Signal Processing
A Framework for Geometry-based Statistical Channel Modeling in ISAC Systems
Ali Waqar Azim, Ahmad Bazzi, Theodore S. Rappaport et al. (+1 more)
This paper proposes a comprehensive framework for a {geometry-based statistical model for} integrated sensing and communication (
ISAC) tailored for bistatic systems. Our dual-component model decomposes the
ISAC channel into a target channel encompassing all multipath components produced by a sensing target {parameterized} by the target's radar cross-section and scattering points, and a background channel comprising all other propagation paths that do not interact with the sensing target. The framework extends TR38.901 via a hybrid clustering approach, integrating spatiotemporally consistent deterministic clusters with
stochastic clusters to preserve channel reciprocity and absolute delay
alignment for sensing parameter estimation. Extensive simulations across {urban macro, urban micro, and indoor factory} scenarios demonstrate that the model maintains communication performance parity with the standard TR38.901, validated through bit-error rate analysis obtained via simulated and measured
ISAC channels and channel capacity assessment, while enabling sensing performance evaluation, such as target ranging error for localization and receiver operating characteristic curves for detection probability.
Read the full abstract ▾
All authors Ali Waqar Azim, Ahmad Bazzi, Theodore S. Rappaport, Marwa Chafii
Systems and Control · Cross-listed from cs.RO
Fault-Tolerant MARL for CAVs under Observation Perturbations for Highway On-Ramp Merging
Yuchen Shi, Huaxin Pei, Yi Zhang et al. (+1 more)
Multi-
Agent Reinforcement Learning (MARL) holds significant promise for enabling cooperative driving among Connected and Automated Vehicles (CAVs). However, its practical application is hindered by a critical limitation, i.e., insufficient fault tolerance against observational faults. Such faults, which appear as perturbations in the vehicles' perceived data, can substantially compromise the performance of MARL-based driving systems. Addressing this problem presents two primary challenges. One is to generate
adversarial perturbations that effectively stress the policy during training, and the other is to equip vehicles with the capability to mitigate the impact of corrupted observations. To overcome the challenges, we propose a fault-tolerant MARL method for cooperative on-ramp vehicles incorporating two key
agents. First, an
adversarial fault injection
agent is co-trained to generate perturbations that actively challenge and harden the vehicle policies. Second, we design a novel fault-tolerant vehicle
agent equipped with a self-diagnosis capability, which leverages the inherent spatio-temporal correlations in vehicle state sequences to detect faults and reconstruct credible observations, thereby shielding the policy from misleading inputs. Experiments in a simulated highway merging scenario demonstrate that our method significantly outperforms baseline MARL approaches, achieving near-fault-free levels of safety and efficiency under various observation fault patterns.
Read the full abstract ▾
All authors Yuchen Shi, Huaxin Pei, Yi Zhang, Danya Yao
Signal Processing
Near-Field Channel Estimation and Joint Angle-Range Recovery in XL-MIMO Systems: A Gridless Super-Resolution Approach
Feng Xi and Dehui Yang
Existing near-field channel estimation methods for extremely large-scale
MIMO (XL-
MIMO) typically discretize angle and range parameters jointly, resulting in large polar-domain codebooks. This paper proposes a novel framework that formulates near-field channel estimation as a gridless super-resolution problem, eliminating the need for explicitly constructed codebooks. By employing a second-order approximation of spherical-wave steering vectors, the near-field channel is represented as a superposition of complex exponentials modulated by unknown waveforms. We demonstrate that these waveforms lie tightly in a common discrete chirp rate (DCR) subspace, with a dimension that scales as $\Theta(\sqrt{N})$ for an $N$-element array. By leveraging this structure and applying a lifting technique, we reformulate the non-convex problem as a convex program using regularized atomic norm minimization, which admits an equivalent semidefinite program. From the solution to the convex program, we obtain gridless angle estimates and derive closed-form coarse range estimates, followed by refinement under the exact spherical model using gradient-based nonlinear least squares. The proposed method avoids basis mismatch and exhaustive two-dimensional grid searches while enabling accurate joint angle-range estimation with pilot budgets that scale sublinearly with array size in sparse multipath regimes. Simulations demonstrate accurate channel reconstruction and user localization across representative near-field scenarios.
Read the full abstract ▾
Signal Processing
Data-Efficient Motor Condition Monitoring with Time Series Foundation Models
Deyu Li, Xinyuan Liao, Shaowei Chen et al. (+1 more)
Motor condition monitoring is essential for ensuring system reliability and preventing catastrophic failures. However, data-driven diagnostic methods often suffer from sparse fault labels and severe class imbalance, which limit their effectiveness in real-world applications. This paper proposes a motor condition monitoring framework that leverages the general features learned during pre-training of two time series foundation models, MOMENT and Mantis, to address these challenges. By transferring broad temporal representations from large-scale pre-training, the proposed approach significantly reduces dependence on labeled data while maintaining high diagnostic accuracy. Experimental results show that MOMENT achieves nearly twice the performance of conventional deep learning models using only 1% of the training data, whereas Mantis surpasses state-of-the-art baselines by 22%, reaching 90% accuracy with the same data ratio. These results demonstrate the strong generalization and data efficiency of time series foundation models in fault diagnosis, providing new insights into scalable and adaptive frameworks for intelligent motor condition monitoring.
Read the full abstract ▾
All authors Deyu Li, Xinyuan Liao, Shaowei Chen, Shuai Zhao
Systems and Control · Cross-listed from cs.LG
Automated Discovery of Laser Dicing Processes with Bayesian Optimization for Semiconductor Manufacturing
David Leeftink, Roman Doll, Heleen Visserman et al. (+4 more)
Laser dicing of
semiconductor wafers is a critical step in microelectronic manufacturing, where multiple sequential
laser passes precisely separate individual dies from the wafer. Adapting this complex sequential process to new wafer materials typically requires weeks of expert effort to balance process speed, separation quality, and material integrity. We present the first automated discovery of production-ready
laser dicing processes on an industrial
LASER1205 dicing system. We formulate the problem as a high-dimensional, constrained multi-objective
Bayesian optimization task, and introduce a sequential two-level fidelity strategy to minimize expensive destructive die-strength evaluations. On bare silicon and product wafers, our method autonomously delivers feasible configurations that match or exceed expert baselines in production speed, die strength, and structural integrity, using only technician-level operation. Post-hoc validation of different weight configurations of the utility functions reveals that multiple feasible solutions with qualitatively different trade-offs can be obtained from the final surrogate model. Expert-refinement of the discovered process can further improve production speed while preserving die strength and structural integrity, surpassing purely manual or automated methods.
Read the full abstract ▾
All authors David Leeftink, Roman Doll, Heleen Visserman, Marco Post, Faysal Boughorbel, Max Hinne, Marcel van Gerven
Systems and Control
Targeted-Subharmonic-Eliminating Pulse Density Modulation for Wireless Power Transfer Systems
Songyan Li, Hongchang Li, Haiyue Jiang et al. (+3 more)
This letter proposes a targeted-subharmonic-eliminating pulse density modulation (PDM) method for series-series (SS) compensated wireless power transfer (WPT) systems. The subharmonic frequency components which excite current abnormal oscillations in PDM controlled WPT systems are eliminated through a specially designed noise transfer function (NTF). The proposed method is simple to implement in both primary and secondary sides of WPT systems and exhibits a certain tolerance to deviations caused by inaccurate coupling coefficient identification in NTF design. Experimental results demonstrated the effectiveness and robustness of the proposed method in suppressing current abnormal oscillations and reducing the fluctuations in current amplitudes.
Read the full abstract ▾
All authors Songyan Li, Hongchang Li, Haiyue Jiang, Yudong Zhang, Wenjie Chen, Xu Yang
Signal Processing
Strategies to Minimize Out-of-Distribution Effects in Data-Driven MRS Quantification
Julian P. Merkofer, Antonia Kaiser, Anouk Schrantee et al. (+2 more)
This study systematically compared data-driven and model-based strategies for metabolite quantification in
magnetic resonance spectroscopy (MRS), focusing on resilience to out-of-distribution (OoD) effects and the balance between accuracy, robustness, and generalizability. A neural network designed for MRS quantification was trained using three distinct strategies: supervised regression, self-supervised learning, and test-time adaptation. These were compared against model-based fitting tools. Experiments combined large-scale simulated data, designed to probe metabolite concentration extrapolation and signal variability, with 1H single-voxel 7T in-vivo human
brain spectra. In simulations, supervised learning achieved high accuracy for spectra similar to those in the training distribution, but showed marked degradation when extrapolated beyond the training distribution. Test-time adaptation proved more resilient to OoD effects, while self-supervised learning achieved intermediate performance. In-vivo experiments showed larger variance across the methods (data-driven and model-based) due to domain shift. Across all strategies, overlapping metabolites and baseline variability remained persistent challenges. While strong performance can be achieved by data-driven methods for MRS metabolite quantification, their reliability is contingent on careful consideration of the training distribution and potential OoD effects. When such conditions in the target distribution cannot be anticipated, test-time adaptation strategies ensure consistency between the quantification, the data, and the model, enabling reliable data-driven MRS pipelines.
Read the full abstract ▾
All authors Julian P. Merkofer, Antonia Kaiser, Anouk Schrantee, Oliver J. Gurney-Champion, Ruud J. G. van Sloun
Signal Processing
Joint Optimization of Pilot Length, Pilot Assignment, and Power Allocation for Cell-free MIMO Systems with Graph Neural Networks
Yao Peng, Tingting Liu and Chenyang Yang
In user-centric
cell-free multi-
antenna systems, pilot contamination degrades spectral efficiency (SE) severely. To mitigate pilot contamination, existing works jointly optimize pilot assignment and power allocation by assuming fixed pilot length, which fail to balance pilot overhead against the contamination. To maximize net-SE, we jointly optimize pilot length, pilot assignment, and power allocation with deep learning. Since the pilot length is a variable, the size of pilot assignment
matrix is unknown during the optimization. To cope with the challenge, we design size-generalizable graph neural networks (GNNs). We prove that pilot assignment policy is a one-to-many mapping, and improperly designed GNNs cannot learn the optimal policy. We tackle this issue by introducing feature enhancement. To improve learning performance, we design a contamination-aware
attention mechanism for the GNNs. Given that pilot assignment and power allocation respectively depend on large- and small-scale channels, we develop a dual-timescale GNN framework to explore the potential. To reduce inference time, a single-timescale GNN is also designed. Simulation results show that the designed GNNs outperform existing methods in terms of net-SE, training complexity, and inference time, and can be well generalized across problem scales and channels.
