Papers

Distributed multichannel acoustic sensing (DMAS) enables large-scale sound event classification (SEC), but performance drops when many channels are degraded and when sensor layouts at test time differ from training layouts. We propose a learning-free, physics-informed inpainting frontend based on reverse time migration (RTM). In this approach, observed multichannel spectrograms are first back-propagated on a 3D grid using an analytic Green's function to form a scene-consistent image, and then forward-projected to reconstruct inpainted signals before log-mel feature extraction and Transformer-based classification. We evaluate the method on ESC-50 with 50 sensors and three layouts (circular, linear, right-angle), where per-channel SNRs are sampled from -30 to 0 dB. Compared with an AST baseline, scaling-sparsemax channel selection, and channel-swap augmentation, the proposed RTM frontend achieves the best or competitive accuracy across all layouts, improving accuracy by 13.1 points on the right-angle layout (from 9.7% to 22.8%). Correlation analyses show that spatial weights align more strongly with SNR than with channel–source distance, and that higher SNR–weight correlation corresponds to higher SEC accuracy. These results demonstrate that a reconstruct-then-project, physics-based preprocessing effectively complements learning-only methods for DMAS under layout-open configurations and severe channel degradation.

Integrated sensing and communication (ISAC) is a key enabler for future radio networks. This paper presents a sub-band full-duplex (SBFD) ISAC system that assigns non-overlapping OFDM subbands to sensing and communication, enabling simultaneous operation with minimal interference. A distributed testbed with three SIMO nodes is implemented using USRP X410 devices operating at 6.8 GHz with 20 MHz bandwidth per channel. A total of 2048 OFDM subcarriers are partitioned into three subbands: two for sensing using Zadoff-Chu sequences and one for communication using QPSK. Each USRP transmits one subband while receiving signals across all three, forming a 1 x 3 SIMO node. Time synchronization is achieved through host-server coordination without external clock distribution. Indoor measurements, validated against MOCAP ground truth, confirm the feasibility of the SBFD ISAC system. The results demonstrate monostatic sensing with a velocity resolution of 0.145 m/s, and communication under NLoS conditions with a BER of 3.63e-3. Compared with a multiband benchmark requiring three times more spectrum, the SBFD configuration achieves comparable velocity estimation accuracy while conserving resources. The sensing and communication performance trade-off is determined by subcarrier allocation strategy rather than mutual interference.

The transition toward cognitive global navigation satellite system (GNSS) receivers requires accurate interference classification to trigger adaptive mitigation strategies. However, conventional methods relying on Time-Frequency Analysis (TFA) and Convolutional Neural Networks (CNNs) face two fundamental limitations: severe performance degradation in low Jamming-to-Signal Ratio (JSR) regimes due to noise obscuration, and “feature degeneracy” caused by the loss of phase information in magnitude-only spectrograms. Consequently, spectrally similar signals – such as high-order Quadrature Amplitude Modulation versus Band-Limited Gaussian Noise – become indistinguishable. To overcome these challenges, this paper proposes the JSR-Guided Fusion Network (JSR-GFNet). This multi-modal architecture combines phase-sensitive complex In-Phase/Quadrature (IQ) samples with Short-Time Fourier Transform (STFT) spectrograms. Central to this framework is a physics-inspired dynamic gating mechanism driven by statistical signal descriptors. Acting as a conditional controller, it autonomously estimates signal reliability to dynamically reweight the contributions of a Complex-Valued ResNet (IQ stream) and an EfficientNet backbone (STFT stream). To validate the model, we introduce the Comprehensive GNSS Interference (CGI-21) dataset, simulating 21 jamming categories including software-defined waveforms from aerial platforms. Extensive experiments demonstrate that JSR-GFNet achieves higher accuracy across the full 10–50 dB JSR spectrum. Notably, interpretability analysis confirms that the model learns a physically intuitive strategy: prioritizing spectral energy integration in noise-limited regimes while shifting focus to phase precision in high-SNR scenarios to resolve modulation ambiguities. This framework provides a robust solution for next-generation aerospace navigation security.

