Papers

This paper presents Density-based Predictive Control (DPC), a novel multi-agent control strategy for efficient non-uniform area coverage, grounded in optimal transport theory. In large-scale scenarios such as search and rescue or environmental monitoring, traditional uniform coverage fails to account for varying regional priorities. DPC leverages a pre-constructed reference distribution to allocate agents' coverage efforts, spending more time in high-priority or densely sampled regions. We analyze convergence conditions using the Wasserstein distance, derive an analytic optimal control law for unconstrained cases, and propose a numerical method for constrained scenarios. Simulations on first-order dynamics and linearized quadrotor models demonstrate that DPC achieves trajectories closely matching the non-uniform reference distribution, outperforming existing coverage methods.

This paper addresses the problem of frequency-weighted extended balanced truncation for discrete and continuous-time linear time-invariant plants. We show that the frequency-weighted discrete-time plant admits block-diagonal solutions to both the Lyapunov inequality and its extended form. A recursive algorithm for extended balanced truncation is proposed, together with corresponding a-priori error bounds. Theoretical results are extended to continuous-time systems and validated through numerical examples.

5G and beyond mobile telecommunication networks are increasingly embracing software technologies in their operation and control, similar to what has powered the growth of the cloud. This is most recently seen in the radio access network (RAN). In this new approach, the RAN is increasingly controlled by software applications known as xApps, and opens the door to third party development of xApps bringing diversity to the ecosystem, similar to mobile phone apps. This model aligns closely with the controllers in the ITU-T architecture for autonomous networks, and provides a pathway towards autonomous operation in the RAN. Unfortunately, no marketplace to host or supply xApps currently exists. This work describes our experiences in leveraging open-source O-RAN implementations to design and develop an xApp store.

The adaptation of Large-Scale Language Models (LLMs) to specific domains depends on high-quality fine-tuning datasets, particularly in instructional format (e.g., Question-Answer - Q&A). However, generating these datasets, particularly from unstructured sources such as call center audio recordings, poses a significant challenge due to the noisy and disorganized nature of the data. This paper presents a solution to this challenge by offering an end-to-end automated pipeline for generating Q&A instructional datasets from such recordings. The methodology developed comprises sequential steps of audio processing (including diarization, noise removal and automatic transcription), textual processing (cleaning, normalization, and anonymization), semantic extraction of customer demands and attendant responses using vector embeddings, and matching via semantic search to form the final Q&A pairs. As a result, the complete pipeline was successfully implemented, generating a dataset specifically formatted for Instruct Fine Tuning. The practical value and feasibility of the generated dataset were substantiated and functionally demonstrated through the successful fine-tuning of an LLM model (based on Llama 2 7B). The conclusion of the paper states that the proposed approach is viable for converting unstructured conversational data from call centers into valuable resources for training LLMs. This development has the potential to open up avenues for creating more effective AI systems for Q&A tasks in the customer service domain. The developed codes have been made publicly available to promote reproducibility and future research.

The emerging pinching antenna (PA) technology enables flexible antenna positioning for creating line-of-sight (LoS) links, thus offering substantial potential to facilitate ambient signal-based backscatter communication (BSC). This paper investigates PA-assisted BSC for enhanced communication and covertness in the presence of a randomly distributed eavesdropper. An optimization problem is formulated to maximize the uplink covert transmission rate by jointly optimizing the transmit power and antenna positions while satisfying both communication reliability and covertness constraints. An alternative optimization (AO)-based framework is proposed to solve this problem. Numerical results demonstrate that the proposed PA-BSC effectively mitigates the double near-far problem, where energy harvesting and backscatter transmission degrade simultaneously due to distance disparities, thereby improving downlink energy harvesting and uplink data transmission while maintaining covertness performance under practical deployment scenarios.

