Article(id=1148011771119526817, tenantId=1146029695717560320, journalId=1146119989267898375, issueId=1148087921783075097, articleNumber=null, orderNo=null, doi=10.7654/j.issn.2097-1974.20240513, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1712505600000, receivedDateStr=2024-04-08, revisedDate=1717516800000, revisedDateStr=2024-06-05, acceptedDate=null, acceptedDateStr=null, onlineDate=1751636935262, onlineDateStr=2025-07-04, pubDate=1729785600000, pubDateStr=2024-10-25, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1751636935262, onlineIssueDateStr=2025-07-04, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1751636935262, creator=13701087609, updateTime=1751636935262, updator=13701087609, issue=Issue{id=1148087921783075097, tenantId=1146029695717560320, journalId=1146119989267898375, year='2024', volume='47', issue='5', pageStart='1', pageEnd='106', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1751655090995, creator=13701087609, updateTime=1754895903201, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1161680873427390506, tenantId=1146029695717560320, journalId=1146119989267898375, issueId=1148087921783075097, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1161680873427390507, tenantId=1146029695717560320, journalId=1146119989267898375, issueId=1148087921783075097, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=92, endPage=98, ext={EN=ArticleExt(id=1148011771283104686, articleId=1148011771119526817, tenantId=1146029695717560320, journalId=1146119989267898375, language=EN, title=A Spectrum Sensing Method Integrating DenseNet and MLP-Mixer, columnId=1154067658359165250, journalTitle=Missiles and Space Vehicles, columnName=Simulation and Experimental Research, runingTitle=null, highlight=null, articleAbstract=

Along with the surge of radio application, electronic communication in interference environments has become increasingly important. The spectrum sensing technique matters in surmounting the frequency conflict of radio. However, the complex environment hinders the efficient feature extraction from the received spectrum signal and reduces the signal practicality. Recently, the artificial intelligence has been widespread in communication field and crucially influenced the electronic countermeasures. Consequently, based on the deep learning, this work proposes a spectrum sensing method to mix DenseNet and MLP-Mixer. Firstly, the model processes and transforms the spectrum signal data to feature images by Deepinsight Net and the generative adversarial networks renew an image. After obtaining the feature image, aspectrum sensing method integrating DenseNet and MLP-Mixer is used in order to sense the channel occupancy of primary user. Compared with the existing model through ablation experiments, the proposed method improves the detection probability of spectrum sensing better.

, 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=Zuo TIAN, Jing CAI, Yiyang HUO), CN=ArticleExt(id=1148011775196389657, articleId=1148011771119526817, tenantId=1146029695717560320, journalId=1146119989267898375, language=CN, title=融合稠密连接网络与MLP-Mixer的频谱感知方法, columnId=1154067658497577283, journalTitle=导弹与航天运载技术, columnName=仿真与试验研究专栏, runingTitle=null, highlight=null, articleAbstract=

随着无线应用的持续激增,干扰环境下的通信变得愈发重要,频谱感知技术在解决无线电用频冲突方面发挥了重要作用。然而实际应用环境复杂,获取到的频谱信号不易被高效提取特征,这降低了频谱信号的实用性。如今人工智能在通信领域应用广泛,对通信技术产生重要影响。为此,从深度学习方法入手,提出一种融合稠密连接网络与MLP-Mixer的频谱感知方法。该模型首先通过Deepinsight网络对频谱信号数据实施处理与转换,使其变换为特征图像,再使用生成式对抗网络合成新的特征图,并在得到特征图像后,采用融合稠密连接网络的混合感知器提取特征,从而感知主用户信道占用情况。经过消融试验对比,所提方法相较于已有模型,较好地提升了频谱感知的检测概率。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=4MYrE7X3WQoekd14HhJFOQ==, magXml=ZDSLy1JTYLaWDRcpVz4LpA==, pdfUrl=null, pdf=dcJ5IcGGLVXv3wiDsJok+A==, pdfFileSize=1956925, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=JqkaIfpNIQdO5ygLA8Eefg==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=HTdiMRWQSXrN+/81Hcojcg==, mapNumber=null, authorCompany=null, fund=null, authors=

田左(1995—),男,工程师,主要研究方向为遥感图像处理。

蔡静(1993-),女,博士,工程师,主要研究方向为多源信息融合技术。

霍熠阳(1998—),男,助理工程师,主要研究方向为目标检测与识别。

, authorsList=田左, 蔡静, 霍熠阳)}, authors=[Author(id=1197272483754717400, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, 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=1197272483964432603, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, authorId=1197272483754717400, language=EN, stringName=Zuo TIAN, firstName=Zuo, middleName=null, lastName=TIAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Space Long March Vehicle,Beijing,100076, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1197272484039930076, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, authorId=1197272483754717400, language=CN, stringName=田左, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京航天长征飞行器研究所,北京,100076, bio={"content":"

田左(1995—),男,工程师,主要研究方向为遥感图像处理。

"}, bioImg=null, bioContent=

田左(1995—),男,工程师,主要研究方向为遥感图像处理。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1197272483427561684, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, xref=null, ext=[AuthorCompanyExt(id=1197272483435950293, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Space Long March Vehicle,Beijing,100076), AuthorCompanyExt(id=1197272483452727510, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京航天长征飞行器研究所,北京,100076)])]), Author(id=1197272484283199710, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, 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=1197272484551635168, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, authorId=1197272484283199710, language=EN, stringName=Jing CAI, firstName=Jing, middleName=null, lastName=CAI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Space Long March Vehicle,Beijing,100076, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1197272484794904801, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, authorId=1197272484283199710, language=CN, stringName=蔡静, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京航天长征飞行器研究所,北京,100076, bio={"content":"

蔡静(1993-),女,博士,工程师,主要研究方向为多源信息融合技术。

"}, bioImg=null, bioContent=

蔡静(1993-),女,博士,工程师,主要研究方向为多源信息融合技术。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1197272483427561684, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, xref=null, ext=[AuthorCompanyExt(id=1197272483435950293, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Space Long March Vehicle,Beijing,100076), AuthorCompanyExt(id=1197272483452727510, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京航天长征飞行器研究所,北京,100076)])]), Author(id=1197272485017202915, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, 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=1197272485222723813, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, authorId=1197272485017202915, language=EN, stringName=Yiyang HUO, firstName=Yiyang, middleName=null, lastName=HUO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Space Long March Vehicle,Beijing,100076, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1197272485398884582, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, authorId=1197272485017202915, language=CN, stringName=霍熠阳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京航天长征飞行器研究所,北京,100076, bio={"content":"

