Search Results (615)

Millimeter-wave (mmWave) communication systems are vulnerable to severe attenuation, blockage-induced LOS/NLOS transitions, and time-varying co-channel interference. This paper develops a lightweight distributed power-allocation framework in which each base station independently updates a tabular Q-learning policy using locally observable blockage-ratio, serving-distance, and aggregate-interference information. The proposed state-dependent dynamic reward is recalculated at every decision step, and its coefficients vary explicitly with the instantaneous blockage ratio, QoS satisfaction ratio, and normalized interference level. All learning-based and non-learning baselines are evaluated using the same topology realizations, blockage and mobility traces, and random seeds. Under the reconstructed simulation settings, the proposed method achieves performance comparable to fixed Q-learning while retaining a transparent blockage-aware state and low-complexity distributed implementation. DQN, greedy, and uniform power achieve higher raw capacity or QoS in the considered small-scale network. Results from 30 paired runs with 95% confidence intervals, together with ablation, sensitivity, beam-misalignment, and overhead analyses, clarify the empirical benefits, limitations, and deployment scope of the proposed method. The study focuses on power control after beam establishment; joint beam tracking and power allocation remain outside the present scope. Full article

Post-disaster emergency communication recovery is not merely a link-repair task but a high-level planning problem constrained by service priorities, inter-object dependencies, resource budgets, and time windows. Existing restoration optimization methods generally rely on fully structured inputs, whereas direct large language model (LLM) planning may produce fluent candidates that violate encoded prerequisites, stage-order relations, budget limits, or temporal constraints. To address this challenge, we propose ICG-Restore, an intent-constrained, graph-enhanced LLM planning framework with rule-consistent minimal-edit repair. ICG-Restore transforms mixed restoration requests and structured network observations into task packages that can be checked for validator-level feasibility under an encoded high-level constraint model and evaluated by downstream abstract executors or schedulers. The framework compiles natural-language requests, structured observations, and operational rules into a task-intent object; retrieves task-relevant context from a heterogeneous scenario graph and a restoration knowledge graph; generates stage-wise restoration candidates; and applies bounded local corrections to candidates that violate encoded constraints. In this paper, “minimal-edit” is a descriptive label for a bounded local repair principle that prioritizes less disruptive corrections. Candidates accepted by the validators are evaluated and ranked by a safety-aware agent executor operating in an abstract restoration action space. Experiments on controlled abstract topologies covering three scales, four restoration tasks, and five environmental evolution modes show that ICG-Restore improves validator-level constraint satisfaction and benchmark-estimated recovery utility. Compared with Direct-LLM, it improves CSR and CRS by 1.99% and 24.56%, respectively; benchmark-specific WCTC@5 structural-alignment diagnostic increases by 38.87%. Full article

Infrared object detection from unmanned aerial vehicles (UAVs) is critically challenged by multi-type composite degradation—including noise, blur, and low contrast—which severely undermines feature discriminability and multi-scale target perception. This paper proposes SGW-DETR (Spectral-Guided Graph-structured Wavelet Detection Transformer), which is a novel framework built upon RT-DETR, incorporating three synergistic modules across the backbone, neck, and encoder. FDSANet (Frequency Domain Spectral Awareness Network) replaces the conventional ResNet backbone, integrating the Multi-Scale Frequency Perception Module (MSFPM), Selective Channel Frequency Decomposition (SCFD), and Dynamic Kernel Spectral Modulation (DKSM) to achieve instance-level adaptive spectral feature extraction without degradation-type supervision. The Graph-Structured Fusion Network (GSFN) combines the Adaptive Semantic Fusion Module (ASFM) with the Graph Structure Perception Module (GSPM), employing Gaussian kernel soft membership and two-stage message passing to explicitly model spatial topological dependencies among object components. The Wavelet-guided Contrast Feature Aggregation module (WCFA) restructures the Attention-based Intra-scale Feature Interaction (AIFI) encoder via a Haar-based Frequency Decomposition Unit (HFDU), decomposing features into foreground-edge and background-thermal components and achieving hierarchical foreground–background decoupling through nested dual-path causal contrastive attention. A UAV infrared degradation dataset comprising 4686 images spanning six degradation types with component-level annotations was constructed for evaluation. SGW-DETR achieves 75.2% mAP₅₀, outperforming RT-DETR by 3.5%, while simultaneously reducing GFLOPs and parameter count by 16.8% and 9.9% at an inference speed of 85.5 FPS. Sustained performance gains on the M3FD and IndraEye benchmarks further demonstrate the framework’s cross-domain generalization capability, offering practical value for UAV-based surveillance, search-and-rescue, and border monitoring under adverse imaging conditions. Full article

