Article 8318

Title of the article

ESTIMATION OF A CORRELATION FUNCTION OF RANDOM INTERFERENCE 

Authors

Popov Dmitriy Ivanovich, Doctor of engineering sciences, professor, sub-department of radio systems, Ryazan State Radio Engineering University (59/1 Gagarina street, Ryazan, Russia), E-mail: adop@mail.ru 

Index UDK

621.391:621.396 

DOI

10.21685/2072-3059-2018-3-8 

Abstract

Background. The object of the study is the meters of a correlation function of random interference. The aim of the work is to construct analog and digital recurrent algorithms and the corresponding structural schemes for estimating the parameters of the correlation function.
Materials and methods. Approximation of the correlation function in the form of an expansion in a series of orthogonal polynomials is introduced. Estimation of the correlation function on the basis of its expansion in a series of orthogonal polynomials is carried out by determining the desired coefficients of expansion by the method of stochastic approximation from the input data using digital orthogonal filters determined by the chosen system of polynomials.
Results. Analog and digital recurrent algorithms for estimating the current values of the expansion coefficients are obtained as the input data arrive. The required estimates converge to their true values with probability one. The estimates of the coefficients of the expansion together with the chosen system of orthonormal polynomials uniquely determine the analytic approximation of the unknown correlation function of the random noise or its discrete values.
Conclusions. Current estimates of the expansion coefficients obtained both in the transient and steady state can be directly used in the adaptation of the parameters or structure of the signal processing system. In this case, the estimates obtained in the transient regime should be used with a certain weight, the value of which before the beginning of adaptation is established on the basis of the available a priori information and varies during the adaptation process depending on the rate of convergence of the considered algorithms. 

Key words

correlation function, expansion coefficients, stochastic approximation method, orthogonal polynomials, orthogonal filters, recurrent estimation algorithms, random interference, convergence of estimates 

 Download PDF
References

1. Repin V. G., Tartakovskiy G. P. Statisticheskij sintez pri apriornoj neopredelennosti i adaptacija informacionnyh sistem [Statistical synthesis under a priori uncertainty and adaptation of information systems]. Moscow: Sovetskoe radio Publ., 1977, 432 p.
2. Rodionov V. V. Antennas. 2014, no 1 (200), pp. 023–029.
3. Popov D. I. Cifrovaja obrabotka signalov [Digital signal processing]. 2014, no. 4, pp. 32–37.
4. Popov D. I. Radiopromyshlennost [Radio industry]. 2017, no. 3, pp. 13–18.
5. Uidrou B., Stirnz S. Adaptivnaja obrabotka signalov [Adaptive signal processing]. Transl. from Engl. Moscow: Radio i svjaz' Publ., 1989, 440 p.
6. Nevel'son M. B., Has'minskij R. Z. Stohasticheskaja approksimacija i rekurrentnoe ocenivanie [Stochastic approximation and recurrent estimation]. Moscow: Nauka Publ., 1972, 304 p.

 

Дата создания: 19.04.2019 14:04
Дата обновления: 22.04.2019 08:23