Papers

State departments of transportation (DOTs) in the United States increasingly rely on statewide continuously operating reference station (CORS) networks to support high-precision Global Navigation Satellite System (GNSS)-based positioning and timing for intelligent transportation systems. These networks also provide continuous observations that can support regional GNSS integrity monitoring. This study develops and demonstrates a framework that treats a statewide CORS network as a spatially distributed sensor system for identifying unintentional (environmental) and intentional (cyber) interference when GNSS measurements deviate from expected spatial patterns. We develop a graph-based Network Consistency Framework (NCF) that evaluates each station against its spatial neighborhood using four metrics: neighborhood residual, spatial gradient, residual, and graph smoothness. These metrics are combined into a Network Consistency Index (NCI). The framework is demonstrated using two consecutive days of four-constellation observations from 50 stations in the Alabama DOT-maintained CORS network, using changes in vertical total electron content ({\Delta}VTEC) and the Rate of TEC Index (ROTI) as spatially coherent observables. The framework quantified network-wide spatial consistency and identified localized anomalies. Detected anomalies indicate stations whose observations deviated from the surrounding regional network, signaling potential integrity issues. Determining whether anomalies result from receiver faults, localized interference, spoofing, or other causes requires further investigation. This study introduces statewide CORS networks as regional GNSS integrity observatories and presents the NCF and NCI for graph-based spatial integrity monitoring. Transportation agencies can implement the framework using existing CORS observations to monitor network integrity and identify localized anomalies.

With advancements in long-distance wireless power transfer (WPT) and space-based energy technologies, integrating WPT into non-terrestrial networks (NTNs), referred to as NTN-WPT, is emerging as a promising approach for next-generation wireless networks. This paper proposes an energy-scheduling approach that jointly optimizes energy efficiency, task completion rate, and task waiting time for power transfer from low Earth orbit satellites to terrestrial mobile user devices (UDs). To address scheduling challenges caused by satellite and UD mobility and channel uncertainty from stochastic propagation effects, we decompose the problem into three subproblems within a three-layer predictive framework: 1) a state prediction layer forecasts UD and satellite states; 2) an interaction mapping layer uses a graph neural network (GNN) to model energy transfer efficiency; and 3) a decision-making layer determines the energy allocation plan. Distinct machine learning (ML) methods are tailored to each layer. To balance the competing objectives, we adopt a multi-objective reinforcement learning (MORL) technique that scalarizes them into a weighted-sum reward, transforming the multi-objective problem into a tractable single-objective problem. We further introduce a multi-agent deep learning model integrating self-attention with multi-agent proximal policy optimization (MAPPO) to improve objective balancing. Simulation results show that the proposed approach achieves a better overall trade-off than baseline methods, maintaining competitive task completion rates and energy efficiency while reducing task waiting times, and remains robust under highly variable conditions.

Koopman theory offers a linear-operator view of nonlinear sequence dynamics by lifting observations into a space where evolution is governed by a linear time-invariant Koopman operator. While the Koopman operator provides a linear representation of nonlinear dynamics, it is generally infinite dimensional and defined under time-invariant assumptions. To model non-stationary signals with frequency-dependent behavior, a frequency-varying extension is required. In recent years, deep learning has been increasingly employed to exploit its powerful function-approximation ability for learning the Koopman operator. In this study, we propose a novel approach called neural decomposition Koopman (NDKoop), an end-to-end architecture that integrates a learnable signal decomposition module with both frequency-independent and frequency-dependent Koopman based networks for sequence forecasting. To the best of our knowledge, this is the first work to jointly realize end-to end Koopman modeling and signal decomposition within a unified neural framework. We demonstrate that decomposing a signal into a frequency-independent trend component and a frequency-dependent periodic component, each governed by a corresponding Koopman operator, improves prediction accuracy when perfect linearization is unattainable. Numerical experiments across several forecasting benchmarks indicate that the proposed NDKoop provides strong performance.

