The Domain Name System (DNS) is a very valuable component in the Internet infrastructure as it provides a mapping between domain names and IP addresses for nearly all major Internet applications. Unfortunately, DNS was not designed with security in mind and has numerous vulnerabilities. Insecurity of the proxy servers and the inherent hierarchical structure of DNS makes it susceptible to the poisoning attacks. Contemporary techniques for securing DNS operations are predominantly based on cryptographic solutions. In this project, the Principal Investigator (PI) is examining cooperative techniques for enhancing security of DNS as an alternative to cryptographically secured communications as in DNSSEC. Ongoing efforts of the PI on the Domain Name Cross-referencing (DoX) have validated the potential benefits of the cooperation-based security approach. Leveraging on this effort, he is focusing on the development of an enhanced framework for guarding against the poisoning attacks on DNS. The research covers a variety of areas including a comprehensive measurement based evaluation of cooperative security, selection and maintenance of peer-groups used for cooperative security, definition and evaluation of DNS quality of service metrics through analytical modeling, and selective use of cryptographic techniques to make cooperative security mechanisms more robust. These efforts will help in developing techniques to reduce security threats such as phishing, pharming, and denial of service attacks on the DNS. In the context of broader impact, the proposed research will have a very positive impact on the Internet community in general by reducing the attacks and security threat against DNS.
