Papers

A method of information transmission using visual markers has been widely studied. In this approach, information or identifiers (IDs) are encoded in the black-and-white pattern of each marker. By analyzing the geometric properties of the marker frame - such as its size, distortion, and coordinates - the relative position and orientation between the camera and the marker can be estimated. Furthermore, by associating the positional information of each marker with its corresponding ID, the position of the camera that takes the image picture can be calculated. In the field of mobile robotics, such markers are commonly utilized for robot localization. As mobile robots become more widely used in everyday environments, such visual markers are expected to be utilized across various contexts. In environments where robots collaborate with humans - such as in cell-based manufacturing systems in factories or in domestic settings with partner robots - it is desirable for such markers to be designed in a manner that appears natural and unobtrusive to humans. In this paper, we propose a method for implementing an ArUco marker in the form of illumination. In the proposed method, LEDs are arranged in accordance with the grid pattern of the marker, and the blinking frequency of each LED is determined based on the corresponding black or white cell. As a result, the illumination appears uniformly bright to the human eye, while the camera can capture variations in the blinking frequency. From these differences, the black-and-white pattern can be reconstructed, enabling the identification of the marker's tag information. We develop a prototype system, and conduct experiments which are conducted to evaluate its performance in terms of recognition accuracy under varying distances and viewing angles with respect to the ArUco marker.

Power amplifiers (PA) are essential for microwavecontrolled trapped-ion and semiconductor spin based quantum computers (QC). They adjust the power level of the control signal and therefore the processing time of the QC. Their nonlinearities and memory effects degrade the signal quality and, thus, the fidelity of qubit gate operations. Driving the PA with a significant input power back-off reduces nonlinear effects but is neither power-efficient nor cost-effective. To overcome this limitation, this letter augments the conventional signal generation system applied in QCs by digital predistortion (DPD) to linearize the radio frequency (RF) channel. Numerical analysis of the qubit behavior based on measured representative control signals indicates that DPD improves its fidelity.

Consumer-grade biosensors offer a cost-effective alternative to medical-grade electromyography (EMG) systems, reducing hardware costs from thousands of dollars to approximately $13. However, these low-cost sensors introduce significant signal instability and motion artifacts. Deploying machine learning models on resource-constrained edge devices like the ESP32 presents a challenge: balancing classification accuracy with strict latency (<100ms) and memory (<320KB) constraints. Using a single-subject dataset comprising 1,540 seconds of raw data (1.54M data points, segmented into ~1,300 one-second windows), I evaluate 18 model architectures, ranging from statistical heuristics to deep transfer learning (ResNet50) and custom hybrid networks (MaxCRNN). While my custom "MaxCRNN" (Inception + Bi-LSTM + Attention) achieved the highest safety (99% Precision) and robustness, I identify Random Forest (74% accuracy) as the Pareto-optimal solution for embedded control on legacy microcontrollers. I demonstrate that reliable, low-latency EMG control is feasible on commodity hardware, with Deep Learning offering a path to near-perfect reliability on modern Edge AI accelerators.

The weighted median mechanism provides a robust alternative to weighted averaging in opinion dynamics. Existing models, however, are predominantly formulated on pairwise interaction graphs, which limits their ability to represent higher-order environmental effects. In this work, a generalized weighted median opinion dynamics model is proposed by incorporating high-order interactions through a simplicial complex representation. The resulting dynamics are formulated as a nonlinear discrete-time system with synchronous opinion updates, in which intrinsic agent interactions and external environmental influences are jointly modeled. Sufficient conditions for asymptotic consensus are established for heterogeneous systems composed of opinionated and unopinionated agents. For homogeneous opinionated systems, convergence and convergence rates are rigorously analyzed using the Banach fixed-point theorem. Theoretical results demonstrate the stability of the proposed dynamics under mild conditions, and numerical simulations are provided to corroborate the analysis. This work extends median-based opinion dynamics to high-order interaction settings and provides a system-level framework for stability and consensus analysis.

