Article(id=1251226684486271365, tenantId=1146029695717560320, journalId=1251194772300279900, issueId=1251226682309423223, articleNumber=null, orderNo=null, doi=10.20079/j.issn.1001-893x.240530003, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1716998400000, receivedDateStr=2024-05-30, revisedDate=1727193600000, revisedDateStr=2024-09-25, acceptedDate=null, acceptedDateStr=null, onlineDate=1776245288249, onlineDateStr=2026-04-15, pubDate=1764259200000, pubDateStr=2025-11-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1776245288249, onlineIssueDateStr=2026-04-15, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1776245288249, creator=13041195026, updateTime=1776245288249, updator=13041195026, issue=Issue{id=1251226682309423223, tenantId=1146029695717560320, journalId=1251194772300279900, year='2025', volume='65', issue='11', pageStart='1729', pageEnd='1954', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1776245287729, creator=13041195026, updateTime=1776246742124, updator=13041195026, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1251232782568080068, tenantId=1146029695717560320, journalId=1251194772300279900, issueId=1251226682309423223, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1251232782568080069, tenantId=1146029695717560320, journalId=1251194772300279900, issueId=1251226682309423223, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=1766, endPage=1772, ext={EN=ArticleExt(id=1251226686419845516, articleId=1251226684486271365, tenantId=1146029695717560320, journalId=1251194772300279900, language=EN, title=A UWB Non-Line-of-Sight Recognition Method Based on LightGBM, columnId=1251226683223781499, journalTitle=Telecommunication Engineering, columnName=Application Fundamental Research and Advanced Technology, runingTitle=null, highlight=null, articleAbstract=

For the optimal feature subset selection and model parameter optimization in ultra-wideband non-line-of-sight(NLOS) recognition,a new NLOS recognition method based on the cross-validation recursive feature elimination algorithm of Light Gradient Boosting Machine(LightGBM) and Optuna parameter tuning is proposed. First,six important features,including the difference between the first path signal and the total received signal power,and the maximum noise,are selected as the optimal feature subset using the recursive feature elimination and cross-validation algorithm. Then,Optuna is used to optimize the hyperparameters of LightGBM model. Line-of-sight and non-line-of-sight feature data is collected,and the Support Vector Machine,Extreme Gradient Boosting algorithm,and parameter-optimized LightGBM model are trained and tested. The results demonstrate that the selected features exhibit excellent discriminative ability,with the optimized LightGBM model achieving a recognition accuracy of 95.28% .

, 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=Qian LI, Zhuolun LIU, Xiaoyun SUN, Yong CHEN, Shiji SONG, Xinglong ZHANG), CN=ArticleExt(id=1251226691604005370, articleId=1251226684486271365, tenantId=1146029695717560320, journalId=1251194772300279900, language=CN, title=一种基于LightGBM的UWB非视距识别方法, columnId=1251226683383165054, journalTitle=电讯技术, columnName=应用基础与前沿技术, runingTitle=null, highlight=null, articleAbstract=

针对超宽带非视距(Non-Line-of-Sight,NLOS)识别中最优特征子集选取与模型参数优化问题,提出了一种基于轻量级梯度提升机(Light Gradient Boosting Machine,LightGBM)的交叉验证递归特征消除算法与Optuna参数调优相结合的NLOS识别方法。首先通过递归特征消除加交叉验证算法分析选取首径信号与总信号接收功率差值、噪声最大值等6个重要特征作为最优特征子集,之后使用Optuna调参框架优化LightGBM模型超参数。采集视距与非视距特征数据,使用支持向量机、极限梯度提升算法和参数优化后的LightGBM等模型进行训练与测试,结果表明,所选取特征具有良好区分性,参数优化后的LightGBM模型识别准确率达95.28%。

, correspAuthors=null, authorNote=null, correspAuthorsNote=
孙晓云 Email:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=8DZfRe7MVbc3gFc4dz4DGg==, magXml=Moj73d6sEXXszjzhbRjXLA==, pdfUrl=null, pdf=DWm54OkJ7HoGgzvrspynXA==, pdfFileSize=2642376, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=D8bDJ4I3UJIx1FV+DIyxxg==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=XcYP1z/lRTACBbFTay9/Lw==, mapNumber=null, authorCompany=null, fund=null, authors=

李乾 男,1984年生于吉林磐石,2006年获学士学位,现为高级工程师,主要从事电力电缆研究与应用、电缆隧道智能监测技术方面的工作。

刘卓伦 男,1999年生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

孙晓云 女,1971年生于河北晋州,2000年获博士学位,现为教授,主要研究方向为电缆隧道环境智能监测技术、大型电器设备无损检测、电机电器智能控制。

陈勇 男,1978年生于河北,博士,副研究员,主要研究方向为超宽带室内精确定位技术、超宽带非视距识别与缓解等。

宋士济 男,1999年出生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为目标检测技术。

张醒龙 男,2001年出生于天津,2023年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

, authorsList=李乾, 刘卓伦, 孙晓云, 陈勇, 宋士济, 张醒龙)}, authors=[Author(id=1251226692098933269, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, 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=1251226692178625050, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692098933269, language=EN, stringName=Qian LI, firstName=Qian, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1Shijiazhuang Power Supply Branch of State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251226692287676959, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692098933269, language=CN, stringName=李乾, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1国网河北省电力有限公司石家庄供电分公司,石家庄 050004, bio={"content":"

李乾 男,1984年生于吉林磐石,2006年获学士学位,现为高级工程师,主要从事电力电缆研究与应用、电缆隧道智能监测技术方面的工作。

"}, bioImg=null, bioContent=

李乾 男,1984年生于吉林磐石,2006年获学士学位,现为高级工程师,主要从事电力电缆研究与应用、电缆隧道智能监测技术方面的工作。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251226691830497796, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=1, ext=[AuthorCompanyExt(id=1251226691847275016, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691830497796, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Shijiazhuang Power Supply Branch of State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050004, China), AuthorCompanyExt(id=1251226691859857927, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691830497796, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国网河北省电力有限公司石家庄供电分公司,石家庄 050004)])]), Author(id=1251226692384145958, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, 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=1251226692468032045, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692384145958, language=EN, stringName=Zhuolun LIU, firstName=Zhuolun, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251226692560306739, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692384145958, language=CN, stringName=刘卓伦, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2石家庄铁道大学 电气与电子工程学院,石家庄 050043, bio={"content":"

