Search Results (864)

Improving the low-carbon operational performance of wastewater treatment facilities (WWTFs) is important for coordinated pollution reduction and carbon mitigation in China. The interprovincial carbon emission efficiency (CEE) of WWTFs reflects both the low-carbon performance of regional wastewater governance systems and cross-regional linkages shaped by technology diffusion, governance experience, and spatial adjacency. Using operational data from WWTFs in 30 provincial-level regions of China from 2010 to 2023, this study first measures provincial CEE with a super-efficiency slack-based measure (SBM) model incorporating undesirable outputs. It then integrates a modified gravity model, social network analysis, and the quadratic assignment procedure (QAP) to examine the evolution, structure, and formation mechanisms of the association network. The results show that the CEE of provincial WWTFs remained below the efficiency frontier overall, with periodic fluctuations and persistent structural differences. The model-inferred association network first contracted and then expanded, showing dense connections in eastern and central China and sparse connections in the northwest. Its node structure evolved from a dispersed multicore pattern to hub-centered concentration and then back toward a multicore configuration, while inter-block linkages showed a hierarchical transmission structure involving core spillovers, absorptive transmission, and peripheral reception. The QAP results indicate that interprovincial adjacency is the most stable correlate of network formation, whereas differences in technological innovation and capacity utilization show only stage-specific associations. These findings suggest that the CEE of WWTFs is not only a local performance outcome, but also a relational outcome shaped by regional linkages and structural positions. Therefore, low-carbon governance of WWTFs should move beyond single-region efficiency improvement and place greater emphasis on cross-regional collaboration, role-specific policy design, and leadership by core regions. Full article

Rock image recognition based on deep learning serves as a critical task in intelligent geological analysis and core documentation. Although recent advances in deep learning have established the technical foundation for rock image identification, its practical deployment remains constrained by two main bottlenecks: intricate feature representations and ubiquitous label noise. Complexity of rock image features limits the effectiveness of intelligent model applications. Additionally, manual mislabeling and coarse marking issues during lithology annotation are often overlooked, leading to label noise in training datasets. To resolve these challenges, this study proposes a robust learning framework that integrates a novelty-driven feature extraction module with noise-resistant loss functions. Specifically, a Rock Hierarchical Heterogeneous Attention (RHHA) module is designed for enhancing rock feature extraction capabilities by applying distinct spatial and channel attention biases at shallow and deep layers of the network. Furthermore, Symmetric Cross Entropy (SCE) loss is used for loss calculation to enhance the robustness of models under noisy conditions. Comparative experiments were carried out on a real-world dataset of metamorphic rock images from northern Jiangsu Province, China. The results show that the proposed training framework achieves the best performance in rock image identification tasks, outperforming baseline and other comparative methods. The RHHA module outperforms other existing attention modules in capturing the texture and semantic features of rocks. In particular, the SCE loss effectively mitigates overfitting and maintains excellent generalization capability in the presence of noisy labels. The proposed framework holds promise for providing new insights into rock image recognition tasks. Full article

This study assesses the multi-dimensional impacts of large-scale urban mass transit infrastructure on network sustainability and individual mobility behavior, using the newly operational Riyadh Metro as a case study. Adopting a qualitative interpretive synthesis method, the research integrates peer-reviewed literature, institutional policy reports, and early operational datasets to evaluate transit outcomes across network accessibility, spatial equity, and localized environmental constraints. The study shows the interactions between infrastructure provision, urban form, and climatic conditions. The results reveal that while the infrastructure significantly enhances raw corridor accessibility, its systemic impact is heavily constrained by extreme climatic conditions, institutional fragmentation, and critical first- and last-mile gaps. Extreme heat significantly reduces effective walking. Additionally, continued reliance on private vehicles, resulting in a hybrid mobility regime rather than a full modal transition. The analysis also further reveals uneven spatial distribution of the metro for easy accessibility. To overcome these barriers, this paper introduces the Climate-Constrained Sustainable Transit Transition Framework (CSTTF) as an analytical tool. The study concludes that infrastructure-led urban transformation in hyper-arid megacities requires a comprehensive policy shift that integrates land-use reform, behavioral incentives, and climate-adaptive urban design. The proposed Climate-Constrained Sustainable Transit Transition Framework (CSTTF) provides a transferable decision-support framework that can assist policymakers, urban planners, and transit agencies in designing climate-resilient metro systems, optimizing first- and last-mile connectivity, and informing sustainable transport planning in Riyadh and other hot-arid metropolitan regions facing similar environmental and urbanization challenges. Full article

