Papers

The beyond-diagonal reconfigurable intelligent surface (BD-RIS) is a recent architecture in which scattering elements are interconnected to enhance the degrees of freedom for wave control, yielding performance gains over traditional single-connected RISs. For BD-RIS, channel estimation, which is well studied for conventional RIS, becomes more challenging due to complex connections and a larger number of coefficients. Prior works have relied on pilot-assisted estimation followed by data decoding. This paper introduces a semi-blind tensor-based approach to joint channel and symbol estimation that reduces the need for dedicated training sequences by directly leveraging data symbols. We consider a practical scenario with time-varying user terminal-RIS channels under mobility. By reformulating the received signal from a tensor decomposition perspective, we develop two semi-blind receivers: a two-stage method that transforms the fourth-order PARATUCK model into a third-order PARAFAC model, and a single-stage iterative process based on the fourth-order TUCKER decomposition. Identifiability conditions for reliable joint recovery are derived, and numerical results demonstrate the performance advantages and trade-offs of the proposed schemes over existing solutions.

Designing lithium-ion batteries for long service life remains a challenge, as most cells are optimized for beginning-of-life metrics such as energy density, often overlooking how design and operating conditions shape degradation. This work introduces a degradation-aware design framework built around finite, interacting reservoirs (lithium, porosity, and electrolyte) that are depleted over time by coupled degradation processes. We extend a physics-based Doyle-Fuller-Newman model to include validated mechanisms such as SEI growth, lithium plating, cracking, and solvent dry-out, and simulate how small design changes impact lifetime. Across more than 1,000 cycles, we find that increasing electrolyte volume by just 1% or porosity by 5% can extend service life by over 30% without significantly affecting cell energy density. However, lithium excess, while boosting initial capacity, can accelerate failure if not supported by sufficient structural or ionic buffers. Importantly, we show that interaction between reservoirs is crucial to optimal design: multi-reservoir tuning yields either synergistic benefits or compound failures, depending on operating conditions. We also quantify how C-rate and operating temperature influence degradation pathways, emphasizing the need for co-optimized design and usage profiles. By reframing degradation as a problem of managing finite internal reservoirs, this work offers a predictive and mechanistic foundation for designing lithium-ion batteries that balance energy, durability, and application-specific needs.

Channel charting creates a low-dimensional representation of the radio environment in a self-supervised manner using manifold learning. Preserving relative spatial distances in the latent space, channel charting is well suited to support user localization. While prior work on channel charting has mainly focused on two-dimensional scenarios, real-world environments are inherently three-dimensional. In this work, we investigate two distinct three-dimensional indoor localization scenarios using simulated, but realistic ray tracing-based datasets: a factory hall with a three-dimensional spatial distribution of datapoints, and a multistory building where each floor exhibits a two-dimensional datapoint distribution. For the first scenario, we apply the concept of augmented channel charting, which combines classical localization and channel charting, to a three-dimensional setting. For the second scenario, we introduce multistory channel charting, a two-stage approach consisting of floor classification via clustering followed by the training of a dedicated expert neural network for channel charting on each individual floor, thereby enhancing the channel charting performance. In addition, we propose a novel feature engineering method designed to extract sparse features from the beamspace channel state information that are suitable for localization.

Spectroscopic photoacoustic (sPA) imaging can potentially estimate blood oxygenation saturation (sO2) in vivo noninvasively. However, quantitatively accurate results require accurate optical fluence estimates. Robust modeling in heterogeneous tissue, where light with different wavelengths can experience significantly different absorption and scattering, is difficult. In this work, we developed a deep neural network (Hybrid-Net) for sPA imaging to simultaneously estimate sO2 in blood vessels and segment those vessels from surrounding background tissue. sO2 error was minimized only in blood vessels segmented in Hybrid-Net, resulting in more accurate predictions. Hybrid-Net was first trained on simulated sPA data (at 700 nm and 850 nm) representing initial pressure distributions from three-dimensional Monte Carlo simulations of light transport in breast tissue. Then, for experimental verification, the network was retrained on experimental sPA data (at 700 nm and 850 nm) acquired from simple tissue mimicking phantoms with an embedded blood pool. Quantitative measures were used to evaluate Hybrid-Net performance with an averaged segmentation accuracy of >= 0.978 in simulations with varying noise levels (0dB-35dB) and 0.998 in the experiment, and an averaged sO2 mean squared error of <= 0.048 in simulations with varying noise levels (0dB-35dB) and 0.003 in the experiment. Overall, these results show that Hybrid-Net can provide accurate blood oxygenation without estimating the optical fluence, and this study could lead to improvements in in-vivo sO2 estimation.