The integration of autonomous unmanned aerial vehicles (UAVs) into large-scale artistic projects has emerged as a new application in robotics. This paper presents the design, deployment, and testing of a novel multi-drone system for automated mural painting in outdoor settings. This technology makes use of new software that coordinates multiple drones simultaneously, utilizing state-machine algorithms for task execution. Key advancements are the complex positioning system that combines 2D localization using a single motion tracking camera with onboard LiDAR for precise positioning, and a novel flight control algorithm, which works differently along the trajectory and normally to it, ensuring smoothness and high precision of the drawings at the same time. A 100 square meters mural was created using the developed multi-drone system, validating the system's efficacy. Compared to single-drone approaches, our multi-UAV solution significantly improves scalability and operational speed while maintaining high stability even in harsh weather conditions. The findings highlight the potential of autonomous robotic swarms in creative applications, paving the way for further advancements in large-scale robotic art.

Cell-free massive MIMO (multiple-input multiple-output) enhances spectral and energy efficiency compared to conventional cellular networks by enabling joint transmission and reception across a large number of distributed access points (APs). Since these APs are envisioned to be low-cost and densely deployed, hardware impairments, stemming from non-ideal radio-frequency (RF) chains, are unavoidable. While existing studies primarily address hardware impairments on the access side, the impact of hardware impairments on the wireless fronthaul link has remained largely unexplored. In this work, we fill this important gap by introducing a novel amplify-and-forward (AF) based wireless fronthauling scheme tailored for cell-free massive MIMO. Focusing on the uplink, we develop an analytical framework that jointly models the hardware impairments at both the APs and the fronthaul transceivers, derives the resulting end-to-end distorted signal expression, and quantifies the individual contribution of each impairment to the spectral efficiency. Furthermore, we design distortion-aware linear combiners that optimally mitigate these effects. Numerical results demonstrate significant performance gains from distortion-aware processing and illustrate the potential of the proposed AF fronthauling scheme as a cost-effective enabler for future cell-free architectures.

Wave Energy Converters, particularly point absorbers, have emerged as one of the most promising technologies for harvesting ocean wave energy. Nevertheless, achieving high conversion efficiency remains challenging due to the inherently complex and nonlinear interactions between incident waves and device motion dynamics. This study develops an optimal adaptive damping control model for the power take-off (PTO) system by coupling Smoothed Particle Hydrodynamics (SPH) with multi-agent deep reinforcement learning. The proposed framework enables real-time communication between high-fidelity SPH simulations and intelligent control agents that learn coordinated policies to maximise energy capture. In each training episode, the SPH-based environment provides instantaneous hydrodynamic states to the agents, which output continuous damping actions and receive rewards reflecting power absorption. The Multi-Agent Soft Actor Critic algorithm is employed within a centralised-training and decentralised-execution scheme to ensure stable learning in continuous, multi-body systems. The entire platform is implemented in a unified GPU-accelerated C++ environment, allowing long-horizon training and large-scale three-dimensional simulations. The approach is validated through a series of two-dimensional and three-dimensional benchmark cases under regular and irregular wave conditions. Compared with constant PTO damping, the learned control policy increases overall energy capture by 23.8% and 21.5%, respectively, demonstrating the strong potential of intelligent control for improving the performance of wave energy converter arrays. The developed three-dimensional GPU-accelerated multi-agent platform in computational hydrodynamics, is extendable to other fluid-structure interaction engineering problem that require real-time, multi-body coordinated control.

In the uplink of a cell-free massive MIMO system, quantization affects performance in two key domains: the time-domain distortion introduced by finite-resolution analog-to-digital converters (ADCs) at the access points (APs), and the fronthaul quantization of signals sent to the central processing unit (CPU). Although quantizing twice may seem redundant, the ADC quantization in orthogonal frequency-division duplex (OFDM) systems appears in the time domain, and one must then convert to the frequency domain, where quantization can be applied only to the signals at active subcarriers. This reduces fronthaul load and avoids unnecessary distortion, since the ADC output spans all OFDM samples while only a subset of subcarriers carries useful information. While both quantization effects have been extensively studied in narrowband systems, their joint impact in practical wideband OFDM-based cell-free massive MIMO remains largely unexplored. This paper addresses the gap by modeling the joint distortion and proposing a fronthaul strategy in which each AP processes the received signal to reduce quantization artifacts before transmission. We develop an efficient estimation algorithm that reconstructs the unquantized time-domain signal prior to fronthaul transmission and evaluate its effectiveness. The proposed design offers new insights for implementing efficient, quantization-aware uplink transmission in wideband cell-free architectures.

