Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02213770 1997-08-2~
W O 96~1016 PCT/GB9S/00718
I\~lETHOD AND APPARATUS FOR LlMiTlNG INTERFERENCE
BETWEEN SATELLITE SYSTEMS
TECHNICAL FIELD
The present invention relates generally to satellite communication systems,
5 and particularly to methods and systems for limiting signal interference
between satellite communication systems.
BACKGROUND ART
The recent proliferation of satellite communication systems has increased the
likelihoocl of interference between signals associated with neighboring
10 satellites. Such interference can take place, for example, when a non-
geostationary satellite comes within the field of view of a geostatior.ary
satellite. As is well kncwn, geostationary satellites remain fixed in eauatcrialorbits over particular locations on the surface of the earth. Sir.ce
geostationary satellites ordinarily exhibit some minor variation in latitude
15 relative to the equatori_l arc, there exists a narrow "geostationary bar.d"
centered about the equatorial arc corresponding to the set of orbital locations
potentially occupied by geostationary satellites. Unlike geostationary
satellites, the orbits of non-geostationary satellites continuously vary \,vith
respect to the earth's surface. Non-geostationary satellites typically traverse
20 low and medium altitude orbits below the geostationary band.
Signal interference between geostationary and non-geostationary
communication systems can result when non-geostationary satellites move
into the l'ield of view of ground stations oriented toward a particular satellite
within the geostationary band. The potential for such interference arises
25 whenever a non-geostationary satellite becomes located proximate the feeder
link path between a geostationary satellite and one of its ground stations,
hereinah:er referred to as GSY ground stations. Such interference can occur
because non-geostationary satellite systems are generally allocated, on a
secondary basis, the same feeder link frequency bands primarily earmarked
CA 02213770 1997-08-2~
.~, .
to geostationary systems.
Consequently, it is incumbent upon the
operators of non-geostationary systems to avoid disrupting communication
within geostationary systems. Although it is conceivable that the feeder link
5 band could be shared by geostationary and non-geostationary systems, the
frequency separation required between the channels allocated to each system
in order to ensure acceptable interference levels would make this approach
unfeasible under most circumstances.
Since geostationary satellites are distributed throughout the geostatiorlary
10 band above the surface of the equator, the points on the surface of the earthin appr oximate alignment with the geostationary band and a non-
geostationary satellite form a range of "in-line" latitudes across the earth's
surface The position of this terrestrial in-line latitude range will vary with
changes in the latitude of the non-geostationary satellil~. Yet non-
15 geostationary satellites may interfere with geostationary systems even whennot so aligned between a geostationary satellite and a GSY ground station,
since the anl:enna of the GSY ground station projects a radiation pattern
across c3 finite discrimination angle relative to its beam axis. Accordingly, ithas generally been necessary for non-geosynchronous satellites to cease
20 signal transmission when in orbit above GSY ground stations in the vicinity
of this in-line latitude range. This restriction on transmission range has
hindered the performance of non-geostationary satellite systems coordinated
in frequency with geostationary systems.
One Wcly of minimizing interference between satellite systems would be
25 simply to operate one system over frequency bands not aiready allocated to
the other systems. Unfortunately, the limited frequency spectrum available
for sat~ellite communication systems renders this solution untenable.
Moveover, well-established technology is available for implementing
communications eqùipment designed to process signals over the frequency
30 bands primarily allocated to geosynchronous systems.
~;?~ S~T
CA 02213770 1997-08-2~
WO 96/31016 pcrlGBs5loo7l8
- 3 -
While most satellite systems have in the past used geostationary or
geosynchrono~Js satellites, the future development of systems using low and~
medium earth orbits is likely to increase the problem of interference, not only
between 3eost~tionary and non-geostationary systems, but also between two
or more non-geostalionary systems.
US Patent No. 5,227,802 discloses a method of controlling the amount of
overlap betwei~n cells projected by different satellites, by turning off cells
from one satellite when they overlap by more than a predetermined amount
with cells from another satellite.
10 DISCLOSURE OF THE INVENT10~1
Accordinlg to one aspect of the present invention, there is provided a method
of reducing interference between transmissions from a first satellite and from
one or more second satellites located within one or more orbital locations,
comprisins: determining a forbidden area of the surface of the earth within
15 which said first satellite and each of said orbital localions are separated by
less than a predetermined minimum discrimination angle, and inhibiting
transmission by the first satellite to said foroidden area.
According to another aspect of the present invention, there is provided a
method of reducing interference in a link via a first satellite from
20 transmissions from one or more ground stations to one or more second
satellites located within one or more orbital locations, comprising: determininga forbidden area of the surface of the earth within which said first satellite
and each of said orbital locations are separated by less then a predetermined
minimum discrimination angle, and inhibiting reception via the first satellite
Z5 from saicl forbidden area.
According to another aspect of the present inve~ntion, there is provided
apparatus for reducing interference between transmissions from a first
satellite and from one or more second satellites located within one or more
CA 02213770 1997-08-2~
WO 96131016 PCrIGB~5/00718
orbital locations, comprising: means operabte to determine a forbidden area
of the surface of the earth within which said first satellite and each of said
orbital locations are separated by less than a predetermined minimum
discrimination angle, and means operable to inhibit transmission by the first
5 satellite to saicl forbidden area.
Accordin!~ to another aspect of the present invention, there is provided
apparatus for reducing interference in a link via a first satellite from
transmissions from one or more ground stations to one or more second
satellites located within one or more orbital locations, comprising: means
10 operable to determine a forbidden area of the surface of the earth within
which said firsl: satellite at said orbital locations are separated by less than a
predetermined minimum discrimination angle and means operable to inhibit
reception via the first satellite from said forbidden area.
An advantage of the present invention is that the firs~ satellite is only
15 inhibited from transmitting to or receiving from those areas in which
unacceptable interference will occur, insteaa of inhibiting
transmission/reception whenever the beams of the first sateliite and the
second satellite or satellites overlap. In this way, disruption of the service
provided by the first satellite is reduced.
