Noise Detection Research Articles

Implementation of supervised machine learning classification for detection and severity determination of electrified power train noise and vibration fault diagnosis

Vehicles equipped with electrified powertrains produce lower sound and vibration levels compared to those equipped with internal combustion engine powertrains. This makes noise and vibration (N&V) from other non-engine components more perceptible. Gear growl is one of the newly observed N&V that brings concerns by the passengers and manufacturers. The understanding of signal characteristics and the threshold for determining whether gear growl requires attention remains limited. To address this, supervised machine learning classification is employed. Root-mean-square (RMS) and spectral entropy values are sufficient for the classification of vibration data with test accuracy of 0.983. However, the acoustic signal required more features due to background noise, making data linearly inseparable. Features that describe the characteristics of acoustic data are studied, extracted, and selected. Utilizing a support vector machine (SVM) for classification, the study achieves an average test accuracy of 0.918. Further, a multi-class classification model is implemented based on preliminary subjective listening studies, classifying different severities of gear growl. Further listening studies are suggested for improving multi-class classification performance. Methods described in this study mainly focus on the analysis of gear growl, but they can be generalized for N&V signal-based fault diagnosis applications.

Analisa Kebisingan Pada Alat Berat Wheel Loader Wa-350 Berdasarkan Perbandingan Jarak 50, 100, 150 Cm Dan Kapasitas

Wheel loader is a tool that facilitates work by using hydraulic energy. Wheel loader testing is carried out by looking for noise data arising in the combustion chamber. There are several stages that must be done, namely the first step is to prepare a sound level meter as a noise detection device for heavy equipment, wheel loader, then take measurements with several variations in the distance between the combustion chamber and the sound level meter. In the next step we can find out the noise level read at the sound level meter with ideal, medium, and maximum rotation on the WA-350 wheel loader. So that the Noise Threshold Value according to Permenaker No. per-51 / MEN / 1999, ACGIH, 2008 and SNI 16-7063-2004 is 85 dB. So the noise produced by the Wheel Loader above the NAB Permenaker 104.7dB can cause tingling, unwell, hearing saturation, stomach pain, and circulatory problems. It is recommended that workers are required to use earmuffs or ear muffs.

Lyman Continuum Leakers in the AstroSat Ultraviolet Deep Field: Extreme-ultraviolet Emitters at the Cosmic Noon

We report the direct detection of Lyman continuum (LyC) emission from nine galaxies and one active galactic nucleus at z ∼ 1.1–1.6 in the GOODS-North field using deep observations from the Ultraviolet Imaging Telescope (UVIT) on board AstroSat. The absolute escape fraction of the sources estimated from the far-ultraviolet and Hα-line luminosities using Monte Carlo analysis of two intergalactic medium models span a range ∼10%–55%. The rest-frame UV wavelength of the sources falls in the extreme-ultraviolet regime ∼550–700 Å, the shortest LyC wavelength range probed so far. This redshift range remains devoid of direct detections of LyC emission due to the instrumental limitations of previously available facilities. With UVIT having very low detector noise, each of these sources is detected with an individual signal-to-noise ratio (S/N) > 3, while for the stack of six sources, we achieve an S/N ∼ 7.4. The LyC emission is seen to be offset from the optical centroids and extended beyond the UVIT point-spread function of 1.″6 in most of the sources. This sample fills an important niche between GALEX and Cosmic Origins Spectrograph at low z and Hubble Space Telescope's Wide Field Camera 3 at high z and is crucial in understanding the evolution of LyC leakers.

Performance evaluation of deep learning image reconstruction algorithm for dual-energy spectral CT imaging: A phantom study.

