Article(id=1241697945901396222, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241697942122328272, articleNumber=null, orderNo=null, doi=10.12347/j.ycyk.20240606002, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1717603200000, receivedDateStr=2024-06-06, revisedDate=1720368000000, revisedDateStr=2024-07-08, acceptedDate=null, acceptedDateStr=null, onlineDate=1773973459935, onlineDateStr=2026-03-20, pubDate=1726329600000, pubDateStr=2024-09-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1773973459935, onlineIssueDateStr=2026-03-20, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1773973459935, creator=13701087609, updateTime=1773973459935, updator=13701087609, issue=Issue{id=1241697942122328272, tenantId=1146029695717560320, journalId=1238841944844054536, year='2024', volume='45', issue='5', pageStart='1', pageEnd='128', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1773973459034, creator=13701087609, updateTime=1773973945698, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1241699983414580120, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241697942122328272, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1241699983414580121, tenantId=1146029695717560320, journalId=1238841944844054536, issueId=1241697942122328272, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=73, endPage=81, ext={EN=ArticleExt(id=1241697947671392535, articleId=1241697945901396222, tenantId=1146029695717560320, journalId=1238841944844054536, language=EN, title=Automatic Modulation and Recognition of Communication Signals Based on CBAM-GRU, columnId=1239133503154081962, journalTitle=Journal of Telemetry, Tracking and Command, columnName=TT & C Communication and Navigation, runingTitle=null, highlight=null, articleAbstract=

A CBAM-GRU classification model based on the combination of Convolutional Attention Mechanism Module(CBAM) and Gated Recurrent Unit (GRU) network is investigated for automatic modulation identification in non-cooperative communication systems. The pre-processed time-domain amplitude, phase and I/Q values of the signal are combined and converted into a matrix of input sample values, which are entered into the network for signal classification and identification. Simulations are conducted using the RadioML2016. 10a radio dataset, and the CBAM-GRU model are compared with the Convolutional Neural Network (CNN), Long Short-Term Memory network (LSTM), GRU, and Convolutional Long Deep Neural Network (CLDNN). The results indicates that the classification accuracy of the CBAM- GRU model reaches 92.79%, showing improvements of 8.52%, 1.84%, 1.75%, and 8. 61% over the comparison models respectively. Compared to traditional CNN or LSTM models, the CBAM-GRU model is more effective in capturing spatio-temporal features of sig- nals, thereby enhancing recognition accuracy.

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Xiao YANG, Aiqin YAO, Yunqiang SUN, Xiling SHI, Wanting ZHANG), CN=ArticleExt(id=1241697953308537315, articleId=1241697945901396222, tenantId=1146029695717560320, journalId=1238841944844054536, language=CN, title=基于CBAM-GRU的通信信号自动调制识别, columnId=1239133503338631357, journalTitle=遥测遥控, columnName=测控通信与导航, runingTitle=null, highlight=null, articleAbstract=

本文研究了一种基于卷积注意力机制模块(CBAM)与门控循环单元网络(GRU)结合的CBAM-GRU分类模型,用于非合作通信系统中的自动调制识别技术。将信号预处理后的时域幅度值、相位值以及I/Q值合并,转换为输入采样值矩阵,进入网络进行信号分类识别。使用无线电数据集RadioML2016.10a进行仿真实验,并将CBAM-GRU模型与卷积神经网络(CNN)、长短期记忆网络(LSTM)、GRU、卷积长短时深度神经网络(CLDNN)进行比较。实验结果表明:CBAM-GRU模型的分类识别率达到92.79%,相较于对比模型分别提高了8.52%、1.84%、1.75%、8.61%,比传统的CNN或LSTM模型,在处理信号时能够更有效地捕捉时空特征,从而提高识别精度。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=S8YxzlS3Di4Q9aA8V1MYIw==, magXml=SvkE75nHTrvR/t5QpGtjgw==, pdfUrl=null, pdf=m6Z0tnm2gGiyH888jZV+Aw==, pdfFileSize=2531576, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=1yFqGtuUIJCA/m2PUlVYlw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=ftEp+lkGXirT6D/Qg2CDoA==, mapNumber=null, authorCompany=null, fund=null, authors=

杨宵 1999年生,硕士,研究员。

姚爱琴 1969年生,副教授,研究生导师。

孙运强 1963年生,教授,博士生导师。

石喜玲 1980年生,副教授,研究生导师。

张婉婷 1999年生,硕士,研究员。

, authorsList=杨宵, 姚爱琴, 孙运强, 石喜玲, 张婉婷)}, authors=[Author(id=1241712914801939089, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241712914978099867, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712914801939089, language=EN, stringName=Xiao YANG, firstName=Xiao, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Information and Communication, North University of China, Taiyuan 030051, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241712915095540383, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712914801939089, language=CN, stringName=杨宵, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中北大学 信息与通信学院 太原 030051, bio={"content":"

杨宵 1999年生,硕士,研究员。

"}, bioImg=null, bioContent=

杨宵 1999年生,硕士,研究员。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241712914642555527, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, xref=null, ext=[AuthorCompanyExt(id=1241712914659332745, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Information and Communication, North University of China, Taiyuan 030051), AuthorCompanyExt(id=1241712914671915658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中北大学 信息与通信学院 太原 030051)])]), Author(id=1241712915254923941, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241712915368170154, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712915254923941, language=EN, stringName=Aiqin YAO, firstName=Aiqin, middleName=null, lastName=YAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Information and Communication, North University of China, Taiyuan 030051, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241712915493999283, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712915254923941, language=CN, stringName=姚爱琴, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中北大学 信息与通信学院 太原 030051, bio={"content":"

姚爱琴 1969年生,副教授,研究生导师。

"}, bioImg=null, bioContent=

姚爱琴 1969年生,副教授,研究生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241712914642555527, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, xref=null, ext=[AuthorCompanyExt(id=1241712914659332745, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Information and Communication, North University of China, Taiyuan 030051), AuthorCompanyExt(id=1241712914671915658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中北大学 信息与通信学院 太原 030051)])]), Author(id=1241712915586273977, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241712915653382847, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712915586273977, language=EN, stringName=Yunqiang SUN, firstName=Yunqiang, middleName=null, lastName=SUN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Information and Communication, North University of China, Taiyuan 030051, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241712915766629064, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712915586273977, language=CN, stringName=孙运强, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中北大学 信息与通信学院 太原 030051, bio={"content":"

孙运强 1963年生,教授,博士生导师。

"}, bioImg=null, bioContent=

孙运强 1963年生,教授,博士生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241712914642555527, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, xref=null, ext=[AuthorCompanyExt(id=1241712914659332745, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Information and Communication, North University of China, Taiyuan 030051), AuthorCompanyExt(id=1241712914671915658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中北大学 信息与通信学院 太原 030051)])]), Author(id=1241712915917624013, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241712917343687380, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712915917624013, language=EN, stringName=Xiling SHI, firstName=Xiling, middleName=null, lastName=SHI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Information and Communication, North University of China, Taiyuan 030051, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241712917473710807, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712915917624013, language=CN, stringName=石喜玲, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中北大学 信息与通信学院 太原 030051, bio={"content":"

