Simulation of the Cassini solar conjunction experiment with GRETCHEN

According to general relativity, photons are deflected and delayed by the curvature of space-time produced by any mass. The bending and delay are proportional to γ+1, where γ is a parameterised postNewtonian (PPN) relativity parameter which can be estimated using radiometric tracking data collected over a time span of a few weeks around the epoch of a solar superior conjunction. In recent years, the Cassini space mission provided an accurate estimation (to the level of 2×10-5) of this relativity parameter using a novel technique where Doppler data simultaneously acquired in three bands (X/X, X/Ka and Ka/Ka) were efficiently combined in order to eliminate the effects of solar plasma. As the future ESA BepiColombo mission will improve this result by making use of newly developed ground instrumentation and an enhanced version of Cassini's Ka-band transponder, ESA has developed a software tool (the GRETCHEN simulator) to assess the achievable level of accuracy of the Radio Science mission objectives (Relativity Study, Mercury Gravity and Rotation State Experiment). The validation of such tools becomes crucial for the proper selection of experiment scenarios. Regarding the Relativity Experiment, experimental results obtained from the 2002 Cassini Solar Conjunction Experiment were used for testing the mentioned simulator. 23 days of Cassini tracking data around a superior conjunction were simulated with the GRETCHEN software. Tracking data comprised Doppler measurements from one tracking station, which were affected by measurement noises similar to those actually experienced during the Cassini experiment. Uncertainties in solar radiation pressure and accelerations due to radioisotope thermoelectric generators (RTG) were also accounted for according to experimental data. In spite of the difference between the operational processing (performed using JPL's Orbit Determination Program - ODP) and the theoretical covariance analysis performed by GRETCHEN, the two results in the estimation of γ agree well within the achievable accuracy. As expected, the estimation of γ is mainly achieved when the impact parameter is small and the relativity effect is the highest. The obtained accuracy in Cassini's state vector is also similar to that obtained by the ODP, thus confirming the good agreement of the two solutions. Monte Carlo simulations of the estimation process are also performed to confirm the correctness of the theoretical covariance analysis results.