无线通信系统由发射机和接收机组成，信息经过信道编码、调制等处理后通过天线发射出去。由于信道衰落、噪声和干扰的影响，信号会存在较严重的失真，接收机需要从失真的信号中尽可能恢复出原始信息。本文提出基于深度学习的多输入多输出（Multiple-Input Multiple-Output，MIMO）通信系统接收方法，利用深度神经网络替代接收端整个信息恢复环节，实现比特信息恢复。与传统MIMO接收方法相比，可以达到更低的误比特率以及更好的抗干扰性能。本文主要研究内容如下：

（1）针对传统Alamouti算法在复杂环境下的高误比特率问题，提出一种基于深度学习的多标签MIMO智能接收方法，实现从接收到的失真信号中恢复出原始比特信息的目的，替代了传统物理层通信接收机端整个信息恢复流程，尽可能克服无线信道衰落、噪声、干扰等因素。实验表明在瑞利信道下的可以得到更低的误比特率。

（2）目前基于深度学习信号接收方法的研究仅适用于满足小比特数据的信号接收，为了适应更大比特信号接收的需求，进一步提出了一种基于多输出分类的MIMO智能接收方法。设计了一种多个并行子网络结构，每个子网络来实现一位比特信息恢复，解决比特信息位数增大时采用多标签接收方法出现高误比特率的问题。实验结果表明，所提方法在瑞利信道环境下，恢复128位比特信号的误比特率性能优于Alamouti算法。

（3）为了解决MIMO在矿井环境中存在误比特率较高的问题，本文提出了矿井MIMO智能接收模型。该模型基于多标签MIMO智能接收方法，实现原始比特信息恢复。相对于传统的MIMO接收方法，该模型不存在由解码、解调等过程造成的误差累积。实验结果表明，在不同的调制、编码、发射端天线数等场景下，该模型相对于传统的接收方式，具有更低的误比特率。此外，本文证明当接收数据出现一定损失时，该模型仍可实现有效接收，恢复出原始信息。

The wireless communication system is composed of a transmitter and a receiver, and the information is transmitted through the antenna after channel coding, modulation and other processing. Due to the influence of channel fading, noise and interference, the signal will have serious distortion, and the receiver needs to recover the original information as much as possible from the distorted signal. A receiver algorithm for multiple-input multiple-output (MIMO) communication systems based on deep learning is proposed. The deep neural network is used to replace the entire information recovery link at the receiver side to achieve bit information recovery. Compared with the traditional MIMO receiving algorithm, it can achieve lower bit errors. rate and better anti-interference performance. In summary, the main research contents and results of this paper are as follows:

（1）Aiming at the high bit error rate of the traditional Alamouti algorithm in complex environments, a multi-label MIMO intelligent receiving algorithm based on deep learning is proposed to achieve the purpose of recovering the original bit information from the received signal, replacing the traditional physical communication layer. The entire information recovery process at the communication receiver end can overcome factors such as wireless channel fading, noise, and interference as much as possible. Experiments show that a lower bit error rate can be obtained under the Rayleigh channel.

（2）The current research based on deep learning signal receiving algorithm is only suitable for the signal receiving of small-bit data. In order to meet the needs of receiving larger-bit signals, a MIMO intelligent receiving algorithm based on multi-output classification is further proposed. A multi-parallel sub-network structure is designed, each sub-network realizes the recovery of one-bit information, and solves the problem of high bit error rate when the number of bits of bit information increases by using the multi-lable receiving method. The experimental results show that, The bit error rate performance of the received 128-bit signal is better than that of the Alamouti algorithm.

（3）In order to solve the problem of high bit error rate of MIMO in mine environment, a mine MIMO intelligent receiving model is proposed. The model is based on the multi-tag MIMO intelligent receiving algorithm to realize the recovery of the original bit information. Compared with the traditional MIMO receiving algorithm, the model does not have the error accumulation caused by decoding, demodulation and other processes. The experimental results show that the model has a lower bit error rate than the traditional receiving method under different modulation, coding, number of transmitter antennas and other scenarios. In addition, it is proved that when there is a certain loss of received data, the model can still achieve effective reception and restore the original information.

[1] 唐杰, 李凯, 林楚婷, 宋彦, 周恩丞. OAM-MIMO通信系统的信道容量研究[J].北京邮电大学学报, 2020,43(06):27-35.

[2] 王毅, 陈启鑫, 张宁, 冯成, 滕飞, 孙铭阳, 康重庆. 5G通信与泛在电力物联网的融合:应用分析与研究展望[J].电网技术,2019,43(05):1575-1585.

[3] 韩雨, 葛悦涛. 2020年人工智能领域科技发展综述[J].飞航导弹,2021(04):1-6+53.

[4] 丁青锋, 连义翀, 邓玉前. 空间相关信道下大规模MIMO系统频谱效率分析[J].北京邮电大学学报, 2020,43(03):45-50+98.

[5] Ye H, Li G Y, JUANG B H F. Power of Deep learning for channel estimation and signal detection in OFDM systems [J]. IEEE Wireless Communications Letters, 2018, 7(1):114-117.

