Papers

Three-dimensional (3D) radio map (RM) is a key enabler for environment-aware communications in low-altitude wireless scenarios by providing site-specific channel priors indexed by spatial locations. However, existing 3D RM construction methods lack effective modeling of the height dimension, which limits their generalization to unseen height configurations and degrades construction coherence across height layers. In this paper, we propose BDFlow-3DRM, a bi-dynamical flow matching framework for 3D RM construction. Specifically, the 3D RM construction problem is formulated as a deterministic probability flow in a semantic latent space. BDFlow-3DRM learns continuous height-aware representations from flexible transceiver height inputs, enhancing geometric awareness. Meanwhile, its bi-dynamical design explicitly models bidirectional dependencies across neighboring height layers, so that RMs at different height layers can be constructed jointly. Extensive experiments on multiple datasets, covering diverse simplified and realistic urban scenarios, validate the effectiveness of BDFlow-3DRM. Compared with diffusion-based baselines, it reduces the normalized mean square error (NMSE) by 28.6% and attains a 180-fold reduction in inference complexity. More importantly, with only 20 training receiver-height layers, BDFlow-3DRM maintains accurate prediction over a wide continuous receiver-height range from 1 to 120 m under variable transmitter heights, highlighting its practical potential for large-scale 3D RM construction.

Perfusion imaging guides clinical evaluation of stroke and brain tumors by characterizing tissue-level hemodynamics. Routine quantification relies on manual arterial input function (AIF) selection followed by deconvolution, producing summary maps without an executable temporal model for simulation or mechanistic insight. Tracer-dynamics-based models infer transport or compartmental parameters from perfusion time series, but do not yield clinically actionable perfusion indices (e.g., CBF, CBV, MTT) that inform diagnosis and treatment decisions. In this work, we propose HemoPIC, a physics-informed cerebral hemodynamics digital twin that explains perfusion time series through tracer mass conservation and a lumped parameter hemodynamic model. Specifically, HemoPIC solves a constrained inverse problem that jointly estimates digital twin parameters and latent states from perfusion imaging, eliminating manual AIF selection and deconvolution from routine perfusion quantification while directly producing clinically actionable perfusion summary maps. Experiments demonstrate that HemoPIC reconstructs tracer dynamics, generates physiologically consistent perfusion maps with lesion hypoperfusion patterns, satisfies central volume consistency, and yields a mechanistic hemodynamic digital twin that enables forward simulation and counterfactual intervention analysis. Code is publicly available at https://github.com/jhuldr/HemoPIC.

Reconfigurable Intelligent Surface (RIS) has emerged as a promising next-generation technology that improves the throughput and coverage of a wireless system. The realization of the full potential of RISs in a wireless system is tied to their strategic spatial deployment. While existing literature on RIS placement primarily focuses on maximizing coverage, when multiple RIS placements guarantee the required coverage (happens quite often), these approaches fail to exploit prior user trends to choose the one that is most probable to maximize throughput. Thus, to enable throughput maximization while guaranteeing fairness, we formulate a novel hierarchical problem that maximizes the expected sum rate of the system while guaranteeing a certain probabilistic coverage, with the requisite minimum number of RISs deployed. To solve this multi-layered non-convex problem, firstly, we obtain a set of optimal points where we can deploy RISs to provide the coverage guarantee. Then, the least number of RISs that can guarantee the required coverage is obtained by a greedy minimum partitioning. Finally, a Bayesian optimization based approach is used to compute the optimal RIS placement. Numerical results are provided to show that the proposed framework consistently identifies placements that jointly achieve good coverage and throughput, without impractical assumptions.

We report the empirical reliability of Gemini models as audio judges that score full-duplex agent conversations directly from the raw stereo waveform, tested across three models in the Gemini family: 2.5 Flash, 3.5 Flash, and 3.1 Pro. Our primary evidence base uses Gemini 2.5 Flash as the ground-truth model, validated against three calibrated human raters on 209 stereo sessions, scored on 8 production dimensions: 152 full-duplex conversations across 13 accent-and-condition strata, together with 57 adversarial defect-injected clips. The evidence for Gemini 2.5 Flash is consistent across three tests. (i) On 5 of 8 dimensions the LALM-human Spearman rho departs from the pairwise human-human rho by at most 0.07, and on 7 of 8 dimensions the two quantities 95 percent bootstrap confidence intervals overlap. (ii) The LALM agrees with the three-rater human mean within 1 point on 60 to 92 percent of sessions on 6 of 8 dimensions. (iii) On 45 of 48 (defect, dimension) cells the LALM is as sensitive as humans or better under Newcombe-Wilson 95 percent confidence intervals, though most of these are underpowered nulls rather than demonstrated parity. Rank-ordering ability transfers across the Gemini family: 3.5 Flash improves simple agreement to 8 of 8 dimensions, while 3.1 Pro rates several dimensions markedly lower than humans despite comparable rank correlation. A model swap should be re-validated on calibration specifically, not assumed from rank-correlation alone. We identify four areas where deployment requires care, and we estimate that human rating alone for our current evaluation cadence costs roughly two orders of magnitude more than the equivalent LALM workload. The data presented here provides a defensible empirical basis for deploying the LALM as a substitute or fourth rater on the dimensions where the evidence supports it.

Grid codes increasingly require grid-forming (GFM) inverters to demonstrate prescribed active-power response to phase-angle jumps at the point of interconnection (POI). This paper shows that such requirements embed an implicit current-overload mandate whose severity depends on the test parameters but is nowhere made explicit in the specifications. First, an analytic expression for the instantaneous power is derived at an arbitrary measurement point, establishing that a momentary power excursion in the non-opposing direction is an inevitable physical consequence of the phase jump itself, independent of control action. Second, the phase-jump recovery is formulated as a constrained optimal control problem with the characteristic GFM objective of minimizing terminal voltage deviation from the pre-disturbance value while subject to a hard current limit. As the plant dynamics are linear and the constraints are convex, the solution constitutes a controller-architecture-independent physical bound on the achievable power-recovery trajectory. Sweeping the current limit, the phase-jump acceptance criterion is converted into an equivalent minimum overload ratio, making the implicit hardware mandate quantitative. The bound is validated against three WECC generic GFM inverter models (REGFM_A1, B1, C1) in electromagnetic transient simulations, confirming both validity and tightness of the bound. Recommendations are offered for interpreting compliance test results and for structuring test specifications to distinguish physical hardware limitations from control deficiencies.

Hamilton-Jacobi (HJ) reachability provides rigorous safety and reachability guarantees for continuous-time dynamical systems, but its numerical solution suffers from the curse of dimensionality. Deep reinforcement learning (DRL), by contrast, offers scalable sample-based methods. However, RL is typically built around additive cumulative rewards; whereas, reachability objectives are inherently non-additive. This mismatch makes a direct bridge between HJ reachability and RL nontrivial. Recent discounted formulations have either introduced contraction by altering the original reachability semantics, or preserved exact semantics on the HJ side without a corresponding Bellman fixed-point characterization. In this paper, we close this gap by building on a semantics-preserving discounted reach-based value function and deriving a non-additive Bellman operator whose unique fixed point exactly matches the value function in the HJ formulation. We prove that discounting makes this operator contractive, yielding existence, uniqueness, and convergence of value iteration. Furthermore, we establish the equivalence between the HJ and Bellman characterizations, and show that RL can be interpreted as a sample-based approximation scheme for the same fixed-point equation. This yields a principled and semantically exact connection between HJ reachability and RL, enabling learning-based methods to approximate reachability value functions while preserving their safety-critical meaning. As a result, the proposed framework opens the door to scalable, data-driven computation of reachable sets and safety certificates in high-dimensional systems. Numerical experiments demonstrate close agreement with HJ solutions, confirm preservation of reachability semantics via alignment of zero level sets, and support the interpretation of reinforcement learning as a sample-based solver of the proposed Bellman operator.

State estimates used in sampled monitoring and automation need bounds that remain valid between measurements. We develop a finite-horizon input-to-state-stability tube and observer co-design framework for continuous-time observers driven by sampled and held outputs. The sampled-data error model separates process disturbance, sampled measurement noise, and intersample mismatch. A horizon-level disturbance-envelope event is transferred through an ISS estimate to simultaneous containment of the complete error trajectory. Quadratic dissipation inequalities yield ellipsoidal and componentwise tubes, and semidefinite co-design minimizes normalized tube width across the three channels. A structured nonlinear extension preserves known nonlinear channels. Co-design reduces the worst normalized half-width by 31% in a linear compartment benchmark and by a factor of 22.4 in a flexible-joint benchmark.