The growing penetration of renewable and distributed generation is transforming power systems and challenging conventional protection schemes that rely on fixed settings and local measurements. Machine learning (ML) offers a data-driven alternative for centralized fault classification (FC) and fault localization (FL), enabling faster and more adaptive decision-making. However, practical deployment critically depends on robustness. Protection algorithms must remain reliable even when confronted with missing, noisy, or degraded sensor data. This work introduces a unified framework for systematically evaluating the robustness of ML models in power system protection. High-fidelity EMT simulations are used to model realistic degradation scenarios, including sensor outages, reduced sampling rates, and transient communication losses. The framework provides a consistent methodology for benchmarking models, quantifying the impact of limited observability, and identifying critical measurement channels required for resilient operation. Results show that FC remains highly stable under most degradation types but drops by about 13% under single-phase loss, while FL is more sensitive overall, with voltage loss increasing localization error by over 150%. These findings offer actionable guidance for robustness-aware design of future ML-assisted protection systems.

The success of machine learning for real-world robotic systems has created a new form of intellectual property: the trained policy. This raises a critical need for novel methods that verify ownership and detect unauthorized, possibly unsafe misuse. While watermarking is established in other domains, physical policies present a unique challenge: remote detection. Existing methods assume access to the robot's internal state, but auditors are often limited to external observations (e.g., video footage). This “Physical Observation Gap” means the watermark must be detected from signals that are noisy, asynchronous, and filtered by unknown system dynamics. We formalize this challenge using the concept of a glimpse sequence, and introduce Colored Noise Coherency (CoNoCo), the first watermarking strategy designed for remote detection. CoNoCo embeds a spectral signal into the robot's motions by leveraging the policy's inherent stochasticity. To show it does not degrade performance, we prove CoNoCo preserves the marginal action distribution. Our experiments demonstrate strong, robust detection across various remote modalities, including motion capture and side-way/top-down video footage, in both simulated and real-world robot experiments. This work provides a necessary step toward protecting intellectual property in robotics, offering the first method for validating the provenance of physical policies non-invasively, using purely remote observations.

Predictive maintenance of rotating machinery increasingly relies on vibration signals, yet most learning-based approaches either discard phase during spectral feature extraction or use raw time-waveforms without explicitly leveraging phase information. This paper introduces two phase-aware preprocessing strategies to address random phase variations in multi-axis vibration data: (1) three-axis independent phase adjustment that aligns each axis individually to zero phase (2) single-axis reference phase adjustment that preserves inter-axis relationships by applying uniform time shifts. Using a newly constructed rotor dataset acquired with a synchronized three-axis sensor, we evaluate six deep learning architectures under a two-stage learning framework. Results demonstrate architecture-independent improvements: the three-axis independent method achieves consistent gains (+2.7% for Transformer), while the single-axis reference approach delivers superior performance with up to 96.2% accuracy (+5.4%) by preserving spatial phase relationships. These findings establish both phase alignment strategies as practical and scalable enhancements for predictive maintenance systems.

