Papers

We introduce Latent Secret Spin (LSS), a blind speech watermarking method based on geometric operations in codec latent space. Based upon orthogonal rotations to principal components, LSS induces imperceptible but detectable covariance signatures according to a pseudo-random watermarking schedule. The scheme generalises across datasets, preserves perceptual quality and, unlike some learned, neural watermarking schemes, it does not require neural network training, is resistant to common signal manipulations and is flexible to payload size. Analyses show that structured latent-space watermarking is a promising and interpretable alternative to existing approaches.

Solving optimal control problems via large-scale NLP solvers depends on discretizing continuous dynamics. Yet, this transcription step hides critical vulnerabilities-most notably truncation error and invariant drift-that can drive solvers toward dynamically infeasible or suboptimal trajectories. To expose these hidden failures, we introduce a problem- and transcription-agnostic adversarial objective that leverages the structure of local truncation-error bounds to aggressively amplify such defects. When applied to a 6-DOF rocket-landing problem, we reveal a stark reliability gap: of fourteen transcription methods tested, only three satisfy rigorous validation criteria. These results also expose a striking performance inversion: even in the absence of classical stiffness, a fourth-order implicit scheme (GL2) matches the fidelity of a sixth-order explicit method (RK6). Using B-series expansions and symplectic Runge-Kutta theorems, we isolate the specific truncation errors and quaternion-invariant drift responsible for these failures. Crucially, these theoretical vulnerabilities dictate operational performance: in practical lateral-divert scenarios, the implicit GL2 consistently outperforms the explicit RK6 in both end-to-end solve speed and robustness.

Refining visual representations by eliminating their internal feature-level redundancy is crucial for simultaneously optimizing the performance and computational cost of models in visual tracking. To enhance their performance, many contemporary Transformer-based trackers leverage a larger number of historical template frames to capture richer spatio-temporal cues. However, this strategy leads to a massive number of input visual tokens. This creates two critical issues: it imposes a quadratic computational burden and can also degrade the tracker's overall performance. To bridge this gap, we propose a compress-then-interact tracking framework, ETCTrack, that learns to efficiently compress template tokens from historical template frames into a robust target representation, moving beyond handcrafted rules. Our method first employs the Adaptive Token Compressor to dynamically construct compact yet highly discriminative template tokens by filtering out redundant visual tokens. These refined template tokens are then processed by our Hierarchical Interaction Encoder to achieve a deep, adaptive interaction with the search features. Refined search features ensure subsequent precise target localization. Experiments on seven benchmarks demonstrate that our method outperforms current state-of-the-art trackers. ETCTrack-B224 reduces the number of template tokens by 60%, leading to a 21.4% reduction in MACs with only a 0.4% drop in accuracy. The source code are available at https://github.com/PJD-WJ/ETCTrack.

Many vision datasets now provide segmentation masks in addition to annotated images to support a wide range of tasks. In this work, we propose Class Activation Map Attention Learning (CAMAL), an efficient and scalable method that utilizes segmentation masks to improve attention alignment and faithfulness in vision models. Specifically, attention alignment refers to the degree to which a model's attention aligns with ground-truth discriminative regions, while attention faithfulness refers to the degree to which a model's attention influences its decision. Improving both attention alignment and faithfulness is essential for ensuring that model attention is both spatially accurate and causally meaningful. To improve attention alignment and faithfulness in vision models, CAMAL first extracts the model's attention for each image during training and then compares the attention to ground-truth discriminative regions obtained from the corresponding segmentation masks. CAMAL then acts as an auxiliary regularizer, encouraging attention that aligns with ground-truth discriminative regions, while suppressing attention elsewhere. We evaluated CAMAL across two learning paradigms – Deep Learning (DL) and Deep Reinforcement Learning (DRL) – and observed consistent, significant improvements in both attention alignment and faithfulness. In particular, CAMAL yields statistically significant gains in attention alignment across all settings, and improves attention faithfulness by over 35% compared to recent work. Moreover, we show that improved attention alignment and faithfulness enhance explainability, while yielding improved or comparable generalization performance without increasing inference cost. These findings demonstrate that the spatial information contained within segmentation masks can be effectively leveraged to guide model attention across learning tasks.

