Deep Learning for the Gaussian Wiretap Channel

TL;DR

This paper applies end-to-end deep learning with a modified loss function to optimize secure communication over Gaussian wiretap channels, balancing reliable message transmission and secrecy against eavesdroppers.

Contribution

It introduces a novel secure loss function for neural networks to learn secure encoding strategies in wiretap channels, demonstrating the neural network's ability to cluster symbols for secrecy.

Findings

Neural networks learn a trade-off between reliability and secrecy.

Higher noise levels hinder eavesdropper's ability to distinguish symbols.

The approach effectively balances communication quality and security.

Abstract

End-to-end learning of communication systems with neural networks and particularly autoencoders is an emerging research direction which gained popularity in the last year. In this approach, neural networks learn to simultaneously optimize encoding and decoding functions to establish reliable message transmission. In this paper, this line of thinking is extended to communication scenarios in which an eavesdropper must further be kept ignorant about the communication. The secrecy of the transmission is achieved by utilizing a modified secure loss function based on cross-entropy which can be implemented with state-of-the-art machine-learning libraries. This secure loss function approach is applied in a Gaussian wiretap channel setup, for which it is shown that the neural network learns a trade-off between reliable communication and information secrecy by clustering learned constellations. As a result, an eavesdropper with higher noise cannot distinguish between the symbols anymore.