Entropy-Based Evaluation Method of Signal Ensembles Using LPT-τ Permutations and Markov Models.

І. Tulenko; О.  Shevchenko

doi:10.36910/6775-2524-0560-2025-61-39

І. Tulenko https://orcid.org/0009-0007-2847-483X
О. Shevchenko https://orcid.org/0000-0002-8321-3583

DOI: https://doi.org/10.36910/6775-2524-0560-2025-61-39

Keywords: telecommunication networks, cognition, signal ensembles, time domain, entropy, multiscale analysis, correlation, interference immunity, uncertainty, optimization, signal-to-noise ratio (SNR).

Abstract

The article proposes an entropy-oriented method for comprehensive assessment of the structural order of signal ensembles optimized by time domain permutations. The method is based on the integration of multiscale entropy indicators, namely permutation entropy, sample entropy and fuzzy entropy, into a unified ordering criterion that enables a quantitative evaluation of the relationship between entropy-based complexity, correlation properties and the stability of signal ensembles under complex interference conditions. To evaluate the effectiveness of the proposed entropy-oriented method, experimental modeling was carried out for two approaches to time segment permutations: the deterministic LPT tau permutation method (LPT-TP) and the forecast oriented method based on Markov models. For both approaches, a multiscale entropy analysis was performed, covering the study of signal dynamics at different time scales and the evaluation of the influence of signal-to-noise ratio variations over a wide range of levels. This made it possible to analyze how changes in transmission energy conditions affect the entropy-based complexity and structural order of signal ensembles. A comparative analysis of permutation-based approaches showed that the application of LPT tau permutations increases structural order and reduces entropy measures by an average of 12–18%, whereas the Markov based permutation method demonstrates a smoother but more stable entropy reduction of 8–15%, accompanied by a 35–40% decrease in the mean mutual correlation. The obtained results confirm the efficiency of the proposed entropy-oriented method and the feasibility of its application for assessing the interference immunity, dynamic stability and structural consistency of signal ensembles in cognitive telecommunication networks

References

1. Ahmed M. U., Mandic D. P. (2011) Multivariate multiscale entropy: A tool for complexity analysis of multichannel data, Physical Review E, vol. 84, no. 6, pp. 061918, 2011.
2. Azami H., Fernández A., and Escudero J. (2019) Multivariate multi-scale dispersion entropy of biomedical times series, Entropy, vol. 21, no. 9, pp. 913, 2019. DOI: https://doi.org/10.3390/e21090913.
3. Bandt C. and Pompe B. (2002) Permutation entropy: a natural complexity measure for time series, Physical Review Letters, vol. 88, no. 17, p. 174102, 2002.
4. Chen W. et al. (2007) Characterization of surface EMG signal based on fuzzy entropy, Neural Systems and Rehabilitation Engineering, IEEE Transactions on, vol. 15, no. 2, pp. 266–272, 2007.
5. Chen, Z., Zhou, D., & Zio, E. (2019) Hidden Markov model with auto-correlated observations for degradation modeling of manufacturing systems. Reliability Engineering & System Safety, P.224–234, 2019.