Analyzing underwater fish imagery is critical for ecological monitoring but remains difficult due to visual degradation and costly annotations. We introduce FishDetector-R1, a unified MLLM-based framework for fish detection, segmentation, and counting under weak supervision. On the DeepFish dataset, our framework achieves substantial gains over baselines, improving AP by 20% and mIoU by 10%, while reducing MAE by 30% and GAME by 35%. These improvements stem from two key components: a novel detect-to-count prompt that enforces spatially consistent detections and counts, and Reinforcement Learning from Verifiable Reward (RLVR) with a complementary scalable paradigm leveraging sparse point labels. Ablation studies further validate the effectiveness of this reward design. Moreover, the improvement generalizes well to other underwater datasets, confirming strong cross-domain robustness. Overall, FishDetector-R1 provides a reliable and scalable solution for accurate marine visual understanding via weak supervision. The project page for FishDetector-R1 is https://umfieldrobotics.github.io/FishDetector-R1.

With the rapid advancement of conversational and diffusion-based AI, there is a growing adoption of AI in educational services, ranging from grading and assessment tools to personalized learning systems that provide targeted support for students. However, this adaptability has yet to fully extend to the domain of children's speech, where existing models often fail due to their reliance on datasets designed for clear, articulate adult speech. Children, particularly those in early developmental stages or with speech and language pathologies, present unique challenges that current AI models and datasets are ill-equipped to handle. To address this, we introduce KidSpeak, a multi-task speech-enhanced Foundation Model capable of both generative and discriminative tasks specifically tailored to children's speech patterns. Our framework employs a two-stage training process that incorporates phonetic knowledge into the speech encoder, achieving an average accuracy of 87% across four separate tasks. Furthermore, recognizing the limitations of scalable human annotation and existing speech alignment tools, we propose the Flexible and Automatic Speech Aligner (FASA) and leverage the method to construct high quality datasets for training and evaluation. This novel alignment tool significantly improves the quality of aligned children's speech from noisy data, enhancing data quality by 13.6x compared to human annotations, as demonstrated on the CHILDES dataset. To the best of our knowledge, KidSpeak and FASA represent the first comprehensive solution designed for speech and language therapy in children, offering both a multi-purpose speech LLM and a robust alignment tool.

Wildlife monitoring with drones must balance competing demands: approaching close enough to capture behaviorally-relevant video while avoiding stress responses that compromise animal welfare and data validity. Human operators face a fundamental attentional bottleneck: they cannot simultaneously control drone operations and monitor vigilance states across entire animal groups. By the time elevated vigilance becomes obvious, an adverse flee response by the animals may be unavoidable. To solve this challenge, we present an edge-native, behavior-adaptive drone system for wildlife monitoring. This configurable decision-support system augments operator expertise with automated group-level vigilance monitoring. Our system continuously tracks individual behaviors using YOLOv11m detection and YOLO-Behavior classification, aggregates vigilance states into a real-time group stress metric, and provides graduated alerts (alert vigilance to flee response) with operator-tunable thresholds for context-specific calibration. We derive service-level objectives (SLOs) from video frame rates and behavioral dynamics: to monitor 30fps video streams in real-time, our system must complete detection and classification within 33ms per frame. Our edge-native pipeline achieves 23.8ms total inference on GPU-accelerated hardware, meeting this constraint with a substantial margin. Retrospective analysis of seven wildlife monitoring missions demonstrates detection capability and quantifies the cost of reactive control: manual piloting results in 14 seconds average adverse behavior duration with 71.9% usable frames. Our analysis reveals operators could have received actionable alerts 51s before animals fled in 57% of missions. Simulating 5-second operator intervention yields a projected performance of 82.8% usable frames with 1-second adverse behavior duration,a 93% reduction compared to manual piloting.

