Systems and Control
Simple and Combination Parametric Resonances of an Electromagnetically Suspended Vehicle subject to Base Excitation
Jithu Paul, Karel N. van Dalen, Andrei B. Faragau et al. (+3 more)
This paper investigates the dynamic stability of an electromagnetically suspended vehicle, encountered in Hyperloop and Maglev systems, subject to periodic excitations caused by surface irregularities or vibration of the support induced by external noise. The narrow clearance between the vehicle and the support can make it highly sensitive to small oscillations, since the admissible amplitudes of the vehicle oscillations can be comparable to external excitation amplitude. The vehicle is modelled as a three-degree-of-freedom model where the vehicle is suspended via two identical electromagnetic actuators from a rigid support that oscillates. The governing equations are derived using force and torque balances, incorporating nonlinear electromagnetic forces, and Kirchhoffs law for the electromagnets with PD control strategy on the airgap. The equations of motion are linearized around the steady state induced by the surface oscillation, yielding a system with time-periodic coefficients. We analytically explore both principal and combination parametric resonances using an extended Hills method, and Floquet theory is used for numerical validation. The stability boundaries are obtained as ellipses in control gain parameter space, and the influence of system parameters on these boundaries is characterized. For the principal parametric resonance, the ratio of the sizes of the two obtained ellipses is three to one, whereas for the combination parametric resonance, the ratio is fourteen to one. When all ellipses are simultaneously present, one of the ellipses associated with the combination parametric resonance is the largest.
Read the full abstract ▾
All authors Jithu Paul, Karel N. van Dalen, Andrei B. Faragau, Rens J. van Leijden, Biagio Carboni, Andrei V. Metrikine
Audio and Speech Processing · Cross-listed from cs.LG
Mitigating Sample-Level Imbalance via Probabilistic Separation for Adaptive Multimodal Fusion
Zhiwen Yu, Zhaocheng Liu, Xiaoqing Liu et al. (+2 more)
Multimodal learning faces modality imbalance, where dominant modalities suppress weaker ones due to inconsistent convergence rates. Existing static or heuristic methods overlook sample-level variations in prediction bias and fail to isolate low-quality outlier samples. To address this, we propose a novel framework to quantitatively diagnose and dynamically mitigate modality imbalance at the sample level. We first introduce a Modality Gap metric to quantify prediction discrepancies between unimodal branches. Empirical analysis reveals a distinct bimodal distribution, reflecting the natural coexistence of balanced and imbalanced sample subgroups. We then employ a Gaussian Mixture Model (GMM) to model this gap distribution, leveraging
Bayesian posterior probabilities for probabilistic soft separation of subgroups. Next, we construct a two-stage training framework comprising a Warm-up stage and an Adaptive Training stage. In the Adaptive Training stage, a GMM-guided Adaptive Loss dynamically reallocates optimization priorities, imposing stronger modality
alignment penalties on imbalanced samples while prioritizing
multimodal fusion for balanced ones. Experimental results demonstrate that our method significantly outperforms current state-of-the-art baselines. Furthermore, fine-tuning on a high-quality balanced subset filtered by the GMM serves as an effective data purification strategy.