Read the full abstract ▾
Audio and Speech Processing
Group-Aware Partial Model Merging for Children's Automatic Speech Recognition
Thomas Rolland and Alberto Abad
While supervised fine-tuning of adult pre-trained models for children's ASR has shown promise, it often fails to capture group-specific characteristics and variations among children. To address this, we introduce GRoup-Aware PARtial model Merging, a parameter-efficient approach that combines unsupervised clustering, partial fine-tuning, and model merging. Our approach adapts adult-pre-trained models to children by first grouping the children's data based on acoustic similarity. Each group is used to partially fine-tune an adult pre-trained model, and the resulting models are merged at the parameter level. Experiments conducted on the MyST children's
speech corpus indicate that GRAPAM achieves a relative WER improvement of 6%, using the same amount of data, outperforming full fine-tuning while training fewer parameters.
Read the full abstract ▾
Signal Processing
Physical Layer Security with Artificial Noise in MIMO Pinching-Antenna Systems
Pigi P. Papanikolaou, Dimitrios Bozanis, Sotiris A. Tegos et al. (+3 more)
As next-generation wireless networks emerge, security is becoming a critical performance metric. However, conventional multiple-input-multiple-output (
MIMO) systems often suffer from severe path loss and are vulnerable to nearby eavesdroppers due to their fixed-
antenna configurations. Pinching-
antenna systems (PASs) offer a promising alternative, leveraging reconfigurable pinching
antennas (PAs) positioned along low-loss dielectric waveguides to enhance channel conditions and dynamically mitigate security threats. In this paper, we propose an artificial noise (AN)-aided beamforming framework for the PAS downlink that maximizes the secrecy rate (SR) by jointly optimizing the information beams, the AN covariance, and the PA positions. We examine both perfect and imperfect channel state information (CSI) for the eavesdropper's channel. For the latter, location errors are mapped via a Jacobian into an ellipsoidal channel uncertainty set to accurately formulate the problem. We derive a closed-form solution for the single-waveguide scenario, yielding the optimal PA location and an information/AN power-splitting rule. For multiple waveguides and users, we develop a deep neural network (DNN)-aided joint optimizer that outputs beams, AN, and PA placements. Numerical results demonstrate that the proposed scheme improves SR consistently over PAS baselines in single- and multi-user settings under both perfect and imperfect CSI.
Read the full abstract ▾
All authors Pigi P. Papanikolaou, Dimitrios Bozanis, Sotiris A. Tegos, Panagiotis D. Diamantoulakis, Panagiotis Sarigiannidis, George K. Karagiannidis
Signal Processing
What If They Took the Shot? A Hierarchical Bayesian Framework for Counterfactual Expected Goals
Mikayil Mahmudlu, Oktay Karakuş, Hasan Arkadaş et al. (+1 more)
How would the probability of scoring change if a different player took the same shot? Although expected goals models describe the difficulty of scoring opportunities, comparing players on identical chances requires accounting for uncertainty in their estimated conversion probabilities. We address this question through a hierarchical
Bayesian framework that evaluates alternative shooters under the same observed circumstances. Using 35,700 eligible shots from four leagues during the 2015/16 season, we compare a contextual
benchmark with a primary player-intercept model and a provisional extension that allows player differences to vary with shooting conditions. Paired posterior draws preserve shared uncertainty when individual probability differences are aggregated across opportunities. Chronological validation demonstrates improved predictive performance relative to a constant-rate baseline, but introducing player effects provides negligible additional gains. Controlled single-seed experiments show that greater information per player improves estimation accuracy and narrows uncertainty intervals, although distinguishing small differences remains difficult and confident conclusions about near-null effects can become more frequent. Applications involving Giroud–Vardy and Berardi–Sansone similarly yield estimated differences that are small relative to their posterior uncertainty. These findings highlight an important distinction between predicting the outcome of a shot and establishing whether one player would be more likely to convert it. The proposed framework makes this distinction explicit by evaluating alternative shooters on fixed recorded opportunities while accounting for their contextual exposure. The resulting comparisons concern conditional conversion rather than the causal consequences of replacing a player in a match.
