{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/139855"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/139855","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"On the Use of Deep Learning Models for Interference Detection and Mitigation","abstract":"Interference and noise are two of the primary factors that degrade the reliability of wireless communication systems. With the rapid expansion of satellite communication networks, the radio frequency spectrum has become increasingly congested, creating severe challenges not only for general wireless communication but also for highly sensitive applications such as radio astronomy. Weak astronomical signals, often several orders of magnitude below the thermal noise floor, are particularly vulnerable to disruption from anthropomorphic signals that overlap in frequency and time. This thesis addresses the problem of detecting and suppressing interference in modern wireless environments under a wide range of noise and interference conditions. Building on recent advancements in deep learning, which have shown strong capabilities in learning patterns directly from data without detailed prior models, we propose a framework composed of two major modules: interference detection and interference mitigation. Detection is performed using advanced deep learning architectures such as using a hybrid Convolutional Neural Network(CNN) models knows as InceptionTimePlus and MiniRocketPlus, with results compared to classical methods including matched filtering, energy detection, and FFT-based thresholding. Interference suppression is achieved through a hybrid approach that combines a two-stage convolutional autoencoder pipeline with adaptive successive interference cancellation, each optimized for different bandwidth interference. To further enhance adaptability, a recommender system is introduced that leverages the parameter estimates to dynamically select the most effective mitigation strategy for the given interference scenario. Experimental evaluations demonstrate that the proposed framework significantly reduces bit error rates across diverse interference conditions, providing a flexible and data-driven solution to interference management. For the radio astronomy use case, the methods achieve improvements in interference mitigation for the scenarios involving non-AWGN or frequency offsets. Additionally, CNN-based classifiers were also developed for classification of noise and interference in the mixed signal which can be then used for selecting accurate mitigation models. While motivated by the need to protect the integrity of radio astronomy observations in the presence of satellite-based interference, the techniques developed are broadly applicable to wireless communication systems operating in congested spectrum environments.","abstract_html":"Interference and noise are two of the primary factors that degrade the reliability of wireless communication systems. With the rapid expansion of satellite communication networks, the radio frequency spectrum has become increasingly congested, creating severe challenges not only for general wireless communication but also for highly sensitive applications such as radio astronomy. Weak astronomical signals, often several orders of magnitude below the thermal noise floor, are particularly vulnerable to disruption from anthropomorphic signals that overlap in frequency and time. This thesis addresses the problem of detecting and suppressing interference in modern wireless environments under a wide range of noise and interference conditions. Building on recent advancements in deep learning, which have shown strong capabilities in learning patterns directly from data without detailed prior models, we propose a framework composed of two major modules: interference detection and interference mitigation. Detection is performed using advanced deep learning architectures such as using a hybrid Convolutional Neural Network(CNN) models knows as InceptionTimePlus and MiniRocketPlus, with results compared to classical methods including matched filtering, energy detection, and FFT-based thresholding. Interference suppression is achieved through a hybrid approach that combines a two-stage convolutional autoencoder pipeline with adaptive successive interference cancellation, each optimized for different bandwidth interference. To further enhance adaptability, a recommender system is introduced that leverages the parameter estimates to dynamically select the most effective mitigation strategy for the given interference scenario. Experimental evaluations demonstrate that the proposed framework significantly reduces bit error rates across diverse interference conditions, providing a flexible and data-driven solution to interference management. For the radio astronomy use case, the methods achieve improvements in interference mitigation for the scenarios involving non-AWGN or frequency offsets. Additionally, CNN-based classifiers were also developed for classification of noise and interference in the mixed signal which can be then used for selecting accurate mitigation models. While motivated by the need to protect the integrity of radio astronomy observations in the presence of satellite-based interference, the techniques developed are broadly applicable to wireless communication systems operating in congested spectrum environments.","abstract_has_math":false,"creators":["Kothari, Hiten Prakash"],"institution":"Virginia Tech","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Computer Engineering","degree_department":"Electrical and Computer Engineering","school":null,"contributors":[],"advisors":[],"committee_chairs":["Buehrer, Richard M."],"committee_members":["Jones, Creed Farris","Dhillon, Harpreet Singh"],"year":2025,"date_issued":"2025-12-09","date_published":"2025-12-09","updated_at":"2026-07-22T22:19:35Z","subjects":["Radio Frequency Interference","Interference Mitigation","Autoencoders","Signal Detection","U-Net"],"languages":["en"],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:44919"],"render_values":[{"text":"vt_gsexam:44919","href":null,"code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/10919/139855","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Buehrer, Richard M."]