AI4Emergency – AI-Optimized Operational Communication
The development and testing of AI-supported communication technologies based on 5G mobile networks will provide emergency services with a powerful and reliable means of communication.
Due to climate change, disaster scenarios have become more frequent in recent years. Rapid situation assessment and reliable communication are crucial for effective decision-making in such situations. Although digital broadband communication networks enable rapid information transfer, the communication infrastructure can quickly become overloaded in the event of partial failure or in less developed areas. Networks that are specifically optimized and prepared for such scenarios can significantly increase resilience.
The AI4Emergency project aims to develop and test AI-supported communication technologies for use in safety-critical operations. The focus is on operational scenarios such as large-scale and high-performance communication, which is essential. With the help of gigabit-capable 5G mobile communication technologies and intelligent network configurations, innovative AI applications such as drone control, voice interaction with chatbots, and decision support systems are to be integrated, with the aim of reducing information overload for decision-makers.
This is achieved by modelling real-world deployment processes and translating them into technical load models, which makes data-intensive services such as video transmission or AI-based command output more controllable. Simulations and AI-supported optimization approaches are employed to dynamically manage network resources, for example, through network slicing, guaranteed bitrate configurations, or MCX communication. These results are then tested in realistic laboratory experiments and validation scenarios to create practical, scalable solutions.
Salzburg Research is responsible for the technical analysis of the network requirements that arise from AI-supported applications in safety-critical scenarios. Based on the developed application processes, Salzburg Research creates structured load models that map typical communication patterns, such as voice, video, and traffic data, as well as AI services. The relevant network parameters are derived from these process models and synthetic application data.
The program “Broadband Austria 2030: GigaApp” is funded by the Austrian Research Promotion Agency (FFG) and the Federal Ministry for Housing, Arts, Culture, Media, and Sport (BMKMS).
