This paper presents an improved framework for physical-layer authentication (PLA) in multiuser wireless environments, where channel variations, measurement noise, and interference can make it difficult to reliably distinguish legitimate users from attackers. Traditional CSI-based authentication methods often suffer in such conditions because channel estimates collected at different times may not align closely enough to validate identity. To address this issue, the authors introduce an information reconciliation mechanism that refines and stabilizes channel measurements before authentication. Using Polar codes and Slepian–Wolf decoding, the system reconciles CSI observed at two time instants, correcting small discrepancies caused by noise or temporal changes. This process makes legitimate CSI measurements much more consistent, while an attacker’s measurements remain mismatched.
The method also incorporates Lloyd–Max quantization with Gray coding, allowing flexible bit resolutions and improved robustness. After reconciliation, authentication is performed by comparing CSI vectors through Hamming distance–based hypothesis testing, enabling the receiver to decide whether the transmitting user matches the stored legitimate profile. The study demonstrates that this reconciliation-enhanced approach significantly improves authentication accuracy in multiuser scenarios. It achieves high detection performance even with very low false-alarm requirements and outperforms several existing PLA techniques across different system configurations. Overall, the paper shows that combining CSI-based authentication with information reconciliation provides a practical and effective solution for securing future wireless networks, especially those serving large numbers of low-power IoT devices.
Enhanced Multiuser CSI-Based Physical Layer Authentication Based on Information Reconciliation