Accurate prediction of joint kinematics and kinetics is essential for advancing gait analysis and developing intelligent assistive systems such as prosthetics and exoskeletons. This study presents a hybrid LSTM-UKF framework for estimating ankle angle and ground reaction force (GRF) across varying walking speeds. A multimodal sensor fusion strategy integrates force plate data, knee angle, and GRF signals to enrich biomechanical context. Model performance was evaluated using RMSE and $R^2$ under subject-specific validation. The LSTM-UKF consistently outperformed standalone LSTM and UKF models, achieving up to 18.6% lower RMSE for GRF prediction at 3 km/h. Additionally, UKF integration improved robustness, reducing ankle angle RMSE by up to 22.4% compared to UKF alone at 1 km/h. These results underscore the effectiveness of hybrid architectures for reliable gait prediction across subjects and walking conditions.

WiFi sensing has suffered from the limited bandwidths designated for its original communication purpose, leading to fundamental limits in multipath resolution and thus multi-user sensing. Unfortunately, it is practically prohibitive to obtain large bandwidths on commercial WiFi, considering the conflict between the limited spectrum and the crowded networks. In this paper, we present Neuro-Wideband (NWB), a completely different paradigm that enables wideband WiFi sensing without specialized hardware or extra channel measurements. Our key insight is that any physical measurement of channel state information (CSI) inherently encapsulates multipath parameters, which, while unsolvable in isolation, can be transformed into an expanded form of CSI (eCSI) approximating measurements over a broader bandwidth. To ground this insight, we propose WUKONG to address NWB as a unique self-conditioned learning problem that can be trained by using any existing CSI data as self-labeled samples. WUKONG introduces a novel deep learning framework by integrating Transformer and Diffusion models, which captures sample-specific multipath parameters and transfers this sample-level knowledge to the outcome eCSI. We conduct real-world experiments to evaluate WUKONG on diverse WiFi signals across protocols and bandwidths. The results show the promising effectiveness of NWB, which is further demonstrated through case studies on localization and multi-person breathing monitoring using eCSI. Overall, the proposed NWB promises a practical pathway toward realizing wideband WiFi sensing on commodity hardware, expanding the design space of wireless sensing systems.

Millimeter-wave radar enables robust environment perception in autonomous systems under adverse conditions yet suffers from sparse, noisy point clouds with low angular resolution. Existing diffusion-based radar enhancement methods either incur high learning complexity by modeling full LiDAR distributions or fail to prioritize critical structures due to uniform regional processing. To address these issues, we propose R3D, a regional-guided residual radar diffusion framework that integrates residual diffusion modeling-focusing on the concentrated LiDAR-radar residual encoding complementary high-frequency details to reduce learning difficulty-and sigma-adaptive regional guidance-leveraging radar-specific signal properties to generate attention maps and applying lightweight guidance only in low-noise stages to avoid gradient imbalance while refining key regions. Extensive experiments on the ColoRadar dataset demonstrate that R3D outperforms state-of-the-art methods, providing a practical solution for radar perception enhancement. Our anonymous code and pretrained models are released here: https://anonymous.4open.science/r/r3d-F836

Loss-of-control (LOC) remains a leading cause of fixed-wing aircraft accidents, especially in post-stall and flat-spin regimes where conventional gain-scheduled or logic-based recovery laws may fail. This study formulates spin-recovery as a continuous-state, continuous-action Markov Decision Process and trains a Proximal Policy Optimization (PPO) agent on a high-fidelity six-degree-of-freedom F-18/HARV model that includes nonlinear aerodynamics, actuator saturation and rate coupling. A two-phase potential-based reward structure first penalizes large angular rates and then enforces trimmed flight. After 6,000 simulated episodes, the policy generalities to unseen upset initializations. Results show that the learned policy successfully arrests the angular rates and stabilizes the angle of attack. The controller performance is observed to be satisfactory for recovery from spin condition which was compared with a state-of-the-art sliding mode controller. The findings demonstrate that deep reinforcement learning can deliver interpretable, dynamically feasible manoeuvres for real-time loss of control mitigation and provide a pathway for flight-critical RL deployment.

In this work, we investigate a blockage-aware pinching antenna (PA) system designed for secure and robust wireless communication. The considered system comprises a base station equipped with multiple waveguides, each hosting multiple PAs, and serves multiple single-antenna legitimate users in the presence of multi-antenna eavesdroppers under imperfect channel state information (CSI). To safeguard confidential transmissions, artificial noise (AN) is deliberately injected to degrade the eavesdropping channels. Recognizing that conventional linear CSI error bounds become overly conservative for spatially distributed PA architectures, we develop new geometry aware uncertainty sets that jointly characterize eavesdropper position and array-orientation errors. Building upon these sets, we formulate a robust joint optimization problem that determines per waveguide beamforming and AN covariance, individual PA power ratio allocation, and PA positions to maximize the system sum rate subject to secrecy constraints. The highly nonconvex design problem is efficiently addressed via a low computational complexity iterative algorithm that capitalizes on block coordinate descent, penalty based methods, majorization minimization, the S procedure, and Lipschitz based surrogate functions. Simulation results demonstrate that the sum rate achieved by the proposed algorithm outperforms conventional fixed-antenna systems by 4.7 dB, offering substantially improved rate and secrecy performance. In particular, (i) adaptive PA positioning preserves LoS to legitimate users while effectively exploiting waveguide geometry to disrupt eavesdropper channels, and (ii) neglecting blockage effects in the PA system significantly impacts the system design, leading to performance degradation and inadequate secrecy guarantees.

Accurate wireless localization underpins applications from autonomous systems to smart infrastructure. We study the mean-squared error (MSE) and conditional MSE (CMSE) of a practical fusion-based estimator in d-dimensional, stationary isotropic (translation- and rotation-invariant) random sensor networks, where a central processor combines received-signal-strength (RSS) and angle-of-arrival (AOA) measurements to infer a target's position. Our contributions are twofold. First, we establish an approximation theorem: when measurement noise is sufficiently large, the joint law of RSS and AOA observations under a broad class of stationary isotropic deployments is, in distribution, indistinguishable from that induced by a homogeneous Poisson point process (PPP). Second, leveraging this equivalence, we investigate a homogeneous PPP-based sensor network. We propose a fusion-based estimator in which a central processor aggregates RSS and AOA measurements from a set of spatially distributed sensors to infer the target position. For this PPP deployment within a finite observation region, we derive tractable analytical upper bounds for both the MSE and CMSE, establishing explicit scaling laws with respect to sensor density, observation radius, and noise variance. The approximation theorem then certifies these PPP-based bounds as reasonable proxies for non-Poisson deployments in noisy regimes. Overall, the results translate deployment and sensing parameters into achievable accuracy targets and provide robust, cost-aware guidance for the design of next-generation location-aware wireless networks.

We introduce CL-QAS, a continual quantum architecture search framework that mitigates the challenges of costly amplitude encoding and catastrophic forgetting in variational quantum circuits. The method uses Tensor-Train encoding to efficiently compress high-dimensional stochastic signals into low-rank quantum feature representations. A bi-loop learning strategy separates circuit parameter optimization from architecture exploration, while an Elastic Weight Consolidation regularization ensures stability across sequential tasks. We derive theoretical upper bounds on approximation, generalization, and robustness under quantum noise, demonstrating that CL-QAS achieves controllable expressivity, sample-efficient generalization, and smooth convergence without barren plateaus. Empirical evaluations on electrocardiogram (ECG)-based signal classification and financial time-series forecasting confirm substantial improvements in accuracy, balanced accuracy, F1 score, and reward. CL-QAS maintains strong forward and backward transfer and exhibits bounded degradation under depolarizing and readout noise, highlighting its potential for adaptive, noise-resilient quantum learning on near-term devices.

