Papers

The history of extremum seeking is not merely the history of an algorithm; it is the history of an idea. Few ideas in control engineering have demonstrated the remarkable longevity of extremum seeking control (ESC). Invented more than one hundred years ago, ESC has continually reinvented itself while remaining faithful to its original objective: enabling systems to optimize their performance without relying on accurate mathematical models. This perspective article proposes that the evolution of ESC is best understood through three scientific revolutions. The first Engineering Revolution (1922-1999) established the engineering principles of model-free optimization; the second Mathematical Revolution (2000-2010) provided the rigorous mathematical foundations that transformed ESC into a mature discipline of nonlinear control; and the third ongoing Infinite-Dimensional and Cyber-Physical Revolution (2010-present) continues to expand its scope toward delays, partial differential equations, distributed optimization, event-triggered implementations, and increasingly complex cyber-physical systems. Beyond recounting this historical evolution, we offer a personal perspective on why ESC has remained relevant across successive technological eras. We argue that its enduring influence arises because the fundamental engineering challenge has never changed: "how can a dynamical system learn to improve its own performance when the optimum is unknown?" As optimization, learning, and feedback control become increasingly intertwined, ESC appears uniquely positioned to contribute to a new generation of intelligent autonomous systems, pointing toward what may become the field's fourth scientific revolution.

Manual annotation remains a major bottleneck in ultrasound (US) bone segmentation, where experts typically iteratively refine rough brush masks rather than delineating precise contours in a single pass. We present ExiL, a mask-conditioned progressive learning framework that models annotation as a structured refinement trajectory. ExiL combines a synthetic expert-like brush simulator based on signed distance fields with a lightweight 7.8M-parameter U-Net that learns to complete and refine imperfect masks from US images. During deployment, an expert mode updates the model directly from accepted refinements, enabling continual adaptation to expert behavior. Evaluated using UltraBones100k cadaver data for quantitative segmentation and a prospective volunteer dataset for annotation-efficiency analysis, ExiL reduced single-expert average annotation time from 60 to 20 seconds per frame (66.7%) and improved mean Dice by approximately 0.045 over non-progressive training, while achieving 0.87 Dice and 2.7 px boundary error in the best trajectory-aware setting. With 10–50 ms inference, ExiL enables real-time, self-improving annotation for US-guided orthopedic workflows in practical clinical labeling.

This paper introduces a structural taxonomy for constrained non-convex optimization based on the signature of Lagrange multipliers at KKT stationary points. Leveraging a unified game-theoretic interpretation of eight classical algorithm families–including block coordinate descent, ADMM, generalized Benders decomposition, successive convex approximation, interior-point methods, mirror descent, Frank-Wolfe, and Riemannian gradient descent–we show that the normalized multiplier vector carries an algorithm-independent structural fingerprint. Four scale-free shape features of this vector partition the dual space into five operational regimes: Unconstrained, Resource-Limited, Saturation, Strongly-Coupled, and Hybrid. We establish four structural theorems characterizing the partition: invariance under natural KKT symmetries, local stability under data perturbation with explicit Lipschitz margins from Robinson's strong regularity, codimension-one regime transitions, and the topological identification of the Hybrid regime as the Lebesgue-null boundary of the core regimes. A linear-time classifier is proposed with provable guarantees on correctness, iteration stabilization, sample complexity, and online tracking under data drift. Numerical experiments on 104 mixed-integer nonlinear programs and a downlink beamforming instance validate the theoretical predictions. The framework provides a foundational tool for regime-aware algorithm design and robustness analysis in non-convex optimization.

Orthogonal time frequency space (OTFS) modulation has emerged as a promising candidate for high-mobility wireless communication systems due to the diversity it offers across both time and frequency. Reliable OTFS detection, however, remains challenging under doubly-dispersive channels, where delay and Doppler dispersion induce structured interference between transmitted symbols and complicate symbol recovery. To address these challenges, we propose a two-stage iterative OTFS detector that integrates a physics-informed learned initializer with an iterative refinement network, enabling progressively more accurate symbol estimates in doubly-dispersive channels. The initializer incorporates the known delay-Doppler input-output relationship to produce a robust first-stage estimate, while the refinement stage iteratively suppresses residual symbol interference in the delay-Doppler domain. Simulation results demonstrate that the proposed detector achieves consistent performance gains over conventional and existing learning-based detectors across a variety of channel conditions. These results highlight the effectiveness of incorporating known channel structure into the detection process and using iterative refinement for improved and robust OTFS detection.

