МЕТОД ОБРАБОТКИ ИНФОРМАЦИИ ДЛЯ АНАЛИЗА СТАТИСТИЧЕСКИ НЕРАЗЛИЧИМЫХ ФУНКЦИЙ НА ОСНОВЕ МОДОВОГО РАЗЛОЖЕНИЯ В ОРТОГОНАЛЬНОМ БАЗИСЕ АДАМАРАINFORMATION PROCESSING METHOD FOR THE ANALYSIS OF STATISTICALLY INDISTINGUISHABLE FUNCTIONS BASED ON MODAL DECOMPOSITION

Authors

  • Nikolaj V. Belyi Saint Petersburg State University of Aerospace and Instrumentation image/svg+xml
  • Mihail B. Sergeev Saint Petersburg State University of Aerospace and Instrumentation image/svg+xml

DOI:

https://doi.org/10.17308/sait/1995-5499/2026/2/160-174

Keywords:

Hadamard matrices, statistically indistinguishable functions, orthogonal basis, local decomposition, anomalous component, latent fingerprint

Abstract

Current signal analysis methods are unable to detect hidden structural differences between deterministic functions that are statistically indistinguishable in terms of mean, variance, correlation, and energy. While these functions share identical global dynamics, they contain subtle local modifications that remain invisible to conventional integral statistics. Traditional approaches — including principal component analysis — fail to resolve these distinctions, as local features are systematically suppressed by the dominant global component. To address this limitation, we introduce a novel method based on local signal decomposition using an orthogonal basis derived from Hadamard matrices. The signal is partitioned into short temporal segments, each transformed into a set of latent components that capture local waveform morphology. By accumulating statistical properties of these components over time, persistent yet weak structural deviations — undetectable by global measures — are revealed. We further define an anomalous component, the latent feature most divergent from the dominant global pattern, yet statistically inert at the integral level. This component serves as a unique, intrinsic fingerprint of individual function identity, enabling discrimination without cross-comparison. The method requires no training data, makes no assumptions about signal form, and avoids reliance on moment-based statistics. Experiments on synthetic data demonstrate high-fidelity detection of hidden structural patterns inaccessible to classical techniques. The approach holds significant promise for biomedical, financial, and engineering applications, where individual signatures are diagnostically critical but masked by statistical equivalence.

Author Biographies

  • Nikolaj V. Belyi, Saint Petersburg State University of Aerospace and Instrumentation

    1st-year postgraduate student of Department №44 SUAI

  • Mihail B. Sergeev, Saint Petersburg State University of Aerospace and Instrumentation

    DSc in Technical Sciences, Professor, Head of the Department № 44 of SUAI

References

1. International Encyclopedia of Statistical Science / ed. by M. Lovric. – 2nd ed. – Berlin, Heidelberg : Springer, 2025. – LXXVI, 2917 p. – (Mathematics and Statistics; Reference Module Computer Science and Engineering). – DOI: 10.1007/978-3-662-69359-9.

2. Acar-Denizli N., Delicado P. Functional data analysis for wearable sensor data: a systematic review. AStA Adv Stat Anal 109. –2025. – P. 591–631. https://doi.org/10.1007/s10182-025- 00531-8

3. Sözen Ç., Sağlam F., Sözen M. Comparative Study of Missing Data Imputation Methods in Functional Data Analysis. Iranian Journal of Science and Technology, Transactions A: Science. – 2025. https://doi.org/10.1007/s40995- 025-01858-2

4. Saeidi H., Aminghafari M. FunCWM: Functional Cluster-Weighted Models and TheirApplications to Climatological Functional Data. J Classif. – 2026. https://doi.org/10.1007/s00357- 025-09530-5

5. Valla R., Mozharovskyi P., d’Alché-Buc F. Anomaly-Driven Visualization of Functional Data. In: Aneiros G., Bongiorno E. G., Goia A., Hušková M. (eds) New Trends in Functional Statistics and Related Fields. IWFOS 2025. Contributions to Statistics. Springer, Cham. https://doi. org/10.1007/978-3-031-92383-8_64

6. Cleophas T. J. Functional Data Analysis (FDA) Including Regularized Regressions / T. J. Cleophas, A. H. Zwinderman // Application of Regularized Regressions to Identify Novel Predictors in Clinical Research. – Cham : Springer. – 2024. – Ch. 17. – DOI: 10.1007/978-3-031- 72247-9_17.

7. Сергеев М. Б., Белая Т. И., Белый Н. В. Биометрическая идентификация на основе искусственного интеллекта: разработка с использованием нейронных дифференциальных уравнений // Прикладной искусственный интеллект: перспективы и риски: II Международная научная конференция (Санкт-Петербург, 21 октября 2025 г.): Сборник статей. Санкт-Петербург : ГУАП, 2025. – C. 293–296.

