Linear Jamming Bandits: Learning to Jam 5G-based Coded Communications Systems

TL;DR

This paper explores how reinforcement learning with contextual bandits can be used to effectively jam 5G-based wireless communication systems, revealing vulnerabilities and learning dynamics under imperfect feedback conditions.

Contribution

It introduces a novel approach combining reinforcement learning and jamming strategies to analyze vulnerabilities in 5G communications with unreliable feedback.

Findings

Reinforcement learning can successfully learn to jam 5G signals.

Imperfect ACK/NACK feedback impacts the jammer's learning speed.

The study highlights vulnerabilities in 5G protocols to RL-based jamming.

Abstract

We study jamming of an OFDM-modulated signal which employs forward error correction coding. We extend this to leverage reinforcement learning with a contextual bandit to jam a 5G-based system implementing some aspects of the 5G protocol. This model introduces unreliable reward feedback in the form of ACK/NACK observations to the jammer to understand the effect of how imperfect observations of errors can affect the jammer's ability to learn. We gain insights into the convergence time of the jammer and its ability to jam a victim 5G waveform, as well as insights into the vulnerabilities of wireless communications for reinforcement learning-based jamming.