Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input

Volterra-series-type analyses to communication systems with stochastic resonance driven by sine waves result to. The n -hold Fourier transform of the n -th Volterra kernel plays an important role in the analysis. Methods of computing transfer functions from the system equation are described and ob...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Вопросы атомной науки и техники
Datum:2018
Hauptverfasser: Kharchenko, O. I., Lonin, Yu.F., Ponomarev, A.G.
Format: Artikel
Sprache:English
Veröffentlicht: Національний науковий центр «Харківський фізико-технічний інститут» НАН України 2018
Schlagworte:
Online Zugang:https://nasplib.isofts.kiev.ua/handle/123456789/147447
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input / O. I. Kharchenko, Yu.F. Lonin, A.G. Ponomarev // Вопросы атомной науки и техники. — 2018. — № 4. — С. 249-251. — Бібліогр.: 9 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-147447
record_format dspace
spelling Kharchenko, O. I.
Lonin, Yu.F.
Ponomarev, A.G.
2019-02-14T18:47:09Z
2019-02-14T18:47:09Z
2018
Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input / O. I. Kharchenko, Yu.F. Lonin, A.G. Ponomarev // Вопросы атомной науки и техники. — 2018. — № 4. — С. 249-251. — Бібліогр.: 9 назв. — англ.
1562-6016
PACS: 05.45
https://nasplib.isofts.kiev.ua/handle/123456789/147447
Volterra-series-type analyses to communication systems with stochastic resonance driven by sine waves result to. The n -hold Fourier transform of the n -th Volterra kernel plays an important role in the analysis. Methods of computing transfer functions from the system equation are described and obtaining results are considered. It is shown, if the transfer functions are known, then the output signal can be obtained by substitution the transfer functions in general formulas derived from Volterra series representation. The obtained results showed that the amplitude of the sinusoid should not be greater than one to obtain a minor third harmonic. Besides, with an increase in the frequency of the sinusoid, the power of the output signal decreases sharply. Numerical calculations of the output signal driven by sinusoidal input were made to improve the accuracy and reliability of the obtained results. Comparative analysis showed the coincidence of the results of the calculation by different methods.
Наведено аналіз на основі рядів Вольтера стосовно телекомунікаційних систем, що володіють ефектом стохастичного резонансу, які збуджуються синусоїдальним сигналом. Важливу роль при такому аналізі грає багатовимірне перетворення Фур'є. Описано способи обчислення передаточних функцій, виходячи з рівняння системи, і розглянуто результати розрахунків. Порівняльний аналіз чисельних розрахунків і розрахунків на основі рядів Вольтера показав збіг результатів розрахунку.
Приведен анализ на основе рядов Вольтера применительно к телекоммуникационным системам, обладающим эффектом стохастического резонанса, возбуждаемым синусоидальным сигналом. Важную роль при таком анализе играет многомерное преобразование Фурье. Описаны способы вычисления передаточных функций, исходя из уравнения системы, и рассмотрены результаты расчетов. Сравнительный анализ численных расчетов и расчетов на основе рядов Вольтера показал совпадение результатов расчета.
en
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
Вопросы атомной науки и техники
Нелинейные процессы
Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
Частотний аналіз стохастичної фільтрації за допомогою передаточних функцій. Частина i: Синусоїдальний вхідний сигнал
Частотный анализ стохастической фильтрации с помощью передаточных функций. Часть I: синусоидальный входной сигнал
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
spellingShingle Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
Kharchenko, O. I.
Lonin, Yu.F.
Ponomarev, A.G.
Нелинейные процессы
title_short Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
title_full Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
title_fullStr Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
title_full_unstemmed Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input
title_sort frequency analysis of the stochastic filtering using transfer functions. part i: sinusoidal input
author Kharchenko, O. I.
Lonin, Yu.F.
Ponomarev, A.G.
author_facet Kharchenko, O. I.
Lonin, Yu.F.
