kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Evaluating the Efficiency and Error Resistance of Vector Databases: Utilizing a naive solution with generated radar signal data
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Architecture and the Built Environment (ABE).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Utvärdering av effektiviteten och feltolerans hos vektordatabaser : En naiv lösning med genererad radarsignaldata (Swedish)
Abstract [en]

This paper examines and compares the efficiency of storing and retrieving similar radar track reports in a vector database, as well as the error resistance of the database, using various indexing techniques available in Pgvector and PostgreSQL's native indexing methods. By using generated dummy data and open-source tools such as Pgvector, PostgreSQL, and the Rust programming language, the study demonstrates the implementation's high level of error resistance and rapid performance.

The results highlight the potential of vector databases in handling radar signal data, showcasing their robustness and speed in various scenarios, but fails to conclusively answer whether vector databases can reliably identify radar signals from similar sources. This suggests that further research and optimization are necessary to fully understand and enhance the capabilities of vector databases in this application.

Abstract [sv]

Denna uppsats undersöker och jämför hur effektivt en vektordatabas kan spara och hämta liknande data från signalrapporter, samt hur resistent vektordatabasen är mot felaktigheter i datan, genom att använda olika indexeringstekniker som finns tillgängliga i PostgreSQL och Pgvector. Via användning av genererad data och verktyg med öppen källkod så som PostgreSQL, Pgvector och programmeringsspråket Rust demonstrerar studien implementationens höga nivå av felresistens och höga prestanda. Resultaten påvisar implementationens stabilitet och hastighet i olika scenarion men lyckas inte get ett slutgiltigt svar på om vektordatabaser kan pålitligt indentifiera radarsignaler från liknande källor. Detta tyder på att det krävs fortsatt forksning och optimisering för att skapa gedigen förståelse i ämnet och förbättra vektordatabasers förmågor inom området.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology , 2024. , p. 56
Series
TRITA-EECS-EX ; 2024:292
Keywords [en]
Vector database, indexation, benchmark, high dimensional data, time complexity, PostgreSQL
Keywords [sv]
Vektordatabas, indexering, utvärderingsprov, högdimensionell data, tidskomplexitet, PostgreSQL
National Category
Computer and Information Sciences Computer Engineering
Identifiers
URN: urn:nbn:se:kth:diva-350915OAI: oai:DiVA.org:kth-350915DiVA, id: diva2:1885504
External cooperation
saab AB
Educational program
Bachelor of Science - Information and Communication Technology
Presentation
2024-05-30, Ka-304, Isafjordsgatan 22 (Kistagången 16), Stockholm, 15:00 (English)
Supervisors
Examiners
Available from: 2024-08-14 Created: 2024-07-23 Last updated: 2024-08-14Bibliographically approved

Open Access in DiVA

fulltext(805 kB)158 downloads
File information
File name FULLTEXT01.pdfFile size 805 kBChecksum SHA-512
cfacf46a8343855d02d6956ff660a57b100872e8db162a0216e473f0bf4d816781e9fdefc710f4bc78237843c7830a9d526d9dd53175a10a5c131f9f31f18405
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Dåderman, JonasSaskin, Deniel
By organisation
School of Electrical Engineering and Computer Science (EECS)School of Architecture and the Built Environment (ABE)
Computer and Information SciencesComputer Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 158 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 403 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf