Search Results (266)

6G low-power and lossy network (6G LLN) is a kind of distributed network designed for IoT and edge computing scenarios of the sixth-generation mobile communication technology. Its routing technologies should fully consider characteristics of green and low carbon, constrained nodes, lossy links, etc. This paper proposes an improved routing protocol for low-power and lossy networks (I-RPL) to better suit the characteristics of 6G LLN and meet its application requirements. I-RPL has designed new context-aware routing metrics, which include the residual energy indicator, buffer utilization ratio, ETX, delay, and hop count to meet multi-dimensional network QoS requirements. The candidate parent and its preferred parent’s residual energy indicator and buffer utilization ratio are calculated recursively to reduce the effect of upstream parents. ETX and delay calculating methods are improved to ensure a better performance. Moreover, I-RPL has optimized the network construction process to improve energy and protocol efficiency. I-RPL has designed scientific multiple routing metrics evaluation theories (Lagrangian multiplier theories), proposed new rank computing and optimal route selecting mechanisms to simplify protocol, and optimized broadcast suppression and network reliability. Finally, theoretical analysis and experiment results show that the average end-to-end delay of I-RPL is 13% lower than that of RPL; the average alive node number increased 11% and so on. So, I-RPL can be applied well to the 6G LLN and is superior to RPL and its improvements. Full article

Vegetation phenology (VP) is a crucial biological indicator for monitoring terrestrial ecosystems and global climate change. However, VP monitoring using traditional remote sensing vegetation indices has significant limitations in precise analysis. Furthermore, most studies have overlooked the distinction between stable and short-term VP in relation to climate change and have failed to clearly identify the seasonal variation in the impact of climatic factors on stable VP (SVP). This study compared the accuracy of solar-induced chlorophyll fluorescence (SIF) and three traditional vegetation indices (e.g., Normalized Difference Vegetation Index) for estimating SVP in China, using ground-based data for validation. Additionally, this study employs Sen’s slope, the Mann–Kendall (MK) test, and the Hurst index to reveal the spatiotemporal evolution of the Start of Season (SOS), End of Season (EOS), and Length of Growing Season (LOS) over the past two decades. Partial correlation analysis and random forest importance evaluation are used to accurately identify the key climatic drivers of SVP across different climate zones and to assess the seasonal contributions of climate to SVP. The results indicate that (1) phenological metrics derived from SIF data showed the strongest correlation coefficients with ground-based observations, with all correlation coefficients (R) exceeding 0.69 and an average of 0.75. (2) The spatial distribution of SVP in China has revealed three primary spatial patterns: the Tibetan Plateau, and regions north and south of the Qinling–Huaihe Line. From arid, cold-to-warm, and humid regions, the rate of SOS advancement gradually increases; EOS transitions from earlier to nearly unchanged; and the rate of LOS delay increases accordingly. (3) The spring climate primarily drives the advancement of SOS across China, contributing up to 70%, with temperatures generally having a negative effect on SOS (r = −0.53, p < 0.05). In contrast, EOS is regulated and more complex, with the vapor pressure deficit exerting a dual ‘limitation–promotion’ effect in autumn (r = −0.39, p < 0.05) and summer (r = 0.77, p < 0.05). This study contributes to a deeper scientific understanding of the interannual variability in SVP under seasonal climate change. Full article

Forest fires pose significant threats to ecosystems, human life, and the global climate, necessitating rapid and reliable detection systems. Traditional fire detection approaches, including sensor networks, satellite monitoring, and centralized image analysis, often suffer from delayed response, high false positives, and limited deployment in remote areas. Recent deep learning-based methods offer high classification accuracy but are typically computationally intensive and unsuitable for low-power, real-time edge devices. This study presents an autonomous, edge-based forest fire and smoke detection system using a lightweight MobileNetV2 convolutional neural network. The model is trained on a balanced dataset of fire, smoke, and non-fire images and optimized for deployment on resource-constrained edge devices. The system performs near real-time inference, achieving a test accuracy of 97.98% with an average end-to-end prediction latency of 0.77 s per frame (approximately 1.3 FPS) on the Raspberry Pi 5 edge device. Predictions include the class label, confidence score, and timestamp, all generated locally without reliance on cloud connectivity, thereby enhancing security and robustness against potential cyber threats. Experimental results demonstrate that the proposed solution maintains high predictive performance comparable to state-of-the-art methods while providing efficient, offline operation suitable for real-world environmental monitoring and early wildfire mitigation. This approach enables cost-effective, scalable deployment in remote forest regions, combining accuracy, speed, and autonomous edge processing for timely fire and smoke detection. Full article

