{"id":{"repo_id":"auckland-ms","oai_identifier":"oai:researchspace.auckland.ac.nz:2292/74187"},"canonical_url":"https://search.dev.ndltd.org/etd/auckland-ms/oai:researchspace.auckland.ac.nz:2292/74187","repository":{"repo_id":"auckland-ms","name":"University of Auckland","base_url":"https://researchspace.auckland.ac.nz/server/oai/request"},"display":{"title":"Urban Sound and Environmental Monitoring using Distributed IoT Sensor Nodes","abstract":"Urban sound and environmental monitoring are essential in urban planning, noise pollution control, and public safety. By collecting and analysing urban data, cities can make proper decisions to improve the quality of life for their residents. This research proposes a scalable, flexible, and efficient urban sound and environmental monitoring system using distributed Internet of Things (IoT) sensor nodes, including the ability to on-site classify sounds based on their sources. The two key components in the proposed approach are data acquisition and on-site robust sound classification. First, this thesis proposes a novel Wireless Sensor Network (WSN) architecture for urban sound data acquisition, constructed by a multi-tier architecture using cost-effective IoT nodes. It employs various communication protocols for efficient and reliable data transmission. A Jetson Nano was deployed to classify the sounds on-site, thus enabling insight into the acoustic environment. This network can also sample and transmit environmental parameters other than acoustic data. Second, this research focused on enhancing the robustness of urban sound classification in noisy environments. A novel feature extraction method was introduced to represent urban sound in complex scenarios better. A dual-stream classifier was employed to improve decision-making, ensuring accurate source recognition in urban soundscapes, especially in noisy conditions. To solve the latency issue of the proposed classification system, an end-to-end sound classification model was further introduced to enable real-time processing on Jetson Nano. Knowledge and feature distillation techniques were used in training this classifier to learn from the original classification system. A two-stage training strategy was applied to improve the performance of the end-to-end model. The classifier significantly reduced computational overhead while maintaining classification accuracy and robustness. Finally, the data acquisition component and the end-to-end sound classifier were integrated for urban sound and environmental monitoring. The successful real-world deployment validated its functionality, efficiency, and scalability. With superior performance compared to existing technologies, the proposed system emerges as a promising solution for various applications, including urban noise pollution monitoring, smart city infrastructure development, and public safety enhancement.","abstract_html":"Urban sound and environmental monitoring are essential in urban planning, noise pollution control, and public safety. By collecting and analysing urban data, cities can make proper decisions to improve the quality of life for their residents. This research proposes a scalable, flexible, and efficient urban sound and environmental monitoring system using distributed Internet of Things (IoT) sensor nodes, including the ability to on-site classify sounds based on their sources. The two key components in the proposed approach are data acquisition and on-site robust sound classification. First, this thesis proposes a novel Wireless Sensor Network (WSN) architecture for urban sound data acquisition, constructed by a multi-tier architecture using cost-effective IoT nodes. It employs various communication protocols for efficient and reliable data transmission. A Jetson Nano was deployed to classify the sounds on-site, thus enabling insight into the acoustic environment. This network can also sample and transmit environmental parameters other than acoustic data. Second, this research focused on enhancing the robustness of urban sound classification in noisy environments. A novel feature extraction method was introduced to represent urban sound in complex scenarios better. A dual-stream classifier was employed to improve decision-making, ensuring accurate source recognition in urban soundscapes, especially in noisy conditions. To solve the latency issue of the proposed classification system, an end-to-end sound classification model was further introduced to enable real-time processing on Jetson Nano. Knowledge and feature distillation techniques were used in training this classifier to learn from the original classification system. A two-stage training strategy was applied to improve the performance of the end-to-end model. The classifier significantly reduced computational overhead while maintaining classification accuracy and robustness. Finally, the data acquisition component and the end-to-end sound classifier were integrated for urban sound and environmental monitoring. The successful real-world deployment validated its functionality, efficiency, and scalability. With superior performance compared to existing technologies, the proposed system emerges as a promising solution for various applications, including urban noise pollution monitoring, smart city infrastructure development, and public safety enhancement.","abstract_has_math":false,"creators":["Peng, Bo"],"institution":"ResearchSpace@Auckland","degree_name":"PhD","degree_level":"Doctoral","degree_discipline":"Electrical and Electronic Engineering","degree_department":null,"school":null,"contributors":[],"advisors":["Abdulla, Waleed H","Wang, Kevin I-Kai"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-07-24T01:03:37Z","subjects":[],"languages":[],"rights":["Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated."],"rights_urls":["https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2292/74187","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Abdulla, Waleed H","Wang, Kevin I-Kai"]},{"key":"dc:creator","label":"Author","values":["Peng, Bo"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-12-03T01:14:42Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:publisher","label":"Institution","values":["ResearchSpace@Auckland"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical and Electronic Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["PhD"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["The University of Auckland"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated."]