This paper addresses two critical limitations in diagonally loaded (DL) adaptive matched filter (AMF) detector: (1) the lack of CFAR property with respect to arbitrary covariance matrices, and (2) the absence of selection criteria for optimal loading factor from the perspective of maximizing the detection probability (Pd). We provide solutions to both challenges through a comprehensive analysis for the asymptotic performance of DL-AMF under large dimensional regime (LDR) where the dimension N and sample size K tend to infinity whereas their ratio N/K converges to a constant c\in(0,1). The analytical results show that any DL detectors constructed by normalizing the random variable |a|2=|sH(R+{\lambda}IN)-1y0|2 with a deterministic quantity or a random variable that converges almost surely to a deterministic value will exhibit equivalent performance under LDR. Following this idea, we derive two CFAR DL detectors: CFAR DL semi-clairvoyant matched filter (CFAR-DL-SCMF) detector and CFAR DL adaptive matched filter (CFAR-DL-AMF) detector, by normalizing |a|2 with an appropriate deterministic quantity and its consistent estimate, respectively. The theoretical analysis and simulations show that both CFAR-DL-SCMF and CFAR-DL-AMF achieve CFAR with respect to covariance matrix, target steering vector and loading factor. Furthermore, we derive the asymptotically optimal loading factor \lambda_opt by maximizing the explicit expression of asymptotic Pd. For practical implementation, we provide a consistent estimator for \lambda_opt under LDR. Based on \lambda_opt and its consistent estimate, we establish the optimal CFAR-DL-SCMF (opt-CFAR-DL-SCMF) and the optimal CFAR-DL-AMF (opt-CFAR-DL-AMF). Numerical examples demonstrate that the proposed opt-CFAR-DL-SCMF and opt-CFAR-DL-AMF consistently outperform EL-AMF and persymmetric AMF in both full-rank and low-rank clutter plus noise environments.

Direction of Arrival (DOA) estimation is a fundamental problem in signal processing. Diffuse sources, whose power density cannot be represented with a single angular coordinate, are usually characterized based on prior assumptions, which associate the source angular density with a specific set of functions. However, these assumptions can lead to significant estimation biases when they are incorrect. This paper introduces the Moment-Matching Estimation Technique (MoMET), a low-complexity method for estimating the mean DOA, spread, and power of a narrow diffuse source without requiring prior knowledge on the source distribution. The unknown source density is characterized by its mean DOA and its first central moments, which are estimated through covariance matching techniques which fit the empirical covariance of the measurements to that modeled from the moments. The MoMET parameterization is robust to incorrect model assumptions, and numerically efficient. The asymptotic bias and covariance of the new estimator are derived and its performance is demonstrated through simulations.

To enhance coverage and signal quality in millimeter-wave (mmWave) frequencies, reconfigurable intelligent surfaces (RISs) have emerged as a game-changing solution to manipulate the wireless environment. Traditional semiconductor-based RISs face scalability issues due to high power consumption. Meanwhile, liquid crystal-based RISs (LC-RISs) offer energy-efficient and cost-effective operation even for large arrays. However, this promise has a caveat. LC-RISs suffer from long reconfiguration times, on the order of tens of milliseconds, which limits their applicability in dynamic scenarios. To date, prior works have focused on hardware design aspects or static scenarios to address this limitation, but little attention has been paid to optimization solutions for dynamic settings. Our paper fills this gap by proposing a reinforcement learning-based optimization framework to dynamically control the phase shifts of LC-RISs and maximize the data rate of a moving user. Specifically, we propose a Deep Deterministic Policy Gradient (DDPG) algorithm that adapts the LC-RIS phase shifts without requiring perfect channel state information and balances the tradeoff between signal-to-noise ratio (SNR) and configuration time. We validate our approach through high-fidelity ray tracing simulations, leveraging measurement data from an LC-RIS prototype. Our results demonstrate the potential of our solution to bring adaptive control to dynamic LC-RIS-assisted mmWave systems.

Preprocessing is a well-established technique for optimizing compression, yet existing methods are predominantly Rate-Distortion (R-D) optimized and constrained by pixel-level fidelity. This work pioneers a shift towards Rate-Perception (R-P) optimization by, for the first time, adapting a large-scale pre-trained diffusion model for compression preprocessing. We propose a two-stage framework: first, we distill the multi-step Stable Diffusion 2.1 into a compact, one-step image-to-image model using Consistent Score Identity Distillation (CiD). Second, we perform a parameter-efficient fine-tuning of the distilled model's attention modules, guided by a Rate-Perception loss and a differentiable codec surrogate. Our method seamlessly integrates with standard codecs without any modification and leverages the model's powerful generative priors to enhance texture and mitigate artifacts. Experiments show substantial R-P gains, achieving up to a 30.13% BD-rate reduction in DISTS on the Kodak dataset and delivering superior subjective visual quality.

LoRa is one of the most widely used low-power wide-area network technology for the Internet of Things. To achieve long-range communication with low power consumption at a low cost, LoRa uses a chirp spread spectrum modulation and transmits in the sub-GHz unlicensed industrial, scientific, and medical (ISM) frequency bands. Due to the rapid densification of IoT networks, it is crucial to obtain tailored channel models to evaluate the performance of LoRa networks. While channel models for cellular technologies have been investigated extensively, specific characteristics of LoRa transmissions operating at long range with a rather small (~ 250kHz) bandwidth require dedicated measurement campaigns and modeling efforts. In this work, we leverage an SDR-based testbed to gather and publish a dataset of LoRa frames transmitted in a campus environment. The dataset includes IQ samples of the received frames at multiple locations and allows for the evaluation of channel variations with high time resolution. Using the gathered data, we derive empirical propagation channel models for LoRa that include receiver correlation over distance for three scenarios: unmanned aerial vehicle (UAV) line-of-sight (LoS), UAV non-LoS, and pedestrian non-LoS. Furthermore, the dataset is annotated with synchronization information, enabling the evaluation of receiver algorithms using experimental data.

Generative face video coding (GFVC) is vital for modern applications like video conferencing, yet existing methods primarily focus on video motion while neglecting the significant bitrate contribution of audio. Despite the well-established correlation between audio and lip movements, this cross-modal coherence has not been systematically exploited for compression. To address this, we propose an Audio-Visual Cross-Modal Compression (AVCC) framework that jointly compresses audio and video streams. Our framework extracts motion information from video and tokenizes audio features, then aligns them through a unified audio-video diffusion process. This allows synchronized reconstruction of both modalities from a shared representation. In extremely low-rate scenarios, AVCC can even reconstruct one modality from the other. Experiments show that AVCC significantly outperforms the Versatile Video Coding (VVC) standard and state-of-the-art GFVC schemes in rate-distortion performance, paving the way for more efficient multimodal communication systems.

Alzheimer s disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline, where early detection is essential for timely intervention and improved patient outcomes. Traditional diagnostic methods are time-consuming and require expert interpretation, thus, automated approaches are highly desirable. This study presents a novel deep learning framework for AD diagnosis using Electroencephalograph (EEG) signals, integrating multiple feature extraction techniques including alpha-wave analysis, Discrete Wavelet Transform (DWT), and Markov Transition Fields (MTF). A late-fusion strategy is employed to combine predictions from separate neural networks trained on these diverse representations, capturing both temporal and frequency-domain patterns in the EEG data. The proposed model attains a classification accuracy of 87.23%, with a precision of 87.95%, a recall of 86.91%, and an F1 score of 87.42% when evaluated on a publicly available dataset, demonstrating its potential for reliable, scalable, and early AD screening. Rigorous preprocessing and targeted frequency band selection, particularly in the alpha range due to its cognitive relevance, further enhance performance. This work highlights the promise of deep learning in supporting physicians with efficient and accessible tools for early AD diagnosis.