Extra-large scale MIMO (XL-MIMO) is a key technology for meeting sixth-generation (6G) requirements for high-rate connectivity and uniform quality of service (QoS); however, its deployment is challenged by the prohibitive complexity of resource management based on instantaneous channel state information (CSI). To address this intractability, this work derives novel closed-form deterministic signal-to-interference-plus-noise ratio (SINR) expressions for both centralized and distributed uplink operations. Valid for Rician fading channels with minimum mean square error (MMSE) receive combining and MMSE channel estimation, these expressions depend exclusively on long-term channel statistics, providing a tractable alternative to computationally expensive instantaneous CSI-driven optimization. Building on these results, we develop statistical-CSI-based algorithms for joint subarray selection, users scheduling, and pilot assignment, leveraging the derived SINR approximations to maximize the minimum spectral efficiency (SE) among scheduled users while preserving computational tractability. The proposed framework exploits the spatial sparsity of user equipment (UE) visibility regions (VRs) to enable more aggressive pilot reuse than is possible in conventional massive MIMO. Numerical results validate the high accuracy of the derived SINR approximations and demonstrate that the proposed algorithms significantly enhance fairness and throughput in crowded scenarios.

Linearly parametrized models are widely used in control and signal processing, with the least-squares (LS) estimate being the archetypical solution. When the input is insufficiently exciting, the LS problem may be unsolvable or numerically unstable. This issue can be resolved through regularization, typically with ridge regression. Although regularized estimators reduce the variance error, it remains important to quantify their estimation uncertainty. A possible approach for linear regression is to construct confidence ellipsoids with the Sign-Perturbed Sums (SPS) ellipsoidal outer approximation (EOA) algorithm. The SPS EOA builds non-asymptotic confidence ellipsoids under the assumption that the noises are independent and symmetric about zero. This paper introduces an extension of the SPS EOA algorithm to ridge regression, and derives probably approximately correct (PAC) upper bounds for the resulting region sizes. Compared with previous analyses, our result explicitly show how the regularization parameter affects the region sizes, and provide tighter bounds under weaker excitation assumptions. Finally, the practical effect of regularization is also demonstrated via simulation experiments.

Recent years have seen an increased interest in the use of Non-terrestrial networks (NTNs), especially the unmanned aerial vehicles (UAVs) to provide cost-effective global connectivity in next-generation wireless networks. We introduce a resilient, adaptive, self-healing network design (RASHND) to optimize signal quality under dynamic interference and adversarial conditions. RASHND leverages inter-node communication and an intelligent algorithm selection process, incorporating combining techniques like distributed-Maximal Ratio Combining (d-MRC), distributed-Linear Minimum Mean Squared Error Estimation(d-LMMSE), and Selection Combining (SC). These algorithms are selected to improve performance by adapting to changing network conditions. To evaluate the effectiveness of the proposed RASHND solutions, a software-defined radio (SDR)-based hardware testbed afforded initial testing and evaluations. Additionally, we present results from UAV tests conducted on the AERPAW testbed to validate our solutions in real-world scenarios. The results demonstrate that RASHND significantly enhances the reliability and interference resilience of UAV networks, making them well-suited for critical communications.

Speech large language models (LLMs) have driven significant progress in end-to-end speech understanding and recognition, yet they continue to struggle with accurately recognizing rare words and domain-specific terminology. This paper presents a novel fine-tuning method, Reinforcement Learning with Biasing Rewards (RLBR), which employs a specialized biasing words preferred reward to explicitly emphasize biasing words in the reward calculation. In addition, we introduce reference-aware mechanisms that extend the reinforcement learning algorithm with reference transcription to strengthen the potential trajectory exploration space. Experiments on the LibriSpeech corpus across various biasing list sizes demonstrate that RLBR delivers substantial performance improvements over a strong supervised fine-tuning (SFT) baseline and consistently outperforms several recently published methods. The proposed approach achieves excellent performance on the LibriSpeech test-clean and test-other sets, reaching Biasing Word Error Rates (BWERs) of 0.59% / 2.11%, 1.09% / 3.24%, and 1.36% / 4.04% for biasing list sizes of 100, 500, and 1000, respectively, without compromising the overall WERs.

