As 6G networks approach, integrating AI for ubiquitous connected intelligence is paramount. Our paper, “Photonic-accelerated AI for cybersecurity in sustainable 6G networks,” details a photonic-based CNN solution for real-time DoS attack detection, achieving a 99.73% mean F1-score with 4-bit resolution.
We address the computational burden of AI models by leveraging photonic hardware, offering significant advantages in energy efficiency and processing speed. This work highlights how our PANN training strategy overcomes the limitations of analog photonic hardware, ensuring high accuracy even with low bit resolution (e.g., 4-bit).
Explore the future of intelligent and sustainable network security with CLEVER!