霍熠阳(1998—),男,助理工程师,主要研究方向为目标检测与识别。

"}, bioImg=null, bioContent=

霍熠阳(1998—),男,助理工程师,主要研究方向为目标检测与识别。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1197272483427561684, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, xref=null, ext=[AuthorCompanyExt(id=1197272483435950293, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Space Long March Vehicle,Beijing,100076), AuthorCompanyExt(id=1197272483452727510, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京航天长征飞行器研究所,北京,100076)])])], keywords=[Keyword(id=1197272486879473897, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, orderNo=1, keyword=spectrum sensing), Keyword(id=1197272487135326443, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, orderNo=2, keyword=deep learning), Keyword(id=1197272487345041645, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, orderNo=3, keyword=signal transformation), Keyword(id=1197272487575728367, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, orderNo=4, keyword=generative adversarial), Keyword(id=1197272487802220785, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, orderNo=5, keyword=feature extraction), Keyword(id=1197272488053879027, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, orderNo=1, keyword=频谱感知), Keyword(id=1197272488330703093, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, orderNo=2, keyword=深度学习), Keyword(id=1197272488506863862, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, orderNo=3, keyword=信号转换), Keyword(id=1197272488821436663, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, orderNo=4, keyword=生成对抗), Keyword(id=1197272489085677816, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, orderNo=5, keyword=特征提取)], refs=[Reference(id=1197272496517984554, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2018, volume=66, issue=21, pageStart=5634, pageEnd=5647, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=XU Y, CHENG P, CHEN Z, journalName=IEEE Transactions on Signal Processing, refType=null, unstructuredReference=XU Y, CHENG P, CHEN Z, et al. Learning and transferring mid-level image representations using convolutional neural networks[J]. IEEE Transactions on Signal Processing, 2018,66(21): 5634-5647., articleTitle=Learning and transferring mid-level image representations using convolutional neural networks, refAbstract=null), Reference(id=1197272496773837100, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=10, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=PAN G, LI J, LIN F, journalName=International Journal of Digital Multimedia Broadcasting, refType=null, unstructuredReference=PAN G, LI J, LIN F. A cognitive radio spectrum sensing method for an OFDMsignal based on deep learning and cycle spectrum[J]. International Journal of Digital Multimedia Broadcasting, 2020(10): 1-10., articleTitle=A cognitive radio spectrum sensing method for an OFDMsignal based on deep learning and cycle spectrum, refAbstract=null), Reference(id=1197272496958386478, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2019, volume=16, issue=2, pageStart=162, pageEnd=171, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=ZHU Y, LIU J, FENG Z, journalName=Journal of Communications and Networks, refType=null, unstructuredReference=ZHU Y, LIU J, FENG Z, et al. Sensing performance of efficient cyclostationary detector with multiple antennas in multipath fading and lognormal shadowing environments[J]. Journal of Communications and Networks, 2019,16(2): 162-171., articleTitle=Sensing performance of efficient cyclostationary detector with multiple antennas in multipath fading and lognormal shadowing environments, refAbstract=null), Reference(id=1197272497235210544, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=WANG H, WANG J, journalName=2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC), refType=null, unstructuredReference=WANG H, WANG J. Short term wind speed forecasting based on feature extraction by CNN[C]// 2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC). Nanjing: 2021., articleTitle=Short term wind speed forecasting based on feature extraction by CNN, refAbstract=null), Reference(id=1197272497407177010, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2020, volume=32, issue=2, pageStart=604, pageEnd=624, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=OTTER D W, MEDINA J R, KALITA J K, journalName=IEEE transactions on neural networks and learning systems, refType=null, unstructuredReference=OTTER D W, MEDINA J R, KALITA J K. A survey of the usages of deep learning for natural language processing[J]. IEEE transactions on neural networks and learning systems, 2020,32(2): 604-624., articleTitle=A survey of the usages of deep learning for natural language processing, refAbstract=null), Reference(id=1197272497520423220, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=DHANANJAYA S, YUVARAJU B N, journalName=2018 International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT), refType=null, unstructuredReference=DHANANJAYA S, YUVARAJU B N. A novel method in matched filter spectrum sensing to minimize interference from compromised secondary users of cognitive radio networks[C]// 2018 International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT). Msyuru: 2018., articleTitle=A novel method in matched filter spectrum sensing to minimize interference from compromised secondary users of cognitive radio networks, refAbstract=null), Reference(id=1197272497587532085, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=NALLAGONDA S, KUMAR Y R, SHILPA P, journalName=2017 IEEE 7th International Advance Computing Conference (IACC), refType=null, unstructuredReference=NALLAGONDA S, KUMAR Y R, SHILPA P. Analysis of hard-decision and soft-data fusion schemes for cooperative spectrum sensing in rayleigh fading channel[C]// 2017 IEEE 7th International Advance Computing Conference (IACC). Hyderabad: 2017., articleTitle=Analysis of hard-decision and soft-data fusion schemes for cooperative spectrum sensing in rayleigh fading channel, refAbstract=null), Reference(id=1197272497654640951, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2012, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=杜红, journalName=认知无线电中频谱感知优化与无线资源管理的研究, refType=null, unstructuredReference=杜红. 认知无线电中频谱感知优化与无线资源管理的研究[D]. 北京: 北京邮电大学, 2012., articleTitle=null, refAbstract=null), Reference(id=1197272497868550456, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2012, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=8, authorNames=DU Hong, journalName=Research on spectrum sensing optimization and radio resource management in cognitive radio, refType=null, unstructuredReference=DU Hong. Research on spectrum sensing optimization and radio resource management in cognitive radio[D]. Beijing: Beijing University of Posts and Telecommunications, 2012., articleTitle=null, refAbstract=null), Reference(id=1197272498170540346, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=9, authorNames=胡泽锋, journalName=单通道全双工光通信系统及关键器件研究, refType=null, unstructuredReference=胡泽锋. 单通道全双工光通信系统及关键器件研究[D]. 南京: 南京邮电大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1197272498451558716, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=HU Zefeng, journalName=Research on single-channel full-duplex optical communication system and key devices, refType=null, unstructuredReference=HU Zefeng. Research on single-channel full-duplex optical communication system and key devices[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2023., articleTitle=null, refAbstract=null), Reference(id=1197272498745159997, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2024, volume=3, issue=2, pageStart=1, pageEnd=14, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=李柯, 刘文忠, 秦镜淘, journalName=宜宾学院学报, refType=null, unstructuredReference=李柯, 刘文忠, 秦镜淘. 改进TransUNet网络对肝脏肿瘤CT图像的级联分割[J]. 宜宾学院学报, 2024,3(2): 1-14., articleTitle=改进TransUNet网络对肝脏肿瘤CT图像的级联分割, refAbstract=null), Reference(id=1197272500037005631, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2024, volume=3, issue=2, pageStart=1, pageEnd=14, url=null, language=null, rfNumber=[10], rfOrder=12, authorNames=LI Ke, LIU Wenzhong, QIN Jingtao, journalName=Journal of Yibin University, refType=null, unstructuredReference=LI Ke, LIU Wenzhong, QIN Jingtao. Improved cascade segmentation of liver tumor CT images by TransUNet network[J]. Journal of Yibin University, 2024,3(2): 1-14., articleTitle=Improved cascade segmentation of liver tumor CT images by TransUNet network, refAbstract=null), Reference(id=1197272500250915137, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2024, volume=7, issue=1, pageStart=23, pageEnd=30, url=null, language=null, rfNumber=[11], rfOrder=13, authorNames=SHENG H, HUAKUN Q, YINGNAN Z, journalName=Energy Informatics, refType=null, unstructuredReference=SHENG H, HUAKUN Q, YINGNAN Z, et al. Automatic grid topology detection method based on Lasso algorithm and t-SNE algorithm[J]. Energy Informatics, 2024,7(1): 23-30., articleTitle=Automatic grid topology detection method based on Lasso algorithm and t-SNE algorithm, refAbstract=null), Reference(id=1197272500607430979, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2024, volume=7, issue=3, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[12], rfOrder=14, authorNames=ZHAO J, HE D, GE M, journalName=Information Fusion, refType=null, unstructuredReference=ZHAO J, HE D, GE M, et al. GA-GGD: improving semantic discriminability in graph contrastive learning via Generative Adversarial Network[J]. Information Fusion, 2024,7(3): 1-10., articleTitle=GA-GGD: improving semantic discriminability in graph contrastive learning via Generative Adversarial Network, refAbstract=null), Reference(id=1197272500859089221, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=1, pageStart=9, pageEnd=14, url=null, language=null, rfNumber=[13], rfOrder=15, authorNames=MOHAMMADI M, JAMSHIDI S, REZVANIAN A, journalName=Measurement: Sensors, refType=null, unstructuredReference=MOHAMMADI M, JAMSHIDI S, REZVANIAN A, et al. Advanced fusion of MTM-LSTM and MLP models for time series forecasting: an application for forecasting the solar radiation[J]. Measurement: Sensors, 2024,6(1): 9-14., articleTitle=Advanced fusion of MTM-LSTM and MLP models for time series forecasting: an application for forecasting the solar radiation, refAbstract=null), Reference(id=1197272501119136071, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=207, pageStart=108930, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=16, authorNames=ZHENNAN L, JIYU G, XINLIANG C, journalName=Signal Processing, refType=null, unstructuredReference=ZHENNAN L, JIYU G, XINLIANG C, et al. A method for threshold setting and false alarm probability evaluation for radar detectors[J]. Signal Processing, 2023(207): 108930., articleTitle=A method for threshold setting and false alarm probability evaluation for radar detectors, refAbstract=null), Reference(id=1197272501291102537, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2023, volume=39, issue=5, pageStart=159, pageEnd=162, url=null, language=null, rfNumber=[15], rfOrder=17, authorNames=郭腾, journalName=微型电脑应用, refType=null, unstructuredReference=郭腾. 基于改进LSTM的认知无线电频谱感知模型及仿真[J]. 微型电脑应用, 2023,39(5): 159-162., articleTitle=基于改进LSTM的认知无线电频谱感知模型及仿真, refAbstract=null), Reference(id=1197272501521789259, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2023, volume=39, issue=5, pageStart=159, pageEnd=162, url=null, language=null, rfNumber=[15], rfOrder=18, authorNames=GUO Teng, journalName=Journal of Microcomputer Applications, refType=null, unstructuredReference=GUO Teng. Cognitive radio waves perception model and simulation based on improved LSTM[J]. Journal of Microcomputer Applications, 2023,39(5): 159-162., articleTitle=Cognitive radio waves perception model and simulation based on improved LSTM, refAbstract=null), Reference(id=1197272501773447501, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, doi=null, pmid=null, pmcid=null, year=2018, volume=356, issue=3, pageStart=1640, pageEnd=1657, url=null, language=null, rfNumber=[16], rfOrder=19, authorNames=HASSAN S E, journalName=Journal of the Franklin Institute, refType=null, unstructuredReference=HASSAN S E. Adaptive threshold to guarantee both detection and false alarm probabilities in multi-taper based spectrum sensing[J]. Journal of the Franklin Institute, 2018,356(3): 1640-1657., articleTitle=Adaptive threshold to guarantee both detection and false alarm probabilities in multi-taper based spectrum sensing, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1197272483427561684, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, xref=null, ext=[AuthorCompanyExt(id=1197272483435950293, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Space Long March Vehicle,Beijing,100076), AuthorCompanyExt(id=1197272483452727510, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, companyId=1197272483427561684, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京航天长征飞行器研究所,北京,100076)])], figs=[ArticleFig(id=1197272489530274044, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 1, caption=The overall technical route, figureFileSmall=yVH6Z877KRlOgzaNOHstfw==, figureFileBig=JqkaIfpNIQdO5ygLA8Eefg==, tableContent=null), ArticleFig(id=1197272489844846848, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图1, caption=总体技术路线, figureFileSmall=yVH6Z877KRlOgzaNOHstfw==, figureFileBig=JqkaIfpNIQdO5ygLA8Eefg==, tableContent=null), ArticleFig(id=1197272491350601987, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 2, caption=Deepinsight flowchart, figureFileSmall=6HNN0Z3BZqrmTRh9xtyU/w==, figureFileBig=Tage/2FSjyEiodJvQHMkDQ==, tableContent=null), ArticleFig(id=1197272491619037447, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图2, caption=Deepinsight流程, figureFileSmall=6HNN0Z3BZqrmTRh9xtyU/w==, figureFileBig=Tage/2FSjyEiodJvQHMkDQ==, tableContent=null), ArticleFig(id=1197272491874889993, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 3, caption=Work mechanisms of GAN, figureFileSmall=jPETLQQE325YlRg8pMHCfg==, figureFileBig=cbXAkiAywBEmTQb5Urd0uQ==, tableContent=null), ArticleFig(id=1197272492290126091, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图3, caption=生成式对抗网络工作机制, figureFileSmall=jPETLQQE325YlRg8pMHCfg==, figureFileBig=cbXAkiAywBEmTQb5Urd0uQ==, tableContent=null), ArticleFig(id=1197272492529201423, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 4, caption=Feature extraction and perception network, figureFileSmall=GosRDiEEdJTGkZx901sS4g==, figureFileBig=heS2ksydccNPytu6Hvu7Nw==, tableContent=null), ArticleFig(id=1197272492789248273, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图4, caption=特征提取与感知网络, figureFileSmall=GosRDiEEdJTGkZx901sS4g==, figureFileBig=heS2ksydccNPytu6Hvu7Nw==, tableContent=null), ArticleFig(id=1197272493078655252, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 5, caption=Structure of Dense Block, figureFileSmall=tk/7HWLqlvVkSO68HGdLkA==, figureFileBig=NauCZFNCSRFjMl9nu9izcQ==, tableContent=null), ArticleFig(id=1197272493254816022, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图5, caption=密集连接块结构, figureFileSmall=tk/7HWLqlvVkSO68HGdLkA==, figureFileBig=NauCZFNCSRFjMl9nu9izcQ==, tableContent=null), ArticleFig(id=1197272493485502744, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 6, caption=Detection probability of multi-datasets under SNRs, figureFileSmall=jad7GpLoL0NKcoMKL2JNgQ==, figureFileBig=/q5qzRRH8zkCvau0Ok4WdQ==, tableContent=null), ArticleFig(id=1197272493695217946, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图6, caption=不同训练数据集在不同信噪比下的检测概率, figureFileSmall=jad7GpLoL0NKcoMKL2JNgQ==, figureFileBig=/q5qzRRH8zkCvau0Ok4WdQ==, tableContent=null), ArticleFig(id=1197272493816852766, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 7, caption=Detection probability of MLP-Mixer-DenseNet and MLP-Mixer and other existing methods under SNRs, figureFileSmall=Zxpk6BVK7L4vLlJsjGFO3g==, figureFileBig=6kD82679AZCefA0RdlbZug==, tableContent=null), ArticleFig(id=1197272494211117344, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图7, caption=融合稠密连接网络与MLP-Mixer的频谱感知方法与现有方法在不同信噪比下的检测概率, figureFileSmall=Zxpk6BVK7L4vLlJsjGFO3g==, figureFileBig=6kD82679AZCefA0RdlbZug==, tableContent=null), ArticleFig(id=1197272494404055330, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Fig. 8, caption=Detection probability of MLP-Mixer-DenseNet and MLP-Mixer and other existing methods under different false alarm probality, figureFileSmall=GBBnhlWgnbsF5egDBiN0xg==, figureFileBig=MubAA2b4FsIXMoXGYC0hMQ==, tableContent=null), ArticleFig(id=1197272495628792099, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=图8, caption=融合稠密连接网络与MLP-Mixer的频谱感知方法与现有方法的感知曲线, figureFileSmall=GBBnhlWgnbsF5egDBiN0xg==, figureFileBig=MubAA2b4FsIXMoXGYC0hMQ==, tableContent=null), ArticleFig(id=1197272495846895909, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=EN, label=Tab. 1, caption=Model training parameters, figureFileSmall=null, figureFileBig=null, tableContent=