石喜玲 1980年生,副教授,研究生导师。

"}, bioImg=null, bioContent=

石喜玲 1980年生,副教授,研究生导师。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241712914642555527, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, xref=null, ext=[AuthorCompanyExt(id=1241712914659332745, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Information and Communication, North University of China, Taiyuan 030051), AuthorCompanyExt(id=1241712914671915658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中北大学 信息与通信学院 太原 030051)])]), Author(id=1241712917637288669, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1241712917767312102, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712917637288669, language=EN, stringName=Wanting ZHANG, firstName=Wanting, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Information and Communication, North University of China, Taiyuan 030051, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1241712917867975402, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, authorId=1241712917637288669, language=CN, stringName=张婉婷, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=中北大学 信息与通信学院 太原 030051, bio={"content":"

张婉婷 1999年生,硕士,研究员。

"}, bioImg=null, bioContent=

张婉婷 1999年生,硕士,研究员。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1241712914642555527, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, xref=null, ext=[AuthorCompanyExt(id=1241712914659332745, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Information and Communication, North University of China, Taiyuan 030051), AuthorCompanyExt(id=1241712914671915658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中北大学 信息与通信学院 太原 030051)])])], keywords=[Keyword(id=1241712918027358962, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, orderNo=1, keyword=Automatic modulation recognition), Keyword(id=1241712918107050744, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, orderNo=2, keyword=Non-cooperative communication systems), Keyword(id=1241712918178353918, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, orderNo=3, keyword=Convolutional block attention mechanism), Keyword(id=1241712918253851393, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, orderNo=4, keyword=Gated recurrent unit network), Keyword(id=1241712918367097605, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, orderNo=1, keyword=自动调制识别), Keyword(id=1241712918463566603, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, orderNo=2, keyword=非合作通信系统), Keyword(id=1241712918547452689, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, orderNo=3, keyword=卷积注意力机制), Keyword(id=1241712918627144471, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, orderNo=4, keyword=门控循环单元网络)], refs=[Reference(id=1241712923479954337, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2019, volume=23, issue=3, pageStart=454, pageEnd=457, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=CHEN W, XIE Z, MA L, journalName=IEEE Communications Letters, refType=null, unstructuredReference=CHEN W, XIE Z, MA L, et al. A faster maximum-likelihood modulation classification in flat fading non Gaussian channels[J]. IEEE Communications Letters, 2019, 23(3): 454-457., articleTitle=A faster maximum-likelihood modulation classification in flat fading non Gaussian channels, refAbstract=null), Reference(id=1241712923551257507, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2018, volume=22, issue=7, pageStart=1390, pageEnd=1393, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=ZHANG J, WANG F, ZHONG Z, journalName=IEEE Communications Letters, refType=null, unstructuredReference=ZHANG J, WANG F, ZHONG Z, et al. Continuous phase modulation classification via Baum-Welch algorithm[J]. IEEE Communications Letters, 2018, 22(7): 1390-1393., articleTitle=Continuous phase modulation classification via Baum-Welch algorithm, refAbstract=null), Reference(id=1241712923630949285, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=XIE X, NI Y, PENG S, journalName=null, refType=null, unstructuredReference=XIE X, NI Y, PENG S, et al. Deep learning based automatic modulation classification for varying SNR environment[C]//Proceedings of the 2019 28th Wireless and Optical Communications Conference (WOCC). 2019: 1-5., articleTitle=Deep learning based automatic modulation classification for varying SNR environment, refAbstract=null), Reference(id=1241712923765167017, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=213, pageEnd=226, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=O'SHEA T J, CORGAN J, CLANCY T C, journalName=null, refType=null, unstructuredReference=O'SHEA T J, CORGAN J, CLANCY T C. Convolutional radio modulation recognition networks[C]// Proceedings IEEE Int. Conf. Eng. Appl. Neural Netw., 2016: 213-226., articleTitle=Convolutional radio modulation recognition networks, refAbstract=null), Reference(id=1241712923899384748, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=695, pageEnd=700, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=HONG D, ZHANG Z, XU X, journalName=null, refType=null, unstructuredReference=HONG D, ZHANG Z, XU X. Automatic modulation classification using recurrent neural networks[C]//2017 3rd IEEE International Conference on Computer and Communications (ICCC). 2017: 695-700., articleTitle=Automatic modulation classification using recurrent neural networks, refAbstract=null), Reference(id=1241712923987465135, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2018, volume=3, issue=4, pageStart=433, pageEnd=445, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=RAJENDRAN S, MEERT W, GIUSTINIANO D, journalName=null, refType=null, unstructuredReference=RAJENDRAN S, MEERT W, GIUSTINIANO D, et al. Deep learning models for wireless signal classification with distributed low-cost spectrum sensors[C]// IEEE Transactions on Cognitive Communications and Networ-king. 2018, (4)3: 433-445., articleTitle=Deep learning models for wireless signal classification with distributed low-cost spectrum sensors, refAbstract=null), Reference(id=1241712924079739827, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=WEST N E, SHEA T O, journalName=null, refType=null, unstructuredReference=WEST N E, SHEA T O. Deep architectures for modulation recognition[C]//2017 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN).2017: 1-6., articleTitle=Deep architectures for modulation recognition, refAbstract=null), Reference(id=1241712924192986036, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=TEKBYK K, EKTI A R, GÖRÇIN A, journalName=null, refType=null, unstructuredReference=TEKBYK K, EKTI A R, GÖRÇIN A, et al. Robust and fast automatic modulation classification with CNN under multipath fading channels[C]//2020 IEEE 91st Vehicular Technology Conference(VTC2020-Spring). 2020: 1-6., articleTitle=Robust and fast automatic modulation classification with CNN under multipath fading channels, refAbstract=null), Reference(id=1241712924297843640, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=1, pageStart=73, pageEnd=79, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=葛战, 孙磊, 李兵, journalName=无线电工程, refType=null, unstructuredReference=葛战, 孙磊, 李兵, 等. 数据驱动的双通道CNN-LSTM调制分类算法[J]. 无线电工程, 2023, 53(1): 73-79., articleTitle=数据驱动的双通道CNN-LSTM调制分类算法, refAbstract=null), Reference(id=1241712924398506940, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2021, volume=61, issue=11, pageStart=1339, pageEnd=1343, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=向建, 高勇, journalName=电讯技术, refType=null, unstructuredReference=向建, 高勇. 基于GRU-CNN并联神经网络的自动调制识别[J]. 电讯技术, 2021, 61(11): 1339-1343., articleTitle=基于GRU-CNN并联神经网络的自动调制识别, refAbstract=null), Reference(id=1241712924494975934, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2021, volume=61, issue=11, pageStart=1339, pageEnd=1343, url=null, language=null, rfNumber=[10], rfOrder=10, authorNames=XIANG Jian, GAO Yong, journalName=null, refType=null, unstructuredReference=XIANG Jian, GAO Yong. Automatic modulation recognition based on GRU-CNN parallel neural network[J]. 2021, 61(11): 1339-1343., articleTitle=Automatic modulation recognition based on GRU-CNN parallel neural network, refAbstract=null), Reference(id=1241712924604027840, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2020, volume=69, issue=11, pageStart=13521, pageEnd=13531, url=null, language=null, rfNumber=[11], rfOrder=11, authorNames=ZHANG Z, LUO H, WANG C, journalName=IEEE Transactions on Vehicular Techno-logy, refType=null, unstructuredReference=ZHANG Z, LUO H, WANG C, et al. Automatic modulation classification using CNN-LSTM based dual-stream structure[J]. IEEE Transactions on Vehicular Techno-logy, 2020, 69(11): 13521-13531., articleTitle=Automatic modulation classification using CNN-LSTM based dual-stream structure, refAbstract=null), Reference(id=1241712924708885443, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2021, volume=57, issue=20, pageStart=150, pageEnd=156, url=null, language=null, rfNumber=[12], rfOrder=12, authorNames=付国栋, 黄进, 杨涛, journalName=计算机工程与应用, refType=null, unstructuredReference=付国栋, 黄进, 杨涛, 等. 改进CBAM的轻量级注意力模型[J]. 计算机工程与应用, 2021, 57(20): 150-156., articleTitle=改进CBAM的轻量级注意力模型, refAbstract=null), Reference(id=1241712924809548743, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2021, volume=57, issue=20, pageStart=150, pageEnd=156, url=null, language=null, rfNumber=[12], rfOrder=13, authorNames=FU Guodong, HUANG Jin, YANG Tao, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=FU Guodong, HUANG Jin, YANG Tao, et al. Improved lightweight attention model based on CBAM[J]. Computer Engineering and Applications, 2021, 57(20):150-156., articleTitle=Improved lightweight attention model based on CBAM, refAbstract=null), Reference(id=1241712926298526665, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2022, volume=52, issue=9, pageStart=1519, pageEnd=1524, url=null, language=null, rfNumber=[13], rfOrder=14, authorNames=高振斌, 张毅, 宿绍莹, journalName=无线电工程, refType=null, unstructuredReference=高振斌, 张毅, 宿绍莹. 