The goal of this work is to demonstrate how the use of short-term radio-meteorological forecasts can aid the optimization of transferred data volumes from deep-space (DS) satellite payloads to Earth receiving stations. To this aim, a weather forecast (WF) numerical model is coupled with a microphysically oriented radiopropagation scheme in order to predict the atmospheric effects on Ka-band signals in DS links. A regional WFs model is exploited to obtain short-term predictions of the atmospheric state. The microphysically oriented radiopropagation scheme consists in a 3-D radiative transfer model which is used to compute the slant path attenuation and the antenna noise temperature at Ka-band in order to predict the signal-to-noise ratio at the receiving station. As a baseline, the BepiColombo mission to Mercury is chosen. Two prediction methods, statistical and maximization, are introduced and tested in two scenarios: 1) full-numerical scenario, where simulated data are used for evaluating the performances of the prediction techniques; 2) semiempirical scenario, where measured meteorological data are exploited to simulate beacon measurements in clear and rainy conditions. The results are shown in terms of received and lost data volumes and compared with benchmark scenarios. Using short-term radio-meteorological forecasts, yearly data volume return can be increased more than 20% if daily WFs, rather than monthly climatological statistics, are exploited.

The aim of this paper is to investigate the usability of high-frequency channels for deep-space (DS) transmissions exploiting radiometeorological forecast modeling. A previously developed model chain for DS link-budget optimization, based on numerical weather forecasts (WFs), is adopted. The latter, already tested at Ka-band, exploits the combination of a high-resolution mesoscale forecast model and a radiative transfer model to predict the atmospheric scenario and optimize received data volume (DV) during DS transmissions. To shift available Ka-band results to other frequencies, we apply frequency-scaling laws to extrapolate forecast path attenuation, link parameters, and maximum allowed bit-rate for data transmission. Exploiting the available WF-based methodology, we compute DV return for DS missions operating at X-, K-, Ka-, Q-, and W-bands in order to make a comparative study of the behavior of DS transmission-channels at these frequencies. Results show that, in terms of received DV, an innovative WF-based approach is more convenient than traditional methodologies and exhibits a trend similar to the benchmark (ideal case). Increasing link frequency, received DV increases up to Q-band. From Qto W-band, despite received DV does not increase significantly, lost data remain under reasonable values, thus making the W-band suitable if coupled with a WF-based technique.

An improved weather-forecast based link-budget design technique for space-to-Earth links is described. The aim is the stochastic optimization of both transmission symbol-rate and received signal-to-noise ratio. The proposed Radio Meteorological Operations Prediction (RadioMetOP) model takes into account the forecast uncertainty by a space-time ensemble method exploiting the temporal evolution of the predicted radio-meteorological variables over the weather-forecast spatial grid. The unique possibility of testing and validating the RadioMetOP model is presented thanks to the Ka-band downlink measurements available from the support of the European Space Agency’s antenna tracking network to deep-space Hayabusa-2 mission, operated by the Japan Aerospace Exploration Agency. First, the RadioMetOP model accuracy is tested by comparing the signal-to-noise ratio, measured during the transmission periods, with the simulated one, properly scaled to the symbol-rate operated by Hayabusa-2, finding correlation values of 0.9 that confirm the effectiveness of the proposed approach. Second, the a posteriori analysis of the optimization process is accomplished showing that, depending on the considered criteria for the link-budget optimization, the use of the RadioMetOP model would have allowed a transmitted data-volume more than doubled and an average signal-to-noise ratio gain between 2.1 and 3.8 dB.

The weather-forecast based Radio Meteorological Operation Planner (RMOP) model for the dynamical link-budget design in satellite communications at Ka-band is described and validated. For the first time, actual received Ka-band data from a deep-space satellite mission (Hayabusa2 mission from JAXA and supported by ESA) were available for operational tests. RMOP-predicted link-budget parameters were delivered in real-time before each scheduled satellite transmission period. After each transmission, received data measured by the ground stations were exploited for the RMOP validation. The data-volume actually received (transmitted and lost) by Hayabusa2 was compared with the one that would have been obtained if the transmission had been configured according to RMOP predictions. The results prove that RMOP model is capable of receiving more than 100% of extra data-volume with respect to the classical link-budget design techniques while keeping data losses under control. A specific approach usable in case of rainy events is described, too. These outstanding results will pave the way to an operative use of the weather-forecast based RMOP model for future satellite missions.

The mean radiative temperature (Tmr) is a key function controlling the sky noise temperature in microwave receiving systems. A generalized parametric prediction (GPP) model of Tmr for microwave slant paths in all-weather conditions is formulated and presented. The proposed GPP model is aimed at being multi-frequency and surface-temperature scaled, valid for elevation angles from 5° to 90° and for frequencies ranging from 5 to 95 GHz within the three transmission windows delimited by the water vapor and the oxygen absorption peaks. The core of the GPP model is a parametrization driven by a physically-based radiative transfer approach taking into account both extinction, emission and multiple scattering. The expression of Tmr is normalized to the surface temperature of the considered site. The GPP model is verified with measurements available from the multi-instrument ITALSAT campaign in Spino d’Adda (Milan, Italy) in 1994_1997, obtaining a fractional mean error ranging from 0.045 to 0.068. A comparison of the GPP model with the current ITU-R model shows a reduction of the root mean square error up to about 20 K and 30 K, depending on the considered frequency.