The Sardinia Space Communication Asset: Performance of the Sardinia Deep Space Antenna X-band Downlink Capability

The Sardinia deep space antenna (SDSA), managed by the Italian Space Agency (ASI) has started its operations in 2017 aiming to provide tracking and communication services for deep space, near earth, and lunar missions, and to support new and challenging radio science experiments. The SDSA shares with the Sardinia Radio Telescope (SRT) a part of the system and infrastructure, but has its own specific equipment and a dedicated control center. The current SDSA capabilities involve the X-band (8.4 GHz - 8.5 GHz) reception of telemetry from deep space probes within interplanetary missions. In this work we describe the development and performance of the X-band receiving system. It was designed and assembled with the cooperation of both the NASA-Jet Propulsion Laboratory (JPL) and the European Space Agency (ESA). Specifically, NASA-JPL provided the X-band feed and the cryogenic receiver installed in a suitable focus of the SRT devoted to space applications, and ESA provided the intermediate frequency modem system (IFMS) for signal processing. The coupling of the X-band feed with the parabolic reflector of the SRT and the radiating features of the SDSA have been evaluated with simulations performed using CST Studio Suite and GRASP by Ticra. The telecommunication performance of the system has been assessed by measurements and experiments showing a good agreement between estimates and simulations.

The SRT is equipped with an active remote controlled optics, composed of an active surface mounted on the M1 backup structure and six electro-mechanical actuators installed behind M2. The M1 active surface consists of 1008 aluminum panels (with a panel manufacturing root-meansquare error (RMSE) less than 70 μm) and 1116 electromechanical actuators, able to compensate the gravitational deformation of the backup structure. The M2 actuators provide six degrees of freedom for an accurate monitoring and control of the sub-reflector position. The primary reflector surface was aligned with an accuracy below 300 μm RMS at 45° elevation by photogrammetry measurements [1]. Then, combined photogrammetric and laser tracker measurements in the elevation range 15°-90° provided the input to two look-up tables for the M1 active surface actuators and for the M2 actuators, respectively. These information are employed to correct both the M1 surface deformations and the M1/M2 axes misalignment as a function of the antenna elevation angle.
The SDSA facilities allow ASI to join the worldwide DSC network dedicated to data acquisition from deep space missions. In the near future, the SDSA is expected to provide a DSC service to the NASA deep space network (DSN) [8] and to the ESA tracking station network (ESTRACK) [9], but also to operate as a stand-alone infrastructure for other space applications.
A feasibility study of a new optical configuration is ongoing to equip the SDSA with a full operational capability (FOC). This perspective would enhance the current SDSA functionality with a new telemetry, tracking and command (TT&C) asset for both deep space and near earth missions at X-as well as K-and Ka-frequency bands allocated to the space research. Also, the SDSA FOC could ensure emergency operations or provide a strategic backup when redundancy is required, as in the case of space missions with high scientific relevance or subjected to potential critical issues.
Moreover, the cooperation between ASI and INAF will enable the SDSA to cover several frequency bands otherwise not available in other facilities of the DSN and ESTRACK, such as, e.g., the P band [5], allowing the tracking of the entry, descent and landing (EDL) phases of space rovers, by processing the information given by the doppler shift of the UHF carrier [10]. Indeed, the current SRT configuration can host up to twenty receivers, which can be installed in four different focal positions: the primary focus (F1) with focal length to diameter ratio (F/D) equal to 0.33, the Gregorian focus (F2) with F/D equal to 2.34, and the BWG foci F3 and F4 with F/D 1.38 and 2.81, respectively [11]. Thanks to the possibility of exploiting the large number of operating frequency bands of the SRT, the SDSA will also be able to investigate new technological and operative solutions.
The SDSA activity has been started recently thanks to a specific agreement of ASI with NASA, aiming to future collaboration within forthcoming interplanetary missions. Technical and operative support provided by the NASA-Jet Propulsion Laboratory (JPL) helped to start the development of the SDSA functionalities. Partnership with ESA enabled the installation of the intermediate frequency modem system (IFMS) backend into the shielded room of SRT, which has allowed SDSA operators to acquire and record the signal from spacecrafts.