Read the full abstract ▾
All authors Mikayil Mahmudlu, Oktay Karakuş, Hasan Arkadaş, Can Eyüpoğlu
Systems and Control
Development of a Load Profile Generator for Non-road Mobile Machinery
Serhiy Kapustyan, Pranav Tetey, Thomas Grube et al. (+1 more)
This research presents a Load Profile Generator model for non-road mobile machinery, which depicts the most common operational profiles that reflect real-world conditions. This technological bottom-up model enables users to parameterize specific machines for simulation and observe their power demand at the actuator interfaces. The application of the Load Profile Generator covers different non-road mobile machinery categories, such as construction, agriculture, industrial and forestry. In this study, the Load Profile Generator was used to demonstrate common operations of material handler and forest forwarder, which have been validated against real-world data to match the results. The usage of Load Profile Generator aids engineers in evaluation machines performance by developing operation strategies. It opens doors to further systems analysis as it can serve as an interface for energy demand calculations. The model's results are accurate enough to provide a sufficient understanding of a wide range of non-road mobile machinery load profiles as well as insights about alternative fuel and power supply concepts helping in optimising the system simulation.
Read the full abstract ▾
All authors Serhiy Kapustyan, Pranav Tetey, Thomas Grube, Jochen Linssen
Signal Processing
Harnessing Chaotic Signals for Wireless Information and Power Transfer
Priyadarshi Mukherjee, Constantinos Psomas and Ioannis Krikidis
Chaotic dynamical systems have attracted considerable
attention due to their inherent randomness and high sensitivity to initial conditions, which makes them ideal for secure wireless communications. Beyond security, these same characteristics also make chaotic signals particularly effective for wireless power transfer (WPT) applications. On the other hand, connectivity along with self-sustainability are the two cornerstones of the upcoming sixth generation (6G) standard for
radio communications. Consequently, with the massive increase in wireless devices and sensors, the concept of self-sustainable wireless networks is becoming more relevant. The aspect of WPT to the widely spread wireless devices and simultaneous wireless information and power transfer (SWIPT) among these devices will play a crucial role in the 6G communication systems. In this context, it has been experimentally observed that chaotic signals result in better WPT performance as compared to the existing
benchmark schemes. Hence, in this paper, we characterize the generalized WPT performance of the multi-dimensional chaotic signals and present the use case of the Lorenz and the Henon chaotic systems. Moreover, we provide a novel differential chaos shift keying (DCSK)-based WPT receiver architecture ideal for enhanced energy harvesting (EH). Furthermore, we propose DCSK-based transmit waveform designs for multi-
antenna SWIPT architectures and investigate the impact of the rate-energy trade-off. Our goal is to explore these aspects of the chaotic signals and discuss their relevance in the context of both WPT and SWIPT.
Read the full abstract ▾
Systems and Control
Physics-Informed Neural Network Models for EMT Simulators
Ignasi Ventura Nadal, Nicolae Darii, Petros Aristidou et al. (+4 more)
This is the first paper, to the best of our knowledge, to propose a framework that integrates Physics-Informed Neural Network (PINN) models in Electromagnetic Transient (EMT) simulations. Both industry and EMT simulation tools face emerging challenges, as the share of inverter-based resources in power grids rapidly grows. Vendors need to protect the Intellectual Property (IP) of their control design when sharing their component models with customers and system operators. EMT simulation tools need to increase computation speed while maintaining numerical stability. NN surrogates offer numerically stable models with higher degrees of IP protection, as they do not expose the underlying model structure and control parameters. We show that NN surrogates can accurately capture the dynamic behavior of both open-loop and closed-loop controllers, producing identical results in both a validated in-house EMT solver and commercial tools, such as PSCAD. Compared with the artificial delay methods used by commercial solvers, we show that NN surrogates are more numerically robust. Compared with iterative numerical solvers, we show that NN surrogates replacing components with algebraic loops (e.g., closed-loop controllers) increase computation speed by 40% on average. We demonstrate our methods on PSCAD and the IEEE 39-bus system. Although the speed advantage vanishes when NN surrogates replace only a few components in large scale systems, NN-based models continue to offer enhanced IP protection and numerical stability. Our source code, in-house EMT solver, and the code for seamless integration to PSCAD are made publicly available.
Read the full abstract ▾
All authors Ignasi Ventura Nadal, Nicolae Darii, Petros Aristidou, Mohammad Kazem Bakhshizadeh, Rahul Nellikkath, Benjamin Vilmann, Spyros Chatzivasileiadis
Signal Processing
DoA Estimation with Sparse Arrays: Effects of Antenna Element Patterns and Nonidealities
Niko Lindvall, Mikko Heino, Robin Rajamäki et al. (+2 more)
This paper studies the effects of directional
antenna element complex gain patterns and nonidealities in direction of arrival (DoA) estimation. We compare sparse arrays and classical uniform linear arrays, harnessing EM simulation tools to accurately model the electromagnetic behavior of both patch and Vivaldi
antenna element including mutual coupling effects. We show that with sparse array configurations, the performance impacts are significant in terms of DoA estimation accuracy and operable SNR ranges. Specifically, in the scenarios considered, both the usage of directional
antenna elements and a sparse array result in over 90% reduction in average direction finding error, compared to a uniform omnidirectional array with the same number of elements (in this case eight), when estimating the directions of two sources using the MUSIC
algorithm. For a fixed angular RMSE, the improvements in array sensitivity are shown to yield a 4 to 15-fold increase in one-way coverage distance (assuming free-space path loss). Among the studied options, the best performance was obtained using sparse arrays with either patch or Vivaldi elements for field of views of 100$^\circ$ or 120$^\circ$, respectively.
