Papers

We introduce a multi-windowed graph Fourier transform (MWGFT) for the joint vertex-frequency analysis of signals defined on graphs. Building on generalized translation and modulation induced by the graph Laplacian, the proposed framework extends the windowed graph Fourier transform by allowing multiple analysis and synthesis windows. Exact reconstruction formulas are derived for complex-valued graph signals, together with sufficient and computable conditions guaranteeing stable invertibility. The associated families of windowed graph Fourier atoms are shown to form frames for the space of graph signals. Numerical experiments on synthetic and real world graphs confirm exact reconstruction up to machine precision and demonstrate improved stability and vertex-frequency localization compared to single-window constructions, particularly on irregular graph topologies.

We propose a novel approach to computationally efficient GP training based on the observation that square-exponential (SE) covariance matrices contain several off-diagonal entries extremely close to zero. We construct a principled procedure to eliminate those entries to produce a banded-matrix approximation to the original covariance, whose inverse and determinant can be computed at a reduced computational cost, thus contributing to an efficient approximation to the likelihood function. We provide a theoretical analysis of the proposed method to preserve the structure of the original covariance in the 1D setting with SE kernel, and validate its computational efficiency against the variational free energy approach to sparse GPs.

This work presents a framework for certifying small-signal frequency stability of a power system with line dynamics and heterogeneous bus dynamics. This framework can certify the stability of systems which include synchronous generators, synchronous condensers, and converter-interfaced resources with a wide range of controls. Moreover, it can do so without detailed or precise knowledge of the network topology. With this framework, we also provide a detailed analysis of how proportional-derivative (PD) droop can improve the stability margin of the frequency response. The stability certificates presented in this work, which extend prior results by incorporating line dynamics, provide insight into how the control parameters for different units in the system impact the overall frequency stability. While damper windings have long been understood to improve the frequency synchronization between machines, the dynamics of the damper windings are complex, making them difficult to analyze. To address this gap, this paper derives a novel reduced-order model of the damper windings in the form of a derivative droop term. Moreover, we show that derivative droop terms used in grid-forming (GFM) control can be understood as a form of damper winding emulation. Our analytical stability conditions highlight the importance of damper windings (or their emulation) in facilitating frequency synchronization and suppressing unstable interactions between GFM converters. These results are validated with electromagnetic-transient (EMT) simulation.

Quantum hardware suffers from intrinsic device heterogeneity and environmental drift, forcing practitioners to choose between suboptimal non-adaptive controllers or costly per-device recalibration. We derive a scaling law lower bound for meta-learning showing that the adaptation gain (expected fidelity improvement from task-specific gradient steps) saturates exponentially with gradient steps and scales linearly with task variance, providing a quantitative criterion for when adaptation justifies its overhead. Validation on quantum gate calibration shows negligible benefits for low-variance tasks but >40% fidelity gains on two-qubit gates under extreme out-of-distribution conditions (10$\times$ the training noise), with implications for reducing per-device calibration time on cloud quantum processors. Further validation on classical linear-quadratic control confirms these laws emerge from general optimization geometry rather than quantum-specific physics. We further introduce a few-shot pre-adaptation protocol that estimates the optimal adaptation budget from $N{=}3$-5 probe steps within 3-19% relative error across out-of-distribution regimes.

For active intervention tasks in underwater environments, the use of autonomous vehicles is just now emerging as an active area of research. During operation, for various reasons, the robot might find itself on a collision course with an obstacle in its environment. In this paper, a switching Nonlinear Model Predictive Control (NMPC) strategy is proposed to safely handle collisions for an Underwater Vehicle-Manipulator System (UVMS). When avoiding the collision is impossible, the control algorithm takes advantage of the manipulator, using it to push against the obstacle, and deflect away from the collision. Virtual experiments are performed to demonstrate the algorithm's capability to successfully detect collisions and either avoid them, or use the manipulator to handle them appropriately without damaging sensitive areas of the vehicle.

