Search Results (33)

This paper presents a lightweight MCU–EdgeTPU platform—a microcontroller unit (MCU) paired with an Edge Tensor Processing Unit (EdgeTPU) accelerator—for onboard drone-perception experiments, extended from an open-source baseline originally limited to Quarter Video Graphics Array (QVGA) single-camera operation. Rather than treating hardware, runtime, model, and data as separate problems, they are developed as parts of the same continuous perception pipeline. The platform extends the hardware baseline toward dual 5 Mpx sensing, onboard inertial measurement unit (IMU) support, real-time embedded inference, and a high-level MicroPython control layer. In parallel, lightweight You Only Look Once (YOLO) detectors are trained and selected on a synthetic aerial-person dataset generated under the visual conditions expected by the drone camera, including target resolution, viewpoint, object scale, weather, lighting, and time-of-day variation. The resulting workflow starts from both ends: the detector must be small and quantization-stable enough for the EdgeTPU path, while the dataset must match the images that the onboard sensor is expected to observe. To evaluate the system, the full path from camera capture and image conversion to TPU transfer, model execution, and post-inference processing is analyzed. In the tested setup, the optimized single-camera pipeline runs stably with no timeouts or inference failures at about 26 detections per second with standard RGB input; because each EdgeTPU invocation is bounded by the USB transfer of the input image, feeding the camera’s native YUV420 format instead halves that transfer and raises throughput to about 40 detections per second at the same accuracy, while the selected 8-bit-integer (INT8) person detector preserves most of its 32-bit floating-point (FP32) accuracy. Detections are exposed to drone-control workflows (MAVLink/PX4 and Crazyflie) through the scriptable layer as an integration interface rather than a validated autonomy stack. The central contribution is therefore a co-designed embedded perception pipeline in which the board, runtime, detector, dataset, and even the camera pixel format are aligned around the same operating conditions. Full article

Intrusion detection in IoT and IIoT networks must operate under tight resource constraints, yet most published machine learning-based IDS solutions report accuracy on held-out data without addressing whether the trained model can actually run on the target hardware. We address this gap with an end-to-end study spanning dataset preprocessing, model training, INT8 quantisation, and physical execution on two real microcontrollers. Five supervised classifiers—Logistic Regression, Decision Tree (depth 5), Random Forest, XGBoost, and LightGBM—plus an MLP deep learning baseline are evaluated on binary and ten-class intrusion detection tasks using the TON_IoT network dataset. A 5-fold stratified cross-validation confirms stable performance across splits, with LightGBM reaching $F_1=0.9993\pm 0.0001$. Models are then exported through three quantisation pipelines: m2cgen C code generation for the two lightest classifiers, TensorFlow Lite Micro full-integer INT8 for the MLP (9.34× size reduction to 13.03 KB), and a custom post-training INT8 binary format for XGBoost and LightGBM (18.91× compression for LightGBM to 73.85 KB). All five quantised models are deployed to an Arduino Mega 2560 (ATmega2560, 16 MHz, 8 KB SRAM) and an ESP32-C3 SuperMini (RISC-V, 160 MHz, 400 KB SRAM) and benchmarked on physical hardware across 500 timed inferences per model (250 per input class), with firmware predictions confirmed to match the Python 3.11 float model on both test vectors. The Decision Tree achieves 5.6 µs inference on the ESP32-C3; LightGBM INT8 ($F_1=0.9992$) provides the best accuracy–size trade-off among ensemble models. Cross-platform comparison reveals that the RISC-V device is 5.8–7.8× faster than the 8-bit AVR for identical model code. A cross-domain evaluation using CIC-IoT-Dataset2023 identifies large normalised distribution shifts (up to $\delta =5.95$ in packet asymmetry), quantifying the generalisation gap that remains an open challenge. Full article