Diabetic Retinopathy (DR) is a common complication of diabetes that can lead to blindness of people. Detecting DR at the earliest stage is essential to prevent irreversible eye damage. Microaneurysm dots are the first signs of DR. As the dots are tiny and of low contrast, detecting mild DR is a very challenging task. Federated learning (FL) preserves data privacy, which is a major concern for medical image processing. FL is a collaborative learning method, which shares only the model parameters with a server, without sharing the patient data to a central server. Inspired by classical FL, we propose a federated learning-based quantum neural network (federated QNN) for this task. We implemented the models with limited samples and few learnable parameters from the E-ophtha and Retina MNIST datasets. The crossevaluation efficiency of the proposed federated quantum neural network system for privacy-preserving early detection of diabetic retinopathy (FQPDR) in Kaggle dataset images indicates the robustness of the light weight learning models. FQPDR performances are inspiring while considering existing non-FL and FL methods.

Monocular depth estimation (MDE) with self-supervised training approaches struggles in low-texture areas, where photometric losses may lead to ambiguous depth predictions. To address this, we propose a novel technique that enhances spatial information by applying a distance transform over pre-semantic contours, augmenting discriminative power in low texture regions. Our approach jointly estimates pre-semantic contours, depth and ego-motion. The pre-semantic contours are leveraged to produce new input images, with variance augmented by the distance transform in uniform areas. This approach results in more effective loss functions, enhancing the training process for depth and ego-motion. We demonstrate theoretically that the distance transform is the optimal variance-augmenting technique in this context. Through extensive experiments on KITTI, Cityscapes, Waymo, NYUv2 and ScanNet our model demonstrates robust performance, surpassing competing self-supervised methods in MDE.

Wi-Fi sensing detects human motions and activities by analysing the channel state information (CSI) derived from Wi-Fi transmissions. However, the impact of variable transmission traffic, which dictates the effective sampling rate and interval, is often overlooked. Existing Wi-Fi sensing systems are trained with fixed input size and sampling rate, which suffer from poor sampling rate generalisation. This paper proposes a novel Wi-Fi sensing approach for motion recognition applications, e.g., gesture and activity recognition, under variable traffic patterns. A sampling rate versatile neural network (SRV-NN) based on the transformer is proposed to efficiently handle variable input-sized sensing signals. A dynamic sampling rate augmentation is employed for variable sampling rates and intervals. To validate our approach, we have carried out extensive experimental evaluation, using two self-collected datasets, namely SRV activity and SRV gesture, as well as two publicly available datasets. Our method demonstrated exceptional performance and stability under variable sampling rates, with substantial improvements in average accuracy compared to baseline models without augmentation. The proposed approach significantly enhances stability by greatly reducing accuracy variance across different sampling rates.

Body composition assessment (BCA) provides detailed information about the distribution of different tissue types in the body, enabling more precise characterization of individuals than BMI or weight alone. Consistent and frequent BCA would be valuable for personalized medicine, but the gold standard methods for BCA, such as CT and MRI, are only practical for opportunistic monitoring of patients with clinical indications for imaging and are not suitable for routine use in the general population. Here, we consider an imaging modality which is not currently used in medical applications: millimeter wave (mmWave) radar. Commonly used in security settings, mmWave scans enable fast, non-intrusive, and privacy-preserving reconstruction of full body shape without the need to remove clothing. To demonstrate the feasibility of fast and convenient BCA from mmWave scans, we present a method for BCA value regression using a multi-task learning strategy that leverages synthetic mmWave-like point clouds derived from clinical imaging and parametric human models. We evaluate the model on a pilot cohort of real mmWave scans with bioimpedance-derived body fat measurements, supporting the feasibility of estimating VAT and body fat percentage (BFP) from mmWave data acquired through clothing in a standing posture. We find that the model can predict VAT and BFP with a mean absolute error of 1.0 L and 3.2%, respectively, demonstrating the potential of mmWave scanning for routine BCA in a wide range of settings.

