Search Results (2,259)

Waveguide-like propagation in elongated underground environments—utility corridors, logistics tunnels—generates dense multipath that can cause the earliest or strongest resolvable channel impulse response (CIR) component to originate from a specular reflection rather than the direct line-of-sight (LOS) path. In the single-anchor CIR-tap-based implementations common to practical ultra-wideband (UWB) systems, baseline estimators such as phase-difference-of-arrival (PDOA) and MUSIC rely on selecting a single dominant CIR component, producing large angle-of-arrival (AoA) errors whenever the selected path is a reflection. We propose a multipath credibility selection (MCS) AoA estimator, MCS-AoA, that does not require explicit LOS/NLOS classification. The algorithm scores each resolvable CIR component with four credibility factors—amplitude significance, time-of-flight (TOF) consistency, inter-baseline phase–geometry agreement, and cross-baseline coherence—and fuses retained candidates into a credibility-weighted spatial covariance matrix for 2D MUSIC search. Field experiments on a custom five-channel coherent UWB platform compare MCS-AoA against six baselines—PDOA, MUSIC, MVDR/Capon, TLS-ESPRIT, PwMUSIC, and DNN-AoA. In an underground corridor (5–40 m), MCS-AoA achieves an azimuth/elevation MAE of 1.00${}^{\circ }$/1.46${}^{\circ }$, outperforming all baselines (PDOA: 2.26${}^{\circ }$/2.49${}^{\circ }$; MUSIC: 1.76${}^{\circ }$/2.40${}^{\circ }$; next-best PwMUSIC: 1.44${}^{\circ }$/2.17${}^{\circ }$); in a logistics tunnel (5–80 m), it achieves a 1.19${}^{\circ }$ overall azimuth MAE. Simulations corroborate these gains, with a 0.71${}^{\circ }$ azimuth RMSE at 80 m (69.3% reduction over PDOA) and 86.6% of estimates falling within 1${}^{\circ }$. Full article

Deep neural networks (DNNs) have demonstrated remarkable capability in accurately predicting equilibrium combustion products and thermodynamic properties of diesel combustion. However, the lack of awareness of uncertainty and interpretability has limited their scientific credibility and practical application. In this work, an enhanced DNN framework with uncertainty quantification and explainability is developed. The model achieves high accuracy across all outputs, with R² values exceeding 0.99 for major thermodynamic variables. In this model, Monte Carlo dropout sampling is used to estimate epistemic uncertainty, and prediction confidence intervals are analyzed across all species and thermodynamic outputs, revealing strong correlations for major components. Model explainability is further explored using Shapley additive explanations (SHAP), which attribute the influence of equivalence ratio, temperature, and pressure on each predicted species and combustion characteristics. The combined uncertainty quantification and explainability framework not only enhances confidence in DNN combustion models but also provides physical insight into the relationships between input conditions and equilibrium thermochemistry that are learned by the DNN. Full article

Forest height is a key structural parameter for characterizing forest architecture and estimating carbon storage. However, under complex terrain and heterogeneous forest conditions, Polarimetric synthetic aperture radar (PolSAR)-based forest height inversion using multi-category features still faces several challenges, including feature redundancy, insufficient characterization of the nonlinear couplings among high-dimensional features by deep learning models, and the difficulty of jointly achieving model stability and interpretability. In this paper, to address these issues, we propose a method for SHapley Additive exPlanations (SHAP) interpretability-driven PolSAR forest height inversion based on deep learning and multi-category feature fusion. Firstly, a deep neural network (DNN) is constructed, and SHAP is introduced to interpret the model decision process, enabling the identification of key feature interactions with clear physical significance and guiding the iterative model optimization in an explainability-driven manner. Furthermore, a SHAP-guided feature attention DNN is developed, in which the feature contribution scores are incorporated as prior knowledge for attention weight initialization, thereby establishing a closed-loop modeling framework from “interpretation” to “optimization”. Experiments were conducted at the site of the Huangfengqiao forest farm, Youxian County, Hunan province, China, using ALOS-2 L-band fully polarimetric SAR imagery. The experimental results demonstrated that the proposed method can significantly outperform the conventional machine learning approaches and various deep learning architectures for forest height inversion. The final model achieved a coefficient of determination (R²) score of 0.75 and a root-mean-square error (RMSE) of 1.35 m on the test dataset. These findings indicate that the combination of SHAP-driven multi-category feature fusion and deep learning can effectively enhance both the inversion accuracy and physical interpretability, providing a reliable solution for PolSAR-based forest structural parameter retrieval at the Huangfengqiao study site, with potential applicability to complex terrain conditions. Full article

