WO2001065726A1 - Satellite communications with beams directed to user positions - Google Patents

Abstract

A satellite system comprising at least one satellite (4); at least one Earth station (6), and a plurality of user terminals (2), the satellite (4) being arranged to provide a link between each user terminal (2) and the Earth station (6), via a plurality of user terminal link beams (B1-BN) carrying communications channels, there being fewer independent said channels than the maximum number of beams which said satellite (4) can generate, said channels being reused between said beams (B1-BN); the system comprising means for directing said beams to positions determined in dependence upon the positions of said user terminals (2), and means for allocating said channels to said beams to reduce co-channel interference between said beams at said user terminals (2).

Description

SATELLITE COMMUNICATIONS WITH BEAMS DIRECTED TO USER POSITIONS

This invention relates to satellite communications, and particularly to

multipoint satellite communications, such as mobile communications

Satellite mobile communications systems are well known. In recent

years, a number of certain systems have been proposed, including the recently

launched Iridium system, and the proposed GlobalStar and ICO systems,

which are intended for communications with small mobile terminals such as

handsets.

One problem in such communications is to achieve sufficient link

margin, for which purpose high gain antennas are used in both the user

downlink and user uplink directions. To achieve this whilst covering a large

area of the Earth visible to the satellite (which is necessary in order to reduce

the number of satellites), an array of narrow beams are generated as disclosed

in. for example, WO 95/28747. It is then also possible to re-use frequencies

between non-neighbouring beams, as in terrestrial cellular communications.

A regular array of beams is used, and frequencies are allocated in a repeating

pattern.

In some proposals (e.g. as in EP 0575678, EP 0610789. and the

proposed Odyssey system of TRW), the grid of beams is steered to centre it

on anticipated hotspots such as cities, or to concentrate cover on landmasses. In older satellite for example for military use in

communication with na\al ships, a few separate steered beams were provided

one to each ship the ships havmg steered antennas However, in such

situation-, the bandwidth, and gam constraints were much less onerous than in

modern cellular^" satellite s> stems, which have many more terminals each

with a lower gain antenna

L S 5754P9 discloses a satellite communications system in which

more beams are allocated to areas with higher anticipated demand, and there

is some provision tor individual beams for some users, which beams track the

user position on the Earth as the satellite moves in orbit

As use of information technology increases, there is an increasing

demand tor bandwidth which is, however, a scarce resource for satellite

systems since the} must avoid conflict with any terrestrial usages in many

different countnes

The present inv ention is intended to provide a bandwidth-efficient

satellite communications system, particularly for relatively low power mobile

terminals

This is achieved by determining user positions, providing multiple

independent beams one for each user, which are steered to ground positions

based on the user positions, and selecting the beam frequencies to satisfy

predetermined re-use constraints, based on the beam positions In a preferred embodiment, the ground positions are selected to reduce

interference between beams, whilst being relatively close to the user positions

to maintain gain.

Other aspects and preferred embodiments of the invention, together

with corresponding advantages, will be apparent from the following

description, drawings and claims.

Embodiments of the invention will now be illustrated, by way of

example only, with reference to the accompanying drawings in which:

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described, by way of

example only, with reference to the accompanying drawings, in which:

Figure 1 is a block diagram showing schematically the elements of a

communications system embodying the present invention;

Figure 2 is a block diagram showing schematically the elements of an

Earth station node forming part of the embodiment of Figure 1 ;

Figure 3 illustrates schematically the disposition of satellites forming

part of Figure 1 in orbits around the Earth;

Figure 4 illustrates schematically the beams produced by a satellite in

the embodiment of Figure 1 ;

Figure 5 is a cross section through a beam, showing gain against

angular displacement from the antenna boresight axis; Figure 6 is a block diagram illustrating schematically the structure of

the payload of a satellite transponder:

Figure 7 shows schematically the content of a database forming part of

the satellite payload of Figure 6:

Figure 8 is a flow diagram showing the operation of the satellite

pa> load of Figure 6;

Figure 9 is a diagram illustrating the contents of a database forming

part of the Earth station of Figure 2;

Figure 10 is a flow diagram ^"showing part of the operation of the Earth

station of Figure 2: and

Figure 1 1 (to be read with Figure 10) shows the remaining part of the

operation at the Earth station in a first embodiment;

Figure 12 (corresponding to Figure 4) shows the effect of altering the

beam aim point according to the process of Figure 1 1 ;

Figure 13 (comprising Figures 13a and 13b) is a flow diagram

show ing the operation of the Eaπh station according to a second embodiment

of the invention; and

Figure 14 is a diagram illustrating the contents of a database forming

pan of the Eaπh station of Figure 2.

GENERAL ASPECTS OF EMBODIMENTS

Referring to Figure 1 , a satellite communications network according to

this embodiment comprises mobile user terminal equipment 2a, 2b (e.g. handsets 2a and 2b); orbiting relay satellites 4a. 4b; satellite Eaπh station

nodes 6a. 6b; satellite system gateway stations 8a, 8b; terrestrial (e.g. public

switched) telecommunications networks 10a, 10b; and fixed

telecommunications terminal equipment 12a, 12b.