刘卓伦 男,1999年生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

"}, bioImg=null, bioContent=

刘卓伦 男,1999年生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251226691952132620, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=2, ext=[AuthorCompanyExt(id=1251226691960521229, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China), AuthorCompanyExt(id=1251226691981492751, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2石家庄铁道大学 电气与电子工程学院,石家庄 050043)])]), Author(id=1251226692639998522, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=sunxyheb@stdu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1251226692749050433, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692639998522, language=EN, stringName=Xiaoyun SUN, firstName=Xiaoyun, middleName=null, lastName=SUN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251226692845519431, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692639998522, language=CN, stringName=孙晓云, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2石家庄铁道大学 电气与电子工程学院,石家庄 050043, bio={"content":"

孙晓云 女,1971年生于河北晋州,2000年获博士学位,现为教授,主要研究方向为电缆隧道环境智能监测技术、大型电器设备无损检测、电机电器智能控制。

"}, bioImg=null, bioContent=

孙晓云 女,1971年生于河北晋州,2000年获博士学位,现为教授,主要研究方向为电缆隧道环境智能监测技术、大型电器设备无损检测、电机电器智能控制。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251226691952132620, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=2, ext=[AuthorCompanyExt(id=1251226691960521229, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China), AuthorCompanyExt(id=1251226691981492751, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2石家庄铁道大学 电气与电子工程学院,石家庄 050043)])]), Author(id=1251226692958765650, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, 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=1251226693097177688, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692958765650, language=EN, stringName=Yong CHEN, firstName=Yong, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251226693214618207, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226692958765650, language=CN, stringName=陈勇, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2石家庄铁道大学 电气与电子工程学院,石家庄 050043, bio={"content":"

陈勇 男,1978年生于河北,博士,副研究员,主要研究方向为超宽带室内精确定位技术、超宽带非视距识别与缓解等。

"}, bioImg=null, bioContent=

陈勇 男,1978年生于河北,博士,副研究员,主要研究方向为超宽带室内精确定位技术、超宽带非视距识别与缓解等。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251226691952132620, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=2, ext=[AuthorCompanyExt(id=1251226691960521229, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China), AuthorCompanyExt(id=1251226691981492751, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2石家庄铁道大学 电气与电子工程学院,石家庄 050043)])]), Author(id=1251226694783287910, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, 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=1251226694925894253, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226694783287910, language=EN, stringName=Shiji SONG, firstName=Shiji, middleName=null, lastName=SONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251226695026557554, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226694783287910, language=CN, stringName=宋士济, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2石家庄铁道大学 电气与电子工程学院,石家庄 050043, bio={"content":"

宋士济 男,1999年出生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为目标检测技术。

"}, bioImg=null, bioContent=

宋士济 男,1999年出生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为目标检测技术。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251226691952132620, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=2, ext=[AuthorCompanyExt(id=1251226691960521229, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China), AuthorCompanyExt(id=1251226691981492751, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2石家庄铁道大学 电气与电子工程学院,石家庄 050043)])]), Author(id=1251226695139803772, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, orderNo=5, 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=1251226695248855685, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226695139803772, language=EN, stringName=Xinglong ZHANG, firstName=Xinglong, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251226695391462029, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, authorId=1251226695139803772, language=CN, stringName=张醒龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2石家庄铁道大学 电气与电子工程学院,石家庄 050043, bio={"content":"