As demand-side flexibility becomes increasingly necessary to integrate variable renewable energy, understanding electricity demand composition across different grid levels is essential. However, at regional and national scales, visibility into the relative contributions of different consumer categories remains limited due to the complexity and cost of collecting end-use consumption data. To address this challenge, we propose a blind source separation framework to disaggregate open-access high-voltage grid load measurements into sectoral contributions. The approach relies on a constrained variant of non-negative matrix factorization, termed linearly-constrained non-negative matrix factorization (LCNMF), which allows prior information to be incorporated as linear constraints on the factor matrices, thereby providing weak supervision of the separation process. The framework is evaluated using Italian national load data from 2021 to 2023. Results demonstrate the identifiability of residential, services, and industrial load components and provide monthly sectoral consumption estimates consistent with reported statistics. The proposed method is generalizable and applicable to load disaggregation problems across multiple grid scales where disaggregated measurements are unavailable.

We introduce O-EENC-SD: an end-to-end online speaker diarization system based on EEND-EDA, featuring a novel RNN-based stitching mechanism for online prediction. In particular, we develop a novel centroid refinement decoder whose usefulness is assessed through a rigorous ablation study. Our system provides key advantages over existing methods: a hyperparameter-free solution compared to unsupervised clustering approaches, and a more efficient alternative to current online end-to-end methods, which are computationally costly. We demonstrate that O-EENC-SD is competitive with the state of the art in the two-speaker conversational telephone speech domain, as tested on the CallHome dataset. Our results show that O-EENC-SD provides a great trade-off between DER and complexity, even when working on independent chunks with no overlap, making the system extremely efficient.

This work proposes a cooperative strategy for a group of quadrotors interacting over ring digraphs with macro-vertices of size two. Consensus for a group of general double integrators has been initially investigated, and it has been proved that through a suitable choice of a single controller parameter, consensus and stability of the resulting networked dynamical system can be ensured. This further opens up the possibility of achieving a desired formation and to move a swarm of quadrotors, interacting over ring digraphs, at a desired flight velocity, using a single controller gain. An analysis of achievable velocities is performed. Examples have been provided to offer deeper insights into the obtained analytical results. Simulation studies clearly demonstrate that a desired formation is achieved, starting from arbitrary initial positions, while also ensuring convergence to a final desired flight velocity.

Self-supervised speech models such as wav2vec2.0 and WavLM have been shown to significantly improve the performance of many downstream speech tasks, especially in low-resource settings, over the past few years. Despite this, evaluations on tasks such as Speaker Diarization and Speech Separation remain limited. This paper investigates the quality of recent self-supervised speech representations on these two speaker identity-related tasks, highlighting gaps in the current literature that stem from limitations in the existing benchmarks, particularly the lack of diversity in evaluation datasets and variety in downstream systems associated to both diarization and separation.

Electromagnetic navigation systems (eMNS) are increasingly used in minimally invasive procedures such as endovascular interventions and targeted drug delivery due to their ability to generate fast and precise magnetic fields. In this paper, we utilize the OctoMag and a custom 13-coil eMNS to achieve remote levitation and control of multiple rigid bodies across large air gaps, showcasing the dynamic capabilities of such systems. A compact parametric analytical model maps coil currents to the forces and torques acting on the levitating object, eliminating the need for computationally expensive simulations or lookup tables and establishing a levitator- and platform-agnostic control framework. Translational motion is stabilized using linear quadratic regulators. A nonlinear time-invariant controller is used to regulate the reduced attitude accounting for the inherent uncontrollability of rotations about the dipole axis and stabilizing the full five degrees of freedom controllable pose subspace. We analyze key design limitations and evaluate the approach through trajectory tracking experiments across different objects and actuation platforms. Notably, our proposed controller demonstrates superiority over an equivalent baseline PID formulation, reliably tracking large spatial angles up to 65 degrees. This work demonstrates the dynamic capabilities and potential of feedback control in electromagnetic navigation, which is likely to open up new medical applications.