Ponomarev, A.G.
topic Нелинейные процессы
topic_facet Нелинейные процессы
publishDate 2018
language English
container_title Вопросы атомной науки и техники
publisher Національний науковий центр «Харківський фізико-технічний інститут» НАН України
format Article
title_alt Частотний аналіз стохастичної фільтрації за допомогою передаточних функцій. Частина i: Синусоїдальний вхідний сигнал
Частотный анализ стохастической фильтрации с помощью передаточных функций. Часть I: синусоидальный входной сигнал
description Volterra-series-type analyses to communication systems with stochastic resonance driven by sine waves result to. The n -hold Fourier transform of the n -th Volterra kernel plays an important role in the analysis. Methods of computing transfer functions from the system equation are described and obtaining results are considered. It is shown, if the transfer functions are known, then the output signal can be obtained by substitution the transfer functions in general formulas derived from Volterra series representation. The obtained results showed that the amplitude of the sinusoid should not be greater than one to obtain a minor third harmonic. Besides, with an increase in the frequency of the sinusoid, the power of the output signal decreases sharply. Numerical calculations of the output signal driven by sinusoidal input were made to improve the accuracy and reliability of the obtained results. Comparative analysis showed the coincidence of the results of the calculation by different methods. Наведено аналіз на основі рядів Вольтера стосовно телекомунікаційних систем, що володіють ефектом стохастичного резонансу, які збуджуються синусоїдальним сигналом. Важливу роль при такому аналізі грає багатовимірне перетворення Фур'є. Описано способи обчислення передаточних функцій, виходячи з рівняння системи, і розглянуто результати розрахунків. Порівняльний аналіз чисельних розрахунків і розрахунків на основі рядів Вольтера показав збіг результатів розрахунку. Приведен анализ на основе рядов Вольтера применительно к телекоммуникационным системам, обладающим эффектом стохастического резонанса, возбуждаемым синусоидальным сигналом. Важную роль при таком анализе играет многомерное преобразование Фурье. Описаны способы вычисления передаточных функций, исходя из уравнения системы, и рассмотрены результаты расчетов. Сравнительный анализ численных расчетов и расчетов на основе рядов Вольтера показал совпадение результатов расчета.
issn 1562-6016
url https://nasplib.isofts.kiev.ua/handle/123456789/147447
citation_txt Frequency analysis of the stochastic filtering using transfer functions. Part I: Sinusoidal input / O. I. Kharchenko, Yu.F. Lonin, A.G. Ponomarev // Вопросы атомной науки и техники. — 2018. — № 4. — С. 249-251. — Бібліогр.: 9 назв. — англ.
work_keys_str_mv AT kharchenkooi frequencyanalysisofthestochasticfilteringusingtransferfunctionspartisinusoidalinput
AT loninyuf frequencyanalysisofthestochasticfilteringusingtransferfunctionspartisinusoidalinput
AT ponomarevag frequencyanalysisofthestochasticfilteringusingtransferfunctionspartisinusoidalinput
AT kharchenkooi častotniianalízstohastičnoífílʹtracíízadopomogoûperedatočnihfunkcíičastinaisinusoídalʹniivhídniisignal
AT loninyuf častotniianalízstohastičnoífílʹtracíízadopomogoûperedatočnihfunkcíičastinaisinusoídalʹniivhídniisignal
AT ponomarevag častotniianalízstohastičnoífílʹtracíízadopomogoûperedatočnihfunkcíičastinaisinusoídalʹniivhídniisignal
AT kharchenkooi častotnyianalizstohastičeskoifilʹtraciispomoŝʹûperedatočnyhfunkciičastʹisinusoidalʹnyivhodnoisignal
AT loninyuf častotnyianalizstohastičeskoifilʹtraciispomoŝʹûperedatočnyhfunkciičastʹisinusoidalʹnyivhodnoisignal
AT ponomarevag častotnyianalizstohastičeskoifilʹtraciispomoŝʹûperedatočnyhfunkciičastʹisinusoidalʹnyivhodnoisignal
first_indexed 2025-11-25T01:48:18Z
last_indexed 2025-11-25T01:48:18Z
_version_ 1850504047554985984
