2023, issue 1, p. 35-47
Received 20.04.2023; Revised 24.04.2023; Accepted 25.04.2023
Published 28.04.2023; First Online 23.05.2023
https://doi.org/10.34229/2707-451X.23.1.4
Previous | FULL TEXT (in Ukrainian) | Next
Computer Technology for Construction ε-Solution of the Problem
Valerii Zadiraka *
, Inna Shvidchenko *
V.M. Glushkov Institute of Cybernetics of the NAS of Ukraine, Kyiv
* Correspondence: This email address is being protected from spambots. You need JavaScript enabled to view it., This email address is being protected from spambots. You need JavaScript enabled to view it.
Introduction. Issues of selection and construction of computing resources and methods of their effective use to calculate an approximate solution of the problem with the given accuracy in a limited processor time are considered.
The purpose of the article. Based on the analysis of the total error of the computational algorithm, to develop a computer technology (CT) for the construction and solution of the ε-problem and illustrate it on the problems of integrating rapidly oscillating functions and digital signal processing.
The results. The general scheme for solving problems of applied and computational mathematics using computer technology is presented.
Computer technology includes the formation of a certain set of computing resources, the method and conditions of their use to build a computing process, the result of which should be the solution of the problem.
The main attention is paid to the choice of a computer calculations model (input data about the problem, the class of problems of computational mathematics, the class of computational algorithms for calculating the solution, evaluation of quality characteristics and parameters of the computational process, computer architecture, software, restrictions on the value of quality characteristics), the construction of quality characteristic estimates, the possibility of adjusting the mathematical model, the construction of the computational process and the performance of calculations.
The issues of determining the optimal parameters of computational algorithms-programs for finding ε-solution are investigated, which are demonstrated on two classes of problems – integration of rapidly oscillating functions and correlation analysis of random processes.
Conclusions. The issues of selection and construction of computing resources and methods of their effective use to calculate an approximate solution of the problem with the given accuracy in a limited processor time are considered. The elements of computer technology for solving specific problems on the examples of problems of integration of rapidly oscillating functions and correlation analysis of random processes are applied.
Keywords: total error, computer technology, rounding error, rapidly oscillating functions.
Cite as: Zadiraka V., Shvidchenko I. Computer Technology for Construction ε-Solution of the Problem. Cybernetics and Computer Technologies. 2023. 1. P. 35–47. (in Ukrainian) https://doi.org/10.34229/2707-451X.23.1.4
References
1. Sergienko I.V., Zadiraka V.K., Lytvyn O.M. Elements of the general theory of optimal algorithms. Springer, 2021. P. 378. https://doi.org/10.1007/978-3-030-90908-6
2. Sergienko I.V., Zadiraka V.K., Babich M.D., Berezovskiy A.I., Besarab P.N., Lyudvichenko V.A. Computer technology of constructing T-efficient algorithms for computing ε-solutions of computational and applied mathematical problems. Cybernetics and systems analysis. 2002. 38. P. 837–845. https://doi.org/10.1023/A:1022987803904
3. Ivanov V.V., Babich M.D., Berezovskiy A.I., Besarab P.N., Zadiraka V.K., Lyudvichenko V.A. Characteristics of problems, algorithms and computers in software complexes of computational mathematics. Kyiv, 1984. 53 p. (Preprint/ AN Ukrainian SSR, Institute of Cybernetics; 84–36).
4. Babich M.D., Zadiraka V.K., Sergienko I.V. Computing experiment in the problem of optimization of computations. Part II. Cybernetics and systems analysis. 1999. 35. P. 221–239. https://doi.org/10.1007/BF02733470
5. Zadiraka V.K. The theory of calculating the Fourier transform. Kyiv: Naukova dumka, 1983. 216 p.
6. Zadiraka V.K., Tereshchenko A.M. Computer arithmetic of multi-bit numbers in serial and parallel computing models. Kyiv: Naukova dumka, 2021. 152 p.
7. Himich A.N., Molchanov I.N., Popov A.V., Chistyakova T.V., Yakovlev M.F. Parallel algorithms for solving problems of computational mathematics. Kyiv: Naukova dumka, 2008. 246 p.
8. Serhiienko I.V., Zadiraka V.K., Lytvyn O.M., Melnykova S.S., Nechuiviter O.P. Optimal algorithms for calculating integrals from rapidly oscillating functions and their application. T. 2. Application. Kyiv: Naukova dumka, 2011. 348 p.
9. Zadiraka V.K., Melnikova S.S. Digital signals processing. Kyiv: Naukova dumka, 1993. 294 p.
10. Babich M.D. An approximation-iteration method for solving nonlinear operator equations. Cybernetics and systems analysis. 1991. 27. P. 26–38. https://doi.org/10.1007/BF01068644
11. Zadiraka V.K., Khimich O.M., Shvidchenko I.V. Models of computer calculations. Cybernetics and computer technologies. 2022. 2. P. 38–51. https://doi.org/10.34229/2707-451X.22.2.4
ISSN 2707-451X (Online)
ISSN 2707-4501 (Print)
Previous | FULL TEXT (in Ukrainian) | Next