张醒龙 男,2001年出生于天津,2023年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

"}, bioImg=null, bioContent=

张醒龙 男,2001年出生于天津,2023年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251226691952132620, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=2, ext=[AuthorCompanyExt(id=1251226691960521229, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China), AuthorCompanyExt(id=1251226691981492751, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2石家庄铁道大学 电气与电子工程学院,石家庄 050043)])])], keywords=[Keyword(id=1251226695529874074, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, orderNo=1, keyword=UWB NLOS recognition), Keyword(id=1251226695651508898, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, orderNo=2, keyword=light gradient boosting machine(LightGBM)), Keyword(id=1251226695756366505, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, orderNo=3, keyword=recursive feature elimination with cross validation(RFECV)), Keyword(id=1251226695827669678, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, orderNo=4, keyword=hyperparameter optimization), Keyword(id=1251226695903167155, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, orderNo=1, keyword=超宽带非视距识别), Keyword(id=1251226695978664633, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, orderNo=2, keyword=轻量级梯度提升机(LightGBM)), Keyword(id=1251226696066745024, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, orderNo=3, keyword=交叉验证递归特征消除算法(RFECV)), Keyword(id=1251226696175796937, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, orderNo=4, keyword=超参数优化)], refs=[Reference(id=1251226700118442883, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2019, volume=11, issue=22, pageStart=1, pageEnd=7, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=LIU F, LI X, WANG J, journalName=Remote Sensing, refType=null, unstructuredReference=LIU F, LI X, WANG J, et al. An adaptive UWB/MEMSIMU complementary Kalman filter for indoor location in NLOS environment[J]. Remote Sensing, 2019, 11(22):1-7., articleTitle=An adaptive UWB/MEMSIMU complementary Kalman filter for indoor location in NLOS environment, refAbstract=null), Reference(id=1251226700214911882, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2019, volume=21, issue=3, pageStart=2568, pageEnd=2599, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=ZAFARI F, GKELIAS A, LEUNG K K, journalName=IEEE Communications Surveys & Tutorials, refType=null, unstructuredReference=ZAFARI F, GKELIAS A, LEUNG K K. A survey of indoor localization systems and technologies[J]. IEEE Communications Surveys & Tutorials, 2019, 21(3):2568-2599., articleTitle=A survey of indoor localization systems and technologies, refAbstract=null), Reference(id=1251226700332352402, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2023, volume=63, issue=10, pageStart=1596, pageEnd=1602, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=戢静红, 张振宇, 邓平, journalName=电讯技术, refType=null, unstructuredReference=戢静红, 张振宇, 邓平, .一种基于随机森林的LOS/NLOS基站识别方法[J].电讯技术, 2023, 63(10):1596-1602., articleTitle=一种基于随机森林的LOS/NLOS基站识别方法, refAbstract=null), Reference(id=1251226700420432793, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2022, volume=40, issue=4, pageStart=865, pageEnd=874, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=畅田田, 王威, 高婧洁, journalName=西北工业大学学报, refType=null, unstructuredReference=畅田田, 王威, 高婧洁, .基于信道测量实验的NLOS误差消除方法对比研究[J].西北工业大学学报, 2022, 40(4):865-874., articleTitle=基于信道测量实验的NLOS误差消除方法对比研究, refAbstract=null), Reference(id=1251226700558844836, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=ZENG Z Q, LIU S, WANG L, journalName=null, refType=null, unstructuredReference=ZENG Z Q, LIU S, WANG L. UWB NLOS identification with feature combination selection based on genetic algorithm[C]//2019 IEEE International Conference on Consumer Electronics. Las Vegas:IEEE, 2019:1-5., articleTitle=UWB NLOS identification with feature combination selection based on genetic algorithm, refAbstract=null), Reference(id=1251226700655313835, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2021, volume=70, issue=null, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=ANDRE A G F, FERNANDES D, BRANCO S, journalName=IEEE Transactions on Instrumentation and Measurement, refType=null, unstructuredReference=ANDRE A G F, FERNANDES D, BRANCO S, et al. Feature selection for real-time NLOS identification and mitigation for body-mounted UWB transceivers[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70:1-10., articleTitle=Feature selection for real-time NLOS identification and mitigation for body-mounted UWB transceivers, refAbstract=null), Reference(id=1251226700722422706, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=5, pageStart=83, pageEnd=89, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=袁驰, 黄波, journalName=移动通信, refType=null, unstructuredReference=袁驰, 黄波. UWB定位中基于XGBoost的NLOS识别方法[J].移动通信, 2023, 47(5):83-89., articleTitle=UWB定位中基于XGBoost的NLOS识别方法, refAbstract=null), Reference(id=1251226700814697400, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=SU C, ZHANG F S, JIN B H, journalName=null, refType=null, unstructuredReference=SU C, ZHANG F S, JIN B H, et al. UWBLoc: NLoS mitigation for practical UWB localization[C]//2023 IEEE Smart World Congress. Portsmouth:IEEE, 2023:1-8., articleTitle=UWBLoc: NLoS mitigation for practical UWB localization, refAbstract=null), Reference(id=1251226700940526531, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2024, volume=137, issue=2, pageStart=1141, pageEnd=1160, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=CUI X R, LI Y X, LI J, journalName=Wireless Personal Communications, refType=null, unstructuredReference=CUI X R, LI Y X, LI J, et al. UWB wireless positioning method based on LightGBM[J]. Wireless Personal Communications, 2024, 137(2):1141-1160., articleTitle=UWB wireless positioning method based on LightGBM, refAbstract=null), Reference(id=1251226701045384143, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2021, volume=54, issue=3, pageStart=1937, pageEnd=1967, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=BENTÉJAC C, CSÖRGO A, MARTÍNEZ-MUÑOZ G, journalName=Artificial Intelligence Review, refType=null, unstructuredReference=BENTÉJAC C, CSÖRGO A, MARTÍNEZ-MUÑOZ G. A comparative analysis of gradient boosting algorithms[J].Artificial Intelligence Review, 2021, 54(3):1937-1967., articleTitle=A comparative analysis of gradient boosting algorithms, refAbstract=null), Reference(id=1251226701154436055, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=3149, pageEnd=3157, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=KE G L, MENG Q, FINLEY T, journalName=null, refType=null, unstructuredReference=KE G L, MENG Q, FINLEY T, et al. LightGBM:a highly efficient gradient boosting decision tree[C]//The 31st Conference on Neural Information Processing Systems. New York:ACM, 2017:3149-3157., articleTitle=LightGBM:a highly efficient gradient boosting decision tree, refAbstract=null), Reference(id=1251226701259293662, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2023, volume=15, issue=1, pageStart=789, pageEnd=798, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=YIN Z Q, SHI L, LUO J R, journalName=Sustainability, refType=null, unstructuredReference=YIN Z Q, SHI L, LUO J R, et al. Pump feature construction and electrical energy consumption prediction based on feature engineering and LightGBM algorithm[J]. Sustainability, 2023, 15(1):789-798., articleTitle=Pump feature construction and electrical energy consumption prediction based on feature engineering and LightGBM algorithm, refAbstract=null), Reference(id=1251226701368345575, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=12, pageStart=54, pageEnd=63, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=徐国天, 沈耀童, journalName=信息网络安全, refType=null, unstructuredReference=徐国天, 沈耀童.