Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Radio frequency interference mitigation using higher order statistics
Thesis (PhD)--Stellenbosch University, 2024.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613907653230592 |
|---|---|
| access_status_str | Open Access |
| author | Van Tonder, Vereese |
| author2 | Gilmore, Jacki |
| author_browse | Gilmore, Jacki Van Tonder, Vereese |
| author_facet | Gilmore, Jacki Van Tonder, Vereese |
| author_sort | Van Tonder, Vereese |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description |
Thesis (PhD)--Stellenbosch University, 2024. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/131961 |
| institution | Stellenbosch University (South Africa) |
| last_indexed | 2026-06-10T12:43:36.390Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/131961 Radio frequency interference mitigation using higher order statistics Van Tonder, Vereese Gilmore, Jacki Schwardt, L. Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Radio -- Interference Electromagnetic interference Fourier transform infrared spectroscopy Radio astronomy UCTD Thesis (PhD)--Stellenbosch University, 2024. ENGLISH ABSTRACT: Radio frequency interference (RFI) deteriorates the scientific observation at hand. It can be artificial or natural. We live in an era where more devices are equipped with an ever-increasing wide bandwidth. New-generation radio telescopes, such as the MeerKAT radio telescope, have an increased sensitivity to astronomical signals and RFI; this exacerbates the problem RFI poses. Traditionally, received random processes are characterised using the power spectrum, the Fourier transform of the signal’s autocorrelation. This is known as the Wiener-Khintchine theorem and is widely used in the radio astronomy community. Most radio astronomical sources are considered to have a Gaussian distribution, which is fully characterised using second-order power statistics. From third-order, statistics are classified as higher-order statistics (HOS) and are not commonly used within the field. HOS is a tool that can distinguish the non-Gaussian components from the Gaussian components of a signal. Therefore, when a signal of interest (SOI) has non-Gaussian components, it can be detected using HOS. Generally, RFI also has a non-Gaussian distribution. This work investigates the usage of HOS as a tool for RFI mitigation. Initially, the project investigates the use of the bispectrum as an RFI mitigation technique. The bispectrum is a third-order statistic. The technique is applied to simulated GPS data. Energy is found within specific GPS codes; however, it is not present in all due to experimental limitations. This, combined with the fact that the technique is computationally costly, led to the investigation of another HOS-based technique. The spectral kurtosis (SK) estimator uses power statistics, though it is based on the fourth-order moment, namely kurtosis. The effects of data clipping, zeros in the data, quantisation, and gain variation on the SK estimator are investigated. The SK estimator is initially applied to Vela pulsar data. Afterwards, the thresholds are optimised for the fastest millisecond pulsar, J0437-4715. Therefore, an estimate of the timing uncertainty in pulsar time of arrival (TOA) is used as a performance metric. The SK estimator is compared to other RFI mitigation strategies, highlighting its strengths and weaknesses. Using the best SK estimator decreases the timing uncertainty for J0437-4715 by 53.57% compared to not conducting any RFI mitigation. Compared to a median-smoothing filter, the best SK estimator yields a 28.34% lower uncertainty. The SK estimator performs even better when a multiscale version is used. The MSK achieves a 2.95% lower timing uncertainty compared to the best single-scale SK estimator. Therefore, an overall 54.94% improvement in timing uncertainty can be achieved using SK. AFRIKAANSE OPSOMMING: Radiofrekwensiesteuring (RFI) verswak ’n wetenskaplike waarneming. Dit kan mensgemaak of natuurlik wees. Ons leef in ’n era waar meer toestelle toegerus is met ’n toenemende bandwydte. Nuwegenerasie-radioteleskope, soos die MeerKAT-radioteleskoop, het ’n verhoogde sensitiwiteit vir astronomiese seine en RFI; dit vererger die probleem wat RFI voorstel. Tradisioneel word ewekansige prosesse gekenmerk deur gebruik te maak van die drywingspektrum, wat die Fourier transformasie van die outokorrelasie van ’n sein is. Dit staan bekend as die Wiener-Khintchine-stelling en word wyd gebruik in die radio sterrekunde gemeenskap. Die meeste radio-astronomiese bronne word beskou as ’n Gaussiese verspreiding, wat volledig gekenmerk word deur gebruik te maak van tweede-orde statistieke. Vanaf derde-orde, word statistieke geklassifiseer as ho¨er-orde statistieke (HOS). Dit word nie algemeen binne die radio-astronomie veld gebruik nie. HOS is ’n instrument wat die nie-Gaussiese komponente van ’n sein kan onderskei van die Gaussiese komponente. Daarom, wanneer ’n sein van belang nie-Gaussiese komponente het, kan dit deur middel van HOS opgespoor word. Gewoonlik het RFI ook ’n nie-Gaussiese verspreiding. Hierdie werk ondersoek die gebruik van HOS as ’n instrument vir RFI versagting. Aanvanklik ondersoek die projek die gebruik van die bispektrum as ’n RFI versagtingstegniek. Die bispektrum is ’n derde-orde statistiek. Die tegniek word toegepas op gesimuleerde GPS-data. Energie word binne sekere GPS-kodes gevind, maar is nie in almal teenwoordig nie. Dit, tesame met die feit dat die tegniek rekenaarmatig baie duur is, het gelei tot die ondersoek van ’n ander HOS-gebaseerde tegniek. Die spektrale kurtose (SK) tegniek gebruik drywingstatistieke, alhoewel dit gebaseer is op die vierde moment, naamlik kurtose. Die effekte van data knipsels, nulle in die data, kwantiseering, en aanwins variasie op SK word ondersoek. Die SK tegniek word aanvanklik toegepas op Vela pulsar data. Daarna word die drempels geoptimeer vir die vinnigste millisekonde pulsar, J0437-4715. Die tydsberekeningonsekerheid word gebruik as die prestasiemetriek. SK word vergelyk met ander RFI versagting tegnieke, waar SK se swakhede en sterktes beklemtoon word. Die gebruik van die beste SK beramer verminder die tydsberekeningonsekerheid vir J0437-4715 met 53.57% in vergelyking met geen RFIversagting nie. In vergelyking met ’n mediaan filter, lewer die beste SK tegniek ’n 28.34% laer onsekerheid. Die SK presteer selfs beter wanneer ’n meerskaalse weergawe daarvan gebruik word, wat tydsberekeningonsekerheid met 2.95% verlaag, in vergelyking met die beste SK tegniek. Daarom kan ’n algehele verbetering van 54.94% in tydsberekeningonsekerheid bereik word deur gebruik te maak van SK. Doctoral 2025-05-02T09:15:31Z 2025-05-02T09:15:31Z 2024-12 Thesis https://scholar.sun.ac.za/handle/10019.1/131961 Stellenbosch University xv, 122 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Radio -- Interference Electromagnetic interference Fourier transform infrared spectroscopy Radio astronomy UCTD Van Tonder, Vereese Radio frequency interference mitigation using higher order statistics |
| title | Radio frequency interference mitigation using higher order statistics |
| title_full | Radio frequency interference mitigation using higher order statistics |
| title_fullStr | Radio frequency interference mitigation using higher order statistics |
| title_full_unstemmed | Radio frequency interference mitigation using higher order statistics |
| title_short | Radio frequency interference mitigation using higher order statistics |
| title_sort | radio frequency interference mitigation using higher order statistics |
| topic | Radio -- Interference Electromagnetic interference Fourier transform infrared spectroscopy Radio astronomy UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/131961 |
| work_keys_str_mv | AT vantondervereese radiofrequencyinterferencemitigationusinghigherorderstatistics |