Generative models have the potential to transform the way we emulate Earth's changing climate. Previous generative approaches rely on weather-scale autoregression for climate emulation, but this is inherently slow for long climate horizons and has yet to demonstrate stable rollouts under nonstationary forcings. Here, we introduce Spatiotemporal Pyramid Flows (SPF), a new class of flow matching approaches that model data hierarchically across spatial and temporal scales. Inspired by cascaded video models, SPF partitions the generative trajectory into a spatiotemporal pyramid, progressively increasing spatial resolution to reduce computation and coupling each stage with an associated timescale to enable direct sampling at any temporal level in the pyramid. This design, together with conditioning each stage on prescribed physical forcings (e.g., greenhouse gases or aerosols), enables efficient, parallel climate emulation at multiple timescales. On ClimateBench, SPF outperforms strong flow matching baselines and pre-trained models at yearly and monthly timescales while offering fast sampling, especially at coarser temporal levels. To scale SPF, we curate ClimateSuite, the largest collection of Earth system simulations to date, comprising over 33,000 simulation-years across ten climate models and the first dataset to include simulations of climate interventions. We find that the scaled SPF model demonstrates good generalization to held-out scenarios across climate models. Together, SPF and ClimateSuite provide a foundation for accurate, efficient, probabilistic climate emulation across temporal scales and realistic future scenarios. Data and code is publicly available at https://github.com/stanfordmlgroup/spf .

We address the problem of verifying closed-loop contraction in nonlinear control systems whose controller and contraction metric are both parameterized by neural networks. By leveraging interval analysis and interval bound propagation, we derive a tractable and scalable sufficient condition for closed-loop contractivity that reduces to checking that the dominant eigenvalue of a symmetric Metzler matrix is nonpositive. We combine this sufficient condition with a domain partitioning strategy to integrate this sufficient condition into training. The proposed approach is validated on an inverted pendulum system, demonstrating the ability to learn neural network controllers and contraction metrics that provably satisfy the contraction condition.

We propose a low-power mobile low earth orbit (LEO) satellite communication architecture, employing double reconfigurable intelligent surfaces (RIS) to enhance energy efficiency and signal performance. With a distance between RISs that satisfies the far-field requirement, this architecture positions one small RIS each in the near-field of the satellite's antenna and the user on the ground. Moreover, we develop a path loss model for the double-RIS communication link, considering the near-field and far-field effects. Further, with the help of dual-stage beamforming, the proposed system maximizes the signal power and minimizes power consumption. Simulation results show that the proposed architecture can reduce the power consumption with 40 dB in the uplink, with a small $0.25^2$ $\text{m}^2$ RIS near the user, to communicate in energy-constrained LEO satellite communication circumstances.

Wavenumber-division multiplexing (WDM) was introduced as a counterpart of orthogonal frequency-division multiplexing in the spatial-frequency domain for line-of-sight holographic multiple-input multiple-output (MIMO) systems. In this paper, we extend WDM to holographic MIMO channels with non-line-of-sight (NLoS) propagation. We show that applying WDM to the NLoS channel yields the corresponding angular-domain representation, which we characterize through the power spectral factor and power spectral density. We further obtain a closed-form characterization for the case of isotropic scattering, recovering Jakes' isotropic model. The analysis is complemented by numerical results evaluating the degrees of freedom and ergodic capacity under both isotropic and non-isotropic scattering.

The impact of hardware impairments on the spectral efficiency of communication systems is well studied, but their effect on sensing performance remains unexplored. In this paper, we analyze the influence of hardware impairments on integrated sensing and communication (ISAC) systems in cluttered environments. We derive the sensing signal-to-clutter-plus-noise ratio (SCNR) and show that hardware distortions significantly degrade sensing performance by enhancing clutter-induced noise, which masks target echoes. The isotropic nature of transmit distortion due to multiple stream transmission further complicates clutter suppression. To address this, we propose a distortion- and clutter-aware precoding strategy that minimizes the deviation from the communication-optimized precoder while improving sensing robustness. We also propose an alternative power allocation-based approach that reduces computational complexity. Numerical results confirm the effectiveness of the proposed approaches in overcoming hardware- and clutter-induced limitations, demonstrating significant performance gains over distortion-unaware designs.

