{"id":{"repo_id":"clemson","oai_identifier":"oai:open.clemson.edu:all_theses-4085"},"canonical_url":"https://search.ndltd.org/etd/clemson/oai:open.clemson.edu:all_theses-4085","repository":{"repo_id":"clemson","name":"Clemson University","base_url":"https://open.clemson.edu/do/oai/"},"display":{"title":"Improvements to MLE Algorithm for Localizing Radiation Sources with a Distributed Detector Network","abstract":"Maximum Likelihood Estimation (MLE) is a widely used method for the localization of radiation sources using distributed detector networks. While robust, MLE is computationally intensive, requiring an exhaustive search over parameter space. To mitigate the computational load of MLE, many techniques have been presented, including iterative and multi-resolution methods. In this work, we present two ways to improve the MLE localization of radiation sources. First, we present a method to mitigate the pitfalls of a standard multi-resolution algorithm. Our method expands the search region of each layer before performing the MLE search. Doing so allows the multi-resolution algorithm to correct an incorrect selection made in a prior layer. We test our proposed method against single-resolution MLE and standard multi-resolution MLE algorithms, and find that the use of grid expansion incurs a general decrease in localization error and a negligible increase in computation time over the standard multi-resolution algorithm. Second, we present a method to perform the MLE localization without prior knowledge of the background radiation intensity. We estimate the source and background intensities using linear regression (LR) and then use these estimates to initialize the intensity parameter search for MLE. We test this method using single-resolution, multi-resolution, and multi-resolution with grid expansion MLE algorithms and compare performance to MLE algorithms that don't use the LR initialization method. We found that using the LR estimates to initialize the intensity parameter search caused a marginal increase in both localization error and computation time for the tested algorithms. The technique is only beneficial in the case of an unknown background intensity.","abstract_html":"Maximum Likelihood Estimation (MLE) is a widely used method for the localization of radiation sources using distributed detector networks. While robust, MLE is computationally intensive, requiring an exhaustive search over parameter space. To mitigate the computational load of MLE, many techniques have been presented, including iterative and multi-resolution methods. In this work, we present two ways to improve the MLE localization of radiation sources. First, we present a method to mitigate the pitfalls of a standard multi-resolution algorithm. Our method expands the search region of each layer before performing the MLE search. Doing so allows the multi-resolution algorithm to correct an incorrect selection made in a prior layer. We test our proposed method against single-resolution MLE and standard multi-resolution MLE algorithms, and find that the use of grid expansion incurs a general decrease in localization error and a negligible increase in computation time over the standard multi-resolution algorithm. Second, we present a method to perform the MLE localization without prior knowledge of the background radiation intensity. We estimate the source and background intensities using linear regression (LR) and then use these estimates to initialize the intensity parameter search for MLE. We test this method using single-resolution, multi-resolution, and multi-resolution with grid expansion MLE algorithms and compare performance to MLE algorithms that don&#x27;t use the LR initialization method. We found that using the LR estimates to initialize the intensity parameter search caused a marginal increase in both localization error and computation time for the tested algorithms. The technique is only beneficial in the case of an unknown background intensity.","abstract_has_math":false,"creators":["Cordone, Guthrie"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":"Thesis","degree_discipline":"Electrical and Computer Engineering (Holcomb Dept. of)","degree_department":null,"school":null,"contributors":["Richard Brooks, Committee Chair","Harlan Russell","Adam Hoover","Satyabrata Sen"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:20:11Z","subjects":["distributed sensor network","localization","maximum likelihood estimation","optimization","radiation","real data"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://open.clemson.edu/all_theses/3078","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://open.clemson.edu/all_theses/3078","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.clemson.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aopen.clemson.edu%3Aall_theses-4085","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Richard Brooks, Committee Chair","Harlan Russell","Adam Hoover","Satyabrata Sen"]},{"key":"dc:creator","label":"Author","values":["Cordone, Guthrie"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2019-05-20T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2019-05-01T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical and Computer Engineering (Holcomb Dept. of)"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["distributed sensor network","localization","maximum likelihood estimation","optimization","radiation","real data"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://open.clemson.edu/all_theses/3078"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Maximum Likelihood Estimation (MLE) is a widely used method for the localization of radiation sources using distributed detector networks. While robust, MLE is computationally intensive, requiring an exhaustive search over parameter space. To mitigate the computational load of MLE, many techniques have been presented, including iterative and multi-resolution methods. In this work, we present two ways to improve the MLE localization of radiation sources. First, we present a method to mitigate the pitfalls of a standard multi-resolution algorithm. Our method expands the search region of each layer before performing the MLE search. Doing so allows the multi-resolution algorithm to correct an incorrect selection made in a prior layer. We test our proposed method against single-resolution MLE and standard multi-resolution MLE algorithms, and find that the use of grid expansion incurs a general decrease in localization error and a negligible increase in computation time over the standard multi-resolution algorithm. Second, we present a method to perform the MLE localization without prior knowledge of the background radiation intensity. We estimate the source and background intensities using linear regression (LR) and then use these estimates to initialize the intensity parameter search for MLE. We test this method using single-resolution, multi-resolution, and multi-resolution with grid expansion MLE algorithms and compare performance to MLE algorithms that don't use the LR initialization method. We found that using the LR estimates to initialize the intensity parameter search caused a marginal increase in both localization error and computation time for the tested algorithms. The technique is only beneficial in the case of an unknown background intensity."]},{"key":"dc:title","label":"Title","values":["Improvements to MLE Algorithm for Localizing Radiation Sources with a Distributed Detector Network"]}]}],"canonical_facts":{"dc:contributor":["Richard Brooks, Committee Chair","Harlan Russell","Adam Hoover","Satyabrata Sen"],"dc:creator":["Cordone, Guthrie"],"dc:date.available":["2019-05-20T07:00:00Z"],"dc:date.created":["2019-05-01T07:00:00Z"],"dc:description.abstract":["Maximum Likelihood Estimation (MLE) is a widely used method for the localization of radiation sources using distributed detector networks. While robust, MLE is computationally intensive, requiring an exhaustive search over parameter space. To mitigate the computational load of MLE, many techniques have been presented, including iterative and multi-resolution methods. In this work, we present two ways to improve the MLE localization of radiation sources. First, we present a method to mitigate the pitfalls of a standard multi-resolution algorithm. Our method expands the search region of each layer before performing the MLE search. Doing so allows the multi-resolution algorithm to correct an incorrect selection made in a prior layer. We test our proposed method against single-resolution MLE and standard multi-resolution MLE algorithms, and find that the use of grid expansion incurs a general decrease in localization error and a negligible increase in computation time over the standard multi-resolution algorithm. Second, we present a method to perform the MLE localization without prior knowledge of the background radiation intensity. We estimate the source and background intensities using linear regression (LR) and then use these estimates to initialize the intensity parameter search for MLE. We test this method using single-resolution, multi-resolution, and multi-resolution with grid expansion MLE algorithms and compare performance to MLE algorithms that don't use the LR initialization method. We found that using the LR estimates to initialize the intensity parameter search caused a marginal increase in both localization error and computation time for the tested algorithms. The technique is only beneficial in the case of an unknown background intensity."],"dc:identifier":["https://open.clemson.edu/all_theses/3078"],"dc:subject":["distributed sensor network","localization","maximum likelihood estimation","optimization","radiation","real data"],"dc:title":["Improvements to MLE Algorithm for Localizing Radiation Sources with a Distributed Detector Network"],"thesis:degree_discipline":["Electrical and Computer Engineering (Holcomb Dept. of)"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-11-19T07:20:11Z"}