Read the full abstract ▾
All authors Niko Lindvall, Mikko Heino, Robin Rajamäki, Mikko Valkama, Visa Koivunen
Systems and Control
Resistant Topology Inference in Consensus Networks: A Feedback-Based Design
Yushan Li, Jiabao He and Dimos V. Dimarogonas
Consensus networks are widely deployed in numerous civil and industrial applications. However, the process of reaching a common consensus among nodes can unintentionally reveal the network's
topology to external observers by appropriate inference techniques. This paper investigates a feedback-based resistant inference design to prevent the
topology from being inferred using data, while preserving the original consensus convergence. First, we characterize the conditions to preserve the original consensus, and introduce the ”accurate inference” notion, which accounts for both the uniqueness of the solution to
topology inference (solvability) and the deviation from the original
topology (accuracy). Then, we employ invariant subspace analysis to characterize the solvability. Even when unique inference remains possible, we provide necessary and sufficient conditions for the feedback design to induce inaccurate inference, and give a
Laplacian structure based distributed design. Simulations validate the effectiveness of the method.
Read the full abstract ▾
Systems and Control
Approximation-Free Control Barrier Functions for Prescribed-Time Reach-Avoid of Unknown Systems
Shubham Sawarkar and Pushpak Jagtap
We study the prescribed-time reach-avoid (PT-RA) control problem for nonlinear systems with unknown dynamics operating in environments with moving obstacles. Unlike robust or learning based Control Barrier Function (CBF) methods, the proposed framework requires neither online model learning nor uncertainty bound estimation. A CBF-based Quadratic Program (CBF-QP) is solved on a simple virtual system to generate a safe reference satisfying PT-RA conditions with respect to time-varying, tightened obstacle and goal sets. The true system is confined to a Virtual Confinement Zone (VCZ) around this reference using an approximation-free feedback law. This construction guarantees real-time safety and prescribed-time target reachability under unknown dynamics and dynamic constraints without explicit model identification or offline precomputation. Simulation results illustrate reliable dynamic obstacle avoidance and timely convergence to the target set.
Read the full abstract ▾
Systems and Control
Closed-Loop Control Law for Low Thrust Orbit Transfer with Guaranteed Stability
Suraj Kumar, Aditya Rallapalli, Nivriti Priyadarshini et al. (+2 more)
Electric propulsion is used to maximize payload capacity in communication satellites. These orbit raising maneuvers span several months and hundreds of revolutions, making trajectory design a complex challenge. The literature typically addresses this problem using feedback laws, with Q-law being one of the most prominent approaches. However, Q-law suffers from closed-loop stability issues, limiting its suitability for real-time on-board implementation. In this work, we focus on closed-loop orbit raising rather than offline trajectory planning and address the stability limitations of the Q-law through a Lyapunov based control design. A Lyapunov-guided modification of the classical Q-law is proposed to ensure closed-loop stability and enable real-time implementation. The effectiveness of the proposed method is demonstrated through closed-loop orbit transfers across various scenarios, including co-planar transfers, equatorial to polar orbit transfers, and geostationary transfer orbit (GTO) to geostationary earth orbit (GEO) transfers.
Read the full abstract ▾
All authors Suraj Kumar, Aditya Rallapalli, Nivriti Priyadarshini, Bharat Kumar GVP, Ravi Kumar L
Systems and Control
Identification of contractive Lur'e-type systems via kernel-based Lipschitz design
Cesare Donati, Fabrizio Dabbene, Constantino Lagoa et al. (+2 more)
This paper addresses the problem of identifying contractive Lur'e-type systems. Specifically, it proposes an identification framework that integrates linear prior knowledge with a kernel representation of the nonlinear feedback while systematically enforcing contractivity via Lipschitz constant design. The resulting
algorithms provide models that are accurate in prediction, interpretable, and faithful to the contractive nature of the true system. Numerical experiments demonstrate that enforcing contractivity significantly improves parameter estimation and yields models that are both accurate and physically meaningful.
Read the full abstract ▾
All authors Cesare Donati, Fabrizio Dabbene, Constantino Lagoa, Carlo Novara, Yoshio Ebihara
Systems and Control
The Battle of the Water Futures
Dennis Zanutto, Christos Michalopoulos, Lydia Tsiami et al. (+12 more)
The highly anticipated 'Battle of the Water Networks' is back with a new challenge for the water community. This competition will be hosted at the 4th International Joint Conference on Water Distribution Systems Analysis and Computing and Control in the Water Industry (WDSA/CCWI 2026), taking place in Paphos, Cyprus, from May 18-21, 2026. This competition embodies the core mission of Water-Futures and the theme for WDSA/CCWI 2026: "Designing the next generation of urban water (and wastewater) systems." The objective is to design and operate a water distribution system over a long-term horizon under deep uncertainty, with interventions applied in stages. For the first time, this challenge features a staged-design approach, unobservable and unknown uncertainties, and incorporates elements of policymaking and artificial intelligence. The solutions will be assessed using a transparent and inspectable open-source evaluation framework.