Cardiovascular sensing systems must preserve clinically useful information despite signal degradation, wireless losses, energy constraints, and edge-computation latency. We introduce Physiological Information Reliability (PIR), a cross-layer framework that represents physiological information value jointly with wireless, energy, and computation states and uses a contextual bandit to adapt sensing and communication decisions. We integrate multimodal ECG/PPG signal-quality estimation with physiological information value and an adaptive network-coding layer under burst-erasure conditions. Across controlled multiseed experiments, PIR-LinUCB demonstrates a promising low-energy operating point while maintaining medical latency constraints and competitive physiological estimation performance relative to fixed and heuristic policies. We analyze the resulting accuracy-energy-latency trade-offs and identify limitations of proxy PIV estimation and simulated communication dynamics. These results provide an initial computational demonstration of physiological-information-aware resource allocation and motivate future clinical and real-channel validation.

Convolutional neural networks (CNNs), used widely and successfully in audio enhancement, capture long-range time-frequency dependencies only indirectly, through successive convolution and pooling. Here, we present U-PAST, a hybrid transformer-U-Net architecture that addresses this limitation through self-attention dependency-modeling in the complex spectrogram domain. U-PAST tokenizes a complex STFT representation, similarly to the magnitude spectrogram tokenization of the Audio Spectrogram Transformer (AST), applies a multi-layer transformer encoder, and reconstructs the enhanced complex spectrogram with a U-Net-style decoder. We evaluate four architectural variants with between 1.17M and 2.40M parameters on the DNS Challenge, VoiceBank-DEMAND, and LibriMix corpora under matched, acoustic mismatch, and two-dataset mismatch conditions. U-PAST attains the best SI-SDR of any evaluated model under acoustic mismatch and closely trails substantially larger convolutional and time-domain baselines by 0.26 dB to 0.63 dB SI-SDR under the remaining three conditions while achieving the strongest perceptual (DNSMOS) quality under dataset mismatch. The largest evaluated configuration, U-PAST-H (2.40M parameters), is consistently the strongest variant of the family, offering an attractive performance-to-cost trade-off at a small parameter footprint.

Open synthetic test systems are essential for reproducible power system research, yet the available cases represent almost exclusively North American and European grids. No open transmission test system exists for any Middle Eastern country, whose grids raise different questions such as single corridor fuel-supply resilience and high-renewable operation within small synchronous systems, and established benchmarks cannot demonstrate. This paper presents a synthetic test system for the Jordanian transmission grid, assembled entirely from public sources, where the real topology is reconstructed from a published diagram and restored to its 2018 energized state, with plant-level generation and renewable fleets, per-site hourly profiles, and loads calibrated to the values published by Jordan's grid operators. The case is validated through structural statistics against real-grid criteria, power flow and $N\!-\!1$ screening, an energy-weighted loss decomposition, in addition to a full-year production-cost run compared against the published per-plant energy, and cross-solver verification. The model reproduces the annual energy mix at technology level (largest producers within 4%, system total within 0.1%), and the released dataset includes the bus-identity key, all scripts, and a post-2019 scenario variant. This test case is designed to benchmark system-level resource adequacy, time-series dispatch and unit commitment, renewable integration, fuel-supply resilience scenarios, interconnection studies and steady-state studies.

Large Language Model (LLM) agents increasingly automate multi-step engineering workflows through tool use, interpretation of intermediate results, and iterative planning. Diagnosing and resolving non-convergent power flow cases is a promising yet largely unexplored application, as it requires engineering judgment, experimentation, and decision-making within constrained action spaces. We introduce a benchmark that evaluates these capabilities across multiple LLMs and three architectures: chatbot, single agent, and multi-agent systems. The evaluation covers two power grids and 46 cases per grid, each requiring one or more corrective actions to restore convergence. The benchmark defines the simulation environment, observation and action spaces, and evaluation metrics, providing a reproducible foundation for developing agentic AI systems for power system planning and operation. The code is available at https://github.com/Mansutti081/RestoreBench

A key protection requirement in multiterminal HVDC (MTDC) transmission systems is to selectively isolate faulted areas using DC circuit breakers (DCCBs) during DC faults while keeping the rest of the systems in operation. However, in MTDC systems using half-bridge modular multilevel converters (HB-MMCs), the lack of inherent fault current suppression capability in HB-MMCs can cause converter blocking before fault clearance by DCCBs. This work presents an active fault current limiting (AFCL) control strategy for HB-MMCs to mitigate DC fault currents, thus avoiding converter blocking and reducing the breaking capability required of DCCBs. In addition, a systematic approach is proposed to design the AFCL control strategy under converter and network stability constraints. The effectiveness of the proposed strategy is demonstrated through PSCAD on two typical MTDC systems during DC faults.

The separation of cardiovascular sounds is a critical task in biomedical signal processing. In this paper, we introduce XVAE-WMT1, an unsupervised explainable generative AI algorithm combining a variational autoencoder (VAE) with explainable AI (XAI), wavelet-based inputs, a post-hoc output mask, and temporal consistency (TC) loss. Unlike existing supervised and VAE-based methods that rely on Short-Time Fourier Transform (STFT) and ignore latent interpretability, XVAE-WMT requires no paired clean recordings and integrates a Continuous Wavelet Transform (CWT) front-end for superior time-frequency localization. We assessed the latent space interpretability via different metrics, with SHAP (SHapley Additive exPlanations) enabling dimensionality reduction to the top 75% of latent features while preserving separation quality. Evaluated across two datasets using Signal-to-Distortion Ratio (SDR), Signal-to-Interference Ratio (SIR), and Signal-to-Artifacts Ratio (SAR), XVAE-WMT attains 26.8 dB SDR, 32.8 dB SIR, and 28.6 dB SAR.

Parameter estimation is widely used for condition monitoring of power converters, but most existing methods provide only point estimates and therefore cannot quantify whether an observed parameter change is statistically significant. This paper proposes an uncertainty-aware parameter estimation framework based on Bayesian maximum a posteriori optimization and a differentiable converter model. A Laplace approximation is used to obtain a local Gaussian posterior, enabling uncertainty quantification, consistency testing, estimator-resolution analysis, and precision-weighted pooling across data windows. The method is validated on synthetic and hardware Buck converter. It demonstrates accurate estimation of well-identified parameters and reveal the weak practical identifiability of parameters such as MOSFET on-resistance under the available sensing configuration.

The accelerated growth in data center projects has introduced a demand-driven bottleneck throughout power grids and contributed to a substantial increase in carbon emissions. These concerns are fueling discussions on methods to use existing energy assets to drive operational efficiency. To this end, this paper explores the benefits of Vehicle-to-Building (V2B) applications to support peak shaving of data center cooling loads. Initially, a literature review was conducted considering V2B constraints and optimization methods including SoC limitations, EV participation, tariffs, and building loads. This analysis was then used to develop a conceptual case study of a 10 MW data center in Loudoun County, VA by simulating a temperature-dependent load profile and adjusting the V2B participation of 40 commercial and passenger EVs. Simulation results indicate that, depending on seasonal variations in cooling load demands, strategic deployment of V2B assets between 12-5pm can offset gross cooling loads by 13-36%.

The increasing digitalization of industrial infrastructure and the convergence of information technology (IT) and operational technology (OT) have expanded the cyberattack surface of industrial systems. To address the growing complexity of cyber threats, artificial intelligence (AI) and machine learning (ML) techniques are increasingly deployed within industrial cybersecurity operations, particularly in Security Operations Centers (SOCs). While these approaches improve anomaly detection, threat analysis, and automated response, their opaque decision-making presents challenges for operational trust, regulatory compliance, and incident response. EXplainable Artificial Intelligence (XAI) has emerged as a promising paradigm to improve the transparency and interpretability of AI-driven cybersecurity systems and decisions. This paper provides a comprehensive review of XAI techniques in industrial cybersecurity, focusing on industrial SOC environments and operational security workflows. We examine the role of AI in industrial SOC workflows, the types of operational data leveraged in industrial environments, and the benefits and limitations of AI-based threat detection. We then review major families of XAI approaches, including feature attribution methods, surrogate models, rule-based explanations, and visualization techniques, and analyze their applicability to industrial use cases. We further discuss the operational, regulatory, and safety requirements that distinguish industrial systems from traditional IT environments. Key challenges are examined, including limited labeled datasets, model reliability, explainability-performance tradeoffs, and the integration of XAI tools into SOC workflows. Finally, we identify open research directions and opportunities for developing trustworthy, operationally viable, and domain-specific XAI-enabled cybersecurity solutions for industrial environments.