基于CBAM-CNN-BiGR的Morse信号智能识别译码算法研究[J]. 无线电工程202252(9): 1519-1524., articleTitle=基于CBAM-CNN-BiGR的Morse信号智能识别译码算法研究, refAbstract=null), Reference(id=1241712926378218444, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=1, pageEnd=4, url=null, language=null, rfNumber=[14], rfOrder=15, authorNames=ZHOU C, SHI F, LIU Q, journalName=null, refType=null, unstructuredReference=ZHOU C, SHI F, LIU Q. Research on parameters estimation and suppression for C&I jamming[C]// Proceedings of the 2016 CIE International Conference on Radar(RADAR). 2016: 1-4., articleTitle=Research on parameters estimation and suppression for C&I jamming, refAbstract=null), Reference(id=1241712926478881741, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=16, authorNames=WOO S, PARK J, LEE J Y, journalName=null, refType=null, unstructuredReference=WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[C]//ECCV 2018, Lecture Notes in Computer Science, 2018: 11211., articleTitle=CBAM: Convolutional block attention module, refAbstract=null), Reference(id=1241712926596322255, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2020, volume=4, issue=5, pageStart=1, pageEnd=14, url=null, language=null, rfNumber=[16], rfOrder=17, authorNames=石磊, 王毅, 成颖, journalName=数据分析与知识发现, refType=null, unstructuredReference=石磊, 王毅, 成颖, 等. 自然语言处理中的注意力机制研究综述[J]. 数据分析与知识发现, 2020, 4(5): 1-14., articleTitle=自然语言处理中的注意力机制研究综述, refAbstract=null), Reference(id=1241712926789260242, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2020, volume=4, issue=5, pageStart=1, pageEnd=14, url=null, language=null, rfNumber=[16], rfOrder=18, authorNames=SHI Lei, WANG Yi, CHENG Ying, journalName=Data Analysis and Knowledge Discovery, refType=null, unstructuredReference=SHI Lei, WANG Yi, CHENG Ying, et al. Review of attention mechanism in natural language[J]. Data Analysis and Knowledge Discovery, 2020, 4(5): 1-14., articleTitle=Review of attention mechanism in natural language, refAbstract=null), Reference(id=1241712926877340632, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2019, volume=40, issue=11, pageStart=3151, pageEnd=3157, url=null, language=null, rfNumber=[17], rfOrder=19, authorNames=陈榕, 任崇广, 王智远, journalName=计算机工程与设计, refType=null, unstructuredReference=陈榕, 任崇广, 王智远, 等. 基于注意力机制的CRNN文本分类算法[J]. 计算机工程与设计, 2019, 40(11):3151-3157., articleTitle=基于注意力机制的CRNN文本分类算法, refAbstract=null), Reference(id=1241712926957032410, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=1358, pageEnd=1365, url=null, language=null, rfNumber=[18], rfOrder=20, authorNames=O'SHEA T J, WEST N, journalName=null, refType=null, unstructuredReference=O'SHEA T J, WEST N. Radio machine learning dataset generation with GNU radio[C]//Proceedings of the GNU Radio Conference. 2016: 1358-1365., articleTitle=Radio machine learning dataset generation with GNU radio, refAbstract=null), Reference(id=1241712927045112799, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, doi=null, pmid=null, pmcid=null, year=2018, volume=12, issue=1, pageStart=168, pageEnd=179, url=null, language=null, rfNumber=[19], rfOrder=21, authorNames=O'SHEA T J, ROY T, CLANCY T C, journalName=IEEE Journal of Selected Topics in Signal Processing, refType=null, unstructuredReference=O'SHEA T J, ROY T, CLANCY T C. Over-the-air deep learning based radio signal classification[J]. IEEE Journal of Selected Topics in Signal Processing, 2018, 12(1):168-179., articleTitle=Over-the-air deep learning based radio signal classification, refAbstract=null)], funds=[Fund(id=1241712923312182174, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, awardId=20210302123062, language=CN, fundingSource=山西省基础研究计划资助项目(20210302123062), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1241712914642555527, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, xref=null, ext=[AuthorCompanyExt(id=1241712914659332745, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Information and Communication, North University of China, Taiyuan 030051), AuthorCompanyExt(id=1241712914671915658, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, companyId=1241712914642555527, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=中北大学 信息与通信学院 太原 030051)])], figs=[ArticleFig(id=1241712918857831204, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.1, caption=System framework of AMR, figureFileSmall=bi5ybAxrrCdpK3ZTqazy/A==, figureFileBig=QOnYMVV8+Utp+5cOZdkyjA==, tableContent=null), ArticleFig(id=1241712918950105895, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图1, caption=AMR的系统框架, figureFileSmall=bi5ybAxrrCdpK3ZTqazy/A==, figureFileBig=QOnYMVV8+Utp+5cOZdkyjA==, tableContent=null), ArticleFig(id=1241712919059157806, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.2, caption=CBAM model structure, figureFileSmall=obfV2VCt1zv17Wxt2ob8ug==, figureFileBig=UBOlz9vOL+SlyqNVBloyBA==, tableContent=null), ArticleFig(id=1241712919138849584, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图2, caption=CBAM模型结构, figureFileSmall=obfV2VCt1zv17Wxt2ob8ug==, figureFileBig=UBOlz9vOL+SlyqNVBloyBA==, tableContent=null), ArticleFig(id=1241712919231124278, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.3, caption=GRU network structure, figureFileSmall=RnE/fm8ZxK9SYTtwb7/jXA==, figureFileBig=vwhnUkr+FZS41vBUkArB2Q==, tableContent=null), ArticleFig(id=1241712919302427451, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图3, caption=GRU网络结构, figureFileSmall=RnE/fm8ZxK9SYTtwb7/jXA==, figureFileBig=vwhnUkr+FZS41vBUkArB2Q==, tableContent=null), ArticleFig(id=1241712919407285054, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.4, caption=CBAM-GRU model structure, figureFileSmall=Zoy1mOTPUXVdQ6Hed5Xm6A==, figureFileBig=WJzcY2FYpPWLnJ2Yqj8M0Q==, tableContent=null), ArticleFig(id=1241712919516336961, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图4, caption=CBAM-GRU模型结构, figureFileSmall=Zoy1mOTPUXVdQ6Hed5Xm6A==, figureFileBig=WJzcY2FYpPWLnJ2Yqj8M0Q==, tableContent=null), ArticleFig(id=1241712919621194566, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.5, caption=Comparative experimental models structure, figureFileSmall=6h93YbQKCQ5ISSLZD9UMWg==, figureFileBig=tgbVDoyIoSWnvdN38gR4XQ==, tableContent=null), ArticleFig(id=1241712919734440778, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图5, caption=对比实验模型结构, figureFileSmall=6h93YbQKCQ5ISSLZD9UMWg==, figureFileBig=tgbVDoyIoSWnvdN38gR4XQ==, tableContent=null), ArticleFig(id=1241712919805743953, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.6, caption=Recognition curves of three types of networks, figureFileSmall=mQ8b69QYV/fzr/A2mqZNqw==, figureFileBig=sKIALMaIGBopY8VgL31qmg==, tableContent=null), ArticleFig(id=1241712919889630036, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图6, caption=三种网络的识别曲线, figureFileSmall=mQ8b69QYV/fzr/A2mqZNqw==, figureFileBig=sKIALMaIGBopY8VgL31qmg==, tableContent=null), ArticleFig(id=1241712919977710425, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.7, caption=Different models' accuracy curves, figureFileSmall=KZItEn3olY0tr4YqEIBiSg==, figureFileBig=m5G2Fm+xMiS+Z8nGTvBmOg==, tableContent=null), ArticleFig(id=1241712920074179422, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图7, caption=不同模型准确率曲线, figureFileSmall=KZItEn3olY0tr4YqEIBiSg==, figureFileBig=m5G2Fm+xMiS+Z8nGTvBmOg==, tableContent=null), ArticleFig(id=1241712920225174368, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.8, caption=Confusion matrix of CBAM-GRU, figureFileSmall=TQDYeigsmJ/WgaNG5lcXwg==, figureFileBig=VechGEnGnD68HJDmFdMDeA==, tableContent=null), ArticleFig(id=1241712920304866151, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图8, caption=CBAM-GRU混淆矩阵, figureFileSmall=TQDYeigsmJ/WgaNG5lcXwg==, figureFileBig=VechGEnGnD68HJDmFdMDeA==, tableContent=null), ArticleFig(id=1241712921806426985, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.9, caption=Confusion matrix of LSTM, figureFileSmall=Tw+IFqGeqFce6ty3/8SgFw==, figureFileBig=Od5543Jg2JqZ6T7WAPKIMg==, tableContent=null), ArticleFig(id=1241712921902895981, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图9, caption=LSTM混淆矩阵, figureFileSmall=Tw+IFqGeqFce6ty3/8SgFw==, figureFileBig=Od5543Jg2JqZ6T7WAPKIMg==, tableContent=null), ArticleFig(id=1241712922003559281, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.10, caption=Confusion matrix of GRU, figureFileSmall=f+I48z3ibFa+/ztjLZ0r+A==, figureFileBig=Q6yYAa1VyCyTb7oqcqP01Q==, tableContent=null), ArticleFig(id=1241712922074862451, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图10, caption=GRU混淆矩阵, figureFileSmall=f+I48z3ibFa+/ztjLZ0r+A==, figureFileBig=Q6yYAa1VyCyTb7oqcqP01Q==, tableContent=null), ArticleFig(id=1241712922183914357, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.11, caption=Confusion matrix of CLDNN, figureFileSmall=XFsg+Qfgh8SMMr42P/idBQ==, figureFileBig=KCFXpB4SPoLz9yMHMDpLrA==, tableContent=null), ArticleFig(id=1241712922255217528, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图11, caption=CLDNN混淆矩阵, figureFileSmall=XFsg+Qfgh8SMMr42P/idBQ==, figureFileBig=KCFXpB4SPoLz9yMHMDpLrA==, tableContent=null), ArticleFig(id=1241712922376852350, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Fig.12, caption=Confusion matrix of CNN, figureFileSmall=wJfC5ukaq/V84B5GyIWmNA==, figureFileBig=5ByLXI1EORLgTavn8H3QGQ==, tableContent=null), ArticleFig(id=1241712922469127040, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=图12, caption=CNN混淆矩阵, figureFileSmall=wJfC5ukaq/V84B5GyIWmNA==, figureFileBig=5ByLXI1EORLgTavn8H3QGQ==, tableContent=null), ArticleFig(id=1241712922532041603, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Table 1, caption=