Renewable energy system (RES)-based electric vehicle (EV) charging infrastructure enhances energy security. The need for intelligent routing techniques that achieve low latency, high reliability and adaptive path selection under extremely dynamic traffic situations has increased owing to the quick growth of Internet of Things (IoT)-enabled vehicular networks. With an emphasis on recurrent neural networks (RNNs), deep belief networks (DBNs), radial basis function neural networks (RBFNNs), and long short-term memory (LSTM) networks, in addition to convolutional neural networks (CNNs), this analysis looks at cutting-edge AI-based models used for shortest-path routing in IoT-driven vehicular ad hoc networks (VANETs). The paper examines how various designs handle issues such as connection instability, heterogeneous sensor data, quick topological changes, and real-time decision making. A comparative analysis shows that DBN and CNN display strong feature learning for intricate mobility patterns and congestion recognition, while sequence-aware techniques like RNN and LSTM advance spatiotemporal traffic estimation. For low-latency route evaluation, RBFNN compromises rapid nonlinear representation. The examination shows that CNN models greatly improve the scalability, adaptability and optimality of routing, confirming AI-enabled structures as a promising path for next-generation IoT-based vehicular routing methods. Full article

Automated 3D city generation, focusing on road networks and building layouts, is in high demand for applications in urban planning, analysis, and simulations. The surge in deep generative models has facilitated automated design in recent years. However, the lack of high-quality datasets and benchmarks hinders the progress of these data-driven methods in generating city configurations. To fill this gap, this study introduces a multimodal dataset designed for the controllable generation of road networks and building layouts (named RoBus), whose public project repository provides release materials, and constitutes a large-scale resource in the field of generative city design. The RoBus dataset comprises aligned images, graphics, labels, and texts, with 72,400 paired samples that cover around 80,000 km² globally. Besides utilizing prevalent generative models, we also introduce baseline models that leverage the multimodal features of RoBus. The experiments establish the dataset’s usability while revealing complementary trade-offs rather than uniform superiority. ControlNet obtains the lowest road network FID ($20.78$), whereas our topology-aware road baseline obtains the highest traffic-convenience score ($0.83$) at the cost of lower fidelity and diversity. For building layouts, our multimodal baseline reduces FID to $17.42$ and building-density Wasserstein distance from $6.12$ to $3.37$ but produces lower diversity and a higher invalid-sample rate than the strongest comparison methods. Full article

Future sixth-generation (6G)-oriented networks require programmable control that can adapt routing to latency and congestion without unsafe online exploration. This study evaluates offline multi-agent deep deterministic policy gradient (MADDPG) with behavior-adjusted training rewards for latency-aware path control in software-defined networking (SDN). Each traffic pair is modeled as an agent selecting one of three retained candidate paths, while centralized critics learn coordinated decisions from topology-specific Ryu–Mininet transition datasets. Nine policies are compared using ten paired seeds on fat-tree, mesh-grid, and WAN-corridors topologies under a deployed utilization–latency weighting of $0.60/0.40$, together with flow-completion, latency, congestion, architectural-comparison, sensitivity, robustness, statistical, and controller-overhead analyses. The utilization-aware path heuristic achieves the strongest overall reward ranking. MADDPG is the strongest learned policy on fat-tree, is not significantly outperformed by any evaluated policy on mesh-grid, and remains statistically tied with completion-matched policies on WAN-corridors. Behavior adjustment is topology-dependent rather than uniformly beneficial. The exported policy requires approximately $52\mathsf{\mu }\mathrm{s}$ per joint decision, whereas complete control-loop timing is dominated by network-statistics polling. These results support offline multi-agent SDN control as a competitive, low-overhead option when interpreted jointly with topology structure, flow completion, and strong heuristic baselines. Full article

