Influence Of Random Noise Research Articles

Effect of Q-learning on the evolution of cooperation behavior in collective motion: An improved Vicsek model

There have been numerous studies on collective behavior, among which communication between agents can have a great impact on both the payoff and the cost of making decisions. Research usually focuses on how to improve the collective synchronization rate or accelerate the process of cooperation under given communication cost constraints. In this context, evolutionary game theory (EGT) and reinforcement learning (RL) arise as essential frameworks for tackling this intricate problem. In this study, an adapted Vicsek model is introduced, wherein agents exhibit varying movement patterns contingent on their chosen strategies. Each agent gains a payoff determined by the advantages of collective motion juxtaposed with the cost of communicating with neighboring agents. Individuals choose the objective agents based on the Q-learning strategy and then adapt their strategies following the Fermi rule. The research reveals that the utmost level of cooperation and synchronization can be attained at an optimal communication radius after applying Q-learning. Similar conclusions have been drawn from research on the influence of random noise and relative cost. Different cost functions were considered in the study to demonstrate the robustness of the proposed model and conclusions under a wide range of conditions. (https://github.com/WangchengjieT/VM-EGT-Q)

Calculation Method and Application of Basic Regeneration Number for a Class of Stochastic Systems

Considering the influence of random noise on SIR, SEIR and SEIAR infectious disease models, we establish SIR, SEIR and SEIAR models with random disturbance, and deduce the calculation formula of the basic regeneration number of the random infectious disease model in the sense of mean value by using Itô formula. The effectiveness of the basic regeneration number calculation method is verified by numerical simulation of the system evolution process.

Random noise suppression and super-resolution reconstruction algorithm of seismic profile based on GAN

In this paper, we propose a random noise suppression and super-resolution reconstruction algorithm for seismic profiles based on Generative Adversarial Networks, in anticipation of reducing the influence of random noise and low resolution on seismic profiles. Firstly, the algorithm used the residual learning strategy to construct a de-noising subnet to accurate separate the interference noise on the basis of protecting the effective signal. Furthermore, it iterated the back-projection unit to complete the reconstruction of the high-resolution seismic sections image, while responsed sampling error to enhance the super-resolution performance of the algorithm. For seismic data characteristics, designed the discriminator to be a fully convolutional neural network, used a larger convolution kernels to extract data features and continuously strengthened the supervision of the generator performance optimization during the training process. The results on the synthetic data and the actual data indicated that the algorithm could improve the quality of seismic cross-section, make ideal signal-to-noise ratio and further improve the resolution of the reconstructed cross-sectional image. Besides, the observations of geological structures such as fractures were also clearer.

Simultaneous estimation of heat generation and magnetic field in a radial porous fin from surface temperature information

In this paper, for the first time an inverse methodology is demonstrated for simultaneously predicting the internal heat generation and magnetic field strength in a radial porous fin using the surface temperature response. The operation of the system is considered under an imposed magnetic field and all modes of heat transfer. Initially, validated direct solutions are acquired for calculating the temperature field, and thereafter the unknowns are estimated using an inverse method assisted by the Artificial Bee Colony (ABC) algorithm. Numerical case studies are done to find an appropriate relationship among the given unknowns. The present analysis highlights that while many possible combinations exist satisfying the given thermal profile, however, the magnetic field strength and heat generation always vary linearly for a given distribution of temperature. Even under the influence of random noise, the ABC assisted algorithm is found to accurately reconstruct the available condition and excellently establish the mutual relationship between the parameters with an accuracy within 2%. For the purpose of a required heat transfer from porous fins, the present methodology is concluded to be beneficial in accurately controlling the magnetic field against an unknown condition of internal heat generation.

Firefly Algorithm for Transient Electromagnetic Inversion

Transient electromagnetic (TEM) method is widely used in regional mineral resources surveys, environmental engineering geological surveys and shallow surface geophysical exploration, and so on. However, interpretation and inversion of TEM data is a complicated process. The traditional algorithm of TEM inversion employs the “smoke ring” fast imaging method, which can only reflect the approximate morphology of the stratigraphic model, and the inversion accuracy is low. Therefore, this method cannot meet the requirements of high-precision inversion. In this article, we present the firefly algorithm (FA) technology for TEM inversion. First of all, the response of the rectangular loop source TEM based on electric dipole integration was calculated and compared with the analytical solution results of the rectangular loop source and the accuracy of the algorithm was verified. Then, a layered medium model was established. The FA technology and “smoke ring” fast imaging method were used to perform inversion calculation, and the influence of random noise on the accuracy of the FA algorithm was analyzed. The results show that the FA has a high degree of fitting to the model, good anti-noise property, and fast search speed. Next, to illustrate the application effect of the FA algorithm in pseudo 2-D inversion, the 2-D model was established. The results show that the FA algorithm can reflect the distribution of the anomalous body more accurately, especially for the low-resistance anomalous body. Finally, we examined the effectiveness of the FA for TEM data processing by inverting survey data and comparing the results with those from the “smoke ring” fast imaging and particle swarm optimization (PSO) algorithms. The research works provide new methods and techniques for TEM data processing.