Read the full abstract ▾
All authors Zhiwen Yu, Zhaocheng Liu, Xiaoqing Liu, Huanqiang Zeng, C. L. Philip Chen
Image and Video Processing · Cross-listed from cs.CV
2D_3D Feature Fusion via Cross-Modal Latent Synthesis and Attention Guided Restoration for Industrial Anomaly Detection
Usman Ali, Ali Zia, Abdul Rehman et al. (+5 more)
Industrial anomaly detection (IAD) increasingly benefits from integrating 2D and 3D data, but robust cross-modal fusion remains challenging. We propose a novel unsupervised framework, Multi-Modal
Attention-Driven Fusion Restoration (MAFR), which synthesises a unified latent space from RGB images and point clouds using a shared fusion encoder, followed by
attention-guided, modality-specific decoders. Anomalies are localised by measuring reconstruction errors between input features and their restored counterparts. Evaluations on the MVTec 3D-AD and Eyecandies
benchmarks demonstrate that MAFR achieves state-of-the-art results, with a mean I-AUROC of 0.972 and 0.901, respectively. The framework also exhibits strong performance in few-shot learning settings, and ablation studies confirm the critical roles of the fusion architecture and composite loss. MAFR offers a principled approach for fusing visual and
geometric information, advancing the robustness and accuracy of industrial anomaly detection. Code is available at https://github.com/adabrh/MAFR
Read the full abstract ▾
All authors Usman Ali, Ali Zia, Abdul Rehman, Umer Ramzan, Zohaib Hassan, Talha Sattar, Jing Wang, Wei Xiang
Audio and Speech Processing
Hearing Health in Home Healthcare: Leveraging LLMs for Illness Scoring and ALMs for Vocal Biomarker Extraction
Yu-Wen Chen, William Ho, Sasha M. Vergez et al. (+8 more)
The growing demand for home healthcare calls for tools that can support care delivery. In this study, we explore automatic health assessment from voice using real-world home care visit data, leveraging the diverse patient information it contains. First, we utilize Large
Language Models (LLMs) to integrate Subjective, Objective, Assessment, and Plan (SOAP) notes derived from unstructured audio transcripts and structured vital signs into a holistic illness score that reflects a patient's overall health. This compact representation facilitates cross-visit health status comparisons and downstream analysis. Next, we design a multi-stage preprocessing pipeline to extract short
speech segments from target speakers in home care recordings for acoustic analysis. We then employ an Audio
Language Model (ALM) to produce plain-
language descriptions of vocal biomarkers and examine their association with individuals' health status. Our experimental results
benchmark both commercial and open-source LLMs in estimating illness scores, demonstrating their
alignment with actual clinical outcomes, and revealing that SOAP notes are substantially more informative than vital signs. Building on the illness scores, we provide the first evidence that ALMs can identify health-related acoustic patterns from home care recordings and present them in a human-readable form. Together, these findings highlight the potential of LLMs and ALMs to harness heterogeneous in-home visit data for better patient monitoring and care.
Read the full abstract ▾
All authors Yu-Wen Chen, William Ho, Sasha M. Vergez, Grace Flaherty, Pallavi Gupta, Zhihong Zhang, Maryam Zolnoori, Margaret V. McDonald, Maxim Topaz, Zoran Kostic, Julia Hirschberg
Systems and Control · Cross-listed from cs.RO
ANGEL: A Novel Gripper for Versatile and Light-touch Fruit Harvesting
Dharmik Patel, Antonio Rafael Vazquez Pantoja, Jiuzhou Lei et al. (+3 more)
Fruit harvesting remains predominantly a labor-intensive process, motivating the development of research for robotic grippers. Conventional rigid or
vacuum-driven grippers require complex mechanical design or high energy consumption. Current enveloping-based fruit harvesting grippers lack adaptability to fruits of different sizes. This paper introduces a drawstring-inspired, cable-driven soft gripper for versatile and gentle fruit harvesting. The design employs 3D-printed Thermoplastic Polyurethane (TPU) pockets with integrated steel wires that constrict around the fruit when actuated, distributing pressure uniformly to minimize bruising and allow versatility to fruits of varying sizes. The lightweight structure, which requires few components, reduces mechanical complexity and cost compared to other grippers. Actuation is achieved through servo-driven cable control, while motor feedback provides autonomous grip adjustment with tunable grip strength. Experimental validation shows that, for tomatoes within the gripper's effective size range, harvesting was achieved with a 0% immediate damage rate and a bruising rate of less than 9% after five days, reinforcing the gripper's suitability for fruit harvesting.
Read the full abstract ▾
All authors Dharmik Patel, Antonio Rafael Vazquez Pantoja, Jiuzhou Lei, Kiju Lee, Xiao Liang, Minghui Zheng
Systems and Control · Cross-listed from cs.LG
Provably Optimal Reinforcement Learning under Safety Filtering
Donggeon David Oh, Duy P. Nguyen, Haimin Hu et al. (+1 more)
Recent advances in reinforcement learning (RL) enable its use on increasingly complex tasks, but the lack of formal safety guarantees still limits its application in safety-critical settings. A common practical approach is to augment the RL policy with a safety filter that overrides unsafe actions to prevent failures during both training and deployment. However, safety filtering is often perceived as sacrificing performance and hindering the learning process. We show that this perceived safety-performance tradeoff is not inherent and prove, for the first time, that enforcing safety with a sufficiently permissive safety filter does not degrade asymptotic performance. We formalize RL safety with a safety-critical Markov decision process (SC-MDP), which requires categorical, rather than high-probability, avoidance of catastrophic failure states. Additionally, we define an associated filtered MDP in which all actions result in safe effects, thanks to a safety filter that is considered to be a part of the environment. Our main
theorem establishes that (i) learning in the filtered MDP is safe categorically, (ii) standard RL convergence carries over to the filtered MDP, and (iii) any policy that is optimal in the filtered MDP, when executed through the same filter, achieves the same asymptotic return as the best safe policy in the SC-MDP, yielding a complete separation between safety enforcement and performance optimization. We validate the theory on Safety Gymnasium with representative tasks and constraints, observing zero violations during training and final performance matching or exceeding unfiltered baselines. Together, these results shed light on a long-standing question in safety-filtered learning and provide a simple, principled recipe for safe RL: train and deploy RL policies with the most permissive safety filter that is available.