Interconnecting the satellite system gateways 8a. 8b with the Eaπh

station nodes 6a, 6b, and interconnecting the nodes da, 6b with each other, is a

dedicated ground-based network comprising channels 14a, 14b. 14c. The

satellites 4, Eaπh station nodes 6 and lines 14 make up the infrastructure of

the satellite communications network, for communication with the mobile

terminals 2, and accessible through the gateway stations 8.

A terminal location database station 15 (equivalent to a GSM HLR) is

connected, via a signalling link 60 (e.g. within the channels 14 of the

dedicated network), to the gateway station and Eaπh stations 6.

The PSTNs 10a. 10b comprise, typically, local exchanges 16a, 16b to

which the fixed terminal equipment 12a. 12b is connected via local loops 18a,

18b; and international switching centres 20a, 20b connectable one to another

via transnational links 21 (for example, satellite links or subsea optical fibre

cable links). The PSTNs 10a, 10b and fixed terminal equipment 12a, 12b

(e.g. telephone instruments) are well known and almost universally available

today.

For voice communications, each mobile terminal apparatus is in

communication with a satellite 4 via a full duplex channel (in this embodiment) comprising a downlink channel and an uplink channel, for

example (in each case) a TDMA time slot on a paHicular frequency allocated

on initiation of a call, as disclosed in UK patent applications GB 2288913

and GB 2293725. The satellites 4 in this embodiment are non geostationary,

and thus, periodically, there is handover of each user from one satellite 4 to

another.

Terminal 2

The user terminals (UT's) 2a. 2b may be similar to those presently

available for use with the GSM system, comprising a digital low rate

coder/decoder, together with conventional microphone, loudspeaker, battery

and keypad components, and a radio frequency (RF) interface and antenna

suitable for satellite communications.

Each UT 2 comprises an omnidirectional antenna, i.e. an antenna

having generally satisfactory communications performance at all directions

above a certain minimum elevation above the horizon (such as ten degrees) so

as not to require pointing or steering to a satellite.

Small satellite communications terminals are currently available (with

omnidirectional antennas) for the Iridium system from Motorola Inc, and

(with steered antennas) for the Inmarsat-M and mini-M systems, for example.

Terminals may be connected, as shown, to data terminal equipment

160a, 160b such as a facsimile machine or a personal computer.

Earth Station Node 6 The Earth station nodes 6 are arranged for communication with the

satellites.

Each Earth station node 6 comprises, as shown in Figure 2, a

conventional satellite Earth station 22 (functioning somewhat equivalently to

the Base Station of a cellular system) consisting of at least one satellite

tracking antenna 24 arranged to track at least one moving satellite 4, RF

power amplifiers 26a for supplying a signal to the antenna 24, and 26b for

receiving a signal from the antenna 24; and a control unit 28 for storing the

satellite ephemera data, controlling the steering of the antenna 24, and

effecting any control of the satellite 4 that may be required (by signalling via

the antenna 24 to the satellite 4).

The Earth station node 6 further comprises a mobile satellite switching

centre 42 comprising a network switch 44 connected to the trunk links 14

forming part of the dedicated network. It may be, for example, a commercially

available mobile switching centre (MSC) of the type used in digital mobile

cellular radio systems such as GSM systems.

A multiplexer 46 is arranged to receive switched calls from the switch

44 and multiplex them into a composite signal for supply to the amplifier 26

via a low bit-rate voice codec 50. Finally, the Earth station node 6 comprises

a local store 48 storing details of each mobile terminal equipment 2a within

the area served by the satellite 4 with which the node 6 is in communication.

The local store 48 acts to fulfil the functions of a visited location register (VLR) of a GSM system, and may be based on commercially available GSM

products.

Alternatively, satellite control may be provided from a separate control

station.

Other Network Elements

The gateway stations 8a, 8b comprise,, in this embodiment,

commercially available mobile switch centres (MSCs) of the type used in

digital mobile cellular radio systems such as GSM systems. They could

alternat ely comprise a part of an^" international or other exchange forming

one of the PSTNs 10a, 10b operating under software control to interconnect

the networks 10 with the satellite system trunk lines 14.

The gateway stations 8 comprise a switch arranged to interconnect

incoming PSTN lines from the PSTN 10 with dedicated service lines 14

connected to one or more Earth station nodes 6.

The database station 15 comprises a digital data store which contains,

for every subscriber terminal apparatus 2. a record showing the identity (e.g.

the International Mobile Subscriber Identity or IMSI); the service provider

station 8 with which the apparatus is registered (to enable billing and other

data to be collected at a single point) and the currently active Eaπh station

node 6 with which the apparatus 2 is in communication via the satellite 4. Thus, in this embodiment the database station 15 acts to fulfil the

functions of a home location register (HLR) of a GSM system, and may be

based on commercially available GSM products.

Periodically, the Eaπh station nodes measure the delay and Doppler

shift of communications from the terminals 2 and calculate the rough

terrestrial position of the mobile terminal apparatus 2 using the differential

arrival times and/or Doppler shifts in the received signal. The position is then

stored in the database 48.

The Eaπh stations 6 are positioned dispersed about the Eaπh such that

for any orbital position, at least one Earth station 6 is in view of a satellite 4.

Referring to Figure 3, a global coverage constellation of satellites is

provided, consisting of a pair of orbital planes each inclined at 45 degrees to

the equatorial plane, spaced apaπ by 90 degrees around the equatorial plane,

each comprising ten pairs of satellites 4a. 4b, (i.e. a total of 20 operational

satellites) the pairs being evenly spaced in orbit, with a phase interval of zero

degrees between the planes (i.e. a 10/2/0 constellation in Walker notation) at

an altitude of about 10.000 km.

Thus, neglecting blockages, a UT at any position on Eaπh can always

have a communications path to at least one satellite 4 in orbit ("global

coverage").

Satellites 4 The satellites 4 comprise a bus module and a payload module. The

bus module comprises the elements of the satellite which are common to all

satellite applications.

Specifically, the bus module comprises a propulsion system

comprising thrusters for maintaining the satellite in its assigned orbital

position: a power subsystem comprising, for example, a pair of solar power-

wings pointed at the sun and a storage batten' charged from the solar panel

and discharged when the satellite is not in view of the sun; and a thermal

control subsystem to dissipate heat.

Also provided are an attitude control subsystem arranged, in this case,

to direct the body of the satellite towards the Eaπh and the solar cells towards

the sun as described in our earlier application No. GB 2320232; and a

telemetry and command system by which the satellite transmits data

concerning its operating conditions and receives commands from a satellite

control centre causing it to, for example, adjust its position in orbit.

The bus may be, for example, the HS601 or HS601 high power

satellites, or the HS702 satellite, all available from Hughes Space and

Communications Company, in California, US.

Satellite Payload

Each satellite payload generates a plurality of spatially separated user

link radio frequency beams, B 1 -BN in a manner described in more detail below.

Each satellite also has an array of radiation reception directions which intercept the surface of the Eaπh; the reception directions roughly coincide with the

beams. The directions of the beams are at defined stereo angles with the

antenna centre axis or "boresight", which (in this embodiment) is directed

vertically towards the centre of the Eaπh. Each beam is directed towards a

respective user terminal. Thus, as shown in Figure 4. the beams are unevenly

distributed over the satellite footprint - i.e. the portion of the Eaπh visible from

the satellite (or which has visibility of the satellite above some minimum

elevation angle such as 10°).

Any beam which is not centred on the boresight axis will have a non-

circular profile, derived as the intersection of the conical beam with the

spherical surface of the Earth. The sizes and shapes of the beams therefore vary

with their positions (i.e. the position of the beam centre, referred to here as the

'beam aim point') on Eaπh.

The satellite also generates global uplink and downlink beams (e.g. a

beam covering the whole satellite footprint area of the Eaπh) for carrying

signalling traffic for setting up and pulling down calls, and requesting changes

to allocated channel capacity. These may be generated by the same array

antennas or by additional antennas (not shown).

Figure 5 illustrates the beam profile in section. The gain falls away

from a maximum value at the beam centre. Beyond some point (e.g. IdB or

3dB down) the beam may be unsuitable for use; this therefore defines the

"edge" of the beam. However, the beam continues to have an amplitude, and thus to be capable of interfering with other co-channel users, beyond this

"edge".

The satellite payload comprises at least one steerable high gain spot

beam antenna 3 providing a feeder link for communicating with one or more

fixed Earth stations 6 connected to telecommunications networks: a receive

array antenna 1 for receiving the plurality of reception directions Rl-RN; and a

transmit array antenna 200 for generating the plurality of beams B l -BN. The

antennas 1 -3 are provided on the side of the satellite which is maintained facing

the Eaπh.

The transmit and receive antennas each comprise two dimensional array

antennas with, for example, a few hundred elements each.

A brief explanation of the access methods employed will now be given.

The feeder link antenna 3 operates at a transmit frequency of 7 GHz and a

receive frequency of 5 GHz. The receive array antenna operates at a frequency

of 2 GHz and the transmit array antenna at a frequency of 2.2 GHz.

The bandwidth available for each channel is 4 KHz, which is adequate

for speech. Time Division Multiple Access (TDMA) is employed, with 40mS

frames. In the to-mobile direction, there are 36 timeslots in each repeating

frame, on frequency subcarriers each of 150KHz bandwidth. In the from-

mobile direction, there are 6 timeslots in each frame, on frequency subcarriers

each of 25KHz bandwidth. Conveniently, the frequencies allocated to different satellites are such

that no two satellites whose footprints overlap (i.e. who can be seen

simultaneously from any point on the ground) share any common frequencies.

This is conveniently achieved by partitioning the available frequencies

between the two planes of satellites (or, in general. N planes) and then, within

each plane, re-using frequencies only on every alternate satellite (or, where

levels of coverage higher than double coverage are provided by the

constellation, on every Nth satellite).

Referring to Figure 6, the electrical arrangement provided within the

satellite payload comprises a forward link, for communicating from an Earth

station to a terminal, and a return link, for communicating from the terminal to

the Earth station.

The forward link begins at the feeder link antenna 3, the signals from

which are bandpass filtered by respective filters 206a-206d and amplified by

respective low noise amplifiers 207a-207d. The amplified signals are combined

and down-converted to an intermediate frequency (IF) by a combiner/IF

downconverter circuit 208. This IF signal is digitised by an analogue to digital

converters (ADCs) 210.

The digitised IF signals are each then frequency-demultiplexed into

separate 150KHz frequency slots by a frequency demultiplexer 211.

Under the control of a digital control circuit 113, a routing network 212

routes each of the frequency slots to one of the input ports of a digital beαmformer 220. which generates a plurality of energising signals for

energising respective radiating elements 200a-200M of the transmit array

antenna 200.

The digital beamformer network comprises a Fast Fourier Transform

processor, which accepts, from the digital control circuit 13, a set of control

parameters for each of the frequency and time channels. The control parameters

comprise:

• The amplitude for the user channel;

• The subcarrier frequency;

• The Doppler shift offset; and

• The angular direction of the beam.

The beamformer is arranged to synthesise beams each in the specified

angular direction with respect to the antenna boresight, at the specified

frequency, with the desired amplitude, by multiplying the signal by the

subcarrier frequency (including Doppler offset) .

The energising signals are each converted to an analogue signal by a

respective digital to analogue converter (DAC) 215a-215N, the outputs of which

are up-convened to a beam frequency lying within a 30 MHz range in the 2.2

GHz band by an array of IF/S band converters, amplified by a bank of M RF

power amplifiers 217a-217M. and bandpass filtered by a bank of filters 218a-

218M, prior to being supplied to the respective radiating elements 200a-200M. The components of the return link are. in general, the reverse of those in

the forward link. A plurality P of receiving elements 1 18a-l 18P receive

incoming radio signals in the 2 GHz band from user terminals 2 on the Earth.

The signal from each element is filtered and amplified by respective filters

1 18a-l 18P and low noise amplifiers 1 17a- 117P. down-converted to a 5 MHz IF

signal by an array 106 of down converters, and digitised by a respective ADC

1 15a-l 15N and fed to the input ports of a digital beamformer 120.

The uplink beamformer 120 is arranged to apply the same direction

control data as the downlink beamformer 220, and amplitude and frequency

offset control data supplied from the digital control circuit 13 (in the latter case,

the Doppler offset is the same, but the frequency channel is different).

The signals at each of the N output ports of the beamformer 120

comprise 25KHz frequency channels each carrying 40mS TDMA frames

divided into 6 timeslots. They are routed, under control of the control circuit

13. through a switch 1 12 to a predetermined input (corresponding to a particular

frequency) of a frequency multiplexer 1 1 1 generating 25 MHz output signals

which are converted to analogue signals by a DAC 110. The analogue signals

are up-converted into 7 GHz signals by an up converter and RF divider network

1 18.

Each RF signal is amplified by an RF power amplifier (e.g. a travelling

wave tube or solid state amplifier device) 107a-107d; filtered by a bandpass filter 106a-106d: and supplied to a feeder link antenna 3 for transmission to a

respective Earth station.

Thus, the system shown will be seen to consist of a feeder link

communication subsystem comprising the elements 3. and 106-109 and 206-

209; a channel separation and combination subsystem comprising the elements

211-214 and 1 1 1-1 14: and a mobile link communication subsystem comprising

the elements 215-218. 115- 1 18. and antennas 100 and 200.

The digital control circuit 13 comprises a store 502 and a digital

processor 504. The store 502 is shown in Figure 7, and comprises static store

502a and a dynamically updated store 502b, each of which has an entry for each

beam. Each beam is associated with a user terminal 2, with which each entry is

therefore also associated. Each entry in the static table 502a comprises fields

storing: the beam number; data defining the position of the beam aim point on

Earth; the channels used (defined as frequency subcarriers of the forward and

reverse link channels for the user the timeslots of the forward and reverse link

channels: and the power for the forward and reverse link channels for the user).

Each entry in the dynamic table 502b comprises the beam number; data

defining the beam direction (relative to the antenna) and the Doppler shift to

apply.

The digital processor 504 connected is to the store 502, and receives

control data from the Earth Station 6. The control data specifies the user terminal positions, and time and frequencies to be used for each, to be written to

the store 502

Referring to Figure 9. the database 48 of the Eaπh station node 6 in

this embodiment, comprises, for each terminal 2. a field defining the terminal

position on Earth (e.g. in latitude and longitude, or as a three dimensional

position rela e to the centre of the Earth ), a beam aim point position (which

will be discussed in greater detail below ) in similar dimensional co-ordinates:

a beam pow er level specifying the pow er transmitted towards the user

terminal; a beam frequency field specifying the

of the beam

transmitted to the terminal (for example

specify ing the frequency channel

used): and a time slot field specifying the time slot used for communication

by the terminal.