RML2016.10a dataset parameters

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集RML2016.10.a
调制类型数字调制:BPSK, QPSK, 8PSK, 16QAM, 64QAM, BFSK, CPFSK, PAM4,
模拟调制:WBFM, AM-SSB, AM-DSB
信号格式同相和正交(IQ)[128,2]
噪声设置加性高斯白噪声,信噪比范围
[-20 dB,-18 dB,… ,18 dB]
信道环境选择性衰落(莱斯+瑞利)、
中心频率偏移、采样率偏移
总信号样本220 000
), ArticleFig(id=1241712922615927686, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=表1, caption=

RML2016.10a 数据集参数

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集RML2016.10.a
调制类型数字调制:BPSK, QPSK, 8PSK, 16QAM, 64QAM, BFSK, CPFSK, PAM4,
模拟调制:WBFM, AM-SSB, AM-DSB
信号格式同相和正交(IQ)[128,2]
噪声设置加性高斯白噪声,信噪比范围
[-20 dB,-18 dB,… ,18 dB]
信道环境选择性衰落(莱斯+瑞利)、
中心频率偏移、采样率偏移
总信号样本220 000
), ArticleFig(id=1241712922708202376, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Table 2, caption=

Comparison of the complexities of three types of networks

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型Acc_lowAcc_highAcc_all
CNN-GRU17.74%88.33%60.95 %
CNN-CBAM17.92%88.92%61.41%
CBAM-GRU18.69%91.74%62.03%
), ArticleFig(id=1241712922792088460, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=表2, caption=

三种网络实验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型Acc_lowAcc_highAcc_all
CNN-GRU17.74%88.33%60.95 %
CNN-CBAM17.92%88.92%61.41%
CBAM-GRU18.69%91.74%62.03%
), ArticleFig(id=1241712922888557453, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Table 3, caption=

Comparison of the complexities of three types of networks

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型参数量FLOPs/M
CNN-GRU564 9071 952.36
CNN-CBAM937 1312 045.05
CBAM-GRU492 4452 142.86
), ArticleFig(id=1241712922968249232, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=表3, caption=

三种网络的复杂度对比

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型参数量FLOPs/M
CNN-GRU564 9071 952.36
CNN-CBAM937 1312 045.05
CBAM-GRU492 4452 142.86
), ArticleFig(id=1241712923060523924, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=EN, label=Table 4, caption=

Compare network experimental results

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型Acc_lowAcc_highAcc_all参数量
CNN15.74%83.50%57.30%858 123
GRU16.95%89.62%60.63%151 179
LSTM17.24%89.76%60.54%201 099
CLDNN17.05%83.00%57.05%164 433
CBAM-GRU18.69%91.74%62.03%492 445
), ArticleFig(id=1241712923136021400, tenantId=1146029695717560320, journalId=1238841944844054536, articleId=1241697945901396222, language=CN, label=表4, caption=

对比网络实验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型Acc_lowAcc_highAcc_all参数量
CNN15.74%83.50%57.30%858 123
GRU16.95%89.62%60.63%151 179
LSTM17.24%89.76%60.54%201 099
CLDNN17.05%83.00%57.05%164 433
CBAM-GRU18.69%91.74%62.03%492 445
)], attaches=null, journal=Journal(id=1238822691516493831, delFlag=0, nameCn=遥测遥控, nameEn=Journal of Telemetry, Tracking and Command, nameHistory1=null, nameHistory2=null, issn=2095-1000, eissn=null, cn=11-1780/TP, coden=null, periodic=1, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=uUqWBk5pV4VWPULNoAHkuQ==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Telemetry, Tracking and Command, journalRemark=null, publicationField=null, createdTime=1773287945871, updatedTime=1773299995823, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=uUqWBk5pV4VWPULNoAHkuQ==, picEn=RDDslHHvDBDNnJMjk6BxdA==, jcr=null, cjcr=null, exts=[JournalExt(id=1238873232732909900, language=CN, name=遥测遥控, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1773299995835, updatedTime=1773299995835, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://manuscript.spacejournal.cn/ycyk, submissionEditorUrl=https://manuscript.spacejournal.cn/ycyk, submissionReviewUrl=https://manuscript.spacejournal.cn/ycyk, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1238873232779047245, language=EN, name=Journal of Telemetry, Tracking and Command, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1773299995846, updatedTime=1773299995846, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://manuscript.spacejournal.cn/ycyk, submissionEditorUrl=https://manuscript.spacejournal.cn/ycyk, submissionReviewUrl=https://manuscript.spacejournal.cn/ycyk, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1238841944844054536, websiteList=[Website(id=1238843830011744952, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1238841944844054536, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/ycyk/CN, language=CN, createTime=1773292985680, createBy=18614031015, updateTime=1773293577303, updateBy=18614031015, name=遥测遥控-中文, tplId=1146099689490845704, title=遥测遥控, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1238846390617568150, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=articleTextType, value=kx, createTime=1773293596176, updateTime=1773293596176, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390596596627, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=banner, value=null, createTime=1773293596171, updateTime=1773293596171, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390638539673, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=grayFlag, value=0, createTime=1773293596181, updateTime=1773293596181, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390588208018, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=logo, value=https://castjournals.cast.org.cn/joweb/ycyk/CN/file/pic?fileId=g9S2GGWx7iZjf51o9vpmmA==, createTime=1773293596169, updateTime=1773293596169, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390651122587, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=minRunFlag, value=0, createTime=1773293596184, updateTime=1773293596184, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390613373845, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/ycyk/CN/file/pic, createTime=1773293596175, updateTime=1773293596175, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390646928282, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=silenceFlag, value=0, createTime=1773293596183, updateTime=1773293596183, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390604985236, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1773293596173, updateTime=1773293596173, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390625956759, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=themeColor, value=null, createTime=1773293596178, updateTime=1773293596178, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846390634345368, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830011744952, code=themeStyle, value=null, createTime=1773293596180, updateTime=1773293596180, creator=18614031015, updator=18614031015)]), Website(id=1238843830078853819, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1238841944844054536, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/ycyk/EN, language=EN, createTime=1773292985696, createBy=18614031015, updateTime=1773293566698, updateBy=18614031015, name=遥测遥控-英文, tplId=1146101810881728533, title=Journal of Telemetry, Tracking and Command, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1238846419235304352, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=articleTextType, value=kx, createTime=1773293602999, updateTime=1773293602999, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419214332829, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=banner, value=null, createTime=1773293602994, updateTime=1773293602994, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419256275875, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=grayFlag, value=0, createTime=1773293603004, updateTime=1773293603004, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419205944220, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=logo, value=https://castjournals.cast.org.cn/joweb/ycyk/EN/file/pic?fileId=g9S2GGWx7iZjf51o9vpmmA==, createTime=1773293602992, updateTime=1773293602992, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419268858789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=minRunFlag, value=0, createTime=1773293603007, updateTime=1773293603007, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419231110047, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/ycyk/EN/file/pic, createTime=1773293602998, updateTime=1773293602998, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419264664484, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=silenceFlag, value=0, createTime=1773293603006, updateTime=1773293603006, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419222721438, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1773293602996, updateTime=1773293602996, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419243692961, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=themeColor, value=null, createTime=1773293603001, updateTime=1773293603001, creator=18614031015, updator=18614031015), WebsiteProps(id=1238846419252081570, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1238843830078853819, code=themeStyle, value=null, createTime=1773293603003, updateTime=1773293603003, creator=18614031015, updator=18614031015)])], journalTitle=遥测遥控, weixinUrl=null, journalUrl=https://ycyk.spacejournal.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Telemetry, Tracking and Command, journalPhotoCn=uUqWBk5pV4VWPULNoAHkuQ==, journalPhotoEn=RDDslHHvDBDNnJMjk6BxdA==, journalFirstLetter=J, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/ycyk/CN/10.12347/j.ycyk.20240606002, detailUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/10.12347/j.ycyk.20240606002, pdfUrlCn=https://castjournals.cast.org.cn/joweb/ycyk/CN/PDF/10.12347/j.ycyk.20240606002, pdfUrlEn=https://castjournals.cast.org.cn/joweb/ycyk/EN/PDF/10.12347/j.ycyk.20240606002, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于CBAM-GRU的通信信号自动调制识别
收藏切换
PDF下载
杨宵 , 姚爱琴 , 孙运强 , 石喜玲 , 张婉婷
遥测遥控 | 测控通信与导航 2024,45(5): 73-81
收起
收藏切换
遥测遥控 | 测控通信与导航 2024, 45(5): 73-81
基于CBAM-GRU的通信信号自动调制识别
全屏
杨宵, 姚爱琴, 孙运强, 石喜玲, 张婉婷
作者信息
  • 中北大学 信息与通信学院 太原 030051
  • 杨宵 1999年生,硕士,研究员。

    姚爱琴 1969年生,副教授,研究生导师。

    孙运强 1963年生,教授,博士生导师。

    石喜玲 1980年生,副教授,研究生导师。

    张婉婷 1999年生,硕士,研究员。

Automatic Modulation and Recognition of Communication Signals Based on CBAM-GRU
Xiao YANG, Aiqin YAO, Yunqiang SUN, Xiling SHI, Wanting ZHANG
Affiliations
  • School of Information and Communication, North University of China, Taiyuan 030051
出版时间: 2024-09-15 doi: 10.12347/j.ycyk.20240606002
文章导航
收藏切换

本文研究了一种基于卷积注意力机制模块(CBAM)与门控循环单元网络(GRU)结合的CBAM-GRU分类模型,用于非合作通信系统中的自动调制识别技术。将信号预处理后的时域幅度值、相位值以及I/Q值合并,转换为输入采样值矩阵,进入网络进行信号分类识别。使用无线电数据集RadioML2016.10a进行仿真实验,并将CBAM-GRU模型与卷积神经网络(CNN)、长短期记忆网络(LSTM)、GRU、卷积长短时深度神经网络(CLDNN)进行比较。实验结果表明:CBAM-GRU模型的分类识别率达到92.79%,相较于对比模型分别提高了8.52%、1.84%、1.75%、8.61%,比传统的CNN或LSTM模型,在处理信号时能够更有效地捕捉时空特征,从而提高识别精度。

自动调制识别  /  非合作通信系统  /  卷积注意力机制  /  门控循环单元网络

A CBAM-GRU classification model based on the combination of Convolutional Attention Mechanism Module(CBAM) and Gated Recurrent Unit (GRU) network is investigated for automatic modulation identification in non-cooperative communication systems. The pre-processed time-domain amplitude, phase and I/Q values of the signal are combined and converted into a matrix of input sample values, which are entered into the network for signal classification and identification. Simulations are conducted using the RadioML2016. 10a radio dataset, and the CBAM-GRU model are compared with the Convolutional Neural Network (CNN), Long Short-Term Memory network (LSTM), GRU, and Convolutional Long Deep Neural Network (CLDNN). The results indicates that the classification accuracy of the CBAM- GRU model reaches 92.79%, showing improvements of 8.52%, 1.84%, 1.75%, and 8. 61% over the comparison models respectively. Compared to traditional CNN or LSTM models, the CBAM-GRU model is more effective in capturing spatio-temporal features of sig- nals, thereby enhancing recognition accuracy.

Automatic modulation recognition  /  Non-cooperative communication systems  /  Convolutional block attention mechanism  /  Gated recurrent unit network
杨宵, 姚爱琴, 孙运强, 石喜玲, 张婉婷. 基于CBAM-GRU的通信信号自动调制识别. 遥测遥控, 2024 , 45 (5) : 73 -81 . DOI: 10.12347/j.ycyk.20240606002
Xiao YANG, Aiqin YAO, Yunqiang SUN, Xiling SHI, Wanting ZHANG. Automatic Modulation and Recognition of Communication Signals Based on CBAM-GRU[J]. Journal of Telemetry, Tracking and Command, 2024 , 45 (5) : 73 -81 . DOI: 10.12347/j.ycyk.20240606002
自动调制识别(Automatic Modulation Recognition,AMR)技术使接收机能够在非合作通信环境中自动识别信号的调制类型。该项技术在军事和民用领域得到了广泛应用,如频谱干扰检测、频谱感知和认知无线电(Cognitive Radio,CR)等。AMR本质上是一个分类问题,传统的AMR方法主要有基于似然比理论的方法[1,2]和基于特征提取[3]的方法。由于无线环境的日益复杂,信号类型多样且同一信号特征之间存在显著差异,传统的模式识别方法在特征提取方面需要大量人力成本,且信号分类的准确率有限。神经网络能够自动从大量信号样本中提取特征,有效解决了传统模式识别的缺点,并且可以联合多个网络进行识别,因此在AMR领域得到了越来越广泛的应用,神经网络已成为AMR研究的热点。
O'Shea等人[4]在2016年创建了RML2016.10a数据集,并引入卷积神经网络(Convolutional Neural Network,CNN)到AMR领域,深度学习逐渐在无线电通信领域得到应用。文献[5]使用循环神经网络(Recurrent Neural Network,RNN),利用接收到的通信信号的时间序列特征,避免了手动提取信号特征。与基于CNN的方法相比,该方法表现出明显优势,并选择了双层门控循环单元(Gated Recurrent Unit,GRU)网络的更有效结构,进一步提高了识别效果。文献[6]使用长短期记忆网络(Long Short Term Memory,LSTM)处理I/Q数据,并将其转换为幅度和相位信息,相较于直接输入I/Q数据,这种数据预处理方法取得了更优异的效果,凸显了数据预处理在自动调制识别中的重要性。文献[7]使用卷积长短时深度神经网络(Convolutional Long Short-Term Deep Neural Networks,CLDNN)进行分类识别,该模型结合了CNN和LSTM,能够处理原始时域波形, 实现更高效的调制识别。文献[8]改进了文献[4]的CNN作为训练模型,其结构包括4个卷积层和2个全连接层,表现出较低的算法复杂性和较短的训练时间。文献[9]引入了一种数据驱动的融合模型,通过将两个CNN通道的空间特征信息与LSTM的时序特征相互串联,实现了对调制信号的更精准分类。这种复合算法[10,11]结合CNN和RNN的优势网络结构,不仅增强了网络的容量和特征提取能力,还有效解决了梯度消失的问题。然而,这也导致了网络参数量和运行时间的增加。
针对此问题,本文将卷积注意力机制模块(Convolutional Block Attention Module,CBAM)[12,13]与门控循环单元相结合。CBAM模块的引入有助于有效降低无效目标的干扰,提升信号的分类识别性能。当CBAM与GRU网络相结合时,CBAM展现出对序列数据特征的优秀提取能力,有助于解决长期依赖问题。尽管已有研究利用深度学习进行自动调制识别取得了一定成果,但大多数方法未能有效结合空间和时间特征。本文通过引入卷积注意力机制,更好地提取了信号的时空特征,从而显著提升了识别率。与现有方法相比,模型性能在高信噪比条件下表现尤为突出。
通信系统中使用的AMR系统框架[14]图1所示,输入信号经过发射机调制成适合无线传输的射频信号,并通过通信信道传输。接收器捕获到信号后,将其解调为基带复数信号序列并进行预处理。
在接收端,已调信号模型见式(1),存在时变载波相位偏移。m(k)代表发送的基带信号序列,n(k)表示加性白高斯噪声,AЄ[0,1]表示信道增益,ϕ(k)表示相位偏移。
x(k) = Aejϕ(k)m(k) + n(k)
接收到的调制信号是IQ信号,包括同相分量XI,正交分量XQ
xk=XI+jXQ
原始信号可以通过其同相XI与正交分量XQ进行表示,同时也可以通过其幅度XA和相位XP来表示,元素可以由下式得到:
将采样点上的两个通道的值合并,使每个采样点信号值变为一个矩阵,此时信号的维度为k×4。
CBAM是一种用于增强CNN网络性能的注意力机制模块,由Sanghyun Woo等人[15]在2018年提出。这种机制模仿了人类视觉感知中的注意力机制,能够聚焦于有用信息、减少无效信息干扰,从而提高模型的效率和准确率[16]。就像人类的视觉一样,CBAM经过训练自动调整其配置参数,实现最佳性能。在图2中可见,CBAM引入通道注意力和空间注意力来增强CNN模型的感知能力,从而在不增加网络复杂性的情况下改善模型性能。通道注意力有助于增强不同通道的特征表示,而空间注意力则有助于提取空间中不同位置的关键信息。
如上图,给定中间特征图F,其中HWC分别代表高度、宽度和维度,F可以表示为FЄRC×H×W。对于通道注意力机制:首先,F进行全局平均池化和全局最大池化操作,计算每个通道的平均值和最大值。其次,分别将输入特征通过多层感知机(Multilayer Perceptron,MLP)进行处理。在此过程中,为了减少参数数量,使用衰减率r=8降低神经元的数量至C/r(其中C为原始通道数),激活函数选择ReLU。最后,通过Sigmoid激活函数融合通道权重,得到最终的通道注意力特征图MC(F),用于加权特征图中的每个通道;对于空间注意力机制: F′同样进行全局平均池化和全局最大池化操作,将通道维度压缩为1,并保留空间信息。随后,将这两个特征图进行拼接,并通过一个卷积层提取特征,最终经过Sigmoid激活,得到最终的空间注意力MS(F′)。
CBAM依次得到一个一维的通道注意力图MC(F)和一个二维的空间注意图MS(F′),整个流程可以概括为:
F′=MC(F)⊗F
F″=MS(F′)⊗F′
其中,表示对应元素相乘。通过对输入特征图进行通道和空间的双重注意力加权,CBAM能够更好地捕获信号中的关键信息,从而提升模型的表征能力和泛化性能。
门控循环单元(Gated Cycle Unit,GRU),是传统RNN的变体[17],与LSTM结构相似,均具有门控机制,用来调节单元内部的信息流动。同时与LSTM相比,GRU没有单独的存储单元,因此其结构和计算更为简洁,提高了运算效率,同时有效降低了训练过程中过拟合的风险。GRU的核心结构包括更新门(zt)和重置门(rt)两个部分,如图3所示。GRU网络状态更新过程如下:
① 更新门(Update Gate):
更新门决定了前一时刻的隐藏状态(记忆)有多少需要传递到当前时刻。它的计算公式为:
zt=σ(Wz·[ht-1,xt])
其中,xt是当前时刻的输入,ht-1是上一时刻的隐藏状态,Wz是权重矩阵,σ是sigmoid激活函数。
② 重置门(Reset Gate):
重置门决定了多少前一时刻的隐藏状态需要被重置,从而计算当前时刻的候选隐藏状态。它的计算公式为:
rt=σ(Wr·[ht-1,xt])
其中,Wr是权重矩阵,其他符号同上。
③ 候选隐藏状态(Candidate Hidden State):
利用重置门控制的上一时刻的隐藏状态来计算当前时刻的候选隐藏状态。计算公式为:
其中,W是权重矩阵,⊙表示元素逐位相乘,tanh是双曲正切激活函数。
④ 隐藏状态更新:
最终的隐藏状态通过更新门来综合前一时刻的隐藏状态和当前时刻的候选隐藏状态。计算公式为:
其中,(1-zt)和zt控制着保留前一时刻隐藏状态和采用候选隐藏状态的比例。