Machine learning approaches to wildfire spread prediction are constrained by the lack of standardized, multi-source, spatiotemporal datasets that fuse terrain, weather, and fire-state information into a single ML-ready format. We present WildfireCube, a reproducible event-centric pipeline and methodology for constructing dense fourth-order spatiotemporal tensors of shape (T, C, H, W) at 30 m spatial and 3 h temporal resolution. Following the analysis-ready data convention established in the Earth Observation community, the pipeline fuses four open data sources: the Copernicus GLO-30 Digital Elevation Model for static terrain derivatives, ERA5-Land reanalysis for hourly weather forcing, Sentinel-2 Level-2A imagery for spectral vegetation and burn-severity indices, and NASA FIRMS active-fire hotspot detections for fire-state reconstruction via ordinary kriging. The resulting 13-channel normalized tensor separates causal drivers into three physically motivated groups: static landscape controls (elevation, slope, aspect, fuel load), dynamic atmospheric forcings (wind components, temperature, precipitation), and evolving fire state (fire-front mask, burn severity, fractional burn, observation confidence). A physics-informed normalization framework maps all channels to bounded ranges using fixed physical constants rather than sample statistics, ensuring cross-event comparability and exact invertibility. We demonstrate the pipeline on 13 wildfire events across the United States, Canada, and Greece (2017–2023), producing a processed catalog exceeding 300 GB compressed and spanning a 14-fold range in burned area, a 27 °C range in mean temperature, and different fire regimes. Event tensors are stored in chunked Zarr archives with Zstandard compression, achieving a 2.58× compression ratio. As future work, the pipeline will be applied to a 40-event target catalog projected to exceed 2 TB of raw data, providing the multi-regime diversity and scale required for training robust deep learning models for spatiotemporal wildfire prediction. Full article

We investigatewhether a finite local information capacity of spacetime can account for the gravitational phenomena commonly attributed to cold dark matter. Starting from a covariant effective-field-theory description, we modelcoarse-grained entropy deposition as a dynamical scalar field $S\left(x\right)$ whose stress–energy tensor contributes to structure formation. The macroscopic action contains a single dimensionless coupling λ multiplying the canonical kinetic term, ensuring ghost-free dynamics and conservation of the associated stress–energy tensor. In a slow-roll regime, defined by a covariant source term $\Gamma \equiv \ddot{S}+3H\dot{S}=0$, where H is the Hubble parameter and overdot denotes derivative with respect to cosmic time, and $|\ddot{S}|\ll H|\dot{S}|$, the entropy sector behaves as pressureless dust at background and in linear order. Implemented in a modified Cosmic Linear Anisotropy Solving System (CLASS) Boltzmann solver, the entropy component fits Planck satellite 2018 cosmic microwave background (CMB) data, baryon acoustic oscillation (BAO) measurements, and the Pantheon + Type Ia supernova sample for $0.5\lesssim \lambda \lesssim 2$, while preserving the linear growth factor to within 0.2% over Euclid space telescope scales. To regulate ultraviolet contributions, we introduce a holographically motivated prescription in which gravitationally active entropy deposition is confined to causal two-surfaces, yielding a $\rho \propto r^{-2}$ halo envelope with a finite-density core determined by local entropy saturation. Fixing the flux scale $\mathcal{A}$ from astrophysical entropy budgets reproduces Milky-Way-mass halos without introducing fine-tuned length scales. Pilot N-body simulations that evolve the entropy field on a staggered grid reproduce the halo mass function down to ${10}^{10.5}{M}_{\odot }$, mitigate the cusp–core and missing-satellite tensions, and remain consistent with cluster lensing constraints. On linear scales, the model predicts percent-level, scale-dependent deviations in the lensing convergence and matter power spectra, testable by Euclid space telescope, the Roman Space Telescope High Latitude Survey, and the CMB-S4 experiment. Full article

Water quality monitoring remains a critical challenge for achieving Sustainable Development Goal 6, particularly in rural and resource-constrained environments where conventional laboratory-based methods are costly and slow. This study presents the development and field validation of an Artificial Intelligence of Things (AIoT) ecosystem for intelligent, low-cost, and real-time water quality assessment using edge computing and generative artificial intelligence. The system integrates a laboratory-developed multiparameter probe measuring temperature, pH, dissolved oxygen, and electrical conductivity with a mobile application and a cloud-based backend. Field validation was conducted in riverine environments in the municipality of Pueblo Bello (Cesar, Colombia), where the system was deployed for in situ data acquisition and real-time inference. A supervised Artificial Neural Network (ANN) was trained to classify water quality based on a Water Quality Index (WQI) ground truth derived from a public dataset, employing KNN-based missing data imputation, interquartile range outlier filtering, stratified balancing, and grid search hyperparameter optimization. The best-performing model achieved 85.1% accuracy and an AUC of 0.87 using only four physical parameters and was successfully deployed in TensorFlow Lite format on both the embedded probe and the mobile application with sub-millisecond inference time. Integration with a generative AI backend provides contextual natural-language interpretations of measurements. These results demonstrate that reduced-parameter edge AI systems can provide reliable environmental diagnostics while enhancing accessibility and citizen engagement for participatory water monitoring. Full article