Human Activity Recognition (HAR) from wearable sensors supports broad healthcare and behavior science applications. However, data heterogeneity and the scarcity of labeled data limit its real-world generalization. Recent advances in self-supervised learning (SSL) in vision and language domains have shown strong capability for learning generalizable representations from unlabeled data. Yet, few studies have systematically compared the generalization performance of SSL methods or explored how to adapt them for generalizable HAR. To address these gaps, we present BenchHAR, a unified framework for evaluating the generalization capability of SSL methods for sensor-based HAR on unseen target distributions. BenchHAR curates a large-scale dataset (~258K samples) and evaluates eight representative SSL methods across 12 encoder-classifier architectures. Our results reveal that existing SSL methods struggle to achieve satisfactory generalization performance. We find that: (1) For HAR models, the hybrid paradigm (combining reconstruction and contrastive pretraining) achieves the best overall performance. The CNN encoder exhibits the strongest ability to learn generalizable representations, while more expressive classifier architectures further improve generalization. (2) For data scale, increasing the amount of pretraining data from downstream activity classes consistently improves generalization, while adding more labeled data yields limited gains. Interestingly, incorporating unlabeled data from non-downstream activity classes does not improve generalization. (3) Sensor data collected from custom-grade devices generalizes better than that from research-grade devices, and data from limb transfers more effectively to trunk positions. BenchHAR provides a unified benchmark and actionable insights for generalizable sensor-based HAR systems. Our code is available at https://github.com/saiketa/HAR-Bench.

Purpose: We aim to enhance the image quality of point-of-care ultrasound (POCUS) devices using deep learning and a novel paired dataset of POCUS and high-end ultrasound images. Approach: We collected the first accurately paired dataset using a custom-built automated gantry system of low-end POCUS and high-end ultrasound images. A conditional generative adversarial network (cGAN) was utilized based on the pix2pix architecture, with a U-Net generator that incorporates both L1 and structural similarity index (SSIM) losses to improve perceptual quality. Pretraining on a simulation dataset further boosts performance. Evaluation was performed on 1064 paired ex vivo tissue and phantom ultrasound image sets. Results: Our approach improves the SSIM from 0.29 to 0.54 and PSNR from 19.16 dB to 22.41 dB. No-reference metrics also indicate substantial enhancement, with the Natural Image Quality Evaluator (NIQE) and Perception-based Image Quality Evaluator (PIQE) scores dropping from 7.95 to 4.44 and 31.12 to 19.99, respectively. Conclusions: This work presents the first publicly available accurately paired dataset of low-end POCUS to high end ultrasound images. Additionally, our results demonstrate the potential of the proposed framework to overcome hardware limitations of handheld POCUS, enhancing its diagnostic value in low-resource and point-of-care settings. The POCUS-IQ Dataset is publicly available at https://github.com/NKI-MedTech-AI/POCUS-IQ.

Conventional MRI reconstruction methods treat images and coil sensitivities as discrete objects, leading to high memory demands and limited structural awareness that hamper effective regularization. These limitations hinder accurate reconstruction in highly undersampled scenarios, such as dynamic 3D cardiac magnetic resonance (CMR). We introduce a discretization-free, memory-efficient, model-based framework for dynamic 2D and 3D MRI reconstruction from highly undersampled data. We represent magnetization and coil sensitivities as continuous objects – differentiable functions – using tensor products of univariate neural fields. This tensor product structure enables scalable optimization in high-dimensional spatiotemporal settings. Our method outperforms state-of-the-art model-based reconstructions in dynamic 2D and 3D MR settings, preserving structure and motion even under aggressive undersampling (e.g., acceleration factor 16).

Wireless capsule endoscopy (WCE) enables painless visualization of the gastrointestinal tract, but its diagnostic potential is limited by incomplete mucosal coverage and poor transferability of existing navigation methods across patient anatomies. We propose a transferable, anatomical landmarkguided deep reinforcement learning (AL-DRL) framework for autonomous gastric navigation. Leveraging a lightweight edgecontour-depth fusion module, our policy operates on stable, lowdimensional landmark coordinates rather than high-dimensional video streams, effectively bridging the sim-to-real gap. In simulations across eight patient-derived models, the method achieves over 97% coverage within 50 seconds, significantly outperforming vanilla PPO, SAC, and DQN agents. A two-stage sim-to-real pipeline with an adaptive dynamic programming controller actively mitigates physical disturbances. Ex-vivo experiments demonstrate a mean coverage of 87% and a 53% reduction in procedure time compared with expert manual control.

Recent advances in learned image compression (LIC) have enabled practical deployments, spurring active research into image compression for machines and progressive coding schemes. However, their integration remains under-explored: prior works on progressive machine codec predominantly target sample-level difficulty adaptation (i.e., easy-to-hard), without considering semantic-level scalability. In this work, we introduce a semantic hierarchy-aware progressive codec that enables semantic scalability (i.e., coarse-to-fine) from a single bitstream. We first systematically categorize ImageNet-1K classes into CLIP embedding-based semantic hierarchies. Based on a channel-wise autoregressive framework, we decompose latent representations into hierarchically ordered channel blocks, each explicitly optimized for a corresponding semantic hierarchy. Extensive experiments demonstrate that our approach substantially improves coarse-level recognition at low bitrates while maintaining fine-grained accuracy at higher bitrates. By reframing progressive transmission through the lens of semantic scalability, our work provides an efficient and interpretable solution for task-adaptive image coding, outperforming existing progressive codecs under hierarchical evaluation.

Decoding imagined speech from non-invasive brain recordings is challenging because imagined datasets are scarce and difficult to align temporally across subjects and sessions In this work, we propose a new approach to the decoding of imagined speech that leverages the richer and more reliably labeled recordings during listening to speech. We collected paired listened and imagined MEG recordings to rhythmic melodic and spoken stimuli from trained musicians. Using trained musicians helped improve temporal alignment across conditions. We then developed a three-stage decoding pipeline that revealed consistent and meaningful relationships between neural activity evoked by imagining and listening to the same stimuli. First, we trained six linear and neural models to map imagined MEG responses to listened responses. We evaluated these models against a null baseline from unseen subjects to validate that the predicted-listening responses preserve stimulus-specific information. In the second stage, we trained a contrastive word decoder exclusively on the listened MEG responses, and evaluated it using four embedding strategies including semantic, acoustic, and phonetic representations. In the third stage, we process the imagined MEG responses from held-out subjects through the mapping pipeline to compute the corresponding listening responses that are then decoded by the listened decoder. Using rank-based analysis, we show that the imagined words are decodable significantly above chance. We shall report here the results of a proof-of-concept implementation to decode imagined speech, where all evaluations are performed on held-out subjects. We also demonstrate that performance improves with training data size, suggesting that this approach is scalable and can directly be made applicable to realistic brain-computer interface scenarios.

Recently, the pinching antenna systems (PASS) have attracted significant attention due to their ability to exploit dynamically reconfigurable pinching points along waveguides for flexible signal transmission. However, existing work largely overlooks user mobility although the optimal pinching configuration is highly dependent on the user's location and must be continuously adjusted. In this work, we investigate a PASS-enabled system model in which a base station (BS) serves a mobile user. We formulate an optimization problem that aims to maximize the user's average sum rate over a predefined time horizon while satisfying quality-of-service (QoS) constraint. This objective is achieved by jointly optimizing the beamforming vector at the BS and the pinching locations along the waveguides. Nevertheless, the resulting problem is highly non-convex and challenging to solve using conventional optimization techniques due to the intricate coupling among variables. The difficulty is further exacerbated by environmental randomness arising from user mobility and a probabilistic blockage model. This reveals a key engineering challenge: the performance gains of PASS critically rely on the ability to track or predict user trajectories in real time. To address these challenges, we adopt a deep deterministic policy gradient (DDPG) approach within a reinforcement learning framework, which is well-suited for continuous state and action spaces. Finally, extensive simulations are conducted to validate the proposed approach and demonstrate the importance of real-time configurability.