[6] 申东, 赵丹, 李强等. 低复杂度的大规模MIMO上行链路软输出信号检测[J].计算机应用研究, 2021,38(05):1524-1528.

[7] Gruber T, Cammerer S, Hoydis J, et al. On deep learning-based channel decoding[C]//2017 51st Annual Conference on Information Sciences and Systems (CISS). IEEE, 2017: 1-6.

[8] 周鑫, 何晓新, 郑昌文.基于图像深度学习的无线电信号识别[J].通信学报, 2019, 40(07): 114-125.

[9] 王希. 混合噪声下基于深度学习的无线通信信号接收技术研究[D]. 成都：电子科技大学, 2020.

[10] T.O’Shea, T.Erpek, T.C.Clancy, Deep learning based MIMO communications[J],ArXiv reprint arXiv, 2017:1707.7980.

[11] Mousavi A, Baraniuk R G. Learning to invert: Signal recovery via deep convolutional networks[C]//2017 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 2017: 2272-2276.

[12] Luo C, Ji J, Wang Q, et al. Channel state information prediction for 5G wireless communications: A deep learning approach[J]. IEEE Transactions on Network Science and Engineering, 2018, 7(1): 227-236.

[13] Xie H, Qin Z, Li G Y, et al. Deep learning enabled semantic communication systems[J]. IEEE Transactions on Signal Processing, 2021, 69: 2663-2675.

[14] He H, Wen C K, Jin S, et al. A model-driven deep learning network for MIMO detection[C]//2018 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE, 2018: 584-588.

[15] Zhang J, He H, Wen C K, et al. Deep learning based on orthogonal approximate message passing for CP-free OFDM[C]//ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019: 8414-8418.

[16] 金石, 张静, 何恒涛, 高璇璇, 温朝凯. 一种基于模型驱动深度学习的无循环前缀OFDM接收方法[P]. 中国专利：CN109617847A,2019-04-12.

[17] He H, Zhang M, Jin S, et al. Model-driven deep learning for massive MU-MIMO with finite-alphabet precoding[J]. IEEE Communications Letters, 2020, 24(10): 2216-2220.

[18] 左崇彦. 超奈奎斯特(FTN)速率传输的递归神经网络解调方法[D].南京：东南大学,2017.

[19] 杨耀栋, 吴乐南.相移键控信号的深度神经网络解调器[J].电子测量与仪器学报,2018,32(04):144-150.

[20] 王海, 刘晶晶, 韩潇, 刘宗延, 李林鹏, 张广睿. 基于1D-CNN的BPSK调制信号相位跳变检测方法[P]. 陕西省：CN107317778B, 2019-11-26.

[21] 王海, 刘宗延, 俞忠伟, 张敏, 秦红波. 基于CNN和LSTM的DQPSK调制信号解调方法[P].中国专利：CN109379318A,2019-02-22.

[22] Alkhateeb A, Alex S, Varkey P, et al. Deep learning coordinated beamforming for highly-mobile millimeter wave systems[J]. IEEE Access, 2018, 6: 37328-37348.

[23] 桂冠, 王禹, 黄浩. 基于深度学习的物理层无线通信技术:机遇与挑战[J].通信学报,2019,40(02):19-23.

[24] Sarieddeen H, Alouini M S, Al-Naffouri T Y. An overview of signal processing techniques for terahertz communications[J]. Proceedings of the IEEE, 2021,109(10):1628-1665.

[25] Xie H, Qin Z, Li G Y, et al. Deep learning enabled semantic communication systems[J]. IEEE Transactions on Signal Processing, 2021, 69: 2663-2675.

[26] Zhang L, Yang L L. Machine learning for joint channel equalization and signal detection[J]. Machine Learning for Future Wireless Communications, 2020: 213-241.

[27] Luan D, Thompson J. Low Complexity Channel estimation with Neural Network Solutions [C]//WSA 2021; 25th International ITG Workshop on Smart Antennas. VDE, 2021: 1-6.

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

[29] He H, Wen C K, Jin S, et al. Deep learning-based channel estimation for beamspace mmWave massive MIMO systems[J]. IEEE Wireless Communications Letters, 2018, 7(5): 852-855.

[30] Yassine T, Le Magoarou L. mpNet: variable depth unfolded neural network for massive MIMO channel estimation[J]. IEEE Transactions on Wireless Communications, 2022:1-1.

[31] Ma X, Gao Z, Gao F, et al. Model-driven deep learning based channel estimation and feedback for millimeter-wave massive hybrid MIMO systems[J]. IEEE Journal on Selected Areas in Communications, 2021, 39(8): 2388-2406.

[32] Fang L, Wu L. Deep learning detection method for signal demodulation in short range multipath channel[C]//2017 IEEE 2nd International Conference on Opto-Electronic Information Processing (ICOIP). IEEE, 2017: 16-20.

[33] Ahmed I, Xu W, Annavajjala R, et al. Joint Demodulation and Decoding with Multi-Label Classification Using Deep Neural Networks[C]//2021 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). IEEE, 2021: 365-370.

