Fading Channel Classification Using Wavelet Transform and LDA Ensemble Learning Algorithm

Abstract :

Channel fading is a critical challenge in wireless communications, for which various solutions have been proposed.  Knowing the type of fading is essential to choosing the right solution for communication links. The fading of communications channels is classified into different forms based on the frequency selectivity and time variations. To find the channel type, a classification problem should be solved. Firstly, appropriate features are extracted from the channel response, then the classification algorithm is run on these features. In this paper, we propose a method for fading channels classification in OFDM systems that extracts features from the scalogram obtained by applying wavelet transform to the channel response, and determines the channel type using the LDA algorithm with a random subspace ensemble learning method. Wavelet transform is applied to the estimated channel response to resemble the realistic conditions. The channel response is estimated using the pilots. The proposed method is simulated to classify 8 channel types, which is the combination of four multipath models based on the COST207 standard and two Doppler shifts of 50 and 200 Hz. In simulations, the effect of various parameters on classification accuracy has been evaluated. Simulation results illustrate that the proposed method is more robust to noise and channel estimation errors in reduced pilot overhead, and its performance in these conditions decreases less than the competing methods, while extracting the smaller feature vector.

References:

1. Rappaport, T.S., Wireless communications: principles and practice. 2024: Cambridge University Press.
2. Tse, D. and P. Viswanath, Fundamentals of wireless communication. 2005: Cambridge university press.
3. Kurniawan, E., L. Zhiwei, and S. Sun. Machine learning-based channel classification and its application to IEEE 802.11 ad communications. in GLOBECOM 2017-2017 IEEE Global Communications Conference. 2017. IEEE.
4. Proakis, J.G. and M. Salehi, Contemporary communication systems using MATLAB. (No Title), 2000.
5. Haykin, S., Communication systems. 2008: John Wiley & Sons.
6. Kurniawan, E., et al. Machine learning-based channel-type identification for IEEE 802.11 ac link adaptation. in 2018 24th Asia-Pacific Conference on Communications (APCC). 2018. IEEE.
7. Wen, C.-K., et al., Channel estimation for massive MIMO using Gaussian-mixture Bayesian learning. IEEE Transactions on Wireless Communications, 2014. 14(3): p. 1356-1368.
8. Yan, D., et al., Deep reinforcement learning based resource allocation for network slicing with massive MIMO. IEEE Access, 2023.
9. Ying, R., T. Jiang, and Z. Xing. Classification of transmission environment in UWB communication using a support vector machine. in 2012 IEEE Globecom Workshops. 2012. IEEE.
10. Eskola, M. and T. Heikkilä, Classification of radio channel disturbances for industrial wireless sensor networks. Ad Hoc Networks, 2016. 42: p. 19-33.
11. Kang, Y.Y., et al. Support vector machine-based wireless channel classification for adaptive AFC in LTE downlink. in 2017 IEEE 85th Vehicular Technology Conference (VTC Spring). 2017. IEEE.
12. Lu, Z., et al. UWB Channel Classification Based on Deep-Learning for Adaptive Transmission. in 2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). 2019. IEEE.
13. Jiang, C., et al., UWB NLOS/LOS classification using deep learning method. IEEE Communications Letters, 2020. 24(10): p. 2226-2230.
14. Li, X., et al. On Wireless Channel Classification Based on CP-OFDM System. in 2021 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB). 2021. IEEE.
15. Zaki, A., et al., 5G and beyond: Channel classification enhancement using VIF-driven preprocessing and machine learning. Electronics, 2023. 12(16): p. 3496.
16. Baldini, G. and F. Bonavitacola. Fading Channel Classification Based on Convolutional Neural Network with General Linear Chirplet Transform. in 2023 17th European Conference on Antennas and Propagation (EuCAP). 2023. IEEE.
17. Yu, Y., F. Liu, and S. Mao, Fingerprint extraction and classification of wireless channels based on deep convolutional neural networks. Neural processing letters, 2018. 48(3): p. 1767-1775.
18. ShirinAbadi, P.A. and A. Abbasi. UWB channel classification using convolutional neural networks. in 2019 IEEE 10th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON). 2019. IEEE.
19. Deng, R., et al., Channel fingerprint based beam tracking for millimeter wave communications. IEEE Communications Letters, 2019. 24(3): p. 639-643.
20. Proakis, J.G. and M. Salehi, Digital communications. 2008: McGraw-hill.
21. Pätzold, M., M. Patzold, and M. Paetzold, Mobile fading channels. 2002: Wiley Online Library.
22. Ho, T.K., The random subspace method for constructing decision forests. IEEE transactions on pattern analysis and machine intelligence, 1998. 20(8): p. 832-844.
23. Molisch, A.F., Wireless communications. Vol. 34. 2012: John Wiley & Sons.
24. Breiman, L., Classification and regression trees. 2017: Routledge.
25. Loh, W.Y., Classification and regression trees. Wiley interdisciplinary reviews: data mining and knowledge discovery, 2011. 1(1): p. 14-23.
26. Breiman, L., Random forests. Machine learning, 2001. 45: p. 5-32.