{"id":{"repo_id":"helsinki","oai_identifier":"oai:helda.helsinki.fi:10138/598119"},"canonical_url":"https://search.dev.ndltd.org/etd/helsinki/oai:helda.helsinki.fi:10138/598119","repository":{"repo_id":"helsinki","name":"University of Helsinki","base_url":"https://helda.helsinki.fi/server/oai/request"},"display":{"title":"Knowledge Distillation in DMRS CHEST ML to Optimize Radio Performance and Hardware Efficiency","abstract":"The rapid increase in data traffic, diverse device connections, and strict latency requirements in mobile communication systems have created high demands on the radio access network (RAN), especially at the physical layer. In 5G New Radio (NR) systems, good channel estimation is crucial for demodulating uplink signals like those sent over the Physical Uplink Shared Channel (PUSCH). Demodulation Reference Signals (DMRS) help channel estimation but are often sparse, needing smart interpolation to accurately recreate full channel matrices in challenging conditions. This thesis explores using knowledge distillation (KD) to improve machine learning-based channel estimation in 5G systems. A convolutional neural network (CNN) acts as a teacher model, trained on synthetic data, to create a lightweight student CNN through response-based and feature-based KD techniques. The distillation is formulated as a supervised regression task to reduce losses. The thesis evaluates the effectiveness of different distillation strategies across various operational scenarios relevant to 5G deployments, from ideal conditions to challenging cell-edge environments. It also explores the impact of hyperparameter selection on distillation quality. It investigates how different channel characteristics influence the knowledge transfer process while assessing computational efficiency through training duration measurements to provide a holistic view of each method's practical feasibility. The results demonstrate that KD enables the student model to achieve performance comparable to the teacher, while significantly reducing resource requirements, thereby making real-time deployment on edge hardware in 5G base stations feasible.","abstract_html":"The rapid increase in data traffic, diverse device connections, and strict latency requirements in mobile communication systems have created high demands on the radio access network (RAN), especially at the physical layer. In 5G New Radio (NR) systems, good channel estimation is crucial for demodulating uplink signals like those sent over the Physical Uplink Shared Channel (PUSCH). Demodulation Reference Signals (DMRS) help channel estimation but are often sparse, needing smart interpolation to accurately recreate full channel matrices in challenging conditions. This thesis explores using knowledge distillation (KD) to improve machine learning-based channel estimation in 5G systems. A convolutional neural network (CNN) acts as a teacher model, trained on synthetic data, to create a lightweight student CNN through response-based and feature-based KD techniques. The distillation is formulated as a supervised regression task to reduce losses. The thesis evaluates the effectiveness of different distillation strategies across various operational scenarios relevant to 5G deployments, from ideal conditions to challenging cell-edge environments. It also explores the impact of hyperparameter selection on distillation quality. It investigates how different channel characteristics influence the knowledge transfer process while assessing computational efficiency through training duration measurements to provide a holistic view of each method's practical feasibility. The results demonstrate that KD enables the student model to achieve performance comparable to the teacher, while significantly reducing resource requirements, thereby making real-time deployment on edge hardware in 5G base stations feasible.","abstract_has_math":false,"creators":["Mann, Shaiza"],"institution":"Helsingin yliopisto","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-06-26","date_published":"2025-06-26","updated_at":"2026-07-27T19:55:53Z","subjects":["mobile communication networks","wireless networks","signal processing","machine learning","radio technology","knowledge distillation","5G new radio","neural networks","channel estimation","hardware efficiency"],"languages":["eng"],"rights":["In Copyright 1.0"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/10138/598119","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Mann, Shaiza"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-06-26T06:53:15Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-06-26T06:53:15Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-06-26"]},{"key":"dc:publisher","label":"Institution","values":["Helsingin yliopisto","University of Helsinki","Helsingfors universitet"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["mobile communication networks","wireless networks","signal processing","machine learning","radio technology","knowledge distillation","5G new radio","neural networks","channel estimation","hardware efficiency"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright 1.0"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/10138/598119"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The rapid increase in data traffic, diverse device connections, and strict latency requirements in mobile communication systems have created high demands on the radio access network (RAN), especially at the physical layer. In 5G New Radio (NR) systems, good channel estimation is crucial for demodulating uplink signals like those sent over the Physical Uplink Shared Channel (PUSCH). Demodulation Reference Signals (DMRS) help channel estimation but are often sparse, needing smart interpolation to accurately recreate full channel matrices in challenging conditions. This thesis explores using knowledge distillation (KD) to improve machine learning-based channel estimation in 5G systems. A convolutional neural network (CNN) acts as a teacher model, trained on synthetic data, to create a lightweight student CNN through response-based and feature-based KD techniques. The distillation is formulated as a supervised regression task to reduce losses. The thesis evaluates the effectiveness of different distillation strategies across various operational scenarios relevant to 5G deployments, from ideal conditions to challenging cell-edge environments. It also explores the impact of hyperparameter selection on distillation quality. It investigates how different channel characteristics influence the knowledge transfer process while assessing computational efficiency through training duration measurements to provide a holistic view of each method's practical feasibility. The results demonstrate that KD enables the student model to achieve performance comparable to the teacher, while significantly reducing resource requirements, thereby making real-time deployment on edge hardware in 5G base stations feasible."]},{"key":"dc:title","label":"Title","values":["Knowledge Distillation in DMRS CHEST ML to Optimize Radio Performance and Hardware Efficiency"]}]}],"canonical_facts":{"dc:creator":["Mann, Shaiza"],"dc:date.accessioned":["2025-06-26T06:53:15Z"],"dc:date.available":["2025-06-26T06:53:15Z"],"dc:date.issued":["2025-06-26"],"dc:description.abstract":["The rapid increase in data traffic, diverse device connections, and strict latency requirements in mobile communication systems have created high demands on the radio access network (RAN), especially at the physical layer. In 5G New Radio (NR) systems, good channel estimation is crucial for demodulating uplink signals like those sent over the Physical Uplink Shared Channel (PUSCH). Demodulation Reference Signals (DMRS) help channel estimation but are often sparse, needing smart interpolation to accurately recreate full channel matrices in challenging conditions. This thesis explores using knowledge distillation (KD) to improve machine learning-based channel estimation in 5G systems. A convolutional neural network (CNN) acts as a teacher model, trained on synthetic data, to create a lightweight student CNN through response-based and feature-based KD techniques. The distillation is formulated as a supervised regression task to reduce losses. The thesis evaluates the effectiveness of different distillation strategies across various operational scenarios relevant to 5G deployments, from ideal conditions to challenging cell-edge environments. It also explores the impact of hyperparameter selection on distillation quality. It investigates how different channel characteristics influence the knowledge transfer process while assessing computational efficiency through training duration measurements to provide a holistic view of each method's practical feasibility. The results demonstrate that KD enables the student model to achieve performance comparable to the teacher, while significantly reducing resource requirements, thereby making real-time deployment on edge hardware in 5G base stations feasible."],"dc:identifier.uri":["http://hdl.handle.net/10138/598119"],"dc:language.iso":["eng"],"dc:publisher":["Helsingin yliopisto","University of Helsinki","Helsingfors universitet"],"dc:rights":["In Copyright 1.0"],"dc:subject":["mobile communication networks","wireless networks","signal processing","machine learning","radio technology","knowledge distillation","5G new radio","neural networks","channel estimation","hardware efficiency"],"dc:title":["Knowledge Distillation in DMRS CHEST ML to Optimize Radio Performance and Hardware Efficiency"]},"updated_at":"2026-07-27T19:55:53Z"}