DDOS Attack Identification Using Photonic Deep Learning Technology

Researchers have demonstrated a method for identifying distributed denial of service (DDoS) attacks using a silicon photonic deep neural network. This system operates at a frequency of 50 GHz and is capable of updating both input signals and weights in real-time. The study reports a Cohen’s κ-score of 0.636, indicating a moderate level of accuracy in attack identification. The technology aims to enhance the detection capabilities of DDoS attacks, which pose significant risks to network infrastructure. The research was presented at the Photonic Processing for Computing and ML session, highlighting advancements in machine learning applications for cybersecurity. This innovative approach could potentially improve response times and accuracy in identifying cyber threats. The findings are relevant for organizations looking to bolster their defenses against DDoS attacks.

Key Points: • A silicon photonic deep neural network was used for DDoS attack identification. • The system operates at 50 GHz and achieved a Cohen’s κ-score of 0.636. • This research enhances detection capabilities for DDoS attacks, crucial for network security.