RTS Products

Combination approach (orbits, clocks and code biases)

Clock combination. The satellite clock corrections of the RTACs enter a Kalman filter as pseudo-observations, system by system. Each RTAC clock is modelled as the sum of three parameter types:

• the combined satellite clock correction (the result of the combination), estimated anew in each epoch;
• an RTAC-specific offset per system and epoch, absorbing the different clock datums of the RTACs;
• an RTAC- and satellite-specific offset, modelled as a random walk with a noise of 1 cm per 10 s

(independent of the combination interval), absorbing slowly varying satellite-specific differences between the RTACs (e.g. hardware delay and antenna model differences).

The solution is regularized by minimal constraints: the satellite-specific offsets of each satellite sum up to zero over the RTACs, and the sum of the combined satellite clocks is constrained in each epoch.

Reference signals. The combination is done for the ionosphere-free clocks of the reference signals specified by the IGS Real-Time Committee:

RTACs may estimate their clocks with other signals. Before the combination, each RTAC clock is therefore referred to the reference signals, epoch by epoch, with the ionosphere-free linear combination of the same RTAC’s code biases of the two reference signals (Banville et al. 2020). The combined clocks are thus consistent with the IGS clock convention, even if the code biases of an RTAC contain contributions from other biases. Clocks of an RTAC without code biases of both reference signals are still combined, but excluded from the clock datum.

Consistency of clocks and code biases. For each RTAC and system, BNC checks whether the satellite-specific offsets follow the ionosphere-free combination of the RTAC’s reference-signal biases (regression slope and correlation). Such a dependency indicates code biases that do not belong to the clocks, e.g. a wrong sign convention or a different clock reference. An RTAC failing the check is excluded from the clock datum; it is used again after one hour of passing checks.

Outlier detection. Orbit corrections deviating from the mean of all RTACs by more than a threshold are excluded. Clock corrections are checked iteratively: the observation with the largest residual is rejected if it exceeds the threshold, and the adjustment is repeated. RTAC satellites rejected in most epochs are reported hourly. Satellites with corrections from fewer than two RTACs are not combined.

Orbits and a priori consistency corrections. Orbits are not combined: the orbit corrections of a system are taken from one RTAC solution (the master orbit). If its corrections are missing, the next RTAC in the priority list provides the orbits; the master orbit RTAC is used again as soon as it has delivered for 60 s without interruption. To keep the combined clocks consistent with the master orbits, a priori corrections are applied to the RTAC clocks before the combination:

• orbit: the radial difference between the RTAC and the master orbit:
  dC_orb = r_AC · (r_AC − r_master) / |r_master|;
• attitude: a different satellite yaw angle changes the phase wind-up, which the RTAC clock absorbs:
  dC_att = λ_IF · (ψ_master − ψ_AC) / 2π,

with the ionosphere-free wavelength λ_IF = c / (f1 + f2) (Loyer et al. 2021). The yaw angles are taken from the SSR stream of RTACs providing them, otherwise from an attitude model including the noon and midnight turns. The yaw difference is accumulated over time rather than reduced to ±180°: if two RTACs model a turn in opposite directions, their phase wind-up, and hence their clocks, differ by one full cycle (about 10 cm) afterwards (Loyer et al. 2021).

Reference point. The combination is done for satellite antenna phase center (APC) corrections. The CoM streams are derived from the combined APC corrections with the satellite antenna offsets of the IGS ANTEX file valid at the epoch, using the attitude model.

Timing. The corrections of an RTAC epoch are passed to the combination as soon as all its messages have arrived (Multiple Message Indicator = 0). The epoch is combined as soon as all RTACs have delivered it, at the latest after the combination interval (minimum 10 s), so that late or missing RTACs do not delay the product.

Code biases of the combined products

With the clocks referred to the reference signals, the ionosphere-free linear combination of the two observable-specific biases (OSBs) of the reference signals is zero. All other OSBs can then be expressed in terms of differential code biases (DCBs). The PCO-corrected satellite DCB product of the Chinese Academy of Sciences (CAS, Wang et al. 2025) is used for this and sent as SSR code biases together with the combined clocks. The SINEX bias files are archived at CDDIS.

The combined clocks and code biases have to be used together: their combination, the clock minus the code bias of a signal, corresponds to that of the individual RTACs for all signals (up to the differences between the DCB estimates), although the clocks and biases of the RTACs may individually differ by decimeters because of different clock reference signals (Banville et al. 2020). Users can therefore switch between the combined and the individual products without jumps, as long as clocks and biases always come from the same product.

References

Banville S., Geng J., Loyer S., Schaer S., Springer T., Strasser S. (2020) On the interoperability of IGS products for precise point positioning with ambiguity resolution. Journal of Geodesy. 94, 10 (2020). https://doi.org/10.1007/s00190-019-01335-w

Liu, Q., Hernández-Pajares, M., Yang, H., Monte-Moreno, E., Roma-Dollase, D., García-Rigo, A., Li, Z., Wang, N., Laurichesse, D., Blot, A., Zhao, Q., Zhang, Q., Hauschild, A., Agrotis, L., Schmitz, M., Wübbena, G., Stürze, A.,  Krankowski, A., Schaer, S., Feltens, J., Komjathy, A.,and Ghoddousi-Fard, R. (2021). The cooperative IGS RT-GIMs: a reliable estimation of the global ionospheric electron content distribution in real time. ESSD, 13 (9), 4567-4582. doi: 10.5194/essd-13-4567-2021 

Loyer, Sylvain & Banville, Simon & Geng, J. & Strasser, Sebastian. (2021). Exchanging satellite attitude quaternions for improved GNSS data processing consistency. Advances in Space Research,Volume 68, Issue 6, 2021, Pages 2441-2452, ISSN 0273-1177, https://doi.org/10.1016/j.asr.2021.04.049

Wang N., Li Y., Li Z., Liu A., Liu B. (2025) Determination of multi-GNSS differential code biases with satellite antenna phase center corrections. GPS Solutions 30, 22 (2026). https://doi.org/10.1007/s10291-025-01983-w

Content Description of the RTS Broadcast Ephemeris Streams

RTS also provides real-time access to broadcast ephemeris. Several streams are generated which carry only ephemeris data and no observations. Incoming ephemeris are checked for errors and inconsistencies and then merged, encoded and uploaded to NTRIP broadcasters with a high repetition rate: