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
Radar detection of aircraft using machine learning
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Radardetektion av flygplan med maskininlärning (Swedish)
Abstract [en]

SAAB is a large Swedish defense and aerospace company developing a large variety of products such as combat aircrafts, guided missiles, military submarines and many more. One important technology which has many applications is radar. Modern radar systems are used to detect and survey the surroundings both in the air and at sea and are used by the military to detect and target hostile targets. In this thesis we explore the possibility of using machine learning for detecting aircrafts on a range-doppler map using Automatic Dependent Surveillance–Broadcast (ADS-B) data as ground truth for training. A passive radar was used in combination with an ADS-B receiver to collect data used for training and evaluation. Several machine learning models were made using a logistic regression algorithm, which were then evaluated based on testing it on different radar data. Training and testing with selected and balanced radar data resulted in a precision score of 86% and recall at 88%. Testing showed that given clean radar data, a machine learning model can be trained to detect aircraft from radar imaging.

Abstract [sv]

SAAB är ett stort svenskt företag inom flyg och försvar vars verksamhet innefattar utveckling av en rad olika produkter som stridsflyg, robotar, militära ubåtar och mycket mer. En viktig teknologi för sådan utveckling av system är radar. Moderna radarsystem används för detektion och span i både luften och på havet och används av militären för att detektera och sikta in sig på fientliga mål. I denna avhandling utforskar vi möjligheten att använda maskininlärning för detektion av flygplan på en avstånd-doppler karta med data från systemet Automatic Dependent Surveillance-Broadcast (ADS-B) som facit för träning. En passiv radar användes tillsammans med en ADS-B mottagare för insamling av data till träning och utvärdering. Flera maskininlärningsmodeller skapades med en logistisk regression algoritm som sedan utvärderades baserat på tester med annan radardata. Träning och testning med utvald och balanserad data resulterade i 86% precision och 88% recall. Testerna visade att givet ren data så kan en maskininlärningsmodell tränas till att detektera flygplan på radarbilder.

Place, publisher, year, edition, pages
2024. , p. 23
Series
TRITA-EECS-EX ; 2024:357
Keywords [en]
Passive radar, logistic regression, ADS-B, radar detection
Keywords [sv]
Passiv radar, logistisk regression, ADS-B, radardetektion
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-351110OAI: oai:DiVA.org:kth-351110DiVA, id: diva2:1886205
External cooperation
SAAB AB
Supervisors
Examiners
Available from: 2024-08-23 Created: 2024-07-30 Last updated: 2024-08-23Bibliographically approved

Open Access in DiVA

fulltext(614 kB)664 downloads
File information
File name FULLTEXT01.pdfFile size 614 kBChecksum SHA-512
21fa3786340e88f69e5b0fccd0e9edf5f7c3c96c7e6a873f8eb3a9e5313d17ae2e73f2261b673c13e8a8bd35925b64e6b8e95c1fcd6d4ac8a68eaeefa30f89f3
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 665 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: 340 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