Random noise suppression of magnetic resonance sounding signal based on modified short-time Fourier transform

Magnetic resonance sounding (MRS) has the advantage of detecting groundwater content directly without drilling, but the signal-to-noise ratio (SNR) is extremely low which limits the application of the method. Most of the current researches focus on eliminating spikes and powerline harmonic noise in the MRS signal, whereas the influence of random noise cannot be ignored even though it is difficult to suppress due to the irregularity. The common method to eliminate MRS random noise is stacking which requires extensive measurement repetition at the cost of detection efficiency, and it is insufficient when employed in a high-level noise surrounding. To solve this problem, we propose a modified short-time Fourier transform(MSTFT) method, in which used is the short-time Fourier transform on the analytical signal instead of the real-valued signal to obtain the high-precision time-frequency distribution of MRS signal, followed by extracting the time-frequency domain peak amplitude and peak phase to reconstruct the signal and suppress the random noise. The performance of the proposed method is tested on synthetic envelope signals and field data. The using of the MSTFT method to handle a single recording can suppress the random noise and extract MRS signals when SNR is more than –17.21 dB. Compared with the stacking method, the MSTFT achieves an 27.88dB increase of SNR and more accurate parameter estimation. The findings of this study lay a good foundation for obtaining exact groundwater distribution by utilizing magnetic resonance sounding.

Modal decomposition of a fibre laser beam based on the push-broom stochastic parallel gradient descent algorithm

Compared with traditional methods, the stochastic parallel gradient descent (SPGD) algorithm used in modal decomposition has its own unique advantages of faster speed, wider application range, and low experimental effort, demonstrating many development prospects. However, the current SPGD algorithm only considers the modal weights of multimode optical fields and neglects the phase. More importantly, this algorithm is usually unsuitable for conducting modal decomposition of incoherent combined laser beams due to its serious ambiguities. In this paper, a novel modal decomposition algorithm, called the push-broom SPGD algorithm, is first proposed to achieve complete modal decomposition of optical fields emerging from multimode fibres. By iteratively evaluating the intensity profiles in different positions using the SPGD algorithm, this approach can obtain both amplitude and phase information with high accuracy. The proposed algorithm is preliminarily applied to the modal decomposition of an incoherent combined laser beam, and its feasibility is demonstrated via numerical simulations. By inserting a parallel plate, ambiguities can be solved and the influence of random noise is suppressed, demonstrating this algorithm’s potential.

A Modeling Method of Electromagnetic Micromirror in Random Noisy Environment

In this paper, a data-driven modeling method for electromagnetically actuated micromirror in random noisy environment is proposed. In this method, the electromagnetic micromirror is considered as a dynamic system with preceded hysteresis. Then, a linear dynamic submodel is used to describe the angular deflection mechanism, while a preceded rate-dependent hysteresis submodel is used to depict the hysteresis phenomenon inherent in the electromagnetic driver. By considering the influence of random noise on the micromirror, an on-line modeling scheme with varying weighting factors is studied to handle the data contaminated by random noises. Subsequently, the convergence of proposed modeling method is analyzed. Finally, the experimental results of the proposed modeling scheme for an electromagnetically actuated micromirror are presented.

Combined impacts of the Allee effect, delay and stochasticity: Persistence analysis

We study a combined influence of the Allee effect, delay and stochasticity on the base of the phenomenological Hassell mathematical model of population dynamics. This bistable dynamical model possesses a wide variety of regimes, both regular and chaotic. In the persistence zone, these regimes coexist with the trivial equilibrium that corresponds to the extinction of the population. It is shown that borders of the persistence zone are defined by the crisis and saddle-node bifurcation points. Noise-induced transitions from the persistence to the extinction are studied both numerically and analytically. Using numerical modeling, we have found that the persistence zone can decrease and even disappear under the influence of random noise. For the theoretical study of this phenomenon, we apply the stochastic sensitivity analysis and Mahalanobis metrics.

Hybrid Monte Carlo algorithm for sampling rare events in space-time histories of stochastic fields.

We introduce a variant of the Hybrid Monte Carlo (HMC) algorithm to address large-deviation statistics in stochastic hydrodynamics. Based on the path-integral approach to stochastic (partial) differential equations, our HMC algorithm samples space-time histories of the dynamical degrees of freedom under the influence of random noise. First, we validate and benchmark the HMC algorithm by reproducing multiscale properties of the one-dimensional Burgers equation driven by Gaussian and white-in-time noise. Second, we show how to implement an importance sampling protocol to significantly enhance, by orders of magnitudes, the probability to sample extreme and rare events, making it possible to estimate moments of field variables of extremely high order (up to 30 and more). By employing reweighting techniques, we map the biased configurations back to the original probability measure in order to probe their statistical importance. Finally, we show that by biasing the system towards very intense negative gradients, the HMC algorithm is able to explore the statistical fluctuations around instanton configurations. Our results will also be interesting and relevant in lattice gauge theory since they provide unique insights into reweighting techniques.