基于XGBoost和LightGBM双层模型的恶意软件检测方法[J].信息网络安全, 2020, 20(12):54-63., articleTitle=基于XGBoost和LightGBM双层模型的恶意软件检测方法, refAbstract=null), Reference(id=1251226701477397488, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=142, pageEnd=152, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=AKKAYA B, journalName=null, refType=null, unstructuredReference=AKKAYA B. The effect of recursive feature elimination with cross-validation method on classification performance with different sizes of datasets[C]//The 4th International Conference on Data Science and Applications. Turkey:IEEE, 2021:142-152., articleTitle=The effect of recursive feature elimination with cross-validation method on classification performance with different sizes of datasets, refAbstract=null), Reference(id=1251226701573866486, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=2623, pageEnd=2631, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=AKIBA T, SANO S, YANASE T, journalName=null, refType=null, unstructuredReference=AKIBA T, SANO S, YANASE T, et al. Optuna: a next-generation hyperparameter optimization framework[C]//The 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Anchorage:ACM, 2019:2623-2631., articleTitle=Optuna: a next-generation hyperparameter optimization framework, refAbstract=null), Reference(id=1251226701649363965, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, doi=null, pmid=null, pmcid=null, year=2024-05-22, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=15, authorNames=DecaWave, journalName=null, refType=null, unstructuredReference=DecaWave. DW1000 user manual[EB/OL]. (2017-10-26)[2024-05-22]. https://www.decawave.com/product-documentation/., articleTitle=DW1000 user manual, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1251226691830497796, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=1, ext=[AuthorCompanyExt(id=1251226691847275016, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691830497796, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1Shijiazhuang Power Supply Branch of State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050004, China), AuthorCompanyExt(id=1251226691859857927, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691830497796, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国网河北省电力有限公司石家庄供电分公司,石家庄 050004)]), AuthorCompany(id=1251226691952132620, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, xref=2, ext=[AuthorCompanyExt(id=1251226691960521229, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China), AuthorCompanyExt(id=1251226691981492751, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, companyId=1251226691952132620, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2石家庄铁道大学 电气与电子工程学院,石家庄 050043)])], figs=[ArticleFig(id=1251226696322597589, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=i5to9MDwQs121iF/5wuvWg==, figureFileBig=D8bDJ4I3UJIx1FV+DIyxxg==, tableContent=null), ArticleFig(id=1251226696423260892, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图1, caption=直方图优化算法示意, figureFileSmall=i5to9MDwQs121iF/5wuvWg==, figureFileBig=D8bDJ4I3UJIx1FV+DIyxxg==, tableContent=null), ArticleFig(id=1251226696712667889, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=ltP4siIEfWVAnr/5INVjbg==, figureFileBig=+Tjb5bL6uvDyB2Gq/lIbDg==, tableContent=null), ArticleFig(id=1251226696846885629, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图2, caption=带深度限制的按叶子生长策略示意, figureFileSmall=ltP4siIEfWVAnr/5INVjbg==, figureFileBig=+Tjb5bL6uvDyB2Gq/lIbDg==, tableContent=null), ArticleFig(id=1251226696943354629, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=SzwyPwRlKM8TcMrNFJE+Xg==, figureFileBig=ZJIKetXt/uaFtouKNA8ihQ==, tableContent=null), ArticleFig(id=1251226697052406543, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图3, caption=RFECV算法流程, figureFileSmall=SzwyPwRlKM8TcMrNFJE+Xg==, figureFileBig=ZJIKetXt/uaFtouKNA8ihQ==, tableContent=null), ArticleFig(id=1251226697148875543, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=dCM3xt8ymntkIgyALbq02Q==, figureFileBig=F1IbgWevLzvoHmrKiA0bwg==, tableContent=null), ArticleFig(id=1251226697316647710, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图4, caption=NLOS识别流程, figureFileSmall=dCM3xt8ymntkIgyALbq02Q==, figureFileBig=F1IbgWevLzvoHmrKiA0bwg==, tableContent=null), ArticleFig(id=1251226697434088227, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=mbbiL5IW/J2XcqprdmMk8w==, figureFileBig=ZoLctIv3mDlKYMGBladVMg==, tableContent=null), ArticleFig(id=1251226697538945835, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图5, caption=特征重要性分数图, figureFileSmall=mbbiL5IW/J2XcqprdmMk8w==, figureFileBig=ZoLctIv3mDlKYMGBladVMg==, tableContent=null), ArticleFig(id=1251226697673163573, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=2zAiaZfnwJ+bAdD7YNPaYw==, figureFileBig=QfurC8ITCAXXAXGOHSGiyg==, tableContent=null), ArticleFig(id=1251226699229250366, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图6, caption=不同特征数量准确率对比, figureFileSmall=2zAiaZfnwJ+bAdD7YNPaYw==, figureFileBig=QfurC8ITCAXXAXGOHSGiyg==, tableContent=null), ArticleFig(id=1251226699329913670, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=JWuiICgDFcsqNe5DReU1Sw==, figureFileBig=ofco4IGY9m8/bUlGIK+9NA==, tableContent=null), ArticleFig(id=1251226699434771276, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=图7, caption=模型分类指标, figureFileSmall=JWuiICgDFcsqNe5DReU1Sw==, figureFileBig=ofco4IGY9m8/bUlGIK+9NA==, tableContent=null), ArticleFig(id=1251226699535434579, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
特征名称特征描述
FP_IDX首径采样点位置
MAX_NOISE噪声最大值
FPAMP1首径采样点1的振幅值
FPAMP2首径采样点+2处的振幅值
FPAMP3首径采样点+3处的振幅值
STDEV_NOISE噪声标准差
maxGrowthCIRCIR最大增幅
rxPreamCount累计前导码数量
LDE_PPINDXLDE峰值路径位置
LDE_PPAMPLLDE峰值路径振幅
FP_L首径功率
RX_L总接收功率
PD首径与接收信号功率差值
), ArticleFig(id=1251226699619320664, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=表1, caption=