训练参数 参数设置
训练周期数/次 90
优化器 Adam
批尺寸/个 32
学习率 ${10}^{-4}$
网络层数 50
损失函数 Cross Entropy Loss
), ArticleFig(id=1197272496115331367, tenantId=1146029695717560320, journalId=1146119989267898375, articleId=1148011771119526817, language=CN, label=表1, caption=模型训练参数, figureFileSmall=null, figureFileBig=null, tableContent=
训练参数 参数设置
训练周期数/次 90
优化器 Adam
批尺寸/个 32
学习率 ${10}^{-4}$
网络层数 50
损失函数 Cross Entropy Loss
)], attaches=null, journal=Journal(id=1146118917132496903, delFlag=0, nameCn=导弹与航天运载技术(中英文), nameEn=Missiles and Space Vehicles, nameHistory1=null, nameHistory2=null, issn=2097-1974, eissn=, cn=10-1807/V, coden=null, periodic=1, language=CN, oaType=否, 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=DdfacUcWE1ibGHaCsbhL8w==, journalPrice=null, startedYear=null, abbrevIsoEn=Miss Space Veh, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1753780023753, createdBy=null, updatedBy=13701087609, firstLetterCn=M, firstLetterEn=M, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=DdfacUcWE1ibGHaCsbhL8w==, picEn=jBog3LY3UHfAAYfYEqIBHg==, jcr=null, cjcr=null, exts=[JournalExt(id=1157000535829205111, language=CN, name=导弹与航天运载技术(中英文), nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1753780023772, updatedTime=1753780023772, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=null, submissionAuthorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-ddht-author&redirect_uri=https%3A%2F%2Fddht.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=fefa73ea-30bb-4b68, submissionEditorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-ddht-author&redirect_uri=https%3A%2F%2Fddht.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=fefa73ea-30bb-4b68, submissionReviewUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-ddht-author&redirect_uri=https%3A%2F%2Fddht.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=fefa73ea-30bb-4b68, submissionCeEditorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-ddht-author&redirect_uri=https%3A%2F%2Fddht.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=fefa73ea-30bb-4b68, submissionAeEditorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-ddht-author&redirect_uri=https%3A%2F%2Fddht.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=fefa73ea-30bb-4b68, option={"copyright":""}), JournalExt(id=1157000535871148152, language=EN, name=Missiles and Space Vehicles, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1753780023782, updatedTime=1753780023782, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=null, submissionAuthorUrl=https://journal.ids.fzyun.cn/auth/realms/journal/protocol/openid-connect/auth?client_id=journal-ddht-author&redirect_uri=https%3A%2F%2Fddht.portal.founderss.cn%2Foauth%2Fcallback&response_type=code&scope=phone+openid+email+profile&state=fefa73ea-30bb-4b68, submissionEditorUrl=, submissionReviewUrl=, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146119989267898375, websiteList=[Website(id=1148243202378817956, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146119989267898375, 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/ddyht/CN, language=CN, createTime=1751692112774, createBy=18614031015, updateTime=1753519037132, updateBy=18614031015, name=导弹与航天运载技术-中文站点, tplId=1146099689490845704, title=导弹与航天运载技术(中英文), delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148620442723704855, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202378817956, code=articleTextType, value=kx, createTime=1751782053881, updateTime=1751782053881, creator=18614031015, updator=18614031015), WebsiteProps(id=1148620442694344724, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202378817956, code=banner, value=null, createTime=1751782053874, updateTime=1751782053874, creator=18614031015, updator=18614031015), WebsiteProps(id=1148620442677567507, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202378817956, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=gfUyzanfTuxQ2yc+L/MeWA==, createTime=1751782053870, updateTime=1751782053870, creator=18614031015, updator=18614031015), WebsiteProps(id=1148620442711121942, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202378817956, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782053878, updateTime=1751782053878, creator=18614031015, updator=18614031015), WebsiteProps(id=1148620442706927637, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202378817956, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782053877, updateTime=1751782053877, creator=18614031015, updator=18614031015)]), Website(id=1155906039850618895, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146119989267898375, 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/ddyht/EN, language=EN, createTime=1753519075604, createBy=18614031015, updateTime=1753519075604, updateBy=18614031015, name=导弹与航天运载技术-英文站点, tplId=1146101810881728533, title=Missiles and Space Vehicles, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155907112585777347, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155906039850618895, code=articleTextType, value=kx, createTime=1753519331356, updateTime=1753519331356, creator=18614031015, updator=18614031015), WebsiteProps(id=1155907112556417216, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155906039850618895, code=banner, value=null, createTime=1753519331349, updateTime=1753519331349, creator=18614031015, updator=18614031015), WebsiteProps(id=1155907112552222911, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155906039850618895, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=gfUyzanfTuxQ2yc+L/MeWA==, createTime=1753519331348, updateTime=1753519331348, creator=18614031015, updator=18614031015), WebsiteProps(id=1155907112577388738, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155906039850618895, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753519331354, updateTime=1753519331354, creator=18614031015, updator=18614031015), WebsiteProps(id=1155907112564805825, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155906039850618895, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753519331351, updateTime=1753519331351, creator=18614031015, updator=18614031015)])], journalTitle=导弹与航天运载技术(中英文), weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Missiles and Space Vehicles, journalPhotoCn=DdfacUcWE1ibGHaCsbhL8w==, journalPhotoEn=jBog3LY3UHfAAYfYEqIBHg==, journalFirstLetter=M, 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=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/ddyht/CN/10.7654/j.issn.2097-1974.20240513, detailUrlEn=https://castjournals.cast.org.cn/joweb/ddyht/EN/10.7654/j.issn.2097-1974.20240513, pdfUrlCn=https://castjournals.cast.org.cn/joweb/ddyht/CN/PDF/10.7654/j.issn.2097-1974.20240513, pdfUrlEn=https://castjournals.cast.org.cn/joweb/ddyht/EN/PDF/10.7654/j.issn.2097-1974.20240513, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
融合稠密连接网络与MLP-Mixer的频谱感知方法
收藏切换
PDF下载
田左 , 蔡静 , 霍熠阳
导弹与航天运载技术 | 仿真与试验研究专栏 2024,47(5): 92-98
收起
收藏切换
导弹与航天运载技术 | 仿真与试验研究专栏 2024, 47(5): 92-98
融合稠密连接网络与MLP-Mixer的频谱感知方法
全屏
田左, 蔡静, 霍熠阳
作者信息
  • 北京航天长征飞行器研究所,北京,100076
  • 田左(1995—),男,工程师,主要研究方向为遥感图像处理。