Recent advances in exoskeleton-assisted rehabilitation have highlighted the significance of lower limb movement intention recognition through deep learning. However, discrete motion phase classification and continuous real-time joint kinematics estimation are typically handled as independent tasks, leading to temporal misalignment or delayed assistance during dynamic movements. To address this issue, this study presents iTransformer-DTL, a dual-task learning framework with an improved Transformer designed to identify end-to-end locomotion modes and predict joint trajectories during sit-to-stand transitions. Employing a learnable query mechanism and a non-autoregressive decoding approach, the proposed iTransformer-DTL can produce the complete output sequence at once, without relying on any previously generated elements. The proposed framework has been tested with a dataset of lower limb movements involving seven healthy individuals and seven stroke patients. The experimental results indicate that the proposed framework achieves satisfactory performance in dual tasks. An average angle prediction Mean Absolute Error (MAE) of 3.84° and a classification accuracy of 99.42% were obtained in the healthy group, while 4.62° MAE and 99.01% accuracy were achieved in the stroke group. These results suggest that iTransformer-DTL could support adaptable rehabilitation exoskeleton controllers, enhancing human–robot interactions. Full article

Rapid access to defibrillators, blood products, and time-critical medicines can improve survival, yet urban congestion and fragmented infrastructure delay deliveries. We present and evaluate an end-to-end framework for beyond-visual-line-of-sight (BVLOS) UAV logistics in Essex (UK), integrating (I) strategic depot placement, (II) a hybrid obstacle-aware route planner, and (III) a time-window-aware (TWA) Mixed-Integer Linear Programming (MILP) scheduler coupled to a battery/temperature feasibility model. Four global planners—Ant Colony Optimisation (ACO), Genetic Algorithm (GA), Particle Swarm Optimisation (PSO), and Rapidly Exploring Random Tree* (RRT*)—are paired with lightweight local refiners, Simulated Annealing (SA) and Adaptive Large-Neighbourhood Search (ALNS). Benchmarks over 12 destinations used real Civil Aviation Authority no-fly zones and energy constraints. RRT*-based hybrids delivered the shortest mean paths: RRT* + SA and RRT* + ALNS tied for the best average length, while RRT* + SA also achieved the co-lowest runtime at $v=60\mathrm{km}{\mathrm{h}}^{-1}$. The TWA-MILP reached proven optimality in 0.11 s, showing that a minimum of seven UAVs are required to satisfy all 20–30 min delivery windows in a single wave; a rolling demand of one request every 15 min can be sustained with three UAVs if each sortie (including service/recharge) completes within 45 min. To validate against a state-of-the-art operations-research baseline, we also implemented a Vehicle Routing Problem with Time Windows (VRPTW) in Google OR-Tools, confirming that our hybrid planners generate competitive or shorter NFZ-aware routes in complex corridors. Digital-twin validation in AirborneSIM confirmed CAP 722-compliant, flyable trajectories under wind and sensor noise. By hybridising a fast, probabilistically complete sampler (RRT*) with a sub-second refiner (SA/ALNS) and embedding energy-aware scheduling, the framework offers an actionable blueprint for emergency medical UAV networks. Full article

The increasing demand for high-performance and low-power hardware in artificial intelligence (AI) applications—such as speech recognition, facial recognition, and object detection—has driven the exploration of advanced memory designs. Convolutional neural networks (CNNs) and deep neural networks (DNNs) require intensive computational resources, leading to significant challenges in terms of memory access time and power consumption. Compute-in-Memory (CIM) architectures have emerged as an alternative by executing computations directly within memory arrays, thereby reducing the expensive data transfer between memory and processor units. In this work, we present a 6T SRAM-based CIM architecture implemented using FinFET technology, aiming to reduce both power consumption and access delay. We explore and simulate three different SRAM cell structures—PLNA (P-Latch N-Access), NLPA (N-Latch P-Access), and SE (Single-Ended)—to assess their suitability for CIM operations. Compared to a reference 10T XNOR-based CIM design, our results show that the proposed structures achieve an average power consumption approximately 70% lower, along with significant delay reduction, without compromising functional integrity. A comparative analysis is presented to highlight the trade-offs between the three configurations, providing insights into their potential applications in low-power AI accelerator design. Full article

