Traffic noise is one of the main sources of environmental problems and a growing challenge for national traffic authorities. It is widely accepted that tire-pavement interaction is the main cause of traffic noise at speeds between 40 and 90 km/h. Typically, noise attenuation strategies include earthworks, tree belts, or noise barriers. However, a solution that is almost always viable is the use of low-noise pavements, which are characterized by their porous macrotexture, such as Stone Mastic Asphalt (SMA) mixtures. These mixtures are increasingly used for heavy traffic volumes because of their many advantages, including drainage properties and mechanical strength. Based on the experimental results obtained on different roads in southern Spain, this paper compares noise reduction in an SMA standard mixture due to the incorporation of different additives, such as crumb rubber and polymeric additives. According to the analysis, increasing the additives content by 1% reduces CPX by 1.18 decibels, approximately, and none of the analyzed sections shows increases greater than 3 dB within 24 months. Additionally, the paper proposes design recommendations regarding macrotexture and the percentage of voids for zones with warm, dry climates, such as Mediterranean Spain.

Abstract Martian exploration imagery is severely degraded by spatially heterogeneous dust, diminishing contrast and obscuring critical terrain details essential for geological analysis and navigation. Martian image dust removal is critically important, yet remains challenged by the failure of Earth-derived physical models, absolute absence of real paired data, and unphysical artifacts from unconstrained unsupervised methods. In this paper, we propose a novel physics-aware dehazing framework integrating: 1) A Multistage Fog Flow Network (MSFF-Net), a haze flow field-guided multi-stage network featuring a HazeFlowAligner module for deformable cross-scale feature alignment to explicitly resolve spatial heterogeneity, and Deformable Enhanced Convolution (Deformable ECV) modules with dynamic receptive fields to balance global haze removal and local detail preservation; 2) a terrain-adaptive data synthesis method using a physics-constrained CycleGAN to generate high-fidelity paired data for network training, which autonomously learns terrain-dependent dust density distributions and incorporates differentiable physical degradation modeling-including Perlin noise for realistic texturing, adaptive blur, and intensity-variant noise attenuation. Extensive experiments demonstrate state-of-the-art performance in perceptual quality and physical consistency compared to existing methods, achieving up to 46.7% and 77.8% improvements in PSNR and SSIM, respectively, on Martian dusty image, significantly outperforming prior arts, and showing exceptional generalization on real Tianwen-1 imagery and cross-domain SOTS benchmarks. This work establishes a new paradigm for physically-grounded visual restoration in planetary exploration and significantly enhances the accuracy of downstream tasks such as mineral mapping and autonomous navigation. Keywords： Mars dust, haze flow，CycleGAN，remote sensing image, dust removal

This paper presents a robust control framework for Boost Direct Current to Direct Current (DC-DC) converters based on µ-synthesis in MATLAB. The approach explicitly models structured parameter uncertainties, including the capacitor Equivalent Series Resistance (ESR) and load variations, and tailors the weighting filters to balance tracking performance, control effort, and noise attenuation. In simulations, the µ-controller achieves zero overshoot, settling time ≈ 1.2 ms, and steady-state error < 0.01 V, while a tuned Proportional-Integral-Derivative (PID) baseline exhibits ~12% overshoot, settling time ≈ 2.5 ms, and ≈ 0.45 V steady-state error under the same uncertainty set. Robust stability is certified by µ-bounds below unity across the design frequency band, and robust performance margins meet the specification. Novelty and contributions: explicit inclusion of capacitor ESR as a structured uncertainty in the modeling and synthesis loop; an implementation-oriented workflow (linearization → weighting design → D-K iteration → realization) with reproducible MATLAB code; and a quantitative benchmark versus a classical PID baseline under identical operating scenarios. The results support the deployment of the proposed controller in renewable and automotive applications that require resilience to parameter variations and fast transients.