Metrological aspects of time-of-flight range imaging

Time-of-flight (ToF) range imaging cameras measure the distance from the camera to each position imaged by each respective pixel. This article summarizes the current state of ToF imaging range measurement. Covering the major forms of biasing error, we review the existing methods to correct them. These error correction methods comprise a mix of camera and data acquisition modifications, coupled with the appropriate signal processing. The pervasive influence of random noise is considered last, but not at all least, and jitter, or random timing error, is highlighted as a special cause of error in ToF imaging of great significance despite the limited attention given to it.

LiftingNet: A Novel Deep Learning Network With Layerwise Feature Learning From Noisy Mechanical Data for Fault Classification

The key challenge of intelligent fault diagnosis is to develop features that can distinguish different categories. Because of the unique properties of mechanical data, predetermined features based on prior knowledge are usually used as inputs for fault classification. However, proper selection of features often requires expertise knowledge and becomes more difficult and time consuming when volume of data increases. In this paper, a novel deep learning network (LiftingNet) is proposed to learn features adaptively from raw mechanical data without prior knowledge. Inspired by convolutional neural network and second generation wavelet transform, the LiftingNet is constructed to classify mechanical data even though inputs contain considerable noise and randomness. The LiftingNet consists of split layer, predict layer, update layer, pooling layer, and full-connection layer. Different kernel sizes are allowed in convolutional layers to improve learning ability. As a multilayer neural network, deep features are learned from shallow ones to represent complex structures in raw data. Feasibility and effectiveness of the LiftingNet is validated by two motor bearing datasets. Results show that the proposed method could achieve layerwise feature learning and successfully classify mechanical data even with different rotating speed and under the influence of random noise.

An Objective Method for Determining Ocean Mixed Layer Depth with Applications to WOCE Data

AbstractA new method is developed to identify the mixed layer depth (MLD) from individual temperature or density profiles. A relative variance profile is obtained that is the ratio between the standard deviation and the maximum variation of the temperature (density) from the sea surface, and the depth of the minimum relative variance is defined as the MLD. The new method is robust in finding the MLD under the influence of random noise (noise level ≤ 5%). A comparison with other available methods, which include the threshold (difference, difference interpolation, gradient, and hybrid methods) and objective (curvature and maximum angle methods) methods, is carried out using the World Ocean Circulation Experiment (WOCE) data. It is found that for a variety of depth sampling resolutions ranging from 0.04 to 25 dbar, the new method and the difference-interpolation method predict MLD values that are closer to the visually inspected ones than those by other methods. Moreover, the quality index (QI) of the MLD that is determined by the new method is the highest when compared with those of the available methods. Also, the application of the new method on the WOCE global dataset yields 94% of MLD values with , substantially higher than those (≤86%) of other methods. Ultimately, it is found that the new method determines very similar MLD values when applied to temperature or density profiles globally because it identifies the base of the mixed layer rather than the uppermost depth of the thermocline. This unique advantage makes the new method applicable in many cases, especially when the density profile is unavailable.

Analysis of the stochastic motion of a charged particle in a magnetic field by the Monte Carlo method on supercomputers

We study the issues of the influence of random noise on the motion of a charged particle in a magnetic field, using the statistical modeling for solving the arising system of stochastic differential equations. We present the results of the numerical experiments carried out on the NKS-30T cluster of the Siberian Supercomputer Center at the Institute of Computational Mathematics and Mathematical Geophysics of the Siberian Division of the Russian Academy of Sciences. To analyze the numerical solutions, we use the frequency characteristics that generalize the integral curve and the phase portrait.

Influence of the mechanics of escape on the instability of von Mises truss and its control

The elastic von Mises truss model is a prototype for bi-stable structures. It allows a deep understanding of the static and dynamic buckling of several planar and spatial truss systems and shallow lattice shell structures, including the geodesic dome, and has a theoretical and practical interest. This structure has a highly nonlinear response in the presence of static and dynamic loads. The geometric nonlinearity is particularly significant even at low load levels when this structure is shallow. This paper presents an exact nonlinear formulation, which is used to investigate the mechanics of erosion and escape from the safe pre-buckling well. As the static pre-load increases, the probability of escape increases in a nonlinear manner. Permanent and transient escapes, as well as the influence of random noise, on the dynamic buckling load are investigated. To increase the load carrying capacity, a method for controlling the global nonlinear dynamics for the elastic von Mises truss is employed. The control method consists of the (optimal) elimination of homoclinic intersection by properly adding superharmonic terms to a given harmonic excitation. Permanent and transient basins of attraction are obtained. The results highlight the complex nonlinear dynamics of this class of structures and the effectiveness of the control method in increasing the integrity of the basins of attraction of the system and its practical safety.

