Papers

The proliferation of wireless devices necessitates more robust and reliable emitter detection and identification for critical tasks such as spectrum management and network security. Existing studies exploring methods for unknown emitters identification, however, are typically hindered by their dependence on labeled or proprietary datasets, unrealistic assumptions (e.g. all samples with identical transmitted messages), or deficiency of systematic evaluations across different architectures and design dimensions. In this work, we present a comprehensive evaluation of unknown emitter detection systems across key aspects of the design space, focusing on data modality, learning approaches, and feature learning modules. We demonstrate that prior self-supervised, zero-shot emitter detection approaches commonly use datasets with identical transmitted messages. To address this limitation, we propose a 2D-Constellation data modality for scenarios with varying messages, achieving up to a 40% performance improvement in ROC-AUC, NMI, and F1 metrics compared to conventional raw I/Q data. Furthermore, we introduce interpretable Kolmogorov-Arnold Networks (KANs) to enhance model transparency, and a Singular Value Decomposition (SVD)-based initialization procedure for feature learning modules operating on sparse 2D-Constellation data, which improves the performance of Deep Clustering approaches by up to 40% across the same metrics comparing to the modules without SVD initialization. We evaluate all data modalities and learning modules across three learning approaches: Deep Clustering, Auto Encoder and Contrastive Learning.

Classical phasor analysis is fundamentally limited to sinusoidal single-frequency conditions, which poses challenges when working in the presence of harmonics. Furthermore, the conventional solution, which consists of decomposing signals using Fourier series and applying superposition, is a fragmented process that does not provide a unified solution in the frequency domain. This paper overcomes this limitation by introducing a complete and direct approach for multi-harmonic AC circuits using Geometric Algebra (GA). In this way, all non-sinusoidal voltage and current waveforms are represented as simple vectors in a $2N$-dimensional Euclidean space. The relationship between these vectors is characterized by a single and unified geometric transformation termed the rotoflex. This operator elevates the concept of impedance from a set of complex numbers per frequency to a single multivector that holistically captures the circuit response, while unifying the magnitude scale (flextance) and phase rotation (rotance) across all harmonics. Thus, this work establishes GA as a structurally unified and efficient alternative to phasor analysis, providing a more rigorous foundation for electrical circuit analysis. The methodology is validated through case studies that demonstrate perfect numerical consistency with traditional methods and superior performance.

This paper presents the first experimental validation of reflective near-field beamfocusing using a reconfigurable intelligent surface (RIS). While beamfocusing has been theoretically established as a key feature of large-aperture RISs, its practical realization has remained unexplored. We derive new analytical expressions for the array gain achieved with a $b$-bit RIS in near-field line-of-sight scenarios, characterizing both the finite depth and angular width of the focal region. The theoretical results are validated through a series of measurements in an indoor office environment at 28 GHz using a one-bit 1024-element RIS. The experiments confirm that beamfocusing can be dynamically achieved and accurately predicted by the proposed simplified analytical model, despite the presence of hardware imperfections and multipath propagation. These findings demonstrate that near-field beamfocusing is a robust and practically viable feature of RIS-assisted wireless communications.

This paper presents a Koopman-based model predictive control (MPC) framework for safe UAV navigation in dynamic environments using real-time LiDAR data. By leveraging the Koopman operator to linearly approximate the dynamics of surrounding objets, we enable efficient and accurate prediction of the position of moving obstacles. Embedding this into an MPC formulation ensures robust, collision-free trajectory planning suitable for real-time execution. The method is validated through simulation and ROS2-Gazebo implementation, demonstrating reliable performance under sensor noise, actuation delays, and environmental uncertainty.

Lithium-ion (Li-ion) batteries are ubiquitous in electric vehicles (EVs) as efficient energy storage devices. The reliable operation of Li-ion batteries depends critically on the accurate estimation of battery capacity. However, conventional estimation methods require extensive training datasets from costly battery tests for modeling, and a full cycle of charge and discharge is often needed to estimate the capacity. To overcome these limitations, we propose a novel capacity estimation method that leverages only one cycle of the open-circuit voltage (OCV) test in modeling and allows for estimating the capacity from partial charge or discharge data. Moreover, by applying it with OCV identification algorithms, we can estimate the capacity from dynamic discharge data without requiring dedicated data collection tests. We observed an invariance property in the OCV versus state of charge relationship across aging cycles. Leveraging this invariance, the proposed method estimates the capacity by solving an OCV alignment problem using only the OCV and the discharge capacity data from the battery. Simulation results demonstrate the method's efficacy, achieving a mean absolute relative error of 0.85% in capacity estimation across 12 samples from 344 aging cycles.

