Stationary Random Process Research Articles

Wide-sense stationary discrete-time random processes: Analyses in the discrete-Fresnel and discrete-frequency domains

This article investigates the expectation and autocorrelation functions of random processes in the discrete-Fresnel and the discrete-frequency domains, originating from wide-sense stationary (WSS) random processes in the discrete-time domain. Through mathematical deduction, it is demonstrated that WSS random processes in the discrete-time domain exhibit the same expectation and autocorrelation functions as transformed to the discrete-Fresnel domain. In contrast, this property is not observed in the discrete-frequency domain. White, pink, and brown random processes are considered to illustrate these behaviors in the discrete-Fresnel and discrete-frequency domains. Furthermore, this study numerically compares data communication systems based on single-carrier with cyclic prefix, orthogonal chirp-division multiplexing, and orthogonal frequency-division multiplexing schemes, which are respectively designed in the discrete-time, discrete-Fresnel, and discrete-frequency domains. The numerical results show that orthogonal chirp-division multiplexing and single-carrier with cyclic prefix schemes achieve comparable communication performance under any additive noise modeled by a wide-sense stationary random process, regardless of the chosen linear equalization or the absence of it.

Stochastic analysis of combined primary-secondary structures subjected to imprecise seismic excitation

The imprecise stochastic process model is able to incorporate both random and epistemic uncertainties, leading to a more accurate description of environmental excitations, such as earthquakes or wind loads, when limited data are available. Epistemic uncertainties in the loading model may significantly affect the performance of structural systems.The present paper addresses the stochastic analysis of combined primary-secondary structures subjected to imprecise seismic excitation modeled as a zero-mean stationary Gaussian random process, characterized by an interval-valued Power Spectral Density (PSD) function. The power and energy content of the imprecise ground motion acceleration may vary, even for the same soil category, affecting the complex dynamic behavior of combined structures and the vibration control capacity of secondary substructures under different frequency tuning conditions. The main purpose of this study is to develop a framework for investigating both the influence of epistemic uncertainties in the loading model and the interaction effects between the subsystems on the seismic performance of combined primary-secondary structures. To this aim, the stochastic analysis of combined structures under imprecise ground motion acceleration is addressed in both the frequency and time domains by an efficient approach capable of decoupling the propagation of interval and random uncertainties. Seismic safety assessment is performed in the framework of the first-passage theory by interval extension.

Determination of the characteristics of drill string vibrations during the drilling process

The object of research is vibration processes of a certain origin in the drill string with typical design deviations depending on the mode parameters of drilling. A drill string is an oscillating system with an infinite number of degrees of freedom of a multifactor system. An exhaustive study of oscillatory processes in the drill string is impossible neither analytically nor experimentally, due to the specifics of the hole deepening in various rocks, the design of the well, its shape, etc. Therefore, in practice, they try to solve the problems of the dynamics of the drill string for an idealized system and, while preserving the main oscillatory properties, solve some problems of the rod system. The work carried out was aimed at experimental studies of vibrations of the drill string during the drilling process. It is shown that the effectiveness of the use of hydrodynamic cavitation requires the development of methods and devices for intensifying the well drilling process. It is proven that the design of the cavitation generator organically fits into the existing well drilling equipment and allows for the intensification of technological processes with lower specific energy consumption. It is found that all oscillatory processes that occur in the drill string are random in nature and must be considered using the mathematical apparatus of the theory of random oscillations. The study of vibrations during well drilling shows that vibrations can be considered as random stationary processes, since transient modes have a sufficiently short duration for homogeneous rocks with fixed drilling modes. The analysis of the vibrations of the drill string elements based on random oscillations in a number of cases allows to increase the reliability of determining the vibration reliability of the drill string elements. It has been proven that the response of drill string elements to broadband random vibration can be defined as the combined effect of several narrowband random vibrations.