Historical information, such as past function values or gradients, has significant potential to enhance decentralized optimization methods for two key reasons: first, it provides richer information about the objective function, which also explains its established success in centralized optimization; second, unlike the second-order derivative or its alternatives, historical information has already been computed or communicated and requires no additional cost to acquire. Despite this potential, it remains underexploited. In this work, we employ a proximal bundle framework to incorporate the function values and gradients at historical iterates and adapt the framework to the proximal decentralized gradient descent method, resulting in a Decentralized Proximal Bundle Method (DPBM). To broaden its applicability, we further extend DPBM to the asynchronous and stochastic setting. We theoretically analysed the convergence of the proposed methods. Notably, both the asynchronous DPBM and its stochastic variant can converge with fixed step-sizes that are independent of delays, which is superior to the delay-dependent step-sizes required by most existing asynchronous optimization methods, as it is easier to determine and often leads to faster convergence. Numerical experiments on classification problems demonstrate that by using historical information, our methods yield faster convergence and stronger robustness in the step-sizes.

I study intertemporal hedging demand in a continuous-time multi-asset long-run risk (LRR) model under Epstein–Zin (EZ) recursive preferences. The investor trades a risk-free asset and several risky assets whose drifts and volatilities depend on an Ornstein–Uhlenbeck type LRR factor. Preferences are described by EZ utility with risk aversion $R$, elasticity of intertemporal substitution $\psi$, and discount rate $\delta$, so that the standard time-additive CRRA case appears as a limiting benchmark. To handle the high-dimensional consumption–investment problem, I use a projected Pontryagin-guided deep policy optimization (P-PGDPO) scheme adapted to EZ preferences. The method starts from the continuous-time Hamiltonian implied by the Pontryagin maximum principle, represents the value and costate processes with neural networks, and updates the policy along the Hamiltonian gradient. Portfolio constraints and a lower bound on wealth are enforced by explicit projection operators rather than by adding ad hoc penalties. Three main findings emerge from numerical experiments in a five-asset LRR economy: (1) the P-PGDPO algorithm achieves stable convergence across multiple random seeds, validating its reliability for solving high-dimensional EZ problems; (2) wealth floors materially reduce hedging demand by limiting the investor's ability to exploit intertemporal risk-return tradeoffs; and (3) the learned hedging portfolios concentrate exposure in assets with high correlation to the LRR factor, confirming that EZ agents actively hedge long-run uncertainty rather than merely following myopic rules. Because EZ preferences nest time-additive CRRA in the limit $\psi \to 1/R$, I use CRRA as an explicit diagnostic benchmark and, when needed, a warm start to stabilize training in high dimensions.

Due to the significant variations in unmanned aerial vehicle (UAV) altitude and horizontal mobility, it becomes difficult for any single network to ensure continuous and reliable threedimensional coverage. Towards that end, the space-air-ground integrated network (SAGIN) has emerged as an essential architecture for enabling ubiquitous UAV connectivity. To address the pronounced disparities in coverage and signal characteristics across heterogeneous networks, this paper formulates UAV mobility management in SAGIN as a constrained multi-objective joint optimization problem. The formulation couples discrete link selection with continuous trajectory optimization. Building on this, we propose a two-level multi-agent hierarchical deep reinforcement learning (HDRL) framework that decomposes the problem into two alternately solvable subproblems. To map complex link selection decisions into a compact discrete action space, we conceive a double deep Q-network (DDQN) algorithm in the top-level, which achieves stable and high-quality policy learning through double Q-value estimation. To handle the continuous trajectory action space while satisfying quality of service (QoS) constraints, we integrate the maximum-entropy mechanism of the soft actor-critic (SAC) and employ a Lagrangian-based constrained SAC (CSAC) algorithm in the lower-level that dynamically adjusts the Lagrange multipliers to balance constraint satisfaction and policy optimization. Moreover, the proposed algorithm can be extended to multi-UAV scenarios under the centralized training and decentralized execution (CTDE) paradigm, which enables more generalizable policies. Simulation results demonstrate that the proposed scheme substantially outperforms existing benchmarks in throughput, link switching frequency and QoS satisfaction.