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2292/74187"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Urban sound and environmental monitoring are essential in urban planning, noise pollution control, and public safety. By collecting and analysing urban data, cities can make proper decisions to improve the quality of life for their residents. This research proposes a scalable, flexible, and efficient urban sound and environmental monitoring system using distributed Internet of Things (IoT) sensor nodes, including the ability to on-site classify sounds based on their sources. The two key components in the proposed approach are data acquisition and on-site robust sound classification. First, this thesis proposes a novel Wireless Sensor Network (WSN) architecture for urban sound data acquisition, constructed by a multi-tier architecture using cost-effective IoT nodes. It employs various communication protocols for efficient and reliable data transmission. A Jetson Nano was deployed to classify the sounds on-site, thus enabling insight into the acoustic environment. This network can also sample and transmit environmental parameters other than acoustic data. Second, this research focused on enhancing the robustness of urban sound classification in noisy environments. A novel feature extraction method was introduced to represent urban sound in complex scenarios better. A dual-stream classifier was employed to improve decision-making, ensuring accurate source recognition in urban soundscapes, especially in noisy conditions. To solve the latency issue of the proposed classification system, an end-to-end sound classification model was further introduced to enable real-time processing on Jetson Nano. Knowledge and feature distillation techniques were used in training this classifier to learn from the original classification system. A two-stage training strategy was applied to improve the performance of the end-to-end model. The classifier significantly reduced computational overhead while maintaining classification accuracy and robustness. Finally, the data acquisition component and the end-to-end sound classifier were integrated for urban sound and environmental monitoring. The successful real-world deployment validated its functionality, efficiency, and scalability. With superior performance compared to existing technologies, the proposed system emerges as a promising solution for various applications, including urban noise pollution monitoring, smart city infrastructure development, and public safety enhancement."]},{"key":"dc:title","label":"Title","values":["Urban Sound and Environmental Monitoring using Distributed IoT Sensor Nodes"]}]}],"canonical_facts":{"dc:contributor.advisor":["Abdulla, Waleed H","Wang, Kevin I-Kai"],"dc:creator":["Peng, Bo"],"dc:date.accessioned":["2025-12-03T01:14:42Z"],"dc:date.issued":["2025"],"dc:description.abstract":["Urban sound and environmental monitoring are essential in urban planning, noise pollution control, and public safety. By collecting and analysing urban data, cities can make proper decisions to improve the quality of life for their residents. This research proposes a scalable, flexible, and efficient urban sound and environmental monitoring system using distributed Internet of Things (IoT) sensor nodes, including the ability to on-site classify sounds based on their sources. The two key components in the proposed approach are data acquisition and on-site robust sound classification. First, this thesis proposes a novel Wireless Sensor Network (WSN) architecture for urban sound data acquisition, constructed by a multi-tier architecture using cost-effective IoT nodes. It employs various communication protocols for efficient and reliable data transmission. A Jetson Nano was deployed to classify the sounds on-site, thus enabling insight into the acoustic environment. This network can also sample and transmit environmental parameters other than acoustic data. Second, this research focused on enhancing the robustness of urban sound classification in noisy environments. A novel feature extraction method was introduced to represent urban sound in complex scenarios better. A dual-stream classifier was employed to improve decision-making, ensuring accurate source recognition in urban soundscapes, especially in noisy conditions. To solve the latency issue of the proposed classification system, an end-to-end sound classification model was further introduced to enable real-time processing on Jetson Nano. Knowledge and feature distillation techniques were used in training this classifier to learn from the original classification system. A two-stage training strategy was applied to improve the performance of the end-to-end model. The classifier significantly reduced computational overhead while maintaining classification accuracy and robustness. Finally, the data acquisition component and the end-to-end sound classifier were integrated for urban sound and environmental monitoring. The successful real-world deployment validated its functionality, efficiency, and scalability. With superior performance compared to existing technologies, the proposed system emerges as a promising solution for various applications, including urban noise pollution monitoring, smart city infrastructure development, and public safety enhancement."],"dc:identifier.uri":["https://hdl.handle.net/2292/74187"],"dc:publisher":["ResearchSpace@Auckland"],"dc:rights":["Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated."],"dc:rights.uri":["https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm"],"dc:title":["Urban Sound and Environmental Monitoring using Distributed IoT Sensor Nodes"],"dc:type":["Thesis"],"thesis:degree_discipline":["Electrical and Electronic Engineering"],"thesis:degree_level":["Doctoral"],"thesis:degree_name":["PhD"],"thesis:institution_name":["The University of Auckland"]},"updated_at":"2026-07-24T01:03:37Z"}