Integrated sensing and communication (ISAC) systems demand precise and efficient target localization, a task challenged by rich multipath propagation in complex wireless environments. This paper introduces MARBLE-Net (Multipath-Aware Rainbow Beam Learning Network), a deep learning framework that jointly optimizes the analog beamforming parameters of a frequency-dependent rainbow beam and a neural localization network for high-accuracy position estimation. By treating the phase-shifter (PS) and true-time-delay (TTD) parameters as learnable weights, the system adaptively refines its sensing beam to exploit environment-specific multipath characteristics. A structured multi-stage training strategy is proposed to ensure stable convergence and effective end-to-end optimization. Simulation results show that MARBLE-Net outperforms both a fixed-beam deep learning baseline (RaiNet) and a traditional k-nearest neighbors (k-NN) method, reducing localization error by more than 50% in a multipath-rich scene. Moreover, the results reveal a nuanced interaction with multipath propagation: while confined uni-directional multipath degrades accuracy, structured and directional multipath can be effectively exploited to achieve performance surpassing even line-of-sight (LoS) conditions.

In this paper, the problem of distributed state estimation of human-driven vehicles (HDVs) by connected autonomous vehicles (CAVs) is investigated in mixed traffic transportation systems. Toward this, a distributed observable state-space model is derived, which paves the way for estimation and observability analysis of HDVs in mixed traffic scenarios. In this direction, first, we obtain the condition on the network topology to satisfy the distributed observability, i.e., the condition such that each HDV state is observable to every CAV via information-exchange over the network. It is shown that strong connectivity of the network, along with the proper design of the observer gain, is sufficient for this. A distributed observer is then designed by locally sharing estimates/observations of each CAV with its neighborhood. Second, in case there exist faulty sensors or unreliable observation data, we derive the condition for redundant distributed observability as a $q$-node/link-connected network design. This redundancy is achieved by extra information-sharing over the network and implies that a certain number of faulty sensors and unreliable links can be isolated/removed without losing the observability. Simulation results are provided to illustrate the effectiveness of the proposed approach.

We demonstrate an experimental phase optical time-domain reflectometry (OTDR) system capable of simultaneous detection and classification of various environmental events, such as wind-induced fiber movement, vehicle movement, and audio signatures, with real-time visualization.

This project investigates traffic congestion within North Campus, Delhi University (DU), using continuous time simulations implemented in UXSim to model vehicle movement and interaction. The study focuses on several key intersections, identifies recurring congestion points, and evaluates the effectiveness of conventional traffic management measures. Implementing signal timing optimization and modest intersection reconfiguration resulted in measurable improvements in simulated traffic flow. The results provide practical insights for local traffic management and illustrate the value of continuous time simulation methods for informing short-term interventions and longer-term planning.

Fleets of autonomous vehicles (AV) often are at the core of intelligent transportation scenarios for smart cities, and may require a wireless Internet connection to offload computer vision tasks to data centers located either in the edge or the cloud section of the network. Cooperation among AVs is successful when the environment is unknown, or changes dynamically, so as to improve coverage and trip time, and minimize the traveled distance. The AVs, while mapping the environment with range-based sensors, move across the wireless coverage areas, with consequences on the experienced access bit rate, latency, and handover rate. In this paper, we propose to modify the cost of common path planning algorithms such as Dijkstra and A*, so that the best path solution takes into account not only the traveled distance, but also the radio coverage experience. To this aim, several radio-related cost-weighting functions are introduced and tested, to assess the performance of the proposed approaches with extensive simulations. The proposed mapping algorithm can achieve a mapping error probability below $2\%$, while the proposed path-planning algorithms extend the radio coverage of the AVs, with only a limited increase in traveled distance with respect to shortest-path existing methods, such as conventional Dijkstra and A* algorithms.