The aim of this work is to describe the set up and the performance of the new X-band Sardinia space communication asset. In Section II, we describe the development, the optical configuration, and the architecture of the current SDSA IOC, i.e. the first step towards the full operational capability of the Sardinia space communication asset. In Sections III and IV, the radiating features and the This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2022 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ downlink performance of the SDSA have been deeply investigated. Simulations have been performed using the commercial software CST Studio Suite and GRASP by Ticra, whereas the experimental validation is provided by observation of known calibrator radio sources. Then, in Section V, the antenna sensitivity has been validated by processing the signal acquired during tracking sessions of the NASA Juno Mission [12]. Finally, conclusion and future perspective are discussed.

II. SDSA X-BAND CONFIGURATION
The present configuration of the SDSA operates in the frequency range 8.4-8.5 GHz, which is the fraction of the Xband allocated to the space research service by the International Telecommunication Union (ITU) for the use in deep space and near-earth communications. The near-earth band, between 8.45 GHz and 8.50 GHz, is used for the reception of signals from space probes at less than 2,000,000 km from Earth [13], whereas the deep space band, between 8.40 GHz and 8.45 GHz, is used for the downlink from space probes at more than 2,000,000 km from Earth [13].  The optical configuration of the SDSA in the X-band is shown in Fig. 3. It is composed of the primary mirror M1, the axially symmetric sub-reflector M2, the rotating mirror M3, and the fixed mirror M5. Mirrors M3 and M5 are portions of ellipsoids with major axis a and minor axis b (see table 1). The focus of the mirror M5 is the focal position F4 of the SRT, where the feed horn devoted to space operations is housed. In table 1 the geometrical parameters of the  mirrors are listed. The optimum theoretical value of the M5 aperture illumination taper is -12 dB at ±10° (see Fig. 3b).  This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. The whole receiving system is represented in Fig. 4, which also depicts the measurement setup, composed of the backend total power and the IFMS backend.
A X-band cryogenic receiver borrowed from JPL-NASA was installed in F4 (see Fig. 4) with a RF bandwidth of 700 MHz centered at 8.45 GHz. Basically, it consists of a corrugated circular feed horn (see [14]) providing the required illumination for M5, a waveguide polarizer and an orthomode transducer providing right-hand circular polarization (RHCP) or left-hand circular polarization (LHCP), and a cryostat hosting the cryogenic LNA that operates at the physical temperature of 10 K. A down conversion section amplifies, filters, and converts the RF band (8.  Fig.  4), either using a total power backend or using a dedicated backend for space applications, named Intermediate Frequency Modem System (IFMS), borrowed from ESA [15]. The total power section is used to calibrate the system, whereas the IFMS operates both in close loop mode (mainly for demodulation and decoding) and in open loop mode. The latter allows to acquire the baseband in-phase and quadrature (IQ) data of the received signal for post-processing. The baseband signal (which is divided into 4 sub-bands/channels of equal size) is recorded into binary files that contain all the available information (i.e., sample rate, quantization, frequency, sample time tags, gain configuration, and so on) aiming to reconstruct the detected signal. Then, from the IFMS output, the IQ data can be used for different types of signal processing, such as fast Fourier transform (FFT) or Doppler shift calculation.

III. RADIATING FEATURES OF THE SYSTEM
A detailed description of the X-band feed horn is available in [14], whereas the coupling between this feed and the SRT BWG optical configuration has been modeled and evaluated using the commercial software GRASP by Ticra. GRASP is a reliable tool for the analysis and design of reflector antenna systems that employs a physical optics (PO) algorithm as the baseline analysis method, supported by Moment Method (MoM) and Geometrical Theory of Diffraction (GTD) solvers for advanced applications.
First, we have modeled and simulated the feed [14] using CST Studio Suite: the simulated radiation pattern in the principal planes at 8.45 GHz is reported in Fig. 5 and the main results of the CST simulation are reported in table 2. The edge taper at ±10° is -12 dB as required.
Then, the CST-simulated far field pattern has been imported as input of the GRASP project (see Fig. 3a) to evaluate the SDSA radiation performance.  Let D the actual antenna directivity and DM the maximum directivity of the aperture [16], defined as: wherein RA is the radius of the M1 reflector aperture and  is the free space wavelength. In our case DM = 75.07 dBi. Then, the aperture efficiency A is defined as D =A DM and the antenna gain is: wherein  is the overall efficiency and i) S is the spillover efficiency; ii) B is the blockage efficiency, including the effect of both the sub-reflector M2 and the quadrupods (Fig. 1); iii) RMS is the Ruze RMS efficiency due to the manufacturing RMSE of mirrors M1, M2, M3, and M5; and to the alignment RMSE of M1 and M2 panels and active surface (see table 3).
iv) AM accounts for the misalignment of the optical path in the BWG.
v) O represents the return loss efficiency, the ohmic efficiency, and the cross-polarization efficiency.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. Using the GRASP model in Fig. 2a we have computed the aperture efficiency A, equal to 0.775.