Analysis of the influence of random noise on self-oscillating chemical reactions by the Monte-Carlo method on supercomputers

Under study is the influence of external and internal random noises on the behavior of concentrations of the components of chemical reactions. We investigate the use of statistical modeling for solving the relevant stochastic differential equations (SDEs). Some results of numerical experiments are presented. To analyze numerical solutions, we use the frequency characteristics that generalize the integral curve and the phase portrait. Numerical experiments show that the random noise even with low intensity causes all kinds of transitions from one oscillation mode to another in the solutions of SDEs.

Real-time non-contact optical detection system for roller deformation based on artificial neural network

A non-contact real-time optical fiber measurement system for roller deformation is designed based on the principle of reflective displacement fiber-optic sensor. The theory and mathematical model of the sensor are deduced in detail. By using a novel probe of three optical fibers with equal transverse space, the effects of fluctuations in light source, reflective changing of target surface and the light intensity losses in fiber lines are automatically compensated. Meantime, an optical fiber sensor model of linearizing output signal and correcting static error based on artificial neural network (ANN) is set up. By using interpolation method and value filtering to process the signals based on ANN, the influence of random noise and the vibration of roller bearing can be reduced effectively. Therefore the linearity, accuracy and resolution are enhanced remarkably. Experimental results proved the accuracy of the system could reach to the demand of practical production process. It provides a new method for high speed, accurate and real-time detection of roller deformation.

An improved OFDM chirp waveform used for MIMO SAR system

Special attention has been devoted to multi-input multi-output (MIMO) synthetic aperture radar (SAR) systems in recent years. The applications of MIMO SAR systems which involve 3D imaging, highresolution wide-swath remote sensing, and multi-baseline interferometry are seriously limited to the orthogonal waveforms. Although orthogonal frequency division multiplexing (OFDM) chirp waveforms can be used for MIMO SAR systems to avoid intra-pulse interferences, there is a small frequency shift between the transmitted OFDM pulses. This vulnerable shift, which can not only affect the waveform orthogonality, but also introduce residual phase error, renders the OFDM waveforms impractical. In this paper, an improved OFDM chirp waveform which works without the mentioned shift is presented, along with the novel modulation and efficient demodulation procedures. Comparison between the improved and the conventional OFDM chirp waveforms is detailed. The influence of random noise, quantization error, and Doppler shift on the orthogonality of OFDM waveform is also investigated in this paper. Theoretical analysis and simulation results illustrate the feasibility of this waveform scheme.

Random noise reduction capability of the hermit polynomial based Least Squares Fourier Transform method

Treating the Fourier transform as an over-determined inverse problem is a new conception for determining the frequency spectrum of a signal. The concept enables us to implement several algorithms depending on the applied inversion tool. One of these algorithms is the Hermit polynomial based Least Squares Fourier Transform (H-LSQ-FT). The H-LSQ-FT is suitable for reducing the influence of random noise. The aim of the investigation presented in the paper was to study the noise reduction capability of the H-LSQ-FT in some circumstances. Four wavelet-like signals with different properties were selected for testing the method. Examinations were completed on noiseless and noisy signals. The H-LSQ-FT provided the best noise reduction for the noisy signal having low peak frequency and wide band width. Finally, the results obtained by the H-LSQ-FT were compared to those of other traditional methods. It is showed that the H-LSQ-FT yields better noise filtering than these methods do.

A time-series modeling method based on the boosting gradient-descent theory

The forecasting of time-series data plays an important role in various domains. It is of significance in theory and application to improve prediction accuracy of the time-series data. With the progress in the study of time-series, time-series forecasting model becomes more complicated, and consequently great concern has been drawn to the techniques in designing the forecasting model. A modeling method which is easy to use by engineers and may generate good results is in urgent need. In this paper, a gradient-boost AR ensemble learning algorithm (AREL) is put forward. The effectiveness of AREL is assessed by theoretical analyses, and it is demonstrated that this method can build a strong predictive model by assembling a set of AR models. In order to avoid fitting exactly any single training example, an insensitive loss function is introduced in the AREL algorithm, and accordingly the influence of random noise is reduced. To further enhance the capability of AREL algorithm for non-stationary time-series, improve the robustness of algorithm, discourage overfitting, and reduce sensitivity of algorithm to parameter settings, a weighted kNN prediction method based on AREL algorithm is presented. The results of numerical testing on real data demonstrate that the proposed modeling method and prediction method are effective.