Operation of Satellite 4

The satellite 4 payload performs, essential!} , two processing loops: a

first in which new beam control data is received from the Earth station 6. and

a second in which beam directions and Doppler compensations are

periodically re-estimated to maintain direction and frequency accuracy .

Accordingly, as shown in Figure 8. in a step 1002 the satellite 4

determines whether new user beam data is being received from the Earth

station 6 (on a suitable control channel) and. if so. in step 1004. data is

received and written to the store 502 in step 1006. In step 1008. a first beam is selected from those listed in the store and

in step 1010 it is determined whether the current beam is the last beam. If not,

in step 1012, the processor 504 calculates the Doppler shift to the user

terminal from the satellite, utilising the user terminal position data stored in

the table 502a, and the current satellite position (calculated from the satellite

orbital data, or from other sources such as a GPS receiver on the satellite) and

the satellite orbital speed (which is calculated from its orbit and position).

The Doppler shift information is then stored in the store 502b.

Next, in step 1016, the direction in which the beam is to be pointed

(from the satellite) is calculated by reading the beam aim point position data

from the table 502a and using the satellite position data as calculated above.

This too is written to the store 502b in step 1018.

On having processed the last beam (step 1010) in the table 502, the

control circuit 504 amends the router 212, 1 12 to take account of any new

beam assignments from the Earth station 4, and sends the Doppler offset,

direction, and power control data to the beamforming network 220, 120. The

process then returns to step 1002 to detect further uplinked beam data from

the Earth station 6.

The beamformers 220. 120 are operative thereafter to synthesise

transmission and reception beams with the designated power, direction and

frequency, towards the user terminal, allowing data transmission to take the

place in conventional fashion. The process of Figure 8 needs to be repeated on each occasion when

data is received from the Earth station 6, since the beam aim points on the

ground may have changed.

It also needs to be repeated sufficiently frequently to track the

movement of the satellite in orbit; in other words, sufficiently frequently that

the movement of the satellite footprint on the ground (determined by satellite

altitude) in-between successive executions of the process of Figure 8 is small

compared to the width of the beams, so that the gain of the link to the user is

essentially unchanged between repetitions.

Operation of Earth Station 6 in a First Embodiment

Referring to Figures 10 and 1 1 the beam allocation and control

processes performed by the Earth station 6 will now be described.

When a new user wishes to make a call, or when call is to be placed to

a user terminal 2 from a terrestrial terminal 12, after initial paging signalling,

the user terminal position is derived in step 2002 for example, as described in

GB 9919568.7.

In this embodiment, the database 48 includes a section 48a storing, for

each user terminal record, a field indicating the beam aim point (in other

words, the co-ordinates of the point on the Earth where the centre of the beam

serving the user terminal falls); a field indicating the beam power (which is

allocated in accordance with an adaptive power strategy as is well known in

cellular and satellite communications systems, based on measurements of signal quality); fields indicating the beam frequency (or frequencies) and

timeslot (or timeslots), in other words, the channel used by the beam.

In step 2004 of Figure 10, the processor reviews the database 48, and

selects those entries which have beam aim positions close to the user position

of the new terminal to be allocated a beam. The processor then calculates

whether the user position falls within each such beam, by calculating the beam

periphery (i.e. 3dB contour) on the Earth, taking into account the satellite

position. If the user uses the same frequency and time slot as an existing

beam which it is within, there will be strong interference and the allocation of

the same channel to the user therefore cannot be made.

Thus, in step 2006. all channels (i.e. frequency/timeslot pairs) not used

in neighbouring and overlapping beams are selected for further processing.

Where two beams are centred within about two beam widths of each

other, and use the same frequency and time slot, interference is inevitable

since beams do not have abrupt edges. However, even where this criterion is

not met. there may still be substantial interference, since the cumulative effect

of several more distant beams may give a sufficiently high interference level

to cause problems.

Accordingly, in step 2008, the processor reviews the database 48 for

all other terminal records which use the same frequency and time slot.

In step 2010. the processor determines what the effect of the other

channels cumulatively would be on the new terminal, and what the effect of allocating a channel to the new terminal would be on the other channel users.

This is achieved by calculating the amplitude (or power), for each of the co-

channel beams, at each of the user terminal locations, and summing all such

beam amplitudes at each user terminal location. The calculation makes use

of:

• the power allocated to each beam

• the gain of each beam at each user terminal location, based on

• the user terminal position, and

• the satellite position in orbit.

If the sum at each user terminal location is below a predetermined

threshold, in step 2010, then in step 2012, the new frequency and time slot are

allocated to the new user terminal in the table 45, and signalled to the new

user terminal on a signalling channel in step 2013.

