A detailed procedure for a downlink design is as follows:

1. Select carrier frequency, based on spectrum availability and FCC allocations. (Refer to Table 13-6 for TT&C, Table 13-12 for communication satellites.)

2. Select the satellite transmitter power, based on satellite size and power limits.

3. Estimate RF losses between transmitter and satellite antennas. (Usually between -1 and -3 dB.)

4. Determine the required beam width for the satellite antenna, depending on the satellite orbit, satellite stabilization, and ground coverage area (see Chap. 7).

5. Estimate the maximum antenna pointing offset angle, based on coverage angle, satellite stabilization error, and stationkeeping accuracy.

6. Calculate transmit antenna gain toward the ground station, using Eqs. (13-20) and (13-21). You might also want to check the antenna diameter, using Eq. (13-19), to see if it will fit on the satellite.

7. Calculate space loss, using Eq. (13-23a). This is determined by satellite orbit and ground-station location.

8. Estimate propagation absorption loss due to the atmosphere using Fig. 13-10, dividing the zenith attenuation by the sine of the minimum elevation angle (e.g. 10 deg) from the ground station to the satellite. (Consider rain attenuation later.) I would also add a loss of 0.3 dB to account for polarization mismatch for large ground antennas. Using a radome adds another 1 dB loss.

9. Select the ground station antenna diameter and estimate pointing error. If autotracking is used, let the pointing error be 10% of the beamwidth. Use Eq. (13-21) to calculate the antenna beamwidth.

10. Calculate the receive antenna gain toward the satellite. For FireSat we used antenna efficiency, tj, of 0-55.

11. Estimate the system noise temperature (in clear weather), using Table 13-10.

12. Calculate E)/N0 for the required data rate, using Eq. (13-14).

13. Using Fig. 13-9, look up required to achieve desired BER for the selected modulation and coding technique. The downlink for FireSat is modulated with BPSK and the uplink is BPSK/PM. See Table 13-11.

14. Add 1 to 2 dB to the theoretical value given in Fig. 13-9 for implementation losses.

15. Calculate the link margin—the difference between the expected value of Ei,/N0 calculated and the required Ei/N0 (including implementation loss).

16. Estimate the degradation due to rain, using Fig. 13-11 and Eq. (13-27).

17. Adjust input parameters until the margin is at least 3 dB greater than the estimated value for rain degradation, depending on confidence in the parameter estimates.

For communications satellites to evaluate a complete communication link (ground-to-ground), you must do the downlink shown above, and also calculate the uplink, and combine their Ej/N0sin order to evaluate the communication link.

The downlink calculation described above provides the signal-to-noise at the ground station based on the assumed parameters for the downlink. In order to establish the performance of a communication link Earth-to-Earth, it is necessary to do the same calculation on the uplink from the ground station to the satellite. The overall link performance can then be predicted based on the design of the satellite communication payload. In a bent-pipe satellite, the signal-to-noise ratio established on the uplink is used as the "signal" input for the downlink, and a final signal-to-noise is calculated based on the noise already on the signal plus the noise gained on the downlink. For a signed processing payload (see Sec. 13.5.2), where signals are demodulated and re-modulated on board the satellite, then the overall signal-to-noise performance is only that of the downlink, because this is a "pure" signal generated on the satellite. Many of the new data satellites, and some of the cellular telephone satellites, use onboard processing.

When digital links are evaluated in terms of their bit error rate, the system for signal processing satellites gets more complicated, because there will be a certain bit error rate on the uplink which becomes the starting point for the downlink. The bit error rate will never be better than the weakest link. For these systems, it is desirable to make the uplink very robust so that bit error rates of 10-9 or 10_1° are achieved on the uplink, in order that, again, the downlink determines the overall performance. This is typical for most satellite links, as the satellite is limited in the amount of transmit power, whereas ground stations are relatively independent of that limitation, at least until frequencies of 30 GHz and above are used, in which case the cost of the transmitters becomes a limiting factor. Transmit power, transmit antenna gain, receiver noise figure, and receive antenna gain establish the maximum signal-to-noise that can be established on any link.

Many of the data handling satellites discussed elsewhere in this chapter expect to deliver 10-10 bit error rates on the entire Earth-Earth link. This is being achieved by using very powerful forward-error-correcting codes (see Sec. 13.3.3). Convolutional coding and Viterbi decoding (rate 1/3, K = 7) allow 10-10 bit error rates with Ef/N0 of only 5 dB for many newer, commercial data satellites.

The question often asked is, "How much margin is enough?" Clearly, too much margin is wasteful and costly, but not enough margin could occasionally lead to excessive bit error rates. Intelsat carries a 4 to 5 dB margin for their C-Band links. At frequencies above 10 GHz the margin should be 6 to 20 dB to accommodate atmospheric and rain losses, the exact amount depending on the required link availability and the amount of rainfall expected.

The order of the steps outlined above will depend on which parameters are specified. For example, one might start with link margin and solve for transmitter power. The uplink design is performed in the same way, except the receive antenna beam width may depend on the Earth-coverage requirement rather than size or pointing limitations.

Figure 13-12 illustrates how the downlink design, using a geostationary satellite, can vary with choice of carrier frequency. In this example, the satellite antenna's beam width is fixed at 6 deg to illuminate a specified Earth coverage area, and the ground-station size is fixed at 0.5 m for ease of transport As the frequency decreases, the satellite antenna's diameter increases to maintain the specified beam width (and gain) until it reaches a maximum size (or mass) limit, which, in this example, is 2 m at 1.75 GHz. Reducing the frequency further requires more transmitter power to compensate for the loss in antenna gain [see Eq. (13-18a)]. On the other hand, going to higher frequencies requires more transmitter power to compensate for increasing receive antenna pointing loss and to provide a margin to operate through rain. The figure shows the preferred frequency is between 1 and 18 GHz.

Frequency, f (GHz)

Fig. 13-12. Example Downlink Design Showing Effect of Frequency Selection on Required Satellite Transmitter Power. A "window" exists between 1-20 GHz. The satellite Is in a geostationary orbit and the ground terminal diameter Is fixed at 0.5 m.

Frequency, f (GHz)

Fig. 13-12. Example Downlink Design Showing Effect of Frequency Selection on Required Satellite Transmitter Power. A "window" exists between 1-20 GHz. The satellite Is in a geostationary orbit and the ground terminal diameter Is fixed at 0.5 m.

Table 13-13 shows we can satisfy the FireSat mission with a 20-W transmitter operating at S-Band (2.2 GHz) with a broad-beam antenna covering the entire Earth. A higher-gain antenna requires continuous steering to point toward the ground station, making the satellite far more complex. The diameter of the ground station antenna is 5.3 m.

0 0

Post a comment