Read the full abstract ▾
All authors Dennis Zanutto, Christos Michalopoulos, Lydia Tsiami, André Artelt, Jasmin Brandt, Demetrios Eliades, Stelios Vrachimis, Stefano Alvisi, Valentina Marsili, Filippo Mazzoni, Panagiotis Smartzis, Barbara Hammer, Phoebe Koundouri, Marios Polycarpou, Dragan Savić
Systems and Control
An LLM-Assisted Multi-Agent Control Framework for Roll-to-Roll Manufacturing Systems
Jiachen Li, Shihao Li, Christopher Martin et al. (+3 more)
Roll-to-roll manufacturing requires precise tension and velocity control to ensure product quality, yet controller commissioning and adaptation remain time-intensive processes dependent on expert knowledge. This paper presents an LLM-assisted multi-
agent framework that automates control system design and adaptation for R2R systems while maintaining safety. The framework operates through five phases: system identification from operational data, automated controller selection and tuning, sim-to-real adaptation with safety verification, continuous monitoring with diagnostic capabilities, and periodic model refinement. Experimental validation on a R2R system demonstrates successful tension regulation and velocity tracking under significant model uncertainty, with the framework achieving performance convergence through iterative adaptation. The approach reduces manual tuning effort while providing transparent diagnostic information for maintenance planning, offering a practical pathway for integrating AI-assisted automation in manufacturing control systems.
Read the full abstract ▾
All authors Jiachen Li, Shihao Li, Christopher Martin, Zijun Chen, Dongmei Chen, Wei Li
Systems and Control
Distribution Network Reconfiguration: A Reproducible Benchmark of Heuristic, Metaheuristic and Mathematical Methods
Ferran Bohigas-Daranas, Oriol Gomis-Bellmunt and Eduardo Prieto-Araujo
Distribution network reconfiguration (DNR) has been studied for five decades, yet published
algorithms are rarely compared under identical assumptions: authors report results on different test systems, with different power-flow models, and almost never release code. This paper re-implements eight representative DNR
algorithms in a single open-source Python framework and
benchmarks them on five networks under identical conditions. The
algorithms span three of the four established paradigms: heuristics (loop cutting, branch exchange, greedy minimum spanning tree, exhaustive search), metaheuristics (a genetic
algorithm and a selective binary particle swarm optimizer) and mathematical programming (a mixed-integer quadratic program built on a convex, simplified DistFlow power-flow model). We report power losses, voltage profile, and, as a hardware-independent cost metric, the number of power-flow evaluations required for convergence. On the 196-bus Simbench urban network, we show that a dynamic DNR policy can save more than 26% of the system losses, weighed against
circuit-breaker maintenance cost. All implementations, test cases, and result scripts are released to allow direct extension and replication
Read the full abstract ▾
Systems and Control
Adaptive Trajectory Bundle Method for Roll-to-Roll Manufacturing Systems
Jiachen Li, Shihao Li, Christopher Martin et al. (+2 more)
Roll-to-roll (R2R) manufacturing requires precise tension and velocity control under operational constraints. Model predictive control demands gradient computation, while sampling-based methods like MPPI struggle with hard constraint satisfaction. This paper presents an adaptive trajectory bundle method that achieves rigorous constraint handling through derivative-free sequential convex programming. The approach approximates nonlinear dynamics and costs via interpolated sample bundles, replacing Taylor-series linearization with function-value interpolation. Adaptive trust region and penalty mechanisms automatically adjust based on constraint violation metrics, eliminating manual tuning. We establish convergence guarantees proving finite-time feasibility and convergence to stationary points of the constrained problem. Simulations on a six-zone R2R system demonstrate that the adaptive method achieves 4.3% lower tension RMSE than gradient-based MPC and 11.1% improvement over baseline TBM in velocity transients, with superior constraint satisfaction compared to MPPI variants. Experimental validation on an R2R dry transfer system confirms faster settling and reduced overshoot relative to LQR and non-adaptive TBM.
Read the full abstract ▾
All authors Jiachen Li, Shihao Li, Christopher Martin, Wei Li, Dongmei Chen
Systems and Control
RDS-DeePC: Robust Data Selection for Data-Enabled Predictive Control via Sensitivity Score
Jiachen Li, Shihao Li, Jian Chu et al. (+1 more)
Data Enabled Predictive Control (DeePC) is an established model free approach to predictive control, but it faces two open challenges: computational complexity that scales cubically with
dataset size and performance degradation when data are corrupted. This paper introduces Robust Data Selection DeePC (RDS DeePC), a framework that addresses both obstacles through influence function analysis. We derive a sensitivity score quantifying the leverage each trajectory segment exerts on the optimization solution and prove that high sensitivity segments correspond to outliers while low sensitivity segments represent consistent data. Selecting low sensitivity segments thus yields both computational efficiency and automatic outlier filtering without requiring data quality labels. For nonlinear systems, we extend the framework via a two stage online selection approach accelerated by the LiSSA
algorithm. Experiments on four systems of increasing complexity including a DC motor, an inverted pendulum, a planar quadrotor
UAV tracking a figure 8 trajectory, and a kinematic bicycle vehicle following a figure 8 path demonstrate that RDS DeePC achieves 94 to 97 percent clean data selection and comparable or better tracking performance under 20 percent data corruption.