Efficient thermal management is critical for ensuring the safety, performance, and durability of lithium ion pouch cells (LIPCs), particularly under high power operating conditions where conventional battery thermal management systems (BTMS) struggle to balance cooling effectiveness, structural simplicity, and weight. Here, we report a lightweight hybrid BTMS that synergistically integrates active liquid cooling with composite phase change material (CPCM) based thermal buffering through a 3D printed hexagonal architecture. The system is fabricated via a two step additive manufacturing process that enables sealed CPCM encapsulation and isolated liquid cooling pathways within a single carbon fiber reinforced nylon module, effectively eliminating leakage risks while allowing precise geometric control. Hexagonally partitioned CPCM cavities maximize the CPCM wall interfacial area and shorten internal conduction paths, accelerating latent heat absorption, while embedded serpentine liquid channels provide continuous convective heat removal and prevent CPCM saturation. A nanocarbon enhanced CPCM is employed to overcome the intrinsic low thermal conductivity of conventional paraffin based materials.

Popularized by their strong image generation performance, diffusion and related methods for generative modeling have found widespread success in visual media applications. In particular, diffusion methods have enabled new approaches to data compression, where realistic reconstructions can be generated at extremely low bit-rates. This article provides a unifying review of recent diffusion-based methods for generative lossy compression, with a focus on image compression. These methods generally encode the source into an embedding and use a diffusion model to iteratively refine it during decoding, so that the reconstruction approximately follows the true data distribution. The embedding can take various forms and is typically transmitted via an auxiliary entropy model, and recent methods also explore the use of diffusion models themselves for information transmission via channel simulation. We review representative approaches through the lens of rate-distortion-perception theory, highlighting the role of common randomness and connections to inverse problems, and identify open challenges.

The pinching-antenna system (PASS), recently proposed as a flexible-antenna technology, has been regarded as a promising solution for several challenges in next-generation wireless networks. It provides large-scale antenna reconfiguration, establishes stable line-of-sight links, mitigates signal blockage, and exploits near-field advantages through its distinctive architecture. This article aims to present a comprehensive overview of the state of the art in PASS. The fundamental principles of PASS are first discussed, including its hardware architecture, circuit and physical models, and signal models. Several emerging PASS designs, such as segmented waveguide-enabled PASS (SWAN), center-fed PASS (C-PASS), and multi-mode PASS (M-PASS), are subsequently introduced, and their design features are discussed. In addition, the properties and promising applications of PASS for wireless sensing are reviewed. On this basis, recent progress in the performance analysis of PASS for both communications and sensing is surveyed, and the performance gains achieved by PASS are highlighted. Existing research contributions in optimization and machine learning are also summarized, with the practical challenges of beamforming and resource allocation being identified in relation to the unique transmission structure and propagation characteristics of PASS. Finally, several variants of PASS are presented, and key implementation challenges that remain open for future study are discussed.

Speech Emotion Recognition systems often use static features like Mel-Frequency Cepstral Coefficients (MFCCs), Zero Crossing Rate (ZCR), and Root Mean Square Energy (RMSE). Because of this, they can misclassify emotions when there is acoustic noise in vocal signals. To address this, we added dynamic features using Dynamic Spectral features (Deltas and Delta-Deltas) along with the Kalman Smoothing algorithm. This approach reduces noise and improves emotion classification. Since emotion changes over time, the Kalman Smoothing filter also helped make the classifier outputs more stable. Tests on the RAVDESS dataset showed that this method achieved a state-of-the-art accuracy of 87% and reduced misclassification between emotions with similar acoustic features

Generative AI for speech and audio is increasingly expected to serve users across languages, cultures, and communities, yet current auditory Large Language Models (LLMs) are still largely trained and evaluated on high-resource data. Globalizing such systems requires handling low-resource settings, where the target speakers, languages, or tasks are poorly represented in training data. In these regimes, collecting enough labeled in-domain data is often impractical, and the small corpora available may still under-represent the test distribution, making direct fine-tuning brittle under domain shift. In-Context Learning (ICL) offers an alternative: instead of updating model parameters for every underserved community, an auditory LLM can adapt at inference time by conditioning on a few local demonstrations. However, vanilla speech ICL remains limited because most auditory LLMs are not explicitly trained to use such demonstrations effectively. We address this gap with Meta Speech In-Context Learning (MetaSICL), a post-training recipe that strengthens an auditory LLM's in-context adaptation ability using only abundant high-resource speech data. Although MetaSICL never trains on the target low-resource domains, it improves performance across two backbones on children's ASR, audio understanding/reasoning, and speech translation and ASR in directions and languages unseen in post-training. We further study the case where some in-domain data is available, using low-resource language ASR as a case study, since recognition for underserved languages is central to globalizing generative AI. Here, using MetaSICL as a warmup for in-domain reinforcement learning yields the strongest results, outperforming direct fine-tuning across five typologically diverse languages. Overall, MetaSICL offers a practical route toward globalizing auditory LLMs by building inference-time adaptation into the model.

