深度学习在认知无线电中的应用研究综述

刘波; 白晓东; 张更新; 沈俊; 谢继东; 赵来定; 洪涛

doi:10.3969/j.issn.1000-5641.201922017

深度学习在认知无线电中的应用研究综述

doi: 10.3969/j.issn.1000-5641.201922017

1.
南京邮电大学通信与信息工程学院，南京　210003
2.
西安空间无线电技术研究所，西安　710000

基金项目: 国家自然科学基金(61701260, 91738201)

详细信息

通讯作者:
白晓东，男，博士，讲师，研究方向为机器学习、模式识别、计算机视觉、数字信号处理. E-mail: xdbai@njupt.edu.cn

中图分类号: TN92
计量
- 文章访问数: 730
- HTML全文浏览量: 927
- PDF下载量: 134
- 被引次数: 0
出版历程
- 收稿日期: 2019-11-16
- 网络出版日期: 2020-06-06
- 刊出日期: 2021-01-27

Review of deep learning in cognitive radio

1.
College of Telecommunication and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing　210003, China
2.
CAST-Xi’an Institute of Space Radio Technology, Xi’an　710000, China

摘要

摘要: 无线通信业务的发展使得频谱资源变得越发紧张, 而现有的频谱利用效率却不高, 这一矛盾很大程度上可归结为频谱的静态分配策略. 认知无线电(Cognitive Radio, CR)技术被广泛认为是解决频谱静态分配问题的可行方案. 深度学习作为机器学习的新兴分支, 近几年在学术界和产业界都取得了许多成果, 成为人工智能的驱动性技术之一. 对深度学习在认知无线电中的应用进行了调研, 简要介绍了认知无线电和深度学习各自的发展, 且着重介绍了深度学习算法在频谱预测、频谱环境感知、信号分析等认知无线电关键技术环节中的应用, 并在最后对此进行了总结和探讨.
- 认知无线电 /
- 深度学习 /
- 卷积神经网络 /
- 循环神经网络 /
- 无线通信
Abstract: The development of wireless communication has made spectrum resources increasingly scarce. Existing spectrum resources, however, are not currently used in an efficient way. This contradiction can usually be attributed to the problem created by static spectrum allocation strategies. Cognitive radio (CR) is widely regarded as a feasible solution to solve the problem of static spectrum allocation. In recent years, deep learning, an emerging field of machine learning, has contributed to a number of notable research and application achievements. It has become one of the driving technologies behind artificial intelligence. In this paper, we investigated the application of deep learning to CR; this includes the development of cognitive radio and deep learning as well as the usage of deep learning models in key technologies for CR (such as spectrum prediction, spectrum environment sensing, signal analysis, etc.). Lastly, we summarize and discuss conclusions from this review.
- cognitive radio(CR) /
- deep learning /
- convolutional neural network /
- recurrent neural network /
- wireless communication

HTML全文

图 1 近年论文发表趋势

Fig. 1 Publication trends in recent years

下载: 全尺寸图片幻灯片

图 2 认知环路^[7]

Fig. 2 Cognitive loop^[7]

下载: 全尺寸图片幻灯片

图 3 认知引擎

Fig. 3 Cognitive engine

下载: 全尺寸图片幻灯片

图 4 CNN模型训练步骤

Fig. 4 CNN model training process

下载: 全尺寸图片幻灯片

图 5 典型CNN结构

Fig. 5 General CNN structure

下载: 全尺寸图片幻灯片

图 6 频谱图像形态学预处理^[11]

Fig. 6 Morphological preprocessing of spectral images^[11]

下载: 全尺寸图片幻灯片

图 7 递归神经网络结构

Fig. 7 Recursive neural network structure

下载: 全尺寸图片幻灯片

图 8 RNN与LSTM单元结构对比

Fig. 8 RNN and LSTM unit structure

下载: 全尺寸图片幻灯片

图 9 基于预测的频谱接入时序结构

Fig. 9 Timing structure based on the predicted spectrum access

下载: 全尺寸图片幻灯片

表 1 深度学习模型在CR中应用概况

Tab. 1 Deep learning model in CR

CR应用	深度学习模型	相关研究及文献
调制识别	CNN	[11-26]
	DBN	[27-29]
	RNN	GRU (Gate Recurrent Unit)^[30], LSTM^[31-32]
	其他	GAN(Generative Adversarial Networks)^[33], AE^[34-35]
频谱预测	LSTM	[36-45]
	CNN	[43,46]
	RNN	[47-48]
频谱感知	CNN	[49-55]
	DBN	[56-57]
	其他	DNN (Deep Neural Networks, DNN)^[58], LSTM+CNN^[59]
资源分配	CNN	[60-62]

