The task of comparison testing of time series power spectrum parametric estimates is considered. It is shown that an urgent problem that arises in solving it is the optimization of the parameters of spectral estimates under conditions of small samples of observations. To overcome this problem, it is proposed to use the system-wide concept of analysis through time series synthesis. Based on this concept, a regular method was developed for comparison testing of parametric estimates of the power spectrum that are obtained from a time series of finite duration. Within the method, solutions are accepted based on the results of testing statistical hypotheses about the homogeneity of two samples: the finite empirical one, compiled based on the results of observations, and an infinite virtual one, mathematically synthesized according to each individual parametric estimate in the series of spectral alternatives being examined. The principle of minimum informational divergence between samples according to the Kullback–Leibler criterion. An example of the practical application of the developed method is presented in the problem of discrete spectral modeling of speech signals. The ability of the method to identify patterns of unstable parametric estimates of the autoregressive type has been shown. The results obtained are intended for use in the field of speech acoustics, as well as technical and medical diagnostics, where parametric methods of spectral analysis are increasingly used in practice.
Similar content being viewed by others
Notes
QATestlab (site) URL:https://blog.qatestlab.com/2012/10/01/types-of-software-testing-comparison-testing/. (Date accessed May 17, 2023).
Since moments no greater than second order are used in spectral analysis of a random process, the assumption regarding the Gaussian distribution law of sample Xr does not limit in any manner the generality of the study performed later.
References
V. Kreinovich, G. P. Dimuro, and A. C. da Rocha Costa, A General Description of Measurement Devices: First Step — Finite Set of Possible Results, in: From Intervals to-?, Stud. Comp. Intell., Springer, Cham., 1041, pp. 9–22 (2023), https://doi.org/10.1007/978-3-031-20569-9_3.
S. I. Marple, Digital Spectral Analysis with Applications, 2nd ed., Mineola, Dover Publications, New York, (2019).
D. Tjøstheim, H. Otneim, and B. Støve, Time Series Dependence and Spectral Analysis, in Statistical Modeling Using Local Gaussian Approximation, pp. 261–299, Academic Press, (2022), https://doi.org/10.1016/B978-0-12-815861-6.00015-8.
V. V. Savchenko, Radiophys. Quantum El., 58, No. 5, 373–379 (2015), https://doi.org/10.1007/s11141-015-9611-4.
Sh. Ando, J. Acoust. Soc. Am., 146, 2846 (2019), https://doi.org/10.1121/1.5136873.
L. Shen and E. A. Maharaj, Comput. Stat. Data Analysis, 60, 32–49 (2013), https://doi.org/10.1016/j.csda.2012.11.014.
R. Shiavi, Random Signal Modeling and Parametric Spectral Estimation, in Biomedical Engineering, Introduction to Applied Statistical Signal Analysis, 3rd ed., Academic Press, pp. 287–330 (2007), https://doi.org/10.1016/B978-012088581-7/50025-X.
C. Candan, Signal Process., 166, Article ID 107256 (2020), https://doi.org/10.1016/j.sigpro.2019.107256.
S. Cui, E. Li, and X. Kang, IEEE Int. Conf. Multimedia and Expo (ICME), 1–6, London, UK (2020), https://doi.org/10.1109/ICME46284.2020.9102765.
V. V. Savchenko, Improvement of the Methodology of the Measurement of an Indicator of the Precision of the Autoregressive Model of a Speech Signal, Izmer. Tekh., No. 10, 58–63 (2022), https://doi.org/10.32446/0368-1025it.2022-10-58-63.
K. V. Mishra, M. Cho, A. Kruger, and W. Xu, IEEE Trans. Signal Proc., 63, No. 20, 5342–5357 (2015), https://doi.org/10.1109/TSP.2015.2452223.
W. Liao and A. Fannjiang, Appl. Comput. Harmon. Analysis, 40, No. 1, 33–67 (2016), https://doi.org/10.1016/j.acha.2014.12.003.