Discrete-time stochastic systems with continuous spaces are hard to verify and control, even with MDP abstractions due to the curse of dimensionality. We propose an abstraction-based framework with robust dynamic programming mappings that deliver control strategies with provable lower bounds on temporal-logic satisfaction, quantified via approximate stochastic simulation relations. Exploiting decoupled dynamics, we reveal a Canonical Polyadic Decomposition tensor structure in value functions that makes dynamic programming scalable. The proposed method provides correct-by-design probabilistic guarantees for temporal logic specifications. We validate our results on continuous-state linear stochastic systems.

Future Internet-of-Things networks are envisioned to use small and cheap sensor nodes with extremely low power consumption to avoid the extensive use of batteries. To provide connectivity to a massive number of these nodes, backscatter communication (BC) is emerging as an energy- and cost-efficient technology exploiting the reflection of radio frequency signals. However, challenges such as round-trip path loss and direct link interference (DLI) between the carrier emitter and the reader limit its performance. To tackle these limitations, this paper proposes a joint access point role selection and a novel beamforming technique for bistatic BC in a distributed multiple-input multiple-output setup. The proposed approach boosts the received backscattered energy while effectively mitigating DLI, thereby reducing the error probability. We also propose a channel estimation method tailored to operate under DLI conditions and propose a mismatch detector using estimated channel coefficients. Furthermore, we derive a closed-form expression for the probability of error for the detectors and model the quantization noise caused by DLI. Finally, comprehensive simulation results show that the proposed method with 1-bit analog-to-digital converters (ADCs) effectively mitigates DLI, reduces the quantization noise, and enhances backscattered signal energy, achieving performance comparable to the benchmark scenario with 8-bit ADCs.

Unmanned Aerial Vehicles (UAVs) are widely used for aerial photography and remote sensing applications. One of the main challenges is to stitch together multiple images into a single high-resolution image that covers a large area. Featurebased image stitching algorithms are commonly used but can suffer from errors and ambiguities in feature detection and matching. To address this, several approaches have been proposed, including using bundle adjustment techniques or direct image alignment. In this paper, we present a novel method that uses a combination of IMU data and computer vision techniques for stitching images captured by a UAV. Our method involves several steps such as estimating the displacement and rotation of the UAV between consecutive images, correcting for perspective distortion, and computing a homography matrix. We then use a standard image stitching algorithm to align and blend the images together. Our proposed method leverages the additional information provided by the IMU data, corrects for various sources of distortion, and can be easily integrated into existing UAV workflows. Our experiments demonstrate the effectiveness and robustness of our method, outperforming some of the existing feature-based image stitching algorithms in terms of accuracy and reliability, particularly in challenging scenarios such as large displacements, rotations, and variations in camera pose.

This paper explores the application of vision-based system identification techniques in quadrotor modeling and control. Through experiments and analysis, we address the complexities and limitations of quadrotor modeling, particularly in relation to thrust and drag coefficients. Grey-box modeling is employed to mitigate uncertainties, and the effectiveness of an onboard vision system is evaluated. An LQR controller is designed based on a system identification model using data from the onboard vision system. The results demonstrate consistent performance between the models, validating the efficacy of vision based system identification. This study highlights the potential of vision-based techniques in enhancing quadrotor modeling and control, contributing to improved performance and operational capabilities. Our findings provide insights into the usability and consistency of these techniques, paving the way for future research in quadrotor performance enhancement, fault detection, and decision-making processes.

This work proposes a two-layered control scheme for constrained nonlinear systems represented by a class of recurrent neural networks and affected by additive disturbances. In particular, a base controller ensures global or regional closed-loop l_p-stability of the error in tracking a desired equilibrium and the satisfaction of input and output constraints within a robustly positive invariant set. An additional control contribution, derived by combining the internal model control principle with a stable operator, is introduced to improve system performance. This operator, implemented as a stable neural network, can be trained via unconstrained optimisation on a chosen performance metric, without compromising closed-loop equilibrium tracking or constraint satisfaction, even if the optimisation is stopped prematurely. In addition, we characterise the class of closed-loop stable behaviours that can be achieved with the proposed architecture. Simulation results on a pH-neutralisation benchmark demonstrate the effectiveness of the proposed approach.