Gaze analysis often relies on hypothesised, subjectively defined regions of interest (ROIs) or heatmaps: ROIs enable condition comparisons but reduce objectivity and exploration; while heatmaps avoid this, they require many pixel-wise comparisons, making differences hard to detect. Here, we propose an advanced data-driven approach for analysing gaze behaviour. We use DNNs (adapted versions of AlexNet) to classify conditions from gaze patterns, paired with reverse correlation to show where and how gaze differs between conditions. We test our approach on data from an experiment investigating the effects of object-specific sounds (e.g., church bell ringing) on gaze allocation. ROI-based analysis shows a significant difference between conditions (congruent sound, no sound, phase-scrambled sound and pink noise), with more gaze allocation on sound-associated objects in the congruent sound condition. However, as expected, significance depends on the definition of the ROIs. Heatmaps show some unclear qualitative differences, but none are significant after correcting for pixelwise comparisons. We showed that, for some scenes, the DNNs could classify the task based on individual fixations with accuracy significantly higher than chance. Our approach shows that sound can alter gaze allocation, revealing task-specific, non-trivial strategies: fixations are not always drawn to the sound source but shift away from salient features, sometimes falling between salient features and the sound source. Crucially, such fixation strategies could not be revealed using a traditional hypothesis-driven approach. Overall, the method is objective, data-driven, and enables clear comparisons of conditions. Full article

Unmanned Aerial Vehicle (UAV)-assisted mobile edge computing is pivotal for the Space–Air–Ground–Sea Integrated Network (SAGSIN) to support heterogeneous task offloading. However, the inherent resource constraints of UAVs limit their ability to support intensive and concurrent task processing in dynamic environments. In such complex scenarios, the dual requirements of discrete model partitioning and continuous bandwidth allocation make it difficult for traditional reinforcement learning algorithms to achieve optimal resource matching. Therefore, in this paper, we design a joint optimization framework based on Asynchronous Advantage Actor-Critic (A3C) and proximal policy optimization (PPO). Specifically, the model partitioning strategy is learned through PPO, which utilizes a clipped objective function to ensure training stability and generalization across complex Deep Neural Network (DNN) structures. Moreover, the framework leverages the asynchronous multi-threaded architecture of A3C to dynamically allocate bandwidth, effectively accommodating rapid fluctuations in terminal access. Finally, to prevent resource monopolization and ensure fairness, a weighted priority scheduling mechanism based on task urgency and computation time is introduced. Extensive simulations show that the proposed algorithm outperforms existing approaches in terms of task completion rate, task processing latency, and resource utilization under dynamic SAGSIN scenarios. Full article

In recent years, video recognition models have witnessed the rapid development of Deep Neural Networks (DNNs). However, these models remain not robust to adversarial examples that are created by adding imperceptible perturbations to clean samples. Recent studies indicate that generating adversarial examples in the hard-label black-box setting is particularly challenging yet highly practical. Compared to image recognition models, there are few hard-label black-box adversarial example generation algorithms for video recognition models. To this end, we propose a hard-label black-box video adversarial example generation algorithm, referred to as Dynamic Black-box Algorithm (DBA). First, DBA uses the binary search algorithm to find the boundary video between two original videos; then, the sampling-based algorithm is used to estimate the gradient on the boundary video; finally, with a dynamic step size adjustment strategy, DBA moves the boundary video towards the direction of the estimated gradient to generate the adversarial video. Additionally, we designed another strategy to skip invalid samples generated during the adversarial example generation process. Experiments demonstrate that DBA attains a superior trade-off between the magnitude of perturbations and query efficiency. Specifically, DBA outperforms state-of-the-art algorithms, achieving an average reduction in Mean Squared Error (MSE) of over 50%. Full article