特征信息

, figureFileSmall=null, figureFileBig=null, tableContent=
特征名称特征描述
FP_IDX首径采样点位置
MAX_NOISE噪声最大值
FPAMP1首径采样点1的振幅值
FPAMP2首径采样点+2处的振幅值
FPAMP3首径采样点+3处的振幅值
STDEV_NOISE噪声标准差
maxGrowthCIRCIR最大增幅
rxPreamCount累计前导码数量
LDE_PPINDXLDE峰值路径位置
LDE_PPAMPLLDE峰值路径振幅
FP_L首径功率
RX_L总接收功率
PD首径与接收信号功率差值
), ArticleFig(id=1251226699724178274, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
参数名称具体值
max_depth9
num_leaves79
learning_rate0.074
reg_alpha0.596
reg_lambda0.363
n_estimators831
), ArticleFig(id=1251226699862590321, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=表2, caption=

优化后超参数值

, figureFileSmall=null, figureFileBig=null, tableContent=
参数名称具体值
max_depth9
num_leaves79
learning_rate0.074
reg_alpha0.596
reg_lambda0.363
n_estimators831
), ArticleFig(id=1251226699938087801, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
项目预测值
NLOSLOS
真实值NLOSTPFN
LOSFPTN
), ArticleFig(id=1251226700017779582, tenantId=1146029695717560320, journalId=1251194772300279900, articleId=1251226684486271365, language=CN, label=表3, caption=