Control Barrier Functions (CBFs), their High-Order variants (HOCBFs) and Exponential CBFs (ECBFs) are standard geometric tools for enforcing nonlinear safety constraints. CBFs, and their variants, offer an elegant geometric framework for nonlinear safety, yet mathematically, they reduce to continuous-time convolutions restricted by zero-memory kernels. In the presence of high-frequency measurement noise, these memoryless operators act as improper filters, leading to significant control chattering and the potential loss of active control authority due to Quadratic Program (QP) infeasibility. To address this structural limitation, this paper introduces a space-time transform that embeds dynamic temporal filtering directly into the safety constraint synthesis. By designing a proper spatio-temporal kernel, this approach inherently attenuates high-frequency noise while preserving affine control authority. Crucially, we prove the robust forward invariance of the designed STT-CBF. Monte Carlo simulations of a third-order system demonstrate that the proposed framework achieves a 100% safety rate while reducing control total variation by over 99% compared to conventional parameterized barrier methods, mitigating hardware hazards and enabling reliable deployment on physical robotic platforms.

We rethink the definition of privacy in multi-server, graph-replicated private information retrieval (PIR) systems, by introducing a novel setting where the user's privacy is governed by the servers' storage structure. In classical graph-replicated PIR, the user retrieves a single message stored at the servers, while hiding the message index from each server. In our proposed privacy setting, the user is concerned with hiding the message index from a particular server, only if that server stores the message being retrieved, and privacy is not imposed otherwise. We coin this relaxed privacy requirement as local user privacy and the resulting PIR problem as local PIR on the graph. Our focus is on two-replicated PIR systems, where every message is replicated twice and stored on two distinct servers. Specifically, we study local PIR systems where the storage is represented by simple graphs, i.e., every pair of vertices is associated with at most one edge, and by their multigraph extension, i.e., $r$ parallel edges replace every edge. For these settings, we establish bounds on the local PIR capacity, defined as the maximum number of message symbols retrieved, per downloaded symbol. The local privacy requirement yields significant capacity gain over the classical PIR capacity under the same storage structure. For instance, in settings where the graph is a disjoint union of multiple identical sub-graphs, the gain in the local PIR capacity over classical PIR capacity is multiplicative in the number of sub-graphs. Further, for connected graphs, we derive capacity lower bounds for edge-transitive and bipartite graphs, which are greater than the best-known PIR capacity bounds. From these and by establishing matching upper bounds, we exactly characterize the capacity for star graphs, cyclic graphs, and path graphs with odd number of vertices. We introduce two local PIR schemes for general graphs.

Analytical signal-processing blocks are interpretable and reliable, but their optimality depends on assumptions that practical hardware and channels violate. Learned replacements can adapt, but often discard the structure that makes the original solution understandable. Here we introduce deformable matched filtering, a platform in which learning steers a bounded deformation of an explicit matched filter instead of replacing the waveform-processing path. In a hardware-in-the-loop optical wireless link, blind state descriptors drive causal, pilotless updates while payload samples, transmitted references, and condition labels remain outside the controller. Receiver deformation generalises across held-out signalling and channel conditions and remains ahead of a span-matched fractionally spaced equaliser after stationary convergence in most tested regimes. A transferable transmitter deformation reduces error-vector magnitude in all 144 held-out evaluations, whereas its additional value after receiver adaptation emerges principally under severe combined distortion. KAN, MLP, and linear controllers provide different condition-dependent advantages within the same filter structure. During 24 hours of uninterrupted changing-condition operation, bounded deformable receivers recover their original operating regime after severe intervening distortion while the persistent conventional equaliser accumulates destructive state. These results establish deformable analytical filters as a reusable middle ground between fixed theory and end-to-end learned signal processing.

