随着移动通信系统的快速发展，不断延伸的应用场景对数据传输的速率、能效等各项指标有了更高的要求。当前MIMO-OFDM以及智能反射面等技术驱动着新一代移动通信系统向着更高传输速率、更大系统容量方向发展。然而，更多的场景需求和更高的性能要求使得无线传输技术（Radio Transmission Technology，RTT）面临新的挑战，如预编码、资源分配、信号检测等关键技术的实现通常以获取信道状态信息（Channel Sate Information，CSI）为前提，因此通过信道估计技术获取CSI在无线通信系统中至关重要。现有信道估计算法的研究普遍存在着导频开销大、算法复杂度高等问题，基于此，本文的主要研究内容和创新点可概况如下：

首先，本文研究了MIMO-OFDM系统下的信道估计算法，在基于导频序列方法进行信道估计的背景下，研究一种联合导频设计和信道估计的深度学习方案，提出一个基于自编码器和生成对抗网络的新型混合网络架构（CAGAN）。该算法首先利用自编码器网络对含噪信道进行特征信息提取，完成导频优化设计，然后利用设计好的优化导频输入条件生成对抗网络完成信道估计。仿真结果表明相比传统的最小二乘（Least Square，LS）、最小均方误差（Minimal Mean Square Error，MMSE）算法以及现有基于深度学习架构的ChannelNet等算法，本文中所提出的CAGAN算法对环境噪声具有更高的鲁棒性，并且能在有效节省导频资源的前提下达到较高的估计精度。

其次，本文进一步研究了基于智能反射面辅助系统的信道估计算法，在时分双工模式下IRS-SIMO系统上行链路的信道估计问题中，将MIMO-OFDM系统中基于条件生成对抗网络（conditional Generative Adversarial Networks，cGAN）的信道估计部分做进一步网络优化。该算法首先采用LS算法进行信道的粗估计，然后设计超分辨率（Super Resolution，SR）深度学习网络作为生成器模型，进一步融合cGAN网络将信道估计建模为低分辨图像到高分辨图像的恢复问题，通过设计网络结构和改进损失函数来提高信道估计的精度。仿真实验表明，相比传统的最小二乘估计算法（LS）以及基于卷积的深度残差网络（CNN-based DRN，CDRN）估计算法，所提出的算法具有更高的估计精度，且可以适应更少导频数目和更复杂的应用场景。

With the rapid evolution of mobile communication systems, the ever-extending application scenarios have higher requirements for data transmission rate, energy efficiency and other indicators. Current technologies such as MIMO-OFDM and intelligent reflective surfaces are driving the development of a new generation of mobile communication systems toward higher transmission rates and greater system capacity. However, more scenario demands and higher performance requirements make radio transmission technology are facing new challenges, such as precoding, resource allocation, signal detection and other key 
technologies are usually implemented with the prerequisite of obtaining Channel Sate Information (CSI), so obtaining CSI through channel estimation techniques is crucial in 
wireless communication systems. The existing research on channel estimation algorithms generally suffers from the problems of high guide frequency overhead and high complexity of algorithms, based on which the main research contents and innovations of this paper can be summarized as follows:
Firstly, in this thesis we study the channel estimation algorithm in MIMO-OFDM system, and in the context of channel estimation based on the guide frequency sequence approach, a deep learning scheme for joint guide frequency design and channel estimation is 
investigated, and a new hybrid network architecture (CAGAN) based on self-encoder and generative adversarial network is proposed. The algorithm first uses the self-encoder network 
to extract the feature information of the noisy channel to complete the optimal design of the guide frequency, and then uses the designed optimized guide frequency input conditions to 
generate the adversarial network to complete the channel estimation. The simulation results show that the proposed CAGAN algorithm is more robust to environmental noise than the traditional Least Square (LS) and Minimal Mean Square Error (MMSE) algorithms and the existing ChannelNet algorithms based on deep learning architecture. The CAGAN algorithm proposed in this paper has higher robustness to environmental noise and can achieve higher estimation accuracy with effective saving of guide frequency resources.
Secondly, in this thesis we further investigate the channel estimation algorithm based on the intelligent reflective surface assisted system, in which the channel estimation problem 
of the uplink of the IRS-SIMO system in the time division duplex mode is further investigated. The channel estimation part of the MIMO-OFDM system based on conditional Generative 
Adversarial Networks (cGAN) is further network optimized. The algorithm first uses the LS algorithm for coarse channel estimation, designs the Super Resolution (SR) deep learning network as the generator model, and then further fuses the cGAN network to model the channel estimation as a low-resolution to high-resolution image recovery problem, and improves the accuracy of channel estimation by designing the network structure and 
improving the loss function. Simulation experiments show that the proposed algorithm has higher estimation accuracy and can be adapted to fewer number of leads and more complex application scenarios than the traditional least squares estimation algorithm (LS) and the convolution-based deep residual network (CNN-based DRN, CDRN) estimation algorithm.