Dynamic obstacle avoidance in unstructured outdoor environments remains a critical challenge for autonomous mobile robots, particularly when large-scale robot-specific training data and simulation-based policies are impractical. We present a data-efficient, interpretable method for vision-based dynamic obstacle avoidance that operates entirely on real-world data, avoiding the sim-to-real transfer problem inherent in simulation-trained policies. Our approach leverages UniDepth, a large pretrained monocular depth estimation model, to produce dense depth maps from RGB video without requiring stereo cameras or LiDAR at inference time. Dynamic obstacle avoidance is achieved by extending the SuperPoint and SuperGlue feature correspondence pipeline to track keypoints across long frame sequences, projecting their 2D pixel-space positions into 3D using camera intrinsics and predicted depth, running bundle adjustment initialized from these 3D keypoints, and computing per-keypoint time-to-collision (TTC). A 2D motion primitive in the ground plane is then selected to move the robot away from the closest point of approach of the minimum-TTC keypoint. Evaluated on real-world data from the M3ED dataset, our pipeline achieves a precision of 0.49 and a recall of 0.38 in identifying frames with a ground truth TTC below 1 second, and correctly generates the evasive motion direction in 84% of true positive detections. Crucially, it detects at least one frame with TTC less than 1 second for 20 out of 22 unique physical obstacles present in our test sequences. Unlike end-to-end learned methods that demand thousands of hours of robot-specific training data, our approach eliminates model training entirely, requiring only 74 seconds of data for hyperparameter tuning. This demonstrates exceptional data efficiency while preserving interpretable and generalizable behavior across diverse obstacle types.

Automated segmentation of cervical-spine MRI is increasingly used in clinical workflows, yet no fairness audit exists for this anatomy. We show that auditing these segmentation tasks is complicated by a common property of modern segmentation datasets: expert-annotated gold labels are expensive, so abundant machine-generated (silver) labels are added to limit annotation cost. This matters because the reference used to judge a model can itself be biased. In this study, we present the first fairness audit of cervical-spine MRI segmentation across sex, age, and race using the CSpineSeg dataset. We observe that the deployed model is demographically fair, but the choice of reference label, however, is not neutral. Because a dataset's silver labels are generated by a model trained on its gold labels, any new model trained on those same gold labels agrees more with the silver labels than with expert truth: scoring identical predictions against silver rather than gold overestimates performance by ~8 Dice points and turns the fairness verdict for age from non-significant to significant - not by the gap inflation Parikh et al. report (which we term false magnitude) but by collapsing within-group variance (which we term false confidence). Reference-label provenance is thus a first-order confounder in segmentation evaluation: performance and fairness should be reported against expert labels, and any fairness claim stated together with the provenance of its reference.

This paper presents a Multi-Map Dynamic-Entropy Intrusion-Aware Chaotic Modulation (MU-DE-IAEACM-MM) framework for adaptive physical-layer security in multi-user wireless systems. Unlike conventional chaos-based schemes that rely on static parameter secrecy, the proposed architecture treats entropy as a regulated security variable and dynamically updates chaotic control parameters using an entropy-driven adaptation law. Heterogeneous chaotic maps, including Logistic, Tent, Chebyshev, and Sine generators, are distributed across users to enlarge entropy dimensionality and reduce cross-user statistical dependence. A correlation-based intrusion metric is incorporated to detect improved adversarial reconstruction coherence and trigger controlled entropy escalation. Stability analysis establishes bounded convergence conditions for the adaptive update process. Monte Carlo simulations under AWGN, fading, impulsive, colored, and narrowband interference demonstrate a persistent BER gap between legitimate and mismatched receivers and measurable secrecy capacity gains over representative fixed-parameter chaotic modulation schemes. The framework maintains positive secrecy rates in dense deployments with up to 64 legitimate users and 30 passive eavesdroppers. The results indicate that entropy-regulated multi-map chaotic modulation provides a scalable and synchronisation-stable approach for adaptive physical-layer security in next-generation wireless and IoT networks.

This paper investigates semantic-aware neural joint source-channel coding (JSCC) for robust video transmission over block erasure channels. We propose a neural video compression framework exploring both spatial-domain and feature-domain designs. In the spatial domain, video frames are partitioned into blocks, enabling localized erasure handling and fine-grained robustness control via uniform erasure and two-level, semantic-guided non-uniform erasure strategies. In the feature domain, latent features are partitioned, enabling missing features to be semantically recovered while maintaining overall spatial consistency. Comprehensive experiments quantify reconstruction quality under varying uniform and non-uniform erasure probabilities. Our results show that spatial-domain JSCC excels at handling random localized losses, whereas feature-domain JSCC provides superior robustness to distributed erasures and maintains fidelity under low-loss scenarios. The analysis highlights the trade-offs between spatial continuity and semantic redundancy, offering insights for designing robust, task-aware video communication systems.

Transport-based generative models, which learn a time-dependent vector field that moves noise to data, have become a dominant paradigm. However, these models typically do not explicitly encode the data distribution. Energy-based models (EBMs) instead represent the data distribution explicitly through a scalar energy landscape, which Energy Matching learns by combining transport learning with contrastive refinement. Its transport objective, however, fits energy gradients to stochastic targets, whose variance degrades the training signal at scale. We introduce Projected Energy Matching, which learns this landscape through a more stable route: we first train a time-independent transport teacher, then freeze it and fit the negative energy gradient to its predicted velocities. This projection replaces noisy transport targets with deterministic supervision, while contrastive refinement shapes the landscape near the data manifold. On CIFAR-10, gradient-noise analysis reveals a cleaner training signal, accompanied by faster convergence than Energy Matching at matched, teacher-free training budgets. In the latent space of CT volumes, our method enables 3D CT generation and achieves better FID scores than flow models. The learned scalar potential serves as a zero-shot prior for the ill-posed inverse problem of sparse-view cone-beam CT reconstruction. By making explicit energy landscapes practical at volumetric scale, this work opens a path to wider adoption of energy-based formulations, bringing their flexibility to high-dimensional generation and inverse problems.

Recent advances in multi-band wireless communication systems have driven the increasing need for dual-band bandpass filters. These types of filters are capable of isolating a small section of the frequency spectrum within a broader spectrum. Over the years, coplanar waveguide, microstrip, slotline, stripline, and other planar transmission line technologies have been widely employed in the design of microwave circuits and systems. This work employs a Folded-Arms Square Open-Loop Resonator (FASOLR) microstrip planar structure, designed and simulated using PathWave Advanced Design System (ADS). A third order (three-pole) single bandpass filter is transformed into a sixth order (six-pole) dual-band bandpass filter. The proposed six-pole dual-band bandpass filter is centred at 2.3 GHz, with a fractional bandwidth of 7% and produces two passbands centred at approximately 2.2 GHz and 2.4 GHz. The design is implemented on a commercially available Rogers RT/Duroid 6010LM substrate with a dielectric constant of 10.7, a loss tangent of 0.0023, a substrate thickness of 1.27 mm, and 35 micro meter copper cladding on both sides. The overall filter component has a compact footprint of 30.56 mm by 20.56 mm. The design is validated through comparison of circuit-level and electromagnetic (EM) simulation results, which show good agreement. The EM simulation responses indicate a return loss better than 10 dB on the first band and better than 19 dB on the second, and an insertion loss better than 1.5 dB, demonstrating its suitability for WiMAX applications.

Electrical distribution networks are regional, medium- and low-voltage power grids connecting energy sources to individual households and businesses with given power demands. While these networks contain redundant power lines for reliability, they are typically operated in a radial (spanning tree) configuration by opening and closing switches on the lines. The challenge is to find a spanning tree that minimizes the sum of the resistive power losses: The power loss of a line $e$ is its resistance $r(e)$ times the squared current $f(e)^2$ flowing across the line. We study approximation algorithms for this problem, known as Distribution Network Reconfiguration (DNR). We give an $n$-approximation algorithm and, via a new NP-hardness for planar Balanced Connected Partition with a fixed number of parts, show that no $n^{1-\varepsilon}$-approximation is possible even on planar graphs unless P $=$ NP, for any $\varepsilon>0$. Since the approximation hardness holds only if there are many sources, we focus on $k$-DNR with $k$ sources; this is motivated by traditional distribution networks, where oftentimes $k = 1$. For $2$-DNR, we give an approximation lower bound of $\Omega(\log^2 n)$ conditioned on P $\neq$ NP. For $1$-DNR, which is equivalent to finding an uncapacitated confluent flow minimizing the squared Euclidean norm, we prove APX-hardness and give an $\mathcal{O}(\sqrt{n})$-approximation for uniform line resistances, answering an open question by Gupta et al. [Math. Program. 2022].