To evaluate the performance of deep learning image reconstruction (DLIR) algorithm in dual-energy spectral CT (DEsCT) as a function of radiation dose and image energy level, in comparison with filtered-back-projection (FBP) and adaptive statistical iterative reconstruction-V (ASIR-V) algorithms. An ACR464 phantom was scanned with DEsCT at four dose levels (3.5 mGy, 5 mGy, 7.5 mGy, and 10 mGy). Virtual monochromatic images were reconstructed at five energy levels (40 keV, 50 keV, 68 keV, 74 keV, and 140 keV) using FBP, 50% and 100% ASIR-V, DLIR at low (DLIR-L), medium (DLIR-M), and high (DLIR-H) settings. The noise power spectrum (NPS), task-based transfer function (TTF) and detectability index (d') were computed and compared among reconstructions. NPS area and noise increased as keV decreased, with DLIR having slower increase than FBP and ASIR-V, and DLIR-H having the lowest values. DLIR had the best 40 keV/140 keV noise ratio at various energy levels, DLIR showed higher TTF (50%) than ASIR-V for all materials, especially for the soft tissue-like polystyrene insert, and DLIR-M and DLIR-H provided higher d' than DLIR-L, ASIR-V and FBP in all dose and energy levels. As keV increases, d' increased for acrylic insert, and d' of the 50 keV DLIR-M and DLIR-H images at 3.5 mGy (7.39 and 8.79, respectively) were higher than that (7.20) of the 50 keV ASIR-V50% images at 10 mGy. DLIR provides better noise containment for low keV images in DEsCT and higher TTF(50%) for the polystyrene insert over ASIR-V. DLIR-H has the lowest image noise and highest detectability in all dose and energy levels. DEsCT 50 keV images with DLIR-M and DLIR-H show potential for 65% dose reduction over ASIR-V50% withhigher d'.

Enhancing in-situ updates of quantized memristor neural networks: a Siamese network learning approach.

Brain-inspired neuromorphic computing has emerged as a promising solution to overcome the energy and speed limitations of conventional von Neumann architectures. In this context, in-memory computing utilizing memristors has gained attention as a key technology, harnessing their non-volatile characteristics to replicate synaptic behavior akin to the human brain. However, challenges arise from non-linearities, asymmetries, and device variations in memristive devices during synaptic weight updates, leading to inaccurate weight adjustments and diminished recognition accuracy. Moreover, the repetitive weight updates pose endurance challenges for these devices, adversely affecting latency and energy consumption. To address these issues, we propose a Siamese network learning approach to optimize the training of multi-level memristor neural networks. During neural inference, forward propagation takes place within the memristor neural network, enabling error and noise detection in the memristive devices and hardware circuits. Simultaneously, high-precision gradient computation occurs on the software side, initially updating the floating-point weights within the Siamese network with gradients. Subsequently, weight quantization is performed, and the memristor conductance values requiring updates are modified using a sparse update strategy. Additionally, we introduce gradient accumulation and weight quantization error compensation to further enhance network performance. The experimental results of MNIST data recognition, whether based on a MLP or a CNN model, demonstrate the rapid convergence of our network model. Moreover, our method successfully eliminates over 98% of weight updates for memristor conductance weights within a single epoch. This substantial reduction in weight updates leads to a significant decrease in energy consumption and time delay by more than 98% when compared to the basic closed-loop update method. Consequently, this approach effectively addresses the durability requirements of memristive devices.

Recurrence Rate spectrograms for the classification of nonlinear and noisy signals

Time series analysis of real-world measurements is fundamental in natural sciences and engineering, and machine learning has been recently of great assistance especially for classification of signals and their understanding. Yet, the underlying system’s nonlinear response behaviour is often neglected. Recurrence Plot (RP) based Fourier-spectra constructed through τ-Recurrence Rate (RR τ ) have shown the potential to reveal nonlinear traits otherwise hidden from conventional data processing. We report a so far disregarded eligibility for signal classification of nonlinear time series by training RESnet-50 on spectrogram images, which allows recurrence-spectra to outcompete conventional Fourier analysis. To exemplify its functioning, we employ a simple nonlinear physical flow of a continuous stirred tank reactor, able to exhibit exothermic, first order, irreversible, cubic autocatalytic chemical reactions, and a plethora of fast-slow dynamics. For dynamics with noise being ten times stronger than the signal, the classification accuracy was up to ≈ 75% compared to ≈ 17% for the periodogram. We show that an increase in entropy only detected by the RR τ allows differentiation. This shows that RP power spectra, combined with off-the-shelf machine learning techniques, have the potential to significantly improve the detection of nonlinear and noise contaminated signals.