Building a site-specific propagation model typically requires either ray-tracing over detailed 3D maps or dense measurement campaigns. Both approaches are expensive and often infeasible for rapid deployments where geographic data is unavailable or outdated. We present PropSplat, a map-free propagation modeling method that reconstructs radio frequency (RF) fields using 3D anisotropic Gaussian primitives. Each Gaussian encodes a scalar path loss offset relative to an explicit baseline path loss model with a learnable path loss exponent. Gaussians are initialized along observed transmitter–receiver paths and optimized end-to-end to learn the propagation environment without external information like floor plans, terrain databases, or clutter data. We evaluate PropSplat against wireless radiance field methods NeRF$^2$, GSRF, and WRF-GS+ on two real-world datasets. On large-scale outdoor drive-tests spanning multiple topographical regions at six sub-6 GHz frequencies, PropSplat achieves 5.38 dB RMSE when training measurements are spaced 300m apart and outperforms WRF-GS+ (5.87 dB), GSRF (7.46 dB), and NeRF$^2$ (14.76 dB). On indoor Bluetooth Low Energy measurements, PropSplat achieves 0.19m mean localization error, an order of magnitude better than NeRF$^2$ (1.84m), while achieving near-identical received signal strength prediction accuracy. These results show that accurate site-specific propagation reconstruction is achievable from sparse RF-native measurements. The need for geographic data as a prerequisite for scalable RF environment modeling is reduced.

Traffic state estimation from sparse fixed sensors is challenging because physics-informed neural networks (PINNs) tend to over-smooth sharp transitions admitted by the Lighthill-Whitham–Richards (LWR) model. This study proposes Two-Stage Domain Decomposition Physics-Informed Neural Networks (TSDD-PINN), an observation-aligned framework for LWR-based offline speed-field reconstruction. The framework supports spatial, temporal, and space–time refinement. Matched direction analysis shows that spatial refinement has the lowest mean error and less than half the training time of space–time refinement in the tested setting, while temporal refinement is faster. A global parent PINN is first trained. In the controlled spatial implementation, its residual profile guides a deterministic partition for warm-started child networks. An optional operational safeguard retains Stage~1 when the prespecified screen does not activate. The primary I-24 MOTION evaluation spans five days, five sensor configurations, and ten seeds per configuration, yielding 1{,}500 runs. Controlled TSDD-PINN attains the lowest relative $L_2$ error in 18 of 25 configurations and 14 of 15 sparse-sensing cases, while training 2.4 times faster than the extended PINN (XPINN) baseline under the evaluated implementations and training budgets. Non-neural comparisons show that the advantage over interpolation is concentrated under sparse sensing, whereas dense sensing often favors interpolation. A separate 250-run operational evaluation finds infrequent activation and motivates the Stage-1-preserving safeguard. The residual is treated as an indicator of model difficulty rather than a validated shock detector. The evidence supports a sensing-density-dependent operating range rather than uniform improvement.

Safe operation of connected vehicle platoons under stochastic disturbances and time-delayed dynamics requires accurate quantification of rare but dangerous events, such as inter-vehicle collisions. We propose a rigorous framework for quantifying the risk of inter-vehicle collisions in connected vehicle platoons subject to time-delayed stochastic dynamics. We adopt the entropic value-at-risk (EVaR) as a conservative metric to capture risk due to extreme events, highlighting its advantages over conventional Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR). By expressing the inter-vehicle distance covariance in terms of the Laplacian eigenvalues of the communication network, we derive network-and time-delay-induced bounds on both the minimum inherent risk and the worst-case risk. Specifically, the algebraic connectivity dictates the maximum EVaR, while the largest Laplacian eigenvalue determines the minimum risk inherently induced by the network structure. Numerical simulations illustrate how network topology and time delay shape collision risk, offering actionable insights for the safe design of vehicle platoons operating under stochastic disturbances.

This paper provides an in-depth analysis on how different aspects of the dynamic operating envelope (DOE) formulation impact the computation and allocation of network capacity. We show that the envelopes are significantly affected by the power flow model (non-linear or linear), binding network constraint (thermal or voltage) and by the calculation case (import or export envelope). We also propose a novel DOE algorithm (LACE) that presents transparent and scalable computation that is useful for larger networks or to act in tandem with other optimization engines. We run numerical simulations with different test feeders, including a realistic low-voltage feeder with real-world data from Belgium. This paper provides crucial insights and tools to distribution system operators (DSOs), stakeholders and academics alike to make sure DOE calculation achieves desirable and efficient outcome.

Atlas-based approaches allow high-quality, patient-specific shape reconstructions of cardiac anatomy from sparse and/or noisy data such as point clouds. However, these methods are mainly prior-driven, so the impact of uncertainty can be large, limiting their clinical reliability. We propose a probabilistic framework for uncertainty-aware cardiac shape reconstruction that combines Deep Signed Distance Functions (DeepSDFs) with Markov Chain Monte Carlo (MCMC) sampling. Cardiac geometries are modeled implicitly as zero-level sets of a neural network conditioned on learned latent codes, enabling multi-surface reconstruction of the left and right ventricles. By interpreting the reconstruction loss as a log-likelihood, we perform Bayesian inference in the latent space to obtain both maximum a posteriori (MAP) and posterior-sampled reconstructions. Experiments on a public cardiac dataset show that our approach produces accurate reconstructions and well-calibrated uncertainty estimates.

This work introduces a fully tunable, ultra-low power unipolar memory cell inspired by the Schmitt-trigger comparator and designed in CMOS using only nine transistors. The proposed circuit operates entirely in the current domain and exploits a novel feedback configuration between two interdependent Heaviside-like thresholding elements to produce tunable bistable switching behavior. Its three key parameters-threshold current, hysteresis width, and output gain-are independently tunable via programmable bias currents, enabling flexibility across diverse analog computing applications. Unlike prior Schmitt-trigger designs, it simultaneously achieves current-mode operation, nanowatt-range power consumption, temperature stability, and full tunability, solely using standard MOSFET elements. Schematic-level simulations in a 180 nm CMOS process confirm robust hysteresis and resilience to device mismatch. Building on this circuit, we develop a complete family of spike-based logic gates using three-level current encoding, where the bistable memory retains the polarity of the last spike on each input indefinitely, enabling asynchronous logic operations without temporal windowing or refresh mechanisms. The same circuit also serves as the primitive for Bistable Memory Recurrent Units in analog neural networks, where the quantized hidden states provide inherent noise immunity. Together, these capabilities position the design as a versatile building block for next-generation neuromorphic processors integrating memory, logic, and recurrent computation.

In power systems, the risk of wildfire ignition has increased significantly in recent years. The impact and severity of these events on energy dispatch, as well as their societal ramifications, make wildfire prevention critical for power system planning and operation. A common intervention by system operators is to de-energize transmission lines to mitigate the risk of fire caused by equipment failures. With the growing integration of variable renewable generation, managing and preparing the system to de-energization under wildfire risk has become even more challenging. In this context, mitigation decisions such as installing battery energy storage systems and undergrounding transmission lines can reduce the risk and adverse effects associated with de-energization and renewable generation variability. This paper presents a robust optimization model to determine the optimal location of battery storage and undergrounding of transmission line investment, utilizing representative weeks and uncertainty sets to capture the temporal relationship of uncertain variables. Specifically, this paper addresses: (i) the worst-case realization of ignition risk leading to the de-energization of transmission lines, combined with the worst-case realization of renewable energy availability, and (ii) the optimal investment decisions for energy storage capacity and undergrounding of transmission lines that are exposed to ignition risk. The proposed model is formulated as a mixed-integer linear programming (MILP) problem, employing duality theory and binary decomposition to address nonlinearities, and is solved using a column-and-constraint generation algorithm. The proposed framework is evaluated on a model of the San Diego power system, demonstrating its practical effectiveness in improving the resilience to wildfire risk.