    蔡静(1993-),女,博士,工程师,主要研究方向为多源信息融合技术。

    霍熠阳(1998—),男,助理工程师,主要研究方向为目标检测与识别。

A Spectrum Sensing Method Integrating DenseNet and MLP-Mixer
Zuo TIAN, Jing CAI, Yiyang HUO
Affiliations
  • Beijing Institute of Space Long March Vehicle,Beijing,100076
出版时间: 2024-10-25 doi: 10.7654/j.issn.2097-1974.20240513
文章导航
收藏切换

随着无线应用的持续激增,干扰环境下的通信变得愈发重要,频谱感知技术在解决无线电用频冲突方面发挥了重要作用。然而实际应用环境复杂,获取到的频谱信号不易被高效提取特征,这降低了频谱信号的实用性。如今人工智能在通信领域应用广泛,对通信技术产生重要影响。为此,从深度学习方法入手,提出一种融合稠密连接网络与MLP-Mixer的频谱感知方法。该模型首先通过Deepinsight网络对频谱信号数据实施处理与转换,使其变换为特征图像,再使用生成式对抗网络合成新的特征图,并在得到特征图像后,采用融合稠密连接网络的混合感知器提取特征,从而感知主用户信道占用情况。经过消融试验对比,所提方法相较于已有模型,较好地提升了频谱感知的检测概率。

频谱感知  /  深度学习  /  信号转换  /  生成对抗  /  特征提取

Along with the surge of radio application, electronic communication in interference environments has become increasingly important. The spectrum sensing technique matters in surmounting the frequency conflict of radio. However, the complex environment hinders the efficient feature extraction from the received spectrum signal and reduces the signal practicality. Recently, the artificial intelligence has been widespread in communication field and crucially influenced the electronic countermeasures. Consequently, based on the deep learning, this work proposes a spectrum sensing method to mix DenseNet and MLP-Mixer. Firstly, the model processes and transforms the spectrum signal data to feature images by Deepinsight Net and the generative adversarial networks renew an image. After obtaining the feature image, aspectrum sensing method integrating DenseNet and MLP-Mixer is used in order to sense the channel occupancy of primary user. Compared with the existing model through ablation experiments, the proposed method improves the detection probability of spectrum sensing better.

