大規模多輸入多輸出(Massive Multiple-Input Multiple-Output , Massive MIMO)技術為第五代行動通訊(5G)系統中一項核心技術。而在Massive MIMO系統下，通道估測(Channel Estimation)儼然成為一個重要的議題。然而傳統的通道估測演算法需要大量複雜之運算，導致基地台在做訊號處理及估測時耗費許多能量。因此本論文採取二階段式接收及估測訊號的模式，利用此模式接收及估測訊號可以大幅減少使用者的發射功率以及基地台做訊號處理時所消耗的能量。在二階段式接收及估測後可以得到兩個次通道，將兩個次通道重建成完整之通道為一項關鍵的技術。近年來深度學習(Deep Learning，簡稱DL)技術廣泛運用於多項通訊領域研究中，從這些研究能夠證實DL技術能夠有效提高系統效能，並且同時降低其運算複雜度。因此本論文提出兩種監督式學習(Supervised-learning)的DL模型算法。其中第一種DL模型是基於深度神經網路所設計的，因此命名為: Deep Neural Network Based Channel Estimation (DB-CE)，第二種DL模型是基於卷積神經網路所設計的，我們命名為: Convolutional Neural Network Based Channel Estimation(CB-CE)。在這兩種模型中完成通道重建。
本論文提出之DB-CE演算法和CB-CE演算法具有以下兩個特徵。第一，其接收訊號及估測方式與傳統模式不同，因此減少許多能量消耗。第二，重建方法與現有數值演算法不同，我們使用深度學習相關技術學習通道間的通道特徵，藉由通道特徵去重建成完整通道。最後，模擬結果顯示，本論文所提出的DB-CE演算法和CB-CE演算法其通道估測誤差都明顯低於現有算法，且通道重建能力也優於現有數值演算法。

關鍵字: 5G，MIMO，低秩數通道，通道估測，深度學習，監督式學習，深度神經網路、卷積神經網路

Massive Multiple-Input Multiple-Output (Massive MIMO) technology is a core technology in the fifth-generation mobile communication (5G) system. In the MIMO system, Channel Estimation has become an important issue. However, the traditional channel estimation algorithm requires a lot of complicated calculations, which causes the base station to consume a considerable amount of energy in signal processing and estimation. Therefore, In this paper, we adopt a two-stage mode, including signal receiving and estimation . This mode can greatly reduce the user's transmission power and the energy consumed by the base station for signal processing. We can obtain two sub-channel after the two-stage reception and estimation mode. Reconstructing two sub-channel into a complete channel is a key technology in this issue. In recent years, Deep Learning (DL) technology has been widely used in communication research. In these studies, it can be confirmed that DL technology can effectively improve system performance and reduce its computational complexity at the same time. Therefore, we propose two Supervised-learning DL model algorithms. The first DL model, which named Deep Neural Network Based Channel Estimation (DB-CE), is designed based on deep neural networks. The second DL model, which named Convolutional Neural Network Based Channel Estimation(CB-CE), is designed based on convolutional neural networks. The channel reconstruction is done in these two models.
The DB-CE algorithm and CB-CE algorithm proposed in this paper have the following two characteristics. First, the method of reception signal and estimation is different from the traditional method, thus reducing a lot of energy consumption. Second, the reconstruction method is different from the existing numerical algorithm.  
We reconstruct the complete channel by using channel features between channels that learned from deep learning related technology. Finally, the simulation results demonstrate that the channel estimation errors of the DB-CE algorithm and the CB-CE algorithm proposed in this paper are significantly lower than those of the existing algorithms, and that the channel reconstruction ability of the DB-CE algorithm and the CB-CE algorithm is also better than that of the existing numerical algorithms.

Keywords: 5G, MIMO, channel estimation, supervised learning, deep neural network, convolutional neural network

[1] J. Wang, A. Jin, D. Shi, L. Wang, H. Shen, D. Wu, L. Hu, L. Gu, L. Lu,Y. Chen, J. Wang, Y. Saito, A. Benjebbour, and Y. Kishiyama, "Spectral efficiency improvement with 5G technologies: Results from field tests, " IEEE J. Sel. Areas Commun., vol. 35, no. 8, pp. 1867–1875, Aug. 2017.