With the rapid development of edge computing and deep learning, the efficient deployment of deep neural networks (DNNs) on resource-constrained terminal devices faces multiple challenges (background), such as execution delay, high energy consumption, and resource allocation costs. This study proposes an improved Multi-Objective Particle Swarm Optimization (MOPSO) algorithm for PSO. Unlike the conventional PSO, our approach integrates a historical optimal solution detection mechanism and a dynamic temperature regulation strategy to overcome its limitations in this application scenario. First, an end–edge–cloud collaborative computing framework is constructed. Within this framework, a multi-objective optimization model is established, aiming to minimize time delay, energy consumption, and cloud configuration cost. To solve this model, an optimization method is designed that integrates a historical optimal solution detection mechanism and a dynamic temperature regulation strategy into the MOPSO algorithm. Experiments on six types of DNNs, including the Visual Geometry Group (VGG) series, have shown that this algorithm reduces execution time by an average of 58.6%, the average energy consumption by 61.8%, and optimizes cloud configuration costs by 36.1% compared to traditional offloading strategies. Its Global Search Capability Index (GSCI) reaches 92.3%, which is 42.6% higher than the standard PSO algorithm. This method provides an efficient, secure, and stable cooperative computing solution for multi-constraint task unloading in an edge computing environment. Full article

As network environments become increasingly dynamic and users’ Quality of Service (QoS) demands grow more diverse, efficient and adaptive routing strategies are urgently needed. However, traditional routing strategies suffer from limitations such as poor adaptability to fluctuating traffic, lack of differentiated service handling, and slow convergence in complex network scenarios. To this end, we propose a routing strategy based on multi-agent deep deterministic policy gradient for differentiated QoS services (RS-MADDPG) in a software-defined networking (SDN) environment. First, network state information is collected in real time and transmitted to the control layer for processing. Then, the processed information is forwarded to the intelligent layer. In this layer, multiple agents cooperate during training to learn routing policies that adapt to dynamic network conditions. Finally, the learned policies enable agents to perform adaptive routing decisions that explicitly address differentiated QoS requirements by incorporating a custom reward structure that dynamically balances throughput, delay, and packet loss according to traffic type. Simulation results demonstrate that RS-MADDPG achieves convergence approximately 30 training cycles earlier than baseline methods, while improving average throughput by 3%, reducing latency by 7%, and lowering packet loss rate by 2%. Full article

Efficient task offloading for delay-sensitive applications, such as autonomous driving, presents a significant challenge in multi-hop Vehicular Edge Computing (VEC) networks, primarily due to high vehicle mobility, dynamic network topologies, and complex end-to-end congestion problems. To address these issues, this paper proposes a graph attention-based reinforcement learning algorithm, named GAPO. The algorithm models the dynamic VEC network as an attributed graph and utilizes a graph neural network (GNN) to learn a network state representation that captures the global topological structure and node contextual information. Building on this foundation, an attention-based Actor–Critic framework makes joint offloading decisions by intelligently selecting the optimal destination and collaboratively determining the ratios for offloading and resource allocation. A multi-objective reward function, designed to minimize task latency and to alleviate link congestion, guides the entire learning process. Comprehensive simulation experiments and ablation studies show that, compared to traditional heuristic algorithms and standard deep reinforcement learning methods, GAPO significantly reduces average task completion latency and substantially decreases backbone link congestion. In conclusion, by deeply integrating the state-aware capabilities of GNNs with the decision-making abilities of DRL, GAPO provides an efficient, adaptive, and congestion-aware solution to the resource management problems in dynamic VEC environments. Full article