Ground-penetrating radar (GPR) combines depth resolution, non-destructive operation, and broad material sensitivity, yet it has seen limited use in diagnosing building envelopes. The compact geometry of wall assemblies, where reflections from closely spaced studs, sheathing, and cladding strongly overlap, has made systematic inversion difficult. Recent advances in data-driven interpretation provide an opportunity to revisit this challenge and assess whether machine learning can reliably extract structural information from such complex signals. Here, we develop a GPR-based inversion framework that decomposes wall diagnostics into classification tasks addressing vertical (stud presence) and lateral (wall-type) variations. Alongside model development, we implement multiple feature minimization strategies - including recursive elimination, agglomerative clustering, and L0-based sparsity - to promote fidelity and interpretability. Among these approaches, the L0-based sparse neural network (SparseNN) emerges as particularly effective: it exceeds Random Forest accuracy while relying on only a fraction of the input features, each linked to identifiable dielectric interfaces. SHAP analysis further confirms that the SparseNN learns reflection patterns consistent with physical layer boundaries. In summary, this framework establishes a foundation for physically interpretable and data-efficient inversion of wall assemblies using GPR radargrams. Although defect detection is not addressed here, the ability to reconstruct intact envelope structure and isolate features tied to key elements provides a necessary baseline for future inversion and anomaly-analysis tasks.

Flexible spacecraft structures present significant challenges for physical and control system design due to nonlinear dynamics, mission constraints, environmental variables, and changing operational conditions. This paper presents a data-driven framework for constructing reduced-order surrogate models of a flexible spacecraft using the method of Dynamic Mode Decomposition (DMD), followed by optimal sensor/actuator pair placement. Highfidelity simulation data from a nonlinear flexible spacecraft model, including coupled rigid-body and elastic modes, are captured by defining a mesh of nodes over the spacecraft body. The data-driven methods are then used to construct a modal model from the time histories of these node points. Optimal sensor/actuator placement for controllability and observability is performed via a nonlinear programming technique that maximizes the singular values of the Hankel matrix. Finally, the sensor placement and dynamics modeling approach is iterated to account for changes in the dynamic system introduced by sensor/actuator physical mass. The proposed methodology enables initialization of physical modeling without requiring a direct analytical model and provides a practical solution for onboard implementation in model-based control and estimation systems. Results demonstrate optimal design methodology with substantial model-order reduction while preserving dynamic fidelity, and provide insight into effective sensor-actuator configurations for estimation and control.

Koopman operator recently gained increasing attention in the control systems community for its abilities to bridge linear and nonlinear systems. Data driven Koopman operator approximations have established themselves as key enablers for system identification and model predictive control. Nonetheless, such methods commonly entail a preselected definition of states in the function space leading to high dimensional, overparameterized models that may suffer from poor numerical conditioning and degraded long term prediction performance. We address this problem by proposing a hierarchical probabilistic approach for the Koopman model identification problem. In our method, elements of the model are treated as random variables and the posterior estimates are found using variational Bayesian (VB) inference updates. Our model distinguishes from others in the integration of inclusion flags. By the help of the inclusion flags, we intuitively threshold the probability of each state in the model. We then propose a graph search based algorithm to reduce the preselected states of the Koopman model. We demonstrate that the proposed reduction improves numerical conditioning and can preserve or improve prediction performance while substantially reducing the dictionary size.

