Path Selection Algorithms for Future Internet Architectures
With the advent of IoT and the support for 4G technologies, the Internet now carries plenty of private information from the cyber-physical devices hosted at homes and organizations alike. The private information includes everything from the rate at which our hearts beat to the movie that we watched last weekend. Therefore, protecting information on the Internet is now more important than ever before. In this research proposal, we focus on aspects of the Internet architecture that will allow us to protect user data from the malicious adversaries around the world. Specifically, our objective in this work is to help end users (and their devices) select end to end Internet paths that can be trusted.
As an example, consider an IoT Home Security System that is designed to report to a police station when a thief breaks in. A smart tech-savvy thief, with the help of a hacker, can selectively drop/delay the data packets sent by the IoT devices on the Internet in order to break into the house’s security system. While this might look like a contrived way to hide his tracks, it is also the one that is least detectable. To illustrate, it is a very difficult problem for the police (or anyone in the computer science research community) to distinguish this attack from the drops/delays due to legitimate network issues such as network congestion. Therefore, it would take a long time until someone even realizes that the house has been broken into.
In summary, our work will enable the end users (the IoT Gateway in this example) to detect the presence of the malicious adversary quickly, and ensure that a different communication path that circumvents the adversary is chosen.