Electric-vehicle gearboxes remain key design elements because they determine how motor speed and torque are converted into wheel speed and tractive effort over a driving cycle. This design-oriented scoping review synthesizes EV gearbox architectures, gear ratio selection, efficiency losses, NVH, planetary and compound planetary systems, lubrication, thermal behavior, reliability, manufacturability, and cost within one evidence-informed architecture-selection perspective. A structured search of Scopus, IEEE Xplore, ScienceDirect, SpringerLink, and MDPI identified 312 records; after removal of 71 duplicates, screening of 241 titles and abstracts, and full-text assessment of 61 articles, 40 sources formed the reproducible structured-search core. A gap-directed supplementary search then added 12 sources in underrepresented areas, producing a 52-source synthesis set. The revised analysis reports publication trends, evidence-level distributions, technical-focus frequencies, and a dimension-separated evidence-count table for ratio count, gear train topology, and integration level. The evidence indicates that single-speed reduction gearboxes remain the mature baseline for many passenger EVs, whereas two-speed, multi-speed, planetary, compound planetary, and integrated e-axle solutions require application-specific justification based on system-level benefits and risks. An illustrative screening calculation demonstrates the framework logic without being presented as production-level validation. The principal gaps are experimentally validated loss and NVH maps, coupled efficiency–thermal–lubrication–durability analysis, reliability-aware mission-profile validation, standardized benchmarks, and transparent comparison of compound planetary and integrated e-axle systems. Across heterogeneous study conditions, reported energy benefits range from 2.4% for fixed-ratio optimization to 15% for selected multi-speed comparisons; these results are not pooled because the vehicles, motor maps, drive cycles, loss models, and validation methods differ. Full article

The increasing penetration of distributed photovoltaic generation is changing voltage behavior in distribution feeders and creating operational challenges related to voltage violations, losses, curtailment, and hosting capacity. This paper proposes a topology-aware analytical and computational framework for assessing voltage regulation and continuous photovoltaic hosting capacity in distribution feeders with smart-inverter controls. The framework combines topology-dependent voltage sensitivities, balanced steady-state power-flow simulations, explicit inverter apparent-power constraints, and gain indices that quantify the contributions of feeder topology, inverter controls, and their interaction. The IEEE 33-bus and IEEE 69-bus feeders are evaluated under radial and meshed configurations, heavy- and light-load conditions, and four photovoltaic operation modes: no control, Volt–Var, Volt–Watt, and combined Volt–Var/Volt–Watt control. A discrete sweep from 20% to 150% PV penetration is used to characterize voltage, active and reactive losses, and curtailment trends. Hosting-capacity boundaries are subsequently determined through an interval-aware procedure consisting of a one-percentage-point scan followed by bisection refinement to 0.1 percentage point, without assuming a globally monotonic feasibility transition. The results show that the feeder topology strongly affects voltage sensitivity and photovoltaic hosting capacity. The meshed operation generally increases voltage margins, while the interval-aware assessment identifies nonzero feasible penetration ranges, even when the zero-PV operating point is constrained by undervoltage. Volt–Watt achieves the largest hosting-capacity gains at the expense of curtailment, whereas Volt–Var preserves photovoltaic injection but may increase feeder losses. Therefore, hosting capacity should be interpreted jointly with topology, inverter controls, injected power, curtailment, and losses. Full article

Grid-forming (GFM) inverters are increasingly being deployed in distribution networks to improve system strength and provide synthetic inertia. However, their steady-state influence on hosting capacity (HC) is not yet fully understood, particularly across different distribution network topologies. This study shows that the relationship between GFM droop control and HC is strongly topology-dependent. In overvoltage-limited feeders, non-optimised GFM reactive power absorption reduces HC, whereas in undervoltage-limited feeders, the same reactive support mechanism significantly enhances HC. To analyse and optimise this behaviour, this study proposes the Coordinated Hosting-Capacity and Stability Algorithm (CHSA), which determines the optimal site-specific GFM droop parameters $(m_p,n_q,{\omega }_c)$ using a steady-state multi-constraint metaheuristic co-simulation framework. In this context, stability refers exclusively to steady-state voltage feasibility, that is, the maintenance of nodal voltages within statutory limits under varying DER injection levels. For the overvoltage-limited Australian-adapted IEEE 123-bus unbalanced multi-phase feeder, mid-range GFM settings introduce a GFM Hosting-Capacity Penalty ($C{P}_{\mathrm{GFM}}$) of 95.19 kW, reducing HC from the 950.00 kW grid-following (GFL) reference ceiling to 854.81 kW. By coordinating the reactive droop parameters, the proposed CHSA recovers 92.65 kW (97.3% of $C{P}_{\mathrm{GFM}}$), increasing HC to 947.46 kW while eliminating voltage violations, with all inverters operating at 33.3% of their rated apparent power capacity. For the undervoltage-limited IEEE 34-bus feeder, the same GFM reactive support mechanism increases HC from 19.90 kW under GFL operation to 237.41 kW with the optimised GFM configuration, representing a nearly twelve-fold improvement. Independent verification using Particle Swarm Optimisation confirms that both algorithms converge to the same operating point: on the representative run (seed 42), HC values differ by only 1.23 kW, while mean HC across 20 independent seeds differs by 0.99 kW (HBO: $947.70\pm 0.72$ kW; PSO: $948.69\pm 0.00$ kW), indicating that the results are governed by network characteristics rather than the choice of optimisation algorithm. These findings demonstrate that the HC limitations often associated with GFM technology arise mainly from suboptimal control coordination rather than inherent architectural constraints. Furthermore, the proposed CHSA provides a general framework for voltage-constraint-aware droop coordination across distribution networks with contrasting topological characteristics. Full article

