Search Results (1,361)

Semantic communication systems leverage deep neural networks to extract and transmit essential information, achieving superior performance in bandwidth-constrained wireless environments. However, their vulnerability to backdoor attacks poses critical security threats, where adversaries can inject malicious triggers during training to manipulate system behavior. This paper introduces Selective Communication Unlearning (SCU), a novel defense mechanism based on Variational Information Bottleneck (VIB) principles. SCU employs a two-stage approach: (1) joint unlearning to remove backdoor knowledge from both encoder and decoder while preserving legitimate data representations, and (2) contrastive compensation to maximize feature separation between poisoned and clean samples. Extensive experiments on the RML2016.10a wireless signal dataset demonstrate that SCU achieves 629.5 ± 191.2% backdoor mitigation (5-seed average; 95% CI: [364.1%, 895.0%]), with peak performance of 1486% under optimal conditions, while maintaining only 11.5% clean performance degradation. This represents an order-of-magnitude improvement over detection-based defenses and fundamentally outperforms existing unlearning approaches that achieve near-zero or negative mitigation. We validate SCU across seven signal processing domains, four adaptive backdoor types, and varying SNR conditions, demonstrating unprecedented robustness and generalizability. The framework achieves a 243 s unlearning time, making it practical for resource-constrained edge deployments in 6G networks. Full article

Real-time estimation of vehicle sideslip angle is essential for both safety and performance applications. This study presents a temperature-adaptive Extended Kalman Filter (EKF) that estimates the sideslip angle of a racing vehicle by integrating dynamic and kinematic information. A temperature-dependent Pacejka tire model, derived directly from track tests, is embedded in a 3-degree-of-freedom dual-track vehicle model and used within the EKF to compensate for temperature-induced variations in tire behavior. The adaptive model parameters are identified from standard on-track maneuvers conducted at different tire temperatures, without the need for additional indoor rig testing. Experimental validation on a race track demonstrates that incorporating tire temperature adaptation and combining dynamic and kinematic estimation significantly enhance estimation accuracy, particularly underow-grip and high-performance driving conditions attested by a reduction of 40–50% in RMS error and a further reduction in maximum absolute error. Full article

Applying coatings that suppress the radiance changes related to temperature-dependent blackbody emission enables temperature-independent optical and sensing systems. Phase-change materials can significantly modify their optical properties within their transition window, but compensating for the large mid-wave infrared (MWIR, 3–5 µm) variation is demanding: blackbody radiance at 3 µm increases nearly 10-fold as the temperature rises from 30 °C to 80 °C. Vanadium dioxide VO₂, whose insulator–metal transition offers a sharp contrast and a low-loss insulating state, is attractive for applications in thermal management, but simple thin-film designs cannot provide full compensation. We demonstrate metasurface coatings that provide this compensation by constructing an array of metal–VO₂–metal antennas tuned to maintain constant thermal emission at a target wavelength over a temperature range of 30 °C to 80 °C. Antennas of several lateral sizes are combined, so their individual resonances collectively track the Planck change. This design provides both optical contrast and the correct temperature derivative, which are unattainable with homogeneous layers. Our approach results in a negligible apparent temperature change of the metasurface across the 30–80 °C range, effectively masking thermal signatures from MWIR detectors stemming from the low losses of VO₂. Full article

Multi-object tracking and segmentation (MOTS) is a critical task in video analysis with applications spanning autonomous driving, robot navigation, and scene understanding. MOTS has made significant progress but still faces persistent challenges, such as crowded scenes, abnormal illumination, and small objects. Several trackers have implemented attention mechanisms to overcome these difficulties. However, many attention mechanisms have quadratic computational complexity and use little spatio-temporal information. This paper proposes a Dual-Linear Attention Network (DLAN), a novel approach that effectively integrates both appearance and spatio-temporal information while maintaining linear attention complexity. DLAN employs recursive linear self-attention to strengthen the appearance representation and prototypical linear cross-attention to condense rich spatio-temporal information, which can compensate for missing pixel information. DLAN optimizes both image features and segmentation, with the refined segmentation guiding frame-level memory updates to improve instance consistency. Extensive experiments on BDD100K MOT, BDD100K MOTS, and KITTI MOTS datasets demonstrate the following: (1) The three main challenges of object occlusion, illumination variation, and distant objects have been successfully mitigated by integrating DLAN. (2) DLAN has achieved an overall competitive performance when compared to state-of-the-art trackers, with a 26% reduction in identity switches (IDS) when compared to QDTrack-mots-fix. Full article

