A Poor’s Man Approach to Solar Radio Emission Characterization

The Sun (both the photosphere and the corona) is a strong source of radiation across the entire RF spectrum. Under some particular circumstances, Sun noise may heavily interfere with ground receivers for communication satellites or deep space probes, causing degradation, or even destruction, of the useful signal. An accurate characterization of the solar noise is therefore of great importance, especially in the design of satellite broadcasting systems or deep space missions. In this paper we experimentally investigate the transit of the Sun behind a broadcasting satellite by resorting to a commercial-grade receiving equipment for satellite TV and we numerically assess the amount of radiated noise. Experimental results are eventually validated by a comparison with data collected by NASA’s deep-space missions.


Introduction
The Sun is a strong source of radiation across the entire radio frequency (RF) spectrum, originating from both the photosphere and the corona.In some particular circumstances, the Sun may heavily interfere with the (weak) microwave (µW) signal received by an Earth station (ES) and arriving from either a communication satellite on a geocentric orbit, or a deep-space probe travelling across the Solar System.For instance, the downlink of a communication geostationary satellite (GEOS) for direct-tohome (DTH) TV broadcasting (or broadband access), in C, or Ku, or Ka bands, experiences interference when the Sun, during its apparent motion across the sky, enters the beam of the ES antenna (a satellite TV dish), and from ground it is seen to pass behind the satellite [1]; this phenomenon is usually termed "Sun transit" (Figure 1).In the case of a deep-space mission, instead, the ES, typically one of NASA's Deep Space Network (DSN) stations equipped with a large-size, high-gain, and narrow-beam antenna, may occasionally see the probe (e.g., the Voyager 1/2) nearly in conjunction with the Sun, i.e., at low Sun Earth probe (SEP) angles.Then, the ES antenna, aimed at the probe, will catch additional RF noise from the solar corona, affecting the reception of the probe's signal in the X, or Ka bands.In both the cases, the Sun causes a noise increase in the ground receiver, termed Sun noise temperature (SNT) T S (in Kelvin), which sums up to the background noise, i.e., cosmic, atmospheric, ground and receiver electronics contributions.The detrimental effects of SNT can be partial degradation, i.e., signal-to-noise ratio (SNR) reduction and error rate increase, or even outage (i.e., total destruction of the useful signal), and this is why this phenomenon is also known as "Sun fade" or "Sun outage".An accurate characterization of the SNT is of great importance in the following applications.i) GEOS-based broadcasting -The SNT can be used in the design of the downlink [2], for Sun fade/outage prediction.ii) Opportunistic rainfall estimate -The attenuation measurement in GEOS downlinks can be used to derive a real-time estimate of rainfall intensity (in mm/h) along the signal path [3].This feature reveals especially appealing for nowcasting services and climate change monitoring (e.g., the H2020 SCORE project, https://score-eu-project.eu) and, as such, it requires accurate real-time measurements of the received SNR and prompt management of spurious, i.e., non-rain related, alterations of SNR which may cause false detection of rain events.iii) Deep-space missions -The SNT is necessary for the evaluation of the lowest SEP angle allowing a reliable link from the probe to the ES through the solar corona, and the duration of the communication blackout during the transit of the probe behind, or in front of, the Sun.
In this paper we experimentally investigate the transit of the Sun behind a broadcasting GEOS, and we numerically assess the SNT in Ku-band, by resorting to a simple commercial-grade DTH receiving equipment.Experimental results are eventually validated by a comparison with data collected by NASA's deep-space missions.

Measurement System
The measurements presented in this work have been made by two ESs located in Northern Tuscany, Italy, (Figure 2, right), which are part of a sensor network deployed across Central Italy during the Nefocast project [4].The standard time of both the ES locations is UTC+1; their ID and coordinates are listed in Table 1.The measurement equipment in each ESs consists of a commercial-grade receiver for DTH digital video broadcasting -satellite 2nd gen.(DVB-S2) and a conventional parabolic dish.Both the Nefocast ESs were aimed at the Eutelsat 10A satellite, in 10°E orbital position (Figure 1).

IoT FIRST Terminal and Platform
The satellite terminal, called IoT FIRST and produced by AYECKA Ltd., Israel (www.ayecka.com), is an innovative low-cost and small-size device, which integrates into the same compact case both the low-noise block converter (LNB) and the decoder functionalities.Thanks to its optimized form-factor, the IoT FIRST device can be mounted in the focus of a conventional offset dish (Figure 2 left).The IoT FIRST is a two-way (i.e., transmit/receive) device that enables the following links, providing support for both interactive TV services and Internet of things (IoT) / machine-to-machine (M2M) applications: i) forward link (FL) reception -satellite to ground down-link in the 10-13 GHz Ku-band, featuring DVB-S2 receiver; ii) return link (RL) transmission -ground to satellite uplink in the 14 GHz Ku-band carrying the receiver status, including the FL SNR plus IoT sensor data, featuring a low-power ground-to-satellite transmitter.
The SNR measurements presented in this work have been made on the FL signal, whose parameters are listed in Table 2.The RL enables the receiver to send these SNR measurements to a remote data collection and fusion center, via the EUROBIS IoT First platform and using the same Eutelsat 10A satellite (Figure 1).[4].

Sun Transit
Twice a year, around equinoxes [1], Sun transits cause deep notches on the SNR measured at the satellite receiver.For a 80 cm-diameter dish, every notch has a duration of about 20 minutes and occurs every day over a period of a dozen of days.In the Northern Hemisphere this phenomenon happens a few days before the March equinox and a few days after the September equinox, while in the Southern Hemisphere it happens after the March equinox and before the September equinox.The actual number of days in advance or delay w.r.t. the equinoxes depends on the ES latitude (see Figure 4).Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.

Datasets
We used the following three different datasets containing SNR readings from the satellite Eutelsat 10A at 10 • E, taken every minute during the March and September equinoxes by the IoT FIRST receivers of the two Nefocast ESs: • The SNR readings are in the form of the ratio η = E s /N 0 , wherein E s is the average RF received energy during one QPSK symbol interval and N 0 is the one-sided noise power spectral density (PSD) [4].For the sake of exemplification, the SNR measurements in dataset A, plotted in Figure 5 exhibit a series of deep notches caused by Sun transits, occurring daily around 11:30 UTC (12:30 local time).Actually, the NEfocast ES and the satellite have almost the same longitude, so the transits occur around local noon.

Solar Noise Temperature Derivation
Let denote with: i) t n = nτ the nth sampling instant of the receiver, where τ is the sampling interval (1 minute); ii) t start = i 0 τ and t stop = i 1 τ, with i 0 and i 1 integers, the start and the end instants of the notch, respectively; iii) E s (nτ) the received energy per symbol (it is timevarying, due to residual orbit inclination); iv) + T G + T R the total noise temperature without extra-terrestrials sources, where T M , T G , and T R are the noise temperatures of meteorological formations, ground and receiver hardware, respectively, and L is the atmospheric attenuation; v) N 0 the one-sided PSD of the noise [4] without the contribution of the Sun, expressed as where k B is the Boltzmann constant and T C is the noise temperature of the cosmic background; vi) N (S) 0 (nτ) the time-varying one-sided PSD of the noise, including the Sun noise from both photosphere and corona, or from corona only, expressed as 0 (nτ) the samples of the SNR during the Sun transit; viii) η (re f ) (nτ) the reference SNR level, which is calculated by linear interpolation between the sample at the start of the transit η(t start ) and the one at the end η(t stop ) (red marks in Figure 6), and can be expressed as η (re f ) (nτ) = E s (nτ)/N 0 .The numerical values of the parameters that appear in PSD expressions (1) and ( 2) are reported in Table 3. From vii) and viii), with some manipulations, we obtain the SNT picked-up by the receiving antenna during the transit According to (3), Figure 7 shows the peak values of the SNT for the ten transits observed in Massa, near the March 2022 equinox (dataset A), while Table 4 reports the peak values of all the three datasets.Notice that, by plotting the apparent trajectory of the Sun in the sky, and taking into account both the apparent diameter of the Sun and the antenna beamwidth, it can be verified that the first and the last peaks of each dataset, reported in bold in the Table, are due to the effect of the solar corona only, as the photosphere appears to transit outside the antenna spot (not shown here due to lack of space).Table 4. Peaks of SNT for the three datasets (in bold, noise peaks due to solar corona only).

Figure 1 .
Figure 1.Sun transit, at 15 minute-steps, behind GEOS Eutelsat 10A, 10°E, in Pisa, Italy.Sun diameter is not to scale.The Dashed line is the arc of the GEOSs.

Figure 2 .
Figure 2. IoT FIRST terminals and offset satellite dishes (left).Location of the 2 Nefocast ESs in Tuscany (right).

Figure 4 .
Figure 4. Start/end dates of Sun transit periods.

Figure 5 .
Figure 5.Time plot of dataset A.

Figure 6 .
Figure 6.Record of SNR from Eutelsat 10A during a Sun transit (blue marks) and reference SNR level for the evaluation of the excess noise (red marks).

Figure 7 .
Figure 7. Peaks of SNT from dataset A.

Table 1 .
ID and coordinates of the Nefocast ESs.

Table 2 .
Main parameters of the FL from Eutelsat 10A.

Table 3 .
Parameters of the noise PSD.