Presentation | 2020-11-26 Evaluation of 5GC Network Analysis Models Using Machine Learning Junichi Kawasaki, Genichi Mouri, Yusuke Suzuki, Tomohiro Otani, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The advances in network technologies such as network function virtualization (NFV) and network slicing enable flexible and quick integration of networks. However, the operation of networks using these new technologies can be challenging due to a greater number of network elements and their more complex composition. It is difficult to maintain a variety of service level agreements (SLAs) in next-generation networks by the conventional manual-based operation. To address this problem, in this paper, we adopt artificial intelligence (AI) which has been applied in many industries in this decade. We propose network analysis models using machine learning (ML) technology, and evaluate the models in the fifth generation (5G) core network. In our approach, the features for building models are derived from the time difference of the performance data collected from each node, and two types of models are created with the training data sets using the original features and those using the combined features. We evaluate the analysis performance of these two models in five failure cases. The experiment on the test network shows that the performance of the models depends on the failure cases. In particular, the analysis model trained with the combined features presents better results for the cases where a failure in one node causes some impacts in other nodes. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Failure Analysis / Root Cause Analysis / Fault Classification / 5G Core / Artificial Intelligence / Machine Leearning |
Paper # | CQ2020-50 |
Date of Issue | 2020-11-19 (CQ) |
Conference Information | |
Committee | NS / ICM / CQ |
---|---|
Conference Date | 2020/11/26(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Virtual Conference |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Network quality, Network measurement/management, Network virtualization, Network service, Blockchain, Security, Network intelligence, etc. |
Chair | Akihiro Nakao(Univ. of Tokyo) / Kazuhiko Kinoshita(Tokushima Univ.) / Hideyuki Shimonishi(NEC) |
Vice Chair | Tetsuya Oishi(NTT) / Yoichi Sato(NEC) / Haruo Ooishi(NTT) / Jun Okamoto(NTT) / Takefumi Hiraguri(Nippon Inst. of Tech.) |
Secretary | Tetsuya Oishi(NTT) / Yoichi Sato(Chuo Univ.) / Haruo Ooishi(NTT) / Jun Okamoto(Bosco) / Takefumi Hiraguri(Doshisha Univ.) |
Assistant | Shinya Kawano(NTT) / Tetsuya Uchiumi(Fujitsu Lab.) / Yoshiaki Nishikawa(NEC) / Takuto Kimura(NTT) / Ryoichi Kataoka(KDDI Research) |
Paper Information | |
Registration To | Technical Committee on Network Systems / Technical Committee on Information and Communication Management / Technical Committee on Communication Quality |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Evaluation of 5GC Network Analysis Models Using Machine Learning |
Sub Title (in English) | |
Keyword(1) | Failure Analysis |
Keyword(2) | Root Cause Analysis |
Keyword(3) | Fault Classification |
Keyword(4) | 5G Core |
Keyword(5) | Artificial Intelligence |
Keyword(6) | Machine Leearning |
1st Author's Name | Junichi Kawasaki |
1st Author's Affiliation | KDDI CORPORATION/KDDI RESEARCH INC.(KDDI/KDDI RESEARCH) |
2nd Author's Name | Genichi Mouri |
2nd Author's Affiliation | KDDI CORPORATION/KDDI RESEARCH INC.(KDDI/KDDI RESEARCH) |
3rd Author's Name | Yusuke Suzuki |
3rd Author's Affiliation | KDDI CORPORATION/KDDI RESEARCH INC.(KDDI/KDDI RESEARCH) |
4th Author's Name | Tomohiro Otani |
4th Author's Affiliation | KDDI CORPORATION/KDDI RESEARCH INC.(KDDI/KDDI RESEARCH) |
Date | 2020-11-26 |
Paper # | CQ2020-50 |
Volume (vol) | vol.120 |
Number (no) | CQ-258 |
Page | pp.pp.16-21(CQ), |
#Pages | 6 |
Date of Issue | 2020-11-19 (CQ) |