Deep learning (DL) models have achieved remarkable progress in natural language processing, computer vision, content generation, and edge intelligence; however, their rapidly increasing computational complexity, memory demand, and deployment diversity pose significant challenges for practical implementation. Field-programmable gate arrays (FPGAs) provide customized low-precision computation, spatial dataflow, on-chip data reuse, reconfigurability, and rich I/O capabilities, making them an important platform for DL inference. This paper presents a systematic review of FPGA-based DL deployment from a cross-layer perspective spanning model, compiler, architecture, runtime, and electronic design automation (EDA). Following a PRISMA-guided evidence synthesis protocol, this review analyzes DL workload characteristics, FPGA architectural optimizations, deployment toolflows, and physical implementation challenges. A unified taxonomy is proposed along the specialization–programmability continuum, including model-fixed accelerators, generator-based accelerators, template-configurable accelerators, and ISA-programmable overlays. These approaches are compared according to hardware regeneration requirements, model adaptability, operator coverage, compilation cost, and deployment flexibility. Furthermore, emerging workloads, including vision Transformers, graph neural networks, large language models, and multimodal models, are analyzed from the perspectives of computation, memory behavior, and runtime coordination. The review shows that FPGA deployment efficiency increasingly depends on memory capacity, mutable state management, operator support, and end-to-end compilation capability rather than peak multiply–accumulate throughput alone. Based on the analysis of 70 primary FPGA implementation studies, this paper highlights that reliable cross-study comparison requires careful consideration of model configuration, precision, execution phase, batch size, memory residency, FPGA platform, and evidence maturity. For multimodal generative models, the current evidence remains limited, with no identified end-to-end FPGA-based vision–language model implementation in the reviewed corpus. This review provides a systematic perspective for future FPGA-based DL deployment research, emphasizing cross-layer optimization, physically aware compilation, extensible accelerator architectures, and practical deployment efficiency. Full article

Accurate vegetation-soil segmentation is a key component of intelligent agricultural vision systems operating under limited computational resources. Existing CNN-based approaches generally improve segmentation accuracy by increasing network complexity or relying on high-resolution imagery, limiting their suitability for real-time embedded applications. This paper proposes a lightweight task-oriented CNN segmentation algorithm that integrates residual structural reconstruction, spectral augmentation, semantic segmentation, and edge-aware refinement within a unified multi-task optimization framework. Unlike conventional super-resolution methods, the residual module is specifically designed to recover segmentation-relevant structural features, including vegetation contours, crop-row geometry, and vegetation-soil transitions, without increasing spatial resolution. The algorithm combines four computational stages-spectral augmentation, pseudo low-resolution generation, residual enhancement, and edge-guided refinement-to improve internal feature representations while preserving computational efficiency. Experiments on the Crop Row Benchmark Dataset demonstrate competitive segmentation performance (BF1 = 0.95, mIoU = 0.92) with real-time inference (54.7 FPS). Comparative experiments, ablation studies, computational complexity analysis, and Grad-CAM-based interpretability analysis demonstrate the effectiveness of the proposed algorithm and the complementary contribution of its computational modules. The proposed formulation provides an efficient and interpretable CNN-based solution for resource-constrained agricultural vision systems. Full article

This paper evaluates the performance of the Orthogonal Frequency Division Multiple Access (OFDMA) technique within dense IEEE 802.11be (Wi-Fi 7) network environments. As traditional CSMA/CA-based networks reach their scalability limits, this work investigates how OFDMA interacts with key MAC layer features, specifically frame aggregation (A-MPDU/A-MSDU) and Spatial Reuse (SR), using BSS Coloring and OBSS/PD thresholds. Through simulation-based analysis, the work demonstrates that OFDMA provides substantial scalability gains, with throughput improvements of up to 49% in Downlink (DL) and 70% in Uplink (UL), while effectively preventing delay escalation. However, the results also reveal that these benefits are highly sensitive to Resource Unit (RU) granularity and scheduler design. In scenarios with frame aggregation, legacy OFDM may still offer higher throughput, and the effectiveness of Spatial Reuse remains highly constrained in UL and DL-OFDMA due to protocol and power density limitations. The paper concludes that maximizing Wi-Fi 7 performance requires precise parameter tuning adjusted to specific traffic types, network topologies, and utilized mechanisms. Full article