The spillover efficiency has been computed enabling a dedicated GRASP subroutine, providing S = 0.909.
Then, the GRASP project has been modified including the quadrupods (not shown in Fig. 3a) and the blockage of the sub-reflector M2 to compute the blockage efficiency. The result of this simulation is B = 0.907.
The Ruze RMS efficiency is 0.97, computed using the Ruze equation [17] considering a total RMS surface error equal to 463 μm (see table 3) and a wavelength equal to 355 mm at 8.45 GHz.
The O has been estimated at about 0.99. Finally, we estimated a AM equal to 0.92. This value accounts for a residual misalignment of the BWG optics (M3-M5 coupling) that we are currently investigating to optimize the system performance.
As a result, the simulated overall efficiency  is 0.56 and the gain G is 72.4 dBi at 8.45 GHz. This value can be considered the best SDSA performance as the angular elevation changes.
In table 4 we summarize the main radiating parameters resulting from the simulation of the SDSA GRASP project. In addition, in fig. 6 we compare the simulated and measured beam pattern of the SDSA. In fig. 6 the E-plane, H-plane, and 45°-cuts of the simulated far-field pattern virtually overlap and have been obtained with all the mirrors aligned according to the SRT optical design. A dedicated experiment was carried out on March 23rd 2021 to measure two orthogonal cuts of the SDSA far-field pattern at 8.45 GHz and compare them with the GRASP simulation (see fig. 6). Such experiment consisted in moving the antenna along two orthogonal scans (also known as cross-scan mode) centered on the radio source calibrator 3C84, which is a strong variable flux radio source suitable for beam pattern measures that allows to highlight the sidelobe level (SLL) and their features [18]. Both scans were performed around the elevation angle of 66° since, as we will show in the next section (see Fig. 10), the optimal performance in the X-band of the SDSA, with the current optics configuration, is achieved between 60° and 80°. An integration time of 40 ms was set in the total power backend (see Fig. 4) with a bandwidth of 28 MHz centered at 8.45 GHz.
The measured SLL is below -19 dB and the difference with simulation is likely due to the residual misalignment of the BWG optics (M3 and M5). It is also worth to notice that fig. 6 shows a slight asymmetry among the measured elevation and azimuth cuts due to a residual squint in the M1-M2 axes alignment. These SRT residual optics aberrations will be soon adjusted after a fine calibration of both the M3 position and the M2 look-up table by dedicated metrological measurements. The measured half power beamwidth (HPBW) is in good agreement with the simulated one (HPBW = 0.039°), as shown in Fig. 7 that reports the measured beam deformation  with respect to the simulated HPBW at different elevation angles.  is defined as: where  and  are, respectively, the HPBWs measured along the azimuth and elevation axis using the calibrator radio source 3C84. A quite stable beam deformation with respect to the elevation angle, with an average value of 0.002 degrees, is achieved thanks to the active optics of the SRT (see section 3.4.3 of [1]).
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication.

IV. DOWNLINK PERFORMANCE
The ratio antenna gain-to-system noise temperature (G/Tsys) is the primary figure of merit in the characterization of the downlink performance [19]. The antenna gain has been computed in the previous subsection, therefore we are left to estimate the system noise temperature that is defined as [20]: wherein -TCMB is the contribution from the cosmic background radiation (CMB), equal to 2.73 K. -Tatm is the effective temperature of the atmosphere and is the zenithal sky opacity, both measured at 8.45 GHz using the taumeter available at the SRT site [21]: during our observations = 0.0077 at elevation  = 66° and Tatm = 265 K. The contribution of the first term of (5) is 2.61 K. -is the spillover temperature that includes the contribution of mirrors M1, M2, M3, and M5, given by (6), wherein  is the spillover efficiency of the mirror Mi computed using GRASP, and Tgnd = 290 K ( ,  ,  , and  are 0.998, 0.957, 0.996, and 0.953, respectively). Then the second term of (4) is estimated at = 15.3 K. -is the noise temperature generated by the active and passive microwave components of the receiver (see Fig.  4). The noise temperature added by this receiver has been measured at the NASA-JPL by using the Keysight PXA Spectrum Analyzer in noise figure analysis mode (see Fig. 8). The measured is about 13 K in the receiver frequency range. In conclusion, the overall system noise temperature can be estimated at about 33.6 K, and the ratio antenna gain-tosystem noise temperature at 57.1 dB/K.
