On the estimation of frequency residuals in a radio occultation experiment

In the radio occultation (RO) experiments, the difference between observed and predicted Doppler, called frequency residuals, form basic input data to retrieve the vertical profiles of atmospheric parameters. In this paper, we compare different methods to estimate frequency residuals in a one-way downlink single frequency RO experiment, and propose a suitable method that not only improves the residual estimates but is easy to implement too. We show that the Doppler broadening in the received radio signals, which causes uncertainties in Doppler estimation, is predominantly due to changes in the line of sight velocity between the receiver and transmitter during the sampling period. Coupled with the random noises in the receiving system and fluctuations in plasma medium along the propagation path, the resultant uncertainty in Doppler estimates can become as large as 60 Hz. The expected phase changes due to the ionospheric dispersion/refraction, on the other hand, are of the order of 0.01-2 Hz only. To circumvent the issue, we re-sample the observed radio signals in smaller time packets. After removing DC constant shift, a narrow bandpass filter is applied to suppress small-scale fluctuations and Doppler and power of the received signals are estimated. Using Akatsuki radio signals, tracked at Indian Deep Space Network as test data sets, we compare our results with the residuals provided by the Akatsuki team and using the method used for analysing Cassini RO data. We show that our method provides a better estimate of frequency residuals with lesser standard deviations.