Markov decision problems are most commonly solved via dynamic programming. Another approach is Bellman residual minimization, which directly minimizes the squared Bellman residual objective function. However, compared to dynamic programming, this approach has received relatively less attention, mainly because it is often less efficient in practice and can be more difficult to extend to model-free settings such as reinforcement learning. Nonetheless, Bellman residual minimization has several advantages that make it worth investigating, such as more stable convergence with function approximation for value functions. While Bellman residual methods for policy evaluation have been widely studied, methods for policy optimization (control tasks) have been scarcely explored. In this paper, we establish foundational results for the control Bellman residual minimization for policy optimization.

A central difficulty in applying reinforcement learning to real-world applications such as autonomous driving is that conventional scalar reward formulations obtained by aggregating multiple often competing objectives may obscure the structure of their trade-offs. These challenges are particularly relevant to decision-making for heavy-duty vehicles, which requires balancing safety, efficiency, and operational costs under demanding commercial requirements. We present a preference-conditioned multi-objective reinforcement learning framework that explicitly represents these objectives and utilizes Generalized Policy Improvement and Proximal Policy Optimization to learn policies capturing different trade-offs between them. The proposed method approximates the Pareto front across three conflicting objectives: safety, quantified in terms of collisions and successful completion; energy efficiency, quantified using energy cost; and time efficiency, quantified using driver cost. We evaluate the framework in a scalable highway traffic simulation environment tailored for tactical decision making in heavy-duty trucks. It yields smooth and interpretable Pareto fronts, and evaluations across traffic densities and against an analytical baseline show that the learned policies capture meaningful trade-offs. The preference-conditioned policy enables transitions between different driving behaviors without retraining, demonstrating the potential of multi-objective reinforcement learning for adaptive decision making in autonomous heavy-duty vehicles.

We study the quantization of real-valued bandlimited signals on graphs, focusing on low-bit representations. We propose iterative noise-shaping algorithms for quantization, including sampling approaches with and without vertex replacement. The methods leverage the spectral properties of the graph Laplacian and exploit graph incoherence to achieve high-fidelity approximations. Theoretical guarantees are provided for the random sampling method, and extensive numerical experiments on synthetic and real-world graphs illustrate the efficiency and robustness of the proposed schemes.

Raga identification in Indian Art Music (IAM) remains challenging due to the presence of numerous rarely performed Ragas that are not represented in available training datasets. Traditional classification models struggle in this setting, as they assume a closed set of known categories and therefore fail to recognise or meaningfully group previously unseen Ragas. Recent works have tried categorizing unseen Ragas, but they run into a problem of catastrophic forgetting, where the knowledge of previously seen Ragas is diminished. To address this problem, we adopt a unified learning framework that leverages both labeled and unlabeled audio, enabling the model to discover coherent categories corresponding to the unseen Ragas, while retaining the knowledge of previously known ones. We test our model on benchmark Raga Identification datasets and demonstrate its performance in categorizing previously seen, unseen, and all Raga classes. The proposed approach surpasses the previous NCD-based pipeline even in discovering the unseen Raga categories, offering new insights into representation learning for IAM tasks.

Most of the contemporary literature on information freshness solely focuses on the analysis of freshness for martingale estimators, which simply use the most recently received update as the current estimate. While martingale estimators are easier to analyze, they are far from optimal, especially in pull-based update systems, where maximum aposteriori probability (MAP) estimators are known to be optimal, but are analytically challenging. In this work, we introduce a new class of estimators called $p$-MAP estimators, which enable us to model the MAP estimator as a piecewise constant function with finitely many stages, bringing us closer to a full characterization of the MAP estimators when modeling information freshness.