[1] Almeida I, Mendes L, Rodrigues J, et al. 5G Waveforms for IoT Applications[J]. IEEE Communications Surveys & Tutorials, 2019, 21(3):2554-2567.

[2] Liu Q, Cheng Y, Liu X, et al. Deep Learning for Massive MIMO: A Survey[J], IEEE Communications Surveys & Tutorials, 2020, 22(4): 2027-2069.

[3] Zhou L, Lin X, Xu J, et al. A Survey on Energy Efficient Resource Allocation for MIMO-OFDM Systems[J], IEEE Communications Surveys & Tutorials, 2020, 22(2): 1148-1172.

[4] Ertugrul Basar. Transmission through large intelligent surfaces: A new frontier in wireless communications[C]//2019 European Conference on Networks and Communications (EuCNC). IEEE, 2019:112–117.

[5] 刘诗雨. 基于深度学习的IRS辅助通信关键技术研究[D]. 浙江大学，2021.

[6] Di Renzo M, Ozcelik A, Debbah M, et al. Smart Radio Environments Empowered by Reconfigurable AI Meta-surfaces: An Idea Whose Time Has Come[J]. EURASIP Journal on Wireless Communications and Networking, 2019. 2019(1):1–20.

[7] Di Renzo M, Huang J, Song H, et al. A New Wireless Communication Paradigm Through Software-controlled Meta-surfaces[J]. IEEE Communications Magazine, 2018, 57(6): 162-169.

[8] 汪冉冉. MIMO系统信道估计与信号检测算法研究及实现[D]. 中国矿业大学, 2022.

[9] Mao Q, Hu F, Hao Q. Deep Learning for Intelligent Wireless Networks: A Comprehensive Survey[J]. IEEE Communications Surveys & Tutorials, 2018, 20(4): 2595-2621.

[10] Sun Z, Wand L, Wang CH, et al. An Adaptive Transmission Scheme for Deep Space Communication[J]. Transactions of Nanjing University of Aeronautics and Astronautics, 2019, 36(1): 26-35.

[11] 李竹一. 基于神经网络的多载波调制信号联合检测技术研究[D]. 浙江大学, 2019.

[12] Zappone A, Di Renzo M, Debbah M. Wireless networks design in the era of deep learning: Model-based, ai-based, or both? [J]. IEEE Transactions on Communications, 2019. 67(10): 7331–7376.

[13] Yang, Yun F, et al. A survey on channel estimation for massive MIMO-OFDM[J]. IEEE Communications Surveys & Tutorials, 2020. 22.2 (2020): 985-1003.

[14] 丁旭，张韵农，李丹. 基于MIMO-OFDM系统的信道估计算法综述[C]// 全国第21届计算机技术与应用(CACIS)学术会议. 2010.