This study investigated the noise reduction performance of porous asphalt concrete (PAC) pavement under tire–pavement coupling conditions, addressing the limitations of field measurements and laboratory testing. First, tire excitation amplitude parameters were determined based on vibrational contact operational scenarios. Then, finite element simulations were conducted to systematically analyzing the tire–pavement coupling noise characteristics of PAC pavement. The results indicate that PAC pavement effectively reduces the air pumping noise due to its highly porous internal structure, leading to significant noise attenuation. Furthermore, the study examined the key factors influencing the tire–pavement coupling noise in PAC pavement. When maintaining constant vehicle parameters (300 kg load, 60 km/h speed), pavement thickness became the critical noise-control variable, achieving minimum vibration at 6 cm surface layer thickness. Additionally, tire tread depth (5–17 mm) and mold release angle (0–30°) had a more pronounced impact on the air pumping noise compared to groove width (20–60 mm). Increasing the mold release angle and reducing tread depth effectively mitigated the air pumping noise. However, the tire–pavement coupling noise in PAC pavement increased considerably with increasing vehicle speed and load. Particularly, as the speed increased from 30 km/h to 60 km/h, the growth of the air pumping noise was most pronounced, revealing an acoustic transition of tire–pavement coupling noise from vibration-dominated to air-pumping-dominated mechanisms.

ABSTRACT This paper presents a new sliding mode filter for noise attenuation and derivative estimation. The proposed filter is an improved version of Kamal et al.'s sliding mode filter by incorporating an exponential approaching term to achieve a rapid response and alleviate overshoot during the response phase to some extent. The stability of the system is analyzed using the Lyapunov approach. In addition, the discrete‐time implementation of the proposed filter employs the semi‐implicit Euler scheme, and by exploiting the equivalence of the signum and saturation functions, chattering‐free outputs are achieved. Its performance is validated through numerical simulations.

Hybrid electromagnetic interference (EMI) filter is combined with passive filtering components and active cancellation part. This kind of EMI filter shows good noise attenuation ability and the total size is tiny. But during design procedure of hybrid EMI filter (HEF), passive part and active part are independent and the interaction between those two parts are not well considered. To cope with this problem, this letter proposes a new negative inductance based HEF (NI-HEF). By constructing a negative inductance based sensing structure, the sensing transformer could not only sense the common mode (CM) current, but also boost the impedance of primary windings, which in turn, helps to increase attenuation capability of CM noise. Together with active cancellation part, the whole filtering system could better block the propagation of CM noise. A prototype of proposed NI-HEF is built. Experimental results show that this new NI-HEF could achieve a dramatic noise attenuation with tiny size and weight.

Detection of deep fake has also become a top priority in digital security and media integrity domains, particularly with the widespread presence of sophisticated generative adversarial network (GAN)-based methods. The current paper presents a novel hybrid detection system that combines graph-based formulations, a diffusion-osmosis filter specifically developed for attenuation of adversarial noise, and a graph convolutional network (GCN) classifier. The proposed system is extensively evaluated on diverse benchmark datasets: DFDC, Celeb A, Face Forensics++ (FF++), and CIFAR-10. By converting the images into a graph structure using an initial processing layer of a diffusion-osmosis filter, the system efficiently nullifies adversarial perturbations induced by GANs, thereby improving the robustness of the detection pipeline. The GCN classifier utilizes the graphstructured data to attain state-of-the-art performance in real-synthetic media discrimination. Our experimental findings indicate that the proposed framework achieves detection accuracy between 98.71% and 99.87% on the tested datasets. These evidences suggest that the proposed framework is capable of addressing a wide range of deep fake cases, including high-quality face manipulations and refined adversarial attacks. Produces a state-of-theart scalable detection system which is immune to noise and may have potential for use in applications for content authentication, forensic analysis, and digital trust protection. The presented approach establishes a sound bench for future development of hybrid deep fake detection systems.

