AI-NET-PROTECT, with a total volume of almost 24 million euros, is embedded in the European research initiative CELTIC-NEXT and is funded in Germany by the Federal Ministry of Education and Research (BMBF). The main focus of the joint project is the security and resilience of critical infrastructure networks. These include companies and institutions in the health and security sectors, energy producers and telecommunications service providers. The aim is to ensure the protection of critical data, network performance (such as latency, throughput, availability) and infrastructure (against manipulation and attacks).
Usability and security through AI-supported analysis of network data
The consistec team has focused on expanding its machine learning expertise in the area of network monitoring in broadband networks with high data rates. Dr. Patrick Michel, Senior Researcher, Alexander Prange, Senior Machine Learning Expert and Dr.-Ing. Thomas Sinnwell, CEO R&D are working on creating interfaces (APIs) and architectures with which AI-based 'state-of-the-art' applications and modules can be integrated into consistec's own monitoring system caplon©. They use machine learning models that detect attacks and anomalies in networks, trigger these events and raise alarms. Once the event has been diagnosed, caplon© takes over the detailed downstream examination and initiates measures.
A resilient foundation for evaluating the research results in partner labs are in place. Now the field trial research begins: AI applications are integrated at selected partner locations and the results are analysed. "Putting the research results to good use 'in the real world' is a priority for us," explains Alexander Prange. "Thanks to the enormous number of cooperative project partners and research institutes, we can test our approaches in a variety of different networks and make data available to partners in real time. This creates a win-win situation for everyone involved," Prange continues. consistec will use the remaining research period to realize and document concrete applications.
Further information on the AI-NET-PROTECT project can be found here: https://protect.ai-net.tech/.