[15] Suh C, Hwang C S, Choi H. Comparative study of time-domain and frequency-domain channel estimation in MIMO-OFDM systems[C]//14th IEEE Proceedings on Personal, Indoor and Mobile Radio Communications, 2003. PIMRC 2003. IEEE, 2003, 2: 1095-1099.

[16] 郄九玲. 新型大规模MIMO-OFDM信道估计方法的研究[D].南京邮电大学, 2019.

[17] Yin J, Zhang Y, and Yang L. Compressive Sensing for Massive MIMO Channel. Estimation: A Survey[J]. IEEE Communications Surveys & Tutorials, 2020, 22(3):2025-2050.

[18] He X, Song R, Zhu W-P. Pilot allocation for sparse channel estimation in MIMO-OFDM systems[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2013, 60(9): 612-616.

[19] Lin B, Ghassemlooy Z, Xu J, et al. Experimental demonstration of compressive sensing-based channel estimation for MIMO-OFDM VLC[J]. IEEE Wireless Communications Letters, 2020, 9(7): 1027-1030.

[20] Yang Y, Gao F, Li Y, et al. Deep learning-based downlink channel prediction for FDD massive MIMO system[J], IEEE Communications Letters, 2019, 23(11): 1994-1998.

[21] Dong P, Hang H. Deep CNN for wideband mmWave massive mimo channel estimation using frequency correlation[C]. //IEEE International Conference on Acoustics, Speech and Signal Processing, Brighton, UK,2019:4529-4533.

[22] He H, Wen C. Deep-learning based channel estimation for beamspace mmWave massive MIMO systems[J]. IEEE Wireless Communication Letters, 2018, 7(5):852-855.

[23] Jin Y, Hang J. Channel estimation for mmWave massive MIMO with convolutional blind denoising network[J]. IEEE Communication Letters, 2020, 24(1):95-98.

[24] Soltani M, Pourahmadi V, Mirzaei A, et al. Deep learning-based channel estimation[J]. IEEE Communications Letters, 2019, 23(4): 652-655.

[25] Sun Y, Shen H. ICINet: ICI-Aware neural network-based channel estimation for rapidly time-varying OFDM system[C]. //IEEE Communication Letters, 2021, 25(9):2973-2977.

[26] Gao S, Dong P, Pan Z, et al. Deep learning-based channel estimation for massive MIMO with mixed-resolution ADCs[J]. IEEE Communications Letters, 2019, 23(11): 1989-1993.

[27] Zhang Y, Alrabeiah M, Alkhateeb A. Deep learning for massive MIMO with 1-bit ADCs: When more antennas need fewer pilots[J]. IEEE Wireless Communications Letters, 2020, 9(8): 1273-1277.

[28] Gabrié M, Manoel A, Luneau C, et al. Entropy and mutual information in models of deep neural networks[J]. Advances in Neural Information Processing Systems, 2018, 31.

[29] Yang Y, Li Y, Zhang W, et al. Generative-adversarial-network-based wireless channel modeling: Challenges and opportunities[J]. IEEE Communications Magazine, 2019, 57(3): 22-27.

[30] Li X, Alkhateeb A, Tepedelenlioğlu C. Generative adversarial estimation of channel covariance in vehicular millimeter wave systems[C]//2018 52nd Asilomar Conference on Signals, Systems, and Computers. IEEE, 2018: 1572-1576.

[31] Ye H, Li G Y, Juang B H F, et al. Channel Agnostic end-to-end Learning based Communication Systems with Conditional GAN[C]//Proceedings of 2018 IEEE Globecom Workshops. IEEE, 2018: 1-5.

[32] Balevie, Doshi A, Jalala, et al. High Dimensional Channel Estimation using Deep Generative Networks[J]. IEEE Journal on Selected Areas in Communications, 2020, 39(1): 18-30.

[33] Yuan X, Zhang Y J A, Shi Y, et al. Reconfigurable-intelligent-surface empowered wireless communications: Challenges and opportunities[J]. IEEE Wireless Communications, 2021, 28(2): 136-143.