Read the full abstract ▾
All authors Donggeon David Oh, Duy P. Nguyen, Haimin Hu, Jaime Fernández Fisac
Systems and Control · Cross-listed from cs.CY
Prompt-to-Primal Teaching
Euzeli dos Santos
This paper introduces
Prompt-to-Primal (P2P) Teaching, an AI-integrated instructional approach that links
prompt-driven exploration with first-principles
reasoning, guided and moderated by the instructor within the classroom setting. In P2P teaching, student-generated AI
prompts serve as entry points for inquiry and initial discussions in class, while the instructor guides learners to validate, challenge, and reconstruct AI responses through fundamental physical and mathematical laws. The approach encourages self-reflective development, critical evaluation of AI outputs, and conceptual foundational knowledge of the core engineering principles. A large
language model (LLM) can be a highly effective tool for those who already possess foundational knowledge of a subject; however, it may also mislead students who lack sufficient background in the subject matter. Results from two student cohorts across different semesters suggest the pedagogical effectiveness of the P2P teaching framework in enhancing both AI literacy and engineering
reasoning.
Read the full abstract ▾
Image and Video Processing · Cross-listed from cs.CV
TriggerNet: A Novel Explainable AI Framework for Red Palm Mite Detection and Multi-Model Comparison and Heuristic-Guided Annotation
Harshini Suresha and Kavitha SH
The red palm mite infestation has become a serious concern, particularly in regions with extensive palm cultivation, leading to reduced productivity and economic losses. Accurate and early identification of mite-infested plants is critical for effective management. The current study focuses on evaluating and comparing the ML model for classifying the affected plants and detecting the infestation. TriggerNet is a novel interpretable AI framework that integrates Grad-CAM, RISE, FullGrad, and TCAV to generate novel visual explanations for deep learning models in plant classification and disease detection. This study applies TriggerNet to address red palm mite (Raoiella indica) infestation, a major threat to palm cultivation and agricultural productivity. A diverse set of RGB images across 11 plant species, Arecanut, Date Palm, Bird of Paradise, Coconut Palm, Ginger, Citrus Tree, Palm Oil, Orchid, Banana Palm, Avocado Tree, and Cast Iron Plant was utilized for training and evaluation. Advanced deep learning models like CNN, EfficientNet, MobileNet, ViT, ResNet50, and InceptionV3, alongside machine learning classifiers such as Random Forest, SVM, and KNN, were employed for plant classification. For disease classification, all plants were categorized into four classes: Healthy, Yellow Spots, Reddish Bronzing, and Silk Webbing. Snorkel was used to efficiently label these disease classes by leveraging heuristic rules and patterns, reducing manual annotation time and improving
dataset reliability.
Read the full abstract ▾
Audio and Speech Processing · Cross-listed from cs.SD
Transformer Redesign for Late Fusion of Audio-Text Features on Ultra-Low-Power Edge Hardware
Stavros Mitsis, Ermos Hadjikyriakos, Humaid Ibrahim et al. (+4 more)
Deploying
emotion recognition systems in real-world environments where devices must be small, low-power, and private remains a significant challenge. This is especially relevant for applications such as tension monitoring, conflict de-escalation, and responsive wearables, where cloud-based solutions are impractical.
Multimodal emotion recognition has advanced through deep learning, but most systems remain unsuitable for deployment on ultra-constrained edge devices. Prior work typically relies on powerful hardware, lacks real-time performance, or uses unimodal input. This paper addresses that gap by presenting a hardware-aware
emotion recognition system that combines acoustic and linguistic features using a late-fusion architecture optimised for Edge TPU. The design integrates a quantised
transformer-based acoustic model with frozen keyword
embeddings from a DSResNet-SE network, enabling real-time inference within a 1.8MB
memory budget and 21-23ms latency. The pipeline ensures spectrogram
alignment between training and deployment using MicroFrontend and MLTK. Evaluation on re-recorded, segmented IEMOCAP samples captured through the Coral Dev Board Micro microphone shows a 6.3% macro F1 improvement over unimodal baselines. This work demonstrates that accurate, real-time
multimodal emotion inference is achievable on microcontroller-class edge platforms through task-specific fusion and hardware-guided model design.
