350 rub
Journal Radioengineering №1 for 2022 г.
Article in number:
Possibilities of application of nuclear functions for approximation of two-dimensional densities of probability distribution of parameters of random signals
Type of article: scientific article
DOI: https://doi.org/10.18127/j00338486-202201-03
UDC: 004.93'1; 51-74; 519.254
Keywords: Signal calculations detection recognition two-dimensional density distributions estimation of distribution parameters approximation
Authors:

L.I. Dvoiris, I.N. Kryukov

Abstract:

Statement of the problem. Detection/recognition of signals from various objects and interference is traditionally associated with the estimation of the distribution densities of the probabilities of signal parameters. Currently, methods of nonparametric evaluation of signals in various subject areas, including in technical systems of protection and security of objects, are actively developing. The possibility of using one-dimensional nonparametric estimates of signals is considered in [1, 2], however, real signals are often multidimensional.

Goal. To consider the possibility of using methods of nonparametric estimation of two-dimensional densities of probability distributions of signal parameters in object detection/recognition tasks.

Results. A method of nonparametric estimates of the probability distribution densities of signal characteristics in the interests of detection/recognition is presented. Its practical application for the case of two-dimensional densities of probability distributions is considered.

Practical significance. The use of the method of multidimensional nonparametric estimation of the probability distribution densities of signal characteristics and interference makes it possible in the future to increase the noise immunity of object detection and recognition systems in various applied fields.

Pages: 16-20
For citation

Dvoiris L.I., Kryukov I.N. Possibilities of application of nuclear functions for approximation of two-dimensional densities of probability distribution of parameters of random signals. Radiotekhnika. 2022. V. 86. № 1. P. 16−20. DOI: https://doi.org/10.18127/j00338486-202201-03 (In Russian)

References
  1. Rasin D. Neparametricheskaja jekonometrika: vvodnyj kurs. Kvantil'. 2008. № 4 (In Russian).
  2. Scott D.W. Multivariate Density Estimation and Visualisation. Papers. No. 2004. 16. Humboldt-Universität zu Berlin. Center for Applied Statistics and Economics (CASE), Berlin.
  3. Weglarczyk S. Kernel density estimation and its application. ITM Web of Conferences 23. 00037 (2018) XLVIII Seminar of Applied Mathematics.
Date of receipt: 10.11.2021
Approved after review: 18.11.2021
Accepted for publication: 14.12.2021