KABLOSUZ İLETİŞİM SİSTEMLERİ İÇİN MAKİNA ÖĞRENİMİ DESTEKLİ ALTERNATİF SEZİCİ TASARIMI

Abstract

Son yıllarda derin öğrenme (deep learning-DL) teknikleri fiziksel seviye iletişim sistemlerinde yaygın olarak kullanılmaktadır. DL teknikleri, modern haberleşme sistemlerindeki işlem karmaşıklığını azaltmasından ve daha iyi başarım sağlaması gibi nedenlerden dolayı hali hazırda var olan haberleşme yöntemlerine alternatif seçenekler sunmaktadır. Bu çalışmada, Rayleigh sönümlemeli kanalda ikili faz kaydırmalı anahtarlama (binary phase shift keying-BPSK) veya dördün faz kaydırmalı anahtarlama (quadrature phase shift keying-QPSK) modülasyonu kullanılması durumunda alıcıya ulaşan işaretin işaret yıldız kümesi görüntüsünden, gönderilen işaretin DL ile kestirimi hedeflenmiştir. DL tekniklerinden olan evrişimli sinir ağı (convolutional neural network -CNN) girişine alıcıya gelen işaretin ve denkleştirilmiş işaretin işaret yıldız kümesi görüntüsü uygulanmıştır. CNN sınıflandırıcı ile bulunan sistemin hata başarımları klasik en büyük olabilirlikli sezici (maximum likehood-ML) başarımları ile karşılaştırılmıştır. İşaret yıldız kümesinde farklı boyutlarda bölgeler seçilmiştir. Bu bölgelerin her biri ayrı senaryo olarak değerlendirilir. Belirli senaryolar altında bu bölgelerin CNN sınıflandırıcı ile elde edilen hata başarımları ile ML hata başarımları ile benzer çıktığı görülmüştür.

Keywords

• CNN
• QPSK
• BPSK
• ML
• Derin Öğrenme

MACHINE-LEARNING AIDED ALTERNATIVE DETECTOR DESIGN FOR WIRELESS COMMUNICATIONS SYSTEMS

Abstract

In recent years, deep learning (DL) techniques are widely used for physical layer solutions in communication systems. DL techniques offer alternative options to existing communication methods since they can reduce the computational complexity and provide better performance in modern communication systems. In this study, we propose a DL-aided signal detections for BPSK (Binary Phase Shift Keying) or QPSK (Quadrature Phase Shift Keying) modulation over Rayleigh fading channel where the DL-aided detection is performed based on the constellation diagram image of the received signal. The constellation diagram image of received signal and the equalized signal are given as inputs to the convolutional neural network (CNN), which is one of the commonly used DL techniques, Regions of different sizes are selected in the constellation diagram. Each of these regions is considered as different scenarios. The error performance of the system obtained with the CNN classifier is compared with the classical maximum likelihood (ML) detector performance. Under certain scenarios, it has been revealed that the DL-aided signal detection could achieve the performance of the ML detector which shows the effectiveness of the proposed solution.

Keywords

• CNN
• QPSK
• BPSK
• ML
• Machine Learning

References

1. Albawi, S., Mohammed, T. A., & Al-Zawi, S,2018. Understanding of a convolutional neural network. Proceedings of 2017 International Conference on Engineering and Technology, ICET 2017, 2018-January, 1–6. https://doi.org/10.1109/ICEngTechnol.2017.8308186
2. Alrabeiah, M., & Alkhateeb, A., 2020. Deep Learning for mmWave Beam and Blockage Prediction Using Sub-6 GHz Channels. IEEE Transactions on Communications, 68(9), 5504–5518. https://doi.org/10.1109/TCOMM.2020.3003670.
3. Chen, L., Zhou, M., Su, W., Wu, M., She, J., & Hirota, K. ,2018. Softmax regression based deep sparse autoencoder network for facial emotion recognition in human-robot interaction. Information Sciences, 428, 49–61. https://doi.org/10.1016/j.ins.2017.10.044.
4. Cun, Y. Le, Guyon, I., Jackel, L. D., Henderson, D., Boser, B., Howard, R. E., Denker, J. S., Hubbard, W., & Graf, H. P. ,1989. Handwritten Digit Recognition: Applications of Neural Network Chips and Automatic Learning. IEEE Communications Magazine, 27(11), 41–46. https://doi.org/10.1109/35.41400.
5. Dorner, S., Cammerer, S., Hoydis, J., & Brink, S. Ten.,2018. Deep Learning Based Communication over the Air. IEEE Journal on Selected Topics in Signal Processing, 12(1), 132–143. https://doi.org/10.1109/JSTSP.2017.2784180.
6. Emir, A., Kara, F., & Kaya, H., 2019. Deep learning-based joint symbol detection for NOMA. In 27th Signal Processing and Communications Applications Conference, SIU 2019. https://doi.org/10.1109/SIU.2019.8806600.
7. Farsad, N., & Goldsmith, A., 2018. Neural network detection of data sequences in communication systems. IEEE Transactions on Signal Processing, 66(21), 5663–5678. https://doi.org/10.1109/TSP.2018.2868322.
8. Gui, G., Huang, H., Song, Y., & Sari, H., 2018. Deep Learning for an Effective Nonorthogonal Multiple Access Scheme. IEEE Transactions on Vehicular Technology, 67(9), 8440–8450. https://doi.org/10.1109/TVT.2018.2848294.