Four-dimensional (4D) Flow MRI can noninvasively measure cerebrovascular hemodynamics but remains underused clinically because current workflows rely on manual vessel segmentation and yield velocity fields sensitive to noise, artifacts, and phase aliasing. We present VAST (Vascular Flow Analysis and Segmentation), an automated, unsupervised pipeline for intracranial 4D Flow MRI that couples vessel segmentation with physics-informed velocity reconstruction. VAST derives vessel masks directly from complex 4D Flow data by iteratively fusing magnitude- and phase-based background statistics. It then reconstructs velocities via continuity-constrained phase unwrapping, outlier correction, and low-rank denoising to reduce noise and aliasing while promoting mass-consistent flow fields, with processing completing in minutes per case on a standard CPU. We validate VAST on synthetic data from an internal carotid artery aneurysm model across SNR = 2-20 and severe phase wrapping (up to five-fold), on in vitro Poiseuille flow, and on an in vivo internal carotid aneurysm dataset. In synthetic benchmarks, VAST maintains near quarter-voxel surface accuracy and reduces velocity root-mean-square error by up to fourfold under the most degraded conditions. In vitro, it segments the channel within approximately half a voxel of expert annotations and reduces velocity error by 39% (unwrapped) and 77% (aliased). In vivo, VAST closely matches expert time-of-flight masks and lowers divergence residuals by about 30%, indicating a more self-consistent intracranial flow field. By automating processing and enforcing basic flow physics, VAST helps move intracranial 4D Flow MRI toward routine quantitative use in cerebrovascular assessment.

In this paper, we present a Transformer-based architecture for 3D radar object detection that uses a novel Transformer Decoder as the prediction head to directly regress 3D bounding boxes and class scores from radar feature representations. To bridge multi-scale radar features and the decoder, we propose Pyramid Token Fusion (PTF), a lightweight module that converts a feature pyramid into a unified, scale-aware token sequence. By formulating detection as a set prediction problem with learnable object queries and positional encodings, our design models long-range spatial-temporal correlations and cross-feature interactions. This approach eliminates dense proposal generation and heuristic post-processing such as extensive non-maximum suppression (NMS) tuning. We evaluate the proposed framework on the RADDet, where it achieves significant improvements over state-of-the-art radar-only baselines.

State Space Models (SSMs) and Hidden Markov Models (HMMs) are foundational frameworks for modeling sequential data with latent variables and are widely used in signal processing, control theory, and machine learning. Despite their shared temporal structure, they differ fundamentally in the nature of their latent states, probabilistic assumptions, inference procedures, and training paradigms. Recently, deterministic state space models have re-emerged in natural language processing through architectures such as S4 and Mamba, raising new questions about the relationship between classical probabilistic SSMs, HMMs, and modern neural sequence models. In this paper, we present a unified and systematic comparison of HMMs, linear Gaussian state space models, Kalman filtering, and contemporary NLP state space models. We analyze their formulations through the lens of probabilistic graphical models, examine their inference algorithms – including forward-backward inference and Kalman filtering – and contrast their learning procedures via Expectation-Maximization and gradient-based optimization. By highlighting both structural similarities and semantic differences, we clarify when these models are equivalent, when they fundamentally diverge, and how modern NLP SSMs relate to classical probabilistic models. Our analysis bridges perspectives from control theory, probabilistic modeling, and modern deep learning.