We study distributed control for a network of nonlinear, differentially flat subsystems subject to dynamic coupling. Although differential flatness simplifies planning and control for isolated subsystems, the presence of coupling can destroy this property for the overall joint system. Focusing on subsystems in pure-feedback form, we identify a class of compatible lower-triangular dynamic couplings that preserve flatness and guarantee that the flat outputs of the subsystems remain the flat outputs of the coupled system. Further, we show that the joint flatness diffeomorphism can be constructed from those of the individual subsystems and, crucially, its sparsity structure reflects that of the coupling. Exploiting this structure, we synthesize a distributed tracking controller that computes control actions from local information only, thereby ensuring scalability. We validate our proposed framework on a simulated example of planar quadrotors dynamically coupled via aerodynamic downwash, and show that the distributed controller achieves accurate trajectory tracking.

Breast cancer is still the second top cause of cancer deaths worldwide and this emphasizes the importance of necessary steps for early detection. Traditional diagnostic methods, such as mammography, ultrasound, and thermography, which have limitations when it comes to catching subtle patterns and reducing false positives. New technologies like artificial intelligence (AI) and deep learning have brought about the revolution in medical imaging analysis. Nevertheless, typical architectures such as Convolutional Neural Networks (CNNs) often have problems with modeling long-range dependencies. It explores the application of visual transformer models (here: Swin Tiny, DeiT, BEiT, ViT, and YOLOv8) for breast cancer detection through a collection of mammographic image sets. The ViT model reached the highest accuracy of 99.32% which showed its superiority in detecting global patterns as well as subtle image features. Data augmenting approaches, such as resizing croppings, flippings, and normalization, were further applied to the model for achieving higher performance. Although there were interesting results, the issues of dataset diversity and model optimization which present new avenues of research are also still present. Through this study, the crystal potential of transformer-based AI models in changing the detecting process of breast cancer and, thus, to patients health, is suggested.

This study compares baseline (J0) and 24-hour (J1) diffusion magnetic resonance imaging (MRI) for predicting three-month functional outcomes after acute ischemic stroke (AIS). Seventy-four AIS patients with paired apparent diffusion coefficient (ADC) scans and clinical data were analyzed. Three-dimensional ResNet-50 embeddings were fused with structured clinical variables, reduced via principal component analysis (<=12 components), and classified using linear support vector machines with eight-fold stratified group cross-validation. J1 multimodal models achieved the highest predictive performance (AUC = 0.923 +/- 0.085), outperforming J0-based configurations (AUC <= 0.86). Incorporating lesion-volume features further improved model stability and interpretability. These findings demonstrate that early post-treatment diffusion MRI provides superior prognostic value to pre-treatment imaging and that combining MRI, clinical, and lesion-volume features produces a robust and interpretable framework for predicting three-month functional outcomes in AIS patients.

Chaos is a fundamental feature of many complex dynamical systems, including weather systems and fluid turbulence. These systems are inherently difficult to predict due to their extreme sensitivity to initial conditions. Many chaotic systems are dissipative and ergodic, motivating data-driven models that aim to learn invariant statistical properties over long time horizons. While recent models have shown empirical success in preserving invariant statistics, they are prone to generating unbounded predictions, which prevent meaningful statistics evaluation. To overcome this, we introduce the Energy-Constrained Operator (ECO) that simultaneously learns the system dynamics while enforcing boundedness in predictions. We leverage concepts from control theory to develop algebraic conditions based on a learnable energy function, ensuring the learned dynamics is dissipative. ECO enforces these algebraic conditions through an efficient closed-form quadratic projection layer, which provides provable trajectory boundedness. To our knowledge, this is the first work establishing such formal guarantees for data-driven chaotic dynamics models. Additionally, the learned invariant level set provides an outer estimate for the strange attractor, a complex structure that is computationally intractable to characterize. We demonstrate empirical success in ECO's ability to generate stable long-horizon forecasts, capturing invariant statistics on systems governed by chaotic PDEs, including the Kuramoto–Sivashinsky and the Navier–Stokes equations.