下载: 导出CSV

表 2 特征选择与适用

Tab. 2 Features selection and their characteristics

特征表示	相关研究文献	特点
星座图	[14,17,20,25]	对高阶调制区分较好
循环平稳	[17,22]	适应噪声环境性能优
时频变换	[11,13,15-16]	覆盖波形种类多
瞬时特征	[24,31]	适用码率多变及大时延场景
特征自学习	[12,18-19,21,32]	预处理简单, 适用范围广

下载: 导出CSV

表 3 一些基于深度神经网络的频谱感知模型需要的先验知识

Tab. 3 Prior knowledges that some DNN based spectrum sensing models needed

相关研究	PU先验信息	PU统计模型
Tang等^[58]	码率, 调制
Han等^[52]	码率
Do等^[49]		HMM
Lee等^[53]	PU发射功率
Liu等^[54], Ke^[59], Xie等^[55]
Liu等^[51]	OFDM帧

下载: 导出CSV

参考文献(97)

[1]	武秀广, 宋旭, 任培明. 认知无线电关键技术及应用 [J]. 数字通信世界, 2014(2): 57-60. DOI: 10.3969/j.issn.1672-7274.2014.02.013.
[2]	TANDRA R, SAHAI A. Fundamental limits on detection in low SNR under noise uncertainty [C]//2005 International Conference on Wireless Networks, Communications and Mobile Computing. IEEE, 2005(1): 464-469. DOI: 10.1109/WIRLES.2005.1549453.
[3]	MITOLA J, MAGUIRE G Q. Cognitive radio: Making software radios more personal [J]. Software Radio Technologies: Selected Readings, 1999, 6(4): 413-418.
[4]	WU Q H, DING G R, WANG J L, et al. Spatial-temporal opportunity detection for spectrum-heterogeneous cognitive radio networks: Two-dimensional sensing [J]. IEEE Transactions on Wireless Communications, 2013, 12(2): 516-526. DOI: 10.1109/TWC.2012.122212.111638.
[5]	HAYKIN S. Cognitive radio: Brain-empowered wireless communications [J]. IEEE Journal on Selected Areas in Communications, 2005, 23(2): 201-220. DOI: 10.1109/JSAC.2004.839380.
[6]	郭永明, 李军芳, 李宁. 未来智能无线电将对频谱管理带来深刻影响 [J]. 电信科学, 2012, 28(7): 14-15.
[7]	DONG X, LI Y, WEI S Q. Design and implementation of a cognitive engine functional architecture [J]. Chinese Science Bulletin, 2012, 57(28): 3698-3704. DOI: 10.1007/s11434-012-5102-6.
[8]	CLANCY C, HECKER J, STUNTEBECK E, et al. Applications of machine learning to cognitive radio networks [J]. IEEE Wireless Communications, 2007, 14(4): 47-52. DOI: 10.1109/MWC.2007.4300983.
[9]	HINTON G, SALAKHUTDINOV R. Reducing the dimensionality of data with neural networks [J]. Science, 2006, 313(5786): 504-508. DOI: 10.1126/science.1127647.
[10]	陈远哲, 匡俊, 刘婷婷, 等. 共指消解技术综述 [J]. 华东师范大学学报(自然科学版), 2019(5): 16-35. DOI: 10.3969/j.issn.1000-5641.2019.05.002.
[11]	ZHANG M, DIAO M, GUO L M. Convolutional neural networks for automatic cognitive radio waveform recognition [J]. IEEE Access, 2017(5): 11074-11082. DOI: 10.1109/ACCESS.2017.2716191.
[12]	YASHASHWI K, SETHI A, CHAPORKAR P. A learnable distortion correction module for modulation recognition [J]. IEEE Wireless Communications Letters, 2019, 8(1): 77-80. DOI: 10.1109/LWC.2018.2855749.
[13]	GAO L, ZHANG X, GAO J, et al. Fusion image based radar signal feature extraction and modulation recognition [J]. IEEE Access, 2019(7): 13135-13148. DOI: 10.1109/ACCESS.2019.2892526.
[14]	WANG Y, LIU M, YANG J, et al. Data-driven deep learning for automatic modulation recognition in cognitive radios [J]. IEEE Transactions on Vehicular Technology, 2019, 68(4): 4074-4077. DOI: 10.1109/TVT.2019.2900460.
[15]	ZHANG Q, XU Z, ZHANG P. Modulation recognition using wavelet-assisted convolutional neural network [C]//International Conference on Advanced Technologies for Communications. IEEE, 2018: 100-104.
[16]	HIREMATH S M, DESHMUKH S, RAKESH R. Blind identification of radio access techniques based on time-frequency analysis and convolutional neural network [C]//TENCON 2018 − 2018 IEEE Region 10 Conference. IEEE, 2018: 1163-1167.
