Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Physical Layer Authentication Using Information Reconciliation

Atsu Kokuvi Angélo Passah, Rodrigo C. de Lamare, Arsenia Chorti

TL;DR

The paper tackles physical layer authentication in large-scale IoT by addressing time-varying wireless channels that degrade PLA performance. It introduces a reconciliation-based PLA leveraging Slepian-Wolf coding with polar codes and CRC-SCL decoding to align channel measurements across adjacent time slots, enabling a robust hypothesis-test authentication. The authors derive closed-form expressions for the false alarm and detection probabilities and show through simulations that the approach outperforms prior CIR-based PLA schemes, particularly at low SNR and for moderate code rates. This method enhances the practicality of PLA for resource-constrained networks by mitigating CSI fluctuations without heavy cryptographic overhead.

Abstract

User authentication in future wireless communication networks is expected to become more complicated due to their large scale and heterogeneity. Furthermore, the computational complexity of classical cryptographic approaches based on public key distribution can be a limiting factor for using in simple, low-end Internet of things (IoT) devices. This paper proposes physical layer authentication (PLA) expected to complement existing traditional approaches, e.g., in multi-factor authentication protocols. The precision and consistency of PLA is impacted because of random variations of wireless channel realizations between different time slots, which can impair authentication performance. In order to address this, a method based on error-correcting codes in the form of reconciliation is considered in this work. In particular, we adopt distributed source coding (Slepian-Wolf) reconciliation using polar codes to reconcile channel measurements spread in time. Hypothesis testing is then applied to the reconciled vectors to accept or reject the device as authenticated. Simulation results show that the proposed PLA using reconciliation outperforms prior schemes even in low signal-to-noise ratio scenarios.

Physical Layer Authentication Using Information Reconciliation

TL;DR

The paper tackles physical layer authentication in large-scale IoT by addressing time-varying wireless channels that degrade PLA performance. It introduces a reconciliation-based PLA leveraging Slepian-Wolf coding with polar codes and CRC-SCL decoding to align channel measurements across adjacent time slots, enabling a robust hypothesis-test authentication. The authors derive closed-form expressions for the false alarm and detection probabilities and show through simulations that the approach outperforms prior CIR-based PLA schemes, particularly at low SNR and for moderate code rates. This method enhances the practicality of PLA for resource-constrained networks by mitigating CSI fluctuations without heavy cryptographic overhead.

Abstract

User authentication in future wireless communication networks is expected to become more complicated due to their large scale and heterogeneity. Furthermore, the computational complexity of classical cryptographic approaches based on public key distribution can be a limiting factor for using in simple, low-end Internet of things (IoT) devices. This paper proposes physical layer authentication (PLA) expected to complement existing traditional approaches, e.g., in multi-factor authentication protocols. The precision and consistency of PLA is impacted because of random variations of wireless channel realizations between different time slots, which can impair authentication performance. In order to address this, a method based on error-correcting codes in the form of reconciliation is considered in this work. In particular, we adopt distributed source coding (Slepian-Wolf) reconciliation using polar codes to reconcile channel measurements spread in time. Hypothesis testing is then applied to the reconciled vectors to accept or reject the device as authenticated. Simulation results show that the proposed PLA using reconciliation outperforms prior schemes even in low signal-to-noise ratio scenarios.
Paper Structure (10 sections, 14 equations, 5 figures)

This paper contains 10 sections, 14 equations, 5 figures.

Table of Contents

  1. Introduction
  2. System Model
  3. Training phase
  4. Authentication phase
  5. Proposed approach
  6. Training phase
  7. Authentication phase
  8. Performance Analysis
  9. Numerical Results
  10. Conclusion

Figures (5)

  • Figure 1: Proposed authentication scheme: the training phase at time $t$ and the authentication phase at time $t+1$, $u\in \{a,e\}$
  • Figure 2: Simulated vs closed-from expression: code rate $= 0.01$, $SNR = 10dB$, $p_0 \approx 0$ and $p_1 \approx 0.5025$
  • Figure 3: $P_D$ vs code rate: $P_{FA} = 0.001$, $SNR = 15dB$
  • Figure 4: ROC curve: code rate $= 0.01$, $SNR = 5dB$
  • Figure 5: $P_D$ vs $SNR$: code rate $= 0.01$ with a $P_{FA} = 0.001$