{"id":{"repo_id":"ets-quebec","oai_identifier":"oai:espace.etsmtl.ca:3859"},"canonical_url":"https://search.ndltd.org/etd/ets-quebec/oai:espace.etsmtl.ca:3859","repository":{"repo_id":"ets-quebec","name":"ETS (Quebec)","base_url":"https://espace.etsmtl.ca/cgi/oai2"},"display":{"title":"Performance prediction of URLLC in interference-limited networks using transfer learning","abstract":"In the emerging context of sixth-generation (6G) networks, a major challenge is ensuring extremely high reliability for short-packet transmissions. This challenge is linked to the fact that classical Shannon capacity limits are no longer applicable in the finite-blocklength regime. The present thesis addresses this issue by presenting a comprehensive performance analysis of a URLLC system deployed in a clustered wireless network using short-packet communication (SPC). In this architecture, multiple ground users affected by heterogeneous interference sources are grouped around a cluster head (CH), which acts as a wireless relay between the base station and the users. A closed-form analytical expression for the global block error rate (BLER) is first derived by considering key constraints such as packet size, blocklength, and achievable rate. In addition, a transfer-learning framework is proposed for real-time performance prediction in environments statistically independent, but not necessarily identically distributed (Non-IID). The approach uses a pre-trained source model that is subsequently fine-tuned using domain-specific data, significantly improving generalization capability under diverse interference scenarios. The results highlight the benefits of combining analytical modeling with deep learning for accurate characterization of URLLC performance in the finite-blocklength regime. The proposed method provides an effective solution for dynamic BLER prediction and paves the way for a more reliable deployment of URLLC systems in practical applications requiring instantaneous performance evaluation.","abstract_html":"In the emerging context of sixth-generation (6G) networks, a major challenge is ensuring extremely high reliability for short-packet transmissions. This challenge is linked to the fact that classical Shannon capacity limits are no longer applicable in the finite-blocklength regime. The present thesis addresses this issue by presenting a comprehensive performance analysis of a URLLC system deployed in a clustered wireless network using short-packet communication (SPC). In this architecture, multiple ground users affected by heterogeneous interference sources are grouped around a cluster head (CH), which acts as a wireless relay between the base station and the users. A closed-form analytical expression for the global block error rate (BLER) is first derived by considering key constraints such as packet size, blocklength, and achievable rate. In addition, a transfer-learning framework is proposed for real-time performance prediction in environments statistically independent, but not necessarily identically distributed (Non-IID). The approach uses a pre-trained source model that is subsequently fine-tuned using domain-specific data, significantly improving generalization capability under diverse interference scenarios. The results highlight the benefits of combining analytical modeling with deep learning for accurate characterization of URLLC performance in the finite-blocklength regime. The proposed method provides an effective solution for dynamic BLER prediction and paves the way for a more reliable deployment of URLLC systems in practical applications requiring instantaneous performance evaluation.","abstract_has_math":false,"creators":["Ghous, Mujtaba"],"institution":"École de technologie supérieure","degree_name":"Mémoire de maîtrise électronique","degree_level":"masters","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026-03-04","date_published":"2026-03-04","updated_at":"2026-11-18T22:11:02Z","subjects":["communication ultra-fiable et à faible latence (URLLC), apprentissage par transfert (TL), taux d’erreur de bloc (BLER), réseau neuronal convolutionnel (CNN), analyse de performances, réseau d’interférences non identiquement distribuées (non-IID), prédiction des performances basée sur l’apprentissage automatique"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE3859"],"render_values":[{"text":"INSERT-YOUR-LIBRARY-ID-HERE3859","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null,"record_url":null,"record_label":null,"record_kind":null,"record_source":null,"download_url":"https://espace.etsmtl.ca/id/eprint/3859/1/GHOUS_Mujtaba.pdf","download_label":"Download PDF","download_source":"dc:identifier","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://espace.etsmtl.ca/cgi/oai2?