In recent years, there has been a growing interest in low-light image enhancement (LLIE) due to its importance for critical downstream tasks. Current Retinex-based methods and learning-based approaches have shown significant LLIE performance. However, computational complexity and dependencies on large training datasets often limit their applicability in real-time applications. We introduce RetinexGuI, a novel and effective Retinex-guided LLIE framework to overcome these limitations. The proposed method first separates the input image into illumination and reflection layers, and iteratively refines the illumination while keeping the reflectance component unchanged. With its simplified formulation and computational complexity of $\mathcal{O}(N)$, our RetinexGuI demonstrates impressive enhancement performance across three public datasets, indicating strong potential for large-scale applications. Furthermore, it opens promising directions for theoretical analysis and integration with deep learning approaches. The source code will be made publicly available at https://github.com/etuspars/RetinexGuI once the paper is accepted.

The Internet of Underwater Things (IoUT) supports marine sensing, environmental monitoring, subsea inspection, and autonomous underwater operations. However, IoUT communication is constrained by limited bandwidth, long propagation delay, time-varying underwater channels, intermittent connectivity, and strict energy budgets. Semantic Communication (SC) offers a promising alternative by transmitting task-relevant meaning rather than raw data, thereby improving communication efficiency in resource-constrained underwater networks. This paper presents a critical and feasibility-aware survey of SC for IoUT, focusing on opportunities, challenges, limitations, and future research directions. We first review the fundamentals of SC-enabled IoUT systems, including semantic representations, layered architectures, semantic channel modeling, and task-oriented evaluation metrics. We then examine learning-driven approaches based on machine learning (ML), knowledge graphs (KGs), vision-language models (VLMs), generative models, and federated learning (FL), with emphasis on their feasibility under underwater edge constraints. Representative applications, including environmental monitoring, marine ecology, subsea infrastructure inspection, disaster response, and autonomous underwater vehicle (AUV) coordination, are analyzed from an SC perspective. Finally, we identify key research directions involving standardized semantic models, reproducible testbeds, compute–communication trade-offs, trustworthy reconstruction, hybrid underwater links, energy-aware edge intelligence, semantic security, digital twins (DTs), and cross-domain interoperability. This survey provides a structured foundation for developing reliable, efficient, and meaning-driven IoUT communication systems.

We provide a motion-planning framework for target reaching and obstacle avoidance in cluttered environments. Given a polyhedral decomposition of the feasible space, the agent is guided through a prescribed sequence of cells by a suitable control policy. For each cell, we construct a potential induced by a harmonic function on the unit disk whose Dirichlet boundary condition is represented by a cardinal B-spline curve. A detailed analysis of the curve periodicity, support, and control-point selection provides an explicit analytical construction of these potentials and the associated vector fields. We further characterize inter-cell control continuity and forward invariance of non-exit facets, and introduce a smoothed control law ensuring continuous switching between independently constructed neighboring cells. Numerical results illustrate that the resulting framework safely guides the agent along the prescribed cell sequence from the initial point to the target.

Autonomous navigation for nano-scale unmanned aerial vehicles (nano-UAVs) is governed by extreme Size, Weight, and Power (SWaP) constraints (with the weight < 50 g and sub-100 mW onboard processor), distinguishing it fundamentally from standard robotic paradigms. This review synthesizes the state-of-the-art in sensing, computing, and control architectures designed specifically for these sub- 100mW computational envelopes. We critically analyse the transition from classical geometry-based methods to emerging "Edge AI" paradigms, including quantized deep neural networks deployed on ultra-low-power System-on-Chips (SoCs) and neuromorphic event-based control. Beyond algorithms, we evaluate the hardware-software co-design requisite for autonomy, covering advancements in dense optical flow, optimized Simultaneous Localization and Mapping (SLAM), and learning-based flight control. While significant progress has been observed in visual navigation and relative pose estimation, our analysis reveals persistent gaps in long-term endurance, robust obstacle avoidance in dynamic environments, and the "Sim-to-Real" transfer of reinforcement learning policies. This survey provides a roadmap for bridging these gaps, advocating for hybrid architectures that fuse lightweight classical control with data-driven perception to enable fully autonomous, agile nano-UAVs in GPS-denied environments.