混淆矩阵

, figureFileSmall=null, figureFileBig=null, tableContent=
项目预测值
NLOSLOS
真实值NLOSTPFN
LOSFPTN
)], attaches=null, journal=Journal(id=1251193998841266264, delFlag=0, nameCn=电讯技术, nameEn=Telecommunication Engineering, nameHistory1=null, nameHistory2=null, issn=1001-893X, eissn=null, cn=51-1267/TN, coden=null, periodic=0, 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=CpBmHoMzpESavU+iEMTBmw==, journalPrice=null, startedYear=null, abbrevIsoEn=Telecommunication Engineering, journalRemark=null, publicationField=null, createdTime=1776237495387, updatedTime=1776238086301, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=T, firstLetterEn=T, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=CpBmHoMzpESavU+iEMTBmw==, picEn=jCOIy2zOaGJZ/y3z2gPZzg==, jcr=null, cjcr=null, exts=[JournalExt(id=1251196477385687352, language=CN, name=电讯技术, 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=1776238086315, updatedTime=1776238086315, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.teleonline.cn/dxjs/ch/author/login.aspx, submissionEditorUrl=https://www.teleonline.cn/dxjs/ch/login.aspx, submissionReviewUrl=https://www.teleonline.cn/dxjs/ch/auditor/login.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1251196477469573433, language=EN, name=Telecommunication Engineering, 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=1776238086335, updatedTime=1776238086335, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.teleonline.cn/dxjs/ch/author/login.aspx, submissionEditorUrl=https://www.teleonline.cn/dxjs/ch/login.aspx, submissionReviewUrl=https://www.teleonline.cn/dxjs/ch/auditor/login.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1251194772300279900, websiteList=[Website(id=1251197148327522670, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1251194772300279900, 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/dxjs/CN, language=CN, createTime=1776238246280, createBy=18614031015, updateTime=1776238378770, updateBy=18614031015, name=电讯技术-中文, tplId=1146099689490845704, title=电讯技术, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1251197904854135502, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=articleTextType, value=kx, createTime=1776238426650, updateTime=1776238426650, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904833163979, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=banner, value=null, createTime=1776238426645, updateTime=1776238426645, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904870912721, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=grayFlag, value=0, createTime=1776238426654, updateTime=1776238426654, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904824775370, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=logo, value=https://castjournals.cast.org.cn/joweb/dxjs/CN/file/pic?fileId=BBd4SC9puESjyaw04bneig==, createTime=1776238426643, updateTime=1776238426643, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904883495635, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=minRunFlag, value=0, createTime=1776238426657, updateTime=1776238426657, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904845746893, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dxjs/CN/file/pic, createTime=1776238426648, updateTime=1776238426648, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904875107026, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=silenceFlag, value=0, createTime=1776238426655, updateTime=1776238426655, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904841552588, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1776238426647, updateTime=1776238426647, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904858329807, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=themeColor, value=null, createTime=1776238426651, updateTime=1776238426651, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197904866718416, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148327522670, code=themeStyle, value=null, createTime=1776238426653, updateTime=1776238426653, creator=18614031015, updator=18614031015)]), Website(id=1251197148512072052, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1251194772300279900, 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/dxjs/EN, language=EN, createTime=1776238246324, createBy=18614031015, updateTime=1776238398944, updateBy=18614031015, name=电讯技术-英文, tplId=1146101810881728533, title=Telecommunication Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1251197930175152619, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=articleTextType, value=kx, createTime=1776238432687, updateTime=1776238432687, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930154181096, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=banner, value=null, createTime=1776238432682, updateTime=1776238432682, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930200318446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=grayFlag, value=0, createTime=1776238432693, updateTime=1776238432693, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930141598183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=logo, value=https://castjournals.cast.org.cn/joweb/dxjs/EN/file/pic?fileId=BBd4SC9puESjyaw04bneig==, createTime=1776238432679, updateTime=1776238432679, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930212901360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=minRunFlag, value=0, createTime=1776238432696, updateTime=1776238432696, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930170958314, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dxjs/EN/file/pic, createTime=1776238432686, updateTime=1776238432686, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930204512751, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=silenceFlag, value=0, createTime=1776238432694, updateTime=1776238432694, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930162569705, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1776238432684, updateTime=1776238432684, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930183541228, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=themeColor, value=null, createTime=1776238432689, updateTime=1776238432689, creator=18614031015, updator=18614031015), WebsiteProps(id=1251197930191929837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1251197148512072052, code=themeStyle, value=null, createTime=1776238432691, updateTime=1776238432691, creator=18614031015, updator=18614031015)])], journalTitle=电讯技术, weixinUrl=null, journalUrl=https://www.teleonline.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Telecommunication Engineering, journalPhotoCn=CpBmHoMzpESavU+iEMTBmw==, journalPhotoEn=jCOIy2zOaGJZ/y3z2gPZzg==, journalFirstLetter=T, 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/dxjs/CN/10.20079/j.issn.1001-893x.240530003, detailUrlEn=https://castjournals.cast.org.cn/joweb/dxjs/EN/10.20079/j.issn.1001-893x.240530003, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dxjs/CN/PDF/10.20079/j.issn.1001-893x.240530003, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dxjs/EN/PDF/10.20079/j.issn.1001-893x.240530003, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
一种基于LightGBM的UWB非视距识别方法
收藏切换
PDF下载
李乾 1 , 刘卓伦 2 , 孙晓云 2 , 陈勇 2 , 宋士济 2 , 张醒龙 2
电讯技术 | 应用基础与前沿技术 2025,65(11): 1766-1772
收起
收藏切换
电讯技术 | 应用基础与前沿技术 2025, 65(11): 1766-1772
一种基于LightGBM的UWB非视距识别方法
全屏
李乾1, 刘卓伦2, 孙晓云2 , 陈勇2, 宋士济2, 张醒龙2
作者信息
  • 1国网河北省电力有限公司石家庄供电分公司,石家庄 050004
  • 2石家庄铁道大学 电气与电子工程学院,石家庄 050043
  • 李乾 男,1984年生于吉林磐石,2006年获学士学位,现为高级工程师,主要从事电力电缆研究与应用、电缆隧道智能监测技术方面的工作。

    刘卓伦 男,1999年生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

    孙晓云 女,1971年生于河北晋州,2000年获博士学位,现为教授,主要研究方向为电缆隧道环境智能监测技术、大型电器设备无损检测、电机电器智能控制。

    陈勇 男,1978年生于河北,博士,副研究员,主要研究方向为超宽带室内精确定位技术、超宽带非视距识别与缓解等。

    宋士济 男,1999年出生于河北石家庄,2021年获学士学位,现为硕士研究生,主要研究方向为目标检测技术。

    张醒龙 男,2001年出生于天津,2023年获学士学位,现为硕士研究生,主要研究方向为超宽带室内定位技术。

通讯作者:

孙晓云 Email:
A UWB Non-Line-of-Sight Recognition Method Based on LightGBM
Qian LI1, Zhuolun LIU2, Xiaoyun SUN2 , Yong CHEN2, Shiji SONG2, Xinglong ZHANG2
Affiliations
  • 1Shijiazhuang Power Supply Branch of State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050004, China
  • 2School of Electrical and Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China
出版时间: 2025-11-28 doi: 10.20079/j.issn.1001-893x.240530003
文章导航
收藏切换

针对超宽带非视距(Non-Line-of-Sight,NLOS)识别中最优特征子集选取与模型参数优化问题,提出了一种基于轻量级梯度提升机(Light Gradient Boosting Machine,LightGBM)的交叉验证递归特征消除算法与Optuna参数调优相结合的NLOS识别方法。首先通过递归特征消除加交叉验证算法分析选取首径信号与总信号接收功率差值、噪声最大值等6个重要特征作为最优特征子集,之后使用Optuna调参框架优化LightGBM模型超参数。采集视距与非视距特征数据,使用支持向量机、极限梯度提升算法和参数优化后的LightGBM等模型进行训练与测试,结果表明,所选取特征具有良好区分性,参数优化后的LightGBM模型识别准确率达95.28%。

超宽带非视距识别  /  轻量级梯度提升机(LightGBM)  /  交叉验证递归特征消除算法(RFECV)  /  超参数优化

For the optimal feature subset selection and model parameter optimization in ultra-wideband non-line-of-sight(NLOS) recognition,a new NLOS recognition method based on the cross-validation recursive feature elimination algorithm of Light Gradient Boosting Machine(LightGBM) and Optuna parameter tuning is proposed. First,six important features,including the difference between the first path signal and the total received signal power,and the maximum noise,are selected as the optimal feature subset using the recursive feature elimination and cross-validation algorithm. Then,Optuna is used to optimize the hyperparameters of LightGBM model. Line-of-sight and non-line-of-sight feature data is collected,and the Support Vector Machine,Extreme Gradient Boosting algorithm,and parameter-optimized LightGBM model are trained and tested. The results demonstrate that the selected features exhibit excellent discriminative ability,with the optimized LightGBM model achieving a recognition accuracy of 95.28% .