We study the optimal joint sampling and transmission policy for a transmitter communicating with two gossiping receivers that share information with each other, with the objective of tracking a source under the Version Age of Information (VAoI) metric. The transmitter can observe source-version changes, but it has to pay a sampling cost to get the current source information content. Similarly, it can communicate with a receiver by paying a transmission cost. Gossiping enables local information exchange and is able to reduce costly direct transmissions. With imperfect communication links, we formulate an infinite-horizon average-cost Markov Decision Process (MDP) to jointly minimize receiver VAoI, sampling cost, and transmission cost. Using Relative Value Iteration (RVI), we evaluate the optimal policy and establish several properties of its structure. We prove that sampling has a threshold structure in the transmitter VAoI. Among direct transmissions, it is optimal to serve the older receiver. We further characterize the transmit or idle decision through the receiver VAoI difference. Our analysis shows that link reliability and receiver VAoI imbalance have a significant effect on the optimal policy structure. Numerical results verify the structural properties and demonstrate the performance gains of the optimal policy over multiple baselines.

We develop a data-driven framework for nonlinear port-Hamiltonian (pH) systems based on interpolation conditions to characterize consistency between observed data and structured dynamical models. Specifically, we derive necessary and sufficient conditions for the existence of a pH system with a smooth (convex) Hamiltonian instantly consistent with a given dataset, without requiring explicit parametrization. We further provide a semidefinite programming formulation to verify consistency with non-degenerate interconnection and dissipation structures. Our results provide a principled approach to assess instant data consistency with physical structure and pave the way for control design directly from data.

We propose a lightweight two-stage framework for real-time video anomaly detection. The first stage employs YOLO v11n-pose to detect persons and extract seventeen skeletal keypoints in a single forward pass. The second stage encodes each cropped person region through CLIP ViT-B/32 and computes cosine similarity against predefined textual descriptions of anomalous behaviors. This architecture eliminates the need for optical flow, standalone pose estimators, and density-based scoring modules. Experiments on CUHK Avenue, ShanghaiTech Campus, and a custom indoor dataset collected at Chulalongkorn University demonstrate an end-to-end throughput of approximately 51 FPS on an NVIDIA Titan XP GPU, a 3.36x speedup over the multi-feature baseline, while maintaining frame-level AUROC values of 89.26%, 70.26%, and 84.13%, respectively.

Whole-heart segmentation (WHS) in computed tomography (CT) and magnetic resonance imaging (MRI) is affected by acquisition shifts and heterogeneous cardiac annotations. Existing WHS systems combine architectural design, transfer learning, and generic spatial or intensity augmentation. We investigate whether changes to data augmentation and training supervision can improve cross-modality WHS while the segmentation architecture is held constant. We present LISynSeg, a data-centric approach that augments real-image nnU-Net training with label-to-image synthesis. Synthetic volumes are generated from cardiac label maps using contrast and acquisition perturbations calibrated to the training cohort, then mixed with real images to retain thoracic context absent from the labels (and thus the synthesized images). We model cardiac label variation through controlled changes in myocardial wall thickness and partial supervision of uncertain vessel endpoints. On the CARE Whole-Heart benchmark, synthetic-only training performs worse than the real-image nnU-Net baseline, whereas calibrated real-synthetic training improves cross-modality segmentation without changing the architecture; the improvement is larger for MRI than for CT. The results show that modifying the training data strategy can benefit model development for heterogeneous cardiac data. Code and trained weights will be released at https://github.com/MedICL-VU/Care26_LISynSeg.

Few foundation models exist for robot-assisted surgery, partly because large robotic-surgery video corpora are difficult to assemble and existing models are evaluated mostly on laparoscopic benchmarks. Further, most existing models are evaluated on a small set of public benchmarks, mostly focused on laparoscopic surgery. We present OmniRAS, a family of 1B- and 2B-parameter V-JEPA-2.1 encoders for robot-assisted surgery, and detail their training. First, we release two densely annotated robotic-cholecystectomy datasets: OmniRAS-PR and a multi-label YT-Chole tool-verb-target task, the first triplet-style annotation for robotic cholecystectomy, together with splits, probe protocols, and an inter-rater study validating the shared phase ontology. Second, we document continued pretraining at up to 256 compute nodes with global batch 6,144 over 19 sources totaling approximately 2,650 hours of surgical video, 51% robotic, and analyze compute and data composition. Third, we evaluate against raw V-JEPA-2.1 and specialized surgical models on six tasks spanning triplet, phase, and step recognition, action segmentation, and detection, under frozen-encoder and final-four-block fine-tuning regimes. Across three seeds, this yields 254 downstream runs, including 109 with partial backbone fine-tuning. The best OmniRAS models achieve the strongest adapted results across all task families, while frozen differences are smaller.