fulltext ISSN 1562-6016. ВАНТ. 2018. №4(116) 249 FREQUENCY ANALYSIS OF THE STOCHASTIC FILTERING USING TRANSFER FUNCTIONS. PART I: SINUSOIDAL INPUT O. I. Kharchenko, Yu.F. Lonin, A.G. Ponomarev National Science Center “Kharkov Institute of Physics and Technology”, Kharkov, Ukraine E-mail: dthnbycrbq@gmail.com; lonin@kipt.kharkov.ua Volterra-series-type analyses to communication systems with stochastic resonance driven by sine waves result to. The n -hold Fourier transform of the n -th Volterra kernel plays an important role in the analysis. Methods of computing transfer functions from the system equation are described and obtaining results are considered. It is shown, if the transfer functions are known, then the output signal can be obtained by substitution the transfer func- tions in general formulas derived from Volterra series representation. The obtained results showed that the ampli- tude of the sinusoid should not be greater than one to obtain a minor third harmonic. Besides, with an increase in the frequency of the sinusoid, the power of the output signal decreases sharply. Numerical calculations of the output signal driven by sinusoidal input were made to improve the accuracy and reliability of the obtained results. Compar- ative analysis showed the coincidence of the results of the calculation by different methods. PACS: 05.45 INTRODUCTION IN VOLTERRA SERIES ANALYSIS In communication systems often it is necessary to deal with the devices executing non-linear conversions. Volterra series are usually used for calculation of such devices. Wiener introduced Volterra series into nonline- ar circuit analysis [1]. The object of this paper is to present results of ap- plying Volterra-series-type analyses to systems driven by sine waves. Volterra series describe the output of a nonlinear system in degrees of input ( )x t . A substantial number of the communication system can be represented as Volterra series. The series for typical system can be writing as [2] ( ) ( )1 1 1 1 1( ) ... , ..., ! n n n n r n r y t du du g u u x t u n ∞ ∞ ∞ −∞ −∞ = = = −∑ ∏∫ ∫ , (1) where ( )y t is the output, ( )x t − the input and the ker- nels ( )1, ...,n ng u u describe the system. The first-order kernel 1 1( )g u is simply the familiar impulse response of linear network. The higher order kernels of higher order impulse responses and serve to characterize the various orders of nonlinearity. The coefficient 1 !n inserted A. Bedrosian and D. Rice [2] it because it simplifies many of equations. The n -fold Fourier transform have the form [2, 3] ( ) ( ) ( ) 1 1 1 1 1 ,..., ... ,..., exp ... . n n n n n n n G f f du du g u u j f u f u ∞ ∞ −∞ −∞ =  = − + ∫ ∫ (2) 0G is identically zero because our Volterra series starts with 1n = . ( )1 1G f is the transfer function of linear network. For linear systems, the possible output frequencies are the same as the frequencies in the input. For non-linear systems, however, the relationship be- tween the input and output frequencies is more compli- cated [4, 5]. Thus the transform of the n th-order Volterra kernel is seen to be analogous to an n th-order Volterra trans- fer function. In many cases Gn can be obtained without first computing gn. The complete formulas are infinite series. Fortunate- ly, in the study of communication system it is often pos- sible to neglect terms of the Volterra series of order higher then the second or third. They are usually used because of fast increase in complexity [2, 3]. The n - fold Fourier transform considerably simplifies the solu- tion of a large number of problems. To calculate the transfer functions, we use the har- monic input method [2]. This method relies on the fact that a harmonic input must result in a harmonic output when (1) holds. System specified by the nonlinear dif- ferential equation considered [2, 3] 2 ( / ) ( )l l l F d dt y a y x t ∞ = + =∑ , (3) with the condition that system causally ( ( )y t vanish identically when ( )x t does). It is assumed that one and only one such solution exists (it is proved in [3]) and the system is stable. ( / )F d dt is a polynomial in /d dt , and the coefficients in ( / )F d dt and the coefficients la are independent of , ,t x and y . The Volterra transfer functions for (3) can be written as [2] ( ) 