The advent of sixth-generation (6G) places intelligence at the core of wireless architecture, fusing perception, communication, and computation into a single closed-loop. This paper argues that large artificial intelligence models (LAMs) can endow base stations with perception, reasoning, and acting capabilities, thus transforming them into intelligent base station agents (IBSAs). We first review the historical evolution of BSs from single-functional analog infrastructure to distributed, software-defined, and finally LAM-empowered IBSA, highlighting the accompanying changes in architecture, hardware platforms, and deployment. We then present an IBSA architecture that couples a perception-cognition-execution pipeline with cloud-edge-end collaboration and parameter-efficient adaptation. Subsequently,we study two representative scenarios: (i) cooperative vehicle-road perception for autonomous driving, and (ii) ubiquitous base station support for low-altitude uncrewed aerial vehicle safety monitoring and response against unauthorized drones. On this basis, we analyze key enabling technologies spanning LAM design and training, efficient edge-cloud inference, multi-modal perception and actuation, as well as trustworthy security and governance. We further propose a holistic evaluation framework and benchmark considerations that jointly cover communication performance, perception accuracy, decision-making reliability, safety, and energy efficiency. Finally, we distill open challenges on benchmarks, continual adaptation, trustworthy decision-making, and standardization. Together, this work positions LAM-enabled IBSAs as a practical path toward integrated perception, communication, and computation native, safety-critical 6G systems.

Target classification is a fundamental task in radar systems, and its performance critically depends on the quantization precision of the signal. While high-precision quantization (e.g. 16-bit) is well established, 1-bit quantization offers distinct advantages by enabling direct sampling at high frequencies and eliminating complex intermediate stages. However, its extreme quantization leads to significant information loss. Although higher sampling rates can compensate for this loss, such oversampling is impractical at the high frequencies targeted for direct sampling. To achieve high-accuracy classification directly from 1-bit radar data under the same sampling rate, this paper proposes a novel two-stage deep learning framework, CF-Net. First, we introduce a self-supervised pre-training strategy based on a dual-branch U-Net architecture. This network learns to restore high-fidelity 16-bit images from their 1-bit counterparts via a cross-feature reconstruction task, forcing the 1-bit encoder to learn robust features despite extreme quantization. Subsequently, this pre-trained encoder is repurposed and fine-tuned for the downstream multi-class target classification task. Experiments on two radar target datasets demonstrate that CF-Net can effectively extract discriminative features from 1-bit imagery, achieving comparable and even superior accuracy to some 16-bit methods without oversampling.

The rapid growth in wireless infrastructure has increased the need to accurately estimate and forecast electromagnetic field (EMF) levels to ensure ongoing compliance, assess potential health impacts, and support efficient network planning. While existing studies rely on univariate forecasting of wideband aggregate EMF data, multivariate narrow-band EMF forecasting is needed to capture the inter-operator and inter-frequency variations essential for proactive network planning. To this end, this paper introduces EMFusion, a conditional diffusion-based EMF forecasting framework that integrates diverse contextual factors, such as time of day, season, and holidays, while providing uncertainty-aware probabilistic forecasts. The proposed architecture features a residual U-Net backbone enhanced by a cross-attention mechanism that dynamically integrates external conditions to guide the generation process. Furthermore, EMFusion integrates an imputation-based sampling strategy that treats forecasting as a structural inpainting task, ensuring temporal coherence even with irregular measurements. Unlike standard point forecasters, EMFusion generates empirical probabilistic prediction intervals from the learned conditional distribution, providing uncertainty-aware probabilistic forecasting rather than simple point estimation. Numerical experiments conducted on the multivariate narrow-band EMF datasets demonstrate that EMFusion with the contextual information of working hours outperforms the baseline models with or without conditions. The proposed EMFusion outperforms the best baseline by 23.85% in continuous ranked probability score (CRPS) and 13.93% in normalized root mean square error.