This paper presents a robust control strategy for a Buck-type DC–DC (BTDCDC) converter under voltage-mode control (VMC) using only output-voltage feedback. Robust control design is challenging due to parametric and unstructured uncertainties, set-point tracking, source-voltage and load disturbances, and high-frequency (HF) sensor noise. To address these issues, a Quantitative Feedback Theory (QFT) based two-degree-of-freedom (2-DOF) control architecture is developed. The advantages of the proposed QFT approach are: (i) 2-DOF control structure comprises PID controller and a pre-filter, which provides additional flexibility to improve the output voltage set-point tracking despite model variations (ii) Low transient inductor current, leads to subsequent reduced losses (iii) Inclusion of disturbance dynamics and an unstructured uncertainty in the controller loop-shaping design itself (iv) Reduced output voltage deviation for source and load disturbances (v) Improved suppression of HF sensor noise with less peak-to-peak inductor current and duty ratio variation. Experimental validation on a BTDCDC prototype confirms superior tracking and regulation performance. For source-voltage disturbances, the proposed method reduces peak voltage deviation by 30–50% and achieves settling times 4–8 faster. HF noise attenuation is also significantly improved.

South Africa’s electrical infrastructure deficit and poor maintenance plans have resulted in frequently occurring power outages, which has caused a decline in productivity, alongside an increase in vandalism. As a result, many households and businesses have resorted to alternative energy sources, such as petrol generators. However, petrol generators are loud, reaching acoustic levels of roughly 90 dB, and have proven negative impacts on society and the environment. To reduce the noise intensity produced by small petrol generators, a noise reduction mechanism has been designed. The mechanism incorporates several noise attenuation materials such as porous absorbers and perforated panels. The design also features both passive and active ventilation features for optimal cooling of the generator during use. With flow simulating software embedded in SolidWorks, the temperature, pressure and acoustic power level within the noise reduction mechanism was investigated. The sound box is shown to reduce the noise of the generator to approximately 10 dB, whereas the muffler, can reduce the noise intensity to 37 dB. The overall temperature within the sound box is roughly 75 ˚C, and the maximum pressure was determined to be 103 kPa, which is sufficient for operation of the generator, without compromising its performance and functionality. A modified experimental model was built to validate the numerical results. The results from the experiment produced similar results in terms of decrease in sound when the insulator is added.

Accurate evaluation of acoustic performance in built environments is challenged by the complex interplay of sound transmission mechanisms across architectural materials and interfaces. Conventional methods often assess airborne, structure-borne, or porous media sound behavior in isolation, neglecting the interdependencies that influence real-world performance. This study presents a multimodal experimental and computational framework for the comprehensive characterization of acoustic behavior in building assemblies. Using hybrid acoustic metrology, the research integrates standardized testing procedures to assess sound insulation, impact noise attenuation, and airflow resistivity across a range of construction materials. A ceramic brick partition wall between 2 classrooms yielded a weighted standardized level difference D nT,w of 42 dB, falling below regulatory thresholds due to low-frequency underperformance and flanking transmission. Impact noise testing revealed that a floating wooden floor significantly outperformed flexible vinyl, achieving a ΔLw of 16 dB and exhibiting improved attenuation above 400 Hz, despite resonance-related dips around 250–315 Hz. Airflow resistivity measurements on melamine foam, mineral wool, and porous concrete demonstrated method-dependent variability, with melamine showing the highest resistance and porous concrete the lowest. The comparison of alternating airflow and differential pressure techniques showed consistent material trends with slight methodological divergence. These findings emphasize the necessity of integrated acoustic diagnostics that consider material properties, frequency-dependent dynamics, and construction-related imperfections. The proposed framework enhances predictive fidelity and offers practical insights for optimizing acoustic design in educational, residential, and commercial environments.