Codec-based text-to-speech (TTS) models make language-model post-training applicable to speech generation, but it remains unclear when learned perceptual predictors can serve as reinforcement learning rewards without losing alignment with human listeners. We study this question with Group Relative Policy Optimization (GRPO) using learned rewards for anime-like speaking style, naturalness, likability, and arousal. To prevent perceptual rewards from being optimized through transcript drift, we introduce a character error rate (CER) zone constraint and compare policy optimization with Best-of-$N$ reranking under the same reward gate. Across single-reward runs, each reward primarily improves its own target metric, showing that subjective predictors are not interchangeable quality surrogates. Multi-rater A/B tests further show uneven human transfer, while a reward-gap analysis separates average transfer from within-axis calibration: signed reward gaps significantly predict listener choices in the pooled analysis, whereas residual CER gaps do not, but per-axis calibration remains heterogeneous. Best-of-8 is a strong human-level baseline and is not clearly worse than GRPO perceptually, suggesting that GRPO should be viewed as amortizing reward-selected behavior into the policy rather than uniformly outperforming reranking. These results support analyzing subjective speech rewards as predictor-axis-base tuples and provide practical diagnostics for selecting rewards before multi-reward speech post-training.

Wi-Fi Channel State Information (CSI) has emerged as a key enabler for device-free Human Activity Recognition (HAR), enabling low-cost, unobtrusive sensing using existing communication infrastructure. However, Deep Learning (DL) models trained on CSI data often struggle to generalize across users, environments, and device setups due to the context sensitivity of wireless propagation. Despite this challenge, limited attention has been devoted to understanding model decisions and generalization failures. This paper introduces XAI2CSI, a framework that leverages eXplainable Artificial Intelligence (XAI) to analyze DL-based CSI sensing systems. XAI2CSI employs SAGE, a model-agnostic explainability method, to quantify temporal, spectral, and spatial CSI contributions to HAR decisions under nominal and cross-context evaluations on IEEE 802.11ax data. Our analysis reveals that the considered DL model exhibits limited robustness to unseen conditions due to an over-reliance on context-specific CSI patterns, causing models to misinterpret the underlying signal dynamics when deployment conditions change. The proposed methodology and findings provide a reference framework to explore alternative solutions and guide the development of robust, transparent Wi-Fi sensing systems.

This paper studies a susceptible–infected–susceptible (SIS) epidemic model coupled with opinion dynamics over a network of communities with higher-order interactions. Unlike standard networked SIS models, which account only for pairwise transmission, the proposed model incorporates group-level infection mechanisms and feedback between epidemic prevalence and community opinions. We establish local stability and instability conditions for a particular healthy equilibrium, derive a sufficient condition for global exponential eradication of the infection state, and identify conditions under which higher-order interactions induce bistability in the reduced dynamics on a positively invariant synchronous set. These results characterize how higher-order interactions alter the dynamics of opinion-dependent epidemic systems. Numerical simulations illustrate the predicted eradication, bistable, and endemic regimes.

This paper addresses disturbance-induced oscillations in high-precision pendulum-based inclinometers, which reduce measurement availability despite the long-term stability of gravity-referenced sensing. To actively suppress these oscillations, a contactless six-coil electromagnetic actuation system is developed, together with a control framework that generates, in real time, the planar damping force required by the controller despite the nonlinear, unilateral, and bounded nature of magnetic actuation. The proposed approach combines a control-oriented nonlinear model of the spherical pendulum, observer-based feedback control, and a constrained force-allocation scheme that maps the desired vector force into feasible coil currents. Experimental validation on a prototype demonstrates attenuation of the dominant oscillatory mode up to 31.11 dB and a marked reduction in transient duration, with the 50% decay time decreasing from 11.10 s without control to 1.01 s at the most favorable operating point. The results also highlight a trade-off between transient speed and residual oscillation, demonstrating both the effectiveness of the proposed damping strategy and its value as a practical design framework for high-precision inclinometer systems.