As a bridge connecting processes in the Earth’s interior and the superficial lithosphere, distribution characteristics of global stress fields could benefit the verification of geodynamical models and reflect spatial variations of lithospheric strength. Numerical simulation of the global stress field could provide the temporal evolution process of the stress fields, and reveal the dynamic process of accumulation and release of the in situ stress fields as well as the quantitative relationship between force sources and the stress fields, which could compensate for the sparsity and insufficient representativeness of in situ stress observation data. To advance the investigation on the global stress fields, we review the state-of-the-art progress of numerical simulation tools for global stress fields and their applications, and show the existing problems as well as future trends. Full article

Surface defect detection is vital in industrial quality control. While deep learning has largely automated inspection, accurately locating defects with large-scale variations or those difficult to distinguish from similar backgrounds remains challenging. Furthermore, achieving high-precision and real-time performance under limited computational resources in deployment environments complicates effective solutions. In this work, we propose SD-GASNet, a network based on a self-distillation model compression strategy. To identify subtle defects, we design an Alignment, Enhancement, and Synchronization Feature Pyramid Network (AES-FPN) fusion network incorporating the Frequency Domain Information Gathering-and-Allocation (FIGA) mechanism and the Channel Synchronization (CS) module for industrial images from different sensors. Specifically, FIGA refines features via the Multi-scale Feature Alignment (MFA) module, then the Frequency-Guided Perception Enhancement Module (FGPEM) extracts high- and low-frequency information to enhance spatial representation. The CS module compensates for information loss during feature fusion. Addressing computational constraints, we adopt self-distillation with an Enhanced KL divergence loss function to boost lightweight model performance. Extensive experiments on three public datasets (NEU-DET, PCB, and TILDA) demonstrate that SD-GASNet achieves state-of-the-art performance with excellent generalization, delivering superior accuracy and a competitive inference speed of 180 FPS, offering a robust and generalizable solution for sensor-based industrial imaging applications. Full article

To address the issue that the output voltage and power of medium- and low-power wireless power transfer (WPT) systems cannot remain constant under coil misalignment, this paper proposes a single-switch WPT system with misalignment-tolerant characteristics. Based on a single-switch topology, the system combines the LCC-S and S-S compensation networks through an input-series and output-series connection, forming a simplified hybrid-compensated single-switch WPT topology. By exploiting the complementary output characteristics of the two compensation networks, a stable output voltage is achieved under varying mutual inductance conditions. To further enhance misalignment adaptability, a grid-type flat spiral (GFSP) coil is designed for the magnetic coupler. This coil configuration avoids magnetic flux cancelation during lateral displacement, while maintaining a consistent mutual inductance variation trend between the dual windings, thereby exhibiting strong tolerance to misalignment along the X-axis. The proposed system is validated through MATLAB/Simulink simulations and experiments on a 50 W prototype. The results demonstrate that the system maintains resonance and achieves zero-voltage switching (ZVS) of the power device under ±60 mm X-axis misalignment, with output voltage fluctuation below 4% and efficiency fluctuation below 3%, verifying the proposed system’s effectiveness in misalignment tolerance. Full article

This work introduces a tunable technique to push the low-frequency corner ($f_L$) of capacitively coupled instrumentation amplifiers (CCIAs) to the sub-mHz range for emerging biosensing applications. The proposed approach combines Complementary Transimpedance Boosting (CTB) to limit the DC feedback current and segmented duty-cycled resistors (SDR) for tunable resistance. The CTB-SDR technique achieves a stable effective post-layout pseudo-resistance of $535.8 \mathrm{T}\Omega$, equivalent to $f_L=660 \mu \mathrm{Hz}$ while occupying $0.062 {\mathrm{mm}}^2$ in a 180 nm process. According to JESD91 standards, it shows a standard deviation of $0.19 \mathrm{mHz}$ under post-layout Monte Carlo + process analysis, $1.1\%$ spread under voltage variations $(\pm 5.56\% V_{DD})$ and $6.2\%$ under temperature variations $(-20 °\mathrm{C}, 27 °\mathrm{C},$ and $60 °\mathrm{C})$. In addition, duty-cycling calibration can compensate for worst-case process corner variations and mismatch-induced feedback instability. Full article

