Using machine learning for the detection of radio frequency interference

Research output: Chapter in Book/Conference paperConference paper

Abstract

Radio Astronomy, by its very nature, detects extremely faint cosmic radio signals and is therefore very susceptible to Radio Frequency Interference (RFI). We present some initial results of our work to identify RFI using a Machine Learning (ML) based approach. The data is taken from VLBI observations data from three well separated observatories in Australia: ATCA, Parkes and Mopra; and we work with the 2-bit data directly from the telescopes. Our approach uses a Generative Adversarial Network (GAN) and an autoencoder to perform unsupervised machine learning on the data.

Original languageEnglish
Title of host publication2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019
Place of PublicationIndia
PublisherIEEE, Institute of Electrical and Electronics Engineers
ISBN (Electronic)9789082598759
DOIs
Publication statusPublished - 20 Jun 2019
Event2019 URSI Asia-Pacific Radio Science Conference - India Habitat Centre, New Delhi, India
Duration: 9 Mar 201915 Mar 2019
http://aprasc2019.com/author/aprasc-2019/

Publication series

Name2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019

Conference

Conference2019 URSI Asia-Pacific Radio Science Conference
Abbreviated titleURSI AP-RASC 2019
CountryIndia
CityNew Delhi
Period9/03/1915/03/19
OtherAP-RASC is a triennial event that brings together radio science specialists from all over the world, especially from the Asia Pacific region, covering all aspects of radio science across URSI’s ten commissions. This includes aspect of radio science related to metrology, radio propagation and communications and signal processing, electronics, photonics, electromagnetic interference, remote sensing, waves in plasma, radio astronomy and applications in biology and medicine. The main objective of the Conference is to review current trends in research, present new discoveries and make plans for future research work or for specific projects, especially where it seems desirable to arrange for cooperation on an international scale. The AP-RASC 2019 features committee meetings, technical sessions, workshops, general lectures as well as an exhibition of products related to radio science. An equally important objective is also to encourage scientific exchange and fellowship amongst industry colleagues and professionals globally. This conference will be an occasion for delegates to make new acquaintances and strengthen existing friendships
Internet address

Fingerprint

radio frequency interference
Radio interference
machine learning
Learning systems
radio signals
radio astronomy
very long base interferometry
Radio astronomy
observatories
telescopes
Observatories
Telescopes

Cite this

Vinsen, K., Foster, S., & Dodson, R. (2019). Using machine learning for the detection of radio frequency interference. In 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019 [8738332] (2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019). India: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.23919/URSIAP-RASC.2019.8738332
Vinsen, Kevin ; Foster, Samuel ; Dodson, Richard. / Using machine learning for the detection of radio frequency interference. 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019. India : IEEE, Institute of Electrical and Electronics Engineers, 2019. (2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019).
@inproceedings{fdfb2b93a00746999de42dd42dbff950,
title = "Using machine learning for the detection of radio frequency interference",
abstract = "Radio Astronomy, by its very nature, detects extremely faint cosmic radio signals and is therefore very susceptible to Radio Frequency Interference (RFI). We present some initial results of our work to identify RFI using a Machine Learning (ML) based approach. The data is taken from VLBI observations data from three well separated observatories in Australia: ATCA, Parkes and Mopra; and we work with the 2-bit data directly from the telescopes. Our approach uses a Generative Adversarial Network (GAN) and an autoencoder to perform unsupervised machine learning on the data.",
author = "Kevin Vinsen and Samuel Foster and Richard Dodson",
year = "2019",
month = "6",
day = "20",
doi = "10.23919/URSIAP-RASC.2019.8738332",
language = "English",
series = "2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
booktitle = "2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019",
address = "United States",

}

Vinsen, K, Foster, S & Dodson, R 2019, Using machine learning for the detection of radio frequency interference. in 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019., 8738332, 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019, IEEE, Institute of Electrical and Electronics Engineers, India, 2019 URSI Asia-Pacific Radio Science Conference, New Delhi, India, 9/03/19. https://doi.org/10.23919/URSIAP-RASC.2019.8738332

Using machine learning for the detection of radio frequency interference. / Vinsen, Kevin; Foster, Samuel; Dodson, Richard.

2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019. India : IEEE, Institute of Electrical and Electronics Engineers, 2019. 8738332 (2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019).

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Using machine learning for the detection of radio frequency interference

AU - Vinsen, Kevin

AU - Foster, Samuel

AU - Dodson, Richard

PY - 2019/6/20

Y1 - 2019/6/20

N2 - Radio Astronomy, by its very nature, detects extremely faint cosmic radio signals and is therefore very susceptible to Radio Frequency Interference (RFI). We present some initial results of our work to identify RFI using a Machine Learning (ML) based approach. The data is taken from VLBI observations data from three well separated observatories in Australia: ATCA, Parkes and Mopra; and we work with the 2-bit data directly from the telescopes. Our approach uses a Generative Adversarial Network (GAN) and an autoencoder to perform unsupervised machine learning on the data.

AB - Radio Astronomy, by its very nature, detects extremely faint cosmic radio signals and is therefore very susceptible to Radio Frequency Interference (RFI). We present some initial results of our work to identify RFI using a Machine Learning (ML) based approach. The data is taken from VLBI observations data from three well separated observatories in Australia: ATCA, Parkes and Mopra; and we work with the 2-bit data directly from the telescopes. Our approach uses a Generative Adversarial Network (GAN) and an autoencoder to perform unsupervised machine learning on the data.

UR - http://www.scopus.com/inward/record.url?scp=85068615455&partnerID=8YFLogxK

U2 - 10.23919/URSIAP-RASC.2019.8738332

DO - 10.23919/URSIAP-RASC.2019.8738332

M3 - Conference paper

T3 - 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019

BT - 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - India

ER -

Vinsen K, Foster S, Dodson R. Using machine learning for the detection of radio frequency interference. In 2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019. India: IEEE, Institute of Electrical and Electronics Engineers. 2019. 8738332. (2019 URSI Asia-Pacific Radio Science Conference, AP-RASC 2019). https://doi.org/10.23919/URSIAP-RASC.2019.8738332