2022 Japan challenge (network failure prediction & location estimation)
* Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay
This webinar, will introduce 2 problem statements for the 2022 ITU AI/ML in 5G challenge hosted in Japan as well as expert talks;
ML5G-PS-005: Network failure prediction on CNFs 5GC with Linux eBPF
(https://challenge.aiforgood.itu.int/match/matchitem/64)
In this challenge, participants will be asked to predict future network failures on the 5GC using AI/ML. This challenge will provide multi-variant time-series data obtained by eBPF and cAdvisor that provides basic metrics for each 5GC CNF under normal and failure conditions. To realize network failure prediction, AI/ML technology is expected to be leveraged. Participants are asked to challenge how early and accurately the future network failures can be predicted using time-series data consisting of thousands of metrics provided by eBPF and cAdvisor. The target value for prediction is the number of UE registration failures in the 5GC.
ML5G-PS-006: Location Estimation Using RSSI of Wireless LAN in NLoS Environment (https://challenge.aiforgood.itu.int/match/matchitem/65)
This challenge explores the possibility that the data-oriented localization technique can replace the model-based localization with the help of powerful AI/ML techniques. Ultimately, this challenge tries to tackle the limitation of AI/ML-based localization using RSS information; Can AI/ML-based localization technique achieve similar accuracy as the GPS-based location technique or even better accuracy?
2 Expert talks: Machine Learning for Wireless LANs
The two expert talk provides applications of machine and deep learning techniques to wireless communication networks.
