Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Electrical Engineering and Systems Science

eess Watch Electrical Engineering and Systems Science

Wednesday 28 January

Systems and Control

A Data-Driven Krasovskii-Based Approach for Safety Controller Design of Time-Delayed Uncertain Polynomial Systems

Omid Akbarzadeh, MohammadHossein Ashoori, Amy Nejati et al. (+1 more)

We develop a data-driven framework for the synthesis of robust Krasovskii control barrier certificates (RK-CBC) and corresponding robust safety controllers (R-SC) for discrete-time input-affine uncertain polynomial systems with (partially) unknown dynamics, while explicitly accounting for unknown-but-bounded disturbances and time-invariant delays using only observed input-state data. Although control barrier certificates have been extensively studied for safety analysis of control systems, existing work on unknown systems with time delays, particularly in the presence of disturbances, remains limited. The challenge of safety synthesis for such systems stems from two main factors: first, the system's mathematical model is unavailable; and second, the safety conditions should explicitly incorporate the effects of time delays on system evolution during the synthesis process, while remaining robust to unknown disturbances. To address these challenges, we develop a data-driven framework based on RK-CBC, extending the classical CBC formulation for delay-free systems to explicitly account for time delays by aggregating delayed components within the barrier construction. The proposed framework relies solely on input-state data collected over a finite time horizon, enabling the direct synthesis of RK-CBC and R-SC from observed trajectories without requiring an explicit system model. The synthesis is cast as a data-driven sum-of-squares (SOS) feasibility program, yielding a structured design methodology. As a result, robust safety is guaranteed in the presence of unknown disturbances and time delays as long as the closed-loop trajectories remain within the state set. The effectiveness of the proposed method is demonstrated through three case studies, including two physical systems.

Read the full abstract

All authors Omid Akbarzadeh, MohammadHossein Ashoori, Amy Nejati, Abolfazl Lavaei

Signal Processing · Cross-listed from cs.LG

A Learning-based Framework for Spatial Impulse Response Compensation in 3D Photoacoustic Computed Tomography

Kaiyi Yang, Seonyeong Park, Gangwon Jeong et al. (+4 more)

Photoacoustic computed tomography (PACT) is a promising imaging modality that combines the advantages of optical contrast with ultrasound detection. Utilizing ultrasound transducers with larger surface areas can improve detection sensitivity. However, when computationally efficient analytic reconstruction methods that neglect the spatial impulse responses (SIRs) of the transducer are employed, the spatial resolution of the reconstructed images will be compromised. Although optimization-based reconstruction methods can explicitly account for SIR effects, their computational cost is generally high, particularly in three-dimensional (3D) applications. To address the need for accurate but rapid 3D PACT image reconstruction, this study presents a framework for establishing a learned SIR compensation method that operates in the data domain. The learned compensation method maps SIR-corrupted PACT measurement data to compensated data that would have been recorded by idealized point-like transducers. Subsequently, the compensated data can be used with a computationally efficient reconstruction method that neglects SIR effects. Two variants of the learned compensation model are investigated that employ a U-Net model and a specifically designed, physics-inspired model, referred to as Deconv-Net. A fast and analytical training data generation procedure is also a component of the presented framework. The framework is rigorously validated in virtual imaging studies, demonstrating resolution improvement and robustness to noise variations, object complexity, and sound speed heterogeneity. When applied to in-vivo breast imaging data, the learned compensation models revealed fine structures that had been obscured by SIR-induced artifacts. To our knowledge, this is the first demonstration of learned SIR compensation in 3D PACT imaging.

Read the full abstract

All authors Kaiyi Yang, Seonyeong Park, Gangwon Jeong, Hsuan-Kai Huang, Alexander A. Oraevsky, Umberto Villa, Mark A. Anastasio

Systems and Control · Cross-listed from cs.LG

Parametric and Generative Forecasts of EPEX Day-Ahead Energy Market Curves

Julian Gutierrez and Redouane Silvente

We propose two methodologies for modelling aggregated supply and demand curves in the EPEX SPOT Day-Ahead market, emphasizing generative models as a way to recover distributional variability. The first is a low-dimensional parametric representation that yields deterministic point forecasts; the second is a high-dimensional order-level representation that samples from a conditional distribution of plausible curves. Both model the full curve structure, enabling the analysis of price sensitivity, volume sensitivity, and price impact. The parametric representation uses plateau levels, elastic-region boundaries, and polynomial coefficients, forecast with eXtreme Gradient Boosting. The main contribution is the generative representation, which uses price arrivals and volume-increment marks and is implemented with conditional Denoising Diffusion Probabilistic Models. Using French EPEX data from 2021 to 2024, we evaluate both approaches through curve reconstruction and a price-maker storage optimization problem. The parametric implementation provides a deterministic reference, while the diffusion-based implementation produces distributions of plausible curves and achieves higher realized profits and smaller gaps to an oracle benchmark in the storage application.