Read the full abstract ▾
All authors Jiachen Li, Shihao Li, Jian Chu, Dongmei Chen
Signal Processing
RAG-Empowered LLM-Driven Dynamic Radio Resource Management in Open 6G RAN
Onur Salan, Burak Ç{\i}rağ, Onur Sever et al. (+3 more)
Implications of the advancements in the area of artificial intelligence to the wireless communications is extremely significant, especially in terms of resource management. In this paper, a
Retrieval-Augmented Generation (RAG)-empowered Large
Language Model (ReLLM)-driven dynamic
radio resource management framework for Open
Radio Access Network (O-RAN) inspired 6G networks is proposed. The introduced methodology leverages the ReLLM framework to interpret both historical and real-time network data, enabling adaptive control of network slices. The ReLLM is founded on two specialized
agents, one is responsible for proactively detecting service level agreement (SLA) violations by continuously monitoring and estimating slice-specific performance metrics, and the other one is responsible for dynamically reallocating physical resource blocks when the SLA violation probability exceeds a pre-defined threshold. The primary objective of this dual-
agent design is to minimize unnecessary LLM inference calls while satisfying the SLA requirements of the slices, thereby improving computational and energy efficiency. The proposed ReLLM framework is implemented and validated on an end-to-end O-RAN testbed built upon open-source OpenAirInterface emulators. The experimental results demonstrate that the LLM approach with its reduced
token consumption feature maintains a near-zero drop ratio for the low-priority slice while simultaneously satisfying acceptable latency performance for the high-priority slice. The ReLLM-driven design improves reliability and SLA compliance, confirming its practicality for real-world O-RAN testbeds and its potential applicability to future 6G networks.
Read the full abstract ▾
All authors Onur Salan, Burak Ç{\i}rağ, Onur Sever, İbrahim Hökelek, Ali Görçin, Hakan Ali Ç{\i}rpan
Signal Processing · Cross-listed from quant-ph
CO-QLink: Cryogenic Optical Link for Scalable Quantum Computing Systems and High-Performance Cryogenic Computing Systems
Zheng Chang, Siqi Zhang, Wenqiang Huang et al. (+8 more)
Cryogenic systems necessitate extensive data transmission between room-temperature and cryogenic environments, as well as within the cryogenic temperature domain. High-speed, low-power data transmission is pivotal to enabling the deployment of larger-scale cryogenic systems, including the scalable
quantum computing systems and the high-performance cryogenic computing systems fully immersed in
liquid nitrogen. In contrast to wireline and microwave links, optical communication links are emerging as a solution characterized by high data rates, high energy efficiency, low signal attenuation, absence of thermal conduction, and superior scalability. In this work, a 4K heat-insulated high-speed (56Gbps) low-power (1.6pJ/b) transceiver (TRX) that achieves a complete link between 4K systems and room temperature (RT) equipment is presented. Copackaged with a PIN photodiode (PD), the RX uses an inverter-based analog front-end and an analog half-rate clock data recovery loop. Connecting to a Mach-Zehnder modulator (MZM), the TX contains a voltage-mode driver with current-mode injection for low-power output-swing-boosting and 3-tap feed-forward equalization (FFE). This link has been demonstrated in the control and readout of a complete superconducting
quantum computing system.
Read the full abstract ▾
All authors Zheng Chang, Siqi Zhang, Wenqiang Huang, Tian Tian, Qichun Liu, Tiefu Li, Nan Qi, Yuanjin Zheng, Zhihua Wang, Yanshu Guo, Hanjun Jiang
Systems and Control
Fast Distributed Algorithm for Aggregative Games in Malicious Environment
Kai-Yuan Guo, Yan-Wu Wang, Xiao-Kang Liu et al. (+1 more)
This paper addresses the distributed Nash
Equilibrium seeking problem for aggregative games, where legitimate players' decisions are affected by potential malicious players. To describe players' behavior, we introduce a novel heterogeneous trustworthiness probabilistic framework by employing
stochastic trust observations. To mitigate the waste of communication and gradient computation, we utilize a compressible unbalanced network information
matrix and a multi-round communication mechanism to develop a fast Nash
equilibrium seeking
algorithm for aggregative games with unbalanced directed networks. By integrating the multi-round communication mechanism and a trustworthiness broadcast mechanism, we embed our fast convergence
algorithm into the heterogeneous trustworthiness probabilistic framework, yielding a resilient fast Nash
equilibrium seeking
algorithm. Theoretical analysis confirms the convergence of the
algorithm. Comparative simulations verify the accuracy of our fast convergence
algorithm, and validation simulations verify the resilience of the
algorithm.
