Signal Noise Reduction Research Articles

Fault diagnosis based on extremely randomized trees in wireless sensor networks

Wireless Sensor Network (WSN) being highly diversified cyber–physical system makes it vulnerable to numerous failures, which can cause devastation towards safety, economy, and systems’ reliability. Precise detection and diagnosis of failures or faults in WSN is a challenging issue due to the diversity of deployment and the limitations in the sensors’ resources. In this paper, supervised machine learning-based technique is considered to scrutinize the behavior of sensors through their data for the detection and diagnosis of faults. Most of the faults that commonly occur in WSN are considered: hardover, drift, spike, erratic, data-loss, stuck, and random fault. A trusted dataset published online by the researchers at the University of North Carolina composed of temperature and humidity sensor measurements of multi-hop scenario was acquired and the aforementioned faults were simulated in non-faulty (normal) data. Events from fault occurrences were generated to replicate realistic scenarios of WSN. To detect and diagnose the faults in timely manner, we adopt an ensemble learning-based lightweight technique called Extremely Randomized Trees or Extra-Trees. The proposed Extra-Trees-based detection scheme has the ability of robustness towards signal noise and strong reduction of bias and variance error. The performances of the proposed scheme were compared with those of the state-of-the-art machine learning algorithms such as support vector machine, random forest, neural network, and decision tree. Performance evaluation shows the efficiency of the proposed scheme in terms of accuracy, precision, and F1-score. In addition, the proposed scheme has low training time compared to state-of-the-art approaches.

Hybrid deep learning for power generation forecasting in active solar trackers

To meet the growing electricity demand for consumers, it is necessary to use more efficient systems. The solar trackers stand out among the applications that can improve the efficiency of photovoltaic panel generation by increasing their solar uptake. For solar trackers to be more efficient, they can base their position update on a generation forecast and thus perform the control action only when there is greater efficiency in this update. For generation forecast, the long–short-term memory (LSTM) can handle a large volume of non-linear data. Furthermore, to improve the analysis, it is possible to apply signal filtering techniques. The wavelet energy coefficient is a technique used to reduce signal noise and extract features; this technique performs the filter and preserves the signal characteristic. In this study, the authors present a combination of wavelet energy coefficient and LSTM, defined as wavelet LSTM, to perform photovoltaic power forecasting in the dual-axis solar trackers.

Research on a Noise Reduction Method Based on DTCWT and the Cyclic Singular Energy Difference Spectrum

The gear is the most important part of the transmission system of mechanical equipment, and the monitoring and diagnosis of it can improve the reliability of mechanical equipment. However, mechanical equipment generally works in harsh working conditions. The gear vibration signal is subjected to strong noise interference in working conditions, which brings great challenges for the effective diagnosis of gear faults. This paper proposed a noise reduction method based on the dual-tree complex wavelet transform (DTCWT) and cyclic singular energy difference spectrum. First, the gear vibration signal containing strong noise interference is decomposed into a series of signal components with different frequency characteristics by using the time-frequency analysis ability of DTCWT. Then, cyclic singular energy difference spectrum is proposed based on the idea of a cascaded cycle and the successive elimination of noise interference to process each signal component with different frequency characteristics, and the termination conditions of cyclic singular energy difference spectrum can be set according to the noise interference distribution characteristics in different frequency bands. The final noise reduction of the original gear vibration signal can be realized based on signal reconstruction after the noise reduction processing of each signal components with different frequency bands. Finally, experiments are carried out to verify the effectiveness of the proposed method, which is effective and suitable for the noise reduction of the vibration signal.

Analysis of balanced and unbalanced signal transmission lines and noise reduction technique for beam-current monitor in linear accelerators.

A signal transmission line and its characteristics were analyzed in detail for a beam-current monitor (CM) for long-pulsed electron beams. The monitor was based on a conventional current transformer, which comprised an induction coil wound around a ferrite core. The signal transmission and noise generation mechanism were analyzed by theoretical models based on balanced and unbalanced coaxial transmission circuits to effectively reduce noises superimposed on the beam-current signal waveform. The results of numerical analyses based on the theoretical models were in good agreement with those of an experiment for verification. It was found that the noise signals were generated by a high-power klystron operation and they were mode-transformed on the way to the transmission line. These results show that the locations of the grounded points in the transmission line were important. Based on a symmetrically configured three-line circuit theory, a new highly symmetrically configured coaxial transmission line is proposed to effectively suppress noises generated in CM. This report describes in detail the numerical and experimental analysis of a signal transmission line with coaxial cables for CM where the noise generation mechanism and its reduction technique are deeply involved.

Analysis of electromagnetic radiation spectrum during the explosion of energetic materials

In order to provide effective reference data for the equivalent experiment of electromagnetic environment adaptability of electric explosive devices, the electromagnetic radiation generated during the explosion of energetic materials is measured. The spectral characteristics of the electromagnetic radiation during the explosion of energetic materials are obtained by processing the data by means of signal noise reduction, windowed Fourier transform and dynamic spectrum analysis. The results indicate that the spectrum of electromagnetic signal during the explosion of energetic materials shows large changes mainly in the frequency band of 0-100MHz, of which the energy change in the low frequency band below 50 MHz is the most obvious; the distance from the explosion center has an obvious influence on the spectrum distribution of electromagnetic signal, where the closer the test point is to the explosion center, the wider the spectrum distribution of electromagnetic signal is and the stronger the energy is; differences in electromagnetic frequency distribution in different directions reflect the uneven and dispersed propagation of electromagnetic radiation.

Use of trainable wavelet transform with adaptive threshold filtering for noise reduction of speech signals

A method of noise reduction based on an adaptive threshold filtering over trainable wavelet transform was proposed. The restrictions on the trainable transform filter parameters were provided by a set of quadratic regularization terms. As an analog of the “hard” and “soft” threshold functions we used their smooth infinitely differentiable versions. The parameters of trainable wavelet transform, and threshold values were estimated by backpropagation and gradient-based optimization algorithm with an adaptive momentum estimation. Results of the proposed method were compared with an approach based on fixed discrete wavelet transform and non-adaptive global and level-depended threshold algorithms on the model problem. We used signal-to-noise ratio between suppressed and clean signals to numerically estimate the efficiency of noise reduction. We showed that the best results were obtained when the proposed trainable method with Daubechies 4 wavelet filters fine-tuning and adaptive level-dependent thresholding were applied.

A New Hybrid Model for Underwater Acoustic Signal Prediction

The prediction of underwater acoustic signal is the basis of underwater acoustic signal processing, which can be applied to underwater target signal noise reduction, detection, and feature extraction. Therefore, it is of great significance to improve the prediction accuracy of underwater acoustic signal. Aiming at the difficulty in underwater acoustic signal sequence prediction, a new hybrid prediction model for underwater acoustic signal is proposed in this paper, which combines the advantages of variational mode decomposition (VMD), artificial intelligence method, and optimization algorithm. In order to reduce the complexity of underwater acoustic signal sequence and improve operation efficiency, the original signal is decomposed by VMD into intrinsic mode components (IMFs) according to the characteristics of the signal, and dispersion entropy (DE) is used to analyze the complexity of IMF. The subsequences (VMD-DE) are obtained by adding the IMF with similar complexity. Then, extreme learning machine (ELM) is used to predict the low-frequency subsequence obtained by VMD-DE. Support vector regression (SVR) is used to predict the high-frequency subsequence. In addition, an artificial bee colony (ABC) algorithm is used to optimize model performance by adjusting the parameters of SVR. The experimental results show that the proposed new hybrid model can provide enhanced accuracy with the reduction of prediction error compared with other existing prediction methods.

Noise Reduction in ECG Signal Using Combined Ensemble Empirical Mode Decomposition Method with Stationary Wavelet Transform

The diagnostic study of electrocardiography (ECG) signals plays a vital role in the diagnosis of cardiac problems. But the powerline interference in ECG causes an artifact in the interpretation of the original signal. In this paper, a new method for the removal of such noise/artifact from the ECG signal by combining stationary wavelet transform with empirical mode decomposition (EMD-SWT) and ensemble empirical mode decomposition (EEMD-SWT) is proposed. SWT is applied after the decomposition of ECG signals into various intrinsic mode functions (IMFs) for further removal of noise. The proposed methods are tested for various datasets available in MIT-BIH Arrhythmia database, and the performance has been validated with existing methods. From the simulated results, it is found that combining SWT with EMD and EEMD yields better SNR enhancement when compared to the traditional methods.