Accurate deformation monitoring is essential for ensuring the stability of deep vertical shafts. In this study, a temperature-compensated fiber Bragg grating (FBG) sensing system was deployed in the 882 m deep Guotun Coal Mine shaft to measure circumferential and vertical strains at six depths. A site-specific mechanical model integrating stratigraphy, dual-layer concrete lining, and the influence radius was developed to analyze shaft wall stresses. The monitoring results reveal pronounced spatial anisotropy, with circumferential compressive and tensile strains at deeper levels nearly twice those at shallow levels. Strain variation also increases over time, reflecting the combined effects of groundwater fluctuations and overburden consolidation. The stresses inferred from measured strains agree well with the analytical solution in both magnitude and depth-dependent trend, with deviations remaining within a reasonable engineering margin. All stresses are below the strength limits of the C70/C50 concrete lining, confirming that the shaft is in a safe stress state. The proposed monitoring–analysis framework provides a reliable basis for evaluating shaft wall behavior under complex hydrogeological conditions. Full article

This paper proposes a dynamic compensation method for laser ranging based on pulse width for the miniaturization and high-precision requirements of the initiation device in precision-guided aircraft. The study aims to improve the measurement accuracy of the laser ranging unit in the initiation device system and ensure the accuracy and reliability of its fixed-distance initiation decision. The variation in echo pulse width is analyzed by studying laser echo characteristics. The pulse width and the detection distance exhibit an approximately linear negative correlation within the middle range of the applicable distance range. A dynamic compensation method is proposed based on a dual-correction approach using a static lookup table and dynamic compensation. This method establishes the mapping relationship between pulse width and distance deviation, and achieves distance correction by adding distance deviation compensation to the basic value from the static lookup table. The dynamic compensation system integrated with calibration and correction is designed and implemented, and the feasibility of the dynamic compensation method is verified by testing. The relative error between the calculated correction distance and the actual distance is small, and the average relative error is about 1.33%. The proposed method provides key technical support for the establishment of miniaturized and intelligent initiation devices. Full article

Dynamic Wireless Power Transfer (DWPT) systems require analytical models capable of capturing time-varying coupling and multi-transmitter interactions. However, most existing formulations address only static Wireless Power Transfer (WPT) or single-transmitter configurations, offering limited applicability to realistic DWPT scenarios. This paper addresses this gap by developing a comprehensive analytical framework for Series–Series (SS) compensated DWPT systems, supporting general n-transmitter/m-receiver architectures. The model is derived from coupled RLC circuit equations and expressed in normalized time- and frequency-domain forms, enabling analysis of resonance behavior, transient dynamics, and mutual-inductance variations during vehicle motion. To represent the continuous receiver motion, we establish a coupling-coefficient distribution covering the operating range of $k=0.11$ to $k=0.581$. The framework is then applied to three representative cases: a dynamic $1\times 1$ baseline, sequential transmitter activation, and simultaneous multi-transmitter activation. The study investigates system performance across varying operating frequencies and receiver positions to evaluate efficiency characteristics for $1\times 1$, $n\times 1$, and $n\times m$ wireless power transfer configurations. The proposed analytical framework provides a scalable basis for control development, transmitter coordination, and future real-time DWPT implementation. Full article

Surface subsidence, a major geological hazard induced by mining activities, severely compromises the sustainable economic development of mining areas and the safety and stability of residents’ livelihoods. Consequently, long-term and effective monitoring and prediction of mining areas are essential. Aiming to identify the key characteristic of narrow and elongated mining areas—where the strike length is significantly greater than the dip length—this study proposes a surface subsidence monitoring method integrating Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) and the Improved Probability Integral Method (IPIM). Specifically, this method utilizes SBAS-InSAR technology to acquire cumulative subsidence results of low-gradient deformation zones in mining areas. To address the issue of excessively fast edge convergence in traditional Probability Integral Method (PIM) applications for narrow and elongated mining areas, the traditional PIM is adjusted by modifying the dip-direction influence radius parameter; this adjustment alters the shape of the dip-direction subsidence curve at the edge of the subsidence basin, thereby resolving the convergence problem. Meanwhile, based on the InSAR deformation gradient theory, the subsidence edge and subsidence center are identified, and the corresponding threshold is determined. The results of SBAS-InSAR and IPIM are then fused via the inverse distance squared weighting (IDSW) method to eliminate discontinuous boundaries in fused results and obtain complete surface subsidence data of the mining area. Taking the 31109-1 working face of the Lijiahao Coal Mine as the study area, 14 scenes of Sentinel-1A imagery and field leveling data of the working face were used to validate the feasibility and accuracy of the proposed method. The results indicate that a distinct rectangular subsidence basin was formed in the working face during the monitoring period. The maximum subsidence measured by the integrated method is 3453 mm, and the location, subsidence curve, and variation trend of the monitored subsidence basin are basically consistent with actual mining conditions. The maximum relative errors of subsidence in the strike and dip directions are 5.2% and 4.1%, respectively. This method can effectively compensate for the limitations of SBAS-InSAR and PIM when applied individually to surface subsidence monitoring in narrow and elongated mining areas, enabling the acquisition of refined subsidence information for the entire mining basin. The research results provide a scientific basis for subsidence monitoring and early warning, disaster prevention and mitigation, and the rational development and utilization of resources in mining areas. Full article