An Approximate Algorithm for Simulating Stationary Discrete Random Processes with Bivariate Distributions of Their Consecutive Components in the Form of Mixtures of Gaussian Distributions

An Approximate Algorithm for Simulating Stationary Discrete Random Processes with Bivariate Distributions of Their Consecutive Components in the Form of Mixtures of Gaussian Distributions

Randomized limit theorems for stationary ergodic random processes and fields

Using the randomization approach, introduced by A. Tempelman in Randomized multivariate central limit theorems for ergodic homogeneous random fields, Stochastic Processes and their Applications. 143 (2022), 89–105, we prove: (a) a randomized version of the invariance principle (the functional CLT); (b) a version the Glivenko–Cantelli theorem; (c) a randomized theorem about convergence of empirical processes to the Brownian bridge. We also weaken the moment condition in the randomized CLTs, proved in the mentioned article. The main point of our work is that most of our theorems are valid for all ergodic homogeneous random fields on Zm and Rm,m≥1.

Research of stochastic stability of constructions parametric vibrations by the Monte-Carlo method

The stochastic parametric vibrations of the elastic systems behave to the statistical dynamics section of the nonlinear systems. Under stochastic parametric influence the vibrations of the elastic systems, which is a random process more frequent are not stabilized, but their amplitudes are fading or unlimited are growing. Therefore important is stability of stochastic vibrations, which can be examined as stability by probability, on average or by the moment functions of different orders. The mathematical models which describe the stochastic parametric vibrations of the elastic systems are building by analytical or numeral approaches. Researches of stochastic stability of parametric vibrations are executing by probabilistic methods. Stability of parametric vibrations of modern constructions under the action of operating random loads are not enough investigated. In the article the numeral method of research of dynamic stability of the elastic systems under stochastic parametric influence was presented. The mathematical model of stochastic parametric vibrations of construction in the form of a reduсed finite element model was formed. Matrixes of the reduced model were obtained using calculated procedures of finite element analysis software NASTRAN. The dynamic loading as stationary ergodic random process with the spectral density was presented in the form of a finite amount of harmonic functions. Stochastic stability of the constructions was investigated using the Monte Carlo method and the Runge Kutta method of direct numeral integration of the equations of reduсed model. Stochastic stability as stability by probability of appearance of tendency to fading or unlimited growth of amplitude of parametric vibrations was examined during the time interval in space of the random parameters of loading. Dynamic stability of a I-beam with corrugated wall under narrow-band parametric influence was investigated using the numeral method.

On the Transformation of a Stationary Fuzzy Random Process by a Linear Dynamic System

In this paper, stationary random processes with fuzzy states are studied. The properties of their numerical characteristics—fuzzy expectations, expectations, and covariance functions—are established. The spectral representation of the covariance function, the generalized Wiener–Khinchin theorem, is proved. The main attention is paid to the problem of transforming a stationary fuzzy random process (signal) by a linear dynamic system. Explicitform relationships are obtained for the fuzzy expectations (and expectations) of input and output stationary fuzzy random processes. An algorithm is developed and justified to calculate the covariance function of a stationary fuzzy random process at the output of a linear dynamic system from the covariance function of a stationary input fuzzy random process. The results rest on the properties of fuzzy random variables and numerical random processes. Triangular fuzzy random processes are considered as examples.

Covariance-analytic performance criteria, Hardy-Schatten norms and Wick-like ordering of cascaded systems

This paper is concerned with robust performance analysis of linear stochastic systems acting as an integral operator on a standard Wiener process at the input and producing a stationary Gaussian random process at the output. We propose a performance criterion which uses the trace of an analytic function of the spectral density of the output process. This class of “covariance-analytic” cost functionals includes the usual mean square and risk-sensitive criteria as particular cases. Due to the “cost-shaping” analytic function, the covariance-analytic performance criterion depends on higher-order Hardy-Schatten norms of the system transfer function. We discuss the links of these norms with the asymptotic behaviour of cumulants of finite-horizon quadratic functionals of the system output and their variational properties pertaining to system robustness to statistically uncertain inputs which differ from the standard Wiener process. In the case of strictly proper finite-dimensional systems, governed in state space by linear stochastic differential equations, we develop a method for recursively computing the Hardy-Schatten norms through a recently proposed technique of rearranging cascaded linear systems, which resembles the Wick ordering of mixed products of noncommuting annihilation and creation operators in quantum mechanics. This computational procedure, which is one of the main results of the paper, involves a recurrence sequence of solutions to algebraic Lyapunov equations and represents the covariance-analytic cost as the squared H2-norm of an auxiliary cascaded system. These results are also compared with an alternative approach, which uses higher-order derivatives of stabilising solutions of parameter-dependent algebraic Riccati equations, and illustrated by a numerical example.