Read the full abstract

Signal Processing

WirelessJEPA: A Multi-Antenna Foundation Model using Spatio-temporal Wireless Latent Predictions

Viet Chu, Omar Mashaal and Hatem Abou-Zeid

We propose WirelessJEPA, a novel wireless foundation model (WFM) that uses the Joint Embedding Predictive Architecture (JEPA). WirelessJEPA learns general-purpose representations directly from real-world multi-antenna IQ data by predicting latent representations of masked signal regions. This enables multiple diverse downstream tasks without reliance on carefully engineered contrastive augmentations. To adapt JEPA to wireless signals, we introduce a 2D antenna time representation that reshapes multi-antenna IQ streams into structured grids, allowing convolutional processing with block masking and efficient sparse computation over unmasked patches. Building on this representation, we propose novel spatio temporal mask geometries that encode inductive biases across antennas and time. We evaluate WirelessJEPA across six downstream tasks and demonstrate it's robust performance and strong task generalization. Our results establish that JEPA-based learning as a promising direction for building generalizable WFMs.

Read the full abstract

Systems and Control

C-AoEI-Aware Cross-Layer Optimization in Satellite IoT Systems: Balancing Data Freshness and Transmission Efficiency

Yuhua Zhao, Tiejun Lv and Ke Wang

Satellite-based Internet of Things (S-IoT) faces a fundamental trilemma: propagation delay, dynamic fading, and bandwidth scarcity. While Layer-coded Hybrid ARQ (L-HARQ) enhances reliability, its backtracking decoding introduces age ambiguity, undermining the standard Age of Information (AoI) metric and obscuring the critical trade-off between data freshness and transmission efficiency. To bridge this gap, we propose a novel cross-layer optimization framework centered on a new metric, the Cross-layer Age of Error Information (C-AoEI). We derive a closed-form expression for C-AoEI, explicitly linking freshness to system parameters, establishing an explicit analytical connection between freshness degradation and channel dynamics. Building on this, we develop a packet-level encoded L-HARQ scheme for multi-GBS scenarios and an adaptive algorithm that jointly optimizes coding and decision thresholds. Extensive simulations demonstrate the effectiveness of our proposed framework: it achieves 31.8% higher transmission efficiency and 17.2% lower C-AoEI than conventional schemes. The framework also proves robust against inter-cell interference and varying channel conditions, providing a foundation for designing efficient, latency-aware next-generation S-IoT protocols.

Read the full abstract

Signal Processing

Coverage Performance Analysis of FAS-enhanced LoRa Wide Area Networks under both Co-SF and Inter-SF Interference

Gaoze Mu, Yanzhao Hou, Mingjie Chen et al. (+4 more)

This paper presents an analytical framework for evaluating the coverage performance of the fluid antenna system (FAS)-enhanced LoRa wide-area networks (LoRaWANs). We investigate the effects of large-scale pathloss in LoRaWAN, small-scale fading characterized by FAS, and dense interference (i.e., packet collisions under the ALOHA protocol) arising from randomly deployed end devices (EDs). Both co-spreading factor (co-SF) interference (with the same SF) and inter-SF interference (with different SFs) are introduced into the network, and their differences in physical characteristics are also considered in the analysis. Additionally, simple yet accurate statistical approximations of the FAS channel envelope and power are derived using the extreme-value theorem. Based on the approximated channel expression, the theoretical coverage probability of the proposed FAS-enhanced LoRaWAN is derived. Numerical results validate our analytical approximations by exhibiting close agreement with the exact correlation model. Notably, it is revealed that a FAS with a normalized aperture of 1 times 1 can greatly enhance network performance, in terms of both ED numbers and coverage range.

Read the full abstract

All authors Gaoze Mu, Yanzhao Hou, Mingjie Chen, Yuanyu Hu, Yongan Zheng, Qimei Cui, Xiaofeng Tao

Systems and Control · Cross-listed from cs.RO

A Taylor Series Approach to Correct Localization Errors in Robotic Field Mapping using Gaussian Processes

Muzaffar Qureshi, Tochukwu Elijah Ogri, Kyle Volle et al. (+1 more)

Gaussian Processes (GPs) are powerful non-parametric Bayesian models for regression of scalar fields, formulated under the assumption that measurement locations are perfectly known and the corresponding field measurements have Gaussian noise. However, many real-world scalar field mapping applications rely on sensor-equipped mobile robots to collect field measurements, where imperfect localization introduces state uncertainty. Such discrepancies between the estimated and true measurement locations degrade GP mean and covariance estimates. To address this challenge, we propose a method for updating the GP models when improved estimates become available. Leveraging the differentiability of the kernel function, a second-order correction algorithm is developed using the precomputed Jacobians and Hessians of the GP mean and covariance functions for real-time refinement based on measurement location discrepancy data. Simulation results demonstrate improved prediction accuracy and computational efficiency compared to full model retraining.

Read the full abstract

All authors Muzaffar Qureshi, Tochukwu Elijah Ogri, Kyle Volle, Rushikesh Kamalapurkar

Audio and Speech Processing · Cross-listed from cs.CL

Mind the Shift: Using Delta SSL Embeddings to Enhance Child ASR

Zilai Wang, Natarajan Balaji Shankar, Kaiyuan Zhang et al. (+2 more)

Self-supervised learning (SSL) models have achieved impressive results across many speech tasks, yet child automatic speech recognition (ASR) remains challenging due to limited data and pretraining domain mismatch. Fine-tuning SSL models on child speech induces shifts in the representation space. We hypothesize that delta SSL embeddings, defined as the differences between embeddings from a finetuned model and those from its pretrained counterpart, encode task-specific information that complements finetuned features from another SSL model. We evaluate multiple fusion strategies on the MyST childrens corpus using different models. Results show that delta embedding fusion with WavLM yields up to a 10 percent relative WER reduction for HuBERT and a 4.4 percent reduction for W2V2, compared to finetuned embedding fusion. Notably, fusing WavLM with delta W2V2 embeddings achieves a WER of 9.64, setting a new state of the art among SSL models on the MyST corpus. These findings demonstrate the effectiveness of delta embeddings and highlight feature fusion as a promising direction for advancing child ASR.

Read the full abstract

All authors Zilai Wang, Natarajan Balaji Shankar, Kaiyuan Zhang, Zihan Wang, Abeer Alwan

Image and Video Processing · Cross-listed from cs.LG

Scaling Next-Brain-Token Prediction for MEG

Richard Csaky

We present a large autoregressive model for source-space MEG that scales next-token prediction to long context across datasets and scanners: handling a corpus of over 500 hours and thousands of sessions across the three largest MEG datasets. A modified SEANet-style vector-quantizer reduces multichannel MEG into a flattened token stream on which we train a Qwen2.5-VL backbone from scratch to predict the next brain token and to recursively generate minutes of MEG from up to a minute of context. To evaluate long-horizon generation, we introduce task-matched tests: (i) on-manifold stability via generated-only drift compared to the time-resolved distribution of real sliding windows, and (ii) conditional specificity via correct context versus prompt-swap controls using a neurophysiologically grounded metric set. We train on CamCAN and Omega and run all analyses on held-out MOUS, establishing cross-dataset generalization. Across metrics, generations remain relatively stable over long rollouts and are closer to the correct continuation than swapped controls. Code available at: https://github.com/ricsinaruto/brain-gen.

Read the full abstract

Tuesday 27 January

Audio and Speech Processing · Cross-listed from cs.SD

Graph Modelling Analysis of Speech-Gesture Interaction for Aphasia Severity Estimation

Navya Martin Kollapally, Christa Akers and Renjith Nelson Joseph

Aphasia is an acquired language disorder caused by injury to the regions of the brain that are responsible for language. Aphasia may impair the use and comprehension of written and spoken language. The Western Aphasia Battery-Revised (WAB-R) is an assessment tool administered by speech-language pathologists (SLPs) to evaluate the aphasia type and severity. Because the WAB-R measures isolated linguistic skills, there has been growing interest in the assessment of discourse production as a more holistic representation of everyday language abilities. Recent advancements in speech analysis focus on automated estimation of aphasia severity from spontaneous speech, relying mostly in isolated linguistic or acoustical features. In this work, we propose a graph neural network-based framework for estimating aphasia severity. We represented each participant's discourse as a directed multi-modal graph, where nodes represent lexical items and gestures and edges encode word-word, gesture-word, and word-gesture transitions. GraphSAGE is employed to learn participant-level embeddings, thus integrating information from immediate neighbors and overall graph structure. Our results suggest that aphasia severity is not encoded in isolated lexical distribution, but rather emerges from structured interactions between speech and gesture. The proposed architecture offers a reliable automated aphasia assessment, with possible uses in bedside screening and telehealth-based monitoring.

Read the full abstract

Signal Processing

Transforming Computational Lithography with AC and AI – Faster, More Accurate, and Energy-efficient

Saumyadip Mukhopadhyay, Kiho Yang, Kasyap Thottasserymana Vasudevan et al. (+8 more)

From climate science to drug discovery, scientific computing demands have surged dramatically in recent years – driven by larger datasets, more sophisticated models, and higher simulation fidelity. This growth rate far outpaces transistor scaling, leading to unsustainably rising costs, energy consumption, and emissions. Semiconductor manufacturing is no exception. Computational lithography – involving transferring circuitry to silicon in diffraction-limited conditions – is the largest workload in semiconductor manufacturing. It has also grown exceptionally complex as miniaturization has advanced in the angstrom-era, requiring more accurate modeling, intricate corrections, and broader solution-space exploration. Accelerated computing (AC) offers a solution by dramatically freeing up the compute and power envelope. AI augments these gains by serving as high-fidelity surrogates for compute-intensive steps. Together, they present a sustainable, next-generation computing platform for scientific workloads. This new paradigm needs a fundamental redesign of the software stack. For computational lithography, NVIDIA cuLitho reinvents the core primitives – diffractive optics, computational geometry, multi-variant optimization, data processing – to achieve a transformative 57X end-to-end acceleration. Beyond dramatically faster cycles, this expanded compute envelope enables more rigorous solutions, including curvilinear masks, high-numerical aperture extreme ultraviolet (high-NA EUV) lithography, and subatomic modeling. We reinvest a small fraction of the freed-up compute to include through-focus correction for better process resilience. Silicon experiments at IMEC show significant benefits compared to conventional methods – 35% better process window and 19% better edge placement error. This is the first quantified chip-scale demonstration of the lithography benefits of AC and AI in silicon.

