UDC 681.518.2, 681.514
|
|
4 Scientific Center of the Military Academy, Odessa, Ukraine
 akorcin@ukr.net
|
SOLVING THE PROBLEM OF GENERATING STABLE AND CONSISTENTESTIMATES
OF THE CORRELATION MATRIX OF OBSERVATIONS BY THE DYNAMIC
REGULARIZATION METHOD
Abstract. The consistency of stable estimates of the correlation matrix of observations with their static and dynamic regularization is analyzed. The advantage of the dynamic regularization method with the optimal parameter in the context of resolving the contradiction of computational stability and consistency of sample estimates of the correlation matrix of observations is proved. An algorithm is obtained for calculating the optimal dynamic regularization parameter, which does not use forecasting data and does not require additional computational costs.
Keywords: regularization, consistency, stability, convergence, evaluation.
FULL TEXT
REFERENCES
- Wirth W.D. Radar techniques using array antennas (2nd ed). London: The Institution of Engineering and Technology, 2013. 530 p.
- Lekhovytskiy D.I. To the theory of adaptive signal processing in systems with centrally symmetric receive channels, EURASIP Journal on Advances in Signal Process. 2016. Article number 33. https://doi.org/10.1186/s13634-016-0329-z.
- Lekhovitskiy D.I., Abramovich Y.I., Zhuga G.A., Rachkov D.S. Band-diagonal regularization of MP estimates of Gaussian interference correlation matrices in antenna array adaptation algorithms. Prikladnaya radioelektronika. 2010. Vol. 9, N 1. P. 107-121.
- Abramovich Yu.I., Spencer N.K., Johnson B.A. Band-inverse TVAR covariance matrix estimation for adaptive detection. IEEE Transactions on Aerospace and Electronic Systems. 2010. Vol. 46, N 1. P. 375-396. https://doi.org/10.1109/TAES.2010.5417169.
- Cheremisin O.P. Efficiency of an adaptive algorithm with regularization of the sample correlation matrix. Radiotekhnika i elektronika. 1982. Vol. 27, N 10. P. 1933-1942.
- Tikhonov A.N., Goncharskiy A.V., Stepanov V.V., Yagola A.G. Numerical methods for solving ill-posed problems [in Russian]. Moscow: Nauka, 1990. 232 p.
- Voskoboinikov Yu.E., Micel A.A. Modern problems of applied mathematics. Part 1. Lecture course [in Russian]. Tomsk: Tomsk state. University of Control Systems and Radioelectronics (TUSUR), 2015. 136 p.
- Osipov Yu.S., Vasiliev F.P., Potapov M.M. Fundamentals of the dynamic regularization method [in Russian]. Moscow: MGU Publishing House, 1999. 237 p.
- Skachkov V., Chepkyi V., Bratchenko H., Tkachuk H., Kazakova N. Development of the method for dynamic regularization of selected estimates in the correlation matrices of observations. Eastern-European Journal of Enterprise Technologies. 2017. Vol. 6, N 4 (90). P. 11-18. https://doi.org/10.15587/1729-4061.2017.119264.
- Skachkov V., Chepkyi V., Efimchikov A., Korkin O., Dudush A. Dynamic regularization parameter optimization of a sample estimate of the correlation matrix of observations by the criterion В«computational stability - consistencyВ». 2019 IEEE 2nd Ukraine Conference on Electrical and Computer Engineering (UKRCON), July 2-6, Lviv, Ukraine. IEEE, 2019. P. 18-23. DOI: 10.1109/ UKRCON.2019.8879946.