[2] L. Lu, G. Y. Li, A. L. Swindlehurst, A. Ashikhmin, and R. Zhang, "An overview of massive MIMO: Benefits and challenges, " IEEE J. Sel. Areas Commun., vol. 8, no. 5, pp. 742–758, Oct. 2014.
[3] A. L. Swindlehurst, E. Ayanoglu, P. Heydari and F. Capolino, "Millimeter-wave massive MIMO: the next wireless revolution?," in IEEE Communications Magazine, vol. 52, no. 9, pp. 56-62, September 2014.
[4] S. Sun, T. S. Rappaport, R. W. Heath, A. Nix, and S. Rangan, "Mimo for millimeter-wave wireless communications: beamforming, spatial multiplexing, or both? " IEEE Commun. Mag., vol. 52, no. 12, pp. 110–121,Dec. 2014
[5] F. W. Vook, A. Ghosh and T. A. Thomas, "MIMO and beamforming solutions for 5G technology," IEEE MTT-S Int. Microw. Symp. Dig., pp. 1-4, Jun. 2014.
[6] S. Kutty and D. Sen, "Beamforming for Millimeter Wave Communications: An Inclusive Survey," in IEEE Communications Surveys & Tutorials, vol. 18, no. 2, pp. 949-973, Secondquarter 2016.
[7] A. Khlifi and R. Bouallegue, "Performance analysis of LS and LMMSE channel estimation techniques for LTE downlink systems, " Int. J. Wireless Mobile Netw., vol. 3, no. 5, pp. 141–149, Oct. 2011.
[8] N. Shariati, J. Wang, and M. Bengtsson, "Robust training sequence design for correlated MIMO channel estimation, " IEEE Trans. Signal Process., vol. 62, no. 1, pp. 107–120, Jan. 2014.
[9] B. Li, S. Wang, J. Zhang, X. Cao, and C. Zhao, "Randomized approximate channel estimator in massive-MIMO communication, " IEEE Commun. Lett., vol. 24, no. 10, pp. 2314–2318, Oct. 2020.
[10] H. Huang, J. Yang, H. Huang, Y. Song, and G. Gui, "Deep learning for super-resolution channel estimation and DOA estimation based massive MIMO system, " IEEE Trans. Veh. Technol., vol. 67, no. 9, pp. 8549–8560, Sep. 2018.
[11] H. He et al., "Deep Learning-Based Channel Estimation for Beamspace mmWave Massive MIMO Systems, " IEEE Wireless Commun. Lett., vol. 7, no. 5, pp. 852–55, Oct. 2018.
[12] C. K. Wen, W. T. Shih, and S. Jin, "Deep Learning for Massive MIMO CSI Feedback, " IEEE Wireless Commun. Lett.,vol. 7, no. 5, Oct. 2018, pp. 748–51.
[13] T. Wang, C. Wen, S. Jin and G. Y. Li, "Deep Learning-Based CSI Feedback Approach for Time-Varying Massive MIMO Channels," in IEEE Wireless Communications Letters, vol. 8, no. 2, pp. 416-419, April 2019.
[14] P. Dong, H. Zhang, G. Y. Li, N. NaderiAlizadeh, and I. Gaspar, "Deep CNN for wideband mmWave massive MIMO channel estimation using frequency correlation, " in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process., pp. 4529–4533,May 2019.
[15] P. Dong, H. Zhang, G. Y. Li, I. S. Gaspar and N. NaderiAlizadeh, "Deep CNN-Based Channel Estimation for mmWave Massive MIMO Systems," in IEEE Journal of Selected Topics in Signal Processing, vol. 13, no. 5, pp. 989-1000, Sep. 2019.
[16] J. G. Andrews et al., "What will 5G be? " IEEE J. Sel. Areas Commun.,vol. 32, no. 6, pp. 1065–1082, Jun. 2014.
[17] D. W. K. Ng, E. S. Lo, and R. Schober, "Energy-efficient resource allocation in OFDMA systems with large numbers of base station antennas, " IEEE Trans. Wireless Commun., vol. 11, no. 9, pp. 3292–3304, Sep. 2012
[18] R. Zi et al., "Energy efficiency optimization of 5G radio frequency chain systems, " IEEE J. Sel. Areas Commun., vol. 34, no. 4, pp. 758–771, Apr. 2016.
[19] H. Huang et al., "Deep-learning-based millimeter-wave massive MIMO for hybrid precoding, " IEEE Trans. Veh. Technol., vol. 68, no. 3, pp. 3027–3032, Mar. 2019.
