Figure 1. Communications System Overview
The goal of this project is the design of a feasible communications system for NASA’s upcoming mission to Neptune. The proposed mission includes three probe satellites that descend through Neptune’s dense atmosphere at +45°, 0°, and -45° latitude, collecting and transmitting data through their decent. A robust communications system is needed to gather as much streaming data from each of the probes as possible before they succumb to the planets harsh conditions. The proposed communication system calls for an additional satellite placed in Neptune-stationary orbit to serve as a relay between the probes and a NASA earth station. NASA’s Deep Space Network (DSN), which will serve as the earth station, is a network of ground stations allowing for continuous uninterrupted reception of the relay satellite’s signal.
Neptune is a gaseous planet, larger than earth, located approximately 4.347 billion km from earth. Planetary data for Neptune is listed below (see Table 1).
| Average distance from the Sun | 4.497×109 km |
| Mass | 1,023.5×1023 kg |
| Equatorial radius | 24,764 km |
| Inclination of orbit to the exliptic | 1.77° |
Due to the attenuation from Neptune's gaseous atmosphere, the fixed positions of the probes latitudes, and the need to move data quickly from them, a relay satellite will need to be placed in orbit around Neptune. As the probes fall, they will rotate around Neptune at the same rate as the planet itself. To keep the probes in view of the relay satellite throughout the entire mission, the relay satellite will be placed in Neptune-stationary orbit, meaning it’s placed over Neptune’s equator, in a circular orbit, at an altitude of 61,090 km (see Equation 1).
Since the satellite is located at the equator, and Neptune is tilted only slightly with respect to the ecliptic, it will be eclipsed by the planet during the mission. Assuming a mission length of 50 hours, which is approximately 2.9 Neptune days, the satellite will be eclipsed 3 times. To handle the outage, the relay satellite must be able to buffer probe data. Calculating that each probe sends at 8 kbits/sec, for three probes at 50 hours, buffering data for the entire mission would take
bytes. While 553 MB’s of solid state space grade storage would not be cheap, it would be obtainable. An added benefit of buffering the entire mission would be if any problem occurred in the relay to earth station link, the relay satellite could be asked to replay any part of the entire mission when the link was back up.
The relay satellite to earth link will be characterized by the distance from Earth to Neptune. The mission should be planned for such a time that Earth is as close as possible to Neptune (not including the eccentricity of the orbits since Neptune’s orbital period is almost 165 years); whenever they are on the same line of sun longitude. The distance from an Earth station near the horizon to a satellite orbiting Neptune, also near the horizon, is approximated as the distance from the center of the two bodies, which is approximately 4.347402 billion km. The figure below illustrates this.
Designing the probe to relay satellite link revolves around a link budget. Before the link budget can be accurately computer however, several factors must be considered and evaluated.
Note: Both link designs in this project will very likely be two-way links, even though the design only specifies one direction. This is always done in the constrained direction, and due to symmetry it is valid to assume if the link works one way, then the more flexible direction will also work. For example, the probe is more limited in power and antenna size then the relay satellite, so the design is specified from the probe to relay.
The figure below shows the basic geometry of the relay satellite/probe system for the worst case, the probes at 45° and -45° latitude.
The first thing to notice is that Neptune’s thick gaseous atmosphere causes a good deal of attenuation, proportional to frequency, in the same way that free space path loss increases with frequency. There is an inverse relation however, between frequency and antenna size. Thus, the optimal choice is the lowest frequency the probe can use and still reasonably accommodate as high a gain antenna for that frequency. A frequency of 400 MHz was chosen as a compromise. Also, access to the relay satellite will be FDMA, the most southern probe will transmit on 399.50 MHz, the equator probe at 400.00 MHz, and the northern probe at 400.50 MHz. The table below shows the attenuation due to the Neptunian atmosphere for a range of frequencies, calculated for a straight path and the worst case path length that will be encountered by the probe/relay link. The interpolated value for 400 MHz is also shown.
| Frequency | 100 MHz | 500 MHz | 1000 MHz | 2300 MHz |
| Straight Path Loss (dB) | 0.21043 | 4.64343 | 16.4778 | 78.6907 |
| Worst Path Loss (dB) | 0.35773 | 7.89198 | 28.0123 | 133.774 |
Also to consider is the antenna placement on the probe. Quarter wave dipole ground plane antennas will be used to maximize the gain while still providing an omni directional pattern on the two probes at 45° latitudes. A quarter wave dipole at 400 MHz is less than .2 meter long. The peak gain of quarter wave dipole is 5.1 dB, but noticing the figure above, the satellite will be just in the dipole’s 39° half-power beamwidth, meaning a worst case value of 2.1 dB will be used in calculations. The use of an omni directional antenna simplifies the overall probe design by relieving any need to keep the probe orientation fixed to keep an antenna pointed at the relay satellite. For the probe on the equator however, the same arrangement will not work. It’s antenna will have to be placed on the side of the probe, or somehow otherwise mounted horizontally to put the relay satellite in the antenna’s view.