UWB NLOS recognition  /  light gradient boosting machine(LightGBM)  /  recursive feature elimination with cross validation(RFECV)  /  hyperparameter optimization
李乾, 刘卓伦, 孙晓云, 陈勇, 宋士济, 张醒龙. 一种基于LightGBM的UWB非视距识别方法. 电讯技术, 2025 , 65 (11) : 1766 -1772 . DOI: 10.20079/j.issn.1001-893x.240530003
Qian LI, Zhuolun LIU, Xiaoyun SUN, Yong CHEN, Shiji SONG, Xinglong ZHANG. A UWB Non-Line-of-Sight Recognition Method Based on LightGBM[J]. Telecommunication Engineering, 2025 , 65 (11) : 1766 -1772 . DOI: 10.20079/j.issn.1001-893x.240530003
超宽带(Ultra Wide Band,UWB)定位技术具有穿透性强、功耗低、抗多径能力强、定位精度较高的优势[1-2],近年来在军事、民生等领域广泛应用。但是超宽带信号容易受到遮挡,产生非视距(Non-Line-of-Sight,NLOS)传播,进而定位精度大幅降低。如何识别非视距信号并减轻对定位精度的影响一直是重点研究方向。
随着机器学习的发展,基于机器学习的超宽带NLOS识别方法受到广泛研究。戢静红等人[3]使用随机森林(Random Forest,RF)算法以发射机和接收机之间的测量距离作为特征来进行NLOS识别。信道脉冲响应(Channel Impulse Response,CIR)的波形中隐含着信号在传输过程中的信道信息[4],因此可以通过提取CIR波形中的特征信息,来区分信号的视距与非视距情况。例如Zeng等人[5]选择接收信号能量、均方根时延扩展等18个特征,基于遗传算法利用支持向量机(Support Vector Machine,SVM)对不同特征组合的数据进行训练,找出了准确率较高的特征组合,最后采集新的测试数据来验证模型在不同特征组合下的性能。在选取CIR特征时,峭度、均方根时延扩展等特征不能直接从超宽带信号收发器获取,需要通过复杂计算得到,因此也有学者根据超宽带收发器中容易获得或计算的特征来进行NLOS识别。Andre等人[6]使用SVM、分类与回归树(Classification and Regression Tree,CART)、RF、K近邻(K-Nearest Neighbor,KNN)、朴素贝叶斯(Naive Bayes,NB)分类算法,根据从超宽带信号收发器获得的14个特征来进行NLOS判别。袁驰等人[7]通过使用极限梯度提升算法(Extreme Gradient Boosting,XGboost),选取CIR部分点或脉冲通道响应动态范围等特征,从两方面对超宽带信号进行NLOS识别,识别准确率可达92%。Su等人[8]根据从CIR提取的最大振幅、平均超额延迟等特征使用Catboost算法对UWB信号进行非视距识别。Cui等人[9]选取噪声标准差等8个信道特征,结合特征相关性分析,使用遗传算法优化轻量级梯度提升机(Light Gradient Boosting Machine,LightGBM)超参数。在选取特征训练模型时,特征的数量与组合对模型的性能影响较大,因此需要选取重要特征子集。另外,模型的超参数需要优化来提升模型的识别性能。
针对超宽带NLOS识别中需要进行特征选择和超参数需要优化的问题,本文提出了一种基于LightGBM的非视距识别方法。本文所提方法的创新点如下:
1)将超宽带信号收发器获得的信号和经过计算的两种接收功率和功率差作为特征,采用交叉验证递归特征消除算法(Recursive Feature Elimination with Cross Validation,RFECV)选择最优特征子集。
2)使用Optuna调优框架优化LightGBM模型超参数,之后根据最优特征子集,应用调优后的LightGBM模型来识别超宽带信号。
LightGBM算法具有运算速度快、占用内存小等优点[10],并且支持高效率并行训练、可分布式处理大量数据,因此选择LightGBM模型进行识别。非视距识别时,所选择的特征可能具备不同的区分能力,需要选择最优特征子集。递归特征消除算法通过反复训练模型并剔除最不重要的特征来选择特征子集。在最优特征子集确定后,模型的不同超参数对模型性能也有较大影响,有必要确定较优的超参数来提高模型性能。Optuna是一个自动搜索超参数的优化框架,其能够大幅减少人工的工作量,有助于进一步提高模型的识别性能。本文基于LightGBM算法,结合RFECV算法进行最优特征子集选择,有助于得到重要特征,使用Optuna调参框架优化超参数,可以进一步提高模型性能。
LightGBM算法是2017年由微软亚洲研究院提出的一款基于梯度提升决策树(Gradient Boosting Decision Tree,GBDT)的机器学习算法。LightGBM的提出是为了解决梯度提升决策树算法在特征维数高、数据量大时效率较低的问题[11]。LightGBM的基本原理是对于一组输入数据,目的是为fx)找到一个近似函数,能够使损失函数Lyfx))期望最小化,如式(1)所示:
最终模型可以由m个决策树Tnx)(n= 1,2,…,m)进行近似,即
在第k次迭代时,模型的总损失为
在每一轮迭代后,模型都会计算损失值,然后构建新的树,新决策树的目标是最小化损失函数,尽可能减少预测值与真实值差异,不断提高性能。
LightGBM算法采用了一些优化策略,主要是直方图优化策略、基于梯度的单边采样策略(Gradient-based One-Side Sampling,GOSS)、互斥特性捆绑策略(Exclusive Feature Bundling,EFB)、带深度限制的按叶子生长策略[12]。直方图算法是将连续的浮点特征值离散成k个整数,并构造一个宽度为k的直方图,之后遍历数据,统计离散值在直方图中的统计量,这样在特征选择时只需要根据直方图的离散值来寻找最优分割点。直方图优化算法如图1所示。
单边梯度采样策略是以样本的梯度作为权重来进行采样,保留梯度大的样本,在梯度小的样本上随机采样。梯度较小的样本数量减少后,计算所需要的内存就能减少。在处理高维稀疏数据时,LightGBM通过特征捆绑来降低内存消耗和计算时间。EFB将相互排斥的特征进行绑定,形成一个新特征,特征捆绑可以减少特征数量[13],从而提升计算效率。LightGBM采用的leaf-wise按叶子生长策略是每次从当前所有叶子中找到分裂增益最大的一个叶子,然后分裂,如此循环,因此与按层生长(level-wise)策略相比可减小误差。带深度限制的按叶子生长策略如图2所示。
特征选取是一个重要步骤,选择最优的特征子集能够有效提升模型准确率。递归特征消除算法属于封装式算法,是一种基于模型的特征选择方法,具有较强泛化能力。
交叉验证(Cross Validation,CV)是一种常用的模型评估方法,其能够充分利用数据集,减少样本选择偏差,提供更加准确的模型评估。交叉验证递归特征消除算法将与递归特征消除相同的处理过程应用在每一折数据中[14],因此将交叉验证与递归特征消除算法相结合能够选择对模型贡献度较大的特征,并提供更可靠的评估结果。使用RFECV算法选取最优特征子集的流程如图3所示。通过逐渐减小特征数量,来找到最优特征子集,在每一折数据上进行验证模型性能,最后将最优特征子集输出并进行后续的模型训练与参数调优任务。
2019年Akiba等人[15]提出了一种超参数优化框架,其能够在各种复杂环境和超参数空间中自动调整机器学习、深度学习模型的参数。Optuna超参数调优框架采用高效的采样策略,例如随机搜索、TPE(Tree-structured Parzen Estimator)等,能够缩短参数调优时间,减少内存占用;Optuna支持分布式计算,可以并行优化任务;Optuna具有较高的灵活性,用户可采用离散或连续的超参数并且自定义优化目标函数;Optuna还提供了许多可视化工具,能够帮助用户更好地理解超参数优化过程和优化结果。
使用Optuna调优框架进行参数优化过程如下:
1)定义目标函数、参数类型和搜索空间。
2)初始化学习,设置优化方式和迭代次数等。
3)进行迭代调优。
4)满足迭代次数或时间限制等条件后终止。
5)输出最优超参数值和目标函数值。
通过上述步骤,Optuna能够有效搜索并输出满足要求的参数组合,之后用户可根据最优参数组合来训练模型。
本文所提出的基于LightGBM的UWB非视距识别算法流程如图4所示。首先采集不同遮挡场景和非遮挡场景的特征数据集,对数据集进行预处理。之后进行阶段判别,如果是离线训练阶段则利用RFECV算法进行最优特征子集选择。然后根据所选择最优特征子集构建LightGBM模型,利用Optuna调参框架对模型6个关键参数进行优化,选取Optuna优化后的参数训练最终模型,最后使用测试集对优化后的模型进行测试评估。如果是在线判别阶段,则直接将采集到的特征数据输入到训练后的模型中进行识别,并给出识别结果。
本文所使用的特征主要是从超宽带收发器的信道诊断中提取的,超宽带收发器中的信道诊断特征可以跟随测距值一起输出,具有较高实时性。表1是直接从DW1000收发器中获得的信道脉冲响应(Channel Impulse Response,CIR)特征和经过原始特征计算得到的信号功率特征。可以通过通用封装函数dwt_readdiagnostics()来获取FP_IDX、MAX_NOISE、FPAMP1、FPAMP2、FPAMP3、STDEV _NOISE、maxGrowthCIR、rxPreamCount这些信号特征的具体值。LDE_PPINDX、LDE_PPAMPL这两个特征值可以通过读取对应寄存器值来获得。
另外的特征FP_L、RX_L、PD可以通过计算来获得。可使用公式(4)计算首径信号功率[16](单位:dBm):
式中:FPAMP1、FPAMP2、FPAMP3是首径采样点处第1、2、3的幅值;A是常数(当脉冲重复频率PRF为16 MHz时,A等于113.77;当脉冲重复频率PRF为64 MHz时A等于121.74);N为对应寄存器中前导码累计计数值。
可使用公式(5)计算接收信号功率[16](单位:dBm):
式中:C为信道脉冲响应功率值,在寄存器中可读取到;AN均可参考公式(4)。
通过使用LightGBM作为有监督学习分类器,结合RFECV算法选取最优特征组合。LightGBM模型提供了feature_importance属性,此属性可在模型训练完成后查看所有特征的重要性分数。特征的重要性分数主要是从这些特征在树中作为划分属性的次数和该特征作为分割带来的总增益来计算的。本文所选择特征的重要性分数如图5所示,并且通过递归特征消除加交叉验证算法最终选择的最优特征子集为FPAMP1、FPAMP2、MAX_NOISE、FP_L、RX_L、PD。
图5可知,RFECV算法所选择的特征也是重要性分数排名前6的特征。为了验证最优特征子集,故按照上述特征重要性顺序,逐渐增加特征数量,进行模型训练与测试。模型参数选择默认值,记录模型准确率,得到不同特征数量准确度对比如图6所示。使用前6个特征时其准确率达到了93%左右,后面随着特征数量的增加模型的准确度提升不明显。因此,为了平衡模型复杂度和模型的性能,选择前6个特征作为最优特征子集。