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Jones, Creed Farris","Dhillon, Harpreet Singh"]},{"key":"dc:contributor.department","label":"Department","values":["Electrical and Computer Engineering"]},{"key":"dc:creator","label":"Author","values":["Kothari, Hiten Prakash"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-12-10T09:00:15Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-12-10T09:00:15Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-12-09"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Radio Frequency Interference","Interference Mitigation","Autoencoders","Signal Detection","U-Net"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:44919"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10919/139855"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Interference and noise are two of the primary factors that degrade the reliability of wireless communication systems. With the rapid expansion of satellite communication networks, the radio frequency spectrum has become increasingly congested, creating severe challenges not only for general wireless communication but also for highly sensitive applications such as radio astronomy. Weak astronomical signals, often several orders of magnitude below the thermal noise floor, are particularly vulnerable to disruption from anthropomorphic signals that overlap in frequency and time. This thesis addresses the problem of detecting and suppressing interference in modern wireless environments under a wide range of noise and interference conditions. Building on recent advancements in deep learning, which have shown strong capabilities in learning patterns directly from data without detailed prior models, we propose a framework composed of two major modules: interference detection and interference mitigation. Detection is performed using advanced deep learning architectures such as using a hybrid Convolutional Neural Network(CNN) models knows as InceptionTimePlus and MiniRocketPlus, with results compared to classical methods including matched filtering, energy detection, and FFT-based thresholding. Interference suppression is achieved through a hybrid approach that combines a two-stage convolutional autoencoder pipeline with adaptive successive interference cancellation, each optimized for different bandwidth interference. To further enhance adaptability, a recommender system is introduced that leverages the parameter estimates to dynamically select the most effective mitigation strategy for the given interference scenario. Experimental evaluations demonstrate that the proposed framework significantly reduces bit error rates across diverse interference conditions, providing a flexible and data-driven solution to interference management. For the radio astronomy use case, the methods achieve improvements in interference mitigation for the scenarios involving non-AWGN or frequency offsets. Additionally, CNN-based classifiers were also developed for classification of noise and interference in the mixed signal which can be then used for selecting accurate mitigation models. While motivated by the need to protect the integrity of radio astronomy observations in the presence of satellite-based interference, the techniques developed are broadly applicable to wireless communication systems operating in congested spectrum environments."]},{"key":"dc:description.abstractgeneral","label":"General Abstract","values":["Interference and Noise pose a primary challenge in wireless communication systems. Radio astronomy relies on detecting extremely faint signals from space, often thousands of times weaker than the signals from nearby satellites. In recent years, the rapid growth of satellite communication networks has made it increasingly difficult for observatories to avoid interference. Even brief or weak transmissions can mask or distort astronomical signals, making it challenging to study the universe with the required precision. For communication systems, interference having similar modulation and power becomes difficult to remove if the source and parameters of the said interference are unknown. This thesis focuses on creating a system that can detect when interference is present and then remove it as effectively as possible. To do this, the work combines ideas from traditional radio engineering with modern deep learning advancements. The system is built in two stages: 1. Detecting whether interference is present in the received signal stream 2. Using deep learning models to estimate and remove the interference A recommender system is also developed for choosing the best system to mitigate the interference given the parameters and scenario. CNN Classifiers are developed for categorizing the noise and interference for effective mitigation. While this research was inspired by the challenges facing radio astronomy, the techniques developed are also useful for many other situations where important signals are potentially masked in noisy, crowded parts of the radio spectrum."