High-throughput Satellite or HTS systems represent a new generation of satellite communications systems, that provide a significant increase in capacity than conventional communication satellites. They utilize frequency reuse and multiple spot beams to deliver vast throughput for the same amount of allocated orbital spectrum. The increase in capacity typically ranges from 2 to more than 100 times as much capacity as the conventional fixed, broadcast, and mobile satellite services (FSS, BSS, and MSS). This significantly reduces the cost per bit.
The one fundamental difference in the architecture of an HTS system is the use of multiple ‘spot beams’ to cover a desired service area, rather than wide beams, which bring a two-fold benefit:
- Higher transmit/receive gain: because of its higher directivity and therefore higher gain, a narrower beam results in increased power (both transmitted and received), and therefore enables the use of smaller user terminals and permits the use of higher order modulations, thus achieving a higher rate of data transmission per unit of orbital spectrum. A better link budget allows the use of higher order modulation and coding schemes, resulting in a higher spectral efficiency, increased throughput and thus more cost-effective Mbit/s.
- Frequency reuse: when a desired service area is covered by multiple spot beams, several beams can reuse the same frequency band and polarization, boosting capacity of the satellite system for a given amount of frequency band allocated to the system. In the case of a multibeam satellite, the isolation resulting from antenna directivity can be exploited to re-use the same frequency band in separate beam coverages. The higher the spectral efficiency, the higher the rate of data transmission per unit of orbital spectrum utilized. This is a very important feature because of the congestion of orbital slots as well as the limitations in the spectrum available.
A satellite payload using multibeam coverage must be in a position to interconnect all network
earth stations and consequently must provide interconnection of coverage areas. Depending on the on-board processing capability and the network layer, different techniques are considered for interconnection of coverage:
—transponder hopping (used when there is no on-board processing);
—on-board switching (used when there is transparent and regenerative processing);
On-board connectivity with transponder hopping
The band allocated to the system is divided into as many sub-bands as there are beams. A set
of filters on board the satellite separates the carriers in accordance with the sub-band occupied.
The output of each filter is connected by a transponder to the antenna of the destination beam.
It is necessary to use a number of filters and transponders at least equal to the square of the number of beams. According to the type of coverage, the earth stations must be able to transmit or receive on several frequencies and polarisations in order to hop from one transponder to another.
The capacity offered to traffic can be varied between beams, within the total capacity
defined by the system bandwidth, by modification of the sub-band assignments and hence by
modification of the connections between input filters and transponders. This operation is realised
by telecommand, from time to time, in accordance with long-term fluctuations in traffic.
Beam switching by transponder hopping is a solution when the number of beams is low. Because
the number of transponders increases at least as the square of the number of beams, with a large
number of beams the satellite payload becomes too complex and too heavy
On-board connectivity using Analog switched time division multiple access (SS-TDMA)
It is therefore necessary to consider on-board switching at a lower granularity, and shift from beam switching to channel switching. Two types of technology can provide this kind of connectivity: analogue technology using an intermediate frequency-switching matrix, one example of which is known as satellite switched/TDMA (SS/TDMA), and digital technology using baseband processing equipment, in particular digital transparent processors (DTP).
The principle of analogue transparent switching is employing a payload that includes a programmable switching matrix having a number of inputs and outputs equal to the number
of beams. This matrix connects each uplink beam to each downlink beam by way of a receiver
and a transmitter. The number of repeaters is thus equal to the number of beams.
Multibeam satellites may require rapid reconfiguration (within a few hundred nanoseconds)
of the interconnections between beams. They must, therefore, be equipped with a fast switching device. Fast switching implies the use of:
—switches using active elements; The first switching matrices developed used PIN diodes as the switching elements. PIN diodes were replaced later by field-effect transistors which provide better isolation (60 dB), a shorter switching time (less than 0.1 ns instead of 10–100 ns) and gain (of the order of 15 dB with two stages in cascade); this enables the losses inherent in the architecture to be partially compensated.
—an on-board device to control the switching sequence (a distribution control unit (DCU)).
The distribution control unit (DCU) associated with the switch matrix establishes the sequence of connection states between each input and output during a period of time in such a way that the carriers arriving at the satellite in each beam are routed to the destination beams.
When the period of time separating two connection states is a frame, since interconnection
between two beams is cyclic, stations must store traffic from users and transmit it in the form of
bursts when the required interconnection between beams is realised.
