149 papers
This paper addresses the challenges of XL-MIMO radiomap prediction in the upper 6 GHz band by introducing a comprehensive multi-configuration dataset, a systematic benchmark framework, and a novel physics-informed "beam map" approach that significantly improves generalization to unseen array configurations and environments by decoupling deterministic radiation patterns from learned multipath propagation.
This paper introduces CELM, the first foundation model for clinical EEG-to-language generation that leverages a large-scale dataset of 9,922 reports and 11,000 hours of recordings to achieve significant improvements in summarizing long-term EEG data and generating comprehensive clinical reports.
LiveSense is a cross-platform system that transforms commercial off-the-shelf Wi-Fi 6E/7 laptops into real-time, centimeter-accurate Range-Doppler sensors capable of extracting synchronized channel state information and performing on-device signal processing to detect distance, velocity, and micro-motions while maintaining simultaneous communication.
StreamVoiceAnon+ is a streaming speaker anonymization system that preserves emotional content by combining supervised finetuning with neutral-emotion pairs and frame-level acoustic distillation, achieving significant improvements in emotion preservation (49.2% UAR) and intelligibility (5.77% WER) while maintaining strong privacy and zero inference latency overhead.
This study proposes and validates a GeoAI hybrid framework integrating MGWR, Random Forest, and ST-GCN to effectively model the spatiotemporal heterogeneity of multimodal traffic flows and their interaction with land use, demonstrating superior predictive accuracy and revealing distinct urban traffic typologies that underscore the critical role of local morphological context in mobility planning.
This paper establishes a unified multi-shot statistical framework for Rydberg-atom quantum receivers that derives optimal and practical phase-averaged likelihood-ratio tests, demonstrating that just 5–10 quantum shots enable detection performance significantly surpassing classical methods despite non-Gaussian noise and phase blindness.
This paper proposes RHOSI, an efficient resource allocation framework that leverages a Reconfigurable Holographic Surface (RHS)-aided UAV to jointly optimize beamforming, phase shifts, and spatial deployment, thereby significantly enhancing the anti-jamming resilience and throughput of Integrated Sensing and Communication (ISAC) systems.
This paper establishes the necessity of cross-epoch Doppler coupling for bounded phase uncertainty in LEO inter-satellite links, derives a corresponding posterior Cramér-Rao bound, and proposes a hybrid robust filtering framework that significantly outperforms standard extended Kalman filtering by effectively mitigating hardware impairments and measurement outliers.
This paper proposes a novel cascaded accumulator method that efficiently computes time-index powered weighted sums in real-time by reducing multiplicative complexity from to constant multiplications while eliminating the need for storing entire data blocks.
This paper proposes a knowledge distillation framework for mmWave beam tracking that leverages a large attention-enhanced neural network as a teacher to train a lightweight student model, achieving state-of-the-art long-term prediction accuracy with significantly reduced parameters, latency, and power consumption.
This paper proposes a novel, information-centric framework for deriving generic Cramér-Rao Lower Bounds (CRLBs) for rigid body localization that explicitly accounts for both translation and rotation errors under arbitrary measurement types and constraints, providing a versatile tool to evaluate and improve state-of-the-art estimation algorithms.
This paper proposes a low-complexity, compressive sensing-based 2D super-resolution technique to accurately estimate target signatures in XL-MIMO FMCW radars by effectively addressing the spatial wideband effect that couples range and angle domains.
This paper proposes an energy efficiency maximization framework for continuous aperture array (CAPA)-based multi-group multicast communications by developing an optimal block coordinate descent algorithm and a low-complexity zero-forcing approach, demonstrating CAPA's superior performance over discrete arrays while revealing nuanced impacts of aperture size and user distribution.
This paper provides a comprehensive survey of the evolution of anti-deception jamming techniques for radar systems, categorizing current strategies into prevention, detection, and mitigation while highlighting future research directions in distributed, cognitive, and AI-enabled radar technologies.
This paper explores the application of Generative Diffusion Models (GDMs) in massive MIMO communications by reviewing their theoretical foundations, analyzing recent advancements including a case study on near-field channel estimation, and outlining future research directions and challenges for enhancing channel state information acquisition in 5G and 6G networks.
This paper proposes a radar target tracking framework utilizing random finite sets and multiple hypothesis tracking to effectively mitigate main-beam range deception jamming, specifically range gate pull-off attacks, by modeling jammer-induced biases and false detections to maintain accurate target tracking and enable attack detection without relying on additional signal features.
This paper introduces the IW-DKF, a deep Kalman filter framework enhanced by importance sampling, which demonstrates improved generative modeling performance and more accurate state and parameter estimation for non-linear physics-based models compared to standard variational approaches.
This paper proposes a novel spatial IDFT technique to eliminate beam squint in massive phased arrays by counteracting inherent array-induced distortions, while also analyzing the underlying causes of squint and demonstrating how OFDM modulation can mitigate resulting signal quality issues.
This paper proposes an approximate expectation-maximization algorithm that directly reconstructs 3-D molecular structures from low signal-to-noise ratio cryo-EM micrographs, bypassing the traditional and often failing step of locating individual projection images.
This paper proposes a dual-fed pinching antenna system (DF-PAS) that mitigates in-waveguide attenuation by employing two feed points per waveguide for dynamic selection, deriving closed-form performance metrics and an efficient optimization framework to demonstrate its superior data rates over conventional single-fed designs in both single- and multi-waveguide scenarios.