If, in step 2010, the new terminal or one of the existing terminals is

found to have an interference level in excess of its threshold, referring to

Figure 1 1 , an attempt is made to resolve this by displacing the aim point of

the new beam away from the location of the new user terminal.

Accordingly, in step 2042, the aim point is displaced randomly from

the user terminal location by a small increment, and in step 2044, it is

calculated whether the gain of the beam when centred on the new position

would be satisfactory at the user terminal position. The processor is arranged to calculate the gain of the beam at the user

terminal position by taking into account the beam shape (i.e. gam profile ) and

the distance from the beam centre of the user terminal position. In this

embodiment, the processor tests whether the gain is in excess of 0.5 B down

from that of the beam centre.

In step 2046. the process of calculating the levels of co-channel

interference at each user terminal (discussed above) is repeated and it is

determined, in step 2048. whether the highest levels of interference at user

terminals have increased or decreased.

If (step 2048) all have decreased, then the processor returns to step

2042 and executes a further displacement of the beam aim point in the same

direction. If some interference levels already - in excess of the acceptable

threshold are rising, then in step 2050 the processor selects a different

direction for displacement and then returns to step 2042.

When (step 2048) the levels of interference are neither rising or

falling, the beam aim point search has reached a local minimum of

interference or. alternatively, a compromise between interference and loss of

gain.

Accordingly, in step 2052, it is determined whether all interference

levels now meet the threshold and. if so. the processor returns to step 2012.

Thus, in this way. the beam aim points (i.e. positions on Eaπh of the

beam centres) can be de-pointed, away from the user terminals to which they are directed, provided that the user terminals still fall w ithin a central region

of the beam guaranteeing reasonable communications link margin, to find

either a local minimum of interference or an acceptable trade-off between ga

and interference level

By comparing Figure 1 with Figure 4. the effect of steering the

beams apart will be apparent, the interference between neighbouring beams is

substantially reduced whilst each beam continues to

er the user terminal

point as shown in Figure 12. rather than being centred on the user terminal

position as in Figure 4

In a preferred form of this embodiment, as shown in Figure 9. for each

user terminal (or. where the user terminal is using more than one channel, for

each channel used by each user terminal) a current total of interference level is

stored, representing the co-channel interference expected at that user terminal

position on the or each channel used by the user terminal

Thus, when a new channel is added, it is merely necessary to calculate

the additional power of the new beam at the position of each user terminal

sharing the same channel, and add this to the stored power lev el at each user

terminal, to assess the effect of adding the new channel, rather than re¬

calculating the power of each beam at each user terminal sharing the same

channel Ti as according to the first embodiment an initial search ol the

av ailable _^nannels is made to exclude tnose channels alreadv used by beams

w hich substantially ov erlapped the new user terminal

Tπ s procedure may be adv antageous w here the number of cnannels is

relativ ely small, and the user 'ermmais are denseh spaced so that

large

numbers or beams ov erlap, in each case it is possible for a substantial fraction

ot the av anable channels to be exhausted in the v icinity of a user terminal and

so the abov e descπbed procedure reduces the number of channels to be

considered runner efficiently and hence impro es the allocation speed

Operation of Earth Station 6 in a Second Embodiment

Under circumstances where the users are more evenly distributed

through the satellite footprint, the second embodiment of tne inv ention to be

descπbed results in more efficient allocation by omitting the abov e described

initial step Instead, available channels are ranked in order of their cuπent

utilisation weighted in accordance w ith the positions of the rjeam aim points

to provide some measure of the av erage lev el interference hkeh to be

experienced on each channel Cnannels are then allocated to new users on the

basis of the lowest lev el of utilisation (and hence a erage interference)

Accordingly , as shown in Figure 14. the database 48 includes a store

48b maintaining, for each channel (I e frequency timeslot combination), an

interference lev el field Referring to Figure 13a. on a new request for channel allocation, in

step 3002. the processor reviews the interference lev el data stored in the store

48b and selects the channel with the lowest interference lev el in step 3004 In

step 3006. using the contents of the store 48a described in relation to Figure 9.

the processor calculates the level of co-channel interference at each of the user

terminal positions using the channel in question and. if in step 3008. it is

determined that the interference at each user terminal position is below an

acceptance threshold then, (shown as step 3010 ) the frequency is allocated to

the requesting user terminal

If not. then in step 3012. it is determined whether the channel is the

last in the list and. if not. the channel from the store 48b is selected which has

the next lowest interference level in step 3014 and the process of Figure 3006

is repeated until the last channel is reached ( step 3012) or an acceptable

channel is located.

If no acceptable channel is found, then, referred to Figure 13b. in step

3016. the channel is selected which produced, at step 3006. the lowest lev els

of co-channel interference Then in step 301 8. the beam de-pomting process

described above in relation to Figure 1 1 is repeated, to locate a set of beam

aim points which result in a local minimum of interference.

