Does the online card payment system unwittingly facilitate fraud?

Abstract

The research work in this PhD thesis presents an extensive investigation into the security settings of Card Not Present (CNP) financial transactions. These are the transactions which include payments performed with a card over the Internet on the websites, and over the phone. Our detailed analysis on hundreds of websites and on multiple CNP payment protocols justifies that the current security architecture of CNP payment system is not adequate enough to protect itself from fraud. Unintentionally, the payment system itself will allow an adversary to learn and exploit almost all of the security features put in place to protect the CNP payment system from fraud. With insecure modes of accepting payments, the online payment system paves the way for cybercriminals to abuse even the latest designed payment protocols like 3D Secure 2.0. We follow a structured analysis methodology which identifies vulnerabilities in the CNP payment protocols and demonstrates the impact of these vulnerabilities on the overall payment system. The analysis methodology comprises of UML diagrams and reference tables which describe the CNP payment protocol sequences, software tools which implements the protocol and practical demonstrations of the research results. Detailed referencing of the online payment specifications provides a documented link between the exploitable vulnerabilities observed in real implementations and the source of the vulnerability in the payment specifications. We use practical demonstrations to show that these vulnerabilities can be exploited in the real-world with ease. This presents a stronger impact message when presenting our research results to a nontechnical audience. This has helped to raise awareness of security issues relating to payment cards, with our work appearing in the media, radio and TV