20 Communications with ground stations in the forbidden area may be handed
over to a third satellite which is able to communicate with the ground
stations without interference.
Alternatively, communications between the first satellite and ground stations
within the~ forbidden area may be handed over to ground stations outside the
25 forbidden area. Where the ground stations provide alternative gateways into
a ground network, communications with the ground network may thereby be
maintained .
CA 02213770 1997-08-2~
~VO 96~1016 PCT/GB95/00718
Communicatiol~ with ground stations in the forbidden area may be inhibited
by adjusling the antenna beam pattern of the first satellite, preferably by
inhibiting spot beams incident on the forbidden area. This provides an
advantageous method of reducing interference, since the carrier frequencies
5 and/or referenc:e signais of the antenna beam pattern are also inhibited in the
forbidden area
The predetermined discrimination angle may correspond to a predetermined
thresholcl of said interference so as to ensure that a desired threshold of
interference is no~ exceeded. The forbidden area may be determined by
10 obtaininsl the present instantaneous position of each second satellite, whichis compllex and requires that information on the position of each second
satellite is availabl,e, but minimizes Ihe size of the forbidden area.
Alternatively, ~he forbidden area rnay be determined as the area in which the
first satellite is seDarated by less than the minimum discrimination angle from
15 any location in an orbital band which encompasses all possible positions of
the second satellite or satellites. An orbital band is easier to determine ~han
the posi~;ions of satellites within the orbitai band, although this approacn
enlarges the forbidden area
The or e!ach second satellite may be geosynchronous or geostationary, in
20 which case the position of the or each second satellite is comparatively easyto deterrnine. In particular, the position of the geostationary orbital banc is
particularly easy to determine, and the majority of existing satellites are
located ~Nithin it.
Functions of ~he present invention may be controlled from the ground and
25 may be implernented to a greater or lesser extent within the satellite.
BRIEF DESCRIPTION OF THE DRAWINGS
Specific embodiments of the invention will now be described with reference
CA 02213770 1997-08-2~
WO 96~1016 PCT/GBg5/00718
~, ,
to the accompanying drawings, in which:
Figure 1 show~ a geostationary satellite and a non-geostatiorlary satellite in
orbit above the earth's surface.
Figure 2 illustratively represents the non-geostationary orbital path followed
5 by a conventionally-equipped non-geostationary satellite relative to a
geostationary satellite occupying an orbital location in the geostationary band
above the equator.
Figure 3 shows a non-geostationary satellite NG' equipped with a forbidden
band antenna t,ontrol system traversing a non-geostationary orbital trajectory
1 0 NG0'.
Figure 4 depicts the relationship of a non-geostationary satellite relative tO ageostationary satellite and to the earth
Figure 5 shows a block diagram of a forbidden band antenna contro'l systern
of the presen1: invention designed for inclusion within a non-geostationary
1 5 satellite.
Figure 6 is a cliagram depicting various geometrical relationships between a
geostationary satellite G and a non-geostationary satellite I relative to the
earth .
Figure 7,a shows an exact representation and an approximation of a forbidden
20 band within the field of view of a non-geostationary satellite.
Figure 7b shows a projection of a forbidden band in the fieLd of view of a
non-geostationary satellite located at a particular orbital position.
Figure ~la includes a representation on the earth's surface of the horizontal
CA 02213770 1997-08-25
W O 96~1016 PCT/GB9~100718
approximation of the forbidden band shown in Figure 7a.
Figure 8b n,hows a representation of a set of forbidden bands on the surface
of the earth corresponding to various non-geostationary orbital positions.
Figure 9 shows a situation in which calls are handed over to a different
5 ground station, to avoid the forbidden band.
Figure 10 shows a situation in which calls are handed over to a different non-
geostationary satellite to avoid the forbidden band.
MODES FOR CARRYING OUT THE INVENTION
Introduction
10 Referring to Figure 1, there is shown a geostatlonary satellite 10 and a non- geostationary satellite 20 in orbit above the earth's surface 30 I he
geostationary orbital trajectory of the satellite 10 passes perpendicularly
through the plane of Figure 1, while the non-geostationary satellite 20 may
be assume!d
CA 02213770 1997-08-25
W O 96~1~16 PCT/GB95100~18
- 8 -
to be in a polar orbit transvers2 to the geostationary band. The geostationary
sate!lite 10 occupies a location wrthin the geostationary band above the e~uatora. and henc~ remains in a fixed positicn relative to ground station 40 In the
system shown in Figure 1 the geostationary s~tellite 10 provides a
5 communication link to the ground station 40. The geostationary sateilite 10
communicates with the ground station 40 via antenna 60 over channsls
included within a feeder link band that may be simultaneously utilized by the
non-geos~ationary satellite 20.
10 As may 5e appreciated by reference to Figure 1, the potential for interference
between the non-geostationary satellite 20 an~ the seostationary satellite 10
arises when the non-geostationary satellite 20 becomes located proxirnate the
feeder link ~ath (F;:) between the geost~ticnary satellite 10 and the grounc
station ~ C . In ac~ordance with the inventicn, sisnals transmitted by the
15 non-gecs;aticr,ary satellite 20 are prevented from interrering, beyond a
predefineo extent, with signal transmission 5etween the geostationary satellite
10 and sround station 40 by modifying the antenna beam pattern 3 radiated by
the non-secstationary satellite 20. As is ~escribed mcre fully below, this
modification involves nulling the portion of the beam pattern incident cn z
20 forbidden bancl~ of latitudes (not shown to scale in F;sure 1) on the surface ¢f
the earth. The forbidden latitude range corresponds to the set of locations fromwhich the ansular separation, i.e., topocentric an~le, between satellite positions
within the geostationary band and the non-geostationary satellite is such that
more than a minimum acceptable level of interference would exist between
25 signals from the geostationary and non-geostationary satellites 10 and 20 if
beam pa~tem E~ from the non-geostationary satellrte 20 was not altsred. The
topocentric angular separation corresponding to this minimum acceptable level
of interference is derived below, and will be referred to hereinafter as the
minimum discrimination angle Dmjn.
CA 02213770 1997-08-25
WO 96/31016 PCIIGB9SI00718
_ g _
The manner in which the present invention allows a non~e~stationary satellite
to rernain operationai even when in view of ground stations in communication
- with geostationary satellites may ba further appreciated with referQnce to
Fia,ures 2 and 3. Figure 2 depicts a non-geostationary orbital path NGO
followed by a conventionally~quippod non~eostatlonary satellite ~G relative
to an orbital location in the geostationary band, above the equator Q, of a
geostation;ary sa.tellite G. When the satellite NG reaches orbital location A, the
topocentric: angular separation between the two satellites from surfacc location- B is equal to the minimum discrimination angle. As the satsllite I~G traverses10 the orbital path NGO between orbital locations A and C, ther~ will t~xist points
on the surface of the earth between locations B and D for which the topocentric
angle ber~een the satellites is less than the minimum discrimination angle Dm,n.This may be seen by orserving that when the non-seostationary satellite NG
cc_~pies crbital positions vetween points _ and F ~here exis. Iocations between
15 points a and D on the suriace of the aarth frcm which ~he two satellites are in
alicnment, i.e., locations at which the tcpt}centric ansle is zero. It is further
o~served that ~or non-gecstationary orbital 'ocations 'oetween positicns A anc
E, and between pcsiticns F and C, the cvrrespondins tcpccentric ant;les at
~cints B and D, respectively, are less than Dm,n.
One way of guaranteeins 'hat all ground stations between surfacP locations B
anc D experienca less than the minimum ac_eptable level of interference would
be to suspend transmission from the satellite NG while it is located between
or~ital positions A and C. rhe infeasibility of this apprcach, however. may be
25 demonstrated by considering the following numerical example. Assuming a
typical minimum discrimination angle D~ n of 3 degrees, a +/- 3 degree latitude
variation of the geostationary satellite orbiting at a heisht of 35,786 km, and a
ncn-geostationary orbit heisht of 1800 km, the non-geostationary satellite wouldneed to be switched off when between latitudes approximately +/~8 degrees
30 from the ec~uatcr. An interruption in communication due to this switch-off could
CA 02213770 1997-08-25
W O 96~016 P~ 9~100718
- 10 -
only be iavoid~d if other satellites having unused feeder link capacity were
available to rolay signals from the non~eostationary satellite to its ground
station. As described in the following section, a non-ge~stationary satellite
designed in aco~ordance with the invention to selectively communicate anly with
5 locations not encompassed by a ~forbidden band~ of latitudes allows for
substantially improved coverage range.
Overview of Forbidden Band ImDlementation
As shown in Figure 3, a non-geostationary satellite NG' equipped with a
10 forbidden band antenna control system is seen to follow a non~eostationary
crbital trajecto~ IGO'. In ac.,ordancs with the invention, the antenna beam
projectecl by the satellite NG' is directed only to those regions on the surfaceof the ealth from which the tcpGcentric angular separaticn between the satelliteNG' and .he gecstationary orbit exceeds the minimum discnmination ansle
15 3mln. In other words, for each latitude position on the orbital trajectcry NGO'
there exist locations on the earth between a forbidden band of latitudes within
: the field of view of the satellite NIG' from whic,h the topocentric angie is less
~han Dmjn. The location of this forbidden band of latitudes will shift as the
satellite NG' ~raverses the orbital trajectory I~IGO'. In keeping with ;ne
20! invention, the ~ntenna beam projected by the satellite NG' is shaped so as 'o
only transmit sisnals to, and receive sisnats from, locations on the surface of
the earth not covered by the forbidden band.
As shown in F:igure 3, the forbidclen band of latitudes corresponding to the
25 indicated orientation of the non~7eostationary satellite NG' relative to a
~eostationary satellite G' extends from latitude L1 to latitude L2 about the
eqùator C~. The forbidden band is positioned such that on the latitudes L1
and L2 the topocentric angle between the satellites is e~uivalent to the
minimurn discrimination angle DT,jn. It follows that the topocentric angle
30 associated with locations within the forbidden band is less than Dmjn, while the
CA 02213770 1997-08-25
WO 96131016 PCIIGB95100718
_ 11 --
topocentric ansular separation between the satellites corresponding to locationsoutsido of the forbidden band is larger than Dm,n. In a preferrad embodiment
of the invention the antenna beam pattem from the satellite NG' is continuously
modified so as not to illuminate the forbidden band corresponding to the
5 instantaneaus latitude of the satellite hlG'. This may be effected by, fcr
example, selectively enersizing a c!uster of spot beams projected by the
antenna of the satellite NG' in ac-ordance ~ith a switching algorKhm. The
switching allgorithm wilî typically be responsiv~ to latitude information rr~ eived
from either the olbit contrcl system of the satellite or a sround station. In this
10 regard a more detaiied description of a forbidden band antenna control system is set forth in a followins section.
C)eterminatilcn cf the ~linir~urn l~iscriminatic~ ~ngle Dm,n
Interference calc~laticns are ?erfcrmec~ by separately censicerins interference
15 be~ween non- ,eostationar~ and seastationary downlink (i.e., satellite to ear~.h)
and uplink (i.e., earth to satellite) frequency bands. I nat is, any interference
between uplink and downlink ban~s is assumed to be cf nomini I magnitude
relative to interference betweer, bands of like type.
20~ In a particular embodiment of the invention the value of the minimum
discnmination angle, and hence the extent of the forbidden band, depends in
part upon the signal to noise ratio required for communications carried by the
feeder link between the geostationary satellite 1 û and ground station 40.
Detemminalion of the impact of signal transmissions from the non-geostationary
2~ satellite 2C upcn this signal to noise ratio will generally require knowled~e of
the frequency and modulation characteristics of the feeder link, as well as of
specific aspec's of the communications hardware incorporated within the
geostationiary and non-geostationary satellites 10 and 20 and within the ground
station 40. In this regard rt will senerally be necessary to be aware of the
30 shape, or type. of the antenna beam ~ nominally projected by the satellite 20,
CA 02213770 1997-08-25
Wo 96131016 PcrIGB95/00718
-- 12 --
the gain of the antennas associated with the earth station 4û and geostationary
satellite 10, and the frequency (e.g., C-E3and. Ku-Band) and carrier
characteristics (e.g., CDMA) of the feeder link.
Derivation of the Topocentric Anqle n:
Fisure 4 is a diagram depic~ing the geometric relationship of a
non-geastationary satellite S' relative to a geostationary satellite S and to the
earth. rne angular separation between the geostationary and
non-gecstationary satellites S and S' as seen from the center of the earth is
10 denoted by a, while the separation between these satellites as seen from the
location of a ground station on the surfacs of the earth is identified as the
~opocentric angle rl. Given a particular value of the angle ~ between the two
satellites, the c ~rresponding topocentric angle Tl will vary in aceordance with the
location cf the ground station on the sur;acP of the ear~h. It is assumed that
15 the worst case interferenc~ situations arise when the tcpocentric ansle is at a
minimurn for a given geccentric ansle a; that is, interference will be most
prcnouncPd when the angular separation between the satellites is at a
minimurn when viewed from a ground staticn. Althoush the distances between
the groLInd station and the satellites vary as a func:icn of ~he latitude of theZQ ground station, such variation is believed to have a negligibls effect on
interferencs levels relative to that arising as a consequence of changes in the
topocentric angle ~.
Referring to l-igure 4, it may be seen that there exis. geocentric angles for
25 which the minimum topocentric angle, i.e., Tlmin, iS larger than zero. This
corresponds to the situation in which the satellites are not aligned from any
location on the surface of the earth. Whether or not the minimum topocentric
angle ~min iS nonzero will depend upon the orbit heights h and h' of the
geostationary and non-geostationary satellites relative to the surface of the
30 earth. In particular, for ori~it heights h and h' resulting in a geocentric angle a
CA 02213770 1997-08-25
wo 96/31016 Pcr/Gsss/0071s
-- 13 --
less than a threshold angle a~in there will exist ground station locations
represented by the vector E for ~hich the minimum topocentric angle ~mi,, is
z~ro. Conv~rsely, for satellite orbit configurations in which a ~ amjn there exist
ground station locations for which the associatèd topocentric angle Tlmm is not
5 zero. The angle ccmjn may be derTved using plano geometry, and is expressed
below as:
= 2rCcos( Ro )-arcc3s( f~~ ) (1)
wher~ Ro is the earth's radius, and h and h' are the respective orbit heights ofthe geostationary and ncn-geostationary satellites S and S' (h>h'). Asain,
instances in whic.'l the geccentric ansle is superior to the limit value of e~uation
10 (1) correspond to situaticns in whic,'l there are no ground s-ation Iccations alisned ~ith the tvvc sate!lites, i.e., ~mln ~ ~
In what follo~vs it is endeavored to detemline an expressicn relating the location
of a ~round station to .he opocPntric angle Tl be~een the satellites S and S'
15 asscciated therewith. It is ?oted that the followins calc ~laticns are referencPc
to the non-gecsta~ionary satellite S', as is indicate~! by the OXYZ c~crdinate
system used in Figure 4.
If ho denotes the earth radius, and as previously mentioned h and h' the ~r~it
20 heights of geosync.llroncus satellite S and non-geosynchronous satellite S'.
respectively, wherein ~ and ~p are the spherical coordinates of grcund static
location E in the C)XYZ coordinate system, one has:
~Ro sin~ ~s~ o+h) sin~~ ' O
E- Ro sin~ sin~ S= 0 S'= ~ (2)
Ro c~s9 ~ ~(Ro+h) c~s~, ~Ro h',
Letting r and r' denote the ranges from ground station location E to the
satellites S and S', respectively, useful vectors are defined by:
CA 02213770 1997-08-25
W O 96~1016 PCTIGB951Q0718
'~ 14 -
~= E~' ~'a' = E~' (3)
so that th~ cosine of the topoce-~tric angle is: .
C~STI = 0.0/' (4)
If one perrn;ts ground station l~cation E to ~move" on the sur~ace of the earth
while fixinq the positions of satellites S and S', the opposite of the differantial
5 of E c~Ln bQ written:
- dE = dn~ + rr~u = dr'O' + r'du' (5)
The calculation of the scalar product of this aquation with the two unit vec:orsleads to:
U~ = dr - ~r~ tsn
J (6)
~~ u = dr - dr~srl
The differential of the cosine of the topocentric angle is then:
d(~srl) = dr- dr'c~s~ + dr'- drcosn
if 3 is fixed, r' does not vary, so that one has:
a(C~S~l) = 1 ar(r-C~ ) (8)
The term in brackets is positive, so that the cosine of the topocentric anqle
follows the variations of r and thus the variations of the square of r, which isequal to:
CA 02213770 1997-08-25
W O 96~1016 PCTIGB95/00718
- 15 -
r2 , Ro2 - (~o ~ h)2 - 2Ro(F~o ~ h) (c~ s - 8in~ sinac~s~) (9)
HencQ, wh~3n only cp varies, the cosine of thQ topocentric angle follows the
variations of Ule cosine of ~p and therefore the topocentric angle, when ~ is
fixed, is minimurrl for ~0.
5 In the rest of th,is ssction the geoc~ ic angle is ~ssumed to be sufficiently low
that there exist ground station posi~ions cor"3sp~nding to ~0 in the field of
view of satl~llite S. This ccndition can b~ writt~n:
( Ro + h) (10)
It should be noted that this condition is not very restrictive since the limit value,
10 for a geostationary satellite S. is 81.3 desrees. The foilcwing notaticns are a~opted for the rest of the section:
~o h , F~o h '
P Ro ~ ( 1 1 )
(P) ( P )
With these notations, the conditions on a can be written:
~ _ ~/ 5 a s ~ (12)
The fact that the! ground station location represented by the vector E must be
in the field of view of both satellites may be expressed using equation (12) as:
9 s ~-a (13)
15 Since the topocentric an~le ~ is the diffsrsnc~ between the elevation of S' and
the elevatiion of S as viewed from ground station location E, it is possible to
represent the angle rt as follows:
CA 02213770 1997-08-25
WO 96131016 PCI/GB9SI00718
-- 16 --
COS~ - 1 Cos~9 + a) - 1 (14)
- ~ sin~ ~ ~ sln(~ + ~) ,
lllus, the topocentric angle Tl decreasas as the angie ~ increases.
Forbidden 1~2nd Antenna Ccntr~l System
5 Referring to F:igure 5, there is shown a block diagram of a forbidcien band
antenna ccntrol system 100 of ths present invention designed to be included
within, for instance, the non-geostationary satellite NG' depicted in Figure 3.
Operation of the control system 100 is coordinatQd by a central processing unit
(CPU) 102. such as a micrcprocessor or the iike, on the basis of instruc ions
10 receivec' from a fo,rbidden-band antenna c~ntrol prosram 1 C~ stored within an
on-boar,~ memory unit 108. The memory unit 108 fur~.her inc!udes ge!leral
purpose memory 112 and a forbidden-band Icok-up table F16.
The look-up table 1 16 c~ntains inforrnation pertaining to ;he pcsi~ion and snape
15 of the lorbid~1en band of locaticns on the surface of the earth. Whiie in
particuk3r embodiments the forbidden band may be approximated by a strip of
surface locations included between a pair of latitude boundaries, more accurate
representations of the forbidden band include latitude boundaries exhibiting a
longituclinal clependence. When the former approximation is employed the
20 look-up table 116 will include pairs of forbidden band latitude boundaries L1and L2 (Fisure 3) indexed as a function of the latitude of the non-geostationarysatellite NG'. It is anticipated that the look-up table would include boundariesof the forbidden latitude corresponding to a set of uniforrnly separated latitudes.
An interpolalion scheme cculd then be used to detemnine forbidden band
25 bounda.ries corresponding to latitudes not included within the look-up table.
CA 02213770 1997-08-25
W O 96~1016 PCT/GB95/00718
- 17 -
In order to approximate the forbidden band using latitude boundaries
independent ot longitude, the entries within the look-up table 116 could be
derivod In a simplified manner, for example, by using the expression for
topocentric anl~le given by equation (14) above. In an initial step using this
apprt~ach the rninimum discrimination angle l~mjn iS determined as described
in the prec~ding section. Following deterrnination of Dmjn, computation of the
top~centric angles corresponding to a selected set of latitudes on the surface
of the ~alrth is generated for each satellite latitude index stored within the
look-up ti~ble 1 16. This step may be perfonrned by, for example, substituting
10 into equation (14) a trial set of latitudes spanning a range believed to
encompass the forbidden latitude band corresponding to the asscciated satellite
latitude irldex. The topcc~ntric ansle corresponding to each latitude within thetrial set is then compared with Dm,n in order to determine the latitude
boundaries L1 and L2 of the for'~idden band. That is, .here will exist a pair cf1_ latitudes L1 and 1 ~ within the trial set sucn that tCr all forbicden latih:ces
therebetween ':he asscc,ated topocentric ansle will be less ~han ~ n ~nis
desc iption of a simplified technique 'or deterrninins the bcl!r;caries of the
forbidden banc! is incluced at this juncture in order to enhance understant~ins
of the control system 100. Ac~ordinsly, a more comFlete discussion cf a
20 method lor deterrninin~ the limits of the forbidden band is detailed in a
subsequent sec~ion.
Peferring to Figure 5, the antenna control system 100 also inc!udes a position
act~uisition subsystem, 20 for supplying to the CPU 102 information relating to
25 the orbit~l location of the non-geostationary satallite. The forbidden band
antenna contrcll prograrn 104 uses the orbit latitude provided by the subsystem
120 to relrieve the boundaries of the forbidden band from the look-up table 116.The CPU 102 then relays the retrieved forbidden band boundaries to an
antenna control subsystern 124. The control subsystem 124 operates in a
30 conventk~nal mannerto configure an antenna beam-forming network 128 used
CA 02213770 1997-08-25
wc) 96131016 Pcr/Gsgsl~o7l8
~ ..
to drive the antenna 1~2 of the non~ostationary satellite. The subsystem 124
controls th,e sha~pe of ~he beam pattem pr~jected by the antenna by determining
which of the feed elements within a feed array 136 are to be ener~ized by the
beam-forrning network 128. Electromagnetic energy is conv~ntionally coupled
from the beam forming networ~< 128 to the fe~d array 136 by waveguide 140,
and is then directed by the feed 136 to a doubly-c~rved shaped reflector 144.
Although Figur~ 5 depicts a pa~ticular systr m for projecting a forbidden-~and
b~am pattem, other means within the scope of the invention may be used to
perfoml this func~.ion.
For example, in an altemate ernbodiment of the antenna control system the
look-up table inc!udes information relating to longitudinal variation in the
pcsitions of ~he forbidden band boundaries. Suc.h more prec,sa forbidden band
representa;ions could eac,h be stored, for example, as a two-dimensional matrix
15 assoc.ated with a particular orbit latitude of the non-gecstationary satellite. A
set of rnathematical expressions from which these two-dimensional forbi~den
~and reF r~sentations could be derived is set forth below. The geometric
parameters included within the following expressions are defined within the
diagram cf Fi~;ure 6, in which are shown various geometrical relationsnips
20 between a geostationary (GSO) satellite G and non-geostationary (non-GSO)
satellite I relative to the earth.
In another implementation of the antenna control system the look-up table will
include c'ata corresponding to amplitude and phase coefficients used in
2~ cn,ntrollirls individual elements within the antenna feed array. Such control data
will generally be transmitted from a ground station and stored within the look-up
table prior l:o initiation of satellite operation. Since the orbit of
non-geostationary satellites is periodic, it is not required to senerate a separate
set of control data based on recurring sets of geometrical satellite position
30 parameters. Although this approach may require ' a laryer memory than
CA 02213770 1997-08-2~
W O 96~1016 PCT/GB95/00~18
- 1g-
implementations involving on-board generation of antenna control signals, it
is simpler ~r~ith regard to processing requirements and the like. Moreover, in
this approach control data corresponding to representations of the forbidden
band exhibiting longitudinal variation could be stored as easily as those in
5 which the shape of the forbidden band is specified simply by a pair of
latitudes.
In another implementation, the look-up table includes data corresponding to
said ampli tude ;and phase coefficients as a function of time, and the non-
geostationary satellite has an accurate clock which can be adjusted or reset
10 by signals from the ground station. In this way, no direct posilion
information need be stored.
While the antenna control system 100 of Figure 5 is designed ~or
incorporation in the non-GS0 satellite, in an alternative embodiment elements
of the con~rol system t 00 are incorporated in a ground-level control station.
15 The control slation includes the posltion acquisition subsystem 120, the C~U
102 and the memory unit 108, and information on the forbidden band
boundaries is transmitted to the antenna control subsystem 124, so that the
control station inhibits communication with the forbidden band.
Geometric,al Context of Forbidden Band Calculations
20 In what follows it is assumed that the position of the non-GS0 satellit~ is
fixed at a given latitude, such that all calculations correspond to a given
instant in the non-GS0 satellite orbital period. This allows longitude to be
determined relative to the non-GS0 satellite. The GS0 satellite considered
can have any longitude and a latitude be-ween -3 and - 3 degrees, a range
25 corresponding to the typical latitude drift of GS0 satellites. In all of the
following calculations the variable of interest corresponds to the location of
the ground station M on the surface of the earth.
The following notations will be used hereinafter:
0, c:enter of the earth:
CA 02213770 1997-08-25
WO 96/31016 PCIIGB95/00718
- 20 -
Ro, radius of the earth;
I, the non-GSO satellite positioned at latitude I and orbit height h';
G, the GSO satellite positions at longitude L, latitude ~1 and orbit
height h wherein longitude is measured relative to the non-GSO satellite;
lo, the point of latitude O and relative longitude O at the same height
as l; and.
M, the instantaneous location of the ground station on the surface of
the earth.
CA 02213770 1997-08-25
W O96~1016 PCT/GB95100718
-- 21 --
A first re~erance fr~me in Figure 6 de~ned relative to the position of the
non-.GSO satellite may be characterized as follows:
Iz is a vector directed from the non-GSO satellite to the cPnter of the
earth;
Iy is a vector in the longitude plane containing 1, is directed
perpendicular to Iz, and points in the north direction; and
Ix is oriented such that Ixyz is orthonormal.
A second fram~ of reference is denoted ~s loXYZ, and is similar to the
reference l:rame Ixyz but uses lo rather than I as an origin. The reférence
frame Ixyz may be obtained by rotatins the reference frame loXYZ thrcush ~n
angle I wit'h respect to the axis OX. The nctations used within the sec-nc
reference frame of Fi~ure 5 are:
OM= Ro= a
Ol = Ro ~ h' = b
OG = Ro~h = c
p b (15)
Gl= d
~G = to~c~n~7c ansl~
1~ The spherical coordinates of ~1 in Ixyz are (r, ~, ~) so that its cartesian
coordinates in l~ yz are ( r sin~ cos(p, r sin~ sin~, r cosa) . A pair of directicns
of interest are given by the following unit vectors:
o= 1 .G'I
G/ (1 6)
V = 1 . ~,~
IM
CA 02213770 1997-08-25
Wo 96131016 PCI/GB95/00718
-- 2 2 --
Determination of Forbidden ~)ire~tions of Transmission from the non-GSO
Satellite
In a pre!ferrecl approach the forbidd~n transmission dir~ctions from the
non-GSC) sate~lite are calculatQd with resp4ct to a set of orbit locations within
~i th~ geostationary band, i.e., potential positions of the GSO satellite. The
forbidden directions derived with respect to oa~-h GSO satellite position may beenvision~d as l~orrning a beam, originating at the non-GSO sateilite, th~ edges
of which illuminate points on thc ~arth's surface from which the topocantric
angles b~tween the non-GSO satellite and the associated GSO satellite are
10 equal to the minimum discrimination anale Dmjn. The following section
describes a method for determining the shape of the for~idden beam in terms
of a set of unit vectors v, wherein each value of v specifies a linear path
between the non-GSO satellite and a Iccation cn the sur,ace of the earth. In
this way ~he contour of the forbidden band of locations on the surface of the
15 ear!h may be cletermined ~ th knowledse of the shape o, ~he fo;r~idden bearn.
~eferrincj to Figure ô, the c-,ordinates of the satellite G ,-nay be ex-~ressed in
terrns of the non-GSO satellite reference Ixyz as follows:
'1 0 0 ' '~ nlccs~ csinLccs~/
C~ O ctx/ sinl ~In,~/ ~ csin~/ca~- c~Lccs~lsln/ (17)
~0 -~nl ccs/, ~-cccsLccs~ -csin,~kW- cccsLccs~lccs/,
so that:
~sL~ss~nlsL~ s~sln~/~sl ~ (18)
c c~sLc~s,~ Icosl + sin~ lsin/ - p
20 By calculating the scalar square of equation (18), one cbtains:
d= cll +p2-2p(c~s~c~s/c3s,~/+sin/sin~/~ (19)
The ranc;e r between a ground station location and the.non-GSO satellite may
be expressed as:
CA 02213770 1997-08-25
W O 96~1016 ~CTJGB95/00718
-- 23 --
r= b~~s~ 2_b2sin2~(20)
F~urthermore,
~ r+ d~ (21)
Ml.h~G Ml.,/(1171+ /~2ld2 ~ r~ +2rdØ9
leads to:
a ~ = - ds~ r2si~ (22)
Combinins1 eq~:cltions (19), (20) and (22), one thus has u .v in fur,c-ion of ~.l~ne two unit vectors siven by equation (16) may also be representec in the xy7
reference as foliows:
t ~ L
C~ ~ 1 ~L~AI~ L !