This letter presents a novel deep reinforcement learning (DRL) approach for joint time allocation and power control in a cognitive Internet of Things (CIoT) system with simultaneous wireless information and power transfer (SWIPT). The CIoT transmitter autonomously manages energy harvesting (EH) and transmissions using a learnable time switching factor while optimizing power to enhance throughput and lifetime. The joint optimization is modeled as a Markov decision process under small-scale fading, realistic EH, and interference constraints. We develop a double deep Q-network (DDQN) enhanced with an upper confidence bound. Simulations benchmark our approach, showing superior performance over existing DRL methods.

This study develops meta-learners for few-shot weakly-supervised segmentation (FWS) to address the challenge of optic disc (OD) and optic cup (OC) segmentation for glaucoma diagnosis with limited labeled fundus images. We significantly improve existing meta-learners by introducing Omni meta-training which balances data usage and diversifies the number of shots. We also develop their efficient versions that reduce computational costs. In addition, we develop sparsification techniques that generate more customizable and representative scribbles and other sparse labels. After evaluating multiple datasets, we find that Omni and efficient versions outperform the original versions, with the best meta-learner being Efficient Omni ProtoSeg (EO-ProtoSeg). It achieves intersection over union (IoU) scores of 88.15% for OD and 71.17% for OC on the REFUGE dataset using just one sparsely labeled image, outperforming few-shot and semi-supervised methods which require more labeled images. Its best performance reaches 86.80% for OD and 71.78%for OC on DRISHTIGS, 88.21% for OD and 73.70% for OC on REFUGE, 80.39% for OD and 52.65% for OC on REFUGE. EO-ProtoSeg is comparable to unsupervised domain adaptation methods yet much lighter with less than two million parameters and does not require any retraining.

In this work, a Janus metasurface based tensor impedance holographic antenna (JHA) is proposed that simultaneously radiates linearly polarized (LP) and circularly polarized (CP) beams from a single aperture excited by a single feed. The proposed design introduces modified tensor impedance equations to significantly reduce cross-polarization at higher radiation angles. It demonstrates broadband operation bandwidth of 0.5 GHz while maintaining high circular polarization purity. The design methodology is verified using aperture field integration theory, ensuring that the impedance distribution produces the desired far-field radiation patterns. Prototypes of three variations of the holographic antenna are fabricated, validating its performance. The radiation characteristics of the proposed antenna make it an attractive choice for advanced broadband communication applications.

Atomic electron tomography (AET) enables the determination of 3D atomic structures by acquiring a sequence of 2D tomographic projection measurements of a particle and then computationally solving for its underlying 3D representation. Classical tomography algorithms solve for an intermediate volumetric representation that is post-processed into the atomic structure of interest. In this paper, we reformulate the tomographic inverse problem to solve directly for the locations and properties of individual atoms. We parameterize an atomic structure as a collection of Gaussians, whose positions and properties are learnable. This representation imparts a strong physical prior on the learned structure, which we show yields improved robustness to real-world imaging artifacts. Simulated experiments and a proof-of-concept result on experimentally-acquired data confirm our method's potential for practical applications in materials characterization and analysis with Transmission Electron Microscopy (TEM). Our code is available at https://github.com/nalinimsingh/gaussian-atoms.

Imaging systems are commonly described using resolution, contrast, and signal-to-noise ratio, but these quantities do not provide a general account of how physical transformations affect the flow of information. This paper introduces an operator-based formulation of information theory for imaging. The approach models the imaging chain as a composition of bounded operators acting on functions, and characterises information redistribution using the spectral properties of these operators. Three measures are developed. Operator entropy quantifies how an operator distributes energy across its singular spectrum. Operator information capacity describes the number of modes that remain recoverable above a noise-dependent threshold. An irreversibility index measures the information lost through suppression or elimination of modes and captures the accumulation of information loss under operator composition. The framework applies to linear, nonlinear, and stochastic operators and does not depend on the specific imaging modality. Analytical examples show how attenuation, blur, and sampling affect entropy, capacity, and irreversibility in different ways. The results provide a general structure for analysing the physical limits of imaging and form the basis for subsequent work on information geometry, spatiotemporal budgets, nonlinear channels, and reconstruction algorithms.