Back to Search Start Over

RAPT: A Robust Attack Path Tracing Algorithm to Mitigate SYN-Flood DDoS Cyberattacks.

Authors :
AlArnaout, Zakwan
Mostafa, Nour
Alabed, Samer
Aly, Wael Hosny Fouad
Shdefat, Ahmed
Source :
Sensors (14248220). Jan2023, Vol. 23 Issue 1, p102. 20p.
Publication Year :
2023

Abstract

In the recent past, Distributed Denial of Service (DDoS) attacks have become more abundant and present one of the most serious security threats. In a DDoS attack, the attacker controls a botnet of daemons residing in vulnerable hosts that send a significant amount of traffic to flood the victim or the network infrastructure. In this paper, a common type of DDoS attacks known as "TCP SYN-Flood" is studied. This type of attack uses spoofed Internet Protocol (IP) addresses for SYN packets by exploiting the weakness in Transmission Control Protocol (TCP) 3-Way handshake used by the TCP/IP suite of protocols, which make the web servers unreachable for legitimate users or even worse, it might lead to server crash. In this paper, a resilient, efficient, lightweight, and robust IP traceback algorithm is proposed using an IP tracing packet for each attack path. The proposed algorithm suggests that edge routers—where the attack starts from—observe the traffic pattern passing through, and if the observed traffic carries the signature of TCP SYN-Flood DDoS attack and a high percentage of it is destined to a particular web server(s), it starts the tracing process by generating an IP trace packet, which accompanies the attack path recording the routers' IP addresses on the path between the attacker/daemon and the victim, which can extract the path and react properly upon receiving it by discarding any SYN packets originating from that attacker/daemon. To our knowledge, this is the first research that efficiently traces these kinds of attacks while they are running. The proposed solution has low computation and message overhead, efficient detection and tracing time, and converges in near optimal time. The results are validated using extensive simulation runs. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
14248220
Volume :
23
Issue :
1
Database :
Academic Search Index
Journal :
Sensors (14248220)
Publication Type :
Academic Journal
Accession number :
161185813
Full Text :
https://doi.org/10.3390/s23010102