Non-Malleable Codes with Leakage and Applications to Secure Communication

TL;DR

This paper introduces non-malleable codes resilient to active physical layer adversaries, providing new bounds, explicit constructions, and applications to secure communication in wiretap and network settings.

Contribution

It defines a new class of tampering functions for communication, derives rate bounds, and offers two modular constructions including an explicit one.

Findings

Derived rate bounds for the new tampering class

Provided two modular constructions of non-malleable codes

Applied results to secure communication in wiretap II and SMT settings

Abstract

Non-malleable codes are randomized codes that protect coded messages against modification by functions in a tampering function class. These codes are motivated by providing tamper resilience in applications where a cryptographic secret is stored in a tamperable storage device and the protection goal is to ensure that the adversary cannot benefit from their tamperings with the device. In this paper we consider non-malleable codes for protection of secure communication against active physical layer adversaries. We define a class of functions that closely model tampering of communication by adversaries who can eavesdrop on a constant fraction of the transmitted codeword, and use this information to select a vector of tampering functions that will be applied to a second constant fraction of codeword components (possibly overlapping with the first set). We derive rate bounds for non-malleable codes for this function class and give two modular constructions. The first construction adapts and provides new analysis for an existing construction in the new setting. The second construction uses a new approach that results in an explicit construction of non-malleable codes. We show applications of our results in securing message communication against active physical layer adversaries in two settings: wiretap II with active adversaries and Secure Message Transmission (SMT) in networks. We discuss our results and directions for future work.