Road infrastructures are fundamental for ensuring connectivity, safety, and the efficient functioning of transportation networks. However, extreme weather conditions, particularly low temperatures and ice formation, pose significant risks to user safety and road conditions. This research focuses on the development of a geospatial model designed to identify road sections exposed to these harsh weather conditions. The model integrates different sources of information, including meteorological and satellite data, topographic information, and road maintenance plans, through a consistent methodology for data processing and analysis. A combination of geospatial analysis and advanced processing techniques was employed to identify and map regions most vulnerable to the formation of ice. The results show good spatial agreement between the areas identified by the model and the available local information, supporting its ability to characterize areas with greater susceptibility to winter-related hazards. This work highlights the potential of the developed geospatial model to support the planning of preventive and maintenance measures on roads. By providing spatial information on susceptibility to low temperatures and ice formation, the model can serve as a decision-support tool for road management, contributing to the planning of targeted interventions and improved road safety. Overall, this study underscores the importance of integrating advanced geospatial techniques into infrastructure management to improve response strategies in the face of extreme weather events. Full article

With the increasing deployment of multi-unmanned aerial vehicle (multi-UAV) systems in dynamic environments, the problem of efficient cooperative path planning has emerged as a critical challenge requiring urgent solutions. To address this issue, this paper proposes a novel joint optimization framework, named spatio-temporal attention-based multi-agent deep deterministic policy gradient (STA-MADDPG). Rather than proposing a new reinforcement learning algorithm in the strict sense, this work integrates advanced spatial-temporal feature extraction with heuristic gradient guidance. First, a cascaded architecture combining multi-head attention and Long Short-Term Memory (LSTM) networks is utilized to extract key local and temporal features, thereby mitigating the dimensionality curse in dense multi-agent observations. Second, an improved dynamic artificial potential field (DAPF) is integrated into the reinforcement learning framework as a state augmentation mechanism, providing heuristic guidance vectors that accelerate convergence and improve obstacle avoidance. Furthermore, to balance computational complexity and adaptive behavior, a rule-based hierarchical formation strategy is designed. The framework maps predefined formations (elliptical, chain, or wedge) to specific environment categories, while the underlying MARL policy governs the dynamic trajectory planning and topology maintenance. Finally, rigorous comparative and ablation experiments are conducted to evaluate path length, search time, and relative position errors. Statistical analysis demonstrates the effectiveness of the proposed framework, achieving up to a 67.3% reduction in search time and a 91.56% search success rate compared with standard MARL baselines in complex environments. Full article

Colorectal cancer remains a leading cause of cancer-related mortality worldwide, with automated polyp classification from endoscopic images offering a promising avenue for improving early detection. Existing approaches rely on single convolutional neural network (CNN) backbones with manually designed classification heads, limiting both representational capacity and deployment flexibility. This paper presents a swarm intelligence-augmented multi-backbone deep learning framework for eight-class gastrointestinal lesion classification on the Kvasir benchmark. Four CNN backbones (ResNet50, DenseNet121, MobileNetV2, EfficientNetB3) are independently fine-tuned using a two-phase transfer learning protocol and their penultimate-layer features concatenated into a 5888-dimensional representation, reduced to 256 dimensions via PCA. Five swarm intelligence algorithms—Particle Swarm Optimization, Artificial Bee Colony, JADE, L-SHADE, and CMA-ES—are benchmarked on the classification head architecture search task; all independently converge to tanh activation, a consistent pattern across independently initialized algorithms that is suggestive of, though not conclusive evidence for, particular geometric properties of PCA-transformed deep feature spaces. The PSO-optimized single-layer head (284 units, tanh) outperforms a manually designed three-layer baseline by 0.75% while using 67% fewer parameters. SI-guided class weight optimization yields targeted F1 improvements on the two most clinically significant classes (polyps: +0.015, ulcerative-colitis: +0.013). The fixed-head classifier trained on fused four-backbone features achieves 91.08% accuracy on Kvasir v2 (multi-seed mean 91.47% ± 0.49 across nine converging seeds; one seed failed to converge and is disclosed rather than excluded), below end-to-end DenseNet121 (92.25%; Wilcoxon p = 0.31, not statistically significant), while enabling classifier updates in under 30 s; a three-backbone subset dropping the weakest backbone (EfficientNetB3) reaches 92.33%, exceeding the full four-backbone fusion. Cross-dataset evaluation on Kvasir v1-to-v2 confirms near-zero generalization gaps across dataset scales; a restricted two-class evaluation on HyperKvasir (the only two of eight classes with usable labeled data) reaches 96.28% accuracy, and dual Grad-CAM with SI minimal sufficient region analysis, validated quantitatively against Kvasir-SEG ground-truth masks, provides spatially grounded, clinically interpretable explanations. Full article

This study develops the Fuzzy Cross-Efficiency Network Design (FCEND) framework—an integrated Geographic Information System (GIS) and Data Envelopment Analysis (DEA) approach for logistics network design under deep uncertainty. Unlike conventional methods that ignore spatial equity and data credibility, FCEND combines GIS-based suitability mapping (30 m resolution, incorporating slope, land use, and floodplains), hybrid efficiency scores (${\mathrm{\Phi }}_{\mathrm{j}}$) integrating Cross-Efficiency DEA (CEDEA) peer evaluation with Fuzzy DEA (FDEA) uncertainty modeling, and a multi-objective function Z(S) = α·Efficiency(S) + β·H(S) − γ·Gini(S) that balances demand-weighted efficiency, portfolio-dependent criterion diversity (represented by the entropy term H(S)), and spatial equity. Applied to Iran’s staple food commodity network—85 million people across 1.65 million km²—FCEND identifies an optimal 15-node portfolio spanning 15 provinces with 74% direct population coverage within 150 km. The portfolio achieves a Gini coefficient of 0.298, and 9 of 15 nodes with excellent rail connectivity, while capturing strategically vital nodes (Borujerd, Bandar Abbas, Zahedan) overlooked by conventional approaches. Nine core sites with stability scores (f_j = 1.0) demonstrate perfect stability across all uncertainty scenarios. The framework’s modular architecture is conceptually transferable to emerging economies, as illustrated through adaptation to Vietnam (70% parameter swap). By integrating GIS-based spatial analysis, peer evaluation, fuzzy uncertainty, portfolio-dependent entropy, and equity constraints within a unified optimization framework, FCEND offers a transferable methodology for evidence-based logistics infrastructure planning—contributing directly to the United Nations Sustainable Development Goals (SDGs): SDG 2 (Zero Hunger), SDG 9 (Resilient Infrastructure), SDG 10 (Reduced Inequalities), and SDG 13 (Climate Action). Full article

Tree-ring chronologies are essential proxies for investigating ecosystem dynamics and reconstructing environmental variability, yet integrative approaches for assessing chronology quality and sampling representativeness across heterogeneous regions remain limited. We analyzed 190 tree-ring chronologies distributed across Mexico and developed two composite indicators: the Signal Quality Index (SQI), integrating internal coherence, interannual sensitivity, common growth signal strength, and the Sampling Representativeness Index (SRI), quantifying the statistical adequacy of sampling efforts. Both indices were standardized and evaluated using Moran’s I, Local Indicators of Spatial Association (LISA), Getis–Ord Gi* hotspot analysis, and correlations with climatic, hydrological, and edaphic variables. Results revealed a marked decoupling between chronology signal quality and sampling representativeness. SQI exhibited significant positive spatial autocorrelation, with clusters of high and low values associated with hydroclimatic gradients. It was strongly related to indicators of water availability and atmospheric evaporative demand, suggesting greater growth coherence under water-limited conditions. In contrast, SRI displayed weak spatial structure and largely non-significant relationships with environmental variables, indicating that representativeness is driven primarily by methodological decisions and sampling design. These findings highlight complementary ecological (SQI) and methodological (SRI) dimensions of dendrochronological networks and provide a practical framework for improving chronology evaluation, comparability, and network development across environmentally heterogeneous regions. Full article

Copenhagen is often regarded as a pioneer city in climate change adaptation and green–blue infrastructure deployment. However, when addressing extreme environmental problems such as heavy rainfall flooding and urban heat island effects, many interventions remain site-specific and insufficiently connected across spatial scales. This article investigates how Nature-Based Solutions (NBS) can be structured as an integrated spatial infrastructure system to enhance climate resilience and urban livability in Copenhagen. The research combines theoretical analysis, comparative project studies, GIS-based spatial analysis, and a multi-criteria evaluation matrix. Building on this method, a multi-scalar framework for NBS interventions is developed, which consists of site prioritization at the Meso scale, development strategies at the Micro scale, and NBS spatial design at the Nano scale. With the assessment criteria adapted from the UNalab project Nature-Based Solutions Technical Handbook Factsheets, the results demonstrate that Nano-scale design strategy can enhance green connectivity, create new inclusive urban public spaces, transform existing industrial areas, and improve new streets and parks for climate resilience. The Meso-scale distributed measures deliver localized benefits such as infiltration and flood buffering, while the most effective strategies integrate cross-scalar NBS interventions within a connected network. The case of Nordhavn is analyzed through multi-criteria selection and multifunctionality assessment as a qualitative process for a better demonstration of the practical application of NBS prioritization and spatial design analysis. Full article