Online platforms need scalable risk-screening methods for rapidly spreading rumors, misleading content, and large-scale user responses. Platform-side content governance often needs to rank suspicious events when only partial propagation evidence is visible, so that limited review resources can be assigned first to high-risk events. This article focuses on that setting. We propose a Structure-Aware Bidirectional Graph Convolutional Network (SA-BiGCN), which combines a BiGCN-style propagation graph encoder with structural statistics computed from the currently visible propagation tree to estimate event-level rumor probability. We construct Main Weibo V2 from 4664 raw Weibo source posts and their propagation structures, encode node text with a unified vocabulary, and use fixed event-level splits. The model evaluates source-post veracity at Top 10, Top 30, Top 50, Top 100, and full observation windows according to a timestamp-audited propagation order, using only the currently visible propagation-tree subgraph. The experimental results show that SA-BiGCN reaches 92.72% Avg F1 and 91.28% Worst F1 on Main Weibo V2 while using one shared checkpoint for all observation windows. Feature-set, leave-one-feature-group, sampling-strategy, topology-corruption, deployment-cost, and transfer analyses indicate that structural statistics computed from the currently visible propagation tree provide auxiliary evidence mainly in early-window and weakest-window settings. We further extend SA-BiGCN with graph-attention, temporal-quality, response-style, window-aware, and RootRoBERTa fusion variants. These additional cues are retained as diagnostic trends and do not replace the main model. Overall, SA-BiGCN is best understood as a lightweight single-checkpoint propagation-tree scorer. Full article

Robust object detection for autonomous driving requires perception models that remain reliable when visible imagery is degraded by darkness, glare, rain, fog, motion blur, or long-range small targets. Visible and thermal infrared cameras provide complementary evidence, yet many RGB–thermal detectors fuse modalities, mainly as aligned tensors, and may underuse relational structure in channel responses, spatial layouts, semantic scales, and modality-specific uncertainty. This paper presents TopoGraph-Fusion, a hierarchical graph-guided dual-modal object detector that formulates fusion as topology-aware reasoning rather than direct feature concatenation. The proposed framework builds a dual-stream backbone for RGB and thermal images, constructs channel-wise topology through a channel-topology graph aggregation module, derives relation-aware spatial and channel global attention from affinity graphs, and replaces fixed feature-pyramid communication with a Graph-Guided Feature-Pyramid Network. A topology-regularized detection objective further encourages stable cross-modal correspondence while suppressing noisy all-to-all connections. Experiments on M3FD, FLIR, RGBTDronePerson, and VEDAI512 cover road scenes, adverse illumination, drone–person perception, and aerial vehicle detection. Within this validation scope, the results and visual analyses indicate that topology-guided fusion improves small-object recall, cross-modal consistency, and robustness under modality imbalance. Full article

To address ambiguities in phase-line identification, errors in user–transformer associations, and inaccurate topology records in complex low-voltage distribution networks, this paper proposes a hierarchical topology identification method based on improved T-type grey relational analysis and Fisher optimal segmentation. In the proposed method, signed voltage increments, a resolution coefficient, and node-distance weighting are introduced to construct distance-aware relational features from user voltage sequences. Fisher optimal segmentation is subsequently applied to identify user–transformer associations, followed by a suspicious-user verification mechanism for local topology correction. Case studies using practical transformer-area data show that the proposed method provides greater discrimination among user voltage sequences than Pearson correlation analysis and conventional T-type grey relational analysis. At a random-noise level of 6%, the phase-line identification accuracy remains 83.7%, while the accuracy of user–transformer association identification reaches approximately 93% under the available field-data conditions. Comparative and robustness analyses further indicate that Fisher optimal segmentation maintains relatively stable performance under the tested missing-entry and reduced class-separation conditions. These results suggest that the proposed framework provides a feasible and interpretable approach to topology identification in complex low-voltage distribution networks. Full article

