An FFT-based method for Doppler observables estimation in Deep Space tracking

Doppler tracking is nowadays an essential tool in deep space orbit determination and radio science investigations. One of the most critical steps in this tracking process is the frequency estimation, and the variance of this estimation process is directly linked to the accuracy of spacecraft range-rate estimation, therefore possibly limiting it. In this paper, an FFT-based algorithm is presented and characterized using signals corrupted by additive noise channels of varying intensity. A comparison to the current operational tracking method for Open Loop data, based on Phase-Locked Loops, is presented and discussed. Monte Carlo simulations confirm that the estimation accuracy of the proposed method is appropriate for radio science experiments and achieves better accuracies than the alternative method already in use, in case of low signal-to-noise ratios.