BROWSE
Detail
Conference Paper
5G Core NF의 로그데이터 분석을 위한 sLLM 학습데이터 생성 파이프라인 연구
Cited - time in 
Share 
Share
- Issue Date
- 2026-05
- Citation
- IEMEK Symposium on Embedded Technology (ISET) 2026, pp.7-11
- Publisher
- 대한임베디드공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- In 5G mobile core networks, various Network Functions (NFs) such as AMF, SMF, and UPF generate massive volumes of log data. Efficient analysis of these logs is critical for fault detection and quality assurance. This paper proposes a preprocessing pipeline for fine-tuning small Large Language Models(sLLM), validated against real commercial AMF operational logs. The proposed system adopts a three-layer architecture comprising a Data Ingestion Layer, a Knowledge Base Layer with Retrieval-Augmented Generation based on 3GPP specifications, and an Application & Feedback Layer for continuous dataset curation.
- Keyword
- 5G Core Network, NF Log Analysis, Data Preprocessing, Anomaly Detection