Federated learning (FL) has emerged as a powerful approach for privacy-preserving model training in autonomous vehicle networks, where real-world deployments rely on multiple roadside units (RSUs) serving heterogeneous clients with intermittent connectivity. While most research focuses on single-server or hierarchical cloud-based FL, multi-server FL can alleviate the communication bottlenecks of traditional setups. To this end, we propose an edge-based, multi-server FL (MS-FL) framework that combines performance-driven aggregation at each server—including statistical weighting of peer updates and outlier mitigation—with an application layer handover protocol that preserves model updates when vehicles move between RSU coverage areas. We evaluate MS-FL on both MNIST and GTSRB benchmarks under shard- and Dirichlet-based non-IID splits, comparing it against single-server FL and a two-layer edge-plus-cloud baseline. Over multiple communication rounds, MS-FL with the Statistical Performance-Aware Aggregation method and Dynamic Weighted Averaging Aggregation achieved up to a 20-percentage-point improvement in accuracy and consistent gains in precision, recall, and F1-score (95% confidence), while matching the low latency of edge-only schemes and avoiding the extra model transfer delays of cloud-based aggregation. These results demonstrate that coordinated cooperation among servers based on model quality and seamless handovers can accelerate convergence, mitigate data heterogeneity, and deliver robust, privacy-aware learning in connected vehicle environments. Full article

Routing in flying ad hoc networks (FANETs) is hindered by high mobility, trajectory-induced topology dynamics, and energy constraints. Conventional topology-based or position-based protocols often fail due to stale link information and limited neighbor awareness. This paper proposes a trajectory-informed routing protocol enhanced by Q-learning: Traj-Q-GPSR, tailored for mission-oriented UAV swarm networks. By leveraging mission-planned flight trajectories, the protocol builds time-aware two-hop neighbor tables, enabling routing decisions based on both current connectivity and predicted link availability. This spatiotemporal information is integrated into a reinforcement learning framework that dynamically optimizes next-hop selection based on link stability, queue length, and node mobility patterns. To further enhance adaptability, the learning parameters are adjusted in real time according to network dynamics. Additionally, a delay-aware queuing model is introduced to forecast optimal transmission timing, thereby reducing buffering overhead and mitigating redundant retransmissions. Extensive ns-3 simulations across diverse mobility, density, and CBR connections demonstrate that the proposed protocol consistently outperforms GPSR, achieving up to 23% lower packet loss, over 80% reduction in average end-to-end delay, and improvements of up to 37% and 52% in throughput and routing efficiency, respectively. Full article

Background/Objectives: The most common limb-salvage procedure for end-stage ankle arthropathy with severe bone defect is arthrodesis. Successful fusion requires rigid metal fixation, effective filling of the bone defect space, and maximal securing of the contact area between the tibia and talus. In cases with severe bone defect, sufficient grafting using autogenous bone alone is limited, and there is still controversy regarding the effectiveness of allogeneic or xenogeneic bone grafting. This study aimed to evaluate the intermediate-term clinical outcomes after shortening arthrodesis using fibular osteotomy for ankle arthropathy with severe bone defect. Methods: Twenty-two patients with shortening ankle arthrodesis were followed up ≥ 3 years. All operations were performed by one senior surgeon and consisted of internal fixation with anterior fusion plate, fibular osteotomy, and autogenous bone grafting. The causes of ankle joint destruction were failed total ankle arthroplasty (7 cases), neglected ankle fracture (6 cases), delayed diagnosis of degenerative arthritis (5 cases), avascular necrosis of talus (2 cases), and diabetic neuroarthropathy (2 cases). Clinical outcomes including daily living and sport activities were evaluated with the Foot and Ankle Outcome Score (FAOS) and the Foot and Ankle Ability Measure (FAAM). Radiological evaluation included fusion rate, time to fusion, leg length discrepancy, and degenerative change in adjacent joints. Results: The FAOS and FAAM scores significantly improved from a mean of 21.8 and 23.5 points preoperatively to 82.2 and 83.4 points at final follow-up, respectively (p < 0.001). Visual analogue scale for pain during walking significantly improved from a mean of 7.7 points preoperatively to 1.4 points at final follow-up (p < 0.001). The average time to complete fusion was 16.2 weeks, and was achieved in all patients. The average difference in leg length compared to the contralateral side was 11.5 mm based on physical examination, and 13.8 mm based on radiological examination. During the average follow-up of 56.2 months, no additional surgery was required due to progression of degenerative arthritis in the adjacent joints, and no cases required the use of height-increasing insoles in daily life. Conclusions: Shortening ankle arthrodesis using fibular osteotomy and anterior fusion plate demonstrated satisfactory intermediate-term clinical outcomes and excellent fusion rate. Advantages of this procedure included rigid fixation, preservation of the subtalar joint, effective filling of the bone defect space, and maximal securing of the contact area for fusion. The leg length discrepancy, which was concerned to be a main shortage, resulted in no significant clinical symptoms or discomfort in most patients. Full article

Knowledge extraction and sharing is one of the biggest challenges organizations face to ensure successful and long-lasting knowledge repositories. The North Carolina Department of Transportation (NCDOT) commissioned a web-based knowledge management program called Communicate Lessons, Exchange Advice, Record (CLEAR) for end-users to promote employee-generated innovation and to institutionalize organizational knowledge. Reusing knowledge from an improperly managed database is problematic and potentially causes substantial financial loss and reduced productivity for an organization. Poorly managed databases can hinder effective knowledge dissemination across the organization. Data-driven dashboards offer a promising solution by facilitating evidence-driven decision-making through increased information access to disseminate, understand and interpret datasets. This paper describes an effort to create data visualizations in Tableau for CLEAR’s gatekeeper to monitor content within the knowledge repository. Through the three web-based strategic dashboards relating to lessons learned and best practices, innovation culture index, and website analytics, the information displays will aid in disseminating useful information to facilitate decision-making and execute appropriate time-critical interventions. Particular emphasis is placed on utility-related issues, as data from the NCDOT indicate that approximately 90% of projects involving utility claims experienced one or two such incidents. These claims contributed to an average increase in project costs of approximately 2.4% and schedule delays averaging 70 days. The data dashboards provide key insights into all 14 NCDOT divisions, supporting the gatekeeper in effectively managing the CLEAR program, especially relating to project performance, cost savings, and schedule improvements. The chronological analysis of the CLEAR program trends demonstrates sustained progress, validating the effectiveness of the dashboard framework. Ultimately, these data dashboards will promote organizational innovation in the long run by encouraging end-user participation in the CLEAR program. Full article

A maritime mobile ad hoc network (M-MANET) is an essential part of the maritime communication network and plays a key role in many maritime scenarios. However, the topology of M-MANET dynamically changes with the movement of vessels, which leads to unstable link states and poses the risk of data transmission interruption. In this paper, a mobility model for small unmanned surface vessels based on smooth Gaussian semi-Markovian and a trajectory prediction method for large vessels based on a bi-directional long short-term memory network are proposed to better simulate the nodes’ movement in the M-MANET. Then, a link available based routing metric is proposed for M-MANET scenarios, which incorporates factors of mobility model and vessel trajectory. Experiments demonstrate that compared with the benchmark methods, the proposed mobility model depicts the movement characteristics of vessels more accurately, the proposed trajectory prediction method achieves higher prediction accuracy and stability, the proposed routing metric scheme has a reduction of 14.59% in end-to-end delay, a 1.54% increase in packet delivery fraction, and a 4.43% increase in network throughput on average. Full article

Time-sensitive applications require the End-to-End (E2E) delay of wireless networks to be deterministic. For example, control signals in industrial automation, intelligent transportation, and telemedicine must be transmitted to their destinations within the millisecond range, with delay jitter controlled within the microsecond range. To formulate effective policies for maintaining E2E delay within a small deterministic range, it is essential to estimate the probability density function (PDF) of E2E delay. Data-driven methods based on mixture density networks have been employed to estimate the PDF of E2E delay in wireless networks. However, in WiFi networks, the estimation results produced by existing methods exhibit significant discrepancies and fluctuations when compared to actual measurements. Motivated by this, an improved estimation method is proposed, where the delay PDF is divided into three segments with different functional expressions that are coupled together. Moreover, the parameter estimation process is implemented in two stages. First, the two division thresholds for the three segments of the PDF are calculated based on the variation trend of E2E delay measurements. Second, the remaining parameters are obtained through training using an improved mixture density network. Experimental results indicate that the E2E delay PDF obtained by the proposed method exhibits a smaller gap compared to actual measurements than existing methods. Specifically, the mean absolute errors and average fluctuation amplitudes of tail probabilities at certain delay values decrease by at least one order of magnitude. Moreover, the multiple-segmentation feature of the proposed method enhances its robustness in situations where measurement data are affected by low levels of Gaussian noise. Full article