Selective hardening is widely employed to improve the reliability of FPGA based soft processors while limiting the overhead of full redundancy. However, existing approaches primarily rely on architectural criticality or functional fault analysis, overlooking the impact of routing dependent timing sensitivity on processor robustness. This paper introduces a timing fragility aware selective hardening methodology for RISCV soft processors implemented on SRAM based FPGAs. Building on recent advances in in situ timing observability, the proposed approach quantifies the statistical timing sensitivity of pipeline components under controlled routing perturbations and uses this information to guide hardening decisions. Experimental results on a RISCV processor implemented on a commercial FPGA platform show that components exhibiting higher timing fragility also demonstrate increased vulnerability to routing induced delay effects. Leveraging this correlation, the proposed selective hardening strategy achieves robustness comparable to full hardening while significantly reducing area and timing overhead. These results demonstrate that timing fragility provides a practical and effective metric for reliability aware design optimization in FPGA based processor architectures.

Block based image compression relies on transform coding to concentrate signal energy into a small number of coefficients. While classical codecs use fixed transforms such as the Discrete Cosine Transform (DCT), data driven methods such as Principal Component Analysis (PCA) are theoretically optimal for decorrelation. This paper presents an experimental comparison of DCT, Hadamard, and PCA across multiple block sizes and compression rates. Using rate distortion and energy compaction analysis, we show that PCA outperforms fixed transforms only when block dimensionality is sufficiently large, while DCT remains near optimal for standard block sizes such as $8\times8$ and at low bit rates. These results explain the robustness of DCT in practical codecs and highlight the limitations of block wise learned transforms.

Embedded vision systems need efficient and robust image processing algorithms to perform real-time, with resource-constrained hardware. This research investigates image processing algorithms, specifically edge detection, corner detection, and blob detection, that are implemented on embedded processors, including DSPs and FPGAs. To address latency, accuracy and power consumption noted in the image processing literature, optimized algorithm architectures and quantization techniques are employed. In addition, optimal techniques for inter-frame redundancy removal and adaptive frame averaging are used to improve throughput with reasonable image quality. Simulations and hardware trials of the proposed approaches show marked improvements in the speed and energy efficiency of processing as compared to conventional implementations. The advances of this research facilitate a path for scalable and inexpensive embedded imaging systems for the automotive, surveillance, and robotics sectors, and underscore the benefit of co-designing algorithms and hardware architectures for practical real-time embedded vision applications.

In Differential GNSS (DGNSS) positioning, differencing measurements between a user and a reference station suppresses common-mode errors but also introduces reference-station noise, which fundamentally limits accuracy. This limitation is minor for high-grade stations but becomes significant when using reference infrastructure of mixed quality. This paper investigates how large-scale user cooperation can mitigate the impact of reference-station noise in conventional (non-cooperative) DGNSS systems. We develop a unified estimation framework for cooperative DGNSS (C-DGNSS) and cooperative real-time kinematic (C-RTK) positioning, and derive parameterized expressions for their Fisher information matrices as functions of network size, satellite geometry, and reference-station noise. This formulation enables theoretical analysis of estimation performance, identifying regimes where cooperation asymptotically restores the accuracy of DGNSS with an ideal (noise-free) reference. Simulations validate these theoretical findings.

We develop a practical framework for distinguishing diffusive stochastic processes from deterministic signals using only a single discrete time series. Our approach is based on classical excursion and crossing theorems for continuous semimartingales, which correlates number $N_\varepsilon$ of excursions of magnitude at least $\varepsilon$ with the quadratic variation $[X]_T$ of the process. The scaling law holds universally for all continuous semimartingales with finite quadratic variation, including general Ito diffusions with nonlinear or state-dependent volatility, but fails sharply for deterministic systems – thereby providing a theoretically-certfied method of distinguishing between these dynamics, as opposed to the subjective entropy or recurrence based state of the art methods. We construct a robust data-driven diffusion test. The method compares the empirical excursion counts against the theoretical expectation. The resulting ratio $K(\varepsilon)=N_{\varepsilon}^{\mathrm{emp}}/N_{\varepsilon}^{\mathrm{theory}}$ is then summarized by a log-log slope deviation measuring the $\varepsilon^{-2}$ law that provides a classification into diffusion-like or not. We demonstrate the method on canonical stochastic systems, some periodic and chaotic maps and systems with additive white noise, as well as the stochastic Duffing system. The approach is nonparametric, model-free, and relies only on the universal small-scale structure of continuous semimartingales.