Microwave imaging is emerging as an alternative to conventional medical diagnostic techniques. Traditional analytical and numerical methods fail to adequately address these fundamental challenges: they often rely on strict linear approximations or simplified physical models, leading to low reconstruction accuracy, poor robustness, and limited generalization ability in complex clinical scenarios. As a result, they cannot meet the high-precision requirements of practical stroke microwave imaging. To further improve the accuracy of microwave imaging algorithms in recognizing stroke regions and solving the backscattering problem, this study employs a combination of methods with deep learning. It presents the Multi-Scale Attention Network (MSA-Net) for microwave imaging. The network is based on the EGE-UNet network structure with improved multi-axis Hadamard attention, incorporating null-space pyramid pooling and introducing a deep supervisory mechanism to improve the network performance further. To combine microwave imaging with deep learning, firstly, a large amount of microwave data need to be simulated with HFSS, in which the simulation model is a human brain stroke model constructed by an HFSS simulation system. Secondly, the microwave data obtained from the simulation are converted into a tensor format. Then, the tensor data are input into the MSA-Net neural network, which generates a binary mask image that can be used to detect the size and location of the stroke. This study also prompts the model to converge faster by sparsifying the microwave data to improve training efficiency. The method has been tested using simulation data, and based on the comparison experiments with other networks, MSA-Net is more accurate in detecting the location and the bleed size. The experimental results show that the proposed method is superior for stroke imaging. The experimental results show that the proposed model achieves a 1.08 improvement in peak signal-to-noise ratio and a 0.017 reduction in learned perceptual image block similarity, fully validating the effectiveness of the structural optimization strategy proposed in this paper. Full article

In practical well-logging datasets, severe missing values, anomalous disturbances, and highly imbalanced lithology classes are pervasive. To address these challenges, this study proposes a well-logging lithology identification framework that combines Robust Feature Engineering (RFE) with quality-aware XGBoost. Instead of relying on interpolation-based data cleaning, RFE uses sentinel values and a meta-information tensor to explicitly encode patterns of missingness and anomalies, and incorporates sliding-window context to transform data defects into discriminative auxiliary features. In parallel, a quality-aware sample-weighting strategy is introduced that jointly accounts for formation boundary locations and label confidence, thereby mitigating training bias induced by long-tailed class distributions. Experiments on the FORCE 2020 lithology prediction dataset demonstrate that, relative to baseline models, the proposed method improves the weighted F1 score from 0.66 to 0.73, while Boundary F1 and the geological penalty score are also consistently enhanced. These results indicate that, compared with traditional workflows that rely solely on data cleaning, explicit modeling of data incompleteness provides more pronounced advantages in terms of robustness and engineering applicability. Full article

Transformer-based time series models are being increasingly employed for time series data analysis. However, their training remains memory intensive, especially with high-dimensional data and extended look-back windows, while model-level memory optimizations are well studied, the batch formation process remains an underexplored factor to performance inefficiency. This paper introduces a memory-efficient batching framework based on view-based sliding windows operating directly on GPU-resident tensors. This approach eliminates redundant data materialization caused by tensor stacking and reduces data transfer volumes without modifying model architectures. We present two variants of our solution: (1) per-batch optimization for datasets exceeding GPU memory, and (2) dataset-wise optimization for in-memory workloads. We evaluate our proposed batching framework systematically using peak GPU memory consumption and epoch runtime as efficiency metrics across varying batch sizes, sequence lengths, feature dimensions, and model architectures. Results show consistent memory savings, averaging 90% and runtime improvements of up to 33% across multiple transformer-based models (Informer, Autoformer, Transformer, and PatchTST) and a linear baseline (DLinear) without compromising model accuracy. We extensively validate our method using synthetic and standard real-world benchmarks, demonstrating accuracy preservation and practical scalability in distributed GPU environments. The proposed method highlights batch formation process as a critical component for improving training efficiency. Full article

