Native AI
- Develop an AI-native network prototype capable of collecting simulated telemetry data from multiple edge devices, detecting anomalies or congestion using AI/ML, and autonomously taking corrective actions such as traffic rerouting, node shutdown, or resource reallocation.
- Design an adaptive network control system that continuously learns from operational data to enhance performance and resilience.
Goal:
Enable self-learning, self-correcting, and energy-efficient networks, reducing human intervention and optimizing real-time decision-making in 6G systems.
AI/ML Compute Acceleration for Edge Networks
- Build a prototype system that accelerates AI/ML workloads at the network edge using hardware accelerators such as GPUs, FPGAs, DPUs, or co-processors.
- Demonstrate real-time AI acceleration for edge applications like autonomous drones, AR/VR systems, and intelligent traffic management.
Goal:
Achieve ultra-fast and energy-efficient AI computation for distributed networks, showcasing how hardware-software co-design can enable smarter and more responsive 6G systems.
Integrated Sensing and Communication (ISAC)
- Develop a prototype demonstrating simultaneous communication and sensing, using simulated signals to detect motion, obstacles, or environmental changes while transmitting data.
- Explore multi-node coordination to enhance sensing accuracy and real-time situational awareness.
Goal:
Show how 6G networks can integrate perception and communication, creating systems that see, sense, and respond to their surroundings—enabling breakthroughs in connected mobility, smart manufacturing, and defense technologies.