[20] A.M.Elbir and A.K.Papazafeiropoulos,"Hybrid precoding for multiuser millimeter wave massive MIMO systems: A deep learning approach, " IEEE Trans. Veh. Technol., vol. 69, no. 1, pp. 552–563, Jan. 2020.
[21] P.Y. Liu and E.Y. Lam. "Image Reconstruction Using Deep Learning. " arXiv preprint arXiv:1809.10410 (2018).c
[22] H. Xie, F. Gao, and S. Jin, "An overview of low-rank channel estimation for massive MIMO systems, " IEEE Access, vol. 4, pp. 7313–7321, 2016.
[23] W. Shen, L. Dai, B. Shim, S. Mumtaz, and Z. Wang, "Joint CSIT acquisition based on low-rank matrix completion for FDD massive MIMO systems, " IEEE Commun. Lett., vol. 19, no. 12, pp. 2178–2181, Dec. 2015.
[24] N. Song, C. Ye, X. Hu and T. Yang, "Deep Learning based Low-Rank Channel Recovery for Hybrid Beamforming in Millimeter-Wave Massive MIMO, "2020 IEEE Wireless Communications and Networking Conference (WCNC), pp. 1-6, May 2020.
[25] B. Sahiner, A. Pezeshk, L. M. Hadjiiski, X. Wang, K. Drukker, K. H. Cha, et al., "Deep learning in medical imaging and radiation therapy", Med. Phys., vol. 46, no. 1, pp. e1-e36, 2018.
[26] A. S. Lundervold and A. Lundervold, "An overview of deep learning in medical imaging focusing on MRI", Zeitschrift für Medizinische Physik, vol. 29, no. 2, pp. 102-127, 2019.
[27] A. Carrio, C. Sampedro, A. Rodriguez-Ramos and P. Campoy, "A review of deep learning methods and applications for unmanned aerial vehicles", J. Sensors, vol. 2017, Aug. 2017.
[28] A.M. Ozbayoglu, M.U. Gudelek, and O.B. Sezer, "Deep learning for financial applications: A survey." Applied Soft Computing, pp. 106384, 2020.
[29] T. Fischer and C. Krauss, "Deep learning with long short-term memory networks for financial market predictions." European Journal of Operational Research, 270(2):654–669, 2018.
[30] J. Duan, "Financial system modeling using deep neural networks (DNNs)
for effective risk assessment and prediction, " J. Franklin Inst., vol. 356,
no. 8, pp. 4716–4731, May 2019.
[31] X. Luo, L. Oyedele, A. Ajayi, O. Akinade, H. Owolabi, A. Ahmed, "Feature extraction and genetic algorithm enhanced adaptive deep neural network for energy consumption prediction in buildings. " Renew. Sustain. Energy Rev,2020.
[32] K. Gong et al., "Iterative PET image reconstruction using convolutional neural network representation, " IEEE Trans. Med. Imag., vol. 38, no. 3, pp. 675–685, Mar. 2019.
[33] K. H. Cha, L. Hadjiiski, R. K. Samala, H.P. Chan, E. M. Caoili and R. H. Cohan, "Urinary bladder segmentation in CT urography using deep-learning convolutional neural network and level sets", Med. Phys., vol. 43, no. 4, pp. 1882-1896, 2016.
[34] H. Chen et al., "Low-dose CT with a residual encoder-decoder convolutional neural network, " IEEE Trans. Image Process., vol. 36, no. 12, pp. 2524–2535, Dec. 2017.
[35] K. A. J. Eppenhof and J. -P. W. Pluim, "Pulmonary CT registration through supervised learning with convolutional neural networks, " IEEE Trans. Med. Imag., vol. 38, no. 5, pp. 1097–1105, May 2019.
[36] K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition. "In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770–778, 2016.
[37] G. Klambauer, T. Unterthiner, A. Mayr, and S. Hochreiter, "Self-normalizing neural networks. "In Advances in Neural Information Processing Systems (NIPS), pp. 971–980, 2017.
[38] D. P. Kingma and J. Ba, "Adam: A method for stochastic optimization." In ICLR, 2015.
[39] H. He, C.-K. Wen, S. Jin, and G. Y. Li, "A model-driven deep learning network for MIMO detection, " in IEEE Global Conference on Signal and Information Processing (GlobalSIP), pp. 584–588 ,Nov. 2018.