[17]	WU H, LI Y, ZHOU L, et al. Convolutional neural network and multi-feature fusion for automatic modulation classification [J]. Electronics Letters, 2019, 55(16): 895-897. DOI: 10.1049/el.2019.1789.
[18]	GU H, WANG Y, HONG S, et al. Blind channel identification aided generalized automatic modulation recognition based on deep learning [J]. IEEE Access, 2019(7): 110722-110729. DOI: 10.1109/ACCESS.2019.2934354.
[19]	YANG X. Fusion methods for CNN-based automatic modulation classification [J]. IEEE Access, 2019(7): 66496-66504. DOI: 10.1109/ACCESS.2019.2918136.
[20]	WANG D, ZHANG M, LI J, et al. Intelligent constellation diagram analyzer using convolutional neural network-based deep learning [J]. Optics Express, 2017, 25(15): 17150-17166. DOI: 10.1364/OE.25.017150.
[21]	O’SHEA T J, CORGAN J, CLANCY T C. Convolutional radio modulation recognition networks[C]//International Conference on Engineering Applications of Neural Networks 2016: Engineering Applications of Neural Networks. Cham: Springer, 2016: 213-226..
[22]	LI R, LI L, YANG S, et al. Robust automated VHF modulation recognition based on deep convolutional neural networks [J]. IEEE Communications Letters, 2018, 22(5): 946-949. DOI: 10.1109/LCOMM.2018.2809732.
[23]	YONGSHI W, JIE G, HAO L, et al. CNN-based modulation classification in the complicated communication channel [C]//2017 13th IEEE International Conference on Electronic Measurement & Instruments (ICEMI). 2017: 512-516.
[24]	NADEEM F, HOON C, GUONG S, et al. Automatic modulation format / bit-rate classification and signal-to-noise ratio estimation using asynchronous delay-tap sampling [J]. Computers and Electrical Engineering, Elsevier Ltd., 2015, 47: 126-133. DOI: 10.1016/j.compeleceng.2015.09.005.
[25]	PENG S L, JIANG H Y, WANG H X, et al. Modulation classification using convolutional neural network based deep learning model [C]//2017 26th Wireless and Optical Communication Conference (WOCC). IEEE, 2017. DOI: 10.1109/WOCC.2017.7929000.
[26]	SADEGHI M, LARSSON E G. Adversarial attacks on deep-learning based radio signal classification [J]. IEEE Wireless Communications Letters, 2019, 8(1): 213-216. DOI: 10.1109/LWC.2018.2867459.
[27]	ZHANG Y, LIU T, ZHANG L B, et al. A deep learning approach for modulation recognition [C]//2018 IEEE 23rd International Conference on Digital Signal Processing (DSP). IEEE, 2018. DOI: 10.1109/ICDSP.2018.8631811.
[28]	MENDIS G J, WEI J, MADANAYAKE A. Deep learning-based automated modulation classification for cognitive radio [C]//2016 IEEE International Conference on Communication Systems (ICCS). IEEE, 2016. DOI: 10.1109/ICCS.2016.7833571.
[29]	MENDIS G J, WEI J, MADANAYAKE A. Deep belief network for automated modulation classification in cognitive radio [C]//2017 Cognitive Communications for Aerospace Applications Workshop (CCAA). IEEE, 2017. DOI: 10.1109/CCAAW.2017.8001609.
[30]	HONG D H, ZHANG Z L, XU X D. Automatic modulation classification using recurrent neural networks [C]//2017 3rd IEEE International Conference on Computer and Communications (ICCC). IEEE, 2017: 695-700. DOI: 10.1109/CompComm.2017.8322633.
[31]	RAJENDRAN S, MEMBER S, MEERT W, et al. Deep learning models for wireless signal classification with distributed low-cost spectrum sensors [J]. IEEE Transactions on Cognitive Communications and Networking, 2018, 4(3): 433-445. DOI: 10.1109/TCCN.2018.2835460.
