Analyzing Internet Background Radiation with Reflective Network Telescopes


연구 분야: Safety



학회: ANRW '25: Proceedings of the 2025 Applied Networking Research Workshop


초록

Traditional darknets rely on unused address space to capture Internet Background Radiation (IBR), but this approach is becoming less viable as IPv4 space exhausts and attackers evade known darknets. This paper explores leveraging routine ICMP error traffic to recover IBR in operational networks using reflective network telescopes, without reserving address space or inspecting user traffic. We deployed two reflective network telescopes that passively record only ICMP Type 3 and Type 11 messages at a transit ISP PoP for 30 days, and captured 900 GB of data containing 14.9M probe packets. Analysis of the extracted payloads revealed 122,730 Internet scanners and 58,333 probable victims of randomly spoofed DoS attacks. Our results show that ICMP Type 3 errors could expose high-volume UDP scanning, while ICMP Type 11 errors provided balanced protocol coverage. As ICMP error traffic is much lower in volume than user traffic, this method offers a scalable and privacy-preserving approach to gathering threat intelligence.


Author Profile
Echo Chan

Akamai Technologies & The Hong Kong Polytechnic University Hong Kong SAR China

China
Author Profile
Ricky K. P. Mok

CAIDA University of California San Diego La Jolla CA USA

Canada
Author Profile
Edmond W Chan

Akamai Technologies Cambridge MA USA

Morocco

📄 논문 정보

발행 연도 2025년
인용수 0
출판 국가 Morocco, China, Canada
사이트 ACM
좋아요 수 0

연관 논문 목록 (124건)