pl-2p(~L~ ~n~ ~ A/~ hln~.L-p~
(23)
" . ~1~h~
C~
10 so that:
( sinLcos~fsln~
sLsk~ c~sls~n~l)sin~ (24)
cos/c~ sin/sin~l- p)cc~)
yl1, p2_2p(c~sLcos/ccs~ tnl~nAr)
Combined with the other expression of u .v' given in eciuation (22), equation (24)
leads to a reiation between ~ and (p, which can be written:
CA 02213770 1997-08-25
wo 9fil31016 Ptcr/Gsss/0071s
-- 24 --
A(~ )sin~ = C(~) (25)
This resultsi in 0, 1 or 2 values of tp for a given value of 6. For a solution (~,(p)
of the equation, the correspcnding unit vQctor v~ corresponds to the direction of
a ray on the periphery of the ~forbidden beam', i.e., the beam defined by the
values of 1" L, ~l and rl. The calculation of ~ values corresponding to a
sufficient number of ~ values thus yields an approximation of the contour of theforbidden beam under consideration. Superposition of a sufficient number of
for~idden beams, with each forbidden beam corraspondin~ to a particular GSO
satellite loci~tion, enables determinaticn of the forbidden band of locations on10 the surface of the earth associated with the location of the non-GSO satellite.
In a particular implementation the GSO satellite locations c~nsidered are at
spec.fiec values of allowec GSO sateilite drift latitude (e.s., -3 and -3 sesrees).
At eac.~ drift latitude calc~ulations are ~erfcrmed from â ?lurâlity of Icncit~des
relative to the lonsitude of the non-GSO satellite.
~.e~r~sent2~icn ~f the Forbidden 3ant' in ~he ~on-(~SO ~atellite -<eierer,ce
Referrins to Fisure 6, each direction from the non-GSO satellite I is representec
by the projection in plane Ixy of the correspondins unit vector. Each forbidden
direction, defined by a value of 9 and a value of tp, is thus represented in a
20 plane by a point of cartesian coordinates (x - sin~ cost~, y= sin~ sint~). In this
re~resentation format the field of view of the ncn-GSO satellite is circ~.lar.
It is noted that this representation forrnat differs from a satellite reference in
which azimuth and elevation are speci~ied. In these two representations the
2~ directions from the satellite are related by:
fx = a ~s '7sinAz (26)
~y= sinE7
and are thus coincident ~hen small ant~les are involved. In a~imuth/elevatien
CA 02213770 1997-08-25
W O 96~1016 PCTIGB95/~0718
- 25 -
representation, the field of view of the satellite will not be circ~lar unless both
the azimuth and the ;elevation angles are small.
An exact repres~ntation of the forbidden band requires consideration of the
forbidden beams corresponding to all positi~ns wfthin the geostationary band
within view of the non-GSO satellr~e undor c~nsideration. Nonetheless, in
particular implementations it may be sufficient to approximate this exact
representation by a horizcntal latitude band encompassing all t,oints within theexact represent~tion. This approximation may be defined in terrns of two
10 values of elevation within the field of view of the non-GSO satellite. Moreover,
this horizontal approximation of the forbidden band allows a straishtfor~ard
realization of thie antenna t ~2 (Fisure 5) of the ncn-GSO satellite as 2
conventicnai linear feed array.
1~ ~eferrins !~c Figure 7a, an exac~ representaticn anc' -n a~~r~ximatlcn of 2
forbidden band are shown within the field of view (FOV) cf a ncn-~eostaticnar~
satellite. In particular, the exact fcrbidden '~and is shown as ~~e c~ ss-hatc,~eo~
re~ion between the dashed lines while the hcrizcntal z,~Frcxi."aticn is cefined
by the elevations E1 and E2. The particular example of .-isure 7a corresoonds
20 to the situation in which the non-GSO satellite is in crbit at an altitude of 1800
km at a latituce of 25 degrees, and in which the minim~m disc,iminaticn ansle
required for acceptable interference is 7 degrees. This resuits in a forbidden
band approximaticn in which the elevations E1 and --2 are at ,~,.1 cegrees and
41.8 desrees. respectively, relative to the center C1 cf the non-GSO field cf
25 view.
An additional example of a projection of the forbid~en i~and within the field ofview of a non-GSO satellite at an altitude of 1800 '~m at a latitude of 10
degress is provided by Fisure 7b. Specifically, forbidden ands are shown for
30 minimum discrirnination angles of 3 ~ ~ (solid line), 5 degrees (dashed
CA 02213770 1997-08-25
wo 96/31016 PCI/GB9S/0071R
-- 26 --
line), and 7 dsgrecs (dotted line). In the rQpressntation of Figure 3 the
forbidden bands cofrospond to orthographic (sine) projections mapped
- according to the function sin(~), where ~ denotes latitude.
