Engineering Proceedings (Oct 2023)

E1-E6 SDR Platform Based on BladeRF for Testing Galileo-Assisted Commercial Authentication Service

  • Rafael Terris-Gallego,
  • Ignacio Fernandez-Hernandez,
  • José A. López-Salcedo,
  • Gonzalo Seco-Granados

DOI
https://doi.org/10.3390/ENC2023-15428
Journal volume & issue
Vol. 54, no. 1
p. 29

Abstract

Read online

To enhance the robustness of the Global Navigation Satellite System (GNSS) against malicious attacks (e.g., spoofing), the European Galileo is working on new services, like Open Service Navigation Message Authentication (OSNMA), which provides authentication on the navigation bits, or Commercial Authentication Service (CAS), which aims to encrypt the spreading code chips. An assisted mode of the latter, named Assisted Commercial Authentication Service (ACAS), is currently under definition by the Galileo program. It uses the Timed Efficient Stream Loss-tolerant Authentication (TESLA) keys provided by OSNMA on the E1-B signal to re-encrypt some fragments of the encrypted E6-C signal, known as Re-Encrypted Code Sequences (RECSs), that are made available in the GNSS Service Centre (GSC). Once downloaded by a compatible receiver, they can be decrypted using the corresponding key and used to perform the correlation with the broadcasted E6-C signal. If that results in a correlation peak, the signal can be authenticated under certain circumstances. However, the probability of detecting this peak depends on the length of these fragments and their periodicity, since they are only provided for certain predefined instants. Indeed, if the receiver relies solely on E6-C signal and has no accurate time reference, this probability is severely degraded. This is why the nominal operating mode proposed for ACAS is to use the estimates provided by E1-B to reduce the uncertainty on the E6-C signal, so that the receiver can know precisely where these fragments are located. In the context of the PAULA project, we have developed a low-cost hardware platform based on bladeRF that allows acquiring both E1-B and E6-C samples synchronously. In this paper, we describe how to set up this platform and we characterise the alignment between the E1-B and E6-C estimates (code phase and Doppler frequency) using the real datasets obtained with such a platform, which is of key importance for the ACAS nominal mode. The results confirm the convenience of using the estimates from the E1-B signal for ACAS.

Keywords