Read the full abstract ▾
All authors Stavros Mitsis, Ermos Hadjikyriakos, Humaid Ibrahim, Savvas Neofytou, Shashwat Raman, James Myles, Eiman Kanjo
Signal Processing
Majority Vote Compressed Sensing
Henrik Hellström, Jiwon Jeong, Ayfer Özgür et al. (+2 more)
We consider the problem of non-coherent over-the-air computation (AirComp), where $n$ devices carry high-dimensional data vectors $\mathbf{x}_i\in\mathbb{R}^d$ of sparsity $\lVert\mathbf{x}_i\rVert_0\leq k$ whose sum has to be computed at a receiver. Previous results on non-coherent AirComp require more than $d$ channel uses to compute functions of $\mathbf{x}_i$, where the extra redundancy is used to combat non-coherent signal aggregation. However, if the data vectors are sparse, sparsity can be exploited to offer significantly cheaper communication. In this paper, we propose to use random transforms to transmit lower-dimensional projections $\mathbf{s}_i\in\mathbb{R}^T$ of the data vectors. These projected vectors are communicated to the receiver using a majority vote (MV)-AirComp scheme, which estimates the bit-vector corresponding to the signs of the aggregated projections, i.e., $\mathbf{y} = \text{sign}(\sum_i\mathbf{s}_i)$. By leveraging 1-bit compressed sensing (1bCS) at the receiver, the real-valued and high-dimensional aggregate $\sum_i\mathbf{x}_i$ can be recovered from $\mathbf{y}$. We prove analytically that the proposed MVCS scheme estimates the aggregated data vector $\sum_i \mathbf{x}_i$ with $\ell_2$-norm error $\epsilon$ in $T=\mathcal{O}(kn\log(d)/\epsilon^2)$ channel uses. Moreover, we specify
algorithms that leverage MVCS for histogram estimation and distributed machine learning. Finally, we provide numerical evaluations that reveal the advantage of MVCS compared to the state-of-the-art.
Read the full abstract ▾
All authors Henrik Hellström, Jiwon Jeong, Ayfer Özgür, Viktoria Fodor, Carlo Fischione
Systems and Control
An Exact Quantile-Energy Equality for Terminal Halfspaces in Linear-Gaussian Control with a Discrete-Time Companion, KL/Schrodinger Links, and High-Precision Validation
Sandro Andric
We prove an exact equality between the minimal quadratic control energy and the squared normal-
quantile gap for terminal halfspaces in linear-Gaussian systems with additive control and quadratic effort $E(u) = \tfrac12\!\int u^\top M u\,dt$ where $M = B^\top\Sigma^{-1}B$. For terminal halfspace events, the minimal energy equals the squared normal-
quantile gap divided by twice a controllability-to-noise ratio $R_T^2(w)=(w^\top W_c^M w)/(w^\top V_T w)$ and is attained by a matched-filter control. We provide an exact zero-order-hold discrete-time companion via block exponentials, relate the result to minimum-energy control, Gaussian isoperimetry, risk-sensitive/KL control, and Schrodinger bridges, and validate to high precision with Monte Carlo. We state assumptions, singular-$M$ handling, and edge cases. The statement is a compact synthesis and design-ready translator, not a universal principle. Novelty: while the ingredients (Gramians, Cauchy-Schwarz, Gaussian isoperimetry) are classical, to our knowledge the explicit
quantile-energy equality with a constructive matched-filter achiever for terminal halfspaces, and its discrete-time companion, are not recorded together in the cited literature.
Read the full abstract ▾
Systems and Control
Admittance Matrix Concentration Inequalities for Understanding Uncertain Power Networks
Samuel Talkington, Cameron Khanpour, Rahul K. Gupta et al. (+5 more)
This paper presents conservative probabilistic bounds for the spectrum of the admittance
matrix and classical linear power flow models under uncertain network parameters; for example, probabilistic line contingencies. Our proposed approach imports tools from probability theory, such as concentration inequalities for random
matrices. This provides a theoretical framework for understanding error bounds of common approximations of the AC power flow equations under parameter uncertainty, including the DC and LinDistFlow approximations. Additionally, we show that the upper bounds scale as functions of nodal criticality. This network-theoretic quantity captures how uncertainty concentrates at critical nodes for use in contingency analysis. We validate these bounds on IEEE test networks, demonstrating that they correctly capture the scaling behavior of spectral perturbations up to conservative constants.