An Efficient Hybrid Technique for Noise Reduction in Optical Gyroscope Signals

Gyroscopes are sensors that are used for motion measurement. They are generally used to measure rotation rate of moving equipment. There are different types of gyroscopes including mechanical, micro-electro-mechanical (MEMS) and optical gyroscopes. Gyroscope signal suffers from internal noise due to internal device operation and external noise of the environment. This paper presents a proposed hybrid technique that includes both Kalman filter and wavelet denoising. Results show the superiority of this proposed technique to the other filters. Arranging the filters in cascaded hybrid structure has an effect on the performance of the hybrid technique. Using Kalman filter as a first stage is better than using the wavelet as a first stage. For the comparison, two evaluation metrics are used: Signal-to-Noise Ratio (SNR) improvement and correlation coefficient.

SIGNAL NOISE REDUCTION AND FILTERING

The paper deals with noise reduction in signal. Normally measured signal very frequently includes noise and data processing includes the activities for its reduction. The best choice is to reduce the source of noise, but often it is not possible to reduce noise source. Filtering is another activity, which helps us to reduce noise in measured signal. Data processing can be executed only with filtered signal.

Application of local mean decomposition and singular value decomposition in noise reduction of GNSS station coordinate time series signal

Application of local mean decomposition and singular value decomposition in noise reduction of GNSS station coordinate time series signal

Comparison of wavelet and correlation indices of cerebral autoregulation in a pediatric swine model of cardiac arrest

Existing cerebrovascular blood pressure autoregulation metrics have not been translated to clinical care for pediatric cardiac arrest, in part because signal noise causes high index time-variability. We tested whether a wavelet method that uses near-infrared spectroscopy (NIRS) or intracranial pressure (ICP) decreases index variability compared to that of commonly used correlation indices. We also compared whether the methods identify the optimal arterial blood pressure (ABPopt) and lower limit of autoregulation (LLA). 68 piglets were randomized to cardiac arrest or sham procedure with continuous monitoring of cerebral blood flow using laser Doppler, NIRS and ICP. The arterial blood pressure (ABP) was gradually reduced until it dropped to below the LLA. Several autoregulation indices were calculated using correlation and wavelet methods, including the pressure reactivity index (PRx and wPRx), cerebral oximetry index (COx and wCOx), and hemoglobin volume index (HVx and wHVx). Wavelet methodology had less index variability with smaller standard deviations. Both wavelet and correlation methods distinguished functional autoregulation (ABP above LLA) from dysfunctional autoregulation (ABP below the LLA). Both wavelet and correlation methods also identified ABPopt with high agreement. Thus, wavelet methodology using NIRS may offer an accurate vasoreactivity monitoring method with reduced signal noise after pediatric cardiac arrest.

A review of underwater acoustic metamaterials

Underwater acoustic materials are the key to the noise and vibration reduction, stealth, reception and amplification of underwater acoustic signal. However, conventional underwater acoustic materials usually suffer from the weakness of great bulk, heavy weight and poor resistance to hydrodynamic pressure. The advancement of the acoustic metamaterials (AMMs) has provided new insights into fixing these problems, and a variety of underwater AMMs are proposed accordingly. This article reviews the history and the developments of underwater AMMs and the difficulties that existed in the developments of underwater AMMs as compared with the AMMs in air. The focuses are the three types of AMMs which are closely associated with the underwater acoustic stealth and signal reception & amplification of hull-mounted sonar, i.e., the AMMs for sound absorption, AMMs for decoupling and those for sound focusing. The classification, physical mechanisms and controlling effects of those three types of underwater AMMs are reviewed in details. The AMMs for underwater sound absorption can be divided into local-resonance (LR) and non-resonance types. The LR types take advantages of the local resonance principle with the controlling wavelengths orders of magnitude greater than the physical dimension of a unit. To overcome the intrinsic narrow-band limitations of LR and broaden the bandwidths, efforts have been made to utilize multilayer LR-AMMs or multi-size LRs, attach/embed elastic plate scatter on/into elastically coated plate, or introduce complex lattices with multiple (coupled) LRs, etc. The non-resonance types are based-on non-resonance mechanisms such as viscous loss, heat conduction, macromolecular relaxation or a combination of them. These AMMs mainly include porous metals and nonmetal materials, gradient AMMs and acoustic meta-surfaces, etc. The AMMs for underwater decoupling mainly use acoustically soft AMMs to decouple the vibration of the base structure from adjacent fluid. These AMMs contain the acoustic coating embedded with various cavities or local resonators, gradient coating, re-entrant honeycomb structure or structure with negative Poisson’s ratio and chiral porous coating, etc. The sound-focusing AMMs are categorized into diffractive, refractive and reflective types. The diffractive types focus underwater sound through constructive interference of diffracted waves, which mainly include all kinds of zone plates and fractal lens. The refractive types are mainly based on various acoustic lens and focus sound by controlling the refraction index of the materials. The reflective types focus sound by modulating the reflective waves, including the recently-proposed reflective acoustic meta-surfaces and acoustic coatings with attached/embedded periodically distributed signal conditioning plates, etc. Finally, the conclusions and drawbacks of the above AMMs are summarized. It is pointed out that although the present underwater AMMs have achieved some encouraging results, they are still confronted with several key problems such as the realizations of low-frequency and broadband, omnidirectional, subwavelength, lightweight and hydrostatic pressure resistant sound manipulations, the manufacture and testing of large AMM samples, and the environmental adaptations requirements. Therefore, there is still a long way for the current underwater AMM toward the real application. The trend and outlook toward future underwater AMMs are also presented. It is hoped that this review could provide guidelines to the design and realization of future materials demanded for underwater acoustic stealth and detection.