Surface electromyography (sEMG) is increasingly used for gesture recognition in prosthetics, rehabilitation, and human–computer interaction. Existing architectures typically force heterogeneous sEMG features into a shared latent representation, limiting their ability to capture complementary temporal, frequency-domain, and inter-electrode spatial dependencies. To better exploit these features, this paper proposes a three-branch fusion network. Unlike many existing multi-branch methods, the proposed network explicitly models the ring arrangement of armband electrodes, capturing the adjacency information in the sensor topology that linear channel representations ignore. The temporal and spectral branches use a compact multi-scale residual structure, so this topology branch is added while maintaining modest model complexity. A reliability-aware routing mechanism then adaptively assigns fusion weights to the three branches for each sample. On NinaPro DB5 Exercise B (eight-channel lower armband), the method reaches 83.99% under subject-dependent training and 85.66% under transfer learning, exceeding prior transfer learning approaches under matched conditions. Ablation experiments confirm that the three branches contribute non-redundant information and that adaptive fusion outperforms fixed combinations. The architecture also generalizes to MyoArmbandDataset under a subject-adaptive transfer learning protocol without dataset-specific hyperparameter retuning, indicating potential for wearable gesture interfaces, rehabilitation, and prosthetic control. Full article

Redundant channels, sparse electrode topology, and insufficient cross-layer feature fusion limit electroencephalography (EEG)-based emotion recognition. This study proposes the Topology-Aware Dual-Bridge Fusion Network (TDBF-Net), a compact framework that integrates adaptive channel selection, topology-aware sparse convolution, and bidirectional bridge fusion. First, sample entropy, dispersion entropy, and fuzzy entropy are fused to estimate channel importance, while particle swarm optimization (PSO) learns the entropy weights and an elbow-based criterion determines the retained channel subset. Second, differential entropy (DE) features from the $\theta$, $\alpha$, $\beta$, and $\gamma$ bands are mapped to an $8\times 9$ sparse topological tensor according to electrode locations. A fixed spatial validity mask is applied before and after convolution to suppress invalid responses from zero-padded regions and preserve real electrode topology. Third, a dual-bridge fusion module recalibrates shallow and deep features in both directions through channel attention and gated fusion, and a bidirectional long short-term memory network (BiLSTM) further captures short-term temporal dependencies. Subject-dependent experiments on the SJTU Emotion EEG Dataset (SEED) and the Database for Emotion Analysis using Physiological Signals (DEAP) show that TDBF-Net achieves $97.62\%$ ± $1.59\%$ accuracy on SEED and $98.46\%$ ± $0.94\%$ and $98.14\%$ ± $0.77\%$ on DEAP valence and arousal, respectively. Paired DEAP ablations support topology and bridge contributions for valence, whereas the corresponding arousal differences are not significant. Selector controls, robustness tests, computational profiling, and held-out visualizations further characterize the method’s compression, cost, and interpretability. The evidence supports TDBF-Net as an effective subject-dependent framework while leaving subject-independent and cross-dataset generalization for future validation. Full article

In ground-based space surveillance, space target attitude estimation is critical for space situational awareness, yet existing methods based on inverse synthetic aperture radar (ISAR) images suffer from three core limitations: phase information is discarded in amplitude-only processing, convolutional neural networks have a restricted global receptive field, and the physical topology of satellite components is not explicitly modeled. To address these issues, we propose a complex-domain Transformer–graph neural network (CD-TrGNN) that unifies global context modeling and adaptive topological reasoning in an end-to-end framework. Specifically, a complex-domain Transformer module (CD-Transformer) with tailored attention captures long-range dependencies among image patches while preserving both amplitude and phase information; a complex-domain graph convolution module (CD-GC) with learnable adjacency matrices and a dual-path update mechanism explicitly encodes the structural relationships among satellite parts. On a self-built ISAR complex image dataset, CD-TrGNN achieves a three-axis mean absolute error of only 1.70°, substantially outperforming six representative baselines. Ablation experiments confirm the effectiveness of complex-domain processing, global attention, and topological reasoning. At a 5 dB signal-to-noise ratio, the error remains at 2.81°, and the accuracy stays below 2° for two different satellite structures. These results demonstrate that CD-TrGNN can fully exploit the information in ISAR complex images, enabling high-accuracy and highly robust attitude estimation. Full article