Edge AI, which brings artificial intelligence to the edge of the network for real-time processing and decision-making, has emerged as a transformative technology across various applications. However, the deployment of Edge AI systems faces significant challenges due to high energy consumption and extended operation time. In this paper, we consider an Edge AI system which integrates the data acquisition, computation and communication processes, and focus on improving learning performance of this system. We model the time and energy consumption of different processes and perform a rigorous convergence analysis to quantify the impact of key system parameters, such as the amount of collected data and the number of training rounds, on the learning performance. Based on this analysis, we formulate a system-wide optimization problem that seeks to maximize learning performance under given time and energy constraints. We explore both homogeneous and heterogeneous device scenarios, developing low-complexity algorithms based on one-dimensional search and alternating optimization to jointly optimize data collection time and training rounds. Simulation results validate the accuracy of our convergence analysis and demonstrate the effectiveness of the proposed algorithms, providing valuable insights into designing energy-efficient Edge AI systems under real-world conditions.

The integration of spiking neural networks (SNNs) with transformer-based architectures has opened new opportunities for bio-inspired low-power, event-driven visual reasoning on edge devices. However, the high temporal resolution and binary nature of spike-driven computation introduce architectural mismatches with conventional digital hardware (CPU/GPU). Prior neuromorphic and Processing-in-Memory (PIM) accelerators struggle with high sparsity and complex operations prevalent in such models. To address these challenges, we propose a memory-centric hardware accelerator tailored for spiking transformers, optimized for deployment in real-time event-driven frameworks such as classification with both static and event-based input frames. Our design leverages a hybrid analog-digital PIM architecture with input sparsity optimizations, and a custom-designed dataflow to minimize memory access overhead and maximize data reuse under spatiotemporal sparsity, for compute and memory-efficient end-to-end execution of spiking transformers. We subsequently propose inference-time software optimizations for layer skipping, and timestep reduction, leveraging Bayesian Optimization with surrogate modeling to perform robust, efficient co-exploration of the joint algorithmic-microarchitectural design spaces under tight computational budgets. Evaluated on both image(ImageNet) and event-based (CIFAR-10 DVS, DVSGesture) classification, the accelerator achieves up to ~467x and ~1.86x energy reduction compared to edge GPU (Jetson Orin Nano) and previous PIM accelerators for spiking transformers, while maintaining competitive task accuracy on ImageNet dataset. This work enables a new class of intelligent ubiquitous edge AI, built using spiking transformer acceleration for low-power, real-time visual processing at the extreme edge.

Background and Objective: Colorectal cancer prevention relies on early detection of polyps during colonoscopy. Existing public datasets, such as CVC-ClinicDB and Kvasir-SEG, provide valuable benchmarks but are limited by small sample sizes, curated image selection, or lack of real-world artifacts. There remains a need for datasets that capture the complexity of clinical practice, particularly in resource-constrained settings. Methods: We introduce a dataset, BUET Polyp Dataset (BPD), of colonoscopy images collected using Olympus 170 and Pentax i-Scan series endoscopes under routine clinical conditions. The dataset contains images with corresponding expert-annotated binary masks, reflecting diverse challenges such as motion blur, specular highlights, stool artifacts, blood, and low-light frames. Annotations were manually reviewed by clinical experts to ensure quality. To demonstrate baseline performance, we provide benchmark results for classification using VGG16, ResNet50, and InceptionV3, and for segmentation using UNet variants with VGG16, ResNet34, and InceptionV4 backbones. Results: The dataset comprises 1,288 images with polyps from 164 patients with corresponding ground-truth masks and 1,657 polyp-free images from 31 patients. Benchmarking experiments achieved up to 90.8% accuracy for binary classification (VGG16) and a maximum Dice score of 0.64 with InceptionV4-UNet for segmentation. Performance was lower compared to curated datasets, reflecting the real-world difficulty of images with artifacts and variable quality.

Computational spectral imaging (CSI) achieves real-time hyperspectral imaging through co-designed optics and algorithms, but typical CSI methods suffer from a bulky footprint and limited fidelity. Therefore, Spectral Deconvolution imaging (SDI) methods based on PSF engineering have been proposed to achieve high-fidelity compact CSI design recently. However, the composite convolution-integration operations of SDI render the normal-equation coefficient matrix scene-dependent, which hampers the efficient exploitation of imaging priors and poses challenges for accurate reconstruction. To tackle the inherent data-dependent operators in SDI, we introduce a Hierarchical Spatial-Spectral Aggregation Unfolding Framework (HSFAUF). By decomposing subproblems and projecting them into the frequency domain, HSFAUF transforms nonlinear processes into linear mappings, thereby enabling efficient solutions. Furthermore, to integrate spatial-spectral priors during iterative refinement, we propose a Spatial-Frequency Aggregation Transformer (SFAT), which explicitly aggregates information across spatial and frequency domains. By integrating SFAT into HSFAUF, we develop a Transformer-based deep unfolding method, Hierarchical Spatial-Frequency Aggregation Unfolding Transformer (HSFAUT), to solve the inverse problem of SDI. Systematic simulated and real experiments show that HSFAUT surpasses SOTA methods with cheaper memory and computational costs, while exhibiting optimal performance on different SDI systems.

Wireless sensor networks (WSNs) face critical challenges in energy management and network lifetime optimization due to limited battery resources and communication overhead. This study introduces a novel hybrid clustering protocol that integrates the Water Strider Algorithm (WSA) with Fuzzy C-Means (FCM) clustering to achieve superior energy efficiency and network longevity. The proposed WSA-FCM method employs WSA for global optimization of cluster-head positions and FCM for refined node membership assignment with fuzzy boundaries. Through extensive experimentation across networks of 200-800 nodes with 10 independent simulation runs, the method demonstrates significant improvements: First Node Death (FND) delayed by 16.1% ($678\pm12$ vs $584\pm18$ rounds), Last Node Death (LND) extended by 11.9% ($1,262\pm8$ vs $1,128\pm11$ rounds), and 37.4% higher residual energy retention ($5.47\pm0.09$ vs $3.98\pm0.11$ J) compared to state-of-the-art hybrid methods. Intra-cluster distances are reduced by 19.4% with statistical significance (p < 0.001). Theoretical analysis proves convergence guarantees and complexity bounds of $O(n\times c\times T)$, while empirical scalability analysis demonstrates near-linear scaling behaviour. The method outperforms recent hybrid approaches including MOALO-FCM, MSSO-MST, Fuzzy+HHO, and GWO-FCM across all performance metrics with rigorous statistical validation.

This paper presents a high-fidelity evaluation framework for machine learning (ML)-based classification of cyber-attacks and physical faults using electromagnetic transient simulations with digital substation emulation at 4.8 kHz. Twelve ML models, including ensemble algorithms and a multi-layer perceptron (MLP), were trained on labeled time-domain measurements and evaluated in a real-time streaming environment designed for sub-cycle responsiveness. The architecture incorporates a cycle-length smoothing filter and confidence threshold to stabilize decisions. Results show that while several models achieved near-perfect offline accuracies (up to 99.9%), only the MLP sustained robust coverage (98-99%) under streaming, whereas ensembles preserved perfect anomaly precision but abstained frequently (10-49% coverage). These findings demonstrate that offline accuracy alone is an unreliable indicator of field readiness and underscore the need for realistic testing and inference pipelines to ensure dependable classification in inverter-based resources (IBR)-rich networks.

Opinion dynamics has recently been modeled from a game-theoretic perspective, where opinion updates are captured by individuals' cost functions representing their motivations. Conventional formulations aggregate multiple motivations into a single objective, implicitly assuming that these motivations are interchangeable. This paper challenges that assumption and proposes an opinion dynamics model grounded in a multi-objective game framework. In the proposed model, each individual experiences two distinct costs: social pressure from disagreement with others and cognitive dissonance from deviation from the perceived truth. Opinion updates are modeled as Pareto improvements between these two costs. This framework provides a parsimonious explanation for the emergence of pluralistic ignorance, where individuals may agree on something untrue even though they all know the underlying truth. We analytically characterize the model, derive conditions for the emergence and prevalence of the truth, and propose an initial-seeding strategy that ensures consensus on truth. Numerical simulations are conducted on how network density and clustering affect the expression of truth. Both theoretical and numerical results lead to clear and non-trivial sociological insights. For example, no network structure guarantees almost-sure consensus on truth if no one initially expresses the truth; moderately sparse but well-mixed networks are most conducive to consensus on truth.