The biggest reason 400 MHz was chosen over a lower frequency was so that the receive antenna on the relay satellite could be a high gain dish type antenna, which become unreasonably large at low frequencies. The gains of a dish antenna will more than make up for the added atmosphere attenuation of using the higher frequency.
Finally, the transmitter and receivers physical parameters must be known. Below is the probe transmitter hardware design (see Figure 3).
The transmitter consists of a baseband processor, a Maxim 2510 IQ Modulator, and a HD Communications HD20327 Solid State Broadband RF Power Amplifier. The baseband processor takes care of two critical encoding tasks, turbo coding and applying a CDMA spreading sequence. These steps are necessary to maintain a reliable long distance link. Modern half rate turbo coding can achieve coding gains of up to 10 dB over an uncoded channel. A CDMA spreading sequence of 20 bits will be used, which yields a spreading gain of 13.0 dB. While CDMA is usually used to multiplex more than one user to a channel, in this case it is used only for its coding gain, as each probe will operate on its own frequency to minimize interference and maximize C/N at the receiver. A short code is used to minimize time spent synchronizing to the transmitter.
| Output Frequency Range | 100 - 600 MHz |
| Output Power | -41 - 1 dBm |
| Output Frequency Range | 100 - 520 MHz |
| Input Power | 0 dBm |
| Output Power | 6.99 dBW |
Now the relay satellite receiver architecture. The receiver consists of a Maxim 2640 Ultra Low Noise Amplifier that feeds a Linear LT5502 Quadrature Demodulator. As in the transmitter, the baseband processor is responsible for the CDMA despreading and turbo code interpretation. Three receivers will be required to receive all three probes simultaneously.
| Noise Figure | 0.9 dB |
| Gain | 12.8 dB |
| Noise Figure | 4.0 dB |
The receiver will use a dish antenna for maximum gain. According to the geometry figure at top, the dish needs a beamwidth of at least 21.3° to see all three probes simultaneously. The dish will also need three feedhorns to feed all three receivers. The calculated gain and verified beamwidth of a dish for 400 MHz with an assumed aperture efficiency of 0.8 and diameter of 2.5 meters is
Putting it all together, we now have enough information to calculate the probe to relay satellite link budget. First the receive power is
and the received noise power is
Then the C/N and corresponding BER at the relay satellite is
As stated above, Neptune is approximately 4.347 billion km from Earth. To communicate over such distances, large dish antennas with high gain are needed. High frequencies give better gains for the size dish, and as such 40 GHz will be used. With a 3 meter dish on the relay satellite, the gain and beamwidth are
Notice a beamwidth of less than two tenths of a degree. This means the antenna will have to be pointed fairly accurately, although at 4.3 billion km away, .18° still covers more than 1000 earth diameters.
The transmitter hardware is similar to that of the probe transmitter, but instead of the power amplifier, a Hittite HMC329 GaAs MMIC Double-Balanced mixer is put in its spot to upconvert the 400 MHz signal to 40 GHz. That is then feed into a AR Worldwide 40T26G40A 40 W power amplifier. The block diagram of the system is below.
The Earth station is NASA’s Deep Space Network (DSN). The DSN has 34 meter dishes with an aperture efficiency of 94%. The gain and beamwidth of one such dish is
The link budget then is
And the received noise power is
Thanks to the outstanding noise temperature of the DSN receivers, the received noise strength is very low. The C/N and BER is then
While the claim that the BER is actually 0 is invalid, it will be extremely low with a C/N of 26 dB. The huge gain made possible with a 32 m dish at 40 GHz is a large reason this is possible. This leaves a huge safeguard against calculation errors, or better yet, makes some optimizations possible. The dish could be shrunk in size if need be, or possibly the frequency lowered. Or, the output level of the power amplifier on the relay satellite could be turned down to save power. The data rate could also be increased to get data back to earth faster, making up for the time the satellite is eclipsed by Neptune.