In step 3022 it is re-determined whether the interference levels thus

produced fall under the threshold and. if so. the channel is used (shown as step

3024). If not, and if further channels remain (step 3028). then in step 3026. ,_rc„ /65726

26 the channel which gave the next lowest levels of co-channel interference in

step 3006 is selected and the process of step 3020 is repeated. If no suitable

channel can be found then no channel allocation is made after step 3028.

After new allocation of a channel, the levels of interference stored in

the store 48b are updated. The levels are calculated as follows:

For each channel, the power levels of all beams using that channel are

each multiplied by a function of the distance of the beam aim point from the

sub-satellite point (i.e. the centre of the area of visibility of the satellite on

Earth).

The function decreases with increasing distance from the sub-satellite

point and may tamper, for example, from a value of unity at the sub-satellite

point to a value of 0.4 at the edge of the satellite- footprint.

Summing together the powers of the beams using each channel gives

an indication of the average level of the channel across the footprint of the

satellite, and hence, of the level of co-channel interference.

Applying the position-dependent factor takes account of the facts that,

for each beam which crosses the edge of the satellite footprint, part of the

beam power cannot interfere within the footprint; and that the further towards

the edge of the satellite footprint a beam lies, the less likely it is to have

neighbours to interfere with. Thus, ranking channels by the magnitude of the

interference factor thus calculated gives, to a first approximation, the likely level of interference on a given channel and therefore represents a suitable

ordering for testing channels to be allocated.

Although not shown in the above flow diagrams, it will be clear that in

this embodiment and the first embodiment, resources may be allocated not

only on initiation of a new call, but also during a call session, if a user wishes

to switch for example from a voice communication requiring only a single

channel to a high bandwidth data communication requiring multiple channels.

Likewise, user position data may be stored for a number of users to

whom no communications channels are currently allocated, but who remain

periodically and signalling contact with the Earth station node 6. In such

cases, it is not necessary to initially determine position since this will already

be known.

Equally, although not shown in the above flow diagrams, a user

terminal may release channel resources where a call terminated, or drops from

a high bandwidth to a lower bandwidth mode (e.g. data to voice). On the

occurrence of this, the channels previously used are released and the

interference levels maintained in the store 48b are re-calculated.

Summary of Embodiments

It will be seen that the present embodiments enable a more flexible

distribution of the satellite power than hitherto with fixed beam arrays. Some

or all of the beams may be concentrated onto "hotspot" concentrations of

users, since beams are provided on demand to users rather than being provided in a fixed array to all areas of the satellite footprint. At the same

time, the heavy co-channel interference this could otherwise cause is

mitigated by the re-use management methods described herein.

Providing the Doppler correction at the satellite enables the channel

spacing on the feeder link to the Earth station node 6 to be reduced, since it is

not necessary to provide for the possibility of Doppler correction within the

feeder link; accordingly, the channels are closely multiplexed together in the

feeder link on adjacent frequency bands without substantial frequency guard

bands.

Since the satellite is calculating the Doppler compensation to be

applied for each channel, it is also convenient for the satellite to calculate the

beam directions as the satellite moves in orbit. Thus, it is only necessary for

the Earth station node 6 to transmit beam aim points on Earth, on a relatively

infrequent basis, rather than continually uplinking beam steering commands.

This reduces the volume of signalling on the uplink control channels from the

Earth station node 6.

In other respects, however, the satellite is able to act as a transparent

transponder, repeating the signal from the feeder link on to the user terminal

beams and vice versa without needing to demodulate the signals (which

would require substantial on-board processing and could introduce additional

signal delays).

Other Embodiments It will be clear from the foregoing that the above described

embodiment is merely one way of putting the invention into effect. Many

other alternatives will be apparent to the skilled person and are within the

scope of the present invention.

It will be apparent that the first and second embodiments could be

used together. It will also be apparent that it would be possible simply to use

the average power level ranking of the second embodiment as a tool for

channel allocation, without further steps.

It would also be possible to use the allocation methods described in the

first and second embodiments where a fixed grid of beams were provided, as

an alternative to a rigid frequency re-use pattern.

Rather than performing an initial test for overlapping beams, it would

be possible to calculate cumulative interference levels at each user terminal

position from all beams at the outset.

Rather than de-pointing just the beam for the new user terminal to be

added, it would equally be possible de-point existing beams which use the

same channel, or to de-point both.

For certain beam shapes it might be possible to directly calculate a set

of beam aim points which jointly achieved acceptable gains and minimise co-

channel interference between a predetermined set of frequencies and time

slots, rather than using an iterative approach as described. Whilst single beams allocated to each user have been described, it

would be possible (for example, where a large number of users are known to

be at almost exactly the same position on Earth) to provide a single beam

serving multiple users on a single or common frequency, allocating different

time slots to each.

The numbers of satellites and satellite orbits indicated are purely

exemplary. Smaller numbers of geostationary satellites, or satellites in higher

altitude orbits, could be used: or larger numbers of low Earth orbit (LEO)

satellites could be used. Equally, different numbers of satellites in

intermediate orbits could be used.

Although TDMA has been mentioned as suitable access protocol, the

present invention is fully applicable to other access protocols, such as code

division multiple access (CDMA) in which a limited number of codes, or non-

orthogonal codes are re-used or pure frequency division multiple access

(FDMA).

Equally, whilst the principles of the present invention are envisaged

above as being applied to satellite communication systems, the possibility of

the extension of the invention to other communications systems (e.g. digital

terrestrial cellular systems such as GSM) is not excluded.

It will be understood that components of embodiments of the

invention may be located in different jurisdictions or in space. For the

avoidance of doubt, the scope of the protection of the following claims extends to any part of a telecommunications apparatus or system or any

method performed by such a part, which contributes to the performance of the

inventive concept.

Claims

Claims

1. A satellite system comprising at least one satellite (4): at least one

Earth station (6). and a plurality of user terminals (2). the satellite (4) being

arranged to provide a link between each user terminal' (2) and the Earth station

(6). via a plurality of user terminal link beams (Bl-BN) carrying

communications channels, there being fewer independent said channels than

the maximum number of beams which said satellite (4) can generate, said

channels being reused between said beams (Bl-BN); the system comprising

means for directing said beams to positions determined in dependence upon

the positions of said user terminals (2), and means for allocating said channels

to said beams to reduce co-channel interference between said beams at said

user terminals (2).

2. A system according to claim 1. in which said directing means is

arranged to determine said positions jointly in dependence upon the position

of the or each user terminal to which it is directed, and in dependence upon

the co-channel interference between that beam and others using the same

channel.

3. A system according to claim 2, in which said directing means is

arranged to determine said positions jointly in dependence upon:

• the position of the or each user terminal to which it is directed,

• the co-channel interference between that beam and others using the same

channel; and

• the gain of the beam at the user terminal.

4. A system according to claim 1. in which a single said beam is

provided for each said user terminal (2).

5. A system according to claim 1, comprising a plurality of said satellites

covering a region of the Earth.

6. A system according to claim 5, in which said satellites form a non-

geostationary constellation.

7. A system according to claim 6, in which said constellation provides

global coverage.

8. A system according to any of claims 5 to 7, in which the or each

satellite (4) comprises means for maintaining each said beam directed to a

position on Earth as the satellite moves in orbit relative to the Earth.

9 λ sy stem according to any of claims 5 to 7, in which the or each

satellite (4) comprises means for applying a Doppler shift correction to each

said beam.

10 A. s stem according to claim 1, in which said channels comprise

different frequencies

1 1 . A. s stem according to claim 1 or claim 10, in which said channels

comprise different timeslots on a common frequency.

12. A. system according "o claim 1 1, in which the number of timeslots on a

common frequency in channels to a user terminal differs from the number in

channels from a user terminal.

13 A sy stem according to any preceding claim, in which said user

terminals (2) compose handreld terminals.

14 A s stem according to any preceding claim, in which said user

terminals (2) compose terminals with omnidirectional antennas.

15. Channel allocation apparatus for use in the system of any preceding

claim.

16. Apparatus according to claim 15, comprising:

• means for calculating interference level data for each said channel, and

• means for allocating channels in accordance with said interference level

data.

17. Apparatus according to claim 15 or claim 16, said apparatus being

provided at a said Earth station (6).

18. A satellite for use in the system of any of claims 1 to 14.

19. A user terminal for use in the system of any of claims 1 to 14.

20. A method of channel allocation of a plurality of satellite

communications channels carried over beams of a multibeam satellite to user

terminals, there being fewer independent said channels than the maximum

number of beams which a said satellite can generate, said channels being

reused between said beams; the method comprising;

• determining the existing level of co-channel interference on said channels

between said beams, and • allocating a new channel which has a low existing level of co-channel

interference.

21. A method of channel allocation of a plurality of satellite

communications channels carried over beams of a multibeam satellite to user

terminals, there being fewer independent said channels than the maximum

number of beams which a said satellite can generate, said channels being

reused between said beams; the method comprising;

• determining, for each channel, the level of co-channel interference on other

beams which would be produced by the additional use of that channel on a

further beam: and

• allocating a new channel which has a low existing level of co-channel

interference.

22. A method according to claim 20 or claim 21. in which the step of

determining comprises utilising beam power data representing the transmitted

pow er on each said beam

23. A method according to any of claims 20 to 22, in which the step of

determining comprises utilising position data representing the relative

position of each said beam.

24. A method according to any of claims 20 to 23, in which said beams

are individually steered to user terminal positions.

25. A method of satellite communications comprising;

• defining individual user beams from a satellite in non-geostationary orbit,

one for each of a plurality of user terminals; so as to serve all user

terminals through respective individual beams;

• allocating channels to said beams so as to minimise co-channel interference

at said user terminals between beams using the same channel; and

• maintaining said beams pointing to positions related to those of said user

terminals.