[32]	WEST N E, O’HEA T. Deep architectures for modulation recognition [C]//2017 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN). IEEE, 2017. DOI: 10.1109/DySPAN.2017.7920754.
[33]	TANG B, TU Y, ZHANG Z, et al. Digital signal modulation classification with data augmentation using generative adversarial nets in cognitive radio networks [J]. IEEE Access, 2018(6): 15713-15722. DOI: 10.1109/ACCESS.2018.2815741.
[34]	LI J, QI L, LIN Y, et al. Research on modulation identification of digital signals based on deep learning [C]//2016 IEEE International Conference on Electronic Information and Communication Technology (ICEICT). IEEE, 2016: 402-405.
[35]	DAI A, ZHANG H, SUN H. Automatic modulation classification using stacked sparse auto-encoders [C]//2016 IEEE 13th International Conference on Signal Processing (ICSP). IEEE, 2016: 248-252.
[36]	YU L, CHEN J, DING G R, et al. Spectrum prediction based on taguchi method in deep learning with long short-term memory [J]. IEEE Access, 2018(6): 45923-45933. DOI: 10.1109/ACCESS.2018.2864222.
[37]	HERNÁNDEZ J, LÓPEZ D, VERA N. Primary user characterization for cognitive radio wireless networks using long short-term memory [J]. International Journal of Distributed Sensor Networks, 2018, 14(11). DOI: 10.1177/1550147718811828.
[38]	ZUO P L, WANG X, LINGHU W D, et al. Prediction-based spectrum access optimization in cognitive radio networks [C]//2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC). IEEE, 2018. DOI: 10.1109/PIMRC.2018.8580726.
[39]	YU L, CHEN J, DING G R. Spectrum prediction via long short term memory [C]//2017 3rd IEEE International Conference on Computer and Communications, ICCC 2017. IEEE, 2018: 643-647.
[40]	AGARWAL A, DUBEY S, KHAN M A, et al. Learning based primary user activity prediction in cognitive radio networks for efficient dynamic spectrum access [C]//2016 International Conference on Signal Processing and Communications, SPCOM 2016. IEEE, 2016. DOI: 10.1109/SPCOM.2016.7746632.
[41]	SHAWEL B S, WOLDEGEBREAL D H, POLLIN S. Deep-learning based cooperative spectrum prediction for cognitive networks [C]//2018 International Conference on Information and Communication Technology Convergence (ICTC). IEEE, 2018: 133-137. DOI: 10.1109/ICTC.2018.8539570.
[42]	SHAWEL B S, WOLDEGEBREAL D H, POLLIN S. Convolutional LSTM-based long-term spectrum prediction for dynamic spectrum access [C]// 2019 27th European Signal Processing Conference (EUSIPCO). IEEE, 2019. DOI: 10.23919/EUSIPCO.2019.8902956.
[43]	OMOTERE O, FULLER J, QIAN L J, et al. Spectrum occupancy prediction in coexisting wireless systems using deep learning [C]//2018 IEEE 88th Vehicular Technology Conference (VTC-Fall). IEEE, 2018. DOI: 10.1109/VTCFall.2018.8690575.
[44]	ROY D, MUKHERJEE T, CHATTERJEE M, et al. Primary user activity prediction in DSA networks using recurrent structures [C]//2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN). IEEE, 2019. DOI: 10.1109/DySPAN.2019.8935716.
[45]	YU L X, WANG Q L, GUO Y F, et al. Spectrum availability prediction in cognitive aerospace communications: A deep learning perspective [C]//2017 Cognitive Communications for Aerospace Applications Workshop (CCAA). IEEE, 2017. DOI: 10.1109/CCAAW.2017.8001877.
[46]	YU L, CHEN J, ZHANG Y, et al. Deep spectrum prediction in high frequency communication based on temporal-spectral residual network [J]. China Communications, 2018, 15(9): 25-34. DOI: 10.1109/CC.2018.8456449.
[47]	AGARWAL A, GANGOPADHYAY R. Predictive spectrum occupancy probability-based spatio-temporal dynamic channel allocation map for future cognitive wireless networks [J]. Transactions on Emerging Telecommunications Technologies, 2018, 29(8): e3442. DOI: 10.1002/ett.3442.
[48]	TANG Z L, LI S M. Deep recurrent neural network for multiple time slot frequency spectrum predictions of cognitive radio [J]. KSII Transactions on Internet and Information Systems, 2017, 11(6): 3029-3045. DOI: 10.3837/tiis.2017.06.013.
[49]	DO V Q, KOO I. Learning frameworks for cooperative spectrum sensing and energy-efficient data protection in cognitive radio networks [J]. Applied Sciences, 2018, 8(5): 722-745. DOI: 10.3390/app8050722.
[50]	MERCHANT K, REVAY S, STANTCHEV G, et al. Deep learning for RF device fingerprinting in cognitive communication networks [J]. IEEE Journal on Selected Topics in Signal Processing, 2018, 12(1): 160-167. DOI: 10.1109/JSTSP.2018.2796446.
[51]	LIU H, ZHU X, FUJII T. Adversarial training for low-complexity convolutional neural networks using in spectrum sensing [C]//2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). IEEE, 2019: 478-483. DOI: 10.1109/ICAIIC.2019.8668844.
[52]	HAN D, SOBABE G C, ZHANG C J, et al. Spectrum sensing for cognitive radio based on convolution neural network [C]//2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2017. DOI: 10.1109/CISP-BMEI.2017.8302117.
[53]	LEE W, KIM M, CHO D H. Deep cooperative sensing: Cooperative spectrum sensing based on convolutional neural networks [J]. IEEE Transactions on Vehicular Technology, 2019, 68(3): 3005-3009. DOI: 10.1109/TVT.2019.2891291.
[54]	LIU C, WANG J, LIU X M, et al. Deep CM-CNN for spectrum sensing in cognitive radio [J]. IEEE Journal on Selected Areas in Communications, 2019, 37(10): 2306-2321. DOI: 10.1109/JSAC.2019.2933892.
[55]	XIE J D, LIU C, LIANG Y C, et al. Activity pattern aware spectrum sensing: A CNN-based deep learning approach [J]. IEEE Communications Letters, 2019, 23(6): 1025-1028. DOI: 10.1109/LCOMM.2019.2910176.
[56]	CUI Y H, JING X J, SUN S L, et al. Deep learning based primary user classification in cognitive radios [C]//2015 15th International Symposium on Communications and Information Technologies (ISCIT). IEEE, 2015: 165-168. DOI: 10.1109/ISCIT.2015.7458333.
[57]	SUN X K, GAO L, LUO X D, et al. RBM based cooperative Bayesian compressive spectrum sensing with adaptive threshold [C]//2016 IEEE/CIC International Conference on Communications in China (ICCC). IEEE, 2016. DOI: 10.1109/ICCChina.2016.7636844
[58]	TANG Y J, ZHANG Q Y, LIN W. Artificial neural network based spectrum sensing method for cognitive radio [C]//2010 6th International Conference on Wireless Communications Networking and Mobile Computing (WiCOM). IEEE, 2010. DOI: 10.1109/WICOM.2010.5601105.
[59]	KE D, HUANG Z T, WANG X, et al. Blind detection techniques for non-cooperative communication signals based on deep learning [J]. IEEE Access, 2019(7): 89218-89225. DOI: 10.1109/ACCESS.2019.2926296.
[60]	LEE W. Resource allocation for multi-channel underlay cognitive radio network based on deep neural network [J]. IEEE Communications Letters, IEEE, 2018, 22(9): 1942-1945. DOI: 10.1109/LCOMM.2018.2859392.
[61]	DU Y H, ZHANG F, XUE L. A kind of joint routing and resource allocation scheme based on prioritized memories-deep Q network for cognitive radio ad hoc networks [J]. Sensors, 2018, 18(7): 2119-2139. DOI: 10.3390/s18072119.