This paper proposes a novel structure-aware matrix completion framework assisted by radial basis function (RBF) interpolation for near-field radio map construction in extremely large multiple-input multiple-output (XL-MIMO) systems. Unlike the far-field scenario, near-field wavefronts exhibit strong dependencies on both angle and distance due to spherical wave propagation, leading to complicated variations in received signal strength (RSS). To effectively capture the intricate spatial variations structure inherent in near-field environments, a regularized RBF interpolation method is developed to enhance radio map reconstruction accuracy. Leveraging theoretical insights from interpolation error analysis of RBF, an inverse {\mu}-law-inspired nonuniform sampling strategy is introduced to allocate measurements adaptively, emphasizing regions with rapid RSS variations near the transmitter. To further exploit the global low-rank structure in the near-field radio map, we integrate RBF interpolation with nuclear norm minimization (NNM)-based matrix completion. A robust Huberized leave-one-out cross-validation (LOOCV) scheme is then proposed for adaptive selection of the tolerance parameter, facilitating optimal fusion between RBF interpolation and matrix completion. The integration of local variation structure modeling via RBF interpolation and global low-rank structure exploitation via matrix completion yields a structure-aware framework that substantially improves the accuracy of near-field radio map reconstruction. Extensive simulations demonstrate that the proposed approach achieves over 10% improvement in normalized mean squared error (NMSE) compared to standard interpolation and matrix completion methods under varying sampling densities and shadowing conditions.

Enhancing the reliability of AI based fault diagnosis in inverter dominated microgrids requires diverse and statistically balanced datasets. However, the scarcity and imbalance of high fidelity fault data, especially for rare inverter malfunctions and extreme external line faults, limit dependable model training and validation. This paper introduces a unified framework that models a detailed inverter dominated microgrid and systematically generates multiple internal and external fault scenarios to mitigate data scarcity and class imbalance. An enhanced generative model called F2GAN (Feature Feedback GAN) is developed to synthesize high dimensional tabular fault data with improved realism and statistical alignment. Unlike conventional GANs, F2GAN integrates multi level feedback based on mean variance, correlation, and feature matching losses, enabling the generator to refine output distributions toward real fault feature spaces. The generated datasets are evaluated through quantitative and qualitative analyses. Train on Synthetic, Test on Real (TSTR) experiments demonstrate strong generalization of machine learning classifiers trained exclusively on F2GAN samples. The framework is validated on a hardware-in-the-loop (HIL) fault diagnosis platform integrated with a real time simulator and graphical interface, achieving 100 % diagnostic accuracy under real-time testing. Results confirm that F2GAN effectively bridges the gap between simulated and real world microgrid fault datasets

Accurate Channel State Information (CSI) is critical for Hybrid Beamforming (HBF) tasks. However, obtaining high-resolution CSI remains challenging in practical wireless communication systems. To address this issue, we propose to utilize Graph Neural Networks (GNNs) and score-based generative models to enable robust HBF under imperfect CSI conditions. Firstly, we develop the Hybrid Message Graph Attention Network (HMGAT) which updates both node and edge features through node-level and edge-level message passing. Secondly, we design a Bidirectional Encoder Representations from Transformers (BERT)-based Noise Conditional Score Network (NCSN) to learn the distribution of high-resolution CSI, facilitating CSI generation and data augmentation to further improve HMGAT's performance. Finally, we present a Denoising Score Network (DSN) framework and its instantiation, termed DeBERT, which can denoise imperfect CSI under arbitrary channel error levels, thereby facilitating robust HBF. Experiments on DeepMIMO urban datasets demonstrate the proposed models' superior generalization, scalability, and robustness across various HBF tasks with perfect and imperfect CSI.

System identification under unknown external excitation is an inherently ill-posed problem, typically requiring additional knowledge or simplifying assumptions to enable reliable state and parameter estimation. The difficulty of the problem is further amplified in structural systems subjected to natural hazards such as earthquakes or windstorms, where responses are often highly transient, nonlinear, and spatially distributed. To address this challenge, we introduce Spectrum and Physics-Informed Neural Networks (SaPINNs) for efficient input–state–parameter estimation in systems under complex excitations characteristic of natural hazards. The proposed model enhances the neural network with governing physics of the system dynamics and incorporates spectral information of natural hazards by using empirically derived spectra as priors on the unknown excitations. This integration improves inference of unmeasured inputs, system states, and parameters without imposing restrictive assumptions on their dynamics. The performance of the proposed framework is demonstrated through comparative studies on both linear and nonlinear systems under various types of excitation, including the El Centro earthquake, where the seismic spectrum is assumed to be not precisely known. To account for predictive uncertainty, the proposed architecture is embedded within a Deep Ensemble (DEns) networks architecture, providing distributions over possible solutions. The results demonstrate that the proposed approach outperforms conventional PINNs, as the incorporation of spectral information introduces an inductive bias that guides the network more effectively through the solution space and enhances its ability to recover physically consistent state and parameter estimates with realistic uncertainty levels.