spectrum sensing  /  deep learning  /  signal transformation  /  generative adversarial  /  feature extraction
田左, 蔡静, 霍熠阳. 融合稠密连接网络与MLP-Mixer的频谱感知方法. 导弹与航天运载技术, 2024 , 47 (5) : 92 -98 . DOI: 10.7654/j.issn.2097-1974.20240513
Zuo TIAN, Jing CAI, Yiyang HUO. A Spectrum Sensing Method Integrating DenseNet and MLP-Mixer[J]. Missiles and Space Vehicles, 2024 , 47 (5) : 92 -98 . DOI: 10.7654/j.issn.2097-1974.20240513
近年来, 随着电子通信在各个领域的大量应用, 认知无线电(Cognitive Radio, CR)进入了一个前所未有的技术发展新阶段, 为占领通信领域高地给予了充分的支撑。其中, 频谱感知是在频谱资源日益紧张且频谱分配方式相对固定的背景下产生的一种技术。 它主要通过识别主用户(Primary User, PU)当前未使用的授权频段, 使得认知用户(Secondary User, SU)能够利用这些空闲频段, 从而大幅增强频谱利用率。实现频谱感知的传统方法有两种, 分别是单用户频谱感知与协作频谱感知。然而无论是哪种传统方法, 在面对复杂多变的通信环境时, 都无法在短时间内实现频谱感知,不能满足真实应用场景。
机器学习是人工智能的一个重要领域, 它通过汇集并分析无线电网络中的环境数据与用户状态, 构建出精确的模型并进行推理学习。这一过程使得无线通信网络的频谱感知具备了智能化的特性, 能够更好地适应并响应实际通信环境的变化。$\mathrm{{Xu}}$[1]设计了一种集成核心演算法, 该算法通过训练新的帧能量检验统计量来推测其判定成效。然而, 目前广泛应用的机器学习频谱感知方法普遍面临着训练耗时较长的挑战, 这使得频谱感知难以迅速实现。 深度学习作为机器学习的延伸与拓展, 凭借其模型和算法的日益精进以及计算能力的不断提升, 在多个领域中都得到了广泛应用。为了进一步提升无线通信网络频谱感知的性能, 研究者们将深度学习的独特优点引入其中, 并因此取得了一些显著的研究成果。
Pan等[2]提出了一种创新的频谱感知方法,该方法结合了深度学习和循环谱分析, 专门用于正交频分复用(Orthogonal Frequency Division Multiplexing, OFDM)信号的检测, 巧妙利用时域平滑快速傅里叶变换累加算法计算出其循环谱。为了更有效地处理这些频谱数据, 他们将循环谱转化为灰度图像, 从而将频谱感知问题转化为图像处理问题。在这一基础上, 进一步采用经过改进的LeNet-5的卷积神经网络, 逐层提取出信号中的深层特征。试验结果显示, 这种方法相较于传统方法展现出了更为优越的感知性能。Zhu等[3]将频谱感知视作一项分类任务, 并据此提出了一种基于分类的频谱感知策略。 在这个方法中, 他们选择将归一化后的信号功率谱作为卷积神经网络(Convolutional Neural Networks, CNN)的输入数据。为训练网络, 他们利用了多种调制信号与噪声样本。最终结果表明,该方法在性能上超越了传统方法, 具备检测未经训练信号的能力,展现出一定的泛化性能。Wang等[4]采用深度卷积神经网络技术, 成功实现了频谱感知的新突破。 在他们的方案中, 使用卷积神经网络进行离线测试设计, 从而研发了一种效率较高的在线检测机制, 该机制基于阈值设定。经过试验验证, 无论主用户信号是独立存在还是相互关联, 这种方法的检测性能均显著优于相关检测器的最优性能界限。Otter等[5]深入探讨了集成深度学习在协作频谱感知中的应用问题, 针对正交频分复用信号的无线网络系统, 设计了一种基于集成学习框架的协作频谱感知方案。 在这个方案中, 他们创新地将循环谱相关特征作为卷积神经网络的输入数据, 借助自助采样策略构建大量训练数据集, 从而提升对认知用户输出的学习能力。与其他已有的方法相比, 这种方案在检测概率和虚警概率方面展现出了显著的优势[6],然而在准确率方面仍有较大的提升空间, 需要进一步地研究和优化。
近年来, 深度学习在频谱感知技术方面取得了显著进展, 但同时也面临诸多难点。现有的频谱感知方法与技术[7],在许多复杂通信环境下,感知能力仍待提高,无法准确地判断出主用户占用信道情况[8]。因此, 探索其他类型的模型以优化频谱感知性能, 是当前亟待深入研究的问题。
本文研究内容主要涵盖两大部分, 总体技术路线如图1所示。
首先利用信号发生器生成频谱信号, 通过引入Deepinsight网络对接受到的信号进行预处理, 将其转换为特征图像, 并采用生成式对抗网络创造出大量特征图像, 为后期网络训练提供样本素材; 其次输入单张特征图像, 开展频谱感知方法研究, 采用融合稠密连接的混合多层感知器架构对特征图进行特征提取, 并对结果分类, 得到频谱感知信道占用情况的结果。
为了更贴合通信环境, 本文拟采用信号发生器生成大量频谱信号, 如何高效地获取频谱信号特征并充分利用其中的信息,成为亟待解决的问题[9]。本文采用Deepinsight网络架构, 避免了传统方法中手动设计特征的不便[10],同时该网络具有强大的非线性映射能力,可以更好地处理频谱信号的复杂性和非线性。 Deepinsight流程如图2所示。
Deepinsight的机制是将非图像数据转换成图像或者二维特征图, 其核心思想是将高度相关的特征放置在紧密的区域, 转换后的图像可以输入到神经网络中进行分类。首先, 对频谱信号进行预处理, 在信号接收器收到频谱信号之后, 对其开展傅里叶变换, 使频域信号取代之前的时域信号, 其可保存为数组。数组中的每个元素对应频域中的一个频率分量, 再使用归一化的方法,将全部数据压缩到0~1之间。
经过上述预处理之后,得到特征向量${\mathbf{f}}_{1}\text{、}{\mathbf{f}}_{2}\text{、}{\mathbf{f}}_{3}\cdots$${\mathbf{f}}_{d}$。采用$t$分布随机近邻嵌入(t-SNE)[11]的降维方法, 产生可视化二维平面上特征点的位置。对于得到的特征点位置, 应用凸包算法寻找到包含所有特征点的最小矩形。
由于得到的矩形不具有水平和垂直的边, 故将其进行旋转。因为图像大小具有像素限制, 这需要再通过平均一些特征, 实现从笛卡尔坐标系到像素帧的转换。最后一步是将特征值与像素坐标相关联。 如果多个要素在像素帧中获得相同的位置, 则在要素映射过程中, 相应的要素将被平均并放置在相同的位置。通过上述流程即可实现频谱信号到特征图的转换。
针对可用于训练的特征图像不足的瓶颈, 本文引入生成对抗的思想, 实现了特征图像相关数据集的扩增。采用生成对抗式网络(Generative Adversarial Networks, GAN)生成特征图像的流程可以概括为判别器与生成器之间的博弈[12]。通过不断的对抗与博弈, GAN逐渐优化生成的数据使其越来越接近真实特征图像。具体工作机制如图3所示。
其目标损失函数公式如下:
$\mathop{\min }\limits_{G}\mathop{\max }\limits_{D}V\left({D, G}\right)= \mathop{\min }\limits_{G}\mathop{\max }\limits_{D}{\mathbb{E}}_{x \sim {p}_{\text{data }}\left( x\right)}\left\lbrack {\log D\left( x\right)}\right\rbrack +\\{\mathbb{E}}_{z \sim {p}_{z}\left( z\right)}\left\lbrack {\log \left({1 - D\left({G\left( z\right)}\right)}\right)}\right\rbrack $
式中$x$为真实图片;$z$为输入生成网络的噪声;$\mathbb{E}$为期望;$G\left( z\right)$为生成网络生成的图片;${p}_{z}$为生成随机噪声的分布;$D\left( x\right)$为鉴别网络判读图片是否是真实的概率。通过生成器与鉴别器的博弈, 生成器的生成技术逐渐提高, 鉴别器的鉴别技术也一次又一次得到完善。直到鉴别器无法确定数据是真实的还是生成器生成时, 网络便可以生成具备相似特征的特征图。由上述步骤训练得到的GAN可将特征图像在数量较为稀缺的前提下,实现对于频谱数据信号的扩增,为后续特征图像训练与特征提取奠定基础。