Parallel imaging techniques reduce magnetic resonance imaging (MRI) scan time but image quality degrades as the acceleration factor increases. In clinical practice, conservative acceleration factors are chosen because no mechanism exists to automatically assess the diagnostic quality of undersampled reconstructions. This work introduces a general framework for pixel-wise uncertainty quantification in parallel MRI reconstructions, enabling automatic identification of unreliable regions without access to any ground-truth reference image. Our method integrates conformal quantile regression with image reconstruction methods to estimate statistically rigorous pixel-wise uncertainty intervals. We trained and evaluated our model on Cartesian undersampled brain and knee data obtained from the fastMRI dataset using acceleration factors ranging from 2 to 10. An end-to-end Variational Network was used for image reconstruction. Quantitative experiments demonstrate strong agreement between predicted uncertainty maps and true reconstruction error. Using our method, the corresponding Pearson correlation coefficient was higher than 90% at acceleration levels at and above four-fold; whereas it dropped to less than 70% when the uncertainty was computed using a simpler a heuristic notion (magnitude of the residual). Qualitative examples further show the uncertainty maps based on quantile regression capture the magnitude and spatial distribution of reconstruction errors across acceleration factors, with regions of elevated uncertainty aligning with pathologies and artifacts. The proposed framework enables evaluation of reconstruction quality without access to fully-sampled ground-truth reference images. It represents a step toward adaptive MRI acquisition protocols that may be able to dynamically balance scan time and diagnostic reliability.

This paper focuses on the asymptotic analysis of a class of nonlinear one-bit precoding schemes under Rayleigh fading channels. The considered scheme employs a convex-relaxation-then-quantization (CRQ) approach to the well-known minimum mean square error (MMSE) model, which includes the classical one-bit precoder SQUID as a special case. To analyze its asymptotic behavior, we develop a novel analytical framework based on approximate message passing (AMP). We show that, the statistical properties of the considered scheme can be asymptotically characterized by a scalar “signal plus Gaussian noise” model. Based on this, we further derive a closed-form expression for the symbol error probability (SEP) in the large-system limit, which quantitatively characterizes the impact of both system and model parameters on SEP performance. Simulation results validate our analysis and also demonstrate that performance gains over SQUID can be achieved by appropriately tuning the parameters involved in the considered model.

We propose a financial liquidity policy sharing method for firm-to-firm supply networks, introducing a scalable autonomous control function for viable complex adaptive supply networks. Cooperation and competition in supply chains is reconciled through overlapping collaborative sets, making firms interdependent and enabling distributed risk governance. How cooperative range - visibility - affects viability is studied using dynamic complex adaptive systems modelling. We find that viability needs cooperation; visibility and viability grow together in scale-free supply networks; and distributed control, where firms only have limited partner information, outperforms centralised control. This suggests that policy toward network viability should implement distributed supply chain financial governance, supporting interfirm collaboration, to enable autonomous control.

Unmanned aerial vehicles (UAVs) enable efficient in-situ radiation characterization of large-aperture antennas directly in their deployment environments. In such measurements, a continuous-wave (CW) probe tone is commonly transmitted to characterize the antenna response. However, active co-channel emissions from neighboring antennas often introduce severe in-band interference, where classical FFT-based estimators fail to accurately estimate the CW tone amplitude when the signal-to-interference ratios (SIR) falls below -10 dB. This paper proposes a lightweight deep convolutional neural network (DC-CNN) that estimates the amplitude of the CW tone. The model is trained and evaluated on real 5~GHz measurement bursts spanning an effective SIR range of –33.3 dB to +46.7 dB. Despite its compact size (<20k parameters), the proposed DC-CNN achieves a mean absolute error (MAE) of 7% over the full range, with <1 dB error for SIR >= -30 dB. This robustness and efficiency make DC-CNN suitable for deployment on embedded UAV platforms for interference-resilient antenna pattern characterization.

Sparse vector transmission (SVT) is a promising candidate technology for achieving ultra-reliable low-latency communication (URLLC). In this paper, a hierarchical SVT scheme is proposed for multi-user URLLC scenarios. The hierarchical SVT scheme partitions the transmitted bits into common and private parts. The common information is conveyed by the indices of non-zero sections in a sparse vector, while each user's private information is embedded into non-zero blocks with specific block lengths. At the receiver, the common bits are first recovered from the detected non-zero sections, followed by user-specific private bits decoding based on the corresponding non-zero block indices. Simulation results show the proposed scheme outperforms state-of-the-art SVT schemes in terms of block error rate.