Target speaker extraction (TSE) aims to recover the speech of a desired speaker from a mixture given a short enrollment utterance, while speech enhancement (SE) focuses on improving speech quality under noisy conditions. Most existing TSE and SE systems are based on discriminative modeling and have shown strong interference suppression ability, but they often remain limited in perceptual quality and naturalness. To address this issue, we first introduce LauraTSE, a generative TSE model built on an autoregressive decoder-only language model. Although generative modeling is promising for quality enhancement, purely generative TSE may suffer from hallucination, content drift, and limited controllability in complex acoustic conditions. We therefore propose a discriminative-generative two-stage framework, where a discriminative front-end first produces target-related representations with strong interference suppression, and a generative back-end then reconstructs high-quality speech in the neural audio codec representation space. This design combines the controllability of discriminative extraction with the reconstruction capability of generative modeling. We further investigate several collaboration strategies for the two-stage framework, including front-end freezing, joint fine-tuning, SI-SDR regularization, and autoregressive/non-autoregressive inference. Experimental results on both TSE and SE benchmarks show that the proposed framework achieves a better balance among perceptual quality, intelligibility, and speaker consistency than purely discriminative or purely generative baselines.

Unmanned aerial vehicles (UAVs) can be critical for time-sensitive data collection missions, yet existing research often relies on simulations that fail to capture real-world complexities. Many studies assume ideal wireless conditions or focus only on path planning, neglecting the challenge of making real-time decisions in dynamic environments. To bridge this gap, we address the problem of adaptive sensor selection for a data-gathering UAV, considering both the buffered data at each sensor and realistic propagation conditions. We introduce the Hover-based Greedy Adaptive Download (HGAD) strategy, designed to maximize data transfer by intelligently hovering over sensors during periods of peak signal quality. We validate HGAD using both a digital twin (DT) and a real-world (RW) testbed at the NSF-funded AERPAW platform. Our experiments show that HGAD significantly improves download stability and successfully meets per-sensor data targets. When compared with the traditional Greedy approach that simply follows the strongest signal, HGAD is shown to outperform in the cumulative data download. This work demonstrates the importance of integrating signal-to-noise ratio (SNR)-aware and buffer-aware scheduling with DT and RW signal traces to design resilient UAV data-mule strategies for realistic deployments.

Curving beams are a promising new method for bypassing obstacles in future millimeter-wave to sub-terahertz (sub-THz) networks but lack a general predictive model for their reflections from arbitrary surfaces. We show that, unfortunately, attempting to "mirror" the incident beam trajectory across the normal of the reflector, as in ray optics, fails in general. Thus, we introduce the first geometric framework capable of modeling the reflections of arbitrary convex sub-THz curving beams from general reflectors with experimental verification. Rather than "mirroring" the trajectory, we decompose the beam into a family of tangents and demonstrate that this process is equivalent to the Legendre transform. This approach allows us to accurately account for reflectors of any shape, size, and position while preserving the underlying physics of wave propagation. Our model is validated through finite element method simulations and over-the-air experiments, demonstrating millimeter-scale accuracy in predicting reflections. Our model provides a foundation for future curving beam communication and sensing systems, enabling the design of reflected curved links and curving radar paths.

We consider a cellular network containing $n$ nodes where nodes within a cell gossip with each other in a fully-connected fashion and a source shares updates with these nodes via a mobile drone. The drone receives source updates and shares them with nodes in the cell where it currently resides. The drone moves between cells according to an underlying continuous-time Markov chain (CTMC). We evaluate the impact of the number of cells $f(n)$, drone speed $\lambda_m(n)$ and drone dissemination rate $\lambda_d(n)$ on the information freshness of nodes in the network. We use the version age of information metric to quantify information freshness. We observe that the expected duration between two drone-to-cell service times depends on the stationary distribution of the underlying CTMC and $\lambda_d(n)$, but not on $\lambda_m(n)$. However, the version age instability makes high probability analysis for a general underlying CTMC difficult. Therefore, we focus on the fully-connected drone mobility model. Under this model, we uncover a dual-bottleneck, by leveraging stochastic equivalence between drone mobility and drone dissemination speed: the version age is constrained by the slower of these two processes. If $\lambda_d(n) \gg \lambda_m(n)$, then the version age scaling of nodes is dominated by the inverse of $\lambda_m(n)$ and is independent of $\lambda_d(n)$. If $\lambda_m(n) \gg \lambda_d(n)$, then the version age scaling of nodes is dominated by the inverse of $\lambda_d(n)$ and is independent of $\lambda_m(n)$.