The above estimates have been assessed by measurements of ratio G/Tsys at different elevation angles of the SDSA using the definition provided in [22] and, as a further validation, by separate measurements of the system temperature Tsys (see [23]).
Let Pn be the noise power at the backend total power (see Fig. 4) corresponding to the system noise temperature, i.e., in our case, the power from the cold sky in a 28 MHz bandwidth centered at 8.45 GHz. Let Pst be the additional noise power when the antenna is pointed towards a known radio source. Then, by measuring the following ratio (7) the value of G/Tsys can be determined using the formula [22]: (8) wherein k (m 2 kg s -2 K -1 ) is the Boltzmann's constant,  (m) is the wavelength, and  (W m -2 Hz -1 ) is the radiation fluxdensity of the selected known radio source as a function of the frequency f (Hz).
The ratio G/Tsys has been computed by selecting the calibrator radio source 3C147, whose flux value is stable at long and short timescales and it is well known and tabulated in the literature. Specifically, the flux value is evaluated to be 4.68±0.05 Jy at 8.45 GHz (1 Jy corresponds to 10 -26 W m -2 Hz -1 ) [24]. We have tracked the calibrator source from an elevation angle of 80° to the minimum elevation available for the SRT (about 6°), measuring both the power from the

  ( )
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. cold sky and the additional power from the calibrator. This allowed us to compute R using (7), and then the G/Tsys using (8).
On the other hand, an independent measurement of the system temperature has been provided relying on a calibrated noise source (or noise diode) that produces a known power per unit bandwidth. The noise diode has a known equivalent temperature, which is called calibration temperature , and the can be determined by the following expression [23]: (9) wherein the has been computed using the Y-Factor method [25] and is the noise power measured at the receiver output when the antenna is pointed towards the cold sky with the noise source switched on.
The measured Tsys is reported in Fig. 9, showing a good agreement with the estimate resulting from (4), i.e. 36.4 K at 66° elevation against 33.6 K. In Fig. 10 we show the G/Tsys ratio, measured according to (7) and (8), for different elevation angles of the SRT. The measured value at the 66° elevation angle is 56.5 dB, 0.6 dB lower than the simulated estimate, mainly due to Tsys value underestimation of about 2.8 K (which contributes to about 0.35 dB). Combining the results obtained by (8) and (9) we can derive an indirect measurement of the antenna Gain, which is also reported in Fig. 10. The measured gain is 72.1 dBi at 66° elevation angle, in good agreement with the estimates reported in Section III.
An important parameter for the characterization of the system performance is provided by the system equivalent flux density (SEFD). It represents the flux density, measured in W/(m 2 Hz), that a point source must have to produce an antenna temperature equal to the system temperature, and it is defined as [23]: (10) wherein Aeff is the antenna effective area.
Using the definition (10), we can compute the sensitivity of the system ∆ , i.e. the minimum flux density that the system can detect: (11) where B is the system frequency bandwidth and T is the integration time [26].
If we consider the measurement setup for the observation of the calibrator radio source 3C84 at 8.45 GHz, employed to compute the antenna beam pattern (see Fig. 6), with a bandwidth of 28 MHz and an integration time of 40 ms, we obtain ∆ = 0.0564 Jy. Despite the radio source shows a variable flux density at long (months) timescales, we can infer a possible value in the range 26 ± 6 Jy [27,28]. Considering this value, the signal-to-noise ratio is 26.5 dB ± 1 dB, which is consistent with the measurement shown in Fig. 6.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication.  Fig. 11 is obtained by processing one second of these data. The open loop configuration was set in single channel mode with a sampling frequency of 15625 Samples/s and 16 bits of quantization for both I and Q samples. In table 6, the downlink budget is reported and the estimated signal-to-noise ratio (about 41 dB) is consistent with the FFT spectrum reported in Fig. 11.  Signal-to-Noise Ratio 40.9 dB

VI. CONCLUSION AND FUTURE PERSPECTIVES
The successful deployment of the SDSA has enabled the radio telescope to operate in the framework of the international deep space network. In this work we have presented the architecture and the performance of the X-band receiving system of the SDSA. Simulations, estimates, and measurements show that the SDSA can play a competitive role in the panorama of the X-band downlink systems for DSC, as apparent from table 7 where we report a performance comparison of antennas for X-band DSC available in the world [29]. However, it should be noted that a direct comparison between the G/Tsys values in table 7 is not fully consistent since they have been measured in different operating conditions, as discussed in [29]. The downlink capability of the SDSA provides a maximum value of the G/Tsys ratio equal to 56.5 dBi, which is achieved with the best alignment of the SRT BWG optics allowed by the current SRT optical configuration. Optimization of the SRT optics will be reached in the next few months after a fine calibration of both the BWG mirrors and the M2 look-up table by dedicated metrological measurements, aiming to reduce the SLL and increase the antenna gain.