Regulators and voluntary corporate sustainability efforts are increasingly adopting time-matching requirements (TMRs) for clean electricity procurement for large loads, such as data centers, and electricity-intensive fuel production, such as hydrogen. We use a stochastic capacity expansion model (CEM) framework to assess how inter-annual weather variability affects the cost, composition, and emissions of procurement-driven infrastructure to meet annual and hourly TMRs using the case study of a grid-connected hydrogen producer in Texas. Our approach, which relies on co-optimizing investments and hourly operations over nine weather scenarios, reveals that hourly TMR comes at a higher cost premium compared to annual TMR than previously estimated by single-scenario deterministic modeling, while emissions outcomes remain directionally consistent. Demand flexibility and partial hourly TMR (80-90%) lower the cost premium while preserving emissions benefits. We further examine how binding renewable portfolio standards (RPS) interact with TMR costs and emissions outcomes. When an RPS is applied to non-H2 electricity demand, annual TMR reduces emissions comparably to hourly TMR at a lower cost. Incorporating H2-related electricity demand directly into the RPS constraint, rather than imposing a separate TMR, achieves similar emissions outcomes at still lower cost, suggesting that TMR-based clean electricity procurement, particularly hourly matching, offers limited additional value in regions with stringent grid decarbonization policies.

In this paper, a wideband cell-free multi-stream multi-user Multiple-Input Multiple-Output (MIMO) Orthogonal Frequency Division Multiplexing (OFDM) system is considered operating within a smart wireless environment enabled by multiple Beyond Diagonal Reconfigurable Intelligent Surfaces (BDRISs). A novel decentralized active and passive beamforming framework, robust to imperfect channel state availability and with minimal cooperation among the system's multiple Base Stations (BSs) for deciding the final configurations of the shared BDRISs, is proposed, which aims to substantially reduce the overhead inherent in centralized solutions necessitating a central processing unit of high computational power. By considering a Dynamic Group-Connected (DGC) BDRIS architecture with frequency-selective responses per unit element, we formulate the system's sum-rate maximization problem with respect to the tunable capacitances and permutation matrices of the BDRISs as well as the precoding matrices of the BSs, which is solved via successive concave approximation and alternating projections as well as consensus-based updates for the BDRISs' design. Through extensive simulation results, it is showcased that the proposed robust decentralized cooperative approach with diverse BDRIS architectures outperforms non-cooperation benchmarks. It is also demonstrated that the considered DGC BDRIS architecture is able to provide sum-rate performance gains sufficiently close to the more complex fully-connected BDRIS structure.

The growing demand for mobile data services in dense urban areas has intensified the need for energy-efficient radio access networks (RANs) in future 6G systems. In this context, one promising strategy is cell switching (CS), which dynamically deactivates underutilized small base stations (SBSs) to reduce power consumption. However, while previous research explored CS primarily based on traffic load, ensuring user quality of service (QoS) under realistic channel conditions remains a challenge. In this paper, we propose a novel optimization-driven CS framework that jointly minimizes network power consumption and guarantees user QoS by enforcing a minimum received power threshold as part of offloading decisions. In contrast to prior load-based or learning-based approaches, our method explicitly integrates channel-aware information into the CS process, thus ensuring reliable service quality for offloaded users. Furthermore, flexibility of the proposed framework enables operators to adapt system behavior between energy-saving and QoS-preserving modes by tuning a single design parameter. Simulation results demonstrate that the proposed approach achieves up to 30% power savings as compared to baseline methods while fully maintaining QoS under diverse network conditions. Scalability and robustness of the proposed method in realistic heterogeneous networks (HetNets) further highlight its potential as a practical solution for sustainable 6G deployments.