Read the full abstract ▾
All authors Samuel Talkington, Cameron Khanpour, Rahul K. Gupta, Sergio A. Dorado-Rojas, Daniel Turizo, Hyeongon Park, Dmitrii M. Ostrovskii, Daniel K. Molzahn
Audio and Speech Processing
AnyRIR: Robust Non-intrusive Room Impulse Response Estimation in the Wild
Kyung Yun Lee, Nils Meyer-Kahlen, Karolina Prawda et al. (+2 more)
We address the problem of estimating room impulse responses (RIRs) in noisy, uncontrolled environments where non-stationary sounds such as
speech or footsteps corrupt conventional deconvolution. We propose AnyRIR, a non-intrusive method that uses music as the excitation signal instead of a dedicated test signal, and formulate RIR estimation as an L1-norm regression in the time-frequency domain. Solved efficiently with Iterative Reweighted Least Squares (IRLS) and Least-Squares Minimal Residual (LSMR) methods, this approach exploits the sparsity of non-stationary noise to suppress its influence. Experiments on simulated and measured data show that AnyRIR outperforms L2-based and frequency-domain deconvolution, under in-the-wild noisy scenarios and codec mismatch, enabling robust RIR estimation for AR/VR and related applications.
Read the full abstract ▾
All authors Kyung Yun Lee, Nils Meyer-Kahlen, Karolina Prawda, Vesa Välimäki, Sebastian J. Schlecht
Signal Processing
Sample Complexity Analysis of Multi-Target Detection via Markovian and Hard-Core Multi-Reference Alignment
Kweku Abraham, Amnon Balanov, Tamir Bendory et al. (+1 more)
Motivated by single-particle cryo-electron microscopy, we study the sample complexity of the multi-target detection (MTD) problem, in which an unknown signal appears multiple times at unknown locations within a long, noisy observation. We propose a patching scheme that reduces MTD to a non-i.i.d. multi-reference
alignment (MRA) model. In the one-dimensional setting, the latent group elements form a Markov chain, and we show that the convergence rate of any
estimator matches that of the corresponding i.i.d. MRA model, up to a logarithmic factor in the number of patches. Moreover, for
estimators based on empirical averaging, such as the method of moments, the convergence rates are identical in both settings. We further establish an analogous result in two dimensions, where the latent structure arises from an exponentially mixing random field generated by a hard-core placement model. As a consequence, if the signal in the corresponding i.i.d. MRA model is determined by moments up to order $n_{\min}$, then in the low-SNR regime the number of patches required to estimate the signal in the MTD model scales as $\sigma^{2n_{\min}}$, where $\sigma^2$ denotes the noise variance.
Read the full abstract ▾
All authors Kweku Abraham, Amnon Balanov, Tamir Bendory, Carlos Esteve-Yagüe
Systems and Control
Data-driven Communication and Control Design for Distributed Frequency Regulation with Black-box Inverters
Michael Nestor, Jiaxin Wang, Ning Zhang et al. (+1 more)
The increasing penetration of inverter-based resources into the power grid, with often only black-box models available, challenges long-standing frequency control methods. Most recent works take a decentralized approach without online device coordination via communication. This paper considers both dynamic behavior and communication within secondary frequency control on an intermediate timescale. We develop a distributed data-driven approach that utilizes peer-to-peer communication between inverters to avoid the need for a central control center. To enable a trade off between communication network requirements and control performance, we present a framework to guide communication
topology design for secondary frequency regulation. Following design of the inter-
agent information exchange scheme, we design a controller that is structured according to the communication
topology with a closed-loop stability guarantee. Case studies on the IEEE 39-bus system validate the framework and illustrate the trade-off between communication requirements and control performance that is enabled by our approach.