The emergence of Quantum Software Engineering (QSE) responds to the need for systematic, disciplined, and quantifiable approaches to the development, operation, and maintenance of quantum software. Within this context, quantum computer certification represents a significant challenge: verifying that quantum devices produce valid entangled states despite hardware imperfections, noise, and decoherence. This paper presents QAccCert, a hybrid certification framework developed following QSE principles, demonstrating how heterogeneous technologies like FPGAs and Artificial Intelligence can be integrated for quantum processing. The framework implements entanglement certification through CHSH inequality violation in ideal quantum simulations using Qiskit AerSimulator. Through LLM-guided optimization, the system achieves 99.94% of the theoretical maximum of $2\sqrt{2}$, evidencing more efficient parameter space exploration than random search. These simulated results illustrate how QSE methodologies, combined with strategic technology interconnection, can be applied for practical and scalable quantum certification on real NISQ hardware in future work. This study provides a concrete case study of systematic quantum software development.

Accurate state-of-charge (SOC) estimation remains a fundamental challenge in lithium-ion battery management systems because battery dynamics are highly nonlinear, operating-condition dependent, and sensitive to parameter variations caused by aging and temperature. Conventional model-based estimators, such as equivalent circuit model (ECM) and Kalman-filter-based approaches, rely heavily on repeated parameter identification and accurate electrochemical modeling, whereas purely data-driven methods often sacrifice physical interpretability. This work proposes a Koopman-theoretic, data-driven framework for SOC estimation using Dynamic Mode Decomposition with control (DMDc) combined with Hankel time-delay embedding. Instead of explicitly identifying ECM parameters, the proposed approach reconstructs a lifted dynamical state space directly from measured terminal voltage and current obtained through Hybrid Pulse Power Characterization (HPPC) testing. Spectral decomposition of the identified DMDc operator reveals intrinsic battery dynamics in terms of Koopman modes and eigenvalues. The SOC dynamics naturally emerge as the slowest marginally stable mode whose eigenvalue lies closest to the unit circle, consistent with the integrator-type behavior of charge conservation. The corresponding modal coordinate is subsequently utilized as an SOC-sensitive observable.

Neural audio codecs (NACs) enable efficient audio compression and have achieved success in downstream tasks such as speech synthesis. However, their discrete representations consistently underperform traditional spectral features in automatic speaker verification (ASV). We empirically demonstrate that speaker cues are implicitly preserved in discrete tokens but remain underutilized by conventional ASV training paradigms. To address this, we propose a Cross-Feature Knowledge Distillation (CFKD) framework. By guiding the codec-based student to mimic the embedding space of a strong Fbank-based teacher, CFKD provides structured supervision for effective utilization of speaker information in tokens. Experiments on the VoxCeleb benchmarks show that CFKD substantially improves the ASV performance of codec-based systems, allowing them to approach the accuracy of Fbank-based teacher models and highlighting the potential of discrete audio tokens for diverse speech tasks.

In nonlinear optimal control, uncertainties in system dynamics may affect not only closed-loop stability but also the achieved optimality properties of the resulting solutions. This paper develops a systematic robustness analysis for nonlinear optimal control beyond the conventional focus on stability in robust control theory. First, we demonstrate that the optimal value function retains its Lyapunov property under a quantifiable criterion, thereby guaranteeing the preservation of closed-loop stability. Building upon this foundation, we establish explicit characterizations for optimality deviations induced by model mismatch in both closed-loop performance and optimal controllers, and further reveal their consistency with classical linear-quadratic regulator (LQR) results. In addition, the robustness analysis admits a unified computational formulation that gives rise to an iterative scheme with guaranteed convergence, enabling quantitative assessment of optimality robustness in nonlinear control systems. Numerical examples validate the theoretical analysis.

Accurate mapping of informal settlements remains a major challenge in Sub-Saharan African (SSA) cities because optical imagery often fails to distinguish Informal Settlements (defined here as LCZ 7) from spectrally similar formal Compact Low-Rise areas (LCZ 3). This study presents a context-aware, reproducible Optical-SAR framework that improves informal settlement delineation using Sentinel-2 spectral features and Sentinel-1 structural information within an adapted Local Climate Zone (LCZ) taxonomy. We implement a three-tier SAR integration strategy: calibrated backscatter, GLCM textures, and a physics-guided feature engineered to capture the high structural disorder and weak radar return characteristic of SSA informal settlements. Using reference data across Nairobi and Eldoret (Kenya), we evaluate performance via a stratified hold-out protocol and a season-aware ablation study. Results show that SAR textures provide the dominant performance gain for LCZ 7 detection. The Optical-SAR model achieves overall accuracy of 0.816 (dry) and 0.807 (wet), significantly outperforming the WUDAPT baseline (OA 0.704) and reducing the critical LCZ 3 - LCZ 7 confusion to 7%. Seasonal analysis indicates that while optical-only separability varies with phenology, SAR-derived textures stabilize informal settlement mapping across seasons. These findings demonstrate that the incorporation of SAR-derived features yields consistent improvements for urban morphology mapping in data-scarce environments across seasons and across the evaluated source cities, while cross-city transfer remains limited without local adaptation strategies.