MV-based relative electron density estimation (iMREDe) for MR-LINAC dose calculation.

MR-LINAC systems have been increasingly utilized for real-time imaging in adaptive treatments worldwide. Challenges in MR representation of air cavities and subsequent estimation of electron density maps impede planning efficiency and may lead to dose calculation uncertainties. To demonstrate the generation of accurate electron density maps using the primary MV beam with a flat-panel imager. The ViewRay MRIdian MR-LINAC system was modeled digitally for Monte Carlo simulations. Iron shimming, the magnetic field, and the proposed flat panel detector were included in the model. The effect of the magnetic field on the detector response was investigated. Acquisition of projections over 360 degrees was simulated for digital phantoms of the Catphan 505 phantom and a patient treated for Head and Neck cancer. Shim patterns on the projections were removed and detector noise linearity was assessed. Electron density maps were generated for the digital patient phantom using the flat-panel detector and compared with actual treatment planning CT generated electron density maps of the same patient. The effect of the magnetic field on the detector point-spread function (PSF) was found to be substantial for field strengths above 50 mT. Shims correction in the projection images using air normalization and in-painting effectively removed reconstruction artifacts without affecting noise linearity. The relative difference between reconstructed electron density maps from the proposed method and electron density maps generated from the treatment planning CT was 11% on average along all slices included in the iMREDe reconstruction. The proposed iMREDe technique demonstrated the feasibility of generating accurate electron densities for the ViewRay MRIdian MR-LINAC system with a flat-panel imager and the primary MV beam. This work is a step towards reducing the time and effort required for adaptive radiotherapy in the current ViewRay MR-LINAC systems.

Joint gravitational wave-short GRB detection of binary neutron star mergers with existing and future facilities

ABSTRACT We explore the joint detection prospects of short gamma-ray bursts (sGRBs) and their gravitational wave (GW) counterparts by the current and upcoming high-energy GRB and GW facilities from binary neutron star (BNS) mergers. We consider two GW detector networks: (1) a four-detector network comprising LIGO Hanford, Livingston, Virgo, and Kagra (IGWN4) and (2) a future five-detector network including the same four detectors and LIGO India (IGWN5). For the sGRB detection, we consider existing satellites Fermi and Swift and the proposed all-sky satellite Daksha. Most of the events for the joint detection will be off-axis, hence, we consider a broad range of sGRB jet models predicting the off-axis emission. Also, to test the effect of the assumed sGRB luminosity function, we consider two different functions for one of the emission models. We find that for the different jet models, the joint sGRB and GW detection rates for Fermi and Swift with IGWN4 (IGWN5) lie within 0.07–0.62 yr−1 0.8–4.0 yr−1) and 0.02–0.14 yr−1 (0.15–1.0 yr−1), respectively, when the BNS merger rate is taken to be 320 Gpc−3 yr−1. With Daksha, the rates increase to 0.2–1.3 yr−1 (1.3–8.3 yr−1), which is 2–9 times higher than the existing satellites. We show that such a mission with higher sensitivity will be ideal for detecting a higher number of fainter events observed off-axis or at a larger distance. Thus, Daksha will boost the joint detections of sGRB and GW, especially for the off-axis events. Finally, we find that our detection rates with optimal SNRs are conservative, and noise in GW detectors can increase the rates further.

Why thermal images are blurry.

The resolution of optical imaging is limited by diffraction as well as detector noise. However, thermal imaging exhibits an additional unique phenomenon of ghosting which results in blurry and low-texture images. Here, we provide a detailed view of thermal physics-driven texture and explain why it vanishes in thermal images capturing heat radiation. We show that spectral resolution in thermal imagery can help recover this texture, and we provide algorithms to recover texture close to the ground truth. We develop a simulator for complex 3D scenes and discuss the interplay of geometric textures and non-uniform temperatures which is common in real-world thermal imaging. We demonstrate the failure of traditional thermal imaging to recover ground truth in multiple scenarios while our thermal perception approach successfully recovers geometric textures. Finally, we put forth an experimentally feasible infrared Bayer-filter approach to achieve thermal perception in pitch darkness as vivid as optical imagery in broad daylight.