This paper presents a 3-bit transmissive intelligent surface (TIS) using a novel technique that employs a unit cell comprising loaded semi-loop dipole resonators. The two resonators are anti-symmetrically oriented along the H-plane, functioning as transmitter and receiver on opposite sides of the TIS. The unit cell, with 13.2 mm periodicity, achieves 360° phase variation in 45° steps while maintaining insertion loss below 2 dB at 10 GHz. A 17 × 17 array TIS is designed using ray tracing and phase shift compensation techniques, with phase profiles distributed across eight discrete varactor states. The implemented TIS demonstrates a 10.8 dB gain enhancement for a horn antenna source at 10 GHz while preserving antenna matching, polarization, and radiation efficiency. The design achieves beam steering capabilities up to 60° with ±2° precision across elevation, azimuth, and inclined angles, maintaining an average steering gain loss of 3 dB over a 400 MHz bandwidth. These characteristics make the proposed design particularly effective for modern wireless coverage extension and tracking applications. Full article

To address the issues of high cost and poor low-temperature adaptability in cement-based backfill materials, this study developed a high-volume fly ash-based solid waste cementitious backfill material (FAPB) along with a specialized low-temperature admixture. Investigated the fundamental properties and microscopic curing mechanisms of the FAPB at different temperatures. The results indicate that the yield stress and plastic viscosity of FAPB slurry increase with higher contents of the curing agent and admixture, and rise as the temperature decreases. The variation in slump flow aligns with the rheological parameters, with the minimum slump flow being 14.5 cm (>10 cm). Bleeding rate increases with decreasing amounts of curing agent and admixture content, as well as lower temperatures, reaching a maximum bleeding rate of 9.26% (T5C10). Setting time decreases with increased amounts of curing agent and admixtures, and significantly increases with decreasing temperature. Strength increases with curing time and curing agent content, but decreases significantly as temperature drops. Adding admixtures can compensate for strength deterioration caused by low temperatures, with an optimal dosage of 3%. Microstructural analysis showed that the main hydration products of hardened backfill include AFt, C-S(A)-H, and Ca(OH)₂. Low temperature (5 °C) restricts hydration product formation, and the admixture facilitates continuous polycondensation of C-S(A)-H gel, resulting in sustained strength gain. This study provides a theoretical basis for the preparation and application of low-temperature-resistant Fly ash-based backfill materials, holding significant importance for advancing green mining practices in cold regions. Full article

Advancements in ship propulsion technologies are essential for improving the efficiency and reliability of maritime transportation. This study introduces a comprehensive approach that integrates motor design with sensorless control strategies, specifically focusing on Dual Three-Phase Permanent Magnet Vernier Motors (DTP-PMVM) for ship propulsion. The initial section of the paper explores the design of a 5-MW DTP-PMVM using finite element method (FEM) analysis in dual three-phase configurations. The subsequent section presents a novel sensorless control technique employing a Prescribed-time Sliding Mode Observer (PTSMO) for accurate speed and position estimation of the DTP-PMSM, eliminating the need for physical sensors. The proposed observer convergence time is entirely independent of the initial estimation guess and observer gains, allowing for pre-adjustment of the estimation error settling time. Initially, the observer is designed for a DTP-PMVM with fully known model parameters. It is then adapted to accommodate variations and unknown parameters over time, achieving prescribed-time observation. This is accomplished by using an adaptive observer to estimate the unknown parameters of the DTP-PMVM model and a Neural Network (NN) to compensate for the nonlinear effects caused by the model’s unknown terms. The adaptation laws are innovatively modified to ensure the prescribed time convergence of the entire adaptive observer. MATLAB (R2023b) Simulink simulations demonstrate the superior speed-tracking accuracy and robustness of the speed and position observer against model parameter variations, strongly supporting the application of these strategies in real-world maritime propulsion systems. By integrating these advancements, this research not only proposes a more efficient, reliable, and robust propulsion motor design but also demonstrates an effective control strategy that significantly enhances overall system performance, particularly for maritime propulsion applications. Full article