Particle Swarm Optimization Algorithm with Mutation Operator for Particle Filter Noise Reduction in Mechanical Fault Diagnosis

In this paper, a new particle swarm optimization particle filter (NPSO-PF) algorithm is proposed, which is called particle cluster optimization particle filter algorithm with mutation operator, and is used for real-time filtering and noise reduction of nonlinear vibration signals. Because of its introduction of mutation operator, this algorithm overcomes the problem where by particle swarm optimization (PSO) algorithm easily falls into local optimal value, with a low calculation accuracy. At the same time, the distribution and diversity of particles in the sampling process are improved through the mutation operation. The defect of particle filter (PF) algorithm where the particles are poor and the utilization rate is not high is also solved. The mutation control function makes the particle set optimization process happen in the early and late stages, and improves the convergence speed of the particle set, which greatly reduces the running time of the whole algorithm. Simulation experiments show that compared with PF and PSO-PF algorithms, the proposed NPSO-PF algorithm has lower root mean square error, shorter running time, higher signal-to-noise ratio and more stable filtering performance. It is proved that the algorithm is suitable for real-time filtering and noise reduction processing of nonlinear signals.

Research on Noise Reduction of Φ-OTDR Signal Based on Blind Source Separation Algorithm

Optical fiber sensing has become an important means of health and safety monitoring of large buildings and facilities due to its better sensing characteristics. Undesirable noise signals are inevitably generated during actual sensor data acquisition. This paper focuses on the noise reduction processing of optical fiber Φ-OTDR vibration signal. Firstly, the acquired signal is preprocessed with data, then the data is independently correlated, and finally the FastICA algorithm in blind source separation is used to filter the noise signal. In this paper, the traditional wavelet transform algorithm is compared to reduce noise. The experimental results show that the blind source separation algorithm has better separation effect on fiber vibration signals.

One-dimensional convolutional auto-encoder-based feature learning for fault diagnosis of multivariate processes

Noise and high-dimension of process signals decrease effectiveness of those regular fault detection and diagnosis models in multivariate processes. Deep learning technique shows very excellent performance in high-level feature learning from image and visual data. However, the large labeled data are required for deep neural networks (DNNs) with supervised learning like convolutional neural network (CNN), which increases the time cost of model construction significantly. A new DNN model, one-dimensional convolutional auto-encoder (1D-CAE) is proposed for fault detection and diagnosis of multivariate processes in this paper. 1D-CAE is utilized to learn hierarchical feature representations through noise reduction of high-dimensional process signals. Auto-encoder integrated with convolutional kernels and pooling units allows feature extraction to be particularly effective, which is of great importance for fault detection and diagnosis in multivariate processes. The comparison between 1D-CAE and other typical DNNs illustrates effectiveness of 1D-CAE for fault detection and diagnosis on Tennessee Eastman Process and Fed-batch fermentation penicillin process. The proposed method provides an effective platform for deep-learning-based process fault detection and diagnosis of multivariate processes.

Noise Reduction of Welding Crack AE Signal Based on EMD and Wavelet Packet.