Hybrid music generators combine the long-range planning of an autoregressive language model with the fidelity of a diffusion- or flow-based acoustic renderer. Yet renderers are trained with clean, target-derived codec tokens but deployed with imperfect language-model predictions, creating codecinterface exposure bias. Rather than treating rendering as a simple reconstruction task,we formulate it as full-context generation from an imperfect discrete plan. We introduce FullDiT, a conditional DiT that fuses eight frame-aligned RVQ streams with independently encoded captions and lyrics and uses non-causal self-attention over the complete acoustic latent sequence. During training, Error-Matched Distractor Conditioning (EMDC) matches per-codebook replacement rates to teacher-forced top-1 error rates and samples near-miss tokens from cosine-KNN neighborhoods without changing the acoustic target. At inference, four-way classifier-free guidance (4-CFG) independently scales codec, lyric, and caption guidance increments. Matched ablations show that EMDC improves ViSQOL by 0.77 under synthetic corruption and is clearly preferred in non-tied comparisons with fixed languagemodel tokens. Further ablations show gains from full-song context and renderer-side text conditioning. The complete system outperforms five commercial systems on 15 of 18 automatic metrics and ranks among the top three on the Artificial Analysis Music with Vocals Leaderboard. The demo page is available at https://selinacloudl.github.io/fulldit-demo/.

Electrical faults in power transmission systems can severely affect grid stability, equipment safety, and operational reliability. Traditional protection schemes, particularly distance relays, depend on apparent impedance computation and may suffer from relay overreach, underreach, or maloperation due to CT/PT saturation and high-impedance conditions. This paper proposes an intelligent fault detection and classification framework based on supervised machine learning. The approach learns nonlinear relationships between three-phase voltage/current patterns and fault types without assuming fixed impedance paths. A derived feature set is used to represent six fault categories. Artificial Neural Networks, Support Vector Machines, Random Forests, XGBoost, Long Short-Term Memory networks, and Physics-Informed Neural Networks (PINNs) are developed using SMOTE-balanced datasets. Robustness is evaluated under different training sizes and Gaussian noise levels. The PINN achieved the highest fault detection accuracy of 99.86% and multiclass classification accuracy of 99.79% on the clean dataset, while maintaining high accuracy under 2-5% noise and 1-60% training data. By embedding power-system equations and providing millisecond-level inference, the proposed framework bridges traditional impedance-based protection and scalable data-driven grid analytics for real-time protection and wide-area monitoring and control.

Emotion is central to human communication, and its expression varies across languages. Large audio-language models (LALMs) achieve strong performance on multilingual speech tasks, yet it remains unclear whether they encode emotion through language-specific correlations or language-agnostic representations. We present the first neuron-level interpretability study of this question. We define Multilingual Emotion Neurons (MLENs) as functional units exhibiting stable emotional selectivity and aligned causal effects across languages, and introduce Consistency-Regularized Fusion (CR-Fusion) to identify them. Across four modern LALMs and 12 typologically diverse languages, emotion-sensitive neurons identified independently per language show minimal overlap, and additional monolingual identification data saturates quickly without isolating more transferable units, motivating identification from pooled cross-lingual evidence. Causal interventions demonstrate that MLENs identified by CR-Fusion provide more precise and transferable affective control than monolingual neuron sets in both zero-shot and low-resource settings. Leave-one-out ablations further reveal asymmetric transfer: individual identification languages, including low-resource ones, contribute non-redundant evidence, while several low-resource languages benefit most from the resulting cross-lingual transfer. Together, our findings provide the first causal, neuron-level account of how LALMs encode emotion across languages, and establish multilingual neuron identification as an effective mechanism for understanding cross-lingual affective behavior.

Controllable music retrieval lets users find music that is, for example, more ambient, less distorted, or without guitar while preserving the other semantic content of an original seed query. Sparse autoencoders (SAEs) are a promising interface for this kind of concept-level control, but a key problem remains: given a free-form text concept, which sparse features should be edited? In shared multimodal embedding spaces, standard attribution methods often select neurons that match the concept's wording but not the audio examples that express it. This leads to weak or unstable edits: relevant features are missed when concepts are distributed across neurons, while others are selected due to text alignment rather than audio-side structure. We address this with a lightweight, training-free method that recovers a sparse set of audio features whose decoded representation reconstructs the target concept while remaining consistent with audio-space geometry. This reframes concept attribution as a sparse inversion problem rather than a text-side neuron-ranking heuristic. The method requires neither paired audio-text supervision nor SAE retraining. We evaluate this approach in steerable music retrieval and show that the recovered supports align more closely with concept-bearing audio examples and achieve a stronger trade-off between edit strength and preservation than alignment baselines, enabling more precise concept amplification and suppression with reduced drift on preservation metrics.