Sustainable Internet of Things (IoT) deployments require network planning strategies that explicitly account for environmental impact and not only traditional performance and energy metrics. This work analyzes how input representations affect a learning-based framework for carbon footprint (CF)-aware gateway placement in LoRa multi-hop IoT networks. Building on a previous CF model and an integer linear programming dataset, a multilayer perceptron is retrained using different input encodings: end-device coordinates, traffic-based weights, spatial sampling regions (SSRs), and a global CF estimate. Their contributions are evaluated through Shapley additive explanations (SHAP)-based explainability analysis, ablation studies, and sensitivity analysis. Results show that the CF estimate is the most influential input, acting as a global guidance signal that drives large gateway relocations. The combination of raw coordinates and SSR-based spatial summaries achieves the best performance by capturing both fine spatial detail and collective relay opportunities, while traffic-based weights mainly contribute through aggregate effects. These findings provide practical guidelines for designing CF-aware learning pipelines and offer insights to support future research on environmentally aware artificial intelligence for IoT network planning. Full article

Accurate network traffic prediction is important for the autonomy, resilience and resource orchestration of 6G and AI-native communication infrastructures, while also supporting green networking and digital twin network applications. However, existing data-driven prediction models face several limitations: over-reliance on massive labeled data, physically implausible predictions, black-box non-interpretability and over-parameterization that impairs edge deployment. To address these issues, this paper proposes a Physics-Informed Network Traffic Prediction (PINTP) framework for graph topology network traffic prediction, which formalizes network traffic evolution as Graph-based Advection–Diffusion–Reaction (ADR) equations and embeds physical regularization into the neural architecture. The framework adopts a hybrid differentiation paradigm unifying automatic differentiation for temporal dynamics and spectral graph theory-derived operators for discrete spatial topologies, and designs a physics-constrained composite loss function with data-driven collocation to balance data fidelity and physical consistency. Experiments are conducted in two complementary settings: a 100-node synthetic random-graph benchmark that evaluates the full graph-topological formulation, and a topology-unavailable real-world telemetry proxy based on Alibaba Cluster Trace v2018 for evaluating sparse-label physics-informed temporal regularization. Comparative analysis with mainstream baselines, including Multilayer Perceptron (MLP), Spatio-Temporal Graph Convolutional Network (STGCN), Graph WaveNet, Transformer, Temporal Convolutional Network (TCN), and XGBoost, shows that the proposed PINTP/PINN implementation achieves a test $R^2$ of 0.898 and MSE of 0.000723 on the 100-node synthetic graph benchmark, close to the strongest Transformer result ($R^2=0.900$, MSE = 0.000710), while using substantially fewer trainable parameters. PINTP/PINN also outperforms Graph WaveNet, STGCN and TCN in this setting, indicating that physics-informed regularization can remain competitive as graph size increases. On the Alibaba proxy task, PINTP/PINN achieves the strongest result among the evaluated models with a test $R^2$ of 0.963. In an independent Alibaba ablation protocol, physical regularization (e.g., $\lambda =10.0$) reduces the mean squared error by 89.15% compared with pure data-driven models and helps mitigate overfitting. This work presents a systematic PINTP framework for graph topology network traffic prediction, achieving competitive prediction accuracy with high parameter efficiency and a degree of physical interpretability. It helps address several limitations of traditional data-driven models, indicates potential for future deployment-oriented studies on real-time network management and resource-constrained edge analytics, and provides an interpretable modeling route for physics-informed network analytics in next-generation communication systems. Full article

In the campus planning of severe cold regions, indoor pedestrian corridors have emerged as an important strategy for mitigating harsh weather and maintaining pedestrian network continuity. These enclosed systems provide thermal comfort and physical convenience for faculty and students. However, reduced visual contact with outdoor natural environments may be associated with less favorable affective experiences. Existing research primarily emphasizes functional connectivity and spatial efficiency, with limited quantitative evidence on the visual characteristics of indoor corridors and their relationship with emotional perception. Drawing upon environmental psychology, this study examines the associations between corridor visual characteristics and five AI-predicted affective perception dimensions. Using deep-learning-based computer vision, semantic segmentation, and statistical analysis, environmental features and predicted perception scores were analyzed. A local human validation showed moderate overall agreement between human consensus ratings and AI-predicted scores, supporting their use as exploratory perception proxies. The findings indicate that indoor-corridor-related visual features were negatively associated with predicted beauty scores and positively associated with predicted depression scores, while predicted boringness was positively associated with wall enclosure. Predicted safety and liveliness showed weaker associations with individual visual features. These results provide exploratory evidence for balancing thermal protection, circulation efficiency, and psychological comfort in future campus corridor design. Full article