Background/Objectives: Cerebral palsy (CP), often caused by early brain injury such as perinatal stroke or hemorrhage, is the most common lifelong motor disability. Early identification of at-risk infants and timely access to rehabilitation interventions are essential for improving long-term outcomes. The General Movements Assessment (GMA), performed in the first months of life, has high sensitivity and specificity to predict CP; however, the neurological correlates of general movements remain unclear. This analysis aimed to investigate the relationship between white matter integrity and general movements in infants with perinatal brain injury using advanced neuroimaging techniques. Methods: Diffusion-weighted MRI data were analyzed in 17 infants, 12 with perinatal brain injury and 5 typically developing infants. Tractography was used to identify the corticospinal tract, a key motor pathway often affected by perinatal brain injury, and tract-based spatial statistics (TBSS) were used to examine broader white matter networks. Diffusion parameters from the diffusion tensor imaging (DTI) and neurite orientation dispersion and density imaging (NODDI) models were compared between infants with and without typical general movements. Results: Corticospinal tract integrity did not differ between groups when averaged across hemispheres. However, infants with asymmetric general movements exhibited greater corticospinal tract asymmetries. A subset of infants with atypical general movement trajectories at <6 weeks and 3–5 months of age showed reduced corticospinal tract integrity compared to those with typical general movements. TBSS revealed significant differences in white matter integrity between infants with typical and atypical general movements in several white matter pathways, including the corpus callosum, the right posterior corona radiata, bilateral posterior thalamic radiations, the left fornix/stria terminalis, and bilateral tapetum. Conclusions: These findings support and expand upon previous research suggesting that white matter integrity across multiple brain regions plays a role in the formation of general movements. Corticospinal integrity alone was not strongly associated with general movements; interhemispheric and cortical-subcortical connectivity appear critical. These findings underscore the need for further research in larger, diverse populations to refine early biomarkers of neurodevelopmental impairment and guide targeted interventions. Full article

In this paper, we present a methodology for the numerical solving of partial differential equations in 2D geometries with piecewise smooth boundaries via finite element method (FEM) using a Quantized Tensor Train (QTT) format. During the calculations, all the operators and data are assembled and represented in a compressed tensor format. We introduce an efficient assembly procedure of FEM matrices in the QTT format for curvilinear domains. The features of our approach include efficiency in terms of memory consumption and potential expansion to quantum computers. We demonstrate the correctness and advantages of the method by solving a number of problems, including nonlinear incompressible Navier–Stokes flow, in differently shaped domains. Full article

Sparse matrix–matrix multiplication (SpMM) is essential for deep learning models and scientific computing. Recently, Tensor Cores (TCs) on GPUs, originally designed for dense matrix multiplication with mixed precision, have gained prominence. However, utilizing TCs for SpMM is challenging due to irregular memory access patterns and a varying number of non-zero elements in a sparse matrix. To improve data locality, previous studies have proposed reordering sparse matrices before multiplication, but this adds computational overhead. In this paper, we propose Tensor Core-Adapted SpMM (TCA-SpMM), which leverages TCs without requiring matrix reordering and uses the compressed sparse row (CSR) format. To optimize TC usage, the SpMM algorithm’s dot product operation is transformed into a blocked matrix–matrix multiplication. Addressing load imbalance and minimizing data movement are critical to optimizing the SpMM kernel. Our TCA-SpMM dynamically allocates thread blocks to process multiple rows simultaneously and efficiently uses shared memory to reduce data movement. Performance results on sparse matrices from the Deep Learning Matrix Collection public dataset demonstrate that TCA-SpMM achieves up to 29.58× speedup over state-of-the-art SpMM implementations optimized with TCs. Full article

Earth data is essential for global environmental studies. Many Earth data types are naturally modeled by multidimensional arrays (tensors). Array (Tensor) DBMSs strive to be the best systems for tensor-related workloads and can be especially helpful for Earth data engineering, which takes up to 80% of Earth data science. We present a new quantum Array (Tensor) DBMS data model and new quantum approaches that rely on the upcoming quantum memory and demonstrate exponential speedups when applied to many of the toughest Array (Tensor) DBMS challenges stipulated by classical computing and real-world Earth data use-cases. We also propose new types of charts: Quantum Gantt (QGantt) Charts and Quantum Network Diagrams (QND). QGantt charts clearly illustrate how multiple operations occur simultaneously across different data items and what are the input/output data dependencies between these operations. Unlike traditional Gantt charts, which typically track project timelines and resources, QGantt charts integrate specific data items and operations over time. A Quantum Network Diagram combines several QGantt charts to show dependencies between multistage operations, including their inputs/outputs. By using a static format, QGantt charts and Quantum Network Diagrams allow users to explore complex processes at their own pace, which can be beneficial for educational and R&D purposes. Full article