Spectral density estimation for random processes with stationary increments

AbstractSpectral density analysis plays an important role in studying a stationary random process on a real line. In this paper, we extend this discussion for the random process with stationary increments. We investigate the properties of the method of moments structure function estimation, and propose a nonparametric spectral density function estimator. Our numerical results show that the proposed spectral density estimator performs comparable with the parametric counterpart when the underlying process is assumed to be band‐limited. Additionally, this method is applied to analyze US Housing Starts Data, where the hidden periodicities are detected, providing consistent conclusions with previous economic studies.

On stationary random processes with fuzzy states

In this paper, continuous random processes with fuzzy states are studied. The properties of their numerical characteristics – fuzzy expectations, expected values and covariance functions – are established. The main attention is paid to the class of stationary fuzzy-random processes. For them, the ergodicity property and the spectral representation of covariance function (generalized Wiener–Khinchin theorem) are substantiated. The results obtained are based on the properties of fuzzy-random variables and numerical random processes. Triangular fuzzy-random processes are considered as examples.

Ensuring Seismic Resistance of Reinforced Concrete Buildings

There are a large number of works on a comprehensive assessment of the seismic resistance of buildings and structures. However, these studies, as a rule, do not take into account the random nature of the seismic impact, which is a pronounced non-stationary random process. An adequate assessment of the seismic resistance of buildings and structures is possible only on the basis of methods that allow taking into account the large variability of seismic impact parameters. The article presents a probabilistic method for calculating multi-storey reinforced concrete buildings designed in seismic regions, taking into account the interaction of a building with a non-linearly deformable foundation. The developed technique makes it possible to provide the required level of seismic resistance for the designed buildings based on the non-collapse criterion. As an ex-ample, the calculation of a multi-storey reinforced concrete building is considered. External seismic action is represented as a non-stationary random process. The external seismic action is considered as a non-stationary random process, which is obtained by multiplying the stationary random process by a deterministic envelope function. The parameters necessary for constructing the envelope and the stationary random process were obtained from the results of processing the available database of intense earthquakes. The stationary random process was generated by the shaping filter method. The impact parameters are based on the results of processing the available database of intense earthquakes. When modeling reinforced concrete structures, a concrete model is used with the function of damage accumulation under cyclic loads, as well as taking into ac-count the degradation of the strength and stiffness of the material during an intense earthquake. Accounting for the interaction of the building with the soil base is implemented using the SSI interface (Soil Structure Interaction). To prevent the influence of waves reflected from the boundaries of a limited ground massif, a PML layer (Perfectly Matched Layer) is used. The calculation was carried out using explicit methods for integrating the equations of motion on a computing cluster using parallel computing technology. The presented technique makes it possible to investigate the nature of the destruction of reinforced concrete structures during intense earthquakes and to identify zones with a deficiency in bearing capacity. The proposed probabilistic approach to modeling seismic impact as an implementation of a non-stationary random process with given parameters, together with taking into account the nonlinear deformation of the reinforced concrete structures of the building and foundation, allows you to control the level of reliability and design buildings with a given seismic resistance.

A hypothesis test for the domain of attraction of a random variable

In this work, we address the problem of detecting whether a sampled probability distribution of a random variable V has infinite first moment. This issue is notably important when the sample results from complex numerical simulation methods. For example, such a situation occurs when one simulates stochastic particle systems with complex and singular McKean–Vlasov interaction kernels. As stated, the detection problem is ill-posed. We thus propose and analyze an asymptotic hypothesis test for independent copies of a given random variable which is supposed to belong to an unknown domain of attraction of a stable law. The null hypothesis H0 is: ‘√V is in the domain of attraction of the Normal law’ and the alternative hypothesis is H1: ‘X is in the domain of attraction of a stable law with index smaller than 2’. Our key observation is that X cannot have a finite second moment when H0 is rejected (and therefore H1 is accepted). Surprisingly, we find it useful to derive our test from the statistics of random processes. More precisely, our hypothesis test is based on a statistic which is inspired by methodologies to determine whether a semimartingale has jumps from the observation of one single path at discrete times. We justify our test by proving asymptotic properties of discrete time functionals of Brownian bridges.