At present, the capabilities of the SDSA are devoted to downlink communication. However, in the near future, ASI will evaluate the development of a transmitting system to enable uplink capabilities for deep space, near-earth, and lunar communication. Possibly, the BWG foci of the SRT devoted to space applications will be used to activate this functionality. In this context, the main future goal is to implement concurrent uplink and downlink transmission in X and Ka bands endowing the SDSA with a triple link X(uplink)/X(downlink), X(uplink)/Ka(downlink), and Ka(uplink)/Ka(downlink), according to the requirements of challenging radio science experiments. Simultaneous This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2022.3183745 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ operation in the X and K band is also planned, mainly for near-earth and lunar communications. In table 8 the possible operating frequency band for the above applications are reported.  After one year working for the Department of Electrical and Electronic Engineering of the University of Cagliari, he joined National Institute for Astrophysics at the Astronomical Observatory of Cagliari for 10 years. He worked there first in the development of the microwave holography system for the Sardinia Radio Telescope. Subsequently, he contributed towards the commissioning of the SRT. In particular, he was involved in the design of the telescope metrology system, of which became then the coordinator, and in the Radio Frequency Interference. Since February 2022, he has been a Full Professor of electromagnetic fields at the University of Cagliari, teaching courses in electromagnetics and microwave engineering. He has authored or coauthored 75 papers in international journals. His current research interests include the analysis and design of waveguide slot arrays, RFID Antennas, wearable antennas, numerical methods in electromagnetics, and microwave circuits and systems. Prof. Montisci is an associate editor of IEEE Access, IET Microwaves, Antennas & Propagation, IET Electronics Letters, and an academic editor of the International Journal of Antennas and Propagation. He was awarded the IEEE Access Outstanding Associate Editor of 2020.
SARA MULAS received her M.S. degree in physics at the University of Cagliari, with a thesis on the spectropolarimetric studies of Supernova Remnants with the Effelsberg and the Sardinia Radio Telescope. She has started the Ph.D. in physics and astrophysics at the University of Cagliari in 2019 and she has also been the recipient of a research grant University of Cagliari-Italian Space Agency in collaboration with the Italian National Institute for Astrophysics. Her work is mainly focused on investigating the scientific requirements to perform a radioscience experiment with a Deep Space Antenna and studying the radio Sun with the Sardinia Radio Telescope in single-dish imaging modality aiming to Space Weather applications. TIMOTHY T. PHAM is the Chief System Engineer of the NASA Deep Space Network. His interest is in system engineering and system development. Besides DSN system engineering work, he also supports the CCSDS activities in Cross Support Transfer Services and the ground system development at the Morehead State University. He has published several papers on antenna arraying, spacecraft tracking, system modeling and performance analysis. He coauthored the book "Antenna Arraying Techniques in the Deep Space Network". He is recipient of the NASA Exceptional Service Medal, NASA Exceptional Achievement Medal, several NASA New Technology and Space Act Awards, and the IARIA Fellow.
JAVIER DE VICENTE Javier de Vicente is a Ground Stations Systems Engineer at ESA since 2000. He has been technical officer for many studies and subsystem developments, as well as project manager for ground station deployments. He is currently responsible for the ESA Delta DOR correlator and the implementation of the RadioScience capability in ESA Deep Space Antennas.
SALVATORE VIVIANO Salvatore Viviano is the manager of the Sardinia Deep Space Antenna. He is graduated in Aeronautical Engineering and attended the School of Aerospace Engineering of Sapienza University of Rome. He has about 35 years of experience in aerospace sector, working for several private companies and public administrations. His main technical and managerial experiences include Flight Dynamics, Control Centers, Operations for S/C and constellation, satellite communication and navigation, in particular managing the development of GPS Signal in Space and differential applications in many areas, EGNOS applications for Civil Aviation and contributing in Galileo system development and regulation for its governmental use.