Accurate node localization in wireless sensor networks (WSNs) is challenging under variable signal propagation and strict energy constraints. This paper presents a hybrid localization framework that combines a deep neural network (DNN) with particle swarm optimization (PSO) to improve accuracy while reducing energy consumption. The DNN learns the non-linear mapping from received signal strength indicator (RSSI) measurements to node coordinates, mitigating propagation effects. PSO jointly optimizes key DNN hyperparameters and selects a minimal subset of anchor nodes that preserve localization performance, thereby lowering communication overhead. Simulation results on 200-node networks show that the proposed DNN–PSO achieves a mean localization error (MLE) of 0.87 m, outperforming a standard DNN (1.32 m) and classical multilateration (3.84 m). The optimized anchor selection reduces per-cycle energy consumption by 23% (239 mJ to 184 mJ) while maintaining sub-meter accuracy. Performance remains stable across diverse propagation conditions and scales well with increasing network size. These results indicate that the proposed approach provides an effective accuracy–energy trade-off for resource-constrained IoT/WSN deployments requiring reliable localization. Full article

The proliferation of next-generation Internet infrastructures and the Internet of Things (IoT) has exponentially increased network traffic complexity. While deep learning (DL)-based intrusion detection systems (IDSs) show immense potential, they persistently suffer from challenges including high computational overhead, vanishing gradients in deep architectures, and acute sensitivity to noise. Consequently, these issues impede their real-time deployment in resource-constrained edge computing environments. To overcome these limitations, we propose a novel, lightweight, and robust intrusion detection framework based on deep neural networks (DNNs). Initially, we employ a Robust Scaler-based statistical preprocessing strategy to supersede traditional Z-score standardization, effectively mitigating the adverse impacts of outliers and burst traffic noise. Subsequently, we design an advanced architecture that integrates self-normalizing residual blocks with a channel attention mechanism. Leveraging compressed hidden layers alongside the Scaled Exponential Linear Unit (SELU) activation function, this architecture not only mitigates the vanishing gradient problem but also amplifies critical traffic features. Concurrently, it achieves a substantial reduction in both parameter count and inference latency. Furthermore, we introduce a cosine annealing strategy to dynamically adjust the learning rate during training, thereby facilitating the model’s escape from local optima and accelerating convergence. Extensive experiments on standard benchmark datasets demonstrate that our proposed framework achieves superior detection accuracy while maintaining exceptional computational efficiency compared to state-of-the-art baselines. Full article

Neuromorphic computing seeks to mimic structure and function of biological neural systems to enable energy-efficient, adaptive information processing. A critical component of this paradigm is neural encoding—the translation of analog or digital input data into spike-based representations suitable for spiking neural networks (SNNs). This paper provides a comprehensive overview of major neural encoding schemes used in neuromorphic systems, including rate and temporal encoding, as well as latency, interspike interval, phase, and multiplexed encoding. The purpose of this paper is to explore the use of encoding techniques for deep learning applications. We discussed the underlying principles of spike encoding approaches, their biological inspiration, computational efficiency, power consumption, integrated circuit design and implementation, and suitability for various neuromorphic applications. We also presented our research on a hardware-and-software co-design platform for different encoding schemes and demonstrated their performance. By comparing their strengths, limitations, and implementation challenges, we aim to provide insights that will guide the development of more efficient and application-specific neuromorphic systems. We also performed an encoder performance analysis via Python 3.12 simulations to compare classification accuracies across these spike encoders on three popular image and video datasets. The performance of neural encoders working with both deep neural networks (DNNs) and SNNs is analyzed. Our performance data is largely consistent with the benchmark data on image classification from other papers, while limited performance data on the University of Central Florida’s 101 (UCF-101) video dataset were found in comparable studies on spike encoders. Based on our encoder performance data, the Interspike Interval (ISI) encoder performs well across all three datasets, preserving continuous, detailed spike timing and richer temporal information for standard classification tasks. Further, for image classification, multiplexing encoders outperform other spike encoders as they simplify timing patterns by enforcing phase locking and improve stability and robustness to noise. Within the SNN testbenches, the ISI-Phase encoder achieved the highest accuracy on the Modified National Institute of Standards and Technology (MNIST) dataset, surpassing the Time-To-First Spike (TTFS) encoder by 1.9%. On the Canadian Institute For Advanced Research (CIFAR-10) dataset, the ISI encoder achieved the highest accuracy. This ISI encoder had 22.7% higher accuracy than the TTFS encoder on the CIFAR-10 dataset. The ISI encoder performed best on the UCF-101 dataset, achieving 12.7% better performance than the TTFS encoder. Full article