In this literature review, synthesized studies on the software of Discrete Cosine Transformation (DCT) model for analysing electrocardiogram (ECG) indicators are articulated in a summary. The techniques found in this study are developed for the purposed (i) noise reduction (ii) function extraction and (iii) data compression that aim to (i) beautify diagnostic accuracy, (ii) tackling challenges inclusive of signal quality and (iii) information quantity management. The objectives of the evaluation included comparing DCT-driven noise attenuation methods, benchmarking the accuracy of function extraction, figuring out compression algorithms that hold scientific information, comparing hybrid approaches, and assessing their impacts on diagnostic accuracy and computational efficiency. A systematic analysis of studies making use of standard ECG databases and various DCTbased methodologies become performed, emphasizing quantitative metrics like compression ratio, percentage root mean rectangular distinction, and classification accuracy. Research by Qin et al. (2017) showcases a robust system for analyzing electrocardiogram (ECG) data. The system first cleans the signal using DCT-based filtering with adaptive thresholds, which effectively removes common noise like baseline wander and muscle interference without distorting the ECG's morphological features. This cleaned data is then used for feature extraction, where DCT coefficients are fed into a machine learning model, achieving exceptional arrhythmia classification accuracy of over 98% and highlighting its strong diagnostic potential. The analysis reveals a key trade-off in ECG signal compression as (i) Compression Performance: DCT and hybrid transforms achieve high compression ratios while preserving signal fidelity; (ii) Computational Cost: A significant drawback is their complexity, raising concerns about real-time use. This issue is exacerbated in integrative models that combine DCT with wavelets or classifiers for better performance; (iii) Overall Significance: DCT is confirmed as a critical component for optimizing ECG processing in telemedicine and real-time monitoring and finally (iv) Future Direction: These results guide the development of standardized, efficient DCT-based systems to enhance cardiac diagnostic workflows and device integration.

Vibration mitigation and noise attenuation in acoustic cavities by use of viscously damped metamaterial plate

Abstract As urbanization accelerates and population density rises, noise pollution increasingly undermines residents’ quality of life. While the link between noise distribution and urban morphology has been widely established, most studies focus on two-dimensional noise patterns and overlook its continuous three-dimensional variation. This study proposes a method that integrates urban section analysis with noise simulation. Taking typical residential forms in Nanjing as case studies, we treat profiles as core analytical units and incorporate noise simulation data from CadnaA to generate section-based maps that illustrate noise propagation paths and intensity variations. Based on a 50 threshold, high-noise and quiet areas are identified, and an area density index is introduced to enable quantitative comparison across residential morphologies. The results reveal that building form and layout significantly influence noise propagation and attenuation. Compact low-rise layouts foster stable quiet areas, whereas high-rise and wide-spaced layouts often create noise corridors that channel noise deeper into the site. Across cases, a consistent spatial pattern emerges: noise attenuates from the perimeter inward; the first building row provides substantial shielding; and noise troughs frequently occur at interfaces between buildings and open spaces. This section- based framework advances the analysis of coupling between urban morphology and acoustic environments and offers a rigorous basis for optimizing residential layouts and guiding urban acoustic environment management.

Noise pollution from construction activities is a major factor that threatens the health of workers, as well as the environment of people. However, previous research has mainly focused on reducing noise pollution around sensitive buildings outside the construction site, rather than for workers on the construction site. Moreover, currently available noise reduction methods tend to be passive and are accompanied by significant additional costs. Since the location of facilities is a critical factor in noise pollution, this study attempts to analyze how it is possible to reduce noise pollution for workers by optimizing the use of existing and innovative personal noise protection systems (PNPS). Given that noise reduction through the use of existing and innovative PSUs can negatively affect safety and costs, the preferred use of PNPS should strike a balance between noise reduction, safety improvement, and cost control. Therefore, functions have been developed that take into account the potential safety risk and transportation costs that arise as a result of interaction between production entities on the construction site. Information on the level of noise exposure was collected and processed; and data on the use of PNPS; on the use of PNPS in the workplace and outside of working hours. In addition, the noise attenuation provided by PSHS in employees in the workplace was measured. According to surveys, PNPS was used infrequently after work. Analysis of the combination of the results of continuous measurements of noise exposure and data on the use of PNPS showed that in fact PNPS reduced the influence of noise by less than 2 dB. This is a small amount, given that the level of noise exposure in this area is very high. The results show that noise protection programs should be implemented more fully and carefully; and that more attention should be paid to noise reduction in construction workers ' workplaces. The proposed research framework can also be used as a reference point for balancing conflicting Sustainable Development Goals at other industrial facilities. Evaluate the effectiveness of existing and introduce innovative hearing protection systems for workers servicing mechanized equipment at construction production sites. According to the workers of the PSZS, it does not always help to make the noise quieter. On average, PNPS reduce noise by less than 3 dB, and at a volume of more than 85 dB, PSUs help in 20 % of cases. In order for PNPS to be used more often, it is necessary to teach workers how to use PSZS correctly and make sure that they are accessible and work as they should. But even after hearing protection training, 25 % of construction workers still don't wear protection. This is confirmed by research in the world. They found that the attitude of builders to hearing protection does not change quickly. To solve the problem, you need to come up with simple and effective collective means of protecting against noise during construction.