Community detection in decentralized social networks with local differential privacy

Community detection under local differential privacy has been a research hotspot recently. Most existing methods primarily rely on the edge local differential privacy model (edge-LDP), compromising privacy strength for high utility. In addition, they adopt a way of agglomerating nodes one by one to construct communities (or clusters) through multiple iterations, suffering from significant noise and low detection efficiency. Moreover, the error accumulation caused by large-scale iterations is not conducive to detection accuracy. To address these issues, we propose a privacy-preserving community detection method, CD-LDP, based on node-LDP, achieving efficient and accurate community construction by aggregating grouped units instead of individual nodes and delimiting the interaction to three rounds. Specifically, we propose a privacy-preserving global social network construction method to gain connectivity information among all nodes through two rounds of interactions. In addition, a novel perturbation mechanism based on the star social network structure is devised, which improves the accuracy of the global social network by reducing the probability of generating fake edges. Secondly, we design a privacy-preserving node grouping approach leveraging clustering coefficients and Haar discrete wavelet transform in the third round, achieving low noise injection and high-quality node grouping through lossless dimensionality reduction. Lastly, we develop a fast unit merging algorithm combining modularity and Huffman trees to realize efficient and accurate community building by reducing the aggregation possibility of nodes with low similarity. Theoretical analysis and experimentation on real-world datasets testify that our solution can efficiently extract high-quality community structures while satisfying node-LDP.

Characterization of external optical crosstalk reduction for SiPM-based scintillation detectors with an optical bandpass filter

External optical crosstalk remains a significant source of correlated noise in scintillation detectors optically coupled to silicon photomultiplier (SiPM) arrays and can influence achievable performance by limiting the maximum overvoltage at which a detector can be operated. This study evaluated the potential benefits of incorporating an optical bandpass filter, which is absorptive to optical crosstalk and transmissive for scintillation photons, between the scintillator and SiPM array. Several SiPM and bandpass configurations were characterized, including single-element and SiPM arrays with bare, filter-coupled, crystal-coupled, and crystal-filter-coupled interfaces. Coupling the optical filter between a 4 × 4 array of Broadcom AFBR-S4N44P164 M (NUV-MT) SiPMs and a 12 × 12 × 20 mm3 LYSO crystal decreased the total number of detected optical photon noise by up to 92%. The impact of the filter on crosstalk probability reduction was also characterized by the single-channel measurements with the same SiPM, suppressing external crosstalk up to 64%. With the filter coupled, SiPMs were operable at very high overvoltage. The results of these studies demonstrate that optical filtering is a promising technique to significantly reduce correlated noise in scintillation detectors employing SiPMs.

Signal detection and material identification method for loose particles inside aerospace relays based on overlapping signals

Loose particle is an important factor affecting the reliability of aerospace relays. Particle Impact Noise Detection method is a commonly used loose particle detection method, but it can generate interference signals. Accurately identifying loose particle signals and component signals becomes the key to accurately detecting loose particles. Meanwhile, deeply exploring the properties of loose particle signals and further provide material information can provide guidance for manufacturing processes. However, the existing loose particle signal and component signal identification research, as well as the loose particle material identification research, has problems such as limited research objects to pure signals, independent identification results, and failure to refer to the detection requirements in real application scenarios. It is difficult to apply and its practicality is not high. Based on this, the authors proposed a signal detection and material identification method for loose particles based on overlapping signals. Specifically, referring to the latest loose particle detection information, aerospace relay samples were made, and the component identification model and material identification model based on parameter-optimized random forest were trained, respectively. They can achieve good classification effects on data sets created from pure loose particle signals, pure component signals, and pure loose particle signals corresponding to different materials, respectively. On this basis, confidence coefficient was proposed to quantify the degradation degree of the classification effects of the two models on the data set created from overlapping signals, thus the component confidence coefficient and the loose particle confidence coefficient were obtained, respectively. They can be used to determine valid pulses from pure loose particle signals, pure component signals, and mixed signals in overlapping signals, completing loose particle detection. Valid pulses from pure loose particle signals were screened for material identification. In this way, for the aerospace relay to be tested, first, the loose particle detection results can be obtained by a comprehensive judgment of the component identification model, component confidence coefficient, and loose particle confidence coefficient. Second, the material identification results can be obtained by combining the material identification model with the majority voting processing. In addition, the definition of identification accuracy applicable to the loose particle detection was proposed to meet the engineering application requirements in real application scenarios. Multiple experiments verified the feasibility, practicality, and stability of the proposed method.