Accurate electricity load forecasting has become increasingly challenging in Türkiye due to rapid structural changes in the power system driven by renewable energy expansion. Between 2016 and 2022, solar capacity increased by 130% and wind generation by 83%, resulting in renewable-induced variability exceeding 160%. To assess how different forecasting approaches respond to this evolving environment, Facebook Prophet, XGBoost, and Deep Neural Networks (DNNs) were evaluated using more than 55,000 hourly load observations under a strictly chronological out-of-sample validation framework. The comparative analysis reveals substantial differences in model performance. XGBoost achieved the highest forecasting accuracy, with a Mean Absolute Error of 981.48 MWh, a Root Mean Squared Error of 1344.15 MWh, and a Mean Absolute Percentage Error of 2.72%, while effectively capturing rapid intraday variations and maintaining peak deviations within ±1100 MWh. DNN models delivered competitive overall accuracy (MAE: 997.82 MWh; MAPE: 2.77%) but exhibited a tendency to smooth temporal variations, leading to an underestimation of extreme winter peaks by up to 4100 MWh. In contrast, Prophet showed limited adaptability to the observed structural volatility, producing errors nearly seven times higher than XGBoost (MAE: 7041.79 MWh; RMSE: 8718.14 MWh). Based on these findings, a layered forecasting framework is proposed, employing XGBoost for short-term operational dispatch and reserving statistical models for long-term planning and policy analysis. Full article

With the widespread application of automatic fire alarm systems in various types of buildings, the problem of fire false alarms has gradually become prominent, which not only causes resource waste, but also may reduce users’ trust in the alarm system, thereby affecting the efficiency of emergency response in actual fires. According to data from a certain fire cloud platform, 99.85% of the suspected fires predicted by its system are false alarms. Although existing models can recognize most fire accidents, the accuracy of fire alarm recognition is only 0.15%, due to loose judgment logic, which still requires a large amount of manpower to verify alarms. This article analyzes a large amount of false alarm data and explores the main causes of false alarms, including environmental interference, equipment failure, and improper human operation. By using a fire dynamics simulator (FDS) to establish fire simulation models under different data settings, horizontal and vertical multi-scene fire simulation data are obtained. The study combines simulation and platform data to form a fire and false alarm dataset using a one-dimensional convolutional neural network (1D-CNN) and deep neural network (DNN) deep learning techniques to learn the deductive rules of the fire scene, establish a two-stage judgment model, and gradually, accurately, judge the results. By quantifying the precision, recall, and F1 score of the model, a deep learning model designed to accurately identify genuine fire alarms while filtering out false ones is proposed that can significantly reduce the false alarm rate. The results indicate that the model can identify 1705 false alarms out of 2255 highly suspected true alarms identified by existing systems in multiple practical scenarios and eliminate 75.61% of false positive alarms. On the premise of ensuring an authenticity recognition rate greater than 98%, the accuracy of fire alarm recognition increased from 0.15% to 28.85%, which will significantly reduce the workload of staff verifying alerts, and has good practical value. Full article