现有方法在提取频谱信号转换成的特征图特征时, 往往需要人工操作复杂的专业硬件设备, 这对于非专业人员而言, 理解并选取特征图的特征成为了一项挑战。此外, 在提取特征图的特征时, 需要选择那些对于目标有用的特征, 错误的特征选择可能会导致不准确的结论。针对上述难点, 本文以混合多层感知器(Multi Layer Perceptrons-Mixer, MLP-Mixer)为主体, 显著减少了卷积操作的数量, 降低了计算复杂度[13]。又通过融合DenseNet架构,将多层网络的输出直接连接起来, 使其具有更强的表示能力, 增加了特征提取准确率。具体的网络架构示意如图4所示。
图4可知, MLP-Mixer结合稠密连接网络的特性, 形成了一种全新的体系结构, 完全依赖于多层感知器, 这些感知器可以在空间方位或特征通道上多次使用, 其通过简单的矩阵乘法完成数据处理和特征提取, 从而实现了高效且灵活的信息处理目标, 广泛应用于图像特征提取任务中。而在全局平均池化层之前添加的稠密连接网络DenseNet, 则有助于防止模型过拟合, 凭借其密集连接模块增强特征的传播和利用, 提高了网络的表达能力, 在特征提取中展现出一定的优势。
MLP-Mixer接受一系列数量为$S$的不重叠图像块作为输入,每个图像块被转换到特定的隐藏维度$C$, 形成一个二维的实值输入表$X \in {\mathbb{R}}^{S \times C}$,图像块数为原始图像分辨率与每个图像块的分辨率之比。所有图像块都通过一个共享的矩阵进行线性投影。每个多层感知器由两个全连接层组成, 并且其输入数据张量的每一行都会独立地应用一个非线性激活函数。这种混合器层包括通道混合与令牌混合, 其实现方式可以表述为
${\mathbf{U}}_{\propto , i}= {\mathbf{X}}_{\propto , i}+ {\mathbf{W}}_{2}^{\prime }\lambda \left({{\mathbf{W}}_{1}^{\prime }\operatorname{LayerNorm}{\left(\mathbf{X}\right)}_{\propto , i}}\right)\text{,其中}i =\\ 1,\cdots , C $
${\mathbf{Y}}_{j,\alpha }= {\mathbf{U}}_{j,\alpha }+ {\mathbf{W}}_{4}^{\prime }\lambda \left({{\mathbf{W}}_{3}^{\prime }\operatorname{LayerNorm}{\left(\mathbf{U}\right)}_{j,\alpha }}\right)\text{,其中}j =\\ 1,\cdots , S $
式中$\mathbf{X}$为输入的特征图;$\lambda$为高斯误差线性单元GELU;${\mathbf{X}}_{\alpha , i},{\mathbf{U}}_{j,\alpha }$为偏置向量;${\mathbf{W}}_{1}^{\prime },{\mathbf{W}}_{2}^{\prime },{\mathbf{W}}_{3}^{\prime },{\mathbf{W}}_{4}^{\prime }$分别为第1层到第4层的权重矩阵; LayerNorm为对该层网络进行标准化。GELU具有平滑性, 使得神经网络能够学习更复杂的映射, 有助于提高训练过程的收敛速度和性能。基于此, 混合多层感知器能够实现特征图像关键信息的粗提取。
尽管MLP-Mixer在特征图特征提取方面足以展现出高效性和平移不变性, 但其摒弃了自注意力机制与卷积结构, 导致缺乏局部感知能力。而本文通过引入稠密连接网络DenseNet, 更好地重用特征, 进而提升单一MLP-Mixer的性能。
每张特征图包含了尺寸大小不同的特征,其中浅层次特征难以表示图像全局信息,但使用DenseNet中的密集连接块Dense Block结构, 可以加强对图像小尺寸特征的挖掘, Dense Block结构如图5所示。通过前馈传递方式, Dense Block把每一层相互连接起来, 使得梯度更好地流动。通过上述有监督的学习方法, 频谱感知问题最终转化成了二分类问题, 所提模型能够出色地感知到主用户信道是否被占用。
综上所述, DenseNet提取到的局部特征可以为MLP-Mixer提供丰富的信息,而MLP-Mixer的全局特征融合能力则可以将这些局部特征整合起来, 进而实现对于频谱信息的全面理解。而且DenseNet的高效性与MLP-Mixer的简洁性相结合, 在保持性能的同时也能够降低计算成本。
为了充分验证所提融合稠密连接网络与混合多层感知器架构的有效性与可行性, 本节将结合信噪比(Signal to Noise Ratio, SNR)、检测概率(Detection Probability, Pd)、虚警概率(False Alarm Probability, Pfa), 对所提模型与混合多层感知器MLP-Mixer、稠密连接网络DenseNet、卷积神经网络CNN做对比试验, 开展多项试验分析与评估[14]。
本试验使用信号发生器生成正交频分复用信号[15],选取的信号载波频率为${10}\mathrm{{MHz}}$,同时设定子载波个数为64,采样频率为${50}\mathrm{{MHz}}$,码片频率为$1\mathrm{{MHz}}$,采样点个数为1000 。为了确保试验的全面性和准确性,特别选取了信噪比范围在$-{20}\sim 0\mathrm{\;{dB}}$之间的信号作为自变量, 将频谱信号转换为特征图即可实施训练。通过多轮试验验证, 网络训练参数的设置如表1所示, 训练集在一定周期后就能达到100%,测试集也有较高的准确率,损失函数可以收敛。
为了验证训练样本对所提模型检测概率的影响, 选取不同信噪比下的不同训练数据量。采用预训练过的模型实施训练,训练集$H$大小分别为500、1000、 2000、4000,信噪比大小为-20~0 dB,每间隔1 dB做一次记录。测试数据则由100组信号数据和100组干扰噪声数据构成。如图6所示, 当训练样本较小(尤其是不足1000)时,本文所提模型在不同信噪比下的变化幅度较大。考虑到训练样本数量不足, 导致其无法充分覆盖到数据的各种变化和复杂性, 进而导致模型在预测新样本时表现不够稳定, 从而使得模型的检测概率不高。随着训练集数量的增加, 所提模型的检测概率逐渐增加,但是当训练样本增加到4 000时, 其检测概率反而会下降, 根本在于所提模型过拟合, 对于测试数据无法表现出良好的感知效果,而当$H ={2000}$时,稠密连接网络与MLP-Mixer的结合能够同时考虑局部和全局的信息,展现出最佳检测效果。
为了更好地验证所提模型检测概率的准确性, 将其与MLP-Mixer、DenseNet以及CNN这3种频谱感知方法在不同信噪比下进行性能检测对比, 见图7。能够明显看出,在信噪比低于$-6\mathrm{\;{dB}}$的情况下,本文所提模型的检测概率始终优于其他3种模型, 展示出一定的准确性, 其具有重用与传播特征的性能, 可以提升模型局部感知的能力与抗干扰能力, 进而提升模型对频谱的检测能力。一旦缺少DenseNet或者MLP-Mixer架构, 所提模型梯度消失问题会加重, 也会出现过拟合现象, 模型无法从多个维度提取特征, 检测的准确率就会受到影响。MLP-Mixer检测概率比DenseNet高的原因在于MLP-Mixer使用的全连接层具有较低的归纳偏置, 这意味着模型在训练过程中更加灵活, 可以学习更加广泛的数据分布。相比之下, DenseNet的卷积操作具有更高的归纳偏置, 这可能限制了其处理不同数据分布的能力, 导致检测概率偏低。但随着信噪比的增加, 信号质量不断提高, 利于不同的模型更准确地识别和提取特征, 所以不同方法的检测概率较为接近。
类似地, 虚警概率也能够反映出频谱感知性能的优劣[16]。统计并记录信噪比为$-{12}\mathrm{\;{dB}}$时,多个试验的虚警概率及其对应的检测概率, 得到了4个模型的感知曲线图, 如图8所示。
图8可知, 本文所提MLP-Mixer-DenseNet具备更佳的检测与感知性能, 其在任何虚警概率下的检测概率都高于另外3种模型。尽管该模型未引入注意力机制, 但融合稠密连接网络能够将不同层的特征串联起来, 使其重复使用共同学习到的特征, 同样能够保证模型具有良好的感知主用户信道占用情况的能力。而无论是CNN网络或是去掉部分架构的模型, 都存在无法充分捕捉全局关系、抗干扰性差的缺点, 这使得它们不能在不同的信噪比下都具备较高的感知精度。
针对复杂通信环境下,不易提取频谱信号特征、 可用于训练的数据样本少、无法高精度感知信道占用情况的问题, 将Deepinsight网络应用于频谱信号向图像转换的工作中, 将信号感知转换成图像识别, 并采用生成式对抗网络GAN产生大量特征图像。通过改进MLP-Mixer网络, 将稠密连接网络DenseNet添加到全局平均池化层之前, 致力于提取信号转换后得到的频谱特征图像的特征, 从而感知主用户信道是否占用。对比试验表明, 在信噪比较低的情况下,所提融合稠密连接网络与混合多层感知器的频谱感知方法比现有的深度学习模型具备更准的检测概率。该方法在相同的虚警概率下,检测概率明显优于其余模型。