Read the full abstract

All authors Saumyadip Mukhopadhyay, Kiho Yang, Kasyap Thottasserymana Vasudevan, Mounica Jyothi Divvela, Selim Dogru, Dilip Krishnamurthy, Fergo Treska, Werner Gillijns, Ryan Ryoung han Kim, Kumara Sastry, Vivek Singh

Signal Processing

Hybrid Deep Learning Framework for CSI-Based Activity Recognition in Bandwidth-Constrained Wi-Fi Sensing

Alison M. Fernandes, Hermes I. Del Monego, Bruno S. Chang et al. (+3 more)

This paper presents a novel hybrid deep learning framework designed to enhance the robustness of CSI-based Human Activity Recognition (HAR) within bandwidth-constrained Wi-Fi sensing environments. The core of our proposed methodology is a preliminary Doppler trace extraction stage, implemented to amplify salient motion-related signal features before classification. Subsequently, these enhanced inputs are processed by a hybrid neural architecture, which integrates Inception networks responsible for hierarchical spatial feature extraction and Bidirectional Long Short-Term Memory (BiLSTM) networks that capture temporal dependencies. A Support Vector Machine (SVM) is then utilized as the final classification layer to optimize decision boundaries. The framework's efficacy was systematically validated using a public dataset across 20, 40, and 80 MHz bandwidth configurations. The model yielded accuracies of 89.27% (20 MHz), 94.13% (40 MHz), and 95.30% (80 MHz), respectively. These results confirm a marked superiority over standalone deep learning baselines, especially in the most constrained low-bandwidth scenarios. This study underscores the utility of combining Doppler-based feature engineering with a hybrid learning architecture for reliable HAR in bandwidth-limited wireless sensing applications.

Read the full abstract

All authors Alison M. Fernandes, Hermes I. Del Monego, Bruno S. Chang, Anelise Munaretto, Hélder M. Fontes, Rui Campos

Signal Processing

Deep Reinforcement Learning for Interference Suppression in RIS-Aided Space-Air-Ground Integrated Networks

Pujitha Mamillapalli, Shikhar Verma, Tiago Koketsu Rodrigues et al. (+1 more)

Future 6G networks envision ubiquitous connectivity through space-air-ground integrated networks (SAGINs), where high-altitude platform stations (HAPSs) and satellites complement terrestrial systems to provide wide-area, low-latency coverage. However, the rapid growth of terrestrial devices intensifies spectrum sharing between terrestrial and non-terrestrial segments, resulting in severe cross-tier interference. In particular, frequency sharing between the HAPS satellite uplink and HAPS ground downlink improves spectrum efficiency but suffers from interference caused by the HAPS antenna back-lobe. Existing approaches relying on zero-forcing (ZF) codebooks have limited performance under highly dynamic channel conditions. To overcome this limitation, we employ a reconfigurable intelligent surface (RIS)-aided HAPS-based SAGIN framework with a deep deterministic policy gradient (DDPG) algorithm. The proposed DDPG framework optimizes the HAPS beamforming weights to form spatial nulls toward interference sources while maintaining robust links to the desired signals. Simulation results demonstrate that the DDPG framework consistently outperforms conventional ZF beamforming among different RIS configurations, achieving up to \(11.3\%\) throughput improvement for a \(4\times4\) RIS configuration, validating its adaptive capability to enhance spectral efficiency in dynamic HAPS-based SAGINs.

Read the full abstract

All authors Pujitha Mamillapalli, Shikhar Verma, Tiago Koketsu Rodrigues, Abhinav Kumar

Signal Processing · Cross-listed from cs.IT

UAV-Mounted Aerial Relays in Military Communications: A Comprehensive Survey

Faisal Al-Kamali, Francois Chan, Hussein A. Ammar et al. (+2 more)

Relays are pivotal in military communication networks, expanding coverage and ensuring reliable connectivity in challenging operational environments. While traditional terrestrial relays (TR) are constrained by fixed locations and vulnerability to physical obstructions, unmanned aerial vehicle (UAV)-mounted aerial relays (AR) offer a dynamic and flexible alternative by operating above obstacles and adapting to changing battlefield conditions. This paper provides a comprehensive survey of AR systems in military communications, presenting a detailed comparison between AR and TR paradigms and examining two specific AR technologies: active aerial relays (AAR) and aerial reconfigurable intelligent surface (ARIS) relays. The survey delves into their operation, benefits, challenges, and military applications, supported by a qualitative analysis across metrics such as coverage, flexibility, security, and cost. A novel multi-dimensional metric, the mission-critical relay effectiveness score (MCRES), is introduced as a quantitative method for evaluating relay suitability based on mission-specific weights for critical attributes like mobility, jamming resilience, deployment speed, stealth, coverage, and autonomy. Furthermore, we present Algorithm 1, a decision-making framework that leverages the MCRES to guide the systematic selection of the optimal relay type, AR or TR, and subsequently AAR or ARIS, tailored to the unique demands of a given military scenario, such as dynamic battlefield operations, electronic warfare, or covert missions. Finally, the paper addresses current implementation challenges and outlines promising future research directions to advance the deployment of robust and resilient UAV-mounted relay systems in contested military environments.

