The exploitation of Optical Inter-Satellite Links (OISLs) has the potential to provide significant benefits to GNSSs, offering clock synchronization via highly accurate time transfer, precise ranging, robustness against jamming and spoofing, high data rates, and freedom from signal frequency regulations. As with any new technology, it is crucial to conduct in-space experiments to demonstrate the capabilities of OISLs before widespread adoption.
In this work, we present preliminary analyses of an in-orbit demonstrator concept, which is being designed under the name Optical Synchronized Time And Ranging (OpSTAR). It involves two satellites in a trailing configuration, each equipped with two laser terminals. On the ground, two co-located Optical Ground Stations (OGSs) are operated. Whenever both satellites are simultaneously visible from the OGSs, an OISL and two additional Optical Satellite-to-Ground Links (OSGLs) are established, forming a closed "measurement loop" between the two satellites and the two ground stations.
We present a functional model for OISL and OSGL pseudorange observables. A cross-link clock observable is formed by differencing two one-way pseudoranges, from which a relative clock offset estimate is obtained. First, we analyze how modelling errors on differential delays in Two-Way Time Transfer (TWTT) -- relativistic effects, atmospheric delays, hardware delays, and satellite dynamics during the exchange -- impact the estimation accuracy. Next, we study the impact of individual error contributions to the overall zero-sum chain of clock offset estimates across the closed-loop. Results show that errors due to mis-modeling of relativistic effects, satellite dynamics and clock instability are negligible, while hardware and atmospheric delays require accurate calibrations to achieve TWTT at picosecond-level accuracy.