Learning mechanism for a collective classifier based on competition driven by training examples

The purpose of this work is to modify the learning mechanism of a collective classifier in order to provide learning by population dynamics alone, without requiring an external sorting device. A collective classifier is an ensemble of non-identical simple elements, which do not have any intrinsic dynamics neither variable parameters; the classifier admits learning by adjusting the composition of the ensemble, which was provided in the preceding literature by selecting the ensemble elements using a sorting device. Methods. The population dynamics model of a collective classifier is extended by adding a “learning subsystem”, which is controlled by a sequence of training examples and, in turn, controls the strength of intraspecific competition in the population dynamics. The learning subsystem dynamics is reduced to a linear mapping with random parameters expressed via training examples. The solution to the mapping is an asymptotically stationary Markovian random process, for which we analytically find asymptotic expectation and show its variance to vanish in the limit under the specified assumptions, thus allowing an approximate deterministic description of the coupled population dynamics based on available results from the preceding literature. Results. We show analytically and illustrate it by numerical simulation that the decision rule of our classifier in the course of learning converges to the Bayesian rule under assumptions which are essentially in line with available literature on collective classifiers. The implementation of the required competitive dynamics does not require an external sorting device. Conclusion. We propose a conceptual model for a collective classifier, whose learning is fully provided by its own population dynamics. We expect that our classifier, similarly to the approaches taken in the preceding literature, can be implemented as an ensemble of living cells equipped with synthetic genetic circuits, when a mechanism of population dynamics with synthetically controlled intraspecific competition becomes available.

Limit theorems for local polynomial estimation of regression for functional dependent data

<p>Local polynomial fitting exhibits numerous compelling statistical properties, particularly within the intricate realm of multivariate analysis. However, as functional data analysis gains prominence as a dynamic and pertinent field in data science, the exigency arises for the formulation of a specialized theory tailored to local polynomial fitting. We explored the intricate task of estimating the regression function operator and its partial derivatives for stationary mixing random processes, denoted as $ (Y_i, X_i) $, using local higher-order polynomial fitting. Our key contributions include establishing the joint asymptotic normality of the estimates for both the regression function and its partial derivatives, specifically in the context of strongly mixing processes. Additionally, we provide explicit expressions for the bias and the variance-covariance matrix of the asymptotic distribution. Demonstrating uniform strong consistency over compact subsets, along with delineating the rates of convergence, we substantiated these results for both the regression function and its partial derivatives. Importantly, these findings rooted in reasonably broad conditions that underpinned the underlying models. To demonstrate practical applicability, we leveraged our results to compute pointwise confidence regions. Finally, we extended our ideas to the nonparametric conditional distribution, and obtained its limiting distribution.</p>

Direct Electromagnetic Wave Scattering Calculation Using Methods of Moments through Layered Rough Surface

This thesis focuses on the direct calculation of electromagnetic wave scattering through layered rough surfaces using the Method of Moments. The study aims to contribute to existing knowledge by addressing the problems associated with accurately predicting scattered electromagnetic fields from rough surfaces with layered structures. By using the Method of Moments, this research seeks to provide insights into the behavior of electromagnetic waves when interacting with multi-layered rough surfaces, offering valuable implications for practical engineering applications and the development of advanced computational tools for electromagnetic wave analysis. In practical terms, the surface is typically divided into small patches, and the electromagnetic field at each patch is determined using MoM. This involves solving integral equations describing the interaction between the induced currents on the surface and the incident field. The layered structure may require additional equations to model the reflection of waves and the transmission at the boundaries between the layers. The direct problem is also examined in a two-dimensional picture, where the rough surface profile and the medium are known parameters, and the electromagnetic waves scattered from the surface to the entire space are the unknowns. The rough surface is considered to have a finite length and can be deterministic or arise from a stationary stochastic random process, with a focus on Gaussian-type rough surfaces. The roughness is considered as two basic types: as a perfect electric conductor (PEC) in free space and as a boundary between two dielectric media. The equations of integration obtained are based on the scalar solution of Green's theory, Helmholtz equations, and using Hankel type green functions as kernels due to the one-dimensional nature. The MoM is extensively used to solve these integral equations and is compared to asymptotic approaches and analytical. Various simulations are conducted to test the feasibility of the algorithms, and the results are thoroughly discussed. The study of electromagnetic wave scattering in one-dimensional rough surfaces involves the direct problem, where the field distribution in the whole space is determined using the equation of integration. These equations take into account the surface roughness and medium parameters they are solved using the Method of Moments (MoM) by expressing the field distribution on the surface using basis functions. The direct calculation of electromagnetic wave scattering through a layered rough surface using the Method of Moments (MoM) is a computationally intensive process and complex and. It contains simulating the interaction of EM waves with a complex surface, which presents challenges due to the multiple layers with different properties, and a roughly surface causing and diffraction of incident waves and scattering. When working with a layered rough surface increases the electromagnetic wave interaction complexity. Additional multiple layers introduce considerations, such as layer interfaces and accounting for different material abilities. This requires careful modeling and analysis to exactly capture the behavior of electromagnetic waves as they traverse through and interact with layers. MoM, a commonly used numerical method for solving electromagnetic scattering problems, discreteness the surface into small segments and solves integral equations to determine the fields that is scattered. This approach allows for a maximum analysis of electromagnetic interactions at a grained level.