This paper proposes a robust control method known as robust generalized incremental predictive control (RGIPC) for uncertain systems characterized by bounded uncertainties in both the system and input matrices. Conventional control strategies, such as H[Formula: see text] and traditional predictive controllers, face considerable challenges in maintaining a satisfactory balance between performance and robustness under such demanding conditions. To address these challenges, the newly framework integrates structured uncertainties into the optimization cost function, improving robustness in practical applications. This approach converts complex differential optimization equations into algebraic forms over a finite horizon, simplifying the derivation of optimal control signals. The RGIPC can be applied to control design by defining the design bounds. The proposed method comprises a two-step process: first, determining the appropriate design boundaries, and second, analytically computing the robust predictive control signal within the defined bounds during finite-horizon optimization. To thoroughly evaluate the proposed method, two benchmark systems are examined: an inverted pendulum, representing unstable and nonminimum phase dynamics, and a nonlinear multiple-input multiple-output (MIMO) flying object system exhibiting highly statically unstable dynamics with cross-coupling effects. Comparative analysis with H[Formula: see text] and generalized incremental predictive control (GIPC) methods confirms the superior performance of the proposed framework across all evaluation metrics. The obtained results demonstrate substantial enhancements in multiple key areas: Reduced overshoot, improved disturbance rejection, strengthened robustness against model uncertainties, enhanced noise attenuation capabilities, and more efficient control signals.

Impulsive noise poses a significant challenge to broadband feedforward active noise control (ANC) systems, particularly in sensitive environments such as infant incubators. This paper presents an adaptive impulsive noise cancellation approach based on the Kalman filter, designed to improve noise attenuation performance under nonstationary and impulsive interference. The proposed framework integrates impulsive noise detection with a Kalman filter-based suppression scheme. Simulation studies are conducted to evaluate the performance of the combined system in comparison to traditional ANC methods, such as Filtered-x Least Mean Square (FxLMS) and Filtered-x Normalized LMS (FxNLMS). Results demonstrate that the Kalman filter can effectively reduce the influence of impulsive disturbances without degrading overall broadband noise cancellation. A case study involving an infant incubator illustrates the practical effectiveness and robustness of the proposed technique in a real-world healthcare application. The findings support the integration of Kalman filter-based adaptive control in future ANC designs targeting impulsive noise environments.

Surface microseismic data (SMD) are usually presented as weak signals affected by strong interference. In this paper, with the purpose of obtaining available SMD, a deep learning framework combining with the complete ensemble empirical mode decomposition (CEEMD) and the multichannel denoising autoencoder (MDAE) is established to strengthen weak signals and suppress strong interference. First of all, a sort of EMD algorithm referred to as CEEMD is employed to decompose each trace of the SMD into the intrinsic mode functions (IMFs) so as to reduce the interference of random noise. Then, an MDAE algorithm is put forward to extract the effective and robust features of the IMFs, where a novel loss function without any label information is designed to achieve unsupervised noise attenuation in the real-world scenario. After that, the decomposed IMFs are reconstructed from high frequency to low frequency by using the extracted features such that the high-frequency part of the microseismic signals is retained effectively. Finally, the proposed CEEMD-MDAE model is applied to the noise attenuation in both synthetic and real-world SMD datasets. Experimental results demonstrate that the CEEMD-MDAE algorithm significantly improves the signal-to-noise ratio and outperforms some existing popular denoising algorithms.