Read the full abstract ▾
All authors Kai-Yuan Guo, Yan-Wu Wang, Xiao-Kang Liu, Zhi-Wei Liu
Image and Video Processing
MICCAI STS 2024 Challenge: Semi-Supervised Instance-Level Tooth Segmentation in Panoramic X-ray and CBCT Images
Yaqi Wang, Zhi Li, Chengyu Wu et al. (+19 more)
Orthopantomogram (OPGs) and Cone-Beam Computed
Tomography (CBCT) are vital for dentistry, but creating large
datasets for automated tooth
segmentation is hindered by the labor-intensive process of manual instance-level annotation. This research aimed to
benchmark and advance semi-supervised learning (SSL) as a solution for this data scarcity problem. We organized the 2nd Semi-supervised Teeth
Segmentation (STS 2024) Challenge at MICCAI 2024. We provided a large-scale
dataset comprising over 90,000 2D images and 3D axial slices, which includes 2,380 OPG images and 330 CBCT scans, all featuring detailed instance-level FDI annotations on part of the data. The challenge attracted 114 (OPG) and 106 (CBCT) registered teams. To ensure algorithmic excellence and full transparency, we rigorously evaluated the valid, open-source submissions from the top 10 (OPG) and top 5 (CBCT) teams, respectively. All successful submissions were deep learning-based SSL methods. The winning semi-supervised models demonstrated impressive performance gains over a fully-supervised nnU-Net baseline trained only on the labeled data. For the 2D OPG track, the top method improved the Instance Affinity (IA) score by over 44 percentage points. For the 3D CBCT track, the winning approach boosted the Instance Dice score by 61 percentage points. This challenge confirms the substantial benefit of SSL for complex, instance-level medical image
segmentation tasks where labeled data is scarce. The most effective approaches consistently leveraged hybrid semi-supervised frameworks that combined knowledge from foundational models like SAM with multi-stage, coarse-to-fine refinement pipelines. Both the challenge
dataset and the participants' submitted code have been made publicly available on GitHub (https://github.com/ricoleehduu/STS-Challenge-2024), ensuring transparency and reproducibility.
Read the full abstract ▾
All authors Yaqi Wang, Zhi Li, Chengyu Wu, Jun Liu, Yifan Zhang, Jiaxue Ni, Qian Luo, Jialuo Chen, Hongyuan Zhang, Jin Liu, Can Han, Kaiwen Fu, Changkai Ji, Xinxu Cai, Jing Hao, Zhihao Zheng, Shi Xu, Junqiang Chen, Qianni Zhang, Dahong Qian, Shuai Wang, Huiyu Zhou
Signal Processing
RIS-Assisted Physical Layer Security: Artificial Noise-Driven Optimization and Measurements
Ahmet Muaz Aktas, Sefa Kayraklik, Sultangali Arzykulov et al. (+3 more)
Reconfigurable intelligent surface (RIS) has emerged as a key enabler for providing signal coverage, energy efficiency, reliable communication, and physical layer security (PLS) in next-generation wireless communication networks. This paper investigates an artificial noise (AN)-driven RIS-assisted secure communication system. The RIS is partitioned into two segments, where the first segment is configured to direct the communication signal (CS) toward the legitimate user (Bob), and the other one is configured to steer the AN toward the eavesdropper (Eve). To this end, iterative and discrete
Fourier transform-based
algorithms are developed for practical RIS phase shift optimization. The power allocation between the CS and the AN signals is optimized in such a way that the secrecy capacity (SC) is maximized while limiting Eve's channel capacity. The proposed PLS framework is evaluated through both simulations and software defined
radio based testbed experiments. The results demonstrate promising improvements in the SC, highlighting the potential of AN-driven RIS-assisted PLS for practical deployments.
Read the full abstract ▾
All authors Ahmet Muaz Aktas, Sefa Kayraklik, Sultangali Arzykulov, Galymzhan Nauryzbayev, Ibrahim Hokelek, Ali Gorcin
Systems and Control
Switching-time bioprocess control with pulse-width-modulated optogenetics
Sebastián Espinel-Ríos
Biotechnology can benefit from dynamic control to improve production efficiency. In this context, optogenetics enables modulation of gene expression using light as an external input, allowing fine-tuning of
protein levels to unlock dynamic metabolic control and regulation of
cell growth. Optogenetic systems can be actuated by light intensity. However, relying solely on intensity-driven control (i.e., signal amplitude) may fail to properly tune optogenetic bioprocesses when the dose-response relationship (i.e., light intensity versus gene-expression strength) is steep. In these cases, tunability is effectively constrained to either fully active or fully repressed gene expression, with little intermediate regulation. Pulse-width modulation can alleviate this issue by alternating between fully ON and OFF light intensity within forcing periods, thereby smoothing the average response and enhancing process controllability. Optimizing pulse-width-modulated optogenetics entails a switching-time optimal control problem with a binary input over multiple forcing periods. While this can be formulated as a mixed-integer optimization problem on a refined control grid with monotonic input constraints, the number of decision variables can grow rapidly with increasing control-grid resolution within forcing periods and with the total number of forcing periods, complicating the task. Here, we propose an alternative solution based on reinforcement learning. We parametrize control actions via the duty cycle, a continuous proxy variable that encodes the ON-to-OFF switching time within each forcing period, thereby respecting the intrinsic binary nature of the light intensity while avoiding fine-grid binary decision variables.
Read the full abstract ▾