Spatial perception is central to auditory intelligence, enabling accurate understanding of real-world acoustic scenes and advancing human-level perception of the world around us. While recent large audio-language models (LALMs) show strong reasoning over complex audios, most operate on monaural inputs and lack the ability to capture spatial cues such as direction, elevation, and distance. We introduce SPUR, a lightweight, plug-in approach that equips LALMs with spatial perception through minimal architectural changes. SPUR consists of: (i) a First-Order Ambisonics (FOA) encoder that maps (W, X, Y, Z) channels to rotation-aware, listener-centric spatial features, integrated into target LALMs via a multimodal adapter; and (ii) SPUR-Set, a spatial QA dataset combining open-source FOA recordings with controlled simulations, emphasizing relative direction, elevation, distance, and overlap for supervised spatial reasoning. Fine-tuning our model on the SPUR-Set consistently improves spatial QA and multi-speaker attribution while preserving general audio understanding. SPUR provides a simple recipe that transforms monaural LALMs into spatially aware models. Extensive ablations validate the effectiveness of our approach.

As large language models transition from text-based interfaces to audio interactions in clinical settings, they might introduce new vulnerabilities through paralinguistic cues in audio. We evaluated these models on 170 clinical cases, each synthesized into speech from 36 distinct voice profiles spanning variations in age, gender, and emotion. Our findings reveal a severe modality bias: surgical recommendations for audio inputs varied by as much as 35% compared to identical text-based inputs, with one model providing 80% fewer recommendations. Further analysis uncovered age disparities of up to 12% between young and elderly voices, which persisted in most models despite chain-of-thought prompting. While explicit reasoning successfully eliminated gender bias, the impact of emotion was not detected due to poor recognition performance. These results demonstrate that audio LLMs are susceptible to making clinical decisions based on a patient's voice characteristics rather than medical evidence, a flaw that risks perpetuating healthcare disparities. We conclude that bias-aware architectures are essential and urgently needed before the clinical deployment of these models.

Traditional statistical optimization-based state estimation (DSSE) algorithms rely on detailed grid parameters and mathematical assumptions of all possible uncertainties. Furthermore, random data missing due to communication failures, congestion, and cyberattacks, makes these methods easily infeasible. Inspired by recent advances in digital twins (DTs), this paper proposes an interactive attention-based DSSE model for robust grid monitoring by integrating three core components: physical entities, virtual modeling, and data fusion. To enable robustness against various data missing in heterogeneous measurements, we first propose physics-informed data augmentation and transfer. Moreover, a state-of-the-art attention-based spatiotemporal feature learning is proposed, followed by a novel cross-interaction feature fusion for robust voltage estimation. A case study in a real-world unbalanced 84-bus distribution system with raw data validates the accuracy and robustness of the proposed DT model in estimating voltage states, with random locational, arbitrary ratios (up to 40% of total measurements) of data missing.

At a wind farm in Croatia, lightning activity is monitored across the entire site using a lightning location system, and on a single wind turbine equipped with a Rogowski-coil-based current measurement system and a high-speed camera, all independently GPS-synchronized. In addition to recording lightning flash currents on the monitored turbine, the system is frequently triggered by electromagnetic disturbances caused by nearby lightning flashes. These include direct flashes to two neighboring turbines that share the same cable connection to the substation and have interconnected grounding systems with buried bare conductor, as well as cloud-to-ground flashes to soil near cable routes, where the resulting electromagnetic fields couple onto the cables, causing surges to propagate to the monitored turbine. The camera occasionally captures failed upward connecting leaders from the monitored turbine during these lightning events. This paper presents three cases of flashes to two neighboring wind turbines and two cases of cloud-to-ground flashes to nearby soil, all of which induced electromagnetic transients that propagated to the monitored turbine. Failed upward connecting leaders were observed in some of these cases. This paper provides observational analysis, providing Rogowski measurements of electromagnetic disturbances and failed leader currents, complemented by high-speed camera and lightning location system data.