3.1.1. Payload Analysis
It was decided that the
3Cat-3/MOTS mission would carry on-board two payloads: a multispectral optical sensor [
27] in the VNIR (Visible and Near-InfraRed) [
28,
29] and a GNSS-Reflectometer. The final product will consist of data fusion from both payloads merging the multispectral image obtained from the optical sensor and the data collected by the geolocated reflectometer. This section details the performance of the optical sensor and the GNSS-Reflectometer from the point of view of the physical constraints and technical requirements to fulfill the mission statement.
• Optical Sensor
The selected COTS (Commercial off the Shelf) optical sensor and telephoto lens (
Table 3) fulfill the mission statement described in
Section 2 in terms of the Ground Sampling Distance (GSD) and swath. The study has thoroughly considered several COTS candidates for the optical sensor and lenses, but the decisive criterion was the solution adopted to provide multispectrality to the optical system with reliable technology. The standard solution of a filter wheel presents a problem for both the size of the wheel and the filter switch delay introduced into the acquisition of the multispectral image. The selected camera [
29] has a set of up to two charge-coupled device (CCD) optical sensors with its own filters, solving the multispectrality issue. The size of the camera allows for a 75 mm optical focal length lens. The resulting GSD at nadir is calculated from the sensor’s pixel size (p), the focal length of the assembly (f), and the platform height (h) given the specifications of the manufacturer:
The predicted GSD values are only correct if the aperture of the optical system is large enough so as to satisfy the Rayleigh’s diffraction criterion. As the wavelength increases, the condition of the minimum aperture diameter (AP) of the optical system becomes more stringent:
where λ
max is the longest wavelength among all bands. Assuming a narrow swath, it is directly calculated from the GSD obtained as:
where #pixels is the number of pixels in the cross-track direction. Another key parameter calculated after the swath is the Field of View (
FOV). The lens manufacturer provides the
FOV defined in the horizontal, vertical, and diagonal directions, which has to be larger than:
so that the whole image is projected on the sensor. In order to measure the feasibility of the optical system in terms of image quality from the orbital configuration described, the signal-to-noise ratio (SNR) must also be calculated. The solar radiation spectrum at the top of the atmosphere (TOA) is not constant in all bands of interest; thus, the reflected electromagnetic (EM) wave received by the optical sensor has different signal power values for each band studied [
5], (pp. 17–20).
where
is the Exo-Atmospheric Irradiance (EAI) in (W
2/m
2·nm),
is the atmospheric transmission coefficient,
is the reflectance over the Earth’s surface, which depends on the albedo value,
is the area of the pixel in the detector provided by the manufacturer,
is the receiver optical spectral width, and the G-number (
), which includes the f-number (
), characterizes the optical system:
where
is defined as the ratio between the focal length and the aperture:
, and
is the optical transmissivity for all of the optical train (optical sensor plus lens). The other factor needed to compute the SNR is the Noise Equivalent Power (
NEP), which is a measure of the goodness of the photodetector in terms of noise:
In Equation (7),
q is the electric charge of the electron (C), h = 6.63·10
−34 J·S is the Planck constant, ŋ is the quantum efficiency, which changes for each band and it is particular for the optical sensor,
and
are the surface and bulk dark currents, respectively (A),
F is the excess noise factor (−),
M is the multiplication factor of the avalanche diode, which in this case is
F =
M = 1, because there is no photo multiplication involved, and
B (Hz) is the inverse of the integration time,
which is related to the
satellite’s ground speed and the
GSD (Equation (1)).
Results are presented in
Table 4 and fulfill the mission statement requirements. The following sections specify the design and simulated performance of all of the required subsystems needed to enable the payloads and the transmission of the data acquired to the ground stations.
• GNSS-Reflectometer
The GNSS payload of the
3CAT-3 is being developed in parallel with the
3Cat-4 ESA Fly Your Satellite project, in which the UPC NanoSat Lab [
12] is also participating. This 1U CubeSat will carry a flexible microwave payload: an Automatic Identification System (AIS), a GNSS Reflectometer, and a microwave Radiometer [
30]. The use of a software-defined radio (SDR) as a data logger is essential to reduce power consumption, the size of the payload, and cost. The scope of the
3Cat-3 is to perform data fusion between the optical payload and the GNSS-R data. The radiometer capabilities of the flexible microwave payload designed for the
3Cat-4 are being evaluated as a possible inclusion in the
3Cat-3 only if they do not jeopardize the development of the main mission in terms of power availability, data budget, and platform design. If the payload were to include a radiometer, the duty cycle of the down-looking antenna (GNSS receivers use GPS, Galileo, GLONASS, and Beidou systems) would increase and thus the power consumed would also increase. There are two possible configurations for the down-looking antenna with a minimum directivity of 12 dB: a 2 × 2 patch array, tested and integrated in the
3Cat-2 (in the
3Cat-2 it was a 2 × 3 patch array because all of the nadir-looking side was dedicated to the antenna), and a helix antenna, which is being developed for the
3Cat-4.
3.1.2. Power Subsystem and Budget
The power subsystem consists of three separated main elements: the Electric Power System (EPS), the batteries, and the solar panels. An overall 80% efficiency has been considered for the Maximum Power Point Tracking (MPPT) and charger. The scheduler defines the set of rules under which the EPS will provide power to every subsystem on demand. On their behalf, each subsystem will try to be powered up when the satellite passes over certain defined areas (TA, GS, or during all orbit).
