geometric relations of sensor signals to guide inter-domain gaps. Next, physical-informed intra-domain alignment is proposed where a similarity matrix is constructed to pursue the consistency between sensor signals and physical knowledge, and it is also served as pseudo-labels to constrain inter-domain adaptation, so that inter- and intra- domain adaptation interact with united strength to enhance information exchange. Finally, the proposed method is systematically validated on energy transportation experimental platform and competitive results are obtained, which shows its promise in real-world industrial processes.