Secure End-to-End Communication with Optimal Throughput in Unreliable Networks

Abstract: We demonstrate the feasibility of end-to-end communication in highly unreliable networks. Modeling a network as a graph with vertices representing nodes and edges representing the links between them, we consider two forms of unreliability: unpredictable edge-failures, and deliberate deviation from protocol specifications by corrupt nodes. We present a robust routing protocol for end-to-end communication that is simultaneously resilient to both forms of unreliability. In particular, we prove rigorously that our protocol is SECURE against the actions of the corrupt nodes, achieves correctness (Receiver gets ALL of the messages from Sender, in order and without modification), and enjoys provably optimal throughput performance, as measured using competitive analysis. Furthermore, our protocol does not incur any asymptotic memory overhead as compared to other protocols that are unable to handle malicious interference of corrupt nodes. In particular, our protocol requires O(n^2) memory per processor, where n is the size of the network. This represents an O(n^2) improvement over all existing protocols that have been designed for this network model.