GRU通过上述过程有效地控制信息在时间步上的流动,能够在减少参数的同时保持对序列数据的强大建模能力,相较于LSTM具有更高的计算效率。
用于调制分类的CBAM-GRU的模型结构如图4所示,主要由4个CNN层、2个GRU层、CBAM层和2个全连接层组成。输入信号是由实部和虚部构成的实序列,即IQ数据,其格式为k×2,经过预处理后,维度转变为k×4的矩阵。数据首先被输入至CNN模块,该模块的前两层均使用256个卷积核,后两层则使用80个卷积核,所有卷积核的大小均为3×3,激活函数均为ReLU。提取空间特征信息后,数据被传递到由128个单元组成的双层GRU模块。为了防止过拟合,每个GRU层应用30%的Dropout。随后,使用两个密集全连接层进行进一步处理,这两个全连接层的神经元数量分别为64和11,最终的激活函数为Softmax,用于多类别分类。损失函数采用分类交叉熵,表示为:
式中,yi表示实际标签;ai表示预测输出;n表示样本总数量。
在通信调制信号的识别研究中,CNN的优势在于其能够有效地捕获局部特征,具有平移不变性,并且适用于二维数据的处理。LSTM的优势在于其能够捕捉长期依赖关系,适用于序列数据的建模和预测,如自然语言处理任务。GRU与LSTM相似,但参数更少,计算效率更高,适用于需要考虑长期依赖性的任务。卷积长短期记忆神经网络(Convolutional Long short term Deep Neural Networks,CLDNN)结合了CNN和LSTM的优点,既能够提取空间特征,又能够捕捉时间序列信息,适用于复杂的音频和语音识别任务。因此,本文选择了这几种常见的网络模型进行对比实验,如CNN[8]、LSTM[6]、GRU[5]、CLDNN[7]等,全面评估不同模型在调制信号识别中的性能。对应的四个模型结构分别如图5(a)图5(b)图5(c)图5(d)所示。
本文所有实验均使用GNU Radio生成的数据集RadioML2016.10a[18]。该数据集包含22万个输入样本,每个样本对应特定信噪比(Signal to Noise Ratio,SNR)下的一种调制类型。数据集中包含了各种实际的信道缺陷,如采样率偏移、信道频率偏移、多径衰落和高斯白噪声。该数据集包括11种调制方式,SNR范围为-20 dB~18 dB,步长为2 dB。每个信号样本的长度为128,数据格式为[128,2],其中2代表IQ两个通道维度。具体细节如表1所示。
在本文实验中,以6:2:2的比例将数据集划分为训练集、测试集和验证集。
实验环境配置如下:中央处理器(CPU)为Intel(R) Core(TM) i7-14700KF 3.40 GHz,图形处理器(GPU)为NVIDIA GeForce RTX 4080;实验在TensorFlow框架下使用Keras 2.6.0进行开发,编程语言为Python 3.8.18。基于训练结果,设置超参数,并使用Adam优化器,将学习率设定为0.001,批量大小设为256,训练周期设置为100个。
为了验证CBAM和GRU对模型性能的影响,本节进行了消融实验。实验分别使用了移除CBAM模块的CNN-GRU模型、移除GRU模块的CNN-CBAM模型以及同时添加了CBAM注意力模块和GRU网络的完整模型CBAM-GRU。表2展示了这三个网络模型在RadioML2016.10a数据集上的识别准确率。
实验结果表明,CBAM-GRU网络在低信噪比条件下的识别率为18.69%,在高信噪比条件下为91.74%,整体识别率为62.03%,显著优于消融CBAM模块或消融GRU模块的模型。移除CBAM模块后,整体识别率下降了1.08%;移除GRU模块后,整体识别率下降了0.62%。这些结果表明CBAM和GRU模块对模型性能的显著影响,且二者结合能够有效提升信号分类的准确性。表3展示了三种模型的参数量和浮点计算量。
通过对三个模型的比较,可以发现,CNN-CBAM模型拥有最多的参数,CBAM-GRU模型则具有最少的参数。尽管CBAM-GRU模型参数较少,但其FLOPs/M值却最高,高达2 142.86。这表明CBAM-GRU模型在相对较少的参数下实现了更高的计算效率,充分利用了注意力机制和循环神经网络的优点,既减少了模型复杂度,又提升了计算效率。
图6展示了三种模型在RadioML2016.10a数据集上的识别准确率曲线。当信号的信噪比超过8 dB时,CBAM-GRU网络的识别准确率稳定在92%以上。
使用CBAM-GRU网络与如CNN网络、LSTM网络、GRU网络、CLDNN网络进行对比实验。表4展示出5个网络的参数量和识别准确率。
实验结果显示,文献[7]设计的CLDNN网络和文献[8]设计的CNN网络对信号的识别准确率明显低于其他网络,这主要是由于这两个网络的卷积层较浅,网络结构相对简单,并且输入数据未进行预处理。相比之下,文献[5]和文献[6]均将I/Q数据进行幅度和相位的转换,并作为输入数据送入网络模型,显著提升了性能。在所有模型中,CBAM-GRU在通信调制信号识别任务中表现出明显优势,整体准确率达到62.03%,为最高。在高信噪比条件下,其准确率高达91.74%,显著优于其他模型;在低信噪比条件下,其准确率也达到18.69%,同样表现最佳。尽管CBAM-GRU的参数量较大,但其性能提升明显,表明适当增加模型复杂度和引入注意力机制能够显著提升识别能力。
图7展示了本文算法以及对比模型在不同信噪比下的识别准确率曲线。当信噪比(SNR)高于-4 dB时,CBAM-GRU模型显著优于其他模型。在4 dB的信噪比下,CBAM-GRU的识别率超过90%;在18 dB时,其最大识别率达到92.79%。在0~18 dB的范围内,CBAM-GRU的平均识别率为91.35%,相比其他方法提升了1%至6%。随着信噪比增加,5种方法的识别性能均逐渐提升,在0 dB时,五种模型的识别准确率稳定在80%左右。CBAM-GRU算法在高信噪比条件下表现尤为突出,优于其他模型。
为了深入分析CBAM-GRU模型在分类任务中的表现,图8图9图10图11图12展示了五种模型在6 dB信噪比下的混淆矩阵。混淆矩阵是评估分类模型性能的重要工具,每一列代表预测类别,每一行代表实际类别。它展示了分类模型在每个类别中的正确分类和错误分类数量,帮助更好地理解模型的准确率等指标,从而优化模型性能。
图8~图12可以看出,CBAM-GRU模型在对角线上的识别精度优于其他模型,且对角线外的杂块较少。相比之下,CLDNN的杂块明显更多,识别能力较低。
通过对比不同算法在信号识别中的表现,得出以下结论:首先,在高信噪比数据集上训练后,CBAM-GRU模型对64QAM信号的识别率相对较高,而其他四种算法则表现出明显的混淆现象;其次,超过一半的WBFM信号被错误归类为AM-DSB信号。这是因为WBFM和AM-DSB信号都属于连续调制,其在复数平面上的特征相似,导致算法无法准确区分。
CBAM-GRU模型相较于其他对比模型具有显著优势。这是由于引入了CBAM模块和双向GRU网络,CBAM能够有效学习信号的空间和通道信息,而GRU结构则能够有效捕获序列数据中的长期依赖关系,使得模型能够更好地分辨不同调制信号之间的细微差异。此外,CBAM-GRU模型的网络结构经过精心设计和调优,通过增加适当的复杂度和注意力机制,提高了模型的泛化能力和识别准确率。尽管参数量较大,但性能明显提升,特别是在高信噪比条件下,其识别准确率显著优于其他模型,展现了其在复杂信号环境中的卓越能力和鲁棒性。
本文研究了一种结合CBAM和GRU的时域联合AMR模型,旨在有效识别常见的无线通信调制信号。该模型利用采样数据的幅度和相位信息作为输入,通过深度学习网络对11种通信信号进行分类。实验结果表明:相较于传统的CNN、LSTM、GRU和CLDNN等网络模型,该模型在高信噪比环境下实现了卓越的信号识别性能。
随着电磁环境复杂度的不断增加,基于深度学习的自动调制识别方法受到了广泛关注。本文使用的数据集是RML 2016.10a,该数据集在调制类别数量上有所限制。因此,进一步研究可以考虑采用RML 2018.01a[19]和HisarMod 2019.1[8]数据集,这些数据集涵盖更多类别的信号,特别设计用于开发相应的联合识别网络。此外,未来研究中如何运用机器学习技术提升无线通信系统数据处理的安全性,是一项具有挑战性的课题。
  • 山西省基础研究计划资助项目(20210302123062)
参考文献 引证文献
排序方式:
[1]
