Performance analysis of probabilistic packet marking in IPv6

Probabilistic packet marking (PPM) has received considerable attention as an IP traceback approach against distributed Denial-of-Service attack, which is one of the most challenging security threat in the Internet. PPM is a technique that seeks to identify the source of such attacks by marking individual packets with portion of the attack path, and then relies on the volume of attack traffic generated to ensure that the whole path can be reconstructed. However, modifying the identification field in the IPv4 packet header to mark packet incurs backward incompatibility for IP fragmented packets. In this paper, we address this issue and analyze the viability of PPM under the next-generation Internet Protocol, IPv6. In doing so, we consider the flaws inherent to IPv4 implementations that limit their backward compatibility, and demonstrate how these shortcomings can be avoided in IPv6. We show that the Flow Label field in the IPv6 datagram header can be safely and effectively overloaded to implement PPM schemes, and present simulation results verifying the applicability and efficiency of this approach.