Kinematically excited beam vibrations under random perturbations

There is numerous information in the technical and scientific literature that often the sources of forced kinematically excited oscillations have a clearly expressed random nature. Kinematically excited vibrations of beams are considered. Their sources are random disturbances. Steady (in a probabilistic sense) vibrations of beams are studied. As a consequence, the transverse vibrations of the beam will also be random, and it becomes necessary and expedient to switch to stochastic motion models.
 Introduction. For a random perturbation process, a spectral matrix is specified. Initial conditions for the equations are not required. The boundary conditions are similar to those used for free and kinematically excited deterministic oscillations. The task is to find the spectral matrix of the random field of beam deviations and dispersion for a given spectral matrix. The question of determining the mathematical expectation is not raised. It is argued that it can be easily reduced to known deterministic problems.
 Methods. To carry out specific calculations, a model of kinematic disturbances is used in the form of a vector N-dimensional stationary random process with stationary connected components that have hidden periodicities (characteristic frequencies). A test example was completed to determine the dispersion and standard deviation of displacements. The numerical integration step was adopted after numerical experiments. When performing calculations, the parity of the integrand was taken into account. The lower limit of integration is taken to be zero, followed by doubling of the final result.
 Results. For small values of the broadband parameter, the results of the stochastic and deterministic problems differ little. Examples are presented that are stochastic analogs of harmonic oscillations. For beams under kinematic harmonic disturbances, test examples were performed to determine dispersions and standard deviations.
 The discussion of the results. The results of calculations in the article are presented by curves having numbers that coincide with the numbers of the matrices. The first matrix corresponds to the absolute correlation of disturbances. The second matrix corresponds to the absolute correlation between the first and third, second and fourth disturbances, while these pairs are absolutely uncorrelated. The third matrix describes random perturbations when the indicated pairs, being perfectly positively correlated within, are absolutely perfectly negatively correlated between pairs. The fourth curve corresponds to the case of ideal correlation of the first three disturbances with each other, while the movements of the fourth support are in antiphase with them. In the latter case, the movements of the outer supports are in antiphase with the movements of the middle ones, which favors the greatest bending of the beam. Therefore, the elastic line has a curvature that is significantly greater than the others.

ON A CLASS OF NONSTATIONARY CURVES IN HILBERT SPACE

Stationary random processes have been studied quite well over recent years starting with the works of A. N. Kolmogorov. The possibility of building nonstationary random process correlation theory was considered in the monographs by M. S. Livshits, A. A. Yantsevich, V. A. Zolotarev and others. Some classes of nonstationary curves were investigated by V. E. Katsnelson. In this paper nonstationary random processes are represented as curves in Hilbert space which "slightly deviate" from random processes with the correlation function of special kind. The infinitesimal correlation function has been introduced; in essence, this function characterizes the deviation from the correlation process with the given correlation function. The paper discusses the cases of nonstationary random processes, the operator of which has one‑dimensional imaginary component. Cases of a dissipative operator with descrete spectrum are also considered in this work. It is shown that the nonstationarity of the random process is closely related to the deviation of the operator from its conjugated operator. Using the triangle and universal models of non‑self‑ajoint operators it is possible to obtain the representation for the correlation function in the case of nonstationary process which replaces the Bochner – Khinchin representation for stationary random processes. The expresson for the infinitesimal correlation function was obtained for different cases of operator spectrum: for the descrete spectrum placed in the upper half‑plane and for the contrast‑free spectrum at zero. In the case of dissipative operator with descrete spectrum the infinitesimal function can be found in terms of special lambda function. For Lebesque spaces of compex‑valued squared integratable functions the expresson of infinitesimal function was found in terms of special zero order modified Bessel function. It was shown that a similar approach can be applied for the evolutionarily represented sequences in Hilbert spaces.

