Multi-Signal GNSS & the Curious Case of Galileo E24: Part 1 - Bert Hubert's writings
a day ago
- Multi-frequency GNSS removes over 99% of ionospheric delay by combining signals at different frequencies.
- Galileo uses an 'ionosphere-free clock combination' as its official time signal, but internal satellite delays cause small clock offsets.
- Single-band receivers (like phones) must apply a correction term (BGDE1E5a/b) to compensate for the combined clock adjustment.
- Galileo has two clock combinations (E1/E5a and E1/E5b), each with separate clock parameters (af0) that receivers must match correctly.
- Satellite E24 shows a persistent ~5 nanosecond (1.5 meter) clock bias in correction feeds, due to a mismatch in clock references.
- NAVCAST corrections convert E1/E5a clocks to E1/E5b clocks, causing the bias, which is resolved by using I/NAV clock parameters.
- The issue arises from an undocumented RTCM standard and the choice of I/NAV as the reference clock for corrections.
- Offline SP3 files also show the bias, but only when compared to the wrong clock reference (I/NAV instead of F/NAV).
- The af0 parameters between F/NAV and I/NAV clocks differ by 5.8789 ns for E24, closely matching the observed bias.