Read the full abstract

All authors Faisal Al-Kamali, Francois Chan, Hussein A. Ammar, James H. Bayes, Claude D'Amours

Image and Video Processing · Cross-listed from cs.CV

D3R-Net: Dual-Domain Denoising Reconstruction Network for Robust Industrial Anomaly Detection

Dmytro Filatov, Valentyn Fedorov, Vira Filatova et al. (+1 more)

Unsupervised anomaly detection (UAD) is a key ingredient of automated visual inspection in modern manufacturing. The reconstruction-based methods appeal because they have basic architectural design and they process data quickly but they produce oversmoothed results for high-frequency details. As a result, subtle defects are partially reconstructed rather than highlighted, which limits segmentation accuracy. We build on this line of work and introduce D3R-Net, a Dual-Domain Denoising Reconstruction framework that couples a self-supervised 'healing' task with frequency-aware regularization. During training, the network receives synthetically corrupted normal images and is asked to reconstruct the clean targets, which prevents trivial identity mapping and pushes the model to learn the manifold of defect-free textures. In addition to the spatial mean squared error, we employ a Fast Fourier Transform (FFT) magnitude loss that encourages consistency in the frequency domain. The implementation also allows an optional structural similarity (SSIM) term, which we study in an ablation. On the MVTec AD Hazelnut benchmark, D3R-Net with the FFT loss improves localization consistency over a spatial-only baseline: PRO AUC increases from 0.603 to 0.687, while image-level ROC AUC remains robust. Evaluated across fifteen MVTec categories, the FFT variant raises the average pixel ROC AUC from 0.733 to 0.751 and PRO AUC from 0.417 to 0.468 compared to the MSE-only baseline, at roughly 20 FPS on a single GPU. The network is trained from scratch and uses a lightweight convolutional autoencoder backbone, providing a practical alternative to heavy pre-trained feature embedding methods.

Read the full abstract

All authors Dmytro Filatov, Valentyn Fedorov, Vira Filatova, Andrii Zelenchuk

Audio and Speech Processing · Cross-listed from math.HO

Proliferating series by Jean Barraqué: a study and classification in mathematical terms

Isabel Tardón and Pablo Martín-Santamaría

Barraqué's proliferating series give an interesting turn on the concept of classic serialism by creating a new invariant when it comes to constructing the series: rather than the intervals between consecutive notes, what remains unaltered during the construction of the proliferations of the given base series is the permutation of the notes which happens between two consecutive series, that is to say, the transformation of the order of the notes in the series. This presents new possibilities for composers interested in the serial method, given the fact that the variety of intervals obtained by this method is far greater than that of classic serialism. In this manuscript, we will study some unexplored possibilities that the proliferating series offer from a mathematical point of view, which will allow composers to gain much more familiarity with them and potentially result in the creation of pieces that take serialism to the next level.

Read the full abstract

Audio and Speech Processing · Cross-listed from cs.SD

VoxMorph: Scalable Zero-shot Voice Identity Morphing via Disentangled Embeddings

Bharath Krishnamurthy and Ajita Rattani

Morphing techniques generate artificial biometric samples that combine features from multiple individuals, allowing each contributor to be verified against a single enrolled template. While extensively studied in face recognition, this vulnerability remains largely unexplored in voice biometrics. Prior work on voice morphing is computationally expensive, non-scalable, and limited to acoustically similar identity pairs, constraining practical deployment. Moreover, existing sound-morphing methods target audio textures, music, or environmental sounds and are not transferable to voice identity manipulation. We propose VoxMorph, a zero-shot framework that produces high-fidelity voice morphs from as little as five seconds of audio per subject without model retraining. Our method disentangles vocal traits into prosody and timbre embeddings, enabling fine-grained interpolation of speaking style and identity. These embeddings are fused via Spherical Linear Interpolation (Slerp) and synthesized using an autoregressive language model coupled with a Conditional Flow Matching network. VoxMorph achieves state-of-the-art performance, delivering a 2.6x gain in audio quality, a 73% reduction in intelligibility errors, and a 67.8% morphing attack success rate on automated speaker verification systems under strict security thresholds. This work establishes a practical and scalable paradigm for voice morphing with significant implications for biometric security. The code and dataset are available on our project page: https://vcbsl.github.io/VoxMorph/

Read the full abstract

Systems and Control

Control systems for synthetic biology and a case-study in cell fate reprogramming

Domitilla Del Vecchio

This paper gives an overview of the use of control systems engineering in synthetic biology, motivated by applications such as cell therapy and cell fate reprogramming for regenerative medicine. A ubiquitous problem in these and other applications is the ability to control the concentration of specific regulatory factors in the cell accurately despite environmental uncertainty and perturbations. The paper describes the origin of these perturbations and how they affect the dynamics of the biomolecular “plant” to be controlled. A variety of biomolecular control implementations are then introduced to achieve robustness of the plant's output to perturbations and are grouped into feedback and feedforward control architectures. Although sophisticated control laws can be implemented in a computer today, they cannot be necessarily implemented inside the cell via biomolecular processes. This fact constraints the set of feasible control laws to those realizable through biomolecular processes that can be engineered with synthetic biology. After reviewing biomolecular feedback and feedforward control implementations, mostly focusing on the author's own work, the paper illustrates the application of such control strategies to cell fate reprogramming. Within this context, a master regulatory factor needs to be controlled at a specific level inside the cell in order to reprogram skin cells to pluripotent stem cells. The article closes by highlighting on-going challenges and directions of future research for biomolecular control design.

Read the full abstract

Signal Processing

Holographic & Channel-Aware Distributed Detection of a Non-cooperative Target

Domenico Ciuonzo, Alessio Zappone, Marco Di Renzo et al. (+1 more)

This work investigates Distributed Detection (DD) in Wireless Sensor Networks (WSNs), where spatially distributed sensors transmit binary decisions over a shared flat-fading channel. To enhance fusion efficiency, a reconfigurable metasurface is positioned in the near-field of a few receive antennas, enabling a holographic architecture that harnesses large-aperture gains with minimal RF hardware. A generalized likelihood ratio test is derived for fixed metasurface settings, and two low-complexity joint design strategies are proposed to optimize both fusion and metasurface configuration. These suboptimal schemes achieve a balance between performance, complexity, and system knowledge. The goal is to ensure reliable detection of a localized phenomenon at the fusion center, under energy-efficient constraints aligned with IoT requirements. Simulation results validate the effectiveness of the proposed holographic fusion, even under simplified designs.

Read the full abstract

All authors Domenico Ciuonzo, Alessio Zappone, Marco Di Renzo, Ciro D'Elia

Audio and Speech Processing

T-Mimi: A Transformer-based Mimi Decoder for Real-Time On-Phone TTS

Haibin Wu, Bach Viet Do, Naveen Suda et al. (+10 more)

Neural audio codecs provide promising acoustic features for speech synthesis, with representative streaming codecs like Mimi providing high-quality acoustic features for real-time Text-to-Speech (TTS) applications. However, Mimi's decoder, which employs a hybrid transformer and convolution architecture, introduces significant latency bottlenecks on edge devices due to the the compute intensive nature of deconvolution layers which are not friendly for mobile-CPUs, such as the most representative framework XNNPACK. This paper introduces T-Mimi, a novel modification of the Mimi codec decoder that replaces its convolutional components with a purely transformer-based decoder, inspired by the TS3-Codec architecture. This change dramatically reduces on-device TTS latency from 42.1ms to just 4.4ms. Furthermore, we conduct quantization aware training and derive a crucial finding: the final two transformer layers and the concluding linear layers of the decoder, which are close to the waveform, are highly sensitive to quantization and must be preserved at full precision to maintain audio quality.

Read the full abstract

All authors Haibin Wu, Bach Viet Do, Naveen Suda, Julian Chan, Madhavan C R, Gene-Ping Yang, Yi-Chiao Wu, Naoyuki Kanda, Yossef Adi, Xin Lei, Yue Liu, Florian Metze, Yuzong Liu

Image and Video Processing

Orthogonal Plane-Wave Transmit-Receive Isotropic-Focusing Micro-Ultrasound (OPTIMUS) with Bias-Switchable Row-Column Arrays

Darren Dahunsi, Randy Palamar, Tyler Henry et al. (+5 more)

High quality structural volumetric imaging is a challenging goal to achieve with modern ultrasound transducers. Matrix probes have limited fields of view and element counts, whereas row-column arrays (RCAs) provide insufficient focusing. In contrast, Top-Orthogonal-to-Bottom-Electrode (TOBE) arrays, also known as bias-switchable RCAs can enable isotropic focusing on par with ideal matrix probes, with a field of view surpassing conventional RCAs. Orthogonal Plane-Wave Transmit-Receive Isotropic-Focusing Micro-Ultrasound (OPTIMUS) is a novel imaging scheme that can use TOBE arrays to achieve nearly isotropic focusing throughout an expansive volume. This approach extends upon a similar volumetric imaging scheme, Hadamard Encoded Row Column Ultrasonic Expansive Scanning (HERCULES), that is even able to image beyond the shadow of the aperture, much like typical 2D matrix probes. We simulate a grid of scatterers to evaluate how the resolution varies across the volume, and validate these simulations experimentally using a commercial calibration phantom. Experimental measurements were done with a custom fabricated TOBE array, custom biasing electronics, and a research ultrasound system. Finally we performed ex-vivo imaging to assess our ability to discern structural tissue information.

Read the full abstract

All authors Darren Dahunsi, Randy Palamar, Tyler Henry, Mohammad Rahim Sobhani, Negar Majidi, Joy Wang, Afshin Kashani Ilkhechi, Roger Zemp

Signal Processing

Electromagnetically Consistent Bounds on Information Transfer in Real-World RIS-Parametrized Wireless Channels

Albert Salmi, Ville Viikari and Philipp del Hougne

A reconfigurable intelligent surface (RIS) endows a wireless channel with programmability that can be leveraged to optimize wireless information transfer. While many works study algorithms for optimizing such a programmable channel, relatively little is known about fundamental bounds on the achievable information transfer. In particular, non-trivial bounds that are both electromagnetically consistent (e.g., aware of mutual coupling) and in line with realistic hardware constraints (e.g., few-bit-programmable, potentially lossy loads) are missing. Here, based on a rigorous multiport network model of a single-input single-output (SISO) channel parametrized by 1-bit-programmable RIS elements, we apply a semidefinite relaxation (SDR) to derive a fundamental bound on the achievable SISO channel gain enhancement. A bound on the maximum achievable rate of information transfer at a given noise level follows directly from Shannon's theorem. We apply our bound to several numerical and experimental examples of different RIS-parametrized radio environments. Compared to electromagnetically consistent benchmark bounding strategies (a norm-inequality bound and, where applicable, a relaxation to an idealized beyond-diagonal load network for which a global solution exists), we consistently observe that our SDR-based bound is notably tighter. We reach at least 64 % (but often 100 %) of our SDR-based bound with standard discrete optimization techniques. The applicability of our bound to concrete experimental systems makes it valuable to inform wireless practitioners, e.g., to evaluate RIS hardware design choices and algorithms to optimize the RIS configuration. Our work contributes to the development of an electromagnetic information theory for RIS-parametrized channels as well as other programmable wave systems such as dynamic metasurface antennas or real-life beyond-diagonal RISs.