参考文献 引证文献
排序方式:
[1]
XU Y, CHENG P, CHEN Z, et al. Learning and transferring mid-level image representations using convolutional neural networks[J]. IEEE Transactions on Signal Processing, 2018,66(21): 5634-5647.
[2]
PAN G, LI J, LIN F. A cognitive radio spectrum sensing method for an OFDMsignal based on deep learning and cycle spectrum[J]. International Journal of Digital Multimedia Broadcasting, 2020(10): 1-10.
[3]
ZHU Y, LIU J, FENG Z, et al. Sensing performance of efficient cyclostationary detector with multiple antennas in multipath fading and lognormal shadowing environments[J]. Journal of Communications and Networks, 2019,16(2): 162-171.
[4]
WANG H, WANG J. Short term wind speed forecasting based on feature extraction by CNN[C]// 2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC). Nanjing: 2021.
[5]
OTTER D W, MEDINA J R, KALITA J K. A survey of the usages of deep learning for natural language processing[J]. IEEE transactions on neural networks and learning systems, 2020,32(2): 604-624.
[6]
DHANANJAYA S, YUVARAJU B N. A novel method in matched filter spectrum sensing to minimize interference from compromised secondary users of cognitive radio networks[C]// 2018 International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT). Msyuru: 2018.
[7]
NALLAGONDA S, KUMAR Y R, SHILPA P. Analysis of hard-decision and soft-data fusion schemes for cooperative spectrum sensing in rayleigh fading channel[C]// 2017 IEEE 7th International Advance Computing Conference (IACC). Hyderabad: 2017.
[8]
杜红. 认知无线电中频谱感知优化与无线资源管理的研究[D]. 北京: 北京邮电大学, 2012.
DU Hong. Research on spectrum sensing optimization and radio resource management in cognitive radio[D]. Beijing: Beijing University of Posts and Telecommunications, 2012.
[9]
胡泽锋. 单通道全双工光通信系统及关键器件研究[D]. 南京: 南京邮电大学, 2023.
HU Zefeng. Research on single-channel full-duplex optical communication system and key devices[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2023.
[10]
李柯, 刘文忠, 秦镜淘. 改进TransUNet网络对肝脏肿瘤CT图像的级联分割[J]. 宜宾学院学报, 2024,3(2): 1-14.
LI Ke, LIU Wenzhong, QIN Jingtao. Improved cascade segmentation of liver tumor CT images by TransUNet network[J]. Journal of Yibin University, 2024,3(2): 1-14.
[11]
SHENG H, HUAKUN Q, YINGNAN Z, et al. Automatic grid topology detection method based on Lasso algorithm and t-SNE algorithm[J]. Energy Informatics, 2024,7(1): 23-30.
[12]
ZHAO J, HE D, GE M, et al. GA-GGD: improving semantic discriminability in graph contrastive learning via Generative Adversarial Network[J]. Information Fusion, 2024,7(3): 1-10.
[13]
MOHAMMADI M, JAMSHIDI S, REZVANIAN A, et al. Advanced fusion of MTM-LSTM and MLP models for time series forecasting: an application for forecasting the solar radiation[J]. Measurement: Sensors, 2024,6(1): 9-14.
[14]
ZHENNAN L, JIYU G, XINLIANG C, et al. A method for threshold setting and false alarm probability evaluation for radar detectors[J]. Signal Processing, 2023(207): 108930.
[15]
郭腾. 基于改进LSTM的认知无线电频谱感知模型及仿真[J]. 微型电脑应用, 2023,39(5): 159-162.
GUO Teng. Cognitive radio waves perception model and simulation based on improved LSTM[J]. Journal of Microcomputer Applications, 2023,39(5): 159-162.
[16]
HASSAN S E. Adaptive threshold to guarantee both detection and false alarm probabilities in multi-taper based spectrum sensing[J]. Journal of the Franklin Institute, 2018,356(3): 1640-1657.
2024年第47卷第5期
PDF下载
330
136
引用本文
BibTeX
文章信息
doi: 10.7654/j.issn.2097-1974.20240513
  • 接收时间:2024-04-08
  • 首发时间:2025-07-04
  • 出版时间:2024-10-25
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-04-08
  • 修回日期:2024-06-05
基金
作者信息
    北京航天长征飞行器研究所,北京,100076
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/ddyht/CN/10.7654/j.issn.2097-1974.20240513
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