Designing drilling mud rheology is a complex task, particularly when it comes to preventing filter cakes from obstructing formation pores and making sure they can be easily decomposed using breakers. Incorporating both multiphysics and data-driven numerical simulations into the design of mud rheology experiments creates an additional challenge due to their symmetrical integration. In this computational intelligence study, we introduced numerical validation techniques using 498 available datasets from mud rheology and images from filter cakes. The goal was to symmetrically predict flow, maximize filtration volume, monitor void spaces, and evaluate formation damage occurrences. A neural-objective and image optimization approach to drilling mud rheology automation was employed using an artificial neural network feedforward (ANN-FF) function, a non-ANN-FF function, an image processing tool, and an objective optimization tool. These methods utilized the Google TensorFlow Sequential API-DNN architecture, MATLAB-nftool, the MATLAB-image processing tool, and a single-objective optimization algorithm. However, the analysis emanating from the ANN-FF and non-ANN-FF (with neurons of 10, 12, and 18) indicated that, unlike non-ANN-FF, ANN-FF obtained the highest correlation coefficient of 0.96–0.99. Also, the analysis of SBM and OBM image processing revealed a total void area of 1790 M µm² and 1739 M µm², respectively. Both SBM and OBM exhibited notable porosity and permeability that contributed to the enhancement of the flow index. Nonetheless, this study did reveal that the experimental-informed single objective analysis impeded the filtration volume; hence, it demonstrated potential formation damage. It is, therefore, consistent to note that automating flow predictions from mud rheology and filter cakes present an alternative intelligence method for non-programmers to optimize drilling productive time. Full article

Classification is an important means of extracting rich information from hyperspectral images (HSIs). However, many HSIs contain shadowed areas, where noise severely affects the extraction of useful information. General noise removal may lead to loss of spatial correlation and spectral features. In contrast, dynamic stochastic resonance (DSR) converts noise into capability that enhances the signal in a way that better preserves the image’s original information. Nevertheless, current one-dimensional and 2D DSR methods fail to fully utilize the tensor properties of hyperspectral data and preserve the complete spectral features. Therefore, a hexa-directional differential format is derived in this paper to solve the system’s output, and the iterative equation for HSI shadow enhancement is obtained, enabling 3D parallel processing of HSI spatial–spectral information. Meanwhile, internal parameters are adjusted to achieve optimal resonance. Furthermore, the residual neural network 152 model embedded with the convolutional block attention module is proposed to diminish information redundancy and leverage data concealed within shadow areas. Experimental results on a real-world HSI demonstrate the potential performance of 3D DSR in enhancing weak signals in HSI shadow regions and the proposed approach’s effectiveness in improving classification. Full article

As a promising data analysis technique, sparse modeling has gained widespread traction in the field of image processing, particularly for image recovery. The matrix rank, served as a measure of data sparsity, quantifies the sparsity within the Kronecker basis representation of a given piece of data in the matrix format. Nevertheless, in practical scenarios, much of the data are intrinsically multi-dimensional, and thus, using a matrix format for data representation will inevitably yield sub-optimal outcomes. Tensor decomposition (TD), as a high-order generalization of matrix decomposition, has been widely used to analyze multi-dimensional data. In a direct generalization to the matrix rank, low-rank tensor modeling has been developed for multi-dimensional data analysis and achieved great success. Despite its efficacy, the connection between TD rank and the sparsity of the tensor data is not direct. In this work, we introduce a novel tensor ring sparsity measurement (TRSM) for measuring the sparsity of the tensor. This metric relies on the tensor ring (TR) Kronecker basis representation of the tensor, providing a unified interpretation akin to matrix sparsity measurements, wherein the Kronecker basis serves as the foundational representation component. Moreover, TRSM can be efficiently computed by the product of the ranks of the mode-2 unfolded TR-cores. To enhance the practical performance of TRSM, the folded-concave penalty of the minimax concave penalty is introduced as a nonconvex relaxation. Lastly, we extend the TRSM to the tensor completion problem and use the alternating direction method of the multipliers scheme to solve it. Experiments on image and video data completion demonstrate the effectiveness of the proposed method. Full article