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["ETD"]},{"key":"dc:title","label":"Title","values":["On the Use of Deep Learning Models for Interference Detection and Mitigation"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Buehrer, Richard M."],"dc:contributor.committeemember":["Jones, Creed Farris","Dhillon, Harpreet Singh"],"dc:contributor.department":["Electrical and Computer Engineering"],"dc:creator":["Kothari, Hiten Prakash"],"dc:date.accessioned":["2025-12-10T09:00:15Z"],"dc:date.available":["2025-12-10T09:00:15Z"],"dc:date.issued":["2025-12-09"],"dc:description.abstract":["Interference and noise are two of the primary factors that degrade the reliability of wireless communication systems. With the rapid expansion of satellite communication networks, the radio frequency spectrum has become increasingly congested, creating severe challenges not only for general wireless communication but also for highly sensitive applications such as radio astronomy. Weak astronomical signals, often several orders of magnitude below the thermal noise floor, are particularly vulnerable to disruption from anthropomorphic signals that overlap in frequency and time. This thesis addresses the problem of detecting and suppressing interference in modern wireless environments under a wide range of noise and interference conditions. Building on recent advancements in deep learning, which have shown strong capabilities in learning patterns directly from data without detailed prior models, we propose a framework composed of two major modules: interference detection and interference mitigation. Detection is performed using advanced deep learning architectures such as using a hybrid Convolutional Neural Network(CNN) models knows as InceptionTimePlus and MiniRocketPlus, with results compared to classical methods including matched filtering, energy detection, and FFT-based thresholding. Interference suppression is achieved through a hybrid approach that combines a two-stage convolutional autoencoder pipeline with adaptive successive interference cancellation, each optimized for different bandwidth interference. To further enhance adaptability, a recommender system is introduced that leverages the parameter estimates to dynamically select the most effective mitigation strategy for the given interference scenario. Experimental evaluations demonstrate that the proposed framework significantly reduces bit error rates across diverse interference conditions, providing a flexible and data-driven solution to interference management. For the radio astronomy use case, the methods achieve improvements in interference mitigation for the scenarios involving non-AWGN or frequency offsets. Additionally, CNN-based classifiers were also developed for classification of noise and interference in the mixed signal which can be then used for selecting accurate mitigation models. While motivated by the need to protect the integrity of radio astronomy observations in the presence of satellite-based interference, the techniques developed are broadly applicable to wireless communication systems operating in congested spectrum environments."],"dc:description.abstractgeneral":["Interference and Noise pose a primary challenge in wireless communication systems. Radio astronomy relies on detecting extremely faint signals from space, often thousands of times weaker than the signals from nearby satellites. In recent years, the rapid growth of satellite communication networks has made it increasingly difficult for observatories to avoid interference. Even brief or weak transmissions can mask or distort astronomical signals, making it challenging to study the universe with the required precision. For communication systems, interference having similar modulation and power becomes difficult to remove if the source and parameters of the said interference are unknown. This thesis focuses on creating a system that can detect when interference is present and then remove it as effectively as possible. To do this, the work combines ideas from traditional radio engineering with modern deep learning advancements. The system is built in two stages: 1. Detecting whether interference is present in the received signal stream 2. Using deep learning models to estimate and remove the interference A recommender system is also developed for choosing the best system to mitigate the interference given the parameters and scenario. CNN Classifiers are developed for categorizing the noise and interference for effective mitigation. While this research was inspired by the challenges facing radio astronomy, the techniques developed are also useful for many other situations where important signals are potentially masked in noisy, crowded parts of the radio spectrum."],"dc:description.degree":["Master of Science"],"dc:format.medium":["ETD"],"dc:identifier.other":["vt_gsexam:44919"],"dc:identifier.uri":["https://hdl.handle.net/10919/139855"],"dc:language.iso":["en"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["Radio Frequency Interference","Interference Mitigation","Autoencoders","Signal Detection","U-Net"],"dc:title":["On the Use of Deep Learning Models for Interference Detection and Mitigation"],"dc:type":["Thesis"],"thesis:degree_discipline":["Computer Engineering"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:19:35Z"}