CHEN W, XIE Z, MA L, et al. A faster maximum-likelihood modulation classification in flat fading non Gaussian channels[J]. IEEE Communications Letters, 2019, 23(3): 454-457.
[2]
ZHANG J, WANG F, ZHONG Z, et al. Continuous phase modulation classification via Baum-Welch algorithm[J]. IEEE Communications Letters, 2018, 22(7): 1390-1393.
[3]
XIE X, NI Y, PENG S, et al. Deep learning based automatic modulation classification for varying SNR environment[C]//Proceedings of the 2019 28th Wireless and Optical Communications Conference (WOCC). 2019: 1-5.
[4]
O'SHEA T J, CORGAN J, CLANCY T C. Convolutional radio modulation recognition networks[C]// Proceedings IEEE Int. Conf. Eng. Appl. Neural Netw., 2016: 213-226.
[5]
HONG D, ZHANG Z, XU X. Automatic modulation classification using recurrent neural networks[C]//2017 3rd IEEE International Conference on Computer and Communications (ICCC). 2017: 695-700.
[6]
RAJENDRAN S, MEERT W, GIUSTINIANO D, et al. Deep learning models for wireless signal classification with distributed low-cost spectrum sensors[C]// IEEE Transactions on Cognitive Communications and Networ-king. 2018, (4)3: 433-445.
[7]
WEST N E, SHEA T O. Deep architectures for modulation recognition[C]//2017 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN).2017: 1-6.
[8]
TEKBYK K, EKTI A R, GÖRÇIN A, et al. Robust and fast automatic modulation classification with CNN under multipath fading channels[C]//2020 IEEE 91st Vehicular Technology Conference(VTC2020-Spring). 2020: 1-6.
[9]
葛战, 孙磊, 李兵, 等. 数据驱动的双通道CNN-LSTM调制分类算法[J]. 无线电工程, 2023, 53(1): 73-79.
[10]
向建, 高勇. 基于GRU-CNN并联神经网络的自动调制识别[J]. 电讯技术, 2021, 61(11): 1339-1343.
XIANG Jian, GAO Yong. Automatic modulation recognition based on GRU-CNN parallel neural network[J]. 2021, 61(11): 1339-1343.
[11]
ZHANG Z, LUO H, WANG C, et al. Automatic modulation classification using CNN-LSTM based dual-stream structure[J]. IEEE Transactions on Vehicular Techno-logy, 2020, 69(11): 13521-13531.
[12]
付国栋, 黄进, 杨涛, 等. 改进CBAM的轻量级注意力模型[J]. 计算机工程与应用, 2021, 57(20): 150-156.
FU Guodong, HUANG Jin, YANG Tao, et al. Improved lightweight attention model based on CBAM[J]. Computer Engineering and Applications, 2021, 57(20):150-156.
[13]
高振斌, 张毅, 宿绍莹. 基于CBAM-CNN-BiGR的Morse信号智能识别译码算法研究[J]. 无线电工程202252(9): 1519-1524.
[14]
ZHOU C, SHI F, LIU Q. Research on parameters estimation and suppression for C&I jamming[C]// Proceedings of the 2016 CIE International Conference on Radar(RADAR). 2016: 1-4.
[15]
WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[C]//ECCV 2018, Lecture Notes in Computer Science, 2018: 11211.
[16]
石磊, 王毅, 成颖, 等. 自然语言处理中的注意力机制研究综述[J]. 数据分析与知识发现, 2020, 4(5): 1-14.
SHI Lei, WANG Yi, CHENG Ying, et al. Review of attention mechanism in natural language[J]. Data Analysis and Knowledge Discovery, 2020, 4(5): 1-14.
[17]
陈榕, 任崇广, 王智远, 等. 基于注意力机制的CRNN文本分类算法[J]. 计算机工程与设计, 2019, 40(11):3151-3157.
[18]
O'SHEA T J, WEST N. Radio machine learning dataset generation with GNU radio[C]//Proceedings of the GNU Radio Conference. 2016: 1358-1365.
[19]
O'SHEA T J, ROY T, CLANCY T C. Over-the-air deep learning based radio signal classification[J]. IEEE Journal of Selected Topics in Signal Processing, 2018, 12(1):168-179.
2024年第45卷第5期
PDF下载
105
43
引用本文
BibTeX
文章信息
doi: 10.12347/j.ycyk.20240606002
  • 接收时间:2024-06-06
  • 首发时间:2026-03-20
  • 出版时间:2024-09-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-06-06
  • 修回日期:2024-07-08
基金
山西省基础研究计划资助项目(20210302123062)
作者信息
    中北大学 信息与通信学院 太原 030051
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/ycyk/CN/10.12347/j.ycyk.20240606002
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