1 2 1 1 ( , ..., ) ( , ..., ) ( ... ) n l l n n l n n n a G f f G f f F j jω ω == − + + ∑ . (4) The last equation is recurrence relation because ( )l nG is given by ( ) 1 1 1 1 ( ; , ) 2 2 1 1 1 2 ( ,..., ) ! ( ,..., ) ( ,..., )... ( ,..., ) n l n n l l n l l n G f f l a G f f G f f G f f n n n n n n n n µ = + + = × × ∑ ∑ , for the n -fold Fourier transform of the n -th kernel in the Volterra series for [ ( )]ly t , l being a positive inte- ger, and 1 l n≤ ≤ . ( ) 1( , ..., )l n nG f f is zero for l n> and ( ) 1( , ..., )n n nG f f is equal to 1 1 1 2 1! ( ) ( )... ( )nn G f G f G f . mailto:dthnbycrbq@gmail.com mailto:lonin@kipt.kharkov.ua ISSN 1562-6016. ВАНТ. 2018. №4(116) 250 SINUSOIDAL INPUT FOR STOCHASTIC RESONANCE Consider a general bistable dynamic system which can be described by the following stochastic differential equation [6, 7] 3 ( )dy ay by x t dt = − + , (5) where a and b are positive, usually given in terms of system parameters, ( ) ( ) ( )x t s t n t= + , 0( ) sin(2 )s t A f tπ ϕ= + is the driving signal, ( )n t is the input noise, [ ( ) ( )] 2 ( )E n t n t Dt δ t+ = , D is the noise intensity. The input signal consists of the driving signal ( )s t and the additive noise ( )n t [6, 8]. This equation describes a stochastic resonance effect (SR) [6 - 8]. Volterra trans- fer function for ( )y t are given in table 1 for the general case (the equation 3) and for SR equation. Table 1 Volterra transfer function Volterra transfer function for eq. 3 [2] Volterra transfer function for eq. 4 1G 11/ ( )F jω 1 1 a jω− + 2G 2 1 1 1 2 1 22 ( ) ( ) / ( )a G f G f F j jω ω− + 0 3G ' 2 1 1 2 2 3 3 1 1 1 2 1 3 3 1 2 3 2 ( ) ( , ) 6 ( ) ( ) ( ) ( ) a G f G f f a G f G f G f F j j jω ω ω + − + + ∑ 1 2 3 1 2 3 6 ( )( )( )( ) b a j a j a j a j j jω ω ω ω ω ω − − + − + − + − + + + When noise is absent, input signal can be writing as 0( ) ( ) sin(2 )x t s t A f tπ ϕ= = + . We will solve SR equation for this case. The leading terms output signal are given in Table 2. Table 2 The leading terms output signal The output signal for eq. 3 [2] The output signal for eq. 4 2 2 ( , ) ... 4 p p P G f f   − +    0 3 1 3( ) ( , , ) ... 2 16 jpt p p p p P Pe G f G f f f   + − +    + 3 1 3( ) ( , , ) ... 2 16 jpt p p p p P Pe G f G f f f−   + − + − − +    2 2 2 0 02 2 2 2 2 0 0 3 ( ) cos ( ) 4( ) A b aA a t a a ω ω ω ω  − + − +   + +  + 2 0 0 02 2 2 2 2 0 0 3 sin ( ) 2( ) A bA t a a ω ω ω ω ω   + +   + +  2 2 2 ( , ) ... 8 j pt p p Pe G f fπ   +    + 2 2 2 ( , ) ... 8 j pt p p Pe G f fπ−   + − − +    0 3 3 3 ( , , ) ... 48 j pt p p p Pe G f f fπ   + +    + 3 3 3 ( , , ) ... ... 48 j pt p p p Pe G f f fπ−   + − − − + +    4 2 2 43 0 0 0 2 2 3 2 2 2 2 0 0 0 0 0 ( 12 3 ) cos 3 4( ) ( 9 ) 2 (5 3 ) sin 3 a a tA b a a a a t ω ω ω ω ω ω ω ω  − − + +    + + + −  We will define the output signal of non-linear sys- tem by the Runge-Kutta method and by Volterra series. Results of calculations are given in Figure for 1,A = 0.5f = Hz. Results from the Figure show, that in the numerical calculation by Runge-Kutta method takes place transient that lasts about two periods of a signal. Further the re- sults of calculation of an output signal received by both methods match. We will determine the power of an output signal. Powers of the first and third harmonica are specified in Table 3. Analyzed results of the output signal using different methods: Volterra series (solid line); Runge-Kutta method (dotted line) ISSN 1562-6016. ВАНТ. 2018. №4(116) 251 Table 3 The power of the output signal The power of the output signal for eq. 3 [2] The power of the output signal for eq. 4 0 Sω 23 1 0 3 0 0 0( ) ( , , ) 2 16 A AG f G f f f+ − 2 22 2 4 6 2 0 2 2 2 2 2 3 2 2 4 2 2 2 0 0 0 0 3 9 4( ) 8( ) 64( ) 4( ) Aa A A ba A b a a a a ω ω ω ω ω − + + + + + + 03S ω 23 3 0 0 0( , , ) 48 P G f f f 6 2 2 2 3 2 2 0 064( ) ( 9 ) A b a aω ω+ + Results of Table 3 show that, with an increase in the frequency of the sinusoid, the power of the output signal decreases drastically. We can calculate the powers relation of the first and third harmonicas as: 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 0 0 0 0 0 0 30 0 4 2 2 2 2 4 2 0 16( )( 9 ) 24 ( 9 ) 9( 9 ) 16 ( )( 9 ) / ( ) a a a a a a a a S S A b A b a A b ω ω ω ω ω ω ω ω ω ω + + + + + + = − + + + . The obtained results showed that the amplitude of the sinusoid A should not be greater than one to obtain a minor third harmonic. CONCLUSIONS The Volterra series is a powerful tool that can be used to describe a wide class of non-linear systems. Results of applying Volterra series analysis to sys- tems with SR effect driven by harmonic input showed what output signal contains the 1st and 3rd harmonicas, that matches the known results [6]. In addition, with an increase in the frequency of the sinusoid, the power of the output signal decreases sharply and with an increase in the frequency of the sinusoid, the power of the output signal decreases sharply. Results of calculations showed that numerical calcu- lation and calculation by Volterra series match. At the same time numerical calculation is followed by transient which lasts about two periods of oscillations. Transient take place in radioengineering devices [9]. In the present paper the method of transfer functions is developed. The received gear Volterra transfer func- tion for the systems with SR effect will allow to receive further expressions for an output signal in case of input white Gaussian noise and an input additive mix of a harmonic signal and white Gaussian noise. REFERENCES 1. N. Wiener. Nonlinear Problems in Random Theory. Cambridge, Mass. Technology Press; and New York Wiley, 1958. 2. E Bedrosian, S. Rice. The Output Properties of Volterra Systems (Nonlinear Systems with Memory) Driven by Harmonic and Gaussian Inputs // IEEE. 1971, v. 59, № 12, p. 58-82. 3. K.A. Pupkov, V.I. Kapalin, A.S. Yushchenko. Funktsional'nyye ryady v teorii nelineynykh sistem. M.: “Nauka” 1978, 448 p. (in Russian). 4. N.V. Zernov, V.G. Karpov. Teoriya radiotekhnich- eskikh tsepey. Moskva-Leningrad: “Energiya”, 1965, 816 p. (in Russian). 5. Yu.I. Voloshchuk. Pidruchnik dla studentiv vich. navch. zakladiv. Kharkiv: TOV “Kompania CMIT”, 2005, v. 3, 228 p. (in Russian). 6. V.S. Anishchenko, A.B. Neiman, F. Moss, L. Schimansky-Geier. Stochastic resonance: noise- enhanced order // Uspekhi Fizicheskikh Nauk, Rus- sian Academy of Sciences. 1999, 42(1)7-36, p. 7-34 (in Russian). 7. O. Kharchenko. Simulation of the Stochastic Reso- nance Effect in a Nonlinear Device // Global Jour- nal of Researches in Engineering -F Volume 15 Is- sue 7 Version 1.0 September 2015, p. 19-23. 8. Xiaofei Zhang, Niaoqing Hu, Zhe Cheng. Stochastic resonance in multi-scale bistable array // Physics Letter. 2013, A 377, p. 981-984. 9. B. Sklar. Digital Communication. Fundamentals and Applications, Second Edition, Prentice Hall PTR, 2003, 1099 p. Article received 04.06.2018 ЧАСТОТНЫЙ АНАЛИЗ СТОХАСТИЧЕСКОЙ ФИЛЬТРАЦИИ С ПОМОЩЬЮ ПЕРЕДАТОЧНЫХ ФУНКЦИЙ. ЧАСТЬ I: СИНУСОИДАЛЬНЫЙ ВХОДНОЙ СИГНАЛ О.И. Харченко, Ю.Ф. Лонин, А.Г. Пономарев Приведен анализ на основе рядов Вольтера применительно к телекоммуникационным системам, обладающим эф- фектом стохастического резонанса, возбуждаемым синусоидальным сигналом. Важную роль при таком анализе играет многомерное преобразование Фурье. Описаны способы вычисления передаточных функций, исходя из уравнения систе- мы, и рассмотрены результаты расчетов. Сравнительный анализ численных расчетов и расчетов на основе рядов Воль- тера показал совпадение результатов расчета. ЧАСТОТНИЙ АНАЛІЗ СТОХАСТИЧНОЇ ФІЛЬТРАЦІЇ ЗА ДОПОМОГОЮ ПЕРЕДАТОЧНИХ ФУНКЦІЙ. ЧАСТИНА I: СИНУСОЇДАЛЬНИЙ ВХІДНИЙ СИГНАЛ О.І. Харченко, Ю.Ф. Лонін, А.Г. Пономарьов Наведено аналіз на основі рядів Вольтера стосовно телекомунікаційних систем, що володіють ефектом стохастично- го резонансу, які збуджуються синусоїдальним сигналом. Важливу роль при такому аналізі грає багатовимірне перетво- рення Фур'є. Описано способи обчислення передаточних функцій, виходячи з рівняння системи, і розглянуто результати розрахунків. Порівняльний аналіз чисельних розрахунків і розрахунків на основі рядів Вольтера показав збіг результатів розрахунку. references