DOI
10.34229/KCA2522-9664.25.4.15
UDC 004.93+621.396
Application of machine learning methods to certain problems
of digital signal processing in telecommunication tasks
Abstract. The article explores the application of machine learning methods to digital signal processing tasks in telecommunication systems. Specifically, it examines the problems of automatic modulation classification and subsequent signal demodulation. For automatic modulation classification, four machine learning methods are investigated: multinomial regression, nearest neighbors’ method, Gaussian mixture modeling, and a convolutional neural network. Experimental results on artificial data demonstrated recognition accuracy for five modulation types ranging from 96% to 99%. The highest accuracy (99%) was achieved by the convolutional neural network. However, the three other methods, which have a simpler structure (and were not considered in previous works), show a satisfactory trade-off between accuracy and implementation complexity. Verification on 89 signals from real modems showed that the nearest neighbors’ method achieves the highest classification accuracy (100%), while the remaining methods provide accuracy at the level of 99%. This indicates that high classification accuracy can be achieved using significantly simpler methods compared to convolutional neural networks. The paper also proposes a method of block demodulation of signals based on multinomial linear regression and a feedforward neural network, which has a simpler practical implementation compared to other known methods. It is shown that at high noise levels, the proposed method provides higher signal recovery accuracy compared to the traditional demodulation method based on Gardner and Costas loops, and also uses fewer parameters compared to other known methods.
Keywords: machine learning, deep learning, digital signal processing, automatic modulation classification, signal demodulation.
full text
REFERENCES
- 1. Rembovsky A.M., Ashikhmin A.V., Kozmin V.A. Radio Monitoring: Tasks, Methods, Tools. 4th ed. corrected [in Russian]. Moscow: Goryachaya Liniya – Telecom, 2015. 640 c.
- 2. Perunov Yu.M., Kupriyanov A.I. Methods and techniques of electronic warfare [in Russian]. Moscow: Infra-Engineering, 2021. 376 p.
- 3. Sergienko A.B. Digital Communications [in Russian]. SPb: ETU "LETI", 2012. 164 p.
- 4. Shermeh A.E., Ghazalian R. Recognition of communication signal types using genetic algorithm and support vector machines based on the higher order statistics. Digital Signal. Proc. 2010. Vol. 20, N 6. P. 1748–1757. https://doi.org/10.1016/.
- 5. Hossen A., Al-Wadahi F., Jervase J.A. Classification of modulation signals using statistical signal characterization and artificial neural networks. Engineering Applications of Artificial Intelligence. 2007. Vol. 20, N 4. P. 463–472. https://doi.org/10.1016/.
- 6. Hassan K., Dayoub I., Hamouda W., Berbineau M. Automatic modulation recognition using wavelet transform and neural networks in wireless systems. EURASIP J. Adv. Signal Process. 2010. Vol. 2010, N 1. Article number 532898. https://doi.org/10.1155/.
- 7. Velampalli C. Hierarchical blind modulation classification in the presence of carrier frequency offset. Master’s Thesis. Hyderabad: Communications Research Center, 2010. P. 1–39.
- 8. Su W., Kosinski J.A. Higher-order blind signal feature separation: An enabling technology for battlefield awareness. U. S. Army CECOM Research Development and Engineering Center, 2006. 7 p.
- 9. Kozminchuk B.W., Huang X. Joint blind synchronization of M-PSK and M-QAM signals. Defense Research Establishment. Ottawa, 1996. 13 p.
- 10. Boiteau D., Le Martret C. Combinaison de moments d’ordre 2 et 4 pour la classification de modulations linБaires. Seizieme Colloque GRETSI. 1997. Р. 981–984.
- 11. Swami A., Sadler B.M. Hierarchical digital modulation classification using cumulants. IEEE Transactions on Communications. 2000. Vol. 48, N. 3. P. 416–429. https://doi.org/10.1109/ 26.837045.
- 12. Li J., Wang J., Fan X., and Zhang Y. Automatic digital modulation recognition using feature subset selection. PIERS Proceedings (March 24–28, 2008, Hangzhou, China). 2008. Part 1. P. 351–354.
- 13. Kurbanaliev V. K. Cumulative signs for determining the type of signal manipulation. Radio electronics. Nanosystems. Information technologies. 2020. Vol. 12, N 3. P. 319–328. https://doi.org/10.17725/.
- 14. Nagornyuk O.A. Using cumulative analysis for recognition of radio signals of digital telecommunication systems. Systemy obrobky informatsiyi. 2018. N 2(153). P. 136–143. https://doi.org/10.30748/.
- 15. Nagornyuk O.A., Pysarchuk O.O., Manoilov V.P. A method of automated recognition of the type of digital linear modulation based on cumulative signal analysis. Bulletin of the ZhDTU. Ser. "Technical Sciences" 2013. Vol. 2 (65). P. 67–76. https://doi.org/10.26642.
- 16. Gaikwad C.J., Samdani H.K., Sircar P. Signal parameter estimation using fourth order statistics: multiplicative and additive noise environment. SpringerPlus. 2015. Vol. 4, Article number 291. https://doi.org/10.1186/.
- 17. Savit R., Green M.. Time series and dependent variables. Physica D: Nonlinear Phenomena. 1991. Vol. 50, N 1. P. 95–116. https://doi.org/10.1016/.
- 18. Kostenko P.Yu., Slobodyanyuk V.V., Alyonkin M.I., Shapovalov O.V. Application of the predictability index to estimate the signal delay time under conditions of multiplicative interference with unknown probability distribution density. Collection of scientific papers of the Kharkiv National University of the Air Force. 2024. Т 1(79). З. 52–59. https://doi.org/10.30748/.
- 19. Guven G. Testing equality of means in one-way ANOVA using three and four moment approximations. Communications Faculty of Sciences University of Ankara. Ser. A1 Mathematics and Statistics. 2023. Vol. 72, N 3. P. 587–605. https://doi.org/10.31801/.