This technique can thus be used in practice only with digital transmission and access of the TDMA type. This is why it is called satellite switched time division multiple access (SS-TDMA). The granularity of the connectivity provided by the satellite-switched technique is a time slot of a high-capacity frequency carrier
On-board connectivity using Digital transparent switching
When connectivity is required at granularity smaller than a channel, analogue technology may not
be efficient, because it leads to an increased payload complexity. Digital technology, relying on
digital filtering and switching, can be introduced. A digital transparent processor can enable the switching of uplink carriers from one spot beam to another spot beam and the transposition of frequency.
The availability on board the satellite of binary digits obtained after demodulation and decoding
offers several opportunities, and in particular, allows the introduction of some layer 2 switching onboard the satellite. The availability of bits onboard the satellite at the output of the uplink carrier demodulators permits switching between receiving and transmitting antennas to be no longer at radio frequency but at baseband.
The constraint of immediate routing of received information to the destination downlink disappears. This permits earth stations to transmit all their information in the same burst and hence to transmit only a single burst per frame. The number of bursts per frame is reduced and the efficiency of the frame increases.
A change of rate between the uplink and the downlink is not possible with a transparent
satellite. Stations can, therefore, be interconnected only by carriers of the same capacity and this
can be restricting. In contrast, by virtue of on-board processing, the traffic between networks with different data rates can be switched at baseband and combined before transmission on the various downlinks in accordance with their destination and independent of the capacity of the carrier.
To avoid bulk and excessive power consumption, only high-rate carriers which contain traffic destined for low-rate stations are routed to a high-rate demodulator. The other high-rate carriers are switched at radio frequency and are, therefore, not demodulated.
Earth station EIRP reduction
The uplink is often over-dimensioned so that the performance in terms of C/N0 of the total link
is determined by that of the downlink, which is limited by the power available on board the
satellite. With a transparent repeater, it is necessary to provide a ratio of typically 10 dB between uplink to downlink carriers.
With a regenerative repeater, the error probability on the uplink becomes negligible in comparison with that of the downlink when the value of the ratio a is greater than around 2 dB. This reduction of the ratio a translates into a reduction in the EIRP of the earth station and, consequently, its cost.
Each coverage area is illuminated cyclically by an antenna beam whose orientation is controlled by a beam-forming network which is part of the antenna subsystem on board the satellite. The area stations transmit or receive their bursts when the area is illuminated by a beam. Interconnection by beam scanning can be considered both with transparent and regenerative payloads.
Scanning beams with transparent payload: In the absence of on-board storage, at least two beams are necessary at a given instant—one to establish the uplink and one to establish the downlink. The illumination duration is proportional to the volume of traffic to be carried between the two areas.
Scanning beams with regenerative payload: Dynamic real-time forming of antenna beams permits consideration of single-beam satellites with a beam which sequentially scans the various regions of the service zone . The set of dwell areas which are covered sequentially by the beam form the coverage area of the system. When the beam is in a given dwell area, the information destined for stations in the area is extracted from the on-board memory and transmitted in multiplexed form.
Simultaneously, these stations transmit information destined for stations in other dwell areas. This is stored in the on-board memory for later transmission at the time when the beam passes over the destination area. An inherent advantage of this type of system is the disappearance of fixed simultaneous beams and hence of co-channel interference (CCI).
Connectivity using Intersatellite links (ISL)
Intersatellite links (ISL) can be considered as particular beams of multibeam satellites; the beams in this case are directed not towards the earth but towards other satellites. For bidirectional
communication between satellites, two beams are necessary—one for transmission and one for
reception. Network connectivity implies the possibility of interconnecting beams dedicated to
intersatellite links and other links at the payload level.
Three classes of intersatellite link can be distinguished:
—links between geostationary earth orbit (GEO) and low earth orbit (LEO) satellites (GEO–LEO
links); also called inter-orbital links (IOL);
—links between geostationary satellites (GEO–GEO);
—links between low orbit satellites (LEO–LEO).
Links between geostationary and low earth orbit satellites (GEO–LEO)
This type of link serves to establish a permanent relay via a geostationary satellite between one or
more earth stations and a group of satellites proceeding in a low earth orbit at an altitude of the order of 500 to 1000 km.
For economic and political reasons, one does not wish to install a network of stations which is so large that at every instant the passing LEO satellites are visible from at least one station. One or more geostationary satellites are therefore used; they are permanently and simultaneously visible both from stations and low earth orbit satellites and serve to relay communications.
This technique also permits overcoming possible limitations of the terrestrial network. This concept is presently operated in the NASA tracking network by means of the tracking and
data relay satellites (TDRS) which, in particular, provide communication with the International
Space Station. A European programme has successfully launched a data relay payload (ARTEMIS
satellite) to provide communications between the ground and low earth orbit spacecrafts.