The whole-system impedance model has proven a powerful tool for assessing the small-signal stability of multi-inverter power systems; however, its application is limited to a small range around a steady-state operating point due to the inherent assumptions of time invariance and linearisation. In this paper, a dedicated physics-informed neural network (PINN) for small-signal stability analysis in high-dimensional multi-inverter power systems is developed. The PINN is trained with step-response data produced from limited sets of system electromagnetic transient (EMT) simulations, and the trained model can predict the poles and residues of the whole-system impedance/admittance model, i.e., the transfer functions, across the full operating space. Such a PINN offers unique insights into system stability that surpass what conventional analytical methods or EMT simulations can achieve. By characterising how the impedance model evolves with power flow variations, it predicts the dynamic behaviour of the time-varying system and reveals oscillation risks that may emerge while identifying their root causes. It also provides direct visualisation of the possible range of oscillatory modes under a given power flow condition, enabling an optimal generation distribution while maintaining safe operation of the system. The proposed PINN is fully validated on a 2-IBR system and a 4-IBR system, with its application details presented.

Virtualized radio access networks (vRAN) run the compute-intensive multiple-input multiple-output (MIMO) baseband as software on shared servers, which makes energy efficiency (EE) a primary design objective. Distributed MIMO vRAN consumes power across virtualized distributed unit (vDU) baseband, fronthaul transport, and per-radio-unit operation. We build a power model that resolves these three components. We then develop a framework that jointly selects modulation, transmission rank, and per-subcarrier power to maximize system EE. Exponential effective SNR mapping induces a convex per-subcarrier power constraint, which yields a convex power minimization problem with a closed-form waterfilling-like solution. We show that radio frequency-only models underestimate the spectral efficiency range where single-input multiple-output (SIMO) transmission saves power, and our power model extends this range by 24%. We further extend the framework to a traffic-aware setting with realistic user trajectories from the multi-agent transport simulator. We propose a traffic-aware strategy that switches each radio unit among MIMO, SIMO, and sleep modes based on demand. Simulation results over 3GPP NR compliant fading channels show that, after a one-time offline calibration, the framework predicts link performance without further link-level simulation. The proposed framework achieves higher average EE than a traffic-agnostic always-on MIMO baseline, while maintaining comparable throughput at peak hours.

Generating signals on graphs requires permutation-equivariant models that exhibit stability with respect to relative structural perturbations. While favorable stability properties of Graph Neural Networks (GNNs) have been well documented, it is unclear how structural errors propagate through the dynamics of continuous generative flow models that are gaining traction for graph signal generation. In this paper, we analyze continuous normalized flow models parameterized by GNNs and show that permutation equivariance is preserved for both the resulting continuous-time ordinary differential equations and their discrete numerical approximations used as graph signal samplers. Our primary contribution is to derive explicit stability bounds on the generated probability distributions, which quantify how relative graph perturbations affect the final sampled signals. Motivated by these theoretical bounds, we introduce a stability-promoting regularized flow matching strategy that actively penalizes the spatial Lipschitz constant of the vector field during model training. Experiments using synthetic smooth signals on stochastic block model graphs and real-world fMRI signals on brain connectomes demonstrate that this bound-oriented approach yields generative models that are more robust to structural noise, without sacrificing output quality.