Table 5 lists each subsystem, the regions over which they need to be powered on, and the typical and maximum power consumption. The battery heater’s schedule is controlled by the thermal budget, which monitors its temperature. The state of the battery heater responds to survival reasons, so it might need to be powered on over any area. Similarly, the ADCS needs to be powered on over TAs (for payload pointing accuracy) and GSs (for antenna pointing reasons), but the power peak when switching it on may indicate that it is preferable to keep the ADCS on during all orbit.
• Solar Panels
The Indium Gallium phosphide/Gallium arsenide/Germanium (GaInP/GaAs/Ge) triple-junction cells solar panels considered have 30% efficiency [
31]. In order to account for the total energy collected by the satellite, both the direct Sun radiation and the radiation scattered on the surface of the Earth (albedo radiation,
Figure 6) have been considered changing along the track. The platform has been modelled as a ~10 × 20 × 30 cm
3 parallelepiped so as to consider the incident angle between both radiation sources and the solar panels mounted on the sides (
Figure 7). It is necessary to detail the orientation of the satellite to interpret the results of the simulation. The nadir direction is aligned with the −
z axis and the linear velocity of the satellite is aligned with the +
y axis.
• BATTERIES
Typically, the manufacturer provides the DoD [
31] of the batteries as a maximum number of cycles of charge/discharge at 25% and 75% of the total charge (A·h) stored. The life cycle of the mission may depend, to a large extent, on the DoD policy. As specified by the scheduler, there are a few survival subsystems that must be powered on all the time, but others may not be activated due to low battery charge values.
• Power Budget
The power budget provides instantaneous battery charge throughout the simulation period (
Figure 8). It shows the state of the batteries’ charge (A·h), already balancing the incoming/outgoing charge from the photovoltaic converter and to the subsystems. The eclipse periods, which account for approximately 30% of the orbit, are automatically treated by the scheduler, which denies all power demands from the subsystems that are not essential for survival. In the simulation, a maximum 20% usage of the total battery charge has been established in order to increase the lifespan of the satellite as much as possible.
3.1.3. Thermal Budget
Temperatures in the thermosphere vary radically depending on the Sun’s illumination. The dependence on solar activity and the cycles of shadowing illumination every day form a highly variable thermal scenario. On the other hand, electronic devices have a well-defined temperature operating range (
Table 6). The platform is exposed to high temperatures when in direct line of sight of the Sun and, if not properly defined, to very low temperatures when shadowed by the Earth. The simulation takes into account as heat sources the direct Sun radiation, the reflected radiation on the Earth’s surface (albedo radiation), the Earth’s radiation or Earthshine, and the internal heat dissipation of the electronic devices when they are turned on. On the other hand, the satellite itself radiates heat to space depending on the surface and emissivity of the solar cells and other materials.
To better control the satellite temperature, a coating should be applied to the surface of the satellite to vary the coefficient of absorptance and emittance. The satellite’s structure is covered with polished beryllium, which has an absorptance coefficient of α = 0.44 and an emittance coefficient of ε = 0.01 [
23], (p. 363). This coating is will help to increase the satellite’s temperature that drops during eclipse periods (simulated instantaneous average temperature of the satellite,
Figure 9). The polished beryllium has a medium absorptance coefficient, but an extremely low emittance coefficient. Another alternative explored was to apply a coating with higher absorptance in order to capture more heat, for example black paint (epoxy), α = 0.95 and ε = 0.85 [
23], (p. 363). The simulations, however, show a much lower thermal budget due also to the higher emittance of the black paint (
Figure 10).
As mentioned in
Section 3.
C. Scheduler, the thermal budget indirectly controls the battery heater which will power on independently of the scheduler if the temperature drops under certain levels to preserve the thermal range tolerance of the devices on-board.
3.1.4. Data Handling and Budget
The data download is performed with an S-band link when there is contact with a ground station and the scheduler powers on the communication subsystem. The amount of data acquired by the optical system can easily get out of hand if the duty cycle is not carefully controlled. The camera acquires images of 1.25 MP. With a digitalization of 12 bits/pixel, each image is in the order of 1.9 MB. Considering the size of the image (1296 × 966 pixels at approximately 26.6 m GSD), the necessary overlapping (5% to 15%) to ease the formation of the mosaic picture, and the ground speed of the satellite (7.06 km/s at an orbital altitude of 500 km), the optical system will generate roughly 4 Mbps of data (neither optical data compression, nor GNSS data acquisition have been considered). This amount of 4 Mbps of data generated at 100% of duty cycle is overwhelming. Typical commercial values of S-band link modules are less than 4 Mbps [
31,
32], and the average contact time with the GSs, located as described in
Figure 2, is around 5.3% of the orbit (
Figure 11). For the simulations, two scenarios have been considered for the S-band link speed: 0.1 Mbps and 0.5 Mbps. The contact time with the GSs is the same for all simulations: 5.3% of the orbital period.
The data download is the final step of the main objective of the mission, so the scheduler gives priority to the data download over the acquisition of new data if the power available is disputed. It is assumed that the OBC has a storage limit of 2 GB, and two cases (a 0.1 Mbps data rate and a 0.5 Mbps data rate) have been assumed in order to perform the data budget. For the 0.1 Mbps downlink (
Figure 12), the amount of data stored in the on-board memory is steadily growing. The transmission rate does not compensate for the data input flux, so given the acquisition configuration over the TAs, the mission will lose data acquired. This situation is corrected by increasing the download data rate to 0.5 Mbps (
Figure 13). The output flux is now higher than the input flux so, all data acquired is rapidly downloaded to the GS’s.