Read the full abstract

Systems and Control

OptAgent: an Agentic AI framework for Intelligent Building Operations

Zixin Jiang, Weili Xu and Bing Dong

The urgent need for building decarbonization calls for a paradigm shift in future autonomous building energy operation, from human-intensive engineering workflows toward intelligent agents that interact with physics-grounded digital environments. This study proposes an end-to-end agentic AI-enabled Physics-Informed Machine Learning (PIML) environment for scalable building energy modeling, simulation, control, and automation. The framework consists of (1) a modular and physics-consistent PIML digital environment spanning building thermal dynamics, Heating, Ventilation, and Air Conditioning (HVAC), and distributed energy resources (DER) for grid-interactive energy management; and (2) an agentic AI layer with 11 specialist agents and 72 Model Context Protocol (MCP) tools that enable end-to-end execution of multi-step energy analytics. A representative case study demonstrates multi-domain, multi-agent coordination for assessing how system and control upgrades affect energy use, operating cost, thermal comfort, and flexibility. In addition, a large-scale benchmark (about 4000 runs) systematically evaluates workflow performance in terms of accuracy, token consumption, execution time, and inference cost. The results quantify the impacts of intelligence mode design, model size, task complexity, and orchestrator-specialist coordination, and provide key lessons for building future agentic AI systems in real-world building energy applications. This work establishes a scalable, physics-grounded foundation for deploying agentic AI in decarbonized and grid-interactive building operations.

Read the full abstract

Audio and Speech Processing

Do we really need Self-Attention for Streaming Automatic Speech Recognition?

Youness Dkhissi, Valentin Vielzeuf, Elys Allesiardo et al. (+1 more)

Transformer-based architectures are the most used architectures in many deep learning fields like Natural Language Processing, Computer Vision or Speech processing. It may encourage the direct use of Transformers in the constrained tasks, without questioning whether it will yield the same benefits as in standard tasks. Given specific constraints, it is essential to evaluate the relevance of transformer models. This work questions the suitability of transformers for specific domains. We argue that the high computational requirements and latency issues associated with these models do not align well with streaming applications. Our study promotes the search for alternative strategies to improve efficiency without sacrificing performance. In light of this observation, our paper critically examines the usefulness of transformer architecture in such constrained environments. As a first attempt, we show that the computational cost for Streaming Automatic Speech Recognition (ASR) can be reduced using deformable convolution instead of Self-Attention. Furthermore, we show that Self-Attention mechanisms can be entirely removed and not replaced, without observing significant degradation in the Word Error Rate.

Read the full abstract

All authors Youness Dkhissi, Valentin Vielzeuf, Elys Allesiardo, Anthony Larcher

Audio and Speech Processing

VoxPrivacy: A Benchmark for Evaluating Interactional Privacy of Speech Language Models

Yuxiang Wang, Hongyu Liu, Dekun Chen et al. (+2 more)

As Speech Language Models (SLMs) transition from personal devices to shared, multi-user environments such as smart homes, a new challenge emerges: the model is expected to distinguish between users to manage information flow appropriately. Without this capability, an SLM could reveal one user's confidential schedule to another, a privacy failure we term interactional privacy. Thus, the ability to generate speaker-aware responses becomes essential for SLM safe deployment. Current SLM benchmarks test dialogue ability but overlook speaker identity. Multi-speaker benchmarks check who said what without assessing whether SLMs adapt their responses. Privacy benchmarks focus on globally sensitive data (e.g., bank passwords) while neglecting contextual privacy-sensitive information (e.g., a user's private appointment). To address this gap, we introduce VoxPrivacy, the first benchmark designed to evaluate interactional privacy in SLMs. VoxPrivacy spans three tiers of increasing difficulty, from following direct secrecy commands to proactively protecting privacy. Our evaluation of nine SLMs on a 32-hour bilingual dataset reveals a widespread vulnerability: most open-source models perform close to random chance (around 50% accuracy) on conditional privacy decisions, while even strong closed-source systems fall short on proactive privacy inference. We further validate these findings on Real-VoxPrivacy, a human-recorded subset, confirming that failures observed on synthetic data persist in real speech. Finally, we demonstrate a viable path forward: by fine-tuning on a new 4,000-hour training set, we improve privacy-preserving abilities while maintaining robustness. To support future work, we release the VoxPrivacy benchmark, the large-scale training set, and the fine-tuned model to foster the development of safer and more context-aware SLMs.

Read the full abstract

All authors Yuxiang Wang, Hongyu Liu, Dekun Chen, Xueyao Zhang, Zhizheng Wu

Systems and Control · Cross-listed from cs.LG

Probabilistic Sensing: Intelligence in Data Sampling

Ibrahim Albulushi, Saleh Bunaiyan, Suraj S. Cheema et al. (+2 more)

Extending the intelligence of sensors to the data-acquisition process - deciding whether to sample or not - can result in transformative energy-efficiency gains. However, making such a decision in a deterministic manner involves risk of losing information. Here we present a sensing paradigm that enables making such a decision in a probabilistic manner. The paradigm takes inspiration from the autonomous nervous system and employs a probabilistic neuron (p-neuron) driven by an analog feature extraction circuit. The response time of the system is on the order of microseconds, over-coming the sub-sampling-rate response time limit and enabling real-time intelligent autonomous activation of data-sampling. Validation experiments on active seismic survey data demonstrate lossless probabilistic data acquisition, with a normalized mean squared error of 0.41%, and 93% saving in the active operation time of the system and the number of generated samples.

Read the full abstract

All authors Ibrahim Albulushi, Saleh Bunaiyan, Suraj S. Cheema, Hesham ElSawy, Feras Al-Dirini

Signal Processing

Generative Latent Alignment for Interpretable Radar Based Occupancy Detection in Ambient Assisted Living

Huy Trinh

In this work, we study how to make mmWave radar presence detection more interpretable for Ambient Assisted Living (AAL) settings, where camera-based sensing raises privacy concerns. We propose a Generative Latent Alignment (GLA) framework that combines a lightweight convolutional variational autoencoder with a frozen CLIP text encoder to learn a low-dimensional latent representation of radar Range-Angle (RA) heatmaps. The latent space is softly aligned with two semantic anchors corresponding to "empty room" and "person present", and Grad-CAM is applied in this aligned latent space to visualize which spatial regions support each presence decision. On our mmWave radar dataset, we qualitatively observe that the "person present" class produces compact Grad-CAM blobs that coincide with strong RA returns, whereas "empty room" samples yield diffuse or no evidence. We also conduct an ablation study using unrelated text prompts, which degrades both reconstruction and localization, suggesting that radar-specific anchors are important for meaningful explanations in this setting.

Read the full abstract

Systems and Control

A Latent Space Framework for Modeling Transient Engine Emissions Using Joint Embedding Predictive Architectures

Ganesh Sundaram, Tobias Gehra, Jonas Ulmen et al. (+3 more)

Accurately modeling and controlling vehicle exhaust emissions during transient events, such as rapid acceleration, is critical for meeting environmental regulations and optimizing powertrains. Conventional data-driven methods, such as Multilayer Perceptrons (MLPs) and Long Short-Term Memory (LSTM) networks, improve upon phenomenological models but often struggle with the complex nonlinear dynamics of emission formation. These monolithic architectures are sensitive to dataset variability and typically require deep, computationally expensive structures to perform well, limiting their practical utility. This paper introduces a novel approach that overcomes these limitations by modeling emission dynamics within a structured latent space. Leveraging a Joint Embedding Predictive Architecture (JEPA), the proposed framework learns from a rich dataset that combines real-world Portable Emission Measurement System (PEMS) data with high-frequency hardware-in-the-loop measurements. The model abstracts away irrelevant noise, encoding only the key factors governing emission behavior into a compact, robust representation. This results in superior data efficiency and predictive accuracy across diverse transient regimes, significantly outperforming high-performing LSTM baselines in generalization. To ensure suitability for real-world deployment, the JEPA framework is structured to support pruning and post-training quantization. This strategy drastically reduces the computational footprint, minimizing inference time and memory demand with negligible accuracy loss. The result is a highly efficient model ideal for on-board implementation of advanced strategies, such as model predictive control or model-based reinforcement learning, in conventional and hybrid powertrains. These findings offer a clear pathway toward more robust emission control systems for next-generation vehicles.

Read the full abstract

All authors Ganesh Sundaram, Tobias Gehra, Jonas Ulmen, Mirjan Heubaum, Daniel Görges, Michael Günthner

Systems and Control · Cross-listed from math.NA

A refined nonlinear least-squares method for the rational approximation problem

Michael S. Ackermann, Linus Balicki, Serkan Gugercin et al. (+1 more)

The adaptive Antoulas-Anderson (AAA) algorithm for rational approximation is a widely used method for the efficient construction of highly accurate rational approximations to given data. While AAA can often produce rational approximations accurate to any prescribed tolerance, these approximations may have degrees larger than what is actually required to meet the given tolerance. In this work, we consider the adaptive construction of interpolating rational approximations while aiming for the smallest feasible degree to satisfy a given error tolerance. To this end, we introduce refinement approaches to the linear least-squares step of the classical AAA algorithm that aim to minimize the true nonlinear least-squares error with respect to the given data. Furthermore, we theoretically analyze the derived approaches in terms of the corresponding gradients from the resulting minimization problems and use these insights to propose a new greedy framework that ensures monotonic error convergence. Numerical examples from function approximation and model order reduction verify the effectiveness of the proposed algorithm to construct accurate rational approximations of small degrees.

Read the full abstract

All authors Michael S. Ackermann, Linus Balicki, Serkan Gugercin, Steffen W. R. Werner

Systems and Control · Cross-listed from cs.LG

Component-Aware Pruning Framework for Neural Network Controllers via Gradient-Based Importance Estimation

Ganesh Sundaram, Jonas Ulmen and Daniel Görges

The transition from monolithic to multi-component neural architectures in advanced neural network controllers poses substantial challenges due to the high computational complexity of the latter. Conventional model compression techniques for complexity reduction, such as structured pruning based on norm-based metrics to estimate the relative importance of distinct parameter groups, often fail to capture functional significance. This paper introduces a component-aware pruning framework that utilizes gradient information to compute three distinct importance metrics during training: Gradient Accumulation, Fisher Information, and Bayesian Uncertainty. Experimental results with an autoencoder and a TD-MPC agent demonstrate that the proposed framework reveals critical structural dependencies and dynamic shifts in importance that static heuristics often miss, supporting more informed compression decisions.

Read the full abstract