Secure Regularized Zero Forcing for Multiuser MIMOME Channels

TL;DR

This paper introduces a secure regularized zero forcing precoding scheme for multiuser MIMOME channels, enhancing secrecy by suppressing eavesdropper information leakage and providing a closed-form ergodic secrecy rate expression.

Contribution

It proposes a novel linear precoding method that improves physical layer security in MIMOME channels and analyzes its performance in large-system regimes.

Findings

High robustness against eavesdropper channel quality

Closed-form ergodic secrecy rate expression derived

Enhanced secrecy performance demonstrated

Abstract

This paper proposes a new linear precoding scheme for downlink transmission in MIMOME channels, referred to as secure regularized zero forcing. The scheme modifies regularized zero forcing precoding, such that the beamformers further suppress the information leakage towards the eavesdroppers. The proposed scheme is characterized in the large-system limit, and a closed-form expression for the achievable ergodic secrecy rate per user is derived. Numerical investigations demonstrate high robustness against the quality of eavesdroppers' channel.