Connection graphs (CGs) extend classical graphs by associating vector-valued signals to nodes and orthogonal transport maps across edges, making them a natural model for synchronization and manifold-based signal processing. Despite their growing use, learning CGs directly from observations remains challenging because the network topology and the underlying geometric structure are coupled through non-Euclidean orthogonality constraints. In this work, we address this inverse problem by learning a consistent connection graph from noisy vector-valued signals. Exploiting the spectral characterization of consistent CGs, we formulate a structured learning problem that jointly estimates a denoised signal, the graph topology, and node-wise local reference frames. The proposed formulation couples the spectrum of the learned connection Laplacian to that of an underlying combinatorial Laplacian, enabling explicit spectral and topological priors while guaranteeing a nontrivial global-section space. We develop Structured Connection Graph Learning (SCGL), a block-coordinate algorithm that combines closed-form updates, manifold projections, and spectral constraints, and converges to stationary points of the resulting nonconvex problem. Numerical experiments show that SCGL improves topology and geometry recovery over competing approaches, while also yielding effective denoising and signal-compression bases.

Every audio generative system makes two coupled decisions: what representation to generate, and how to model its distribution. This paper organizes audio generative modeling around this coupling. For representation design, we compare discrete, continuous, and hybrid latents through four objectives: representation burden, distortion, empirical modelability, and streaming compatibility. For distribution modeling, rather than treating a latent's difficulty as an intrinsic scalar, we use two diagnostic dimensions: dependency horizon, how far useful context extends, and conditional ambiguity, how much uncertainty remains after conditioning. These dimensions refine the common semantic-versus-acoustic intuition: variables with a long dependency horizon should receive global modeling capacity. Conditionally ambiguous detail may be delegated to a local or iterative generator. Applied to representative systems, this view shows that RVQ's residual order gives ordered capacity but not ordered semantics, that AudioLM's semantic-versus-acoustic cascade is one explicit placement of this boundary rather than a universal template, and that autoregression, iterative refinement, and hybrid designs differ chiefly in how they trade dependency horizon against critical-path generation cost. The distinction between discrete and continuous latents describes the output interface; dependency horizon, conditional ambiguity, and streaming determine how that interface should be modeled. Rather than cataloguing individual systems, we provide an evaluation and design framework for comparing representation-model pairs.

A low-Earth-orbit satellite has only minutes to downlink imagery on a limited power budget, and some moments in a pass carry far better channels than others. Spending more power on the good moments and skipping hopeless ones should deliver more images, but the channel cannot be measured in time, so the schedule is planned in advance from weather statistics using a safety margin. That margin is sized so the forecast is statistically reliable, and the scheduler is judged by whether it beats uniform power on average. Both targets are wrong for a satellite that gets one attempt per pass: on real Sentinel-2 imagery, a scheduler gaining one to two images per pass on average still delivers fewer than uniform power on a quarter to a third of individual passes, and a genie knowing the weather exactly does no better. We instead size the margin by measuring, on recorded passes, how often it would have backfired, keeping only settings that provably backfire rarely. The procedure declines to schedule our learned codec and certifies scheduling for codecs with a sharper cliff.

Digital-twin calibration requires interaction data that is expensive to collect. We study two acquisition decisions: which trajectories to generate, and when to spend a limited budget on privileged parameter measurements. Our framework couples an excitation-oriented reinforcement learning controller, a recurrent parameter estimator with predictive uncertainty, and a budgeted query policy. In Pendulum, a Random Forest diagnostic recovers gravity only weakly from task-oriented trajectories and does not recover mass or length, while a GRU trained on excitation-oriented trajectories reaches a mean absolute error of 0.0066 with no queries. We then withdraw continuous oracle access partway through an episode, so that the twin must run on the estimator's output for the remainder. The estimator-plus-policy pipeline achieves a terminal error of 0.0092 under a three-query budget, against 0.2031 for an uncalibrated twin. In partially observable Waterworld, five controllers produce different observed error profiles across three hidden parameters, and an estimator trained on a five-controller mixture reaches online normalized errors of roughly 4-5%. These exploratory case studies are not controlled ablations, but they motivate treating trajectory design and query allocation as explicit design variables in data-scarce calibration.

Electromagnetically reconfigurable fluid antenna system (ER-FAS) provides additional electromagnetic (EM)-domain degrees of freedom by enabling dynamic control of per-element radiation patterns, thereby improving power efficiency in wireless communications. The effective exploitation of these capabilities, however, critically depends on accurate channel estimation, which has received limited attention in prior studies. This paper presents a low-complexity channel parameter estimation framework for downlink wideband systems that leverages available channel sparsity. A synthesis-based reconfigurability model is considered, where each antenna generates desired beampatterns using a finite set of EM basis functions. Based on this model, a joint optimization of digital and EM precoders is formulated with the objective of minimizing the Cramér-Rao lower bound of the channel parameters. Simulation results demonstrate that the proposed hybrid beamforming and estimation approach enables accurate recovery of channel parameters while maintaining low computational complexity.

Power amplifier (PA) nonlinearity and memory effects significantly limit the spectral compliance, reliability, and energy efficiency of communication systems. To address this, we propose a transfer-learning-enabled, fully digital transceiver-cooperative method for amplitude-phase-time block modulation (APTBM)-based nonlinear single-carrier transmission under adjacent channel leakage ratio (ACLR) constraints. At the transmitter, iterative clipping and filtering (ICAF) and static digital pre-distortion (SDPD) act jointly to reduce signal peaks and suppress spectral regrowth without requiring wideband feedback. At the receiver, the inherent amplitude-phase constraints of APTBM provide weakly supervised prior knowledge for offline inverse-model pretraining, which is followed by the online few-shot adaptation of a lightweight digital post-distortion (DPoD) network. Subsequently, a cascaded DPoD and clipping-noise cancellation scheme systematically compensates for residual distortions induced by both the PA and ICAF. Simulation and measurement results demonstrate reliable transmission at an input back-off of approximately 2 dB under a 30-dBc ACLR constraint. Furthermore, the proposed DPoD approach significantly reduces online training time and computational overhead, delivering a performance gain of over 2 dB compared to conventional learning-based DPoD schemes.

Accurate phasor estimation during power system faults is challenging because fault currents contain decaying DC offsets, harmonics, noise, and possible frequency deviations. These distortions can significantly degrade conventional discrete Fourier transform-based estimators, especially during the first cycle after fault inception. This paper presents SymbolicPhasor, a dynamic deep symbolic regression framework for estimating the fundamental component of distorted fault current signals. The method processes the signal through overlapping moving windows, learns interpretable analytical expressions for the full waveform within each window, and then projects the reconstructed signal onto nominal sine and cosine bases to recover the instantaneous fundamental magnitude and phase. By embedding symbolic tokens corresponding to the nominal, third-, and fifth-order harmonic frequencies, the proposed approach is guided toward physically meaningful expressions while preserving data-driven flexibility. The method is evaluated under single decaying-DC, multiple decaying-DC, and off-nominal frequency conditions. Results show consistently high reconstruction accuracy, with coefficient of determination values reaching 0.985, demonstrating that the proposed framework can recover the fundamental component within one cycle for practical protective relaying and measurement applications.

Approximate message passing (AMP)-type algorithms are widely used for signal recovery in high-dimensional noisy linear systems. Recently, a framework called memory AMP (MAMP) was introduced, offering a new approach to incorporating memory terms within AMP algorithms. Building on this, a low-complexity gradient descent MAMP (GD-MAMP) was proposed for right-unitarily invariant matrices. In this paper, we first address an overflow problem in GD-MAMP caused by intermediate variables exceeding the floating-point range, which typically occurs when the condition number is large. Second, we propose two low-complexity variants of GD-MAMP: one replaces full-length memory with partial memory, while the other reduces the number of matrix-vector products per iteration by $1/3$ (from three to two). Neither degrades the convergence speed notably. Third, we develop a general gradient-based formulation for designing MAMP algorithms. This formulation recovers warm-started conjugate gradient VAMP (WS-CG-VAMP) as a special case. Furthermore, we show that the computation of the orthogonalization parameters in this formulation can suffer from catastrophic cancellation, which explains the finite-precision instability of WS-CG-VAMP. Finally, we derive an equivalent reformulation, termed WS-CG-VAMP(r), which reduces the number of matrix-vector products by up to $50\%$. Measured by matrix-vector products, GD-MAMP converges faster for small condition numbers, whereas WS-CG-VAMP(r) converges faster for large ones under high-precision arithmetic but may diverge in IEEE double precision due to catastrophic cancellation.

Hybrid analog-digital beamforming (HBF) has emerged as a key enabling technology for non-terrestrial networks (NTNs), where large antenna arrays are required to compensate for severe propagation loss but fully digital beamforming is often impractical due to radio-frequency (RF) chain cost, power consumption, and payload limitations. Compared with terrestrial networks, NTN platforms such as low Earth orbit (LEO) satellites and unmanned aerial vehicles (UAVs) impose distinctive HBF design challenges, including high mobility, Doppler effects, sparse line-of-sight-dominant channels, stringent on-board energy budgets, and, for UAVs, the additional coupling between beamforming and controllable platform placement or trajectory. This survey provides a systematic review of HBF techniques for NTN systems, with emphasis on LEO satellite and UAV communications. We first introduce common HBF architectures, signal models, channel representations, analog and digital precoder designs, and learning-aided approaches that form the shared technical foundation of existing works. We then survey both platforms under a common set of five categories, which cover system architecture and precoding design, time-varying beam management, network-level cooperation and scheduling, sensing capability and reconfigurable surfaces, and security and multiple access. Their platform-specific content differs most sharply in the second one, since the dominant time-varying mechanism is traffic-driven beam hopping on an LEO payload but mobility-aware beam tracking on a UAV. Finally, we discuss open research challenges and future directions toward scalable, robust, and hardware-efficient HBF in next-generation NTNs.

Pretrained neural speech codecs typically use a fixed residual vector quantization (RVQ) depth for all frames, ignoring temporal variation in quantization difficulty. We propose BAMU, a bitstream-aware dynamic RVQ allocation framework for frozen pretrained codecs. A lightweight, rate-independent predictor estimates frame- and layer-wise marginal latent-distortion reductions, while a constrained allocator selects prefix-valid depths under an exact serialized-size budget. Experiments on EnCodec and DAC over LibriSpeech, together with VCTK evaluation, show consistent EnCodec gains and DAC improvements mainly at medium and high rates. A 30-listener study confirms a MOS improvement from 3.449 to 3.780 over matched fixed-depth coding.

Motivated by the IEEE 802.11bf effort to standardize advanced WLAN sensing, interest in Wi-Fi Channel State Information (CSI) for passive, device-free, and privacy-preserving activity and gesture recognition has grown rapidly. Recent studies have shown that Doppler velocity projections extracted from CSI, which directly reflect human-motion velocity, enable more robust human activity recognition (HAR) and stronger generalization across users and unseen conditions. Nevertheless, reliable generalization under real-world variability remains a major challenge that hinders the adoption of Wi-Fi sensing. To address this challenge, we build on Doppler Radiance Fields (DoRF), adapting the concept of neural radiance fields (NeRF) from computer vision to Wi-Fi sensing. DoRF treats the extracted Doppler velocity projections as sparse virtual-camera views of human motion. From these views it recovers a regularized latent 3-D motion descriptor whose projections along learned effective Doppler directions explain the CSI observations. The recovered motion is then re-projected onto an equiangular grid of directions on the unit sphere, yielding a spherical representation of the underlying activity. Because DoRF produces a signal defined on the sphere, this paper introduces DoRF++, a spherical representation-learning model that applies quadrature-aware spherical Transformers directly to the spherical field for activity classification. Experiments on our collected hand-gesture dataset show that DoRF++ noticeably outperforms state-of-the-art Wi-Fi-based HAR methods in cross-user generalization accuracy, especially for difficult gestures when only a single multi-antenna receiver access point (AP) is available.

Price forecasts are evaluated in EUR/MWh of error, while storage earns euros; their link depends on the decision rule. We separate the optimal value of a signal V(S|C) from the revenue J(g,S) achieved by an implemented policy. For a price-taking asset with daily throughput bound L and a conditionally sub-Gaussian price law of scale s, the optimal gain over climatology is at most L s sqrt(2 I(S;Pi|C)). This is a square-root envelope, not a prediction for an individual policy. On 939 French day-ahead days the tested bound sits at least two orders of magnitude above the reference uplift. Along a nested Gaussian garbling family, however, plug-in policy revenue is non-monotone even though signal information is monotone: at the residual-noise scale the best Gaussian-shrinkage rule achieves -46% of the uplift, below climatology. The mechanism is ordinal. On a relaxation, optimal schedules depend on interval ordering and profitable-pair tests. A constructed rank-only bid captures 90% of the uplift; exchanging the cheapest and dearest entries of an otherwise exact price vector cuts total revenue by 53%. Within the designed sweep, rank agreement has in-sample R^2 = 0.99, against 0.34 for the information upper bound. Climatology already earns 78% of perfect-foresight revenue. Information quantity alone therefore does not value a forecast-driven policy; realised schedules and ordering must be evaluated explicitly.

This paper studies the uplink performance of OFDM-based cell-free massive MIMO systems in the presence of hardware impairments affecting both low-cost access points and wireless fronthaul transceivers. We consider a centralized architecture with functional split option~8, where the sampled baseband signals are forwarded from the access points to the central processing unit through wireless fronthaul links operating at mmWave or sub-THz frequencies. A unified analytical framework is developed to characterize the aggregate impact of hardware distortions across the access and fronthaul links. In particular, closed-form expressions are derived for channel estimation and uplink spectral efficiency under hardware-impaired transmission and reception. To mitigate the resulting performance degradation, distortion-aware linear receive combining is considered for the wireless fronthaul. Moreover, a max-min fair resource allocation framework is developed to jointly optimize the uplink transmit powers, fronthaul transmit powers, and fronthaul time expansion factor.

Fixed coordinate measuring machines (CMMs) and portable articulated-arm CMMs are often assigned to the same inspection task, but their nominal accuracy specifications do not show whether a change of instrument will preserve the disposition of a part. The question is not simply how far the two results differ, but whether that difference crosses the tolerance boundary. We examined this issue with recorded measurements of cylindrical, cubic, and spherical features under nominal 20 {\deg}C and 30 {\deg}C conditions. Repeated records and two roughness profiles without sufficient acquisition information were removed, leaving six dimensional and four form profiles. For each dimensional feature, the distances of the two system means from nominal define the exact tolerance interval in which the systems receive opposite direct labels. The fixed-CMM stream was approximately 11.2 {\mu}m higher than the articulated-arm stream at both conditions. All four form profiles fell on opposite sides of the recorded 10 {\mu}m upper limit. The dimensional disagreement intervals also overlapped strongly; their mean widths were 6.573 {\mu}m at 20 {\deg}C and 4.995 {\mu}m at 30 {\deg}C. The results clarify why an average difference between instruments is not, by itself, a measure of substitution risk. The proposed tolerance map identifies the feature-tolerance combinations for which instrument choice can change the recorded inspection label and, therefore, where a controlled equivalence study and a task-specific uncertainty budget are needed before substitution.

Heterogeneous federated systems require agents to learn and exchange informative representations despite differences in data distributions, sensing modalities, model architectures, latent dimensionalities, and local learning objectives. To address this challenge, we propose Sheaf-based Federated Representation Learning (SFRL), a general framework that jointly optimizes local objectives with a manifold-constrained geometric alignment regularizer based on learnable sheaf restriction maps. Unlike most existing approaches, SFRL does not assume a shared global latent space. Instead, global consistency emerges from the alignment of neighboring latent representations through orthogonal transformations and isometric embeddings. This alignment is enforced by a quadratic gluing regularizer induced by the sheaf Laplacian, whose learnable restriction maps adapt the geometry to the observed data. The penalty is evaluated on a small set of shared pilot samples, ensuring scalability and communication efficiency. We develop a decentralized algorithm for solving SFRL, termed Sheaf-FRL, which alternates between gradient updates of the local models and closed-form Procrustes updates of the edge-wise restriction maps. We further establish convergence of Sheaf-FRL to first-order stationary points in both deterministic and stochastic settings. As an application, we consider a cooperative classification task in the context of semantic communication, under model and data heterogeneity. Our results show that Sheaf-FRL outperforms baseline approaches in terms of local and post-communication classification accuracy across different levels of local distribution shift and exhibits greater robustness to latent-space dimensionality compression.