Industrial-Internet security faces a core challenge: improving detection accuracy for critical minority-class network attacks. The existing intrusion detection methods based on Conditional Generative Adversarial Nets (CGANs) aim to achieve data balance by reconstructing minority-class attack samples. However, they encounter problems such as generating deceptive samples, poor sample quality, vanishing gradients and difficulties in training. This paper proposes an intrusion detection method based on the Multi-Discriminator Conditional Classification Generative Adversarial Network (MDCCGAN), an improved variant of CGAN, which integrates multiple discriminators and an independent classifier into the traditional CGAN framework. The multiple discriminators reduce the probability of generating deceptive samples, the independent classifier decouples the classification loss to clarify the direction of gradient updates, and the introduction of the Wasserstein distance fundamentally addresses the gradient-vanishing problem. Experiments conducted on the NSL-KDD and UNSW-NB15 datasets demonstrate that the proposed method significantly improves the recall, F1-score and accuracy for minority-class attacks. Specifically, on the NSL-KDD dataset, the overall accuracy increases from 74% to 94%, and the F1-score for the extremely rare U2R attack surges from 0% to 77%. Similarly, on the UNSW-NB15 dataset, the accuracy reaches 88%, a 10% improvement over the baseline DNN, and the F1-scores for extreme minority attacks such as Analysis, Backdoor, and Worms improved to 97%, 62%, and 84%, respectively. These results confirm that our method effectively outperforms traditional generation models and common class-balancing methods. It provides reliable technical support for industrial-Internet security. Full article

Underwater acoustic coatings are widely used to suppress low-frequency noise radiation and sonar reflection in underwater vehicles. In this study, an underwater acoustic coating model consisting of viscoelastic rubber layers and micro-perforated panel (MPP) structures is investigated, with particular emphasis on the low-frequency sound absorption mechanism and predictive modeling. Based on an improved transfer function method, a novel Micro-Perforated Panel Acoustic Coating Layer (MPPACL) model is developed to describe the coupled acoustic behavior of multilayer coatings under underwater conditions. The low-frequency sound absorption performance is primarily governed by the viscoelastic characteristics of the rubber layer, including material damping and complex modulus, while the incorporation of the MPP further enhances absorption through resonance effects. To efficiently explore the relationship between structural parameters and acoustic response, an ensemble learning-based deep neural network (ELDNN) is constructed using analytically generated data, enabling both forward prediction of sound absorption performance and inverse prediction of structural design parameters. The results show that the frequency prediction accuracy of the IDNN model is 3.7 times that of the DNN model. Furthermore, the proposed MPPACL model has achieved a significantly enhanced sound absorption effect within the frequency range of 50 to 2000 hertz. This effect has also been further verified through underwater experiments. The proposed framework provides an efficient and reliable approach for the design and optimization of underwater acoustic coatings. Full article

This paper investigates M-PSK symbol detection in Orthogonal Frequency Division Multiplexing (OFDM) systems for wideband Vehicle-to-Vehicle (V2V) communications using lightweight convolutional neural networks. In doubly dispersive channels, Inter-Carrier Interference (ICI) degrades subcarrier orthogonality, rendering conventional equalization ineffective. Current ICI mitigation techniques face a trade-off between Bit-Error Rate (BER) performance and computational complexity, limiting their applicability in dynamic vehicular scenarios. To address this issue, a low-complexity MobileNetV3-based receiver is proposed, incorporating a signal-model-driven preprocessing stage that compensates for Doppler-induced phase distortions responsible for ICI. Simulation results show that the proposed receiver improves BER performance compared to conventional equalizers and recent neural-based schemes in the low-SNR regime (below 15 dB) while maintaining computational complexity close to linear least-squares detection. Full article

Laser powder bed fusion (LPBF) enables controlled gyroid lattices, but mapping both process and design to performance remains challenging when datasets are small and interactions are non-linear. In this study, data-driven models that link energy density and lattice geometry to Young’s modulus and yield strength were established for sheet and network gyroid architectures. To stabilise small-data learning, stacked-autoencoder pre-training was benchmarked against greedy layer-wise pre-training. Compression characterisation data at under-represented energy-density conditions were added to fill data gaps and validate predictions. The models support property-driven design in which given modulus and yield strength targets inform a method that returns feasible combinations of laser powder bed fusion settings and gyroid density and size. Pre-trained models reduced error and captured the relationship between stiffness and density and between strength and density, with yield strength prediction errors of 3.51% for sheet architectures and 8.76% for network architectures. Young’s modulus showed a higher variability that is consistent with sensitivities in LPBF such as surface roughness and thin walls. This work contributes an artificial intelligence method for manufacturing datasets using stacked autoencoder pre-training with fine-tuning, and an inverse-design workflow that maps energy density and gyroid geometry to Young’s modulus and yield strength in titanium lattices. Full article