FFT-based spectral preprocessing of learned query-key (Q/K) projections substantially improves transformer attention on character-level language modelling. On TinyShakespeare: a fixed random spectral filter achieves val=1.031 (Delta=+0.443); a single learned frequency at paragraph scale achieves val=0.608 (Delta=+0.867); and four learned frequencies spanning paragraph to word scale achieve val=0.309 (Delta=+1.166), a 79% reduction over standard dot-product attention. The single-frequency result is confirmed across three random seeds (mean val=0.236, std=0.019). The four frequencies converge to a near-geometric multi-scale ordering (49, 27, 10, 6 tokens/cycle) corresponding to paragraph, sub-paragraph, phrase, and word scales. The gain is specific to spectral preprocessing: random orthogonal and non-orthogonal projections of Q/K produce no measurable improvement, suggesting the benefit comes from global frequency-domain mixing rather than metric distortion. All results are verified by a shuffled-validation diagnostic against positional leakage. Causal filters (Gaussian, Mexican Hat, Morlet) do not improve over standard attention at character-level tokenisation: the bilateral FFT kernel is structurally non-causal, coupling every position to future tokens. This defines an architectural boundary between bilateral spectral attention (this paper) and genuinely causal spectral attention at word-scale tokenisation (companion paper MorletQK). This work is architecturally distinct from FNet (Lee-Thorp et al., 2021), which replaces attention with Fourier mixing of token embeddings. Here, spectral preprocessing applies only to Q/K projections while the full attention score structure is preserved.

5G New Radio (NR) Non-Terrestrial Networks (NTNs) extend cellular connectivity through Low Earth Orbit (LEO) and Medium Earth Orbit (MEO) satellite constellations while enabling the reuse of downlink NR Positioning Reference Signals (PRS) to provide Positioning, Navigation, and Timing (PNT) services alongside broadband communications. However, the large inter-satellite differential propagation delays inherent to NTN geometry may cause PRS transmissions from non-serving satellites to overlap with the serving-satellite data stream. This paper analyzes this coexistence by deriving a statistical model for the slant-range distribution over the visible spherical cap and extending it to dual-shell constellations through a mixture formulation, yielding a closed-form cumulative distribution function (CDF) of the differential delay. The model is validated using a 10-day orbit simulation representative of a dual-shell European NTN constellation. Detection limits of non-serving satellite PRS under interference from the serving-satellite data stream are characterized in terms of the effective carrier-to-noise density ratio. The impact of periodic PRS transmissions on the uncoded bit error rate (BER) is also evaluated for standardized NR Frequency Range 1 (FR1) and Frequency Range 2 (FR2) configurations. Monte Carlo simulations show that the probability of simultaneous multi-PRS overlap remains below a few percent, depending on PRS duration and repetition period, while PRS detection remains feasible despite data interference. When the PRS is received about 25 dB below the data signal, its impact on uncoded BER is negligible over a wide range of repetition periods, whereas BER degradation increases with PRS duty cycle. These results demonstrate that NR-PRS-based PNT can coexist with broadband downlink in NTN under appropriate PRS periodicity design.

Semantic and terahertz (THz)-band communications are algorithmic and spectral enablers of future wireless networks. This work investigates deep learning-based semantic communication (DeepSC) over THz channels. We show that DeepSC models trained solely under additive white Gaussian noise generalize well to the tested THz block- and fast-fading channels when receiver-side compensation is applied. To enable fully data-driven reception, we propose a lightweight neural detector that does not require channel state information (CSI). At 0.3 THz, DeepSC outperforms a throughput-matched traditional coded communication system baseline over 0-12 dB signal-to-noise ratio (SNR), achieving more than 50 percentage-point higher Bilingual Evaluation Understudy unigram (BLEU-1) score. The proposed pilot-free detector outperforms minimum mean square error (MMSE) equalization with both perfect and imperfect CSI and remains robust to frequency offsets up to 50 MHz, highlighting the resilience of semantic communication to THz channel impairments.

Sparse vector transmission (SVT) has emerged as a promising technique for ultra-reliable low-latency short-packet communications. However, existing SVT schemes typically assume negligible phase noise (PN), an assumption that rarely holds in practical wireless systems. In this paper, a dual-mapping sparse vector coding (DM-SVC) scheme is proposed for short-packet communications subject to PN. In DM-SVC, pilot symbols are mapped onto multiple non-zero blocks and data symbols onto isolated non-zero elements within a single sparse vector, thereby enabling pilot-data separation through distinct sparsity patterns rather than explicit resource partitioning. Moreover, the indices of pilot blocks convey additional information bits, further improving spectral efficiency. A basis expansion model is adopted to represent the PN process, substantially reducing the number of parameters to be estimated. Furthermore, an iterative joint PN estimation and data decoding algorithm is developed, where pilot block indices are first detected exploiting block-sparse priors, after which PN estimation and data decoding proceed iteratively. Simulation results show that DM-SVC could achieve block error rate performance close to that of perfect PN compensation, while offering improved spectral efficiency and reduced codebook storage overhead compared to state-of-the-art SVT schemes.

Wireless powered communication networks (WPCNs) are a key enabler for sustainable Internet of Things (IoT) systems, yet their practical performance is constrained by inefficient wireless energy transfer, limited spatial adaptability, and fragile uplink connectivity in blockage-prone and dynamic environments. Emerging reconfigurable antenna technologies, including pinching antenna systems (PASSs), fluid antenna systems (FASs), movable antennas (MAs), and reconfigurable intelligent surfaces (RISs), provide new opportunities to overcome these limitations, but have mostly been studied separately. In this article, we propose a unified architecture for fully reconfigurable WPCNs by jointly integrating PASS-enabled power beacons, FAS-based IoT devices, MA-assisted base stations, and RIS-aided propagation environments. The proposed framework enables end-to-end reconfigurability across downlink energy transfer, device-side spatial adaptation, base-station reception, and uplink information transmission. We further discuss the integration motivation, system architecture, design and optimization framework, illustrative performance evaluation, implementation tradeoffs, and major practical challenges. This article provides a new perspective for designing next-generation fully reconfigurable WPCNs.

Humanoid robots can extend human presence to remote, constrained, or hazardous environments, but existing teleoperation interfaces often require physically demanding motion tracking or cognitively demanding low-level control. This paper presents an immersive teleoperation framework that integrates voice-controlled locomotion, VR-based manipulation, and bidirectional social interaction for whole-body humanoid control. Using Apple Vision Pro, the operator receives egocentric visual feedback, issues natural-language locomotion commands, and teleoperates the robot's arms and dexterous hands through wrist and finger tracking. An LLM-assisted voice-control module converts spoken instructions into high-level locomotion commands, while the manipulation module retargets human hand motions to the robot through inverse kinematics and PD control. The system also records multimodal data, including egocentric RGB observations, voice/text commands, joint states, hand motions, and eye-gaze signals, supporting future imitation learning and autonomy. We evaluate the framework on a Unitree H1 humanoid equipped with dexterous hands in manipulation and social interaction tasks. Results show that novice users can successfully operate the system after brief familiarization, achieving 80% success in object manipulation and 70% success in a social cube-passing task. These results demonstrate the potential of immersive, language-assisted teleoperation as an accessible interface for humanoid interaction, remote assistance, and multimodal data collection.

The unit commitment (UC) problem is paramount for optimal operation of power systems, but it faces computational limitations in large-scale settings, especially in investment or stochastic models, because of the binary variables that it contains. A lot of research has attempted to improve the computational performance of UC models, either by reducing model size, resulting in lower fidelity and accuracy, or by improving the tightness of the formulation. Tightness and model size are the best a priori indicators of the computational performance of UC models, but there is no clear overview of what the best formulation is for different generators. In this research, we define models with different levels of detail, and present a formulation for each level that is based on the convex hull. We show new proofs on the tightness of well-known formulations for ramping, for start-up and shut-down costs and capabilities, and for UC with investment. These models, with a different level of detail, can be incorporated into large-scale problems to reduce the computational burden, as demonstrated in Part II.

Greenhouse fruit-production control aims to maximize the economic performance (fruit revenue minus operating costs) while operating within system constraints under external weather disturbances. Control methods need to balance the delayed economic benefit of fruit yield with current operating costs. For such problems, model predictive control (MPC) can explicitly handle system constraints under future weather disturbances, but can become computationally demanding when using sufficiently long prediction horizons for (relatively large) nonlinear greenhouse fruit production models. In contrast, reinforcement learning (RL) can learn control policies offline while considering longer-term economic performance, but struggles to enforce system constraints, and performance may degrade under unseen weather trajectories. This work proposes trajectory-selection RL-MPC, a framework that incorporates longer-term economic information of fruit yield into a short-horizon MPC optimization problem. The framework uses an RL rollout trajectory to define a terminal region constraint and terminal cost. Next, a nonlinear MPC solves a short-horizon optimization problem with these terminal ingredients to find a local optimum. Finally, the framework selects and executes the first input from the trajectory with the better objective value, either from the MPC-predicted or the RL rollout trajectory. The method is applied to GreenLight, a large-scale greenhouse tomato production model that exhibits stiff dynamics. The simulation results show that trajectory-selection RL-MPC with a one-hour prediction horizon matches the closed-loop performance of a high-performing guiding policy while significantly improving over standalone MPC with the same horizon.