Part IV 51.4.4-51.5( Satellite antennas and footprints)

51.4.4 Satellite antennas and footprints

Acircular beam 17.5° wide is just sufficient to cover the whole disc of the Earth visible from a geostationary satellite. Such a beam has a beam edge gain of about 16dBi and it is generally provided by a horn antenna.

For most purposes, coverage of the whole visible Earth is not necessary, in which case it is also not desirable. Antennas with higher gain which, nevertheless, cover the required service area can provide much greater transmission capacity for given transponder power than global coverage antennas. Furthermore, frequency co­ordination of such satellites with others operating in the same frequency band is made easier, and the possibility may arise for frequency re-use within the satellite network if beams do not over­lap.

The basic high gain antenna is an offset fed reflector, generating an approximately circular beam. The beam edge gain of such an antenna is given approximately by Equation 51.1 where D is the half power beamwidth in degrees.

However, it is so important to obtain the highest feasible gain from a satellite antenna that it is usual to adopt a more complex antenna design, producing a pattern of illumination on the Earth's surface (called the 'footprint') which matches closely the geographical area which is to be served. Thus, for example, INTELSAT V has six antennas for access to and from the transponders and two of them, optimised to generate beams from a mid-Atlantic orbital location which serve areas on both sides of the Atlantic Ocean where large amounts of traffic originate, take the form of front fed reflectors with an array of 88 carefully phased feed horns at their offset foci. (See Figure 51.10.) The corresponding feed horn arrays on INTEL­SAT VI antennas have 146 elements.

Figure 51.10 Approximate aerial beam coverage of INTELSAT V spacecraft for the Atlantic Ocean region

51.4.5 Modulation techniques

Up to the present time the choice of modulation technique for use in satellite communication has been greatly influenced by the cost of carrier power reaching the receiving antenna. This cost is tending to fall, in particular as satellites with higher antenna gain come into use, but another limitation on down-link power levels is likely to remain as long as most spectrum allocated for space services is shared with terrestrial radio services which have equal allocation status. Despite the low receiver noise levels obtainable with even low cost earth station receivers, modulation techniques must be suitable for operation at relatively low carrier to noise ratios. Am­plitude modulation is never used for analogue signals and the use of high order phase shift and hybrid modulation for digital signals is rare. However, modulation techniques which tolerate lower pre-de-modulator carrier to noise ratios (C/N) tend to need wider band­width for a given information capacity. Thus the modulation parameters should be optimised for each situation, to ensure that the best use is made of transponder capacity.

Frequency modulation is most commonly used for signals which are radiated in analogue form and a relatively high index of modu­lation is typical. Ideally, the index of modulation, and therefore the bandwidth occupied before demodulation, and the carrier power level are chosen so that:

1. The threshold of the demodulator under clear sky conditions (that is, in the absence of signal absorption in the troposphere) will be exceeded by a few decibels (related to the required rain and implementation margins) and the necessary post demodu­lator signal to noise ratio (S/N) will be attained for the specified proportion of the time.

2. The power and bandwidth available from the transponder for the carrier, and any other carriers that the transponder may be relaying, will be occupied when the transponder is fully loaded.

The choice involves consideration of many of the characteristics of the satellite network, the most important of which is probably the figure of merit (G/T) of the antenna and receiver combinations of the various earth stations involved.

Wide deviation FM is widely used for analogue television signals and frequency division multiplex (FDM) multi-channel telephone baseband aggregates. When used for single speech channels, a valuable saving in power can be obtained by suppressing the carrier when the speaker is silent. The concentration of spectral energy in the neighbourhood of the carrier frequency of an FM signal may necessitate the application of a carrier energy dispersal waveform to the baseband of a television emission or a high capacity FDM telephone emission, to meet the PFD constraint referred to earlier, if the down-link frequency allocation is shared with terrestrial radio services.

For digital signals, phase shift keying (PSK) is most commonly used, 2-phase or 4-phase. For 2-phase PSK a clear sky C/N ratio of 8.4dB, plus a small rain and implementation margin, is sufficient and for 4-phase PSK the ratio should be 3dB larger. However, some form of forward error correction (FEC) is often used, especially where the G/T of the earth station receivers is low and this permits satisfactory operation with a significantly lower C/N ratio. Such emissions carry all kinds of digital signals, ranging from single speech channels, through time division multiplex (TDM) multi­channel telephone aggregates and data systems of a wide range of information rates to digital television signals, although the last mentioned are usually subjected to some form of bit rate reduction video signal processing.

Wide band digital emissions may exhibit strong spectral lines under idle circuit conditions, and it may be necessary to add to the modulating signal at the transmitting earth station a pseudo- random sequence, to be subtracted at the receiving earth station, to disperse these lines if the down-link frequency allocation is shared with terrestrial radio services. When the G/T of receiving earth stations is very small, as it may be with some very small aperture terminal (VSAT) networks, it may be preferable to disperse the spectral energy of the carrier by using frequency hopping spread spectrum modulation.