Read the full abstract ▾
All authors Michael Nestor, Jiaxin Wang, Ning Zhang, Fei Teng
Systems and Control
Trajectory Optimization for Minimum Threat Exposure using Physics-Informed Neural Networks
Alexandra E. Ballentine and Raghvendra V. Cowlagi
We apply a physics-informed neural network (PINN) to solve the two-point boundary value problem (BVP) arising from the necessary conditions postulated by Pontryagin's Minimum Principle for optimal control. Such BVPs are known to be numerically difficult to solve by traditional shooting methods due to extremely high sensitivity to initial guesses. In the light of recent successes in applying PINNs for solving high-dimensional differential equations, we develop a PINN to solve the problem of finding trajectories with minimum exposure to a spatiotemporal threat for a vehicle kinematic model. First, we implement PINNs that are trained to solve the BVP for a given pair of initial and final states for a given threat field. Next, we implement a PINN conditioned on the initial state for a given threat field, which eliminates the need for retraining for each initial state. We demonstrate that the PINN outputs satisfy the necessary conditions with low numerical error.
Read the full abstract ▾
Signal Processing
Precoding for Uplink RIS-Assisted Cell-Free MIMO-OFDM Systems with Hardware Impairments
Navid Reyhanian, Reza Ghaderi Zefreh, Parisa Ramezani et al. (+1 more)
This paper studies a reconfigurable intelligent surface (RIS)-assisted
cell-free massive multiple-input multiple-output (CF-mMIMO) system with multiple RISs. Joint design of transmit precoding, RIS coefficients, and receive combining is investigated for uplink sum-rate maximization under in-phase and quadrature phase imbalance (IQI) at user equipments (UEs) and access points (APs). A weighted minimum mean squared error (WMMSE) based block coordinate descent (BCD) approach is proposed, where novel iterative methods are developed to efficiently solve the BCD subproblems. The efficiency of proposed approaches is demonstrated relative to heuristic methods via extensive simulations.
Read the full abstract ▾
All authors Navid Reyhanian, Reza Ghaderi Zefreh, Parisa Ramezani, Emil Bjornson
Signal Processing
Beam Index Map Prediction in Unseen Environments from Geospatial Data
Fabian Jaensch, Giuseppe Caire and Begüm Demir
In 5G, beam training consists of the efficient association of users to beams for a given beamforming codebook used at the base station and the given propagation environment in the
cell. We propose a convolutional neural network approach that leverages the position of the base station and geospatial data to predict beam distributions for all user locations simultaneously. Our method generalizes to unseen environments without site-specific training or specialized sensors. The results show that it significantly reduces the number of candidate beams considered, thereby improving the efficiency of beam training.
Read the full abstract ▾
Signal Processing
Semantic Joint Source Channel Coding for Distributed Subsurface Imaging in Multi-Agent Systems
Maximilian H. V. Tillmann, Ban-Sok Shin, Dmitriy Shutin et al. (+1 more)
Multi-
agent systems (MASs) are a promising solution for autonomous exploration tasks in hazardous or remote environments. In such settings, communication among
agents is essential to ensure collaborative task execution, yet conventional approaches treat exploration and communication as decoupled subsystems. This work presents an approach that tightly integrates semantic communication into the MAS exploration process, adapting the communication system to the exploration methodology to improve overall task performance. Specifically, we investigate the application of semantic joint source-channel coding (JSCC) with over-the-air computation (AirComp) for distributed function computation for the application of cooperative subsurface imaging using the adapt-then-combine full waveform inversion (ATC-FWI)
algorithm. Our results demonstrate that semantic JSCC significantly outperforms classical digital communication and conventional JSCC approaches, especially in high-connectivity networks. Furthermore, incorporating side information at the receiving
agent enhances communication efficiency and imaging accuracy, a feature previously unexplored in MAS-based exploration. We validate our approach through a use case inspired by subsurface anomaly detection, showing measurable improvements in imaging performance per
agent. This work underscores the potential of semantic communication in distributed multi-
agent exploration to improve overall exploration accuracy and efficiency.
Read the full abstract ▾
All authors Maximilian H. V. Tillmann, Ban-Sok Shin, Dmitriy Shutin, Armin Dekorsy
Systems and Control
Artificial magnetic conductor backed dual-mode sectoral cylindrical DRA for off-body biomedical telemetry
Nayab Gogosh, Sohail Khalid, Bilal Tariq Malik et al. (+1 more)
This research investigates the potential of a sectoral Cylindrical Dielectric Resonator
Antenna (CDRA) for biomedical telemetry. CDRAs are known for their low loss, ruggedness, and stability, but their limited bandwidth and size make them unsuitable for wearable devices. The research addresses these limitations by proposing a dual mode
antenna that operates in EH110 and TE210 modes. The sectoral CDRA is a quarter segment with Perfect Electric Conductor boundaries, reducing its size by a factor of four. Mathematical derivations of the field components for both modes are derived to support the design. To minimize specific absorption rate (SAR), an Artificial
Magnetic Conductor (AMC) surface is applied to the
antennas backside, enhancing compatibility with the transverse electric modes. The
antenna achieves a bandwidth of 0.7 GHz (5.2-5.9 GHz), suitable for biomedical applications, with a measured peak gain of 7.9 dBi and a SAR of 1.24 W/kg when applied to a human arm.
