147 papers
This paper proposes ALS-IRLS, a novel outlier-robust algorithm that integrates an innovation-level adaptive thresholding mechanism with an IRLS-based Huber cost function to significantly improve the accuracy of noise covariance estimation and state filtering in the presence of measurement outliers compared to conventional methods.
MetaSort is a novel algorithm that integrates an adaptive level crossing compression technique with meta-transfer learning-based feature representation to simultaneously achieve high-fidelity neural spike compression and robust few-shot classification, demonstrating strong potential for ultra-low-power on-chip implementation.
This paper proposes and validates a hardware-efficient, explainable Convolutional Tsetlin Machine (CTM) for real-time 5G jamming detection that achieves comparable accuracy to convolutional neural networks while significantly reducing training time, memory usage, and enabling deterministic FPGA deployment.
This study presents an automated pipeline using a wearable SonicGuard sensor and a pretrained Audio Spectrogram Transformer to accurately segment and classify bowel sounds, significantly reducing manual labeling time while providing clinicians with an objective, quantitative tool for assessing gastrointestinal function.
This paper proposes a fast, deterministic tensor completion method that leverages standard linear algebra to recover tensors from fiberwise observations along a single mode, offering efficient recovery guarantees without relying on random sampling assumptions.
This paper proposes a low-complexity manifold optimization framework that enforces reciprocity constraints on Beyond-Diagonal Reconfigurable Intelligent Surfaces (BD-RIS) to maximize sum-rate performance, demonstrating superior results compared to state-of-the-art approaches.
This paper proposes a learned optimization framework that combines deep unfolding with low-complexity approximated computations and data-driven hyperparameter learning to achieve state-of-the-art performance in hybrid beamforming and robust PCA while reducing computational complexity by over three orders of magnitude.
This paper proposes an analytical framework and receiver design for Frequency Modulated OFDM (FM-OFDM) as a constant-envelope ISAC solution that maximizes power amplifier efficiency by operating in saturation while maintaining robust sensing and communication performance under doubly dispersive channels.
This paper proposes a platform-aware framework that models 3D wireless links as a novel mutual antenna pattern, demonstrating that while individual platform effects are unidentifiable, the coupled pattern can be effectively estimated from noisy measurements to reduce path loss errors by up to 10 dB compared to traditional models.
This paper presents a systematic review of recent advancements in signal processing and learning over time-varying graphs, comparing methodologies across graph time-spectral filtering, multivariate time-series forecasting, and spatiotemporal neural networks while identifying current limitations and outlining future research directions.
This paper proposes a low-complexity phase selection method for multi-frequency multi-user RIS systems that divides the surface into user-specific subsurfaces, deriving exact closed-form mean SNR expressions for correlated Ricean and Rayleigh channels and demonstrating that iterative and converged versions of this approach outperform existing methods in robustness and efficiency despite reduced bandwidth per user.
This paper proposes Sign-Prioritized FL (SP-FL), a novel wireless federated learning framework that enhances model training reliability under resource constraints by prioritizing the transmission of gradient signs through a hierarchical resource allocation scheme, achieving up to 9.96% higher accuracy than existing methods on the CIFAR-10 dataset.
This paper demonstrates that modeling secondary electron counts as a mixture distribution, rather than a simple convolution, enables a maximum likelihood estimator that achieves significant sub-pixel edge localization accuracy—reducing root mean-squared error by approximately five-fold compared to conventional methods in both simulations and real helium ion microscopy datasets.
This paper introduces a novel language-guided framework that leverages pretrained vision-language models and a specialized adapter to achieve zero-shot, generative detection and localization of subsurface defects in carbon fiber-reinforced polymers using active infrared thermography, thereby eliminating the need for costly, task-specific training datasets while significantly improving signal-to-noise ratios and detection accuracy.
This paper proposes a robust 3-D trajectory optimization framework for movable antenna systems that minimizes the worst-case mean square angular error in direction estimation by deriving a closed-form performance bound and solving a min-max problem via successive convex approximation, thereby achieving isotropic sensing superior to fixed-position and 2-D movable antenna schemes.
This paper proposes a phase-noise resilient framework for mmWave massive MIMO systems using generalized receiver spatial modulation, which combines compact MQAM symbol pool design, enhanced spatial mapping strategies, and a practical single-stage compensation architecture to significantly mitigate phase noise effects while maintaining robust spatial detection.
This paper argues that the aggregate distortion model is overly pessimistic for monostatic ISAC sensing because the transmitter can monitor its own waveform, and it derives new PA-aware and PN-aware Cramér–Rao bounds to demonstrate that this approach reveals an irreducible velocity-error floor while significantly overestimating sensing degradation compared to practical scenarios.
This paper derives a novel exact analytical expression for the level crossing rate (LCR) of optimal single-user RIS-aided systems under blocked direct links and proposes a numerically stable approximation for direct-only channels, revealing that RIS systems effectively mitigate temporal channel variations, thereby easing the burden of channel state information acquisition.
This paper proposes a low-complexity RIS design where each user is served by a dedicated subsurface on a unique frequency band, deriving closed-form performance metrics that demonstrate optimal Line-of-Sight operation and remarkable robustness to non-Line-of-Sight conditions while significantly reducing channel estimation and processing requirements.
This paper introduces a novel, low-complexity Human Presence Detection system for commodity laptops that utilizes the built-in Wi-Fi hardware and a new Range-Filtered Doppler Spectrum technique to achieve privacy-preserving, calibration-free occupancy sensing without requiring external sensors or infrastructure.