Federated Learning (FL) on resource-constrained edge devices faces a critical challenge: The computational energy required for training Deep Neural Networks (DNNs) often dominates communication costs. However, most existing Energy-Harvesting FL (EHFL) strategies fail to account for this reality, resulting in wasted energy due to redundant local computations. For efficient and proactive resource management, algorithms that predict local update contributions must be devised. We propose a lightweight client scheduling framework using the Version Age of Information (VAoI), a semantics-aware metric that quantifies update timeliness and significance. Crucially, we overcome VAoI's typical prohibitive computational cost, which requires statistical distance over the entire parameter space, by introducing a feature-based proxy. This proxy estimates model redundancy using intermediate-layer extraction from a single forward pass, dramatically reducing computational complexity. Experiments conducted under extreme non-IID data distributions and scarce energy availability demonstrate superior learning performance while achieving energy reduction compared to existing baseline selection policies. Our framework establishes semantics-aware scheduling as a practical and vital solution for EHFL in realistic scenarios where training costs dominate transmission costs.

The global capacity for mineral processing must expand rapidly to meet the demand for critical minerals, which are essential for building the clean energy technologies necessary to mitigate climate change. However, the efficiency of mineral processing is severely limited by uncertainty, which arises from both the variability of feedstock and the complexity of process dynamics. To optimize mineral processing circuits under uncertainty, we introduce an AI-driven approach that formulates mineral processing as a Partially Observable Markov Decision Process (POMDP). We demonstrate the capabilities of this approach in handling both feedstock uncertainty and process model uncertainty to optimize the operation of a simulated, simplified flotation cell as an example. We show that by integrating the process of information gathering (i.e., uncertainty reduction) and process optimization, this approach has the potential to consistently perform better than traditional approaches at maximizing an overall objective, such as net present value (NPV). Our methodological demonstration of this optimization-under-uncertainty approach for a synthetic case provides a mathematical and computational framework for later real-world application, with the potential to improve both the laboratory-scale design of experiments and industrial-scale operation of mineral processing circuits without any additional hardware.

Fast time-domain algorithms have been developed in signal processing applications to reduce the multiplication complexity. For example, fast convolution structures using Cook-Toom and Winograd algorithms are well understood. Short length fast convolutions can be iterated to obtain fast convolution structures for long lengths. In this paper, we show that well known fast convolution structures form the basis for design of fast algorithms in four other problem domains: fast parallel filters, fast polynomial modular multiplication, and fast pointwise multiplication in the DFT and NTT domains. Fast polynomial modular multiplication and fast pointwise multiplication problems are important for cryptosystem applications such as post-quantum cryptography and homomorphic encryption. By establishing the equivalence of these problems, we show that a fast structure from one domain can be used to design a fast structure for another domain. This understanding is important as there are many well known solutions for fast convolution that can be used in other signal processing and cryptosystem applications.

In a recent paper [8], we introduced a data-based approach to design event-triggered controllers for linear systems directly from data. Here, we extend the results in [8] to a class of nonlinear systems. We provide two data-based designs certified by a (classical) Lyapunov function. For these two designs, we devise event-triggered policies that rely on the previously found Lyapunov function, have parameters tuned from data, ensure a positive minimum inter-event time, and act based either on the state error or on the library error. These two different policies, and their respective advantages, are illustrated numerically.

The integration of photovoltaic (PV) generation and electric vehicle (EV) charging introduces significant uncertainty in electricity consumption patterns, particularly at the distribution level. This paper presents a comparative study for selecting metrics for uncertainty quantification (UQ) for net load profiles of residential, industrial, and office buildings under increased DER penetration. A variety of statistical metrics is evaluated for their usefulness in quantifying uncertainty, including, but not limited to, standard deviation, entropy, ramps, and distance metrics. The proposed metrics are classified into baseline-free, with baseline and error-based. These UQ metrics are evaluated for increased penetration of EV and PV. The results highlight suitable metrics to quantify uncertainty per consumer type and demonstrate how net load uncertainty is affected by EV and PV adoption. Additionally, it is observed that joint consideration of EV and PV can reduce overall uncertainty due to compensatory effects of EV charging and PV generation due to temporal alignment during the day. Uncertainty reduction is observed across all datasets and is most pronounced for the office building dataset.