Current RF machine-learning pipelines rely on task-specific deep networks for modulation classification and related tasks, but these models require custom architectures and labeled datasets for each problem, generalize poorly across channel conditions and SNRs, and offer little interpretability. In contrast, modern multimodal large language models (MLLMs) can integrate heterogeneous visual and textual data and exhibit strong cross-domain generalization and explanation capabilities. Our goal in this work is to explore whether vision-language models (VLMs) can be adapted to directly perceive RF signals and reason about modulation patterns without redesigning their architectures or injecting RF-specific inductive biases. To achieve this, we convert complex IQ streams into time-series, spectrogram, and joint RF visualizations, build a 57-class RF visual question answering benchmark, and show that lightweight parameter-efficient fine-tuning can enhance the accuracy of a general-purpose VLM from around 10% to nearly 90%, while ensuring robustness to noise and out-of-vocabulary modulations and the ability to produce human-readable rationales. The obtained results show that combining RF-to-image conversion with promptable VLMs provides a scalable and practical foundation for RF-aware AI systems in future 6G networks.

Diffusion models have recently achieved impressive results in reconstructing images from noisy inputs, and similar ideas have been applied to speech enhancement by treating time-frequency representations as images. With the ubiquity of multi-microphone devices, we extend state-of-the-art diffusion-based methods to exploit multichannel inputs for improved performance. Multichannel diffusion-based enhancement remains in its infancy, with prior work making limited use of advanced mechanisms such as attention for spatial modeling - a gap addressed in this paper. We propose AMDM-SE, an Attention-based Multichannel Diffusion Model for Speech Enhancement, designed specifically for noise reduction. AMDM-SE leverages spatial inter-channel information through a novel cross-channel time-frequency attention block, enabling faithful reconstruction of fine-grained signal details within a generative diffusion framework. On the CHiME-3 benchmark, AMDM-SE outperforms both a single-channel diffusion baseline and a multichannel model without attention, as well as a strong DNN-based predictive method. Simulated-data experiments further underscore the importance of the proposed multichannel attention mechanism. Overall, our results show that incorporating targeted multichannel attention into diffusion models substantially improves noise reduction. While multichannel diffusion-based speech enhancement is still an emerging field, our work contributes a new and complementary approach to the growing body of research in this direction.

Intent-based networks (IBNs) are gaining prominence as an innovative technology that automates network operations through high-level request statements, defining what the network should achieve. In this work, we introduce IntAgent, an intelligent intent LLM agent that integrates NWDAF analytics and tools to fulfill the network operator's intents. Unlike previous approaches, we develop an intent tools engine directly within the NWDAF analytics engine, allowing our agent to utilize live network analytics to inform its reasoning and tool selection. We offer an enriched, 3GPP-compliant data source that enhances the dynamic, context-aware fulfillment of network operator goals, along with an MCP tools server for scheduling, monitoring, and analytics tools. We demonstrate the efficacy of our framework through two practical use cases: ML-based traffic prediction and scheduled policy enforcement, which validate IntAgent's ability to autonomously fulfill complex network intents.

Speaker anonymization aims to conceal a speaker's identity, without considering the linguistic content. In this study, we reveal a weakness of Librispeech, the dataset that is commonly used to evaluate anonymizers: the books read by the Librispeech speakers are so distinct, that speakers can be identified by their vocabularies. Even perfect anonymizers cannot prevent this identity leakage. The EdAcc dataset is better in this regard: only a few speakers can be identified through their vocabularies, encouraging the attacker to look elsewhere for the identities of the anonymized speakers. EdAcc also comprises spontaneous speech and more diverse speakers, complementing Librispeech and giving more insights into how anonymizers work.