[34] Mishra D, Johansson H. Channel estimation and low-complexity beamforming design for passive intelligent surface assisted MISO wireless energy transfer[C]//ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019: 4659-4663.

[35] Jensen T L, De Carvalho E. An optimal channel estimation scheme for intelligent reflecting surfaces based on a minimum variance unbiased estimator[C]//ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2020: 5000-5004.

[36] Zheng B, Zhang R. Intelligent reflecting surface-enhanced OFDM: Channel estimation and reflection optimization[J]. IEEE Wireless Communications Letters, 2019, 9(4): 518-522.

[37] Taha A, Alrabeiah M, Alkhateeb A. Deep learning for large intelligent surfaces in millimeter wave and massive MIMO systems[C]//2019 IEEE Global Communications Conference (GLOBECOM). IEEE, 2019:1-6.

[38] Chen S, Gao L, and Zhang J, A VAE-Based Channel Estimation Method for Intelligent Reflecting Surface Aided Wireless Communications[C] //in Proceedings of the 2021 IEEE/CIC International Conference on Communications in China (ICCC), Chengdu, China, 2021: 1133-1137.

[39] Guo J, Huang X, Yang Y, et al. Intelligent Reflecting Surface-Assisted Wireless Communications: A Review[J], IEEE Communications Surveys and Tutorials, 2020, 22(4): 2622-2649.

[40] Heath Jr R W, Lozano A. Foundations of MIMO communication[M]. Cambridge University Press, 2018.

[41] Farhang-Boroujeny B, Moradi H. OFDM Inspired Waveforms for 5G[J]. IEEE Communications Surveys & Tutorials, 2017, 18(4):1-1.

[42] Demir A F, Elkourdi M, Ibrahim M, et al. Waveform design for 5G and beyond[J]. arXiv preprint arXiv:1902.05999, 2019.

[43] Holma H, Toskala A, and Nakamura T, 5G Technology: 3GPP New Radio[J]. John Wiley & Sons, 2020.

[44] Elnakeeb A, Mitra U. Bilinear Channel Estimation for MIMO OFDM: Lower Bounds and Training Sequence Optimization[J]. IEEE Trans. on Signal Processing, 2021, pp (99):1-1.

[45] Ming C. A new derivation of least-squares-fitting principle for OFDM channel estimation[J]. IEEE Transactions on Wireless Communications, 2016, 5(4): 726-731.

[46] Minn H, Bhargava V K. An investigation into time-domain approach for OFDM channel estimation[J]. IEEE Transactions on Broadcasting, 2000, 46(4): 240-248.

[47] Wu Q, Zhang R. Towards smart and reconfigurable environment: Intelligent reflecting surface aided wireless network[J]. IEEE communications magazine, 2019, 58(1): 106-112.

[48] Goodfellow I, Bengio Y, Courville A. Deep learning[M]. MIT press, 2016.

[49] Alcantarila P F, Simon S, Roberto A, et al. Street-view Change Detection with Deconvolutional Networks[J]. Autonomous Robots, 2018, 42(7):1-22.

[50] Ye H , Li G Y , Juang B. Bilinear Convolutional Auto-encoder based Pilot-free End-to-end Communication Systems[C]// ICC 2020 - 2020 IEEE International Conference on Communications (ICC). IEEE, 2020.

[51] Chen Y H, Krishna T, Emer J S, et al. Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks[J]. IEEE journal of solid-state circuits, 2016, 52(1): 127-138.

[52] O'Shea T J, Karra K, Clancy T C. Learning to communicate: Channel auto-encoders, domain specific regularizers, and attention[C]//2019 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT). IEEE, 2019: 223-228.

[53] Jabbar A, Li X, Omar B. A survey on generative adversarial networks: Variants, applications, and training[J]. ACM Computing Surveys (CSUR), 2021, 54(8): 1-49.