Medical image registration drives quantitative analysis across organs, modalities, and patient populations. Recent deep learning methods often combine low-level "trend-driven" computational blocks from computer vision, such as large-kernel CNNs, Transformers, and state-space models, with high-level registration-specific designs like motion pyramids, correlation layers, and iterative refinement. Yet, their relative contributions remain unclear and entangled. This raises a central question: should future advances in registration focus on importing generic architectural trends or on refining domain-specific design principles? Through a modular framework spanning brain, lung, cardiac, and abdominal registration, we systematically disentangle the influence of these two paradigms. Our evaluation reveals that low-level "trend-driven" computational blocks offer only marginal or inconsistent gains, while high-level registration-specific designs consistently deliver more accurate, smoother, and more robust deformations. These domain priors significantly elevate the performance of a standard U-Net baseline, far more than variants incorporating "trend-driven" blocks, achieving an average relative improvement of $\sim3\%$. All models and experiments are released within a transparent, modular benchmark that enables plug-and-play comparison for new architectures and registration tasks (https://github.com/BailiangJ/rethink-reg). This dynamic and extensible platform establishes a common ground for reproducible and fair evaluation, inviting the community to isolate genuine methodological contributions from domain priors. Our findings advocate a shift in research emphasis: from following architectural trends to embracing domain-specific design principles as the true drivers of progress in learning-based medical image registration.

Learning site-specific beams that adapt to the deployment environment, interference sources, and hardware imperfections can lead to noticeable performance gains in coverage, data rate, and power saving, among other interesting advantages. This learning process, however, typically requires a large number of active interactions/iterations, which limits its practical feasibility and leads to excessive overhead. To address these challenges, we propose a digital twin aided codebook learning framework, where a site-specific digital twin is leveraged to generate synthetic channel data for codebook learning. We also propose to learn separate codebooks for line-of-sight and non-line-of-sight users, leveraging the geometric information provided by the digital twin. Simulation results demonstrate that the codebook learned from the digital twin can adapt to the environment geometry and user distribution, leading to high received signal-to-noise ratio performance. Moreover, we identify the ray-tracing accuracy as the most critical factor in digital twin fidelity that impacts the learned codebook performance.

Autonomous ground vehicles (AGVs) must navigate safely in cluttered environments while accounting for complex dynamics and environmental uncertainty. Hamilton-Jacobi Reachability (HJR) offers formal safety guarantees through the computation of forward and backward reachable sets, but its application is hindered by poor scalability in environments with numerous obstacles. In this paper, we present a novel framework called NeuroHJR that leverages Physics-Informed Neural Networks (PINNs) to approximate the HJR solution for real-time obstacle avoidance. By embedding system dynamics and safety constraints directly into the neural network loss function, our method bypasses the need for grid-based discretization and enables efficient estimation of reachable sets in continuous state spaces. We demonstrate the effectiveness of our approach through simulation results in densely cluttered scenarios, showing that it achieves safety performance comparable to that of classical HJR solvers while significantly reducing the computational cost. This work provides a new step toward real-time, scalable deployment of reachability-based obstacle avoidance in robotics.

In this paper, we consider the problem of designing input signals for an unknown linear time-invariant system in such a way that the resulting input-state data is suitable for identification or stabilization. We will take into account prior knowledge on system-theoretic properties of the system, in particular, controllability and stabilizability. For this, we extend the notion of universal inputs to incorporate prior knowledge on the system. An input is called universal for identification (resp., stabilization) if, when applied to any system complying with the prior knowledge, it results in data suitable for identification (resp., stabilization) regardless of the initial condition. We provide a full characterization of such universal inputs. In addition, we discuss online experiment design using prior knowledge, and we study cases where this approach results in the shortest possible experiment for identification and stabilization.