8. Gunning E., Golovkine S., Simpkin A. J. [et al.] Analysing kinematic data from recreational runners using functional data analysis. Comput Stat 40. – 2025. – P. 1825–1847. https://doi. org/10.1007/s00180-024-01591-1

9. Bamdad S. Advancing sustainability in complex systems: integrating environmental strategy with statistical modelling. Ann Oper Res. – 2025. https://doi.org/10.1007/s10479-025- 06878-0

10. Nair S. B., Al-Hemyari Z. A., Gountia D. Decidable Regression Techniques for Statistical Modelling with Sustainable Agriculture Operations. SN COMPUT. SCI. 5, 1159. 2024. https:// doi.org/10.1007/s42979-024-03518-5

11. Statistical Modelling for Evidence-Based Public Health Statistical Modelling for Evidence-Based Public Health : Case Studies in HIV/ AIDS Epidemiology and Child and Maternal Health / L. N. Kazembe, T. M. Kaombe (eds.). – Cham : Springer, 2025. – XXV, 551 p. – (Emerging Topics in Statistics and Biostatistics). – DOI: 10.1007/978-3-031-97281-2.

12. Qu Z., Dai W., Euan C. [et al.] Exploratory functional data analysis. TEST 34. – 2025. – P. 459–482. https://doi.org/10.1007/s11749-024- 00952-8

13. Developments in Statistical Modelling / J. Einbeck, H. Maeng, E. Ogundimu, K. Perrakis (eds.). – Cham : Springer, 2024. – X, 270 p. – (Contributions to Statistics). – DOI: 10.1007/978- 3-031-65723-8.

14. Балонин Н. А., Сергеев А. М. Матрицы Адамара как результат произведения Скарпи без циклического смещения блоков // Информационно-управляющие системы. – 2022. – № 3. – C. 2–8.

15. Manjhi P. K. Embedding into Hadamard matrices. Indian J Pure Appl Math 56. – 2025. – P. 1110–1116. https://doi.org/10.1007/s13226- 025-00826-1

16. Sano T. Positive definiteness of Hadamard exponentials and Hadamard inverses. Positivity 28. – 2024. – 52. https://doi.org/10.1007/s11117- 024-01070-3

17. Fewster C. J. Hadamard States for Decomposable Green-Hyperbolic Operators. Commun. Math. Phys. 407. – 2026. – 14. https://doi. org/10.1007/s00220-025-05512-1

18. Shi M., Lu D., Armario J. A. [et al.] Hadamard bent sequences and spherical designs. J Algebr Comb 62. – 2025. – 43. https://doi. org/10.1007/s10801-025-01463-x

19. Morales E. Boundedness of weighted Hadamard — Bergman and weighted variable Hadamard — Bergman convolution operators. Bol. Soc. Mat. Mex. 30. – 2024. – 70. https://doi. org/10.1007/s40590-024-00643-2

20. Zhao C. J. The Hadamard Inequality for Log-Convex Functions. Math Notes 118. – 2025. – P. 871–877. https://doi.org/10.1134/ S0001434625605301

21. Ma L., Shu Y. Lyapunov-Schmidt reduction for Caputo-Hadamard fractional differential system. Nonlinear Dyn 113. – 2025. – P. 29107–29122. https://doi.org/10.1007/s11071-025-11651-w

22. Karapetyants A., Morales E. Boundedness of Hadamard — Bergman and Variable Hadamard — Bergman Convolution Operators. Math Notes 114. – 2023. – P. 804–817. https://doi. org/10.1134/S0001434623110160

23. Suma D., Narendra V. G., Holla M. R. [et al.] Hadamard defense mechanisms for rice grain image privacy. Discov Appl Sci 7. – 2025. – 725. https://doi.org/10.1007/s42452-025-07340-3

24. Malik D., Jabeen Z. Hermite — Hadamard Type Inequalities Involving Fractional Integrals of Exponentially Convex Functions. Ukr Math J 77. – 2025. – P. 802–813. https://doi.org/10.1007/ s11253-025-02491-3

25. Raickwade P. R., Hsu C. C., Sivakumar K. C. Generalized inverses of linear preservers of Hadamard majorization. J Anal. – 2025. https://doi. org/10.1007/s41478-025-00979-6

26. Hsu C. C., Raickwade P. R., Sivakumar K. C. Generalized Inverse Preservers of Hadamard Circulant Majorization. Indian J Pure Appl Math. – 2025. – 56. – P. 883–890. https://doi.org/10.1007/ s13226-025-00808-3

27. Elmi M. Hadamard products and BPS networks. J. High Energ. Phys. – 2024. – 76. https:// doi.org/10.1007/JHEP07(2024)076

Published

2026-06-30

Issue

Section

Intelligent Information Systems, Data Analysis and Machine Learning

How to Cite

Belyi, N. V., & Sergeev, M. B. (2026). МЕТОД ОБРАБОТКИ ИНФОРМАЦИИ ДЛЯ АНАЛИЗА СТАТИСТИЧЕСКИ НЕРАЗЛИЧИМЫХ ФУНКЦИЙ НА ОСНОВЕ МОДОВОГО РАЗЛОЖЕНИЯ В ОРТОГОНАЛЬНОМ БАЗИСЕ АДАМАРАINFORMATION PROCESSING METHOD FOR THE ANALYSIS OF STATISTICALLY INDISTINGUISHABLE FUNCTIONS BASED ON MODAL DECOMPOSITION . Proceedings of Voronezh State University. Series: Systems Analysis and Information Technologies, 2, 160-174. https://doi.org/10.17308/sait/1995-5499/2026/2/160-174