The acoustic emission (AE) signal collected by a sensor in the welding process has an overlapping frequency band and weak characteristics under a complex noise background. It is difficult for the wavelet noise reduction method, with single basis function, to effectively match the different characteristic information of the welding crack AE signal. Taking into account the adaptive decomposition characteristics of Empirical Mode Decomposition (EMD), a novel wavelet packet noise reduction method for welding AE signal was proposed. The welding AE signal was adaptively decomposed into several Intrinsic Mode Functions (IMFs) by the EMD. The effective IMFs were selected by the frequency distribution characteristics of the welding crack AE signal. A wavelet packet, with a specific basis function, was subsequently performed on the effective IMFs, which were reconstructed to be the welding crack AE signal. The simulated and experimental results indicated that the proposed method can effectively achieve noise reduction of the welding crack AE signal, which provided a mean for structure crack detection in the welding process.

Preparation of single- and double-oligonucleotide antibody conjugates and their application for protein analytics

Oligonucleotide-conjugated antibodies have gained importance for their use in protein diagnostics. The possibility to transfer the readout signal from the protein to the DNA level with an oligonucleotide-conjugated antibody increased the sensitivity of protein assays by orders of magnitude and enabled new multiplexing strategies. A bottleneck in the generation of larger oligonucleotide-conjugated antibody panels is the low conjugation yield between antibodies and oligonucleotides, as well as the lack of product purification methods. In this study, we combined a non-site-directed antibody conjugation technique using copper-free click chemistry with ion-exchange chromatography to obtain purified single and double oligonucleotide-conjugated antibodies. We optimized the click conjugation reaction of antibodies with oligonucleotides by evaluating crosslinker, reaction temperature, duration, oligonucleotide length, and secondary structure. As a result, we were able to achieve conjugation yields of 30% at a starting quantity as low as tens of nanograms of antibody, which makes the approach applicable for a wide variety of protein analytical assays. In contrast to previous non-site-directed conjugation methods, we also optimized the conjugation reaction for antibody specificity, confirmed by testing with knockout cell lines. The advantages of using single or double oligonucleotide-conjugated antibodies in regards to signal noise reduction are shown within immunofluorescence, proximity ligation assays, and single cell CITE-seq experiments.

Adaptive Noise Reduction Method of Synchronous Hydraulic Motor Acoustic Signal Based on Improved Dislocation Superposition Method

Acoustic signal contains information on the state of mechanical motion, but the disadvantage of low signal-to-noise ratio (SNR) is that acoustic signal without noise reduction is difficult to apply to fault diagnosis directly. This study proposes an adaptive noise reduction method based on the dislocation superposition method (DSM), which can realize automatic noise reduction for acoustic signals of low SNR synchronous hydraulic motors. First, the theoretical rotation period of the synchronous hydraulic motor is obtained according to the flow meter. Then, the actual signal is subjected to DSM processing according to the measurement accuracy of the flow meter and the theoretical rotation period to adjust the exact superposition length and the initial position of the actual signal. Finally, when the stop condition is not satisfied, the number of superposition is increased and the above process is repeated. The superimposed signal satisfying the stop condition is taken as the noise reduction signal. According to experimental results, the proposed method has good noise reduction effect on the acoustic signals of the synchronous hydraulic motor health state, wear out state of gear and end cover, and rust state of the gear. The noise reduction signal has been verified to have higher accuracy than the actual signal. Therefore, the proposed automatic noise reduction method can be applied to the noise reduction processing of other kinds of rotating mechanical acoustic signals.

Decision tree‐based method for optimum decomposition level determination in wavelet transform for noise reduction of partial discharge signals

Partial discharge (PD) monitoring in high-voltage equipment is one of the effective methods for assessment of its insulation strength. To do this, noise reduction is one of the essential processes on measured PD signals. One of the most popular tools employed for PD de-noising is wavelet transform. To exploit this transformation, three main parameters, including ‘mother wavelet’, ‘number of decomposition level’, and ‘thresholding procedure’, should be assigned. In this study, a novel and also more accurate method for the determination of required decomposition level is suggested. The proposed method employs a decision tree which takes the pattern of energy spectral density of PD signals in the frequency domain and delivers the optimum decomposition level for de-noising by wavelet transform. The results, as compared with others, show the superiority of the proposed method in noise reduction of PD signals, both for simulations and field measured signals.