This project provides communication security in a new way, based on channel noise rather than standard cryptographic constructs and algorithms. The idea is old, from work in the 1970s in the information-theory and coding field, but this classical work used weak security metrics and even then was not able to provide any explicit and practical designs. This research bridges the gap between this classical work and modern cryptography. First, it provides a sound theoretical basis in the form of security metrics that are adapted from modern cryptography and demand much stronger security than the classical metrics. Second, it provides explicit, efficient and optimal methods to achieve security under these metrics based only on the assumption that the channel from sender to adversary is noisier than the channel from sender to receiver. Unlike standard cryptography, this approach is not vulnerable to advances in algorithms and adversary computing power and thus provides a new option to secure highly sensitive information.