Micro-Unmanned Aerial Vehicles (UAVs) are rapidly expanding into tasks from inventory to environmental sensing, yet their short endurance and unreliable navigation in GPS-denied spaces limit deployment. Lighter-Than-Air (LTA) drones offer an energy-efficient alternative: they use a helium envelope to provide buoyancy, which enables near-zero-power drain during hovering and much longer operation. LTAs are promising, but their design is complex, and they lack integrated solutions to enable sustained autonomous operations and navigation with simple, low-infrastructure. We propose a compact, self-sustaining LTA drone that uses light for both energy harvesting and navigation. Our contributions are threefold: (i) a high-fidelity simulation framework to analyze LTA aerodynamics and select a stable, efficient configuration; (ii) a framework to integrate solar cells on the envelope to provide net-positive energy; and (iii) a point-and-go navigation system with three light-seeking algorithms operating on a single light beacon. Our LTA-analysis, together with the integrated solar panels, not only saves energy while flying, but also enables sustainable operation: providing 1 minute of flying time for every 4 minutes of energy harvesting, under illuminations of 80klux. We also demonstrate robust single-beacon navigation towards a light source that can be up to 7m away, in indoor and outdoor environments, even with moderate winds. The resulting system indicates a plausible path toward persistent, autonomous operation for indoor and outdoor monitoring. More broadly, this work provides a practical pathway for translating the promise of LTA drones into a persistent, self-sustaining aerial system.

Clubroot, a major soilborne disease affecting canola and other cruciferous crops, is characterized by the development of large galls on the roots of susceptible hosts. In this study, we present the first application of terahertz time-domain spectroscopy (THz-TDS) as a non-invasive diagnosis tool in plant pathology. Compared with conventional molecular, spectroscopic, and immunoassay-based methods, THz-TDS offers distinct advantages, including non-contact, non-destructive, and preparation-free measurement, enabling rapid in situ screening of plant and soil samples. Our results demonstrate that THz-TDS can differentiate between healthy and clubroot-infected tissues by detecting both structural and biochemical alterations. Specifically, infected roots exhibit a blue shift in the refractive index in the low-frequency THz range, along with distinct peaks-indicative of disruptions in water transport and altered metabolic activity in both roots and leaves. Interestingly, the characteristic root swelling observed in infected plants reflects internal tissue disorganization rather than an actual increase in water content. Furthermore, a physics-constrained neural network is proposed to extract the main feature in THz-TDS. A comprehensive evaluation, including time-domain signals, amplitude and phase images, refractive index and absorption coefficient maps, and principal component analysis, provides enhanced contrast and spatial resolution compared to raw time-domain or frequency signals. These findings suggest that THz-TDS holds significant potential for early, non-destructive detection of plant diseases and may serve as a valuable tool to limit their spread in agricultural systems.

Recent studies on transportation networks have shown that real-time route guidance can inadvertently induce congestion or oscillatory traffic patterns. Nevertheless, such technologies also offer a promising opportunity to manage traffic non-intrusively by shaping the information delivered to users, thereby mitigating congestion and enhancing network stability. A key step toward this goal is to identify information signals that ensure the existence of an equilibrium with desirable stability and convergence properties. This challenge is particularly relevant when traffic density and routing dynamics evolve concurrently, as increasingly occurs with digital signaling and real-time navigation technologies. To address this, we analyze a parallel-path transportation network with a single origin-destination pair, incorporating joint traffic density and logit-based routing dynamics that evolve at the same timescale. We characterize a class of density-dependent traffic information that guarantees a unique equilibrium in the free-flow regime, ensures its asymptotic stability, and keeps traffic densities within the free-flow region for all time. The theoretical results are complemented by a numerical case study demonstrating how the framework can inform the design of traffic information that reduces total travel time without compromising credibility.

We present an unsourced multiple access (UMAC) scheme tailored to high-mobility wireless channels. The proposed construction is based on orthogonal time frequency space (OTFS) modulation and sparse interleaver division multiple access (IDMA) in the delay-Doppler (DD) domain. The receiver runs a compressive-sensing joint activity-detection and channel estimation process followed by a single-user decoder which harnesses multipath diversity via the maximal-ratio combining (MRC) principle. Numerical results show the potential of DD-based uncoordinated schemes in the presence of double selectivity, while remarking the design tradeoffs and remaining challenges introduced by the proposed design.