Stationary, Markov, stochastic processes with polynomial conditional moments and continuous paths

We are studying stationary random processes with conditional polynomial moments that allow a continuous path modification. Processes with continuous path modification are important because they are relatively easy to simulate. One does not have to care about the distribution of their jumps which is always difficult to find. Among those processes with the continuous path are the Ornstein–Uhlenbeck process, the Gamma process, the process with Arcsin or Wigner margins and the Theta functions as the transition densities and others. We give a simple criterion for the stationary process to have a continuous path modification expressed in terms of skewness and excess kurtosis of the marginal distribution.

The Effect of Random Roughness on the Electromagnetic Flow in a Micropipe.

The features of stationary random processes and the small parameter expansion approach are used in this work to examine the impact of random roughness on the electromagnetic flow in cylindrical micropipes. Utilizing the perturbation method, the analytical solution until second order velocity is achieved. The analytical expression of the roughness function ζ, which is defined as the deviation of the flow rate ratio with roughness to the case having no roughness in a smooth micropipe, is obtained by integrating the spectral density. The roughness function can be taken as the functions of the Hartmann number Ha and the dimensionless wave number λ. Two special corrugated walls of micropipes, i.e., sinusoidal and triangular corrugations, are analyzed in this work. The results reveal that the magnitude of the roughness function rises as the wave number increases for the same Ha. The magnitude of the roughness function decreases as the Ha increases for a prescribed wave number. In the case of sinusoidal corrugation, as the wave number λ increases, the Hartmann number Ha decreases, and the value of ζ increases. We consider the λ ranging from 0 to 15 and the Ha ranging from 0 to 5, with ζ ranging from -2.5 to 27.5. When the λ reaches 15, and the Ha is 0, ζ reaches the maximum value of 27.5. At this point, the impact of the roughness on the flow rate reaches its maximum. Similarly, in the case of triangular corrugation, when the λ reaches 15 and the Ha is 0, ζ reaches the maximum value of 18.7. In addition, the sinusoidal corrugation has a stronger influence on the flow rate under the same values of Ha and λ compared with triangular corrugation.

Analysis of the input material flow of the transport conveyor

Purpose. To develop a method for analyzing the material flow entering the input of a conveyor section, based on the decomposition of the input material flow into a deterministic material flow and a stochastic material flow. Methodology. The analysis of experimental data characterizing the input material flow was performed using the methods of the canonical Fourier representation of a random process. Findings. A method for representing a stochastic material flow as a combination of a deterministic process and a stationary random process with ergodic properties is proposed. Originality. The originality of the obtained results lies in the fact that, for the first time, a method of analysis based on the decomposition of the input material flow for a conveyor section has been proposed, which, unlike the existing methods of input flow typing for the mining industry, will allow us to independently perform deterministic flow typing and stochastic material flow typing in transport conveyors. The proposed approach makes it possible to highlight special characteristics separately for deterministic and stochastic material flows. This will make it possible to use the obtained regularities to increase the accuracy of the conveyor model and will accordingly increase the quality of the belt speed control systems and the flow of material coming from the input bunker. The obtained results are of particular importance due to the fact that the characteristics of the deterministic material flow are directly related to the technical or technological factors of material extraction. Practical value. The obtained results allow determining statistically stable regularities for the incoming flow, which makes it possible, based on these regularities from the set of available control algorithms, to choose the optimal control algorithm for the parameters of the operating conveyor section. This allows reducing the enterprise’s energy costs of the transportation of material. The proposed method can be successfully applied to build random number generators simulating the sequence of values of the input flow of material. The developed generators can be used both for validating existing belt speed control systems and creating new control systems based on neural networks. This opens perspectives for the design of effective systems for controlling the flow parameters of transport system, based on the transport conveyor model, which takes into account the stochastic nature of the incoming material flow.