kth.sePublications KTH
Operational message
There are currently operational disruptions. Troubleshooting is in progress.
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
FM Demodulation Using a Convolutional Neural Network
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2025 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

Within the ever-expanding field of machine-learning (ML), new practical applications are found nearly everywhere. This study explores if ML-based demodulation can be used as an alternative to conventional frequency-modulation (FM) radio receivers, especially under non-ideal channel conditions. Specifically, a convolutional neural network architecture for regression is proposed to demodulate FM signals represented by their in-phase and quadrature components.  A synthetic dataset of base-band FM signals, covering 20 Hz–20 kHz and augmented with a variety of typical signal disturbances, is used to train the model. Tests show that, when the incoming signal-to-noise ratio (SNR) is 10 dB or lower (the FM threshold region), the ML-model raises output SNR by up to 6 dB and reduces mean-squared error by an order of magnitude compared with the conventional method. Listening tests with 10 subjects corroborate these numerical gains. Because the model runs in real-time on a laptop CPU, ML-based demodulation could offer a practical route to more robust FM reception in noisy or adverse channels.

Abstract [sv]

Inom det ständigt växande fältet för maskininlärning (ML) upptäcks nya användningsområden nästintill överallt. I denna studie utforskas möjligheten att använda ML-baserad demodulering som alternativ till konventionella radiomottagare för frekvensmodulerade (FM) radiosignaler, särskilt i fall med icke-ideala signalförhållanden. Specifikt föreslås en konvolutionell neuronnätsarkitektur för regression för att demodulera FM-signaler representerade av deras i-fas och kvadratur-komponenter. En syntetiskt datamängd med basbands FM-signaler, som täcker frekvenser mellan 20 Hz–20 kHz och som även förvrängs med typiska störningar, används för att träna ML-modellen. Tester visar att när den inkommande signalen har ett signal-till-brus förhållande (SNR) lägre än 10 dB så ökar ML-modellen utsignalens SNR med upp till 6 dB och minskar minsta-kvadrat-felet (MSE) med cirka en storleksordning jämfört med den konventionella metoden. Lyssningstest med 10 personer bekräftar de numeriska resultaten. Eftersom modellen kan köras i realtid med en laptop-CPU så kan ML-baserad demodulering erbjuda en praktisk väg till en mer robust FM-mottagning i svårare kanalförhållanden.

Place, publisher, year, edition, pages
2025. , p. 529-539
Series
TRITA-EECS-EX ; 2025:153
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:kth:diva-376174OAI: oai:DiVA.org:kth-376174DiVA, id: diva2:2034546
Supervisors
Examiners
Projects
Kandidatexamensarbete i Elektroteknik 2025, EECS, KTHAvailable from: 2026-02-02 Created: 2026-02-02

Open Access in DiVA

fulltext(80627 kB)17 downloads
File information
File name FULLTEXT01.pdfFile size 80627 kBChecksum SHA-512
35ce0a386dafe4649eb99cbe0efdfed651a3c9044e3339612422234d17a7e8ec21d4fd4aa201500c3c7a8f57194994b78b3e0cfbd5319ecd49f18a5d8a7ff775
Type fulltextMimetype application/pdf

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

Search outside of DiVA

GoogleGoogle Scholar
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: 4048 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