Traffic Steering (TS) dynamically allocates user traffic across cells to enhance Quality of Experience (QoE), load balance, and spectrum efficiency in 5G networks. However, TS algorithms remain vulnerable to adversarial conditions such as interference spikes, handover storms, and localized outages. To address this, an AI-driven fuzz testing framework based on the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) is proposed to systematically expose hidden vulnerabilities. Using NVIDIA Sionna, five TS algorithms are evaluated across six scenarios. Results show that AI-driven fuzzing detects 34.2% more total vulnerabilities and 5.8% more critical failures than traditional testing, achieving superior diversity and edge-case discovery. The observed variance in critical failure detection underscores the stochastic nature of rare vulnerabilities. These findings demonstrate that AI-driven fuzzing offers an effective and scalable validation approach for improving TS algorithm robustness and ensuring resilient 6G-ready networks.

We consider time discretization for score-based diffusion models to generate samples from a learned reverse-time dynamic on a finite grid. Uniform and hand-crafted grids can be suboptimal given a budget on the number of time steps. We introduce Adaptive Reparameterized Time (ART), which controls the clock speed of a reparameterized time variable to redistribute computation along the sampling trajectory while preserving the terminal time, with the objective of minimizing the aggregate Euler discretization error. We derive a randomized companion ART-RL that recasts ART as a continuous-time reinforcement learning problem with Gaussian policies, and prove a two-directional bridge between the two: the deterministic ART optimum lifts to an optimal Gaussian policy, and conversely any optimal Gaussian policy must recover the ART control through its mean. This bridge turns continuous-time actor–critic learning into a principled, rather than heuristic, route to the deterministic timestep optimum. Within the official EDM pipeline, ART-RL improves FID on CIFAR–10 across a wide range of budgets; after one-time offline training, the distilled deterministic schedule transfers without retraining to AFHQv2, FFHQ, and ImageNet at no extra inference cost.

Large-scale video streaming events attract millions of simultaneous viewers, stressing existing delivery infrastructures. Client-driven adaptation reacts slowly to shared congestion, while server-based coordination introduces scalability bottlenecks and single points of failure. We present COMETS, a coordinated multi-destination video transmission framework that leverages information-centric networking principles such as request aggregation and in-network state awareness to enable scalable, fair, and adaptive rate control. COMETS introduces a novel range-interest protocol and distributed in-network decision process that aligns video quality across receiver groups while minimizing redundant transmissions. To achieve this, we develop a lightweight distributed optimization framework that guides per-hop quality adaptation without centralized control. Extensive emulation shows that COMETS consistently improves bandwidth utilization, fairness, and user-perceived quality of experience over DASH, MoQ, and ICN baselines, particularly under high concurrency. The results highlight COMETS as a practical, deployable approach for next-generation scalable video delivery.

To enable Integrated Communications and Sensing (ICAS) in a peer-to-peer vehicular network, precise synchronization in frequency and phase among the communicating entities is required. In addition, self-driving cars need accurate position estimates of the surrounding vehicles. In this work, we propose a joint, distributed synchronization and localization scheme for a network of communicating entities. Our proposed scheme is mostly signal-agnostic and therefore can be applied to a wide range of possible ICAS signals. We also mitigate the effect of finite sampling frequencies, which otherwise would degrade the synchronization and localization performance severely.

Event-based vision provides high-speed, energy-efficient sensing for applications such as autonomous navigation and motion tracking. However, implementing this technology in the long-wave infrared remains a significant challenge. Traditional infrared sensors are hindered by slow thermal response times or the heavy power requirements of cryogenic cooling. Here, we introduce the first event-based infrared detector operating in a Poisson-counting regime. This is realized with a spintronic Poisson bolometer capable of broadband detection from 0.8-14$\mu\text{m}$. In this regime, infrared signals are detected through statistically resolvable changes in stochastic switching events. This approach enables room-temperature operation with high timing resolution. Our device achieves a maximum event rate of 1,250 Hz, surpassing the temporal resolution of conventional uncooled microbolometers by a factor of 4. Power consumption is kept low at 0.2$\mu$W per pixel. This work establishes an operating principle for infrared sensing and demonstrates a pathway toward high-speed, energy-efficient, event-driven thermal imaging.

We propose a tensor-based domain alignment (DA) algorithm designed to align source and target tensors within an invariant subspace through the use of alignment matrices. These matrices along with the subspace undergo iterative optimization of which constraint is on oblique manifold, which offers greater flexibility and adaptability compared to the traditional Stiefel manifold. Moreover, regularization terms defined to preserve the variance of both source and target tensors, ensures robust performance. Our framework is versatile, effectively generalizing existing tensor-based DA methods as special cases. Through extensive experiments, we demonstrate that our approach not only enhances DA conversion speed but also significantly boosts classification accuracy. This positions our method as superior to current state-of-the-art techniques, making it a preferable choice for complex domain adaptation tasks.

In the context of integrated sensing and communications (ISAC), this paper presents an experimental investigation of the relationship between monostatic sensing and naturally bistatic communication channels in an indoor millimeter-wave environment. We characterize the propagation channel in the joint delay–angle domain, extract dominant multipath components (MPCs) and associate them with physical scatterers in the environment, and demonstrate how communication MPCs can be explicitly recovered from sensing channels. Finally, the radar cross-sections (RCSs) of two key scatterers, namely the wall and metal plate, are obtained based on calibrated channel power and reconstructed propagation distances.

Recent advancements have underscored the relevance of low-resolution analog-to-digital converters (ADCs) in integrated sensing and communication (ISAC) systems. Nevertheless, their specific impact on hybrid radar fusion (HRF) remains largely unexplored. In HRF systems, where uplink (UL) paths carry direct and reflected signals in the same frequency band, the reflected signal is often significantly weaker, making HRF performance particularly sensitive to ADC resolution. To study this effect, we use the quantized Cramér-Rao bound (CRB) to measure sensing accuracy. This work derives an upper bound on the quantized CRB for angle of arrival (AoA) estimation and explores CRB-rate trade-offs through two formulated optimization problems. Simulation results indicate that HRF becomes infeasible when the dynamic range of the received signal exceeds the dynamic range supported by the ADC, which is inherently limited by its resolution. Furthermore, the UL communication rate does not increase significantly when the ADC resolution is raised beyond a certain threshold. These observations highlight a fundamental trade-off between sensing and communication performance: while HRF performance benefits from higher ADC resolutions, the corresponding gains in communication rate plateau. This trade-off is effectively characterized using CRB-rate boundaries derived through simulation.

We address the problem of time-frequency audio inpainting, where the goal is to fill missing spectrogram portions with consistent information. Despite recent advances, existing approaches still face limitations in both reconstruction quality and computational efficiency. To bridge this gap, we propose a method that utilizes a phase-aware signal prior which exploits estimates of the instantaneous frequency. An optimization problem is formulated and solved using the generalized Chambolle-Pock algorithm. The proposed method is evaluated against other time-frequency inpainting methods, specifically a deep-prior audio inpainting neural network, the autoregression-based approach known as Janssen-TF, and a sparsity-driven baseline. For short gap durations, the proposed approach achieves superior SNR, while performing comparably to Janssen-TF on larger gaps. In terms of perceptual quality (both objective and subjective), the proposed method consistently outperforms existing methods across all gap lengths. In addition, the reconstructions are obtained with a substantially reduced computational cost compared to alternative methods.

Walking is a key movement of interest in biomechanics, yet gold-standard data collection methods are time- and cost-expensive. This paper presents a real-time, multimodal, high sample rate lower-limb motion capture framework, based on wireless wearable sensors and machine learning algorithms. Random Forests are used to estimate joint angles from IMU data, and ground reaction force (GRF) is predicted from instrumented insoles, while joint moments are predicted from angles and GRF using deep learning based on the ResNet-16 architecture. All three models achieve good accuracy compared to literature, and the predictions are logged at 1 kHz with a minimal delay of 23 ms for 20s worth of input data. The present work fully relies on wearable sensors, covers all five major lower limb joints, and provides multimodal comprehensive estimations of GRF, joint angles, and moments with minimal delay suitable for biofeedback applications.

Integrated localization and communication systems aim to reuse communication waveforms for simultaneous data transmission and localization, but delay resolution is fundamentally limited by the available bandwidth. In practice, large contiguous bandwidths are difficult to obtain due to hardware constraints and spectrum fragmentation. Aggregating non-contiguous narrow bands can increase the effective frequency span, but a non-contiguous frequency layout introduces challenges such as elevated sidelobes and ambiguity in delay estimation. This paper introduces a point-spread-function (PSF)-centric framework for dual-band OFDM delay estimation. We model the observed delay profile as the convolution of the true target response with a PSF determined by the dual-band subcarrier selection pattern, explicitly linking band configuration to resolution and ambiguity. To suppress PSF-induced artifacts, we adapt the RELAX algorithm for dual-band multi-target delay estimation. Simulations demonstrate improved robustness and accuracy in dual-band scenarios, supporting ILC under fragmented spectrum.

With the development of the sixth-generation (6G) communication system, Channel State Information (CSI) plays a crucial role in improving network performance. Traditional Channel Charting (CC) methods map high-dimensional CSI data to low-dimensional spaces to help reveal the geometric structure of wireless channels. However, most existing CC methods focus on learning static geometric structures and ignore the dynamic nature of the channel over time, leading to instability and poor topological consistency of the channel charting in complex environments. To address this issue, this paper proposes a novel time-series channel charting approach based on the integration of Long Short-Term Memory (LSTM) networks and Auto encoders (AE) (LSTM-AE-CC). This method incorporates a temporal modeling mechanism into the traditional CC framework, capturing temporal dependencies in CSI using LSTM and learning continuous latent representations with AE. The proposed method ensures both geometric consistency of the channel and explicit modeling of the time-varying properties. Experimental results demonstrate that the proposed method outperforms traditional CC methods in various real-world communication scenarios, particularly in terms of channel charting stability, trajectory continuity, and long-term predictability.

Real-world audio recordings often contain multiple speakers and various degradations, which limit both the quantity and quality of speech data available for building state-of-the-art speech processing models. Although end-to-end approaches that concatenate speech enhancement (SE) and speech separation (SS) to obtain a clean speech signal for each speaker are promising, conventional SE-SS methods suffer from complex degradations beyond additive noise. To this end, we propose Geneses, a generative framework to achieve unified, high-quality SE–SS. Our Geneses leverages latent flow matching to estimate each speaker's clean speech features using multi-modal diffusion Transformer conditioned on self-supervised learning representation from noisy mixture. We conduct experimental evaluation using two-speaker mixtures from LibriTTS-R under two conditions: additive-noise-only and complex degradations. The results demonstrate that Geneses significantly outperforms a conventional mask-based SE–SS method across various objective metrics with high robustness against complex degradations. Audio samples are available in our demo page.

Hybrid reconfigurable intelligent surfaces (HRIS) enhance wireless systems by combining passive reflection with active signal amplification. However, jointly optimizing the transmit beamforming with the HRIS reflection and amplification coefficients to maximize spectral efficiency (SE) is a non-convex problem, and conventional iterative solutions are computationally intensive. To address this, we propose a deep reinforcement learning (DRL) framework that learns a direct mapping from channel state information to the near-optimal transmit beamforming and HRIS configurations. The DRL model is trained offline, after which it can compute the beamforming and HRIS configurations with low complexity and latency. Simulation results demonstrate that our DRL-based method achieves 95% of the SE obtained by the alternating optimization benchmark, while significantly lowering the computational complexity.

This work presents a speech-to-text system "Pisets" for scientists and journalists which is based on a three-component architecture aimed at improving speech recognition accuracy while minimizing errors and hallucinations associated with the Whisper model. The architecture comprises primary recognition using Wav2Vec2, false positive filtering via the Audio Spectrogram Transformer (AST), and final speech recognition through Whisper. The implementation of curriculum learning methods and the utilization of diverse Russian-language speech corpora significantly enhanced the system's effectiveness. Additionally, advanced uncertainty modeling techniques were introduced, contributing to further improvements in transcription quality. The proposed approaches ensure robust transcribing of long audio data across various acoustic conditions compared to WhisperX and the usual Whisper model. The source code of "Pisets" system is publicly available at GitHub: https://github.com/bond005/pisets.