[62]	WANG S, LIU H, GOMES P H, et al. Deep reinforcement learning for dynamic multichannel access in wireless networks [J]. IEEE Transactions on Cognitive Communications and Networking, 2018, 4(2): 257-265. DOI: 10.1109/TCCN.2018.2809722.
[63]	LAWRENCE S, GILES C L, AH CHUNG TSOI, et al. Face recognition: A convolutional neural-network approach [J]. IEEE Transactions on Neural Networks, 1997, 8(1): 98-113. DOI: 10.1109/72.554195.
[64]	LECUN Y, FU JIE HUANG, BOTTOU L. Learning methods for generic object recognition with invariance to pose and lighting [C]//Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004, CVPR 2004. IEEE, 2004(2): 97-104.
[65]	ABDEL-HAMID O, MOHAMED A, JIANG H, et al. Applying convolutional neural networks concepts to hybrid NN-HMM model for speech recognition [C]//2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2012: 4277-4280. DOI: 10.1109/ICASSP.2012.6288864.
[66]	ABDEL-HAMID O, MOHAMED A, JIANG H, et al. Convolutional neural networks for speech recognition [J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014, 22(10): 1533-1545. DOI: 10.1109/TASLP.2014.2339736.
[67]	NIU X X, SUEN C Y. A novel hybrid CNN-SVM classifier for recognizing handwritten digits [J]. Pattern Recognition, 2012, 45(4): 1318-1325. DOI: 10.1016/j.patcog.2011.09.021.
[68]	LAUER F, SUEN C Y, BLOCH G. A trainable feature extractor for handwritten digit recognition [J]. Pattern Recognition, 2007, 40(6): 1816-1824. DOI: 10.1016/j.patcog.2006.10.011.
[69]	LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition [J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324. DOI: 10.1109/5.726791.
[70]	HAUSER S C, HEADLEY W C, MICHAELS A J. Signal detection effects on deep neural networks utilizing raw IQ for modulation classification [C]// MILCOM 2017 - 2017 IEEE Military Communications Conference (MILCOM). IEEE, 2017: 121-127. DOI: 10.1109/MILCOM.2017.8170853.
[71]	O’SHEA T J, HOYDIS J. An introduction to deep learning for the physical layer [J]. IEEE Transactions on Cognitive Communications and Networking, 2017, 3(4): 563-575. DOI: 10.1109/TCCN.2017.2758370.
[72]	SAHAI A, TANDRA R, MISHRA S M, et al. Fundamental design tradeoffs in cognitive radio systems[C]// Proceedings of the 1rst International Workshop on Technology and Policy for Accessing Spectrum. ACM, 2006. DOI: 10.1145/1234388.1234390.
[73]	郭彩丽, 张天魁, 曾志民, 等. 认知无线电关键技术及应用的研究现状 [J]. 电信科学, 2006(8): 50-55. DOI: 10.3969/j.issn.1000-0801.2006.08.013.
[74]	曾英, 李志勇, 张春平, 等. 新型协作频谱感知系统检测性能优化策略[J]. 华东师范大学学报(自然科学版), 2016(3): 84-91.
[75]	JIN S, ZHANG X. Compressive spectrum sensing for MIMO-OFDM based cognitive radio networks [C]//2015 IEEE Wireless Communications and Networking Conference (WCNC). IEEE, 2015: 2197-2202.
[76]	JAFAR A, SRINIVASA S. The throughput potential of cognitive radio [J]. IEEE Communications Magazine, 2007, 45(5): 73-79. DOI: 10.1109/MCOM.2007.358852.
[77]	SON K, JUNG B C, CHONG S, et al. Power allocation for OFDM-based cognitive radio systems under outage constraints [C]//2010 IEEE International Conference on Communications. IEEE, 2010. DOI: 10.1109/ICC.2010.5501790.
[78]	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, 2018, 7(1): 114-117. DOI: 10.1109/LWC.2017.2757490.
[79]	GALINDO-SERRANO A, GIUPPONI L. Distributed Q-learning for aggregated interference control in cognitive radio networks [J]. IEEE Transactions on Vehicular Technology, 2010, 59(4): 1823-1834. DOI: 10.1109/TVT.2010.2043124.
[80]	VENKATRAMAN P, HAMDAOUI B, GUIZANI M. Opportunistic bandwidth sharing through reinforcement learning [J]. IEEE Transactions on Vehicular Technology, 2010, 59(6): 3148-3153. DOI: 10.1109/TVT.2010.2048766.
[81]	LUND J. Reinforcement learning based distributed multiagent sensing policy for cognitive radio networks [C]//2011 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN). IEEE, 2011: 642-646.
[82]	WANG Z Y, SCHAUL T, HESSEL M, et al. Dueling network architectures for deep reinforcement learning [C]//Proceedings of the 33rd International Conference on International Conference on Machine Learning - Volume 48. JMLR, 2016: 1995-2003.
[83]	刘全, 翟建伟, 章宗长, 等. 深度强化学习综述[J]. 计算机学报, 2018, 41(1): 1-27.
[84]	VOEGTLIN T, DOMINEY P F. Linear recursive distributed representations [J]. Neural Networks, 2005, 18(7): 878-895. DOI: 10.1016/j.neunet.2005.01.005.
[85]	LECUN Y, BENGIO Y, HINTON G. Deep learning [J]. Nature, 2015, 521(7553): 436-444. DOI: 10.1038/nature14539.
[86]	HOCHREITER S, SCHMIDHUBER J. Long short-term memory [J]. Neural Computation, 1997, 9(8): 1735-1780. DOI: 10.1162/neco.1997.9.8.1735.
[87]	MACKAY D J C. A practical Bayesian framework for backpropagation networks [J]. Neural Computation, 1992, 4(3): 448-472. DOI: 10.1162/neco.1992.4.3.448.
[88]	UTSUGI A. Hyperparameter selection for self-organizing maps [J]. Neural Computation, 1997, 9(3): 623-635. DOI: 10.1162/neco.1997.9.3.623.
[89]	GRAHAM B M, ADLER A. Objective selection of hyperparameter for EIT [J]. Physiological Measurement, 2006, 27(5): S65-S79. DOI: 10.1088/0967-3334/27/5/S06.
[90]	CHEN Z, GUO N, HU Z, et al. Experimental validation of channel state prediction considering delays in practical cognitive radio [J]. IEEE Transactions on Vehicular Technology, 2011, 60(4): 1314-1325. DOI: 10.1109/TVT.2011.2116051.
[91]	LIN Z J, JIANG X Y, HUANG L F, et al. A energy prediction based spectrum sensing approach for cognitive radio networks [C]//2009 5th International Conference on Wireless Communications, Networking and Mobile Computing. IEEE, 2009. DOI: 10.1109/WICOM.2009.5302514.
[92]	CHEN Z. Channel state prediction in cognitive radio [C]// Cognitive Radio and Interference Management Technology and Strategy, 2012(15): 273-288. DOI: 10.4018/978-1-4666-2005-6.ch015.
[93]	WANG L C, WANG C W, CHANG C J. Modeling and analysis for spectrum handoffs in cognitive radio networks [J]. IEEE Transactions on Mobile Computing, 2012, 11(9): 1499-1513. DOI: 10.1109/TMC.2011.155.
[94]	DING G R, WANG J L, WU Q H, et al. On the limits of predictability in real-world radio spectrum state dynamics: From entropy theory to 5G spectrum sharing [J]. IEEE Communications Magazine, 2015, 53(7): 178-183. DOI: 10.1109/MCOM.2015.7158283.
[95]	XING X, JING T, CHENG W, et al. Spectrum prediction in cognitive radio networks [J]. IEEE Wireless Communications, 2013, 20(2): 90-96. DOI: 10.1109/MWC.2013.6507399.
[96]	ELTOM H, KANDEEPAN S, EVANS R J, et al. Statistical spectrum occupancy prediction for dynamic spectrum access: A classification [J]. EURASIP Journal on Wireless Communications and Networking, 2018(1): Article number 29. DOI: 10.1186/s13638-017-1019-8.
[97]	HINTON G E, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets [J]. Neural Computation, 2006, 18(7): 1527-1554. DOI: 10.1162/neco.2006.18.7.1527.