Secure Transmission in IRS-Assisted MIMO Systems with Active Eavesdroppers

TL;DR

This paper proposes a secure transmission scheme for IRS-assisted MIMO systems that effectively blinds active eavesdroppers performing pilot attacks by jointly designing beamforming and IRS phase-shifts.

Contribution

It introduces a novel algorithm using secure regularized zero forcing to counteract active pilot attacks in IRS-assisted MIMO systems.

Findings

The algorithm effectively suppresses active eavesdroppers.

Secure transmission is achievable when legitimate and malicious terminals are distinguishable.

Numerical results demonstrate the method's robustness against pilot contamination.

Abstract

This work studies secure transmission in intelligent reflecting surfaces (IRS)-assisted MIMO systems when an active eavesdropper is available in the network. We consider a scenario in which the eavesdropper performs an active pilot attack to contaminate the channel estimation at the base station. Invoking the method of secure regularized zero forcing, we develop an algorithm that designs beamforming vectors, as well as phase-shifts at the IRS, such that the active attacker is blinded. Our numerical investigations confirm that the proposed algorithm can suppress the active eavesdropper effectively, as long as legitimate and malicious terminals are statistically distinguishable.