Design and implementation of a seismic Newtonian noise cancellation system for the Virgo gravitational-wave detector

Terrestrial gravity perturbations caused by seismic fields produce the so-called Newtonian noise in gravitational-wave detectors, which is predicted to limit their sensitivity in the upcoming observing runs. In the past, this noise was seen as an infrastructural limitation, i.e., something that cannot be overcome without major investments to improve a detector’s infrastructure. However, it is possible to have at least an indirect estimate of this noise by using the data from a large number of seismometers deployed around a detector’s suspended test masses. The noise estimate can be subtracted from the gravitational-wave data, a process called Newtonian noise cancellation (NNC). In this article, we present the design and implementation of the first NNC system at the Virgo detector as part of its AdV+ upgrade. It uses data from 110 vertical geophones deployed inside the Virgo buildings in optimized array configurations. We use a separate tiltmeter channel to test the pipeline in a proof-of-principle. The system has been running with good performance over months.

Extending loophole-free nonlocal correlations to arbitrarily large distances

Quantum theory allows spatially separated observers to share nonlocal correlations, which enable them to accomplish classically inconceivable information processing and cryptographic feats. However, the distances over which nonlocal correlations can be realized remain severely limited due to their high fragility to noise and high threshold detection efficiencies. To enable loophole-free nonlocality across large distances, we introduce Bell experiments wherein the spatially separated parties randomly choose the location of their measurement devices. We demonstrate that when devices close to the source are perfect and witness extremal nonlocal correlations, such correlations can be extended to devices placed arbitrarily far from the source. To accommodate imperfections close to the source, we demonstrate an analytic trade-off: the higher the loophole-free nonlocality close to the source, the lower the threshold requirements away from the source. We utilize this trade-off and formulate numerical methods to estimate the critical requirements of individual measurement devices in such experiments.

Numerical and experimental research on vibration signal characteristics of water pipeline leakage

In the leak detection of water pipelines, interference and environmental noise directly affect the effectiveness of leak detection and the accuracy of leak point location, so it is necessary to study the real signal. Previous studies mostly rely on a single test to improve the positioning accuracy by collecting vibration signals for denoising, but denoising will also affect the real signal, and the actual detection situation is complex, so it is difficult to deal with all working conditions only through the test. With the rapid development of computer technology, the combination of numerical simulation and experiment is expected to solve this problem. Therefore, this paper takes the acceleration signal as the research object, collects and simulates the leakage signals under different pressure strengths by combining experiments with numerical simulation, and verifies the feasibility and accuracy of a numerical simulation model. The generation of leakage vibration acceleration under different pressure and different leakage hole radius is simulated. Research findings: a. the leakage signal energy has a power function relationship with the pressure and an exponential function relationship with the leakage radius; b. the characteristic frequency is the variable of the characteristics of pipeline and leakage hole, which is independent of the pressure; c. the power of each peak frequency has a step-by-step change trend with the energy of the signal; d. the power of acceleration signal concentrates to a higher frequency with the increase of pressure and the decrease of leakage radius. This study provides a basis for identifying, detecting, and extracting leakage signals in a real noise environment, helps to improve the accuracy of acoustic emission signal detection, and can provide a theoretical basis for pipeline leakage detection research.

Application of cross-correlation-based transimpedance amplifier in InAs and InAsSb IR detectors noise measurements

Application of cross-correlation-based transimpedance amplifier in InAs and InAsSb IR detectors noise measurements

Variational Autoencoders for Biomedical Signal Morphology Clustering and Noise Detection.

Accurate estimation of physiological biomarkers using raw waveform data from non-invasive wearable devices requires extensive data preprocessing. An automatic noise detection method in time-series data would offer significant utility for various domains. As data labeling is onerous, having a minimally supervised abnormality detection method for input data, as well as an estimation of the severity of the signal corruptness, is essential. We propose a model-free, time-series biomedical waveform noise detection framework using a Variational Autoencoder coupled with Gaussian Mixture Models, which can detect a range of waveform abnormalities without annotation, providing a confidence metric for each segment. Our technique operates on biomedical signals that exhibit periodicity of heart activities. This framework can be applied to any machine learning or deep learning model as an initial signal validator component. Moreover, the confidence score generated by the proposed framework can be incorporated into different models' optimization to construct confidence-aware modeling. We conduct experiments using dynamic time warping (DTW) distance of segments to validated cardiac cycle morphology. The result confirms that our approach removes noisy cardiac cycles and the remaining signals, classified as clean, exhibit a 59.92% reduction in the standard deviation of DTW distances. Using a dataset of bio-impedance data of 97885 cardiac cycles, we further demonstrate a significant improvement in the downstream task of cuffless blood pressure estimation, with an average reduction of 2.67 mmHg root mean square error (RMSE) of Diastolic Blood pressure and 2.13 mmHg RMSE of systolic blood pressure, with increases of average Pearson correlation of 0.28 and 0.08, with a statistically significant improvement of signal-to-noise ratio respectively in the presence of different synthetic noise sources. This enables burden-free validation of wearable sensor data for downstream biomedical applications.

A simple and effective method for the removal of gamma white spots noise in fast neutron images

Fast neutron radiography technology has unique application advantages in the field of non-destructive testing. However, due to the interaction between the gamma rays and the charge-coupled device camera, a large amount of gamma white spots noise is attached to the fast neutron images, which affects the subsequent analysis. In this paper, we propose a simple and effective method to remove gamma white spots noise in fast neutron images. First, the image is down-sampled by the dilated down-sampling method to reduce the influence of noise. Then, the existence of gamma white spots noise is accurately identified by noise detection. Finally, the image is denoised using an improved median filtering method. In order to verify the performance of the proposed method, several commonly used noise reduction methods and the proposed method are used to conduct test experiments on both simulated and fast neutron images, and several image quality evaluation indices are used for quantitative assessment. The results show that the proposed method not only performs well in visual effects but also achieves satisfactory results in evaluation indices.

Isolation and Detection of Arc Fault Noise in a Real PV System using Current Demodulation and Autocorrelation Coefficients

Isolation and Detection of Arc Fault Noise in a Real PV System using Current Demodulation and Autocorrelation Coefficients

Feature extraction and pattern recognition of gas pipeline flow noise signals in a strong noisy background.

The purpose of this study is to put forward a feature extraction and pattern recognition method for the flow noise signal of natural gas pipelines in view of the complex situation brought by the rapid development and expansion of urban natural gas infrastructure in China, especially in the case that there are active and abandoned pipelines, metal and nonmetal pipelines, and natural gas, water and power pipelines coexist in the underground of the city. Because the underground situation is unknown, gas leakage incidents caused by natural gas pipeline rupture occur from time to time, posing a threat to personal safety. Therefore, the motivation of this study is to provide a feasible method to accelerate the aging, renewal and transformation of urban natural gas pipelines to ensure the safe operation of urban natural gas pipeline network and promote the high-quality development of urban economy. Through the combination of experimental test and numerical simulation, this study establishes a database of urban natural gas pipeline flow noise signals, and uses principal component analysis (PCA) to extract the characteristics of flow noise signals, and develops a mathematical model for feature extraction. Then, a classification and recognition model based on backpropagation neural network (BPNN) is constructed, which realizes the detection and recognition of convective noise signals. The research results show that the theoretical method based on acoustic feature analysis provides guidance for the orderly and safe construction of urban natural gas pipeline network and ensures its safe operation. The research conclusion shows that through the simulation analysis of 75 groups of gas pipeline flow noise under different working conditions. Combined with the experimental verification of ground flow noise signals, the feature extraction and pattern recognition method proposed in this study has a recognition accuracy of up to 97% under strong noise background, which confirms the accuracy of numerical simulation and provides theoretical basis and technical support for the detection and recognition of urban gas pipeline flow noise.