Continuous hemodynamic monitoring guides treatment decisions in surgery and intensive care. However, gold-standard signals are only measured in severe cases due to risks associated with invasive measurement. In this work, we introduce HIPNO (Hemodynamic Inference via Physics-informed Neural Operators) to recover hemodynamic state from ubiquitous, non-invasive signals and expand access to advanced monitoring. HIPNO addresses a problem of scale symmetry in physics-informed hemodynamic inference, where different combinations of flow, resistance, and compliance can generate the same observed pressure. We identify the symmetry group of the observation model and parameterize the network in its quotient space. For the 3-element Windkessel model, the quotient coordinates are the compliance-normalized flow $U=Q/C$, the decay time constant $\tau_{WK}=R_2 C$, and the characteristic-impedance coordinate $\kappa=R_1 C$. Across 945499 intraoperative windows from 2562 patients, HIPNO predicts $\tau_{wave}$, a proxy for vascular decay derived from pressure, with 32% lower error on the log scale than a population baseline while preserving mean arterial pressure accuracy. Because vascular decay and flow drive occupy separate coordinates, counterfactual perturbations produce the expected directional responses in at least 90% of windows in almost all prespecified scenarios, a separation unavailable to pressure-only baselines. The coordinates are also used as inputs to a calibration model for monitored cardiac output. Finally, the formulation identifies the external compliance or flow reference required to recover absolute physical scale.

Capacity markets award reserve in discrete increments, yet valuation models almost always treat the awarded quantity as continuous. We show that the size dependence of the resulting error is organised by one dimensionless number, the granularity ratio rho = delta/P of award increment to installed power: the joint energy-and-reserve problem is positively homogeneous, so, at fixed market data and operating parameters, asset size enters per-megawatt revenue only through rho and storage duration. Three structural results follow. Rounding a continuous award down to the lattice is always feasible, so the valuation error is O(rho) and its upper bound falls as the inverse of asset size. The largest feasible award is P rho floor(1/rho), so up to half of a connection can be structurally unsellable, worst just below twice the increment. Pledging the whole connection forces a zero net grid position for the entire product, converting a partial capacity commitment into exclusion from the energy market. On 943 days of French day-ahead and frequency containment reserve settlement prices, a 1 MW two-hour battery priced with a continuous award is worth 4.2% more than the same asset on the 1 MW French lattice, and 77% of that difference is displaced arbitrage rather than foregone capacity. The bias is not monotone in size: the largest value on our grid is 21.5% at rho = 0.505. Bid granularity is therefore a market-design variable with a size-dependent incidence, and the measured gap is at once a modelling error, a sizing rule, and a ceiling on aggregation value.

Data-driven health-state estimators for SiC (Silica-Carbide) power modules typically report their performance on a single accelerated-aging campaign, and how that performance transfers to a different failure mechanism is rarely tested. We benchmark five reference methods from the prognostics and condition-monitoring literature against a physics-informed NODE (Neural Ordinary Differential Equation) on two SiC power-cycling campaigns driven by structurally different failure mechanisms, solder-layer fatigue and wire-bond lift-off, under a per-module $k$-fold protocol. The NODE is evaluated under two input regimes that share the rest of the pipeline: the baseline electrical precursors and a set of cumulative thermoelectric features. Every reference method degrades on the wire-bond campaign, with average errors growing and precision decreasing with respect to their performance on the soldered campaign. The NODE fed with the cumulative features keeps its soldered-campaign metrics on both mechanisms, with differences inside the fold-to-fold variance, while the same architecture fed with the baseline precursors falls back to the reference-method cluster. The input representation contributes at least as much as the architecture to failure-mechanism transferability of a health-state estimator.

Silicon carbide (SiC) power modules are increasingly deployed in automotive traction inverters, where condition monitoring is essential to prevent in-service failures. Despite extensive qualification under AQG 324, no consolidated approach exists for in-field health state estimation: physics-of-failure lifetime models lack real-time applicability, purely data-driven architectures require large labeled datasets and generalize poorly, and physics-informed frameworks remain too demanding for embedded deployment. We address SiC MOSFET modules assembled with sintered packaging, which suppresses solder degradation and produces aging behavior distinct from previously studied devices. Instead of the smooth quasi-exponential drift of solder-based modules, the forward voltage drop $V_{DS}$ exhibits multi-regime profiles, with wirebond liftoff events introducing abrupt, non-monotonic perturbations. We propose a condition monitoring framework combining three elements. First, physics-informed features replace raw sensor signals with cumulative damage indicators derived from junction temperature swing, mean junction temperature and a Miner rule accumulator, encoding degradation history in an interpretable form. Second, a monotonicity constraint enforced by gradient penalty regularization embeds the expected degradation direction as a physics-guided prior. Third, a heavy-tailed output distribution replaces the point estimate, giving calibrated uncertainty robust to the out-of-distribution variance introduced by liftoff. On an industrial power cycling dataset from Infineon Technologies, several neural architectures are compared under a strict cross-validation protocol. The full configuration reduces mean absolute error by approximately 70% over purely data-driven baselines and stays stable across all folds, while remaining lightweight enough for embedded deployment.