使用Optuna调参框架对LightGBM模型进行参数调优时,设置目标函数中准确率作为目标进行优化,设置关键参数在合理范围内,根据计算资源限制和模型准确率提升程度,设置n_trials为500,得到6个关键超参数优化后的值,如表2所示。
选择前文的最优特征子集,使用表2参数训练LightGBM模型,并与其他模型进行对比测试。
实验所使用的UWB测距和定位模块是基于Decawave公司的DW1000芯片而设计,该定位模块主控芯片为STM32F103C8T6,使用外置3.3 V电源供电。配置芯片的基本工作参数为脉冲重复频率64 MHz,前导码数量1024,数据速率为110 kb/s。选取5种遮挡环境进行数据采集,即玻璃遮挡、木门遮挡、铁门遮挡、墙壁遮挡和人体遮挡。LOS环境数据选取1、3、5、9、15和1~15 m中随机位置采集。每个NLOS场景标签与基站的距离选取1、3、5、9、15 m进行数据采集。数据采集时通过串口输出所选择的16进制特征数值,并保存在txt文件中,后经过Python程序处理成十进制数,并以表格的形式保存。最后共获得25000组数据,并且保证NLOS环境和LOS环境数据平衡。
在分类模型中常用的评价标准有准确率(Accuracy)、精确率(Precision)、召回率(Recall)和F1分数,这些数据可由混淆矩阵中的数据计算出来。在二分类中,混淆矩阵是一个2×2的表格,包含真正例(TP)、假正例(FP)、真负例(TN)和假负例(FN),混淆矩阵如表3所示。
真正例表示实际信号是NLOS并且预测的也是NLOS,假正例表示实际信号是LOS而预测的是NLOS,真负例表示实际是LOS并且预测的也是LOS,假负例表示实际信号是NLOS而预测是LOS。准确率、精确率、召回率和F1分数的计算公式如式(6)~(9)所示。
准确率是模型整体的分类精度,计算公式如下:
精确率是衡量分类成NLOS的结果中真实是NLOS情况的比例,计算公式如下:
召回率是衡量分类成NLOS的能力,也被称为灵敏度。其计算公式如下:
F1分数是精确率和召回率的调和平均,可以分析模型的综合性能。其计算公式如下:
为验证本文所提UWB非视距识别方法的有效性,将本文所提方法与常用模型SVM[5]、XGboost算法[7]、Catboost算法[8]、经过递归特征消除加交叉验证算法进行特征选择后的LightGBM模型(RFECV-LightGBM)和经过Optuna参数调优后的RFECV-LightGBM模型进行对比分析,评估指标采用准确率、精确率、召回率和F1分数。所有模型都在Python3.11中使用scikit-learn工具箱实现,并在相同数据集下进行训练和测试。数据集按照80%训练集和20%测试集随机分配,设置随机种子为42,得到不同模型分类指标如图7所示。
图7可知,经过RFECV特征选择和Optuna参数调优的LightGBM模型表现最好,准确率达到了95.28%,精确率、召回率和F1分数都在95%左右。采用默认参数的LightGBM模型准确率为92.8%,其余3个参数在92.6%左右。其余模型中,SVM模型准确率最低为88.76%,精确率、召回率、F1分数分别为89.70%、87.04%、88.35%。XGboost模型的准确率为92.06%,精确率为92.43%,召回率和F1分数分别为91.27%和91.84%。Catboost模型的准确率为89.79%,其余3个参数在89%左右。
结果表明,经过递归特征消除算法后选出的6个特征具有良好的区分性,并且该特征子集在基于梯度提升决策树的模型上均具有较高的准确性,对区分超宽带NLOS与LOS信号具有较好的效果。最后,经过Optuna参数调优之后的LightGBM模型准确率提升了2.48%,模型的灵敏度和综合性能均得到一定提升。
本文以LightGBM算法为基础,利用交叉验证递归特征消除算法选出对超宽带NLOS识别影响力较大的6个特征,并且利用Optuna调参框架对模型关键超参数进行调优,提高LightGBM模型的分类性能。通过搭建非视距和视距实验环境采集超宽带NLOS和LOS特征数据集,将SVM、XGboost、Catboost、未进行Optuna参数调优的LightGBM模型、经过Optuna参数调优后的LightGBM模型进行测试对比,结果表明首径信号与总信号接收功率差值等6个关键特征具有较好的区分度,并且通过Optuna调参以后,模型的性能得到显著提升。与此同时,这些特征也为实时判别NLOS和LOS信号提供了更多选择。
本文提出的方法虽然在区分超宽带NLOS和LOS信号方面具有较高的准确度,但是为了提升实际定位精度,如何缓解非视距对测距和定位的影响还需要进一步深入研究。
参考文献 引证文献
排序方式:
[1]
LIU F, LI X, WANG J, et al. An adaptive UWB/MEMSIMU complementary Kalman filter for indoor location in NLOS environment[J]. Remote Sensing, 2019, 11(22):1-7.
[2]
ZAFARI F, GKELIAS A, LEUNG K K. A survey of indoor localization systems and technologies[J]. IEEE Communications Surveys & Tutorials, 2019, 21(3):2568-2599.
[3]
戢静红, 张振宇, 邓平, .一种基于随机森林的LOS/NLOS基站识别方法[J].电讯技术, 2023, 63(10):1596-1602.
[4]
畅田田, 王威, 高婧洁, .基于信道测量实验的NLOS误差消除方法对比研究[J].西北工业大学学报, 2022, 40(4):865-874.
[5]
ZENG Z Q, LIU S, WANG L. UWB NLOS identification with feature combination selection based on genetic algorithm[C]//2019 IEEE International Conference on Consumer Electronics. Las Vegas:IEEE, 2019:1-5.
[6]
ANDRE A G F, FERNANDES D, BRANCO S, et al. Feature selection for real-time NLOS identification and mitigation for body-mounted UWB transceivers[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70:1-10.
[7]
袁驰, 黄波. UWB定位中基于XGBoost的NLOS识别方法[J].移动通信, 2023, 47(5):83-89.
[8]
SU C, ZHANG F S, JIN B H, et al. UWBLoc: NLoS mitigation for practical UWB localization[C]//2023 IEEE Smart World Congress. Portsmouth:IEEE, 2023:1-8.
[9]
CUI X R, LI Y X, LI J, et al. UWB wireless positioning method based on LightGBM[J]. Wireless Personal Communications, 2024, 137(2):1141-1160.
[10]
BENTÉJAC C, CSÖRGO A, MARTÍNEZ-MUÑOZ G. A comparative analysis of gradient boosting algorithms[J].Artificial Intelligence Review, 2021, 54(3):1937-1967.
[11]
KE G L, MENG Q, FINLEY T, et al. LightGBM:a highly efficient gradient boosting decision tree[C]//The 31st Conference on Neural Information Processing Systems. New York:ACM, 2017:3149-3157.
[12]
YIN Z Q, SHI L, LUO J R, et al. Pump feature construction and electrical energy consumption prediction based on feature engineering and LightGBM algorithm[J]. Sustainability, 2023, 15(1):789-798.
[13]
徐国天, 沈耀童.基于XGBoost和LightGBM双层模型的恶意软件检测方法[J].信息网络安全, 2020, 20(12):54-63.
[14]
AKKAYA B. The effect of recursive feature elimination with cross-validation method on classification performance with different sizes of datasets[C]//The 4th International Conference on Data Science and Applications. Turkey:IEEE, 2021:142-152.
[15]
AKIBA T, SANO S, YANASE T, et al. Optuna: a next-generation hyperparameter optimization framework[C]//The 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Anchorage:ACM, 2019:2623-2631.
[16]
DecaWave. DW1000 user manual[EB/OL]. (2017-10-26)[2024-05-22]. https://www.decawave.com/product-documentation/.
2025年第65卷第11期
PDF下载
119
56
引用本文
BibTeX
文章信息
doi: 10.20079/j.issn.1001-893x.240530003
  • 接收时间:2024-05-30
  • 首发时间:2026-04-15
  • 出版时间:2025-11-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-30
  • 修回日期:2024-09-25
基金
作者信息
    1国网河北省电力有限公司石家庄供电分公司,石家庄 050004
    2石家庄铁道大学 电气与电子工程学院,石家庄 050043

通讯作者:

孙晓云 Email:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/dxjs/CN/10.20079/j.issn.1001-893x.240530003
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