Designing a learning solution to detect anomalies in telecommunications networks

Taking a look back at the success stories of 2022, LIST developed a semi-supervised solution based on deep learning methods to enable the detection of anomalies in the central telecommunications network.

"The results of the Secure5GeXP project have been welcomed by the Ministry of Economy, which supports the project. As a result, the Ministry has endorsed an extension of the current project and has encouraged a follow-up project between POST and LIST." Qiang TANG, Leader of the Trustworthy Data Systems Group

In close collaboration with POST within the framework of the Secure5GeXP project, LIST developed a semi-supervised solution based on deep learning methods to enable the detection of anomalies in the central telecommunications network.

The solution was validated with private data from POST as well as with simulated data created by LIST on the basis of an open-source simulator. POST continues to validate the solution before it is put into production.

The flagship field of research and innovation combining machine learning and cybersecurity was also boosted by the launch of the Horizon Europe LAZARUS project. It aims to solve many of the security problems faced by modern software during its development cycle.

Discover more success stories in the digital version of the LIST 2022 annual report.