[34] Nachmani E, Wolf L. Hyper-graph-network decoders for block codes[J]. Advances in Neural Information Processing Systems, 2019:32.

[35] Deng C, Yuan S L B. Reduced-complexity deep neural network-aided channel code decoder: A case study for BCH decoder[C]//ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019: 1468-1472.

[36] Tang Y, Zhou L, Zhang S, et al. Modern-Driven Deep Learning for Belief Propagation LDPC Decoding[M]//Smart Communications, Intelligent Algorithms and Interactive Methods. Springer, Singapore, 2022: 251-257.

[37] Lin Y C, Lee T S, Ding Z. Deep Learning for Partial MIMO CSI Feedback by Exploiting Channel Temporal Correlation[J]. arXiv preprint arXiv, 2022:2201.02790.

[38] Ye H, Liang L, Li G Y. Circular convolutional auto-encoder for channel coding[C]//2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC). IEEE, 2019: 1-5.

[39] Ye H, Li G Y. Initial results on deep learning for joint channel equalization and decoding[C]//2017 IEEE 86th Vehicular Technology Conference (VTC-Fall). IEEE, 2017: 1-5.

[40] Ye H, Li G Y, Juang B H. Power of deep learning for channel estimation and signal detection in OFDM systems[J]. IEEE Wireless Communications Letters, 2017, 7(1): 114-117.

[41] Xu W, Zhong Z, Be'ery Y, et al. Joint neural network equalizer and decoder[C]//2018 15th International Symposium on Wireless Communication Systems (ISWCS). IEEE, 2018: 1-5.

[42] Cheng X, Liu D, Wang C, et al. Deep learning-based channel estimation and equalization scheme for FBMC/OQAM systems[J]. IEEE Wireless Communications Letters, 2019, 8(3): 881-884.

[43] Thaj T, Viterbo E. Low-Complexity Linear Diversity-Combining Detector for MIMO-OTFS[J]. IEEE Wireless Communications Letters, 2021, 11(2):288-292.

[44] Pinchera D, Migliore M D, Schettino F. Optimizing antenna arrays for spatial multiplexing: towards 6G systems[J]. IEEE Access, 2021, 9: 53276-53291.

[45] Perin G, Buratto A, Anselmi N M, et al. Adversarial jamming and catching games over AWGN channels with mobile players[C]//2021 17th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob). IEEE, 2021: 319-324.

[46] Wang M, Gao F, Jin S, et al. An overview of enhanced massive MIMO with array signal processing techniques[J]. IEEE Journal of Selected Topics in Signal Processing, 2019, 13(5): 886-901.

[47] Zhou D, Niu K, Dong C. Construction of polar codes in Rayleigh fading channel[J]. IEEE Communications Letters, 2019, 23(3): 402-405.

[48] Björnson E, Sanguinetti L. Rayleigh fading modeling and channel hardening for reconfigurable intelligent surfaces[J]. IEEE Wireless Communications Letters, 2020, 10(4): 830-834.

[49] 申敏, 李想, 林欢.一种用于电力线通信系统的改进MIMO检测算法[J].重庆邮电大学学报(自然科学版),2019,31(01):50-56.

[50] 陈发堂, 邓青, 石贝贝, 杨恒.一种面向WHT-NOMA的LC-MMSE-SIC检测算法[J].南京邮电大学学报(自然科学版),2020,40(04):11-16.

[51] Shah M H, Dang X. An effective approach for low-complexity maximum likelihood based automatic modulation classification of STBC-MIMO systems[J]. Frontiers of Information Technology & Electronic Engineering, 2020, 21(3): 465-475.

[52] Alamouti S M. A Simple Transmit Diversity Technique for Wireless Communications[J]. IEEE Journal on Selected Areas in Communications, 1998,16(8):1451-1458.

[53] Azevedo F A C, Carvalho L R B, Grinberg L T, et al. Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled‐up primate brain[J]. Journal of Comparative Neurology, 2009, 513(5): 532-541.

[54] Pospíchal J, Kvasnička V. 70th anniversary of publication: Warren mcculloch & walter pitts-a logical calculus of the ideas immanent in nervous activity[M]//Emergent Trends in Robotics and Intelligent Systems. Springer, Cham, 2015: 1-10.

[55] Agarap A F. Deep learning using rectified linear units (relu)[J]. arXiv preprint arXiv, 2018:1803.08375.

[56] Tang W, Long G, Liu L, et al. Rethinking 1d-cnn for time series classification: A stronger baseline[J]. arXiv preprint arXiv:2002.10061, 2020.

[57] Zheng S, Chen S, Yang X. DeepReceiver: a deep learning-based intelligent receiver for wireless communications in the physical layer[J]. IEEE Transactions on Cognitive Communications and Networking, 2020, 7(1): 5-20.

[58] 刘赛男, 李晖, 张蕊. 矿井MIMO-OFDM通信系统信道的建模与仿真[J].大连民族大学学报,2018,20(03):227-233.

[59] 朱静茹, 张育芝, 王安义, 李萍. 基于Q-学习算法的矿井自适应OFDM调制研究[J].工矿自动化,2021,47(06):109-115.