Improving the quality of computed tomography (CT) medical images by reducing noise is important for maintaining important details and consistency for interpretations. Previous filtering techniques have been used to eliminate noise while retaining fine structures, but very often medical transmission images are obscured in noise which degrades the anatomical features of the images and detecting a subtle abnormality of interest becomes difficult for the radiologist's evaluation. Recent work has developed a mathematically optimized filtering method that combines wavelet thresholding and deep convolutional neural networks (CNN) to mitigate Gaussian noise from CT images using image quality index (IQI) and peak signal-to-noise ratio (PSNR). Though the new methods have improved the depiction over more traditional methods, the methods still struggle with information of fine detail and clarity of structures to the point of making them unusable in clinical settings. Herein lies the motivation for this study, to advance a third generation of filtering refinements for CT images that are impacted by additive white Gaussian noise (AWGN) attenuation. A new framework for filtering that uses anisotropic Gaussian filter (GF) and wavelet transforms combined with deep learning-based filtering convolutional networks (DCNN). This study's initial phase is to utilize GF combined with the Daubechies wavelet transform to mitigate AWGN. In the second phase of the study, in a post-processing mode, the DCNN will be combined with GF and wavelet processing. The specific use of GF is due to its ability to adapt to edge orientation and directional features, in order to avoid edge blurring due to an inhomogeneous noise distribution. The algorithm is evaluated using average PSNR and the average structural similarity index (SSIM). The results showed that the proposed ensemble achieved an average PSNR of 36.8 dB. Furthermore, SSIM results near 1.0 indicate high structural similarity from processing to original image. The proposed methodology also outperforms conventional methods with regards to detail preservation and overall image quality, hence making it a viable solution for CT image filtering.

Abstract High-quality seismic data serve as a critical foundation for imaging and interpretation. However, field seismic data are contaminated by noise, leading to numerous spurious seismic signals that adversely affect subsequent processing and interpretation. Currently, supervised deep learning (DL) methods face challenges due to the scarcity of authentic labels for 3D seismic images, while unsupervised approaches suffer from a lack of guidance during training, resulting in blurred denoising outcomes. To deal with these matters, it is essential to propose an adaptive semi-supervised random noise attenuation method that leverages pseudo-labels, which are easier to generate, to guide the model effectively. In this work, we put forward a semi-supervised gradient-penalized generative adversarial network with feature expansion (FEWGAN-SGP). The network first employs a global residual structure to ensure overall convergence during training. Subsequently, a channel attention mechanism with feature expansion is adopted to extract finer details. Finally, a smooth gradient penalty term is applied to enhance the distinction between field and generated images, thereby promoting deeper model learning. Experiments were conducted on both 2D and 3D datasets, and the results indicate that, compared with mainstream traditional methods and unsupervised learning approaches, the proposed method demonstrates excellent signal preservation and noise attenuation capabilities in practical applications.

This research investigates the acoustic wave propagation in one-dimensional quasi-periodic waveguide structures designed using generalized Thue-Morse sequences. The configuration comprises two distinct resonator blocks: one featuring parallel open/closed resonators and the other containing closed resonators arranged in series within a primary air duct. These structures can be fabricated from materials such as aluminum, brass, or steel. Using the transfer matrix and finite element methods, the study calculates acoustic transmittance under normal incidence, revealing the formation of low-frequency band gaps attributed to the local resonance effects of the quasi-periodic arrangement. The analysis highlights how increasing the cross-section ratio significantly broadens the central band gap in the transmittance spectra, enhancing noise attenuation. Furthermore, the quasi-periodic design generates multiple sharp resonance peaks with high transmittance within the band gaps. The findings show that the number and positioning of these peaks can be controlled by modifying the Thue-Morse sequence order and repetition parameters. Additionally, the study explores how expanding the number of blocks influences the reduction of transmittance in pass bands. These insights hold significant promise for applications in acoustic multiplexing devices. Besides, Thue-Morse sequence of lateral resonators demonstrates a record-high sensitivity of 17.2 Hz·s/m for gas detection, significantly outperforming all recent periodic and quasi-periodic designs.