9. Hojatian, H., Nadal, J., Frigon, J. F., & Leduc-Primeau, F., 2020. Unsupervised Deep Learning for Massive MIMO Hybrid Beamforming. In arXiv: https://arxiv.org/pdf/2007.00038.pdf
10. Jiajia Guo, Chao-Kai Wen & Shi Jin ,2020. Deep Learning-Based CSI Feedback for Beamforming in Single- and Multi-cell Massive MIMO Systems. IEEE Journal on Selected Areas in Communications. Submitted 11 Nov 2020.In arXiv: https://arxiv.org/pdf/2011.06099.pdf
11. Jiang, R., Sun, C., Zhang, L., Tang, X., Wang, H., & Zhang, A., 2020. Deep learning aided signal detection for SPAD-Based underwater optical wireless communications. IEEE Access, 8, 20363–20374. https://doi.org/10.1109/ACCESS.2020.2967461.
12. Lin, C., Chang, Q., & Li, X.,2019. A deep learning approach for mimo-noma downlink signal detection. Sensors (Switzerland). https://doi.org/10.3390/s19112526.
13. Luo, B., Peng, Q., Cosman, P. C., & Milstein, L. B.,2019. Robustness of Deep Modulation Recognition under AWGN and Rician Fading. In Conference Record - Asilomar Conference on Signals, Systems and Computers (Vol. 2018-October, pp. 447–450). https://doi.org/10.1109/ACSSC.2018.8645089.
14. Luong, T. Van, Ko, Y., Vien, N. A., Nguyen, D. H. N., & Matthaiou, M.,2019. Deep Learning-Based Detector for OFDM-IM. IEEE Wireless Communications Letters, 8(4), 1159–1162. https://doi.org/10.1109/LWC.2019.2909893.
15. Peng, S., Jiang, H., Wang, H., Alwageed, H., & Yao, Y. D., 2017. Modulation classification using convolutional Neural Network based deep learning model. In 2017 26th Wireless and Optical Communication Conference, WOCC 2017. https://doi.org/10.1109/WOCC.2017.7929000.
16. Sledevic, T.,2019. Adaptation of Convolution and Batch Normalization Layer for CNN Implementation on FPGA. 2019 Open Conference of Electrical, Electronic and Information Sciences, EStream 2019 - Proceedings. https://doi.org/10.1109/eStream.2019.8732160.
17. Soltani, M., Pourahmadi, V., Mirzaei, A., & Sheikhzadeh, H.,2019. Deep Learning-Based Channel Estimation. IEEE Communications Letters, 23(4), 652–655. https://doi.org/10.1109/LCOMM.2019.2898944.
18. Wang, X., Zhang, Y., Shen, R., Xu, Y., & Zheng, F. C.,2020. DRL-Based Energy-Efficient Resource Allocation Frameworks for Uplink NOMA Systems. IEEE Internet of Things Journal, 7(8), 7279–7294. https://doi.org/10.1109/JIOT.2020.2982699.
19. Wu, N., Wang, X., Lin, B., & Zhang, K.,2019. A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems. IEEE Access, 7, 110197–110204. https://doi.org/10.1109/access.2019.2926843.
20. Xiao, L., Sheng, G., Liu, S., Dai, H., Peng, M., & Song, J., 2019. Deep Reinforcement Learning-Enabled Secure Visible Light Communication against Eavesdropping. IEEE Transactions on Communications, 67(10), 6994–7005. https://doi.org/10.1109/TCOMM.2019.2930247.
21. Xu Y, Yang C, Hua M and Zhou W.,2020. Deep Deterministic Policy Gradient (DDPG)-Based Resource Allocation Scheme for NOMA Vehicular Communications. IEEE Access.2020; 8: 18797-18807.doi: 10.1109/ACCESS.2020.2968595.
22. Xu, W., Zhong, Z., Beery, Y., You, X., & Zhang, C., 2018. Joint neural network equalizer and decoder. In Proceedings of the International Symposium on Wireless Communication Systems (Vol. 2018-August). https://doi.org/10.1109/ISWCS.2018.8491056.
23. Ye, H., Li, G. Y., & Juang, B. H., 2018. Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems. IEEE Wireless Communications Letters, 7(1), 114–117. https://doi.org/10.1109/LWC.2017.2757490.
24. Ye, H., Liang, L., Li, G. Y., & Juang, B. H., 2020. Deep Learning-Based End-to-End Wireless Communication Systems with Conditional GANs as Unknown Channels. IEEE Transactions on Wireless Communications, 19(5), 3133–3143. https://doi.org/10.1109/TWC.2020.2970707.
25. Yuan, J., Ngo, H. Q., & Matthaiou, M., 2020. Machine Learning-Based Channel Prediction in Massive MIMO with Channel Aging. IEEE Transactions on Wireless Communications, 19(5), 2960–2973. https://doi.org/10.1109/TWC.2020.2969627.
26. Zhang, J., Tao, X., Wu, H., Zhang, N., & Zhang, X., 2020. Deep Reinforcement Learning for Throughput Improvement of the Uplink Grant-Free NOMA System. IEEE Internet of Things Journal, 7(7), 6369–6379. https://doi.org/10.1109/JIOT.2020.2972274.
27. Zhenyu Liu, Mason del Rosario & Zhi Ding, 2020. A Markovian Model-Driven Deep Learning Framework for Massive MIMO CSI Feedback. IEEE Transactions on Wireless Communications. Submitted 20 Sep 2020. In arXiv: https://arxiv.org/pdf/2009.09468.pdf.