T. C. Mills, Schuster, Beveridge, and Periodogram Analysis, in: The Foundations of Modern Time Series Analysis, pp. 18–29. Palgrave Advanced Texts in Econometrics Series, Palgrave Macmillan, London, (2011), https://doi.org/10.1057/9780230305021_3.
A. I. Wiesman, J. S. Castanheira, and S. Baillet, Neuroimage, 247, Article ID 118823 (2022), https://doi.org/10.1016/j.neuroimage.2021.118823.
V. V. Savchenko and A. V. Savchenko, Radioelectron. Commun. Syst., 62, No. 5, 276–286 (2019), https://doi.org/10.3103/S0735272719050042.
A. A. Borovkov, Mathematical Statistics, Lan’, St. Petersburg (2010).
R. Aichinger and F. Pernkopf, IEEE/ACM Trans. Audio Speech, 29, 914–926 (2021), https://doi.org/l0.1109/TASLP.2021.3053387.
V. V. Savchenko, The Hybrid Technique of Spectral Analysis of Speech Signals Based on an Autoregressive Model and the Schuster Periodogram, Izmer. Tekh., No. 3, 61–66 (2023), https://doi.org/10.32446/0368-1025it.2023-3-61-66.
A. D. Chikmarev, Comparison of Identifi cation Criteria for Mathematical Models in Solving Measurement Problems, Izmer. Tekh., No. 8, 41–45 (2022), https://doi.org/10.32446/0368-1025it.2022-8-41-45.
V. V. Savchenko, The Itakury–Saito Divergence as an Element of the Information Theory of Speech Perception, Radiotekh., Élektron., 64, No. 6, 585–592 (2019), https://doi.org/10.1134/S0033849419060093.
S. Kullback, Information Theory and Statistics, Dover Publications, New York (1997).
V. V. Savchenko, Radiophys. Quantum Electron., 36, No. 11, 763–768 (1993), https://doi.org/10.1007/BF01039709.
B. Wei and J. D. Gibson, IEEE Signal Proc. Lett., 10, No. 4, 101–103 (2003), https://doi.org/10.1109/LSP.2003.808550.
R. Gray, A. Buzo, A. Gray, and Y. Matsuyama, IEEE Trans. Acoust. Speech, 28, No. 4, 367–376 (1980), https://doi.org/10.1109/TASSP.1980.1163421.
A. V. Savchenko and V. V. Savchenko, Radioelectron. Commun. Syst., 64, No. 6, 300–309 (2021), https://doi.org/10.3103/S0735272721060030.
V. V. Savchenko and L V. Savchenko, A Method of Autoregressive Moding of a Speech Signal Based on the Discrete Fourier Transform and a Scale-Invariant Standard of Informational Divergence, Radiotekh. Élektron., 66. No. 11, 1100–1108 (2021), https://doi.org/10.31857/S0033849421110085.
V. V. Savchenko, A Method of Autoregressive Moding of a Speech Signal Using the Envelope of the Schuster Periodogram as the Reference Spectral Sample, Radiotekh. Élektron., 68, No. 2, 138–145 (2023), https://doi.org/10.31857/S0033849423020122.
A. Kazemipour, S. Miran, P. Pal, B. Babadi, and M. Wu, IEEE Trans. Signal Proc., 65, No. 9, 2333–2347 (2017), https://doi.org/10.1109/TSP.2017.2656848.
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated from Izmeritel'naya Tekhnika, No. 6, pp. 56–62, June, 2023. https://doi.org/10.32446/0368-1025it.2023-6-56-62.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Savchenko, V.V. Method for Comparison Testing of Parametric Power Spectrum Estimates: Spectral Analysis Via Time Series Synthesis. Meas Tech 66, 430–438 (2023). https://doi.org/10.1007/s11018-023-02244-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11018-023-02244-3