|
This article is cited in 1 scientific paper (total in 1 paper)
Methods of local analysis and smoothing of time series and discrete signals
V. A. Ermolaev, Yu. A. Kropotov Murom Institute of Vladimir State University, Murom, Vladimir region
Abstract:
Regularization recovery algorithm has been developed in general smooth dependencies based on described time series and discrete signals by the sequences of generalized polynomials under the system of linearly independent functions. The coefficients of generalized polynomials are computed by recurrent way using the equation, which is obtained by least squares and local analysis methods. Verification of algorithm is performed on examples of smoothing the RTS index and the average value of the signal drift recovery. As the basic system is used a power system functions. The simulation showed the efficiency of the local method of smoothing, and established the influence of regularization parameter on stability algorithm.
Keywords:
local approximation, quadratic minimization, regularization, solution stability, restoration of drift.
Received: 08.12.2014 Revised: 18.02.2016
Citation:
V. A. Ermolaev, Yu. A. Kropotov, “Methods of local analysis and smoothing of time series and discrete signals”, Matem. Mod., 29:2 (2017), 119–132
Linking options:
https://www.mathnet.ru/eng/mm3819 https://www.mathnet.ru/eng/mm/v29/i2/p119
|
Statistics & downloads: |
Abstract page: | 299 | Full-text PDF : | 105 | References: | 48 | First page: | 12 |
|