Sampling-based model predictive control (MPC) algorithms, such as model predictive path integral (MPPI), enable approximate, gradient-free solutions to optimal control problems by drawing samples from a proposal distribution, evaluating their trajectory costs, and updating the proposal parameters accordingly. However, these approaches typically rely on heuristics for adjusting hyperparameters, such as temperature or momentum, or manual tuning. We propose a trust region formulation for sampling-based MPC that constrains updates of the proposal distribution via a principled Kullback–Leibler (KL) divergence bound and, optionally, an entropy lower bound. This replaces heuristic hyperparameter adaptation with values that are optimal w.r.t. the underlying Lagrangian. We further improve sample efficiency and convergence by combining the trust region update with deterministic localized cumulative distribution (LCD)-based sampling. Experiments on two benchmark environments demonstrate that the proposed trust region update achieves faster convergence and better sample efficiency in low-sample and low-iteration regimes, especially when paired with deterministic LCD-based sampling.

We investigate interactive trajectory planning subject to uncertainty in the decisions of surrounding agents. To control the ego-agent, we aim to first learn the decision distribution and solve a Stochastic Model Predictive Control (SMPC) problem. To account for errors in the learned distribution, we show that it is possible to utilize Probably Approximately Correct (PAC) learning in combination with Distributionally Robust (DR) optimization to obtain a solution which accounts for the errors induced by the learning model. The results indicate that our PAC learning-based DR-MPC framework provides a method to interpolate between a robust MPC and an omnipotent SMPC, based on the available number of samples.

Dense low-Earth-orbit (LEO) networks increasingly reuse spectrum across moving satellites and overlapping beams, making a spatial radio map essential for interference diagnosis and coordination. Constructing such a map from sparse ground power measurements is difficult because the active satellite subset and its cardinality are unknown, candidate footprints can be highly coherent, and beam pointing, width, effective power, Doppler, and background must be inferred jointly. Known ephemerides do more than enumerate visible candidates. Propagated positions determine candidate-specific ranges, viewing geometry, and spatial footprints, while radial velocities induce distinct time-frequency signatures. Their joint spatial–Doppler structure rejects candidates whose geometry or time–frequency signature conflicts with the observations before continuous beam inference. SateBeam therefore screens all candidates through Doppler-resolved nonnegative fitting. GPU-accelerated grid correlations initialize continuous off-grid refinement; a Gamma-likelihood Bayesian information criterion (BIC) selects model cardinality, and backward deletion and exchange repair greedy support errors. Across paired real-ephemeris scenarios, SateBeam recovers the exact support in every case and reduces independent-query RMSE from $0.655$ to $0.050$ dB relative to orthogonal matching pursuit, with consistent advantages under model mismatch and larger candidate sets. The code is available at https://github.com/UNIC-Lab/SateBeam.

This paper proposes and experimentally validates a two-stage scheduling and control strategy for a behind-the-meter battery energy storage system (BESS) delivering both local and grid services. Considered services are the maximization of PV self-consumption, peak-load reduction, and secondary frequency control (aFRR).The day-ahead stage allocates battery capacity across local and balancing services using a scenario based approach, reflecting potential remuneration from aFRR participation without committing to fixed power availability; in the real-time stage, BESS set-points are computed in a periodic fashion at a high time resolution based on updated information on balancing prices, net load realization and BESS state of charge. The strategy is experimentally validated on a building at the Energypolis Campus of HES-SO Valais (Sion, Switzerland), which exhibits a peak power demand of 300 kW and is equipped with a 264 kWh / 140 kW lithium-ion BESS. The experimental results demonstrate the effectiveness of the proposed framework in scheduling and actuating the provision of both behind-the-meter and front-of-the-meter services.

This paper introduces a novel control strategy to optimize urban network traffic in mixed autonomy settings, featuring Connected and Automated Vehicles (CAVs) alongside Human-Driven Vehicles (HDVs). Unlike previous control strategies, where the impact of driver behaviour of CAVs and HDVs is not explicitly considered, we propose a dynamic, queue-responsive saturation rate to account for autonomy-driven variations in traffic flow characteristics. The proposed method is based on an extended multi-commodity store-and-forward model to a mixed autonomy environment, integrating optimized routing for CAVs via infrastructure-linked connectivity, and signal timings at every signalized intersection. The problem is formulated as a Non-Convex Quadratic Program (NQP), which accounts for queue evolution, spillback, green time allocation, and CAVs routing. To enable computational efficiency for real-time applications, we transform the NQP into a sequence of convex subproblems, leveraging under- and over-estimators to reformulate it as a Mixed Integer Linear Program (MILP). Experimental results via microscopic simulations validate the efficiency and robustness of the proposed methodology. The results reflect that the proposed model outperforms the existing multi-commodity approach, thus demonstrating its potential for real-time traffic optimization in future urban mobility systems.

This study investigates the performance of cascade PID control architecture applied to an inverted pendulum on a cart system through both simulation and experimental implementation. A nonlinear model of the system was developed using Simscape Multibody in Simulink, while a physical prototype was constructed using a DC motor-driven cart, pendulum, rotary encoder, ultrasonic sensor, and an Arduino. The cascade PID control structure consists of an inner loop regulating the pendulum angle and an outer loop controlling the cart position. Simulation results demonstrated effective stabilization of the pendulum and satisfactory position tracking under idealized conditions. Experimental results confirmed successful real-time stabilization but revealed notable differences from simulation, particularly in controller gains, transient behavior, and disturbance response due to sensor noise, unmodeled friction, and implementation constraints. The study also highlights the limitations of cascade PID control in disturbance rejection and large position commands, particularly under limited track length. A comparative analysis using an LQR-based inner loop demonstrated better disturbance rejection and reduced overshoot. The results provide practical insights into the applicability and limitations of cascade PID control of the inverted pendulum system.

Hardware acceleration has emerged as a key research topic for supporting computationally intensive signal processing and artificial intelligence applications in 6G research and development studies. This paper presents an RF Network on Chip (RFNoC) based hardware acceleration framework that offloads key physical layer procedures to a field programmable gate array (FPGA). The proposed design accelerates procedures, including low density parity check codes (LDPC) encoding and decoding, rate matching and unmatching, interleaving and deinterleaving, scrambling and descrambling, and log likelihood ratio estimation. The accelerator is integrated directly into the OpenAirInterface radio access network software, enabling simultaneous use of the FPGA as driver of the radio front end and a high throughput accelerator. The proposed system is validated through real time experiments with a commercial smartphone successfully connecting to the network. The implementation results demonstrate that a throughput of about 900 Mbps is achiievable using a moderate FPGA resource utlization.

A unified framework for the characterization of continuous electromagnetic (EM) manifolds for arbitrary multipleinput multiple-output (MIMO) system geometries is presented. The EM manifold refers to the set of all physically realizable radiated field vectors, parameterized by the array excitation, that encodes the full spatial structure of the antenna system including near-field phase variations, polarization, and mutual coupling. Building upon the discrete moment-matrix formulation, the proposed framework addresses three fundamental limitations simultaneously: (i) point-source near-field modeling errors in the radiation operator; (ii) confinement of the beamforming space to the $N$-dimensional subspace dictated by hardware port count; and (iii) restriction to linear (1D) array geometries. Each mesh element is modeled as a two-dimensional (2D) planar patch, whose spatially averaged Green's function is evaluated via Gauss-Legendre (GL) quadrature, yielding superior nearfield accuracy at negligible additional cost. A continuous feeding function $w(\mathbf{p})\in L^2(\mathcal{S}_\mathrm{T})$ is introduced as the infinite-dimensional limit of the $N$-port network, enabling optimization over a higher dimensional current subspace, decoupled from hardware constraints. Full-wave MATLAB Antenna Toolbox validation confirms near-field accuracy improvements over the state-of-the-art (SotA) baseline for both linear and planar array geometries, while maintaining reasonable computational complexity.