留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于深度神经网络的高频谱效率频分复用系统的信道估计方法

陈嘉润 余宝贤 王剑莹 张涵

陈嘉润, 余宝贤, 王剑莹, 张涵. 基于深度神经网络的高频谱效率频分复用系统的信道估计方法[J]. 华南师范大学学报(自然科学版), 2020, 52(3): 17-21. doi: 10.6054/j.jscnun.2020038
引用本文: 陈嘉润, 余宝贤, 王剑莹, 张涵. 基于深度神经网络的高频谱效率频分复用系统的信道估计方法[J]. 华南师范大学学报(自然科学版), 2020, 52(3): 17-21. doi: 10.6054/j.jscnun.2020038
CHEN Jiarun, YU Baoxian, WANG Jianying, ZHANG Han. A DNN-Based Channel Estimation Method for Spectral Efficient Frequency Division Multiplexing Systems[J]. Journal of South China normal University (Natural Science Edition), 2020, 52(3): 17-21. doi: 10.6054/j.jscnun.2020038
Citation: CHEN Jiarun, YU Baoxian, WANG Jianying, ZHANG Han. A DNN-Based Channel Estimation Method for Spectral Efficient Frequency Division Multiplexing Systems[J]. Journal of South China normal University (Natural Science Edition), 2020, 52(3): 17-21. doi: 10.6054/j.jscnun.2020038

基于深度神经网络的高频谱效率频分复用系统的信道估计方法

doi: 10.6054/j.jscnun.2020038
基金项目: 

国家自然科学基金项目 61471176

教育部蓝火计划(惠州)产学研项目 CXZJHZ201705

广东省科技计划项目 2017A010101015

广东省科技计划项目 2017B030308009

广东省科技计划项目 2017KZ010101

广东省特支计划项目 2016TQ03X100

广东省自然科学基金项目 2018A030313990

广东省自然科学基金项目 2019A1515011940

广州市科技计划项目 2020-02-03-06-3008-0007

华南师范大学青年教师科研培育基金项目 19KJ16

详细信息
    通讯作者:

    余宝贤, 副研究员, Email:yubx@m.scnu.edu.cn

    张涵,教授,Email:zhanghan@scnu.edu.cn

  • 中图分类号: TN929.531

A DNN-Based Channel Estimation Method for Spectral Efficient Frequency Division Multiplexing Systems

  • 摘要: 针对高频谱效率频分复用(SEFDM)系统, 提出了一种基于深度神经网络(DNN)的信道估计方法.该方法使用等间隔相互正交的导频符号, 将其接收信号作为DNN的输入信号, 通过4层的全连接DNN结构提取信道特征, 最终实现了时域上的信道估计.仿真结果表明:提出的信道估计方法在同等条件下的均方误差(MSE)明显比最小二乘法(LS)的低, 对应的解调性能也更优, 且对导频数量具有更强的鲁棒性, 由此反映出该方法的优越性.
  • 图  1  SEFDM系统的框架

    Figure  1.  The framework of SEFDM system

    图  2  DNN的结构

    Figure  2.  The framework of DNN

    图  3  不同压缩因子α取值的MSE曲线

    Figure  3.  The MSE curves for different values of α

    图  4  不同压缩因子α取值时的BER曲线

    Figure  4.  The BER curves for different values of α

    图  5  不同导频数变化时的MSE曲线

    Figure  5.  The MSE curves when the number of pilots changes

    图  6  不同导频数变化时的BER曲线

    Figure  6.  The BER curves when the number of pilots changes

  • [1] YU B X, GUO C J, YI L Y, et al. 150-Gb/s SEFDM IM/DD transmission using log-MAP Viterbi decoding for short reach optical links[J]. Optics Express, 2018, 26(24):31075-31084. doi: 10.1364/OE.26.031075
    [2] 张涵, 郭昌霖, 李家明, 等.基于混合导频辅助的大规模MI-MO上行链路信道估计[J].华南师范大学学报(自然科学版), 2016, 48(6):57-62. http://journal-n.scnu.edu.cn/article/id/3882