Large-scale integration of distributed energy resources has led to a rapid increase in the number of controllable devices and a significant change in system dynamics. This has necessitating the shift towards more distributed and scalable control strategies to manage the increasing system complexity. In this work, we address the problem of partitioning a low-inertia power network into dynamically coherent subsystems to facilitate the utilization of distributed control schemes. We show that an embedding of the power network using the spectrum of the linearized synchronization dynamics matrix results in a natural decomposition of the network. We establish the connection between our approach and the broader framework of spectral clustering using the Laplacian matrix of the admittance network. The proposed method is demonstrated on the IEEE 30-bus test system. We consider the robustness of the clusters by analyzing the sensitivity of the small eigenvalues and their corresponding eigenspaces to perturbations caused by variation in the steady-state operating points of the network.

Functional near-infrared spectroscopy (fNIRS) and diffuse optical tomography (DOT) are rapidly evolving toward wearable, multimodal, and data-driven, AI-supported neuroimaging in the everyday world. However, current analytical tools are fragmented across platforms, limiting reproducibility, interoperability, and integration with modern machine learning (ML) workflows. Cedalion is a Python-based open-source framework designed to unify advanced model-based and data-driven analysis of multimodal fNIRS and DOT data within a reproducible, extensible, and community-driven environment. Cedalion integrates forward modelling, photogrammetric optode co-registration, signal processing, GLM Analysis, DOT image reconstruction, and ML-based data-driven methods within a single standardized architecture based on the Python ecosystem. It adheres to SNIRF and BIDS standards, supports cloud-executable Jupyter notebooks, and provides containerized workflows for scalable, fully reproducible analysis pipelines that can be provided alongside original research publications. Cedalion connects established optical-neuroimaging pipelines with ML frameworks such as scikit-learn and PyTorch, enabling seamless multimodal fusion with EEG, MEG, and physiological data. It implements validated algorithms for signal-quality assessment, motion correction, GLM modelling, and DOT reconstruction, complemented by modules for simulation, data augmentation, and multimodal physiology analysis. Automated documentation links each method to its source publication, and continuous-integration testing ensures robustness. This tutorial paper provides seven fully executable notebooks that demonstrate core features. Cedalion offers an open, transparent, and community extensible foundation that supports reproducible, scalable, cloud- and ML-ready fNIRS/DOT workflows for laboratory-based and real-world neuroimaging.

Orthogonal time frequency space (OTFS) modulation is a robust candidate waveform for future wireless systems, particularly in high-mobility scenarios, as it effectively mitigates the impact of rapidly time-varying channels by mapping symbols in the delay-Doppler (DD) domain. However, accurate frame synchronization in OTFS systems remains a challenge due to the performance limitations of conventional algorithms. To address this, we propose a low-complexity synchronization method based on a coarse-to-fine deep residual network (ResNet) architecture. Unlike traditional approaches relying on high-overhead preamble structures, our method exploits the intrinsic periodic features of OTFS pilots in the delay-time (DT) domain to formulate synchronization as a hierarchical classification problem. Specifically, the proposed architecture employs a two-stage strategy to first narrow the search space and then pinpoint the precise symbol timing offset (STO), thereby significantly reducing computational complexity while maintaining high estimation accuracy. We construct a comprehensive simulation dataset incorporating diverse channel models and randomized STO to validate the method. Extensive simulation results demonstrate that the proposed method achieves robust signal start detection and superior accuracy compared to conventional benchmarks, particularly in low signal-to-noise ratio (SNR) regimes and high-mobility scenarios.