In today's digital age, access to the Internet is essential, yet a significant digital divide exists, particularly in rural areas of developing nations. This paper presents a Delay Tolerant Networking (DTN) framework that utilizes informal public transportation systems, such as minibus taxis, as mobile data mules to enhance connectivity in these underserved regions. We develop a probabilistic model to capture the randomness in vehicle mobility, including travel times and contact durations at bus stops. Key performance metrics are analyzed, including average data transmission rate and Peak Age of Information (PAoI), to assess the effectiveness of the proposed system. An analytical approximation for the Mean PAoI (MPAoI) is derived and validated through simulations. Case studies from real-world datasets in Nouakchott, Accra, and Addis Ababa demonstrate the practical applicability and scalability of our framework. The findings indicate that leveraging existing transportation networks can significantly bridge the digital divide by providing reliable internet-like connectivity to remote areas.

The high penetration of inverter-based resources (IBRs) reduces system inertia, leading to frequency stability concerns, especially during synchronous generator (SG) outages. To maintain frequency dynamics within secure limits while ensuring economic efficiency, frequency-constrained optimal power flow (FCOPF) is employed. However, existing studies either neglect the frequency support capability and allocation of grid-forming (GFM) IBRs or suffer from limited accuracy in representing frequency dynamics due to model simplifications. To address this issue, this paper proposes a deep learning (DL)-based FCOPF (DL-FCOPF) framework. A DL model is first developed as a predictor to accurately estimate frequency-related metrics: the required reserved headroom and allocation of GFM IBRs, the rate of change of frequency and frequency nadir. After being trained with data obtained from electromagnetic transient simulations, the DL model is reformulated and incorporated into FCOPF. Case studies conducted on two test systems demonstrate the effectiveness of the proposed approach. Compared with the traditional OPF and linear FCOPF benchmarks, the DL-FCOPF can optimally coordinate SGs and IBRs with minimum cost, achieving desired frequency response, within an acceptable computing time. Further-more, sensitivity analyses are conducted to identify the most suit-able structure and linearization approach of the DL-based frequency predictor.

We consider a distributed multiple-input multiple-output (D-MIMO) testbed in which, to enable coherent-phase transmission without over-the-air synchronization, the remote radio heads (RRHs) are connected to a central unit via a 1-bit radio-over-fiber fronthaul. Specifically, 1-bit samples of the radio-frequency signal are exchanged over the fronthaul. We investigate via both measurements and simulations based on an accurate model of the testbed hardware, the capability of the proposed architecture to provide uniform quality of services over the coverage area–one of the promises of D-MIMO. Our results are encouraging: for the case in which two user equipments (UEs) communicate over the same 75MHz signal bandwidth, the measured error-vector magnitude meets the 3GPP New Radio specification of 12.5% for 16QAM across all tested DMIMO scenarios. We also determine that uplink transmission is a potential bottleneck, due to the limited dynamic range of the automatic gain controller, which prevents the 1-bit quantizer to benefit from dithering. We show that this issue can be mitigated via UE power control.

Over-the-air (OtA) computation enables scalable analog aggregation by exploiting the superposition property of wireless channels, making it an attractive joint communication and computation paradigm for distributed sensing and learning. However, the uncoded nature of analog transmission exposes the computation result to eavesdropping, and the fundamental security limits of OtA computation against multiple cooperating adversaries remain poorly understood. In this paper, we develop an estimation-theoretic framework for analyzing the security of analog OtA computation in the presence of multiple distributed eavesdroppers that may jointly process their observations. We first derive the optimal estimator for cooperating eavesdroppers and bounds on the achievable estimation accuracy of both the legitimate receiver and the adversaries. Our analysis reveals a key insight: while random channel phase misalignment provides significant inherent MSE-security against individual eavesdroppers, this protection largely disappears once multiple eavesdroppers cooperate. Motivated by this observation, we propose a correlated artificial noise design based on zero-forcing that preserves the aggregation accuracy at the legitimate receiver while maximizing the estimation error at the cooperative eavesdroppers. Numerical results demonstrate that the proposed design substantially reduces the security advantage gained through eavesdropper cooperation and achieves security close to uncorrelated artificial-noise schemes without sacrificing computation accuracy. These results provide both a theoretical characterization of the security limits of analog OtA computation and a practical design guideline for its secure deployment in real-world wireless systems.