MATHEMATICAL MODEL FOR ACOUSTIC SIGNAL IDENTIFICATION IN THE MEL-FREQUENCY CEPSTRAL DOMAIN
DOI:
https://doi.org/10.34185/1991-7848.itmm.2026.01.086Keywords:
moving object identification, mel-frequency cepstral coefficients, clustering, monitoring automation, acoustic signal, digital signal processing, industrial automation, motion phase segmentationAbstract
The implementation of intelligent monitoring systems in industrial automation and advanced transportation networks requires the development of moving object identification methods capable of stable operation under high noise levels and limited visibility. Traditional amplitude-based processing methods are sensitive to the non-stationarity of acoustic signals and changes in distance to the source, which complicates the accurate segmentation of the object's dynamic states. The objective of this work is to develop a mathematical model for motion phase identification based on Mel-frequency cepstral analysis, intelligent K-means clustering, Short-Time Fourier Transform, psychoacoustic filtering, and feature decorrelation in the cepstral domain. The effectiveness of phase segmentation for approach, transit, and departure has been experimentally confirmed with high computational speed, enabling the integration of the method into automated control and technical diagnostic systems in real-time.
References
Polyvoda, O., Rudakova, H., Kondratieva, I., Rozov, Y., & Lebedenko, Y. (2020). Digital Acoustic Signal Processing Methods for Diagnosing Electromechanical Systems. Lecture Notes in Computational Intelligence and Decision Making. ISDMCI 2019. Advances in Intelligent Systems and Computing (Vol. 1020, pp. 97–109). Springer, Cham. https://doi.org/10.1007/978-3-030-26474-1_7
Bondarenko, M. E., & Ivashchenko, H. S. (2025). Vykorystannia poslidovnosti metodiv poperednoi obrobky v systemakh holosovoi identyfikatsii [Using a sequence of preprocessing methods in voice identification systems]. Systems of Control, Navigation and Communication. Collection of Scientific Works, (2), 90–96. https://doi.org/10.26906/SUNZ.2025.2.090
Rudresh, M. D., Latha, A. S., Suganya, J., & Nayana, C. G. (2017). Performance analysis of speech digit recognition using cepstrum and vector quantization. 2017 International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT), 1–6. https://doi.org/10.1109/ICEECCOT.2017.8284580
Abdul, Z., & Al-Talabani, A. (2022). Mel Frequency Cepstral Coefficient and its Applications: A Review. IEEE Access, 10, 122136–122158. https://doi.org/10.1109/ACCESS.2022.3223444