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aespace.etsmtl.ca%3A3859","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Ghous, Mujtaba"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026-03-04"]},{"key":"dc:publisher","label":"Institution","values":["École de technologie supérieure"]},{"key":"dc:type","label":"Dc Type","values":["Electronic Thesis or Dissertation"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Mémoire de maîtrise électronique"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["École de technologie supérieure"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["communication ultra-fiable et à faible latence (URLLC), apprentissage par transfert (TL), taux d’erreur de bloc (BLER), réseau neuronal convolutionnel (CNN), analyse de performances, réseau d’interférences non identiquement distribuées (non-IID), prédiction des performances basée sur l’apprentissage automatique"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE3859","https://espace.etsmtl.ca/id/eprint/3859/1/GHOUS_Mujtaba.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["In the emerging context of sixth-generation (6G) networks, a major challenge is ensuring extremely high reliability for short-packet transmissions. This challenge is linked to the fact that classical Shannon capacity limits are no longer applicable in the finite-blocklength regime. The present thesis addresses this issue by presenting a comprehensive performance analysis of a URLLC system deployed in a clustered wireless network using short-packet communication (SPC). In this architecture, multiple ground users affected by heterogeneous interference sources are grouped around a cluster head (CH), which acts as a wireless relay between the base station and the users. A closed-form analytical expression for the global block error rate (BLER) is first derived by considering key constraints such as packet size, blocklength, and achievable rate. In addition, a transfer-learning framework is proposed for real-time performance prediction in environments statistically independent, but not necessarily identically distributed (Non-IID). The approach uses a pre-trained source model that is subsequently fine-tuned using domain-specific data, significantly improving generalization capability under diverse interference scenarios. The results highlight the benefits of combining analytical modeling with deep learning for accurate characterization of URLLC performance in the finite-blocklength regime. The proposed method provides an effective solution for dynamic BLER prediction and paves the way for a more reliable deployment of URLLC systems in practical applications requiring instantaneous performance evaluation."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Performance prediction of URLLC in interference-limited networks using transfer learning"]}]}],"canonical_facts":{"dc:creator":["Ghous, Mujtaba"],"dc:date":["2026-03-04"],"dc:description":["In the emerging context of sixth-generation (6G) networks, a major challenge is ensuring extremely high reliability for short-packet transmissions. This challenge is linked to the fact that classical Shannon capacity limits are no longer applicable in the finite-blocklength regime. The present thesis addresses this issue by presenting a comprehensive performance analysis of a URLLC system deployed in a clustered wireless network using short-packet communication (SPC). In this architecture, multiple ground users affected by heterogeneous interference sources are grouped around a cluster head (CH), which acts as a wireless relay between the base station and the users. A closed-form analytical expression for the global block error rate (BLER) is first derived by considering key constraints such as packet size, blocklength, and achievable rate. In addition, a transfer-learning framework is proposed for real-time performance prediction in environments statistically independent, but not necessarily identically distributed (Non-IID). The approach uses a pre-trained source model that is subsequently fine-tuned using domain-specific data, significantly improving generalization capability under diverse interference scenarios. The results highlight the benefits of combining analytical modeling with deep learning for accurate characterization of URLLC performance in the finite-blocklength regime. The proposed method provides an effective solution for dynamic BLER prediction and paves the way for a more reliable deployment of URLLC systems in practical applications requiring instantaneous performance evaluation."],"dc:format":["application/pdf"],"dc:identifier":["INSERT-YOUR-LIBRARY-ID-HERE3859","https://espace.etsmtl.ca/id/eprint/3859/1/GHOUS_Mujtaba.pdf"],"dc:language":["en"],"dc:publisher":["École de technologie supérieure"],"dc:subject":["communication ultra-fiable et à faible latence (URLLC), apprentissage par transfert (TL), taux d’erreur de bloc (BLER), réseau neuronal convolutionnel (CNN), analyse de performances, réseau d’interférences non identiquement distribuées (non-IID), prédiction des performances basée sur l’apprentissage automatique"],"dc:title":["Performance prediction of URLLC in interference-limited networks using transfer learning"],"dc:type":["Electronic Thesis or Dissertation"],"thesis:degree_level":["masters"],"thesis:degree_name":["Mémoire de maîtrise électronique"],"thesis:institution_name":["École de technologie supérieure"]},"updated_at":"2026-11-18T22:11:02Z"}