Recent Text-To-Speech (TTS) systems have achieved strong naturalness and zero-shot voice cloning performance, but fine-grained control of expressive speech at the word or phoneme level remains challenging. We propose CtrlSpeech, a controllable, expressive TTS framework with coarse-to-fine control. Built on the DiTAR architecture, CtrlSpeech combines global speaker conditioning with phone-aligned pitch, loudness, and duration signals, enabling localized prosodic control while preserving the target speaker's timbre. This design allows users to adjust expressive attributes at a fine temporal granularity, making speech refinement more flexible and controllable. Experimental results show that CtrlSpeech achieves competitive zero-shot TTS performance and improves controllability over expressive attributes, demonstrating its effectiveness for flexible and practical expressive speech synthesis.

The fluid antenna system (FAS) enables position reconfigurability. A potential drawback of real-time FAS, however, is that it requires complete channel state information (CSI) for each FAS port at every communication time slot, an approach referred to as ideal-FAS. Recognizing the difficulties of achieving ideal-FAS, we propose a FAS scheme based on incomplete CSI, referred to as semi-blind FAS. This paper first introduces the spatial-temporal framework of FAS, upon which the proposed semi-blind FAS is developed. The proposed semi-blind FAS is lightweight and computationally efficient, scalable to an arbitrary number of ports and time slots, and operates without pre-training or deep learning structures. The scheme effectively exploits incomplete historical CSI to estimate the conditional distribution across all FAS ports at the desired time slot, thereby identifying the statistical optimal port for signal reception. Generally, the key idea of semi-blind FAS is to select the optimal port through conditional distribution analysis, from a statistical perspective, with optimality defined according to the scenario of interest. Inspired by information-theoretic entropy, we further develop the residual entropy power ratio to characterize how physical parameters influence the performance gap between semi-blind FAS and ideal-FAS. Our analysis reveals that estimation performance depends not only on the number of sampled ports and time slots, but also on the specific indices of ports with given CSI at each time slot, i.e., the port sampling strategy. This critical factor has been largely overlooked in existing port estimation studies. Numerical results demonstrate that the proposed semi-blind FAS achieves performance comparable to, and in some cases indistinguishable from, that of ideal-FAS, while requiring significantly fewer port CSI measurements and lower port switching speeds.

This paper investigates causal radio scheduling for progressive edge inference, with the goal of maximizing timely inference throughput under job-specific deadlines. Specifically, a multi-tenant system is modeled in which each job alternates between wireless transmission and graphics processing unit (GPU) computation. The model captures time-varying uplink service, inter-stage precedence, variant-aware batching, non-preemptive execution on two GPU streams, and heterogeneous deadlines. To account for delayed radio-GPU coupling, a split-gate pooled-evidence stochastic-rollout (SGPE-SR) scheduler is proposed. Candidate and fallback policies are selected on one set of sampled futures, after which admission is evaluated on an independent held-out set; common random numbers are retained within each held-out candidate-fallback comparison. The override is executed only when its pooled nominal/recent-history held-out advantage is sufficiently positive and an independent recent-history replication statistic is nonnegative. System-level simulations using measured GPU profiles and paired random instances show that, in a prespecified long-horizon evaluation over 30 previously unused seeds and 3,379 offered jobs, SGPE-SR improves timely completions over the strongest rate-based baseline by 4.27%, with a mean paired gain of 3.00 jobs per run and a 95% bootstrap interval of [1.60, 4.43].

This paper investigates the coupling between wireless input completion and graphics processing unit (GPU) execution in centralized multi-agent cooperative perception. Specifically, beyond the conceptual treatment of completion-triggered overlap, a complete execution path is realized and validated on a physical GPU for a cooperative-perception deep neural network (DNN). Each encoder branch is released immediately upon completion of its corresponding input transmission, while the original fusion dependencies and inference mapping are preserved. The resulting release-triggered communication computation coupling (RTCC) propagates validated wireless completion events through host-to-device (H2D) staging, CUDA synchronization, and dependency-preserving branch dispatch, while remaining compatible with causal wireless schedulers. Experiments combining trace-driven wireless arrivals, physical GPU execution, and measured-DAG evaluation show that RTCC reduces complete-detection latency across different communication loads and schedulers, while preserving identical detection outputs and average-precision performance relative to conventional execution.