[54] Wang H W, Wang J L, Wang J, et al. Learning with Generative Adversarial Nets[J]. IEEE Transactions on Knowledge and Data Engineering. 2021, 33(8):3090-3103.

[55] Dong Y, Wang H, Yao Y D. Channel estimation for one-bit multiuser massive MIMO using conditional GAN[J]. IEEE Communications Letters, 2020, 25(3): 854-858.

[56] Chen J C, Wen C K, et al. An Efficient Pilot Design Scheme for Sparse Channel Estimation in OFDM Systems[C]. //IEEE Communications Letters, 2018 :1352-1355.

[57] Qi C, Yue G, Wu L, et al. Pilot Design Schemes for Sparse Channel Estimation in OFDM Systems[J]. Vehicular Technology IEEE Transactions on, 2015, 64(4):1493-1505.

[58] 费洪涛，何雪云，梁彦．基于自适应压缩感知的OFDM稀疏信道估计研究[J]．南京邮电大学学报，2019，39（2）：49-54.

[59] Abid A, Balin M F, Zou J. Concrete autoencoders for differentiable feature selection and reconstruction[J]. arXiv preprint arXiv:1901.09346, 2019.

[60] Zhang Y, Lu Z, Wang S. Unsupervised feature selection via transformed auto-encoder[J]. Knowledge-Based Systems, 2021, 215: 106748.

[61] Soltani M, Pourahmadi V, Sheikhzadeh H. Pilot pattern design for deep learning-based channel estimation in OFDM systems[J]. IEEE Wireless Communications Letters, 2020, 9(12): 2173-2176.

[62] Yang A, Sun P, Rakesh T, et al. Deep learning based OFDM channel estimation using frequency-time division and attention mechanism[C]//2021 IEEE Globecom Workshops (GC Wkshps). IEEE, 2021: 1-6.

[63] Mehlfuhrer C, Colom Ikuno J, M. Simko, et al. The vienna lte simulators - enabling reproducibility in wireless communications research[C], //EURASIP Journal on Advances in Signal Processing, vol. 2011, p. 29, Jul 2011.

[64] Soltani M, Pourahmadi V, Mirzaei A, et al. Deep learning-based channel estimation[J]. IEEE Communications Letters, 2019, 23(4): 652-655.

[65] Chen J, Liang Y C, Cheng H V, et al. Channel estimation for reconfigurable intelligent surface aided multi-user MIMO systems[J]. arXiv preprint arXiv:1912.03619, 2019.

[66] Li S, Yuan J, Yuan W, et al. Performance analysis of coded OTFS systems over high-mobility channels[J]. IEEE transactions on wireless communications, 2021, 20(9): 6033-6048.

[67] Liu C, Wang J, Liu X, et al. Deep CM-CNN for Spectrum Sensing Cognitive Radio[J]. IEEE Journal on Selected Areas in Communications,2019 ,37(10): 2306–2321.

[68] Mishra D, Johansson H. Channel Estimation and Low-complexity Beamforming Design for Passive Intelligent Surface Assisted MISO Wireless Energy Transfer[C]. //ICASSP 2019 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, 2019:4659–4663.

[69] Zheng B X, Zhang R. Intelligent Reflecting Surface-enhanced OFDM: Channel estimation and reflection optimization[J]. IEEE Wireless Communications Letters, 2019. 9(4):518–522.

[70] Gong S, Lu X, Hong D T, et al. Towards Smart Wireless Communications via Intelligent Reflecting Surfaces: A Contemporary Survey[J]. IEEE Communications Surveys & Tutorials, 2019, (99):1-1.

[71] Liu C, Liu X, Ng D W K, et al. Deep residual learning for channel estimation in intelligent reflecting surface-assisted multi-user communications[J]. IEEE Transactions on Wireless Communications, 2021, 21(2): 898-912