    ZHANG H, GUO C L, LI J M, et al. Hybrid pilot-aided channel estimation in massive MIMO uplink[J]. Journal of South China Normal University (Natural Science Edition), 2016, 48(6):57-62. http://journal-n.scnu.edu.cn/article/id/3882
    [3] RODRIGUES M, DARWAZWH I. A spectrally efficient frequency division multiplexing based communications system[C]//2003 Proceedings of the international OFDM workshop. Hamburg: IEEE, 2003: 48-49.
    [4] NOPCHINDA D, XU T, MAHER R, et al. Dual polarization coherent optical spectrally efficient frequency division multiplexing[J]. Photonics Technology Letters, 2016, 28(1):83-86. doi: 10.1109/LPT.2015.2485669
    [5] YU B X, ZHANG H, HONG X D, et al. Channel equalisation and data detection for SEFDM over frequency selective fading channels[J]. IET Communications, 2018, 12:2315-2323. doi: 10.1049/iet-com.2018.5114
    [6] CHORTI A, KANARAS I, RODRIGUES M R, et al. Joint channel equalization and detection of spectrally efficient FDM signals[C]//2010 IEEE 21st International Symposium on Personal Indoor and Mobile Radio Commiunications(PIMRC). Blacksburg: IEEE, 2010: 177-182.
    [7] ISAM S, DARWAZEH I. Robust channel estimation for spectrally efficient FDM system[C]//2012 19th International Conference on Telecommunication(ICT). New York: IEEE, 2010: 177-182.
    [8] YU T, ZHAO M, ZHONG J, et al. Low-complexity graph-based turbo equalization for single-carrier FTN signaling[J]. IET Signal Processing, 2017, 11(7):838-845. doi: 10.1049/iet-spr.2016.0251
    [9] GHANNAM H, DARWAZEH I. Robust channel estimation methods for spectrally efficient FDM systems[C]//2018 IEEE 87th Vehicular Technology Conference(VTC Spring). New York: IEEE, 2018: 1-6.
    [10] 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. http://cn.bing.com/academic/profile?id=c5b0bb6adf4ec51f581a49298f347364&encoded=0&v=paper_preview&mkt=zh-cn
    [11] WANG X, GAO L, MAO S, et al. CSI-based fingerprinting for indoor localization:a deep learning approach[J]. IEEE Transactions on Vehicular Technology, 2017, 66(1):763-776. http://cn.bing.com/academic/profile?id=e1379d28a7ae970be05f46d29233fd93&encoded=0&v=paper_preview&mkt=zh-cn
    [12] CHEN S, GIBSON G J, COWN C F N, et al. Adaptive equalization of finite non-linear channels using multilayer perceptrons[J]. Signal Processing, 1990, 20(2):107-119. https://www.sciencedirect.com/science/article/pii/016516849090122F
    [13] NACHMANI E, BEERY Y, BURSHTEIN D. Learning to decode linear codes using deep learning[C]//54th Annual Allerton Conference on Communication, Control Computing. New York: IEEE, 2016: 341-346.
    [14] MAZO J E. Faster-than-nyquist signaling[J]. Bell System Technical Journal, 1975, 54(8):1451-1462. doi: 10.1002/j.1538-7305.1975.tb02043.x
    [15] YU B X, ZHANG H, DAI X H. A low-complexity demodulation technique for spectrally efficient FDM systems using decision-feedback[J]. IET Communications, 2017, 11:2386-2392. doi: 10.1049/iet-com.2017.0069
    [16] SCHMIDHUBER J. Deep learning in neural networks:an overview[J]. Neural Networks, 2015, 61:85-117. doi: 10.1016/j.neunet.2014.09.003
    [17] GLOROT X, BENGIO Y. Understanding the difficulty of training deep feedforward neural networks[C]//International Conference on Artificial Intelligence and Statistics. Sardinia: Journal of Machine Learning Research, 2010: 249-256.
  • 加载中
图(6)
计量
  • 文章访问数:  813
  • HTML全文浏览量:  536
  • PDF下载量:  54
  • 被引次数: 0
出版历程
  • 收稿日期:  2019-09-21
  • 刊出日期:  2020-06-25

目录

    /

    返回文章
    返回