Simon Wilkinson
The distributed nature of existing peer-to-peer (P2P) systems exposes them to several security vulnerabilities, including observability and availability through the use of both passive and proactive attacks. We propose a new secure, private, and anonymous P2P system that is based on ideas in the field of Artificial Intelligence, and more specifically, a meta-heuristic algorithm known as Ant Colony Optimization (ACO). By utilizing the natural ability of an ant colony to find the shortest path between their nest and a food source through the use of pheromone trails, these algorithms adapt naturally to topology changes, work in a distributed manner, are highly adaptive and robust, and provide automatic load balancing. Our scheme will provide improved confidentiality and integrity in the system, as well as improving on link and end-to-end encryption present in current P2P networks by employing the use of a threshold scheme to break apart messages, and using a set of disjoint paths to send them across.