Read the full abstract ▾
All authors Nayab Gogosh, Sohail Khalid, Bilal Tariq Malik, Slawomir Koziel
Systems and Control · Cross-listed from cs.LG
Towards a Practical Understanding of Lagrangian Methods in Safe Reinforcement Learning
Lindsay Spoor, Álvaro Serra-Gómez, Aske Plaat et al. (+1 more)
Safe reinforcement learning addresses constrained optimization problems where maximizing performance must be balanced against safety constraints, and Lagrangian methods are a widely used approach for this purpose. However, the effectiveness of Lagrangian methods depends crucially on the choice of the Lagrange multiplier $\lambda$, which governs the multi-objective trade-off between return and cost. A common practice is to update the multiplier automatically during training. Although this approach is standard in practice, there remains limited empirical evidence on the optimally achievable trade-off between return and cost as a function of $\lambda$, and there is currently no systematic
benchmark comparing automated update mechanisms to this empirical optimum. Therefore, we study (i) the constraint geometry for eight widely used safety tasks and (ii) the previously overlooked constraint-regime sensitivity of different Lagrange multiplier update mechanisms in safe reinforcement learning. Through the lens of multi-objective analysis, we present empirical Pareto frontiers that offer a complete visualization of the trade-off between return and cost in the underlying optimization problem. Our results reveal the highly sensitive nature of $\lambda$ and further show that the restrictiveness of the constraint cost can vary across different cost limits within the same task. This highlights the importance of careful cost limit selection across different regions of cost restrictiveness when evaluating safe reinforcement learning methods. We provide a recommended set of cost limits for each evaluated task and offer an open-source code base: https://github.com/lindsayspoor/Lagrangian_SafeRL.
Read the full abstract ▾
All authors Lindsay Spoor, Álvaro Serra-Gómez, Aske Plaat, Thomas Moerland
Signal Processing · Cross-listed from cs.LG
Reliable Inference in Edge-Cloud Model Cascades via Conformal Alignment
Jiayi Huang, Sangwoo Park, Nicola Paoletti et al. (+1 more)
Edge intelligence enables low-latency inference via compact on-device models, but assuring reliability remains challenging. We study edge-cloud cascades that must preserve conditional coverage: whenever the edge returns a prediction set, it should contain the true label with a user-specified probability, as if produced by the cloud model. We formalize conditional coverage with respect to the cloud predictive distribution, and introduce a conformal
alignment-based (CAb) cascading mechanism that certifies this property with user control over the risk level. Our method casts escalation from edge to cloud models as a multiple-hypothesis testing (MHT) problem, tailoring conformal
alignment (CA) to select which inputs can be safely handled at the edge. The proposed CAb model cascading method yields statistical guarantees on the average fraction of edge decisions that satisfy cloud-level conditional coverage. The procedure applies to arbitrary edge prediction sets, including variants of conformal prediction (CP), and exposes a tunable trade-off among coverage, deferral rate, and set size. Experiments on CIFAR-100 image classification and the TeleQnA question-answering (QA)
benchmark show that the proposed CAb cascade maintains the target conditional coverage for edge predictions while substantially reducing offloading to the cloud and incurring modest increases in prediction-set size.
Read the full abstract ▾
All authors Jiayi Huang, Sangwoo Park, Nicola Paoletti, Osvaldo Simeone
Audio and Speech Processing · Cross-listed from cs.SD
AWARE: Audio Watermarking with Adversarial Resistance to Edits
Kosta Pavlović, Lazar Stanarević, Petar Nedić et al. (+2 more)
Prevailing practice in learning-based audio watermarking is to pursue robustness by expanding the set of simulated distortions during training. However, such surrogates are narrow and prone to overfitting. This paper presents AWARE (Audio Watermarking with
Adversarial Resistance to Edits), an alternative approach that avoids reliance on attack-simulation stacks and handcrafted differentiable distortions.
Embedding is obtained through
adversarial optimization in the time-frequency domain under a level-proportional perceptual budget. Detection employs a time-order-agnostic detector with a Bitwise Readout Head (BRH) that aggregates temporal evidence into one score per watermark bit, enabling reliable watermark decoding even under desynchronization and temporal cuts. Empirically, AWARE attains high audio quality and
speech intelligibility (PESQ/STOI) and consistently low BER across various audio edits, often surpassing representative state-of-the-art learning-based systems.
Read the full abstract ▾
All authors Kosta Pavlović, Lazar Stanarević, Petar Nedić, Elena Nešović Slavko Kovačević, Igor Djurović
Signal Processing
6D Movable Metasurface (6DMM) in Downlink NOMA Transmissions
Li-Hsiang Shen
This letter proposes a novel six-dimensional movable metasurface (6DMM)-assisted downlink non-orthogonal multiple access (NOMA) system, in which a conventional base station (BS) equipped with fixed
antennas serves multiple users with the assistance of a reconfigurable intelligent surface (RIS) with six-dimensional spatial configurability. In contrast to traditional RIS with static surface, the proposed 6DMM architecture allows each element to dynamically adjust its position and orient the whole metasurface in yaw-pitch-roll axes, enabling both in spatial and electromagnetic controls. We formulate a sum-rate maximization problem that jointly optimizes the BS NOMA-based beamforming, phase-shifts, element positions, and rotation angles of metasurface under constraints of NOMA power levels, unit-modulus of phase-shifts, power budget, inter-element separation and boundaries of element position/orientation. Due to non-convexity and high-dimensionality, we employ a probabilistic cross-
entropy optimization (CEO) scheme to iteratively refine the solution distribution based on maximizing likelihood and elite solution sampling. Simulation results show that the proposed CEO-based 6DMM-NOMA architecture achieves substantial rate performance gains compared to 6DMM sub-structures, conventional static RIS, and other multiple access mechanisms. It also highlights the effectiveness of CEO providing probabilistic optimization for solving high-dimensional scalable metasurface.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.IT
Delay-Doppler Pulse Shaping in Zak-OTFS Using Hermite Basis Functions
Fathima Jesbin and Ananthanarayanan Chockalingam
The performance of Zak-OTFS modulation is critically dependent on the choice of the delay-Doppler (DD) domain pulse shaping filter. The design of pulses for $L^2(\mathbb{R})$ is constrained by the Balian-Low
Theorem, which imposes an inescapable trade-off between time-frequency localization and orthogonality for spectrally efficient systems. In Zak-OTFS, this trade-off requires balancing the need for localization for input/output (I/O) relation estimation with the need for orthogonality for reliable data detection when operating without time or bandwidth expansion. The well-known sinc and Gaussian pulse shapes represent the canonical extremes of this trade-off, while composite constructions such as the Gaussian-sinc (GS) pulse shape offer a good compromise. In this work, we propose a systematic DD pulse design framework for Zak-OTFS that expresses the pulse as a linear combination of Hermite basis functions. We obtain the optimal coefficients for the Hermite basis functions that minimize the inter-symbol interference (ISI) energy at the DD sampling points by solving a constrained optimization problem via singular value decomposition. For the proposed class of Hermite pulses, we derive closed-form expressions for the I/O relation and noise covariance in Zak-OTFS. Simulation results of Zak-OTFS with embedded pilot and model-free I/O relation estimation in Vehicular-A channels with fractional DDs demonstrate that the optimized pulse shape achieves a bit error rate performance that is significantly superior compared to those of the canonical sinc and Gaussian pulses and is on par with that of the state-of-the-art GS pulse, validating the proposed framework which provides greater design flexibility in terms of control of ISI and sidelobe energies.
Read the full abstract ▾
Systems and Control · Cross-listed from math.OC
A condensing approach for linear-quadratic optimization with geometric constraints
Alberto De Marchi
Optimization problems with convex quadratic cost and polyhedral constraints are ubiquitous in signal processing, automatic control and decision-making. We consider here an enlarged problem class that allows to encode logical conditions and cardinality constraints, among others. In particular, we cover also situations where parts of the constraints are nonconvex and possibly complicated, but it is practical to compute projections onto this nonconvex set. Our approach combines the augmented Lagrangian framework with a solver-agnostic structure-exploiting subproblem reformulation. While convergence guarantees follow from the former, the proposed condensing technique leads to significant improvements in computational performance.
Read the full abstract ▾