Close

1. Identity statement
Reference TypeConference Paper (Conference Proceedings)
Sitemtc-m16d.sid.inpe.br
Identifier8JMKD3MGPDW34R/45QLC4P
Repositorysid.inpe.br/mtc-m16d/2021/11.17.19.11
Last Update2021:11.17.19.11.55 (UTC) administrator
Metadata Repositorysid.inpe.br/mtc-m16d/2021/11.17.19.11.55
Metadata Last Update2022:04.03.19.25.44 (UTC) administrator
ISSN2177-3114
Citation KeySilvaJúniorDosS:2021:HyMaLe
TitleA hybrid machine learning process for anomalous satellite telemetry behaviour detection
FormatOn-line.
Year2021
Access Date2024, May 18
Secondary TypePRE CN INT
Number of Files1
Size1330 KiB
2. Context
Author1 Silva Júnior, Márcio Waldir
2 Dos Santos, Walter Abrahão
Resume Identifier1
2 8JMKD3MGP5W/3C9JJC2
Group1 CSE-ETES-DIPGR-INPE-MCTI-GOV-BR
2 DIPST-CGCE-INPE-MCTI-GOV-BR
Affiliation1 Instituto Nacional de Pesquisas Espaciais (INPE)
2 Instituto Nacional de Pesquisas Espaciais (INPE)
Author e-Mail Address1 marciowaldir.sjr@outlook.com
2 walter.abrahao@inpe.br
EditorRodrigues, Aline Castilho
Rodrigues, Italo Pinto
Santos, Walter Abrahão dos
Mateus, Dairo Antonio Cuellar
Machado, Danilo
Diniz, Gledson Hernandes
Dallamuta, João
Carmo, Thiago Augusto do
Siqueli, Guilherme Afonso
Conference NameWorkshop em Engenharia e Tecnologia Espaciais, 12 (WETE)
Conference LocationSão José dos Campos
Date6, 7, 13 e 14 nov. 2021
PublisherInstituto Nacional de Pesquisas Espaciais (INPE)
Publisher CitySão José dos Campos
Tertiary Typefull paper
OrganizationInstituto Nacional de Pesquisas Espaciais (INPE)
History (UTC)2021-11-17 19:12:21 :: simone -> administrator :: 2021
2022-04-03 19:25:44 :: administrator -> simone :: 2021
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
KeywordsMachine Learning
KPCA
DBSCAN
KNN
Anomalous Behaviour
AbstractIn space missions, telemetry is a key source of systems health, and the lack of this may compromise the mission. Even with functional telemetry, there are some difficulties regarding the telemetry analysis. Some satellites have hundreds, even thousands of telemetries, and analyzing that to infer something about the system tends to be quite laborious. In this scenario, it can be difficult to perform in advance the detection, diagnosis, and prevention of anomalies and failures, decreasing the reliability and availability of space systems. Thus, shortening the system life and service continuity. This study proposes a data-driven approach, composed of a hybrid Machine Learning (ML) process capable of detecting anomalous behaviour on telemetry. Through statistics, data science processes, and ML algorithms, the proposed process was capable of classifying, with more than 90% of accuracy, four different circumstances for the telemetry system, sunlight, eclipse, twilight, and anomalous behaviour.
AreaETES
TypeCSE
Arrangement 1urlib.net > BDMCI > Fonds > Produção pgr ATUAIS > CSE > A hybrid machine...
Arrangement 2urlib.net > BDMCI > Fonds > Produção a partir de 2021 > CGCE > A hybrid machine...
Arrangement 3A hybrid machine...
doc Directory Contentaccess
source Directory Contentthere are no files
agreement Directory Contentthere are no files
4. Conditions of access and use
data URLhttp://urlib.net/ibi/8JMKD3MGPDW34R/45QLC4P
zipped data URLhttp://urlib.net/zip/8JMKD3MGPDW34R/45QLC4P
Languagept
Target File46 - [ARTIGO][INPE] MARCIO WALDIR SILVA JUNIOR.pdf
User Groupsimone
Reader Groupadministrator
italo.rodrigues@inpe.br
simone
Visibilityshown
Copyright Licenseurlib.net/www/2012/11.12.15.19
Rightsholderoriginalauthor yes
Read Permissionallow from all
5. Allied materials
Next Higher Units8JMKD3MGPCW/3F35BSP
8JMKD3MGPCW/46KTFK8
8JMKD3MGPDW34P/4627RDH
Citing Item Listsid.inpe.br/bibdigital/2013/10.14.22.20 2
sid.inpe.br/mtc-m21/2012/07.13.15.01.48 2
sid.inpe.br/bibdigital/2022/04.03.17.52 2
Host Collectionsid.inpe.br/mtc-m19@80/2009/08.21.17.02
6. Notes
Empty Fieldsarchivingpolicy archivist booktitle callnumber contenttype copyholder creatorhistory descriptionlevel dissemination documentstage doi e-mailaddress edition holdercode isbn label lineage mark mirrorrepository nextedition notes numberofvolumes orcid pages parameterlist parentrepositories previousedition previouslowerunit progress project schedulinginformation secondarydate secondarykey secondarymark serieseditor session shorttitle sponsor subject tertiarymark url versiontype volume
7. Description control
e-Mail (login)simone
update 


Close