5 F~eDresentation o~ Forbidden B~nd on Sufface of Farth
In this sectiorl the forbidden band within the non-GSO satellite reference
depicted in Figure 7a will be transformed to a corresponding forbidden band on
the suffacs of the earth. The transformation entails calculation of th~ longitude
L~ and latitude l~,~ (with respect to the non-GSO satellite position) of the point
10 M on the earth's surface corrQspondins to a siven direction from the non-GSO
satellite. This calculation requires finding an xyz coordinate representaticn cfthe vector O~ based on the followins ~xpression:
s/"sinL~, = rsin~
~ sinl~, = rsin~ sin~ c~s/- rsinJc~s9 - bsin/ (27)
s/ucssLu= rsin9sin~sin/ rc~ sJ-bc~sl
The values cf IM and L~,~ are then simply derived as func~ions of ~ and (D. rnistrans,omlation from (6,~) to (L~ ) allows the intersection with the ear~.",'s
t5 surface of any ray from the non-GSO satellite to be represented in terms c'
longitude! and latitude upon an earth map. The transfommation is initiated by
considerins (i) the representation of the field of view of the non-GSO satellite:
~ = arcsin(~) (28)
b
and, (ii) lhe two values of elevation (Ei) within the non-GSO reference frame
associated with the boundaries of the horizontal approximation of the forbidden~20 band. The foliowing expression holds at each boundary elevation El:
sin~sin~ = sinEJ (29)
Referrinsl to Figure 8a, there is shown a representation on the earth's surface
CA 02213770 1997-08-2~
Wo 96/31~16 PCI'IGB9~/00718
- 27 -
of the horizontal forbidden band of Figure 7a. Specifically, the field of view
of the non-GSO satellite is denoted in Figure 8a by the solid line FOV', while
the limits of the approximation of the forbidden band are indicated by the
dashed line E1' and E2'. It is noted that the horizontal limited E1 and E2 of
5 the forbi~dden band within the non-GSO satellite reference are transformed to
the curved se~ments Et' and E2' on the earth's surface.
As an a~dditional example, Fiqure 8b shows projections upon the earth's
surface of a set of forbidden bands ~solid lines~ together with the projections
of the associated fields of view (dashed lines). The projections correspond
1 Q to a plurality of latitude positions of a non-GSO satellite in orbit at art altitude
of 1800 km, and assume a minimum discrimination angle of 7 degrees. The
relative longitudin,al positions of the forbidden bands were varied ir~ order tomir iml~ mut~lal sup~rQQsi~ion
Alternati~e Im~lementations
15 Referring again to Figure 5, in other embodiments of the antenna control
system 100 the boundaries of the forbidden band could oe determined in reai-
time rather than using a look-up table 116. For example, latitude and
longitude information from the position acquisition subsystem 120 (Figure 5)
could be used by a microprocessor or the like on-board the non-geostationary
20 satellite to determine the shape of the forbidden band in accordance with the
analytical expressions set forth in the above equations. In another
implementation such real-time calculations would be performed at a ground
station and transmitted to the non-geostationary satellite. In each of these
implementatiolls the control system 100 would then operate as descri~e-'
25 above tc project an antenna beam only to those regions out,ide of the
forbidden band.
The calculation of the forbidden beam is used above to derive the contour of
the forbidden band. However, in an alternative embodiment in which the
precise position of the GSO satellite is known, the instantaneous forbidden
t
CA 02213770 1997-08-2~
WO 96/31016 PCIIGB95100718
- 28 -
beam is ~Ised to define a forbidden area with which the non-GSO satellite
does not comrrlunicate, the forbidden area being smaller than the forbidden
band .
Furthermore, the above equations for the forbidden beam are not specific to
5 a combination of a GSO and a non-GSO satellite, but may be applied to any
two satelllites, so that a forbidden beam may be calculated in any situation
involving interl erence between two satellites and transmission from or
reception by one of the satellites may be inhibited in the forbidden beam.
HANDOVER
10 In each of the above methods, the non-geostationary satellite 20 is prevented from communicating with any ground station within the forbidden band or
beam. However, an object of using non-geostationary satellites is to provide
global or near global coverage. Therefore, calls routed between the non-
geostationary satellite ZO and ground stations within the forbidden band or
15 beam should not be cut off, but should be handed over in a way that avoids
the forbidden band or beam.
Figure 9 shows a situation in which first and second ground stations 50 and
5Z, which are designed for communication with the non-geostationary
satellite 20, are both linked via a ground network 54 to a PSTN 56. Calls are
20 routed between a third ground station 58 and the PSTN 56 via the non-
geostationary satellite 20 and the first ground station 50. However, as the
non-geos1:ationary satellite 20 moves with respect to the earth's surface, the
first ground station 50 falls within a forbidden band or beam with respect to
the geostationary satellite 10, so that communication with the first ground
25 station 5C) is not possible without interference with the feeder-link path FP between lthe geostationary satellite 10 and the ground station 4C.
In this case, calls are handed over from the first ground station 50 to the
second ground station 52, which is not located within the forbidden band or
CA 02213770 1997-08-2~
'
- 29 -
beam, so that the link between the third ground station 58 and the PSTN 56
is maintained. The handover is preferably controlled by determining which
calls are routed throu~h a spot-beam which covers the first ground station 50
and re-routing the calls through a spot-beam which covers the second ground
5 station 52, and may either be controlled by the third earth station 58 or the
non-geostationary satellite 20.
It is to be noted that the extent of the forbidden band or beam depends on
the value of Dmin, which depends partly on the directional properties of the
antennas associated with the ground station 40, the non-geostatioriary
10 satellite 20 and the geostationary satellite 10. The first and second ground
stations 50 and 52 may be equipped with directional or omnidirectional
antennas without affecting the extent of the forbidden band or béam.
An alternative situation will now be described with reference to Figure 10, in
which a communication link is set up between the third ground station 58
15 and a mobile ground station 59. In this case, it is not acceptable to handover
calls to an alternative ground station as in Figure 9, because the mobile
ground station 59 is not connected to a ground network which would allow
it to receive t:he calls after they are handed over.
Instead, when the mobile ground station 59 falls within a forbidden band or
20 beam with respect to the geostationary satellite 10 and the non-geostationarysateiiite 2û, c:aiis are handed over from the non-geostationary sateiiite 2û to
a further non-geostationary satellite 25 for which the mobile station 59 does
not fall within a forbidden band. Preferably, the handover is controlled by the
third ground station 58.
25 While the present invention has been described with reference to a few
specific embodiments, the description is illustrative of the invention and is not
to be construed as limiting the invention. Various modifications may occur
to thosç skilled in the art without departing from the scope of the invention
as defined by the appended claims.
~ T