Prosthetic hands equipped with cameras can use computer vision to plan grasps automatically, reducing cognitive effort. However, running modern vision models on wearable devices is impractical due to power and processing constraints. We present the first prototype of a 5G-connected mobile edge computing (MEC)-enabled semi-autonomous prosthetic hand, which streams RGB-D images to an edge server for real-time grasp planning. Thirteen able-bodied participants performed pick-and-place tasks under six conditions: manual EMG control, on-device inference, wired Ethernet connectivity, and three 5G connectivity configurations (private 20 MHz network, private 100 MHz network, and a commercial 5G link) to the server. All network-based conditions performed similarly, achieving task times around 8.6 s (34% faster than manual control), failure rates of 20-38%, and 62% lower overall workload. On-device processing performed the worst with 10.3 s task time and a 76% failure rate due to slow embedded inference (3 fps vs. 6-20 fps over the network). Network latencies remained below 180 ms for private 5G and 270 ms for commercial 5G. All 5G configurations, including bandwidth-constrained and commercially variable networks, matched wired Ethernet performance while significantly outperforming both manual control and local processing, establishing 5G edge-offloading as a practical path to deploying compute-intensive prosthesis control.

Agentic coding environments give a frontier large language model (LLM) direct access to a workstation's terminal, file system, and software. This work demonstrates they extend to professional RF hardware design: an active GNSS L1-band antenna - a circularly polarized patch, surface acoustic wave (SAW) prefilter, and two-stage low-noise amplifier (LNA) on one printed circuit board (PCB) - was designed, optimized, and made manufacturing-ready. The LLM agent autonomously operated CST Studio Suite, Keysight ADS, and KiCad via scripting interfaces. Engineer input was limited to the specification, trade-off decisions, and design reviews. Workflow, results, and the RF engineer's evolving role are discussed.

This paper studies semantic-aware sub-band al- location for terahertz (THz) communication systems, where frequency-selective molecular absorption creates highly non- uniform sub-band qualities. Unlike conventional formulations, semantic fidelity depends nonlinearly on the signal-to-noise ratio (SNR) and is also sentence-specific, leading to a non-separable assignment problem that is generally not solvable using simple ordering-based policies. To address this, we use a sentence-BERT (SBERT)-based surrogate model that predicts semantic fidelity from the sentence embedding and sub-band SNR. We propose an importance-aware scheduler that assigns sentences to sub- bands based on their semantic contribution using an oracle utility function that captures importance-weighted semantic similarity across sentence-sub-band pairs. A neural scheduler is then trained through imitation learning to approximate the oracle policy at more than 200x lower runtime than full DeepSC- based oracle evaluation. Integrated with a deep-learning-enabled semantic communication (DeepSC) system, the proposed method consistently outperforms all benchmark schemes and approaches oracle-level performance under realistic THz channel conditions.

For systems with unknown parameters, finite excitation and concurrent learning can potentially yield parameter convergence without persistent excitation but the regressor may still depend on inaccessible states, leading to regressor mismatch. In this paper, this problem is addressed for a class of nonlinear systems with one-sided Lipschitz properties and quadratically inner-bounded nonlinearities with bounded disturbances and linearly parametrized uncertainties. To this aim, an output-integral regression is utilized by using measured outputs and estimated states, and history-stack residual is explicitly bounded in terms of state-estimation error and disturbance. Furthermore, a perturbation bound between the estimated-state and true-state information matrices is derived. Additionally, an OSL-QIB LMI condition is applied for the observer design and a projected adaptive law is designed without needing exact output matching. Stability analysis's results indicate the proposed observer and parameter estimation outperform observers without history-stack learning term.

Joint-Embedding Predictive Architecture (JEPA) has shown strong performance in learning rich representations through self-supervised prediction in latent space. However, it typically relies on teacher–student architecture with an EMA to stabilise training, and can tend to yield uninformative representations. Contrastive learning is stable to train and produces strong global representations, but remains limited on local tasks by the global nature of its objective. In this work, we combine both into CoJEPA: a single shared backbone jointly trained with a JEPA objective on masked sequence tokens and a contrastive objective on the class token. The contrastive gradient provides stability, removing the need for an EMA teacher entirely, while JEPA enriches the sequence tokens via local predictions that contrastive learning alone cannot provide. Crucially, no extra parameters are added to the backbone: the same model is guided towards richer representations purely through the design of its training signal. CoJEPA takes the best of both worlds, outperforming or matching both individual methods across global and local MIR tasks, with a particularly strong advantage on tonal and harmonic understanding, and without any task-specific architectural changes. CoJEPA shows that combining objectives with complementary inductive biases can substitute for scale, encouraging future work to invest in smarter training objectives over ever-larger models.