WO2004051805A2

WO2004051805A2 - Rhombic antenna array

Info

Publication number: WO2004051805A2
Application number: PCT/GB2003/005195
Authority: WO
Inventors: Michael John Notter
Original assignee: Eads Astrium Limited
Priority date: 2002-12-04
Filing date: 2003-11-27
Publication date: 2004-06-17
Also published as: GB0228233D0; WO2004051805A3; AU2003285555A8; AU2003285555A1

Abstract

An inventive rhombic antenna array structure is arranged to form a pyramid and is designed to operate with a beam forming network, enabling s shaped radiation patterns to be produced over a predetermined coverage/range. The resultant pyramid configuration advantageously has high inherent rigidity, providing scope for lightweight construction techniques. The antenna array structure finds wide application in space-based to systems as well as in ground-based systems.

Description

IMPROVEMENTS RELATING TO ANTENNAS Field of the Invention

This invention concerns improvements relating to antennas and, more particularly, concerns rhombic array antennas for application in space-based systems.

Background of the Invention

Various kinds of array antenna configurations are known to operate with

2D beam forming networks such as the Butler Matrix. In some such known configurations, a regular planar distribution of radiating elements is excited from a single beamforming network to produce an orthogonal beam set spanning the full hemisphere.

However, the use of such known configurations for space applications is unattractive if the array antenna comprises a large number of distinct radiating elements, as might occur, for example, if it was required to prevent the formation of grating lobes in a suitably high gain system. In such instances, the configuration is typically technically difficult to implement, costly and exhibit unattractive performance characteristics, for example high insertion loss, phase/amplitude errors and narrow bandwidth.

Summary of the Invention The present invention aims to overcome or at least substantially reduce some of the above-mentioned drawbacks.

It is the principal object of the present invention to provide an antenna which is structurally rigid and is compatible with established beam forming network designs for effective application in space-based systems. In broad terms, the present invention resides in the concept of utilising a rhombic array antenna aperture shape in conjunction with a beam forming network to produce shaped radiation patterns extending over a wide coverage zone.

According to a first aspect of the present invention there is provided an antenna system comprising: three rhombic-shaped mutli-element arrays; a beam forming network; each of the arrays being tiltably orientated about a predefined axis associated therewith; said arrays being orientated such as to form a pyramid configuration and wherein the elements of each array are controllably energised by operation of the beam-forming network so that the outputs from the arrays controllably produce a radiation pattern characteristic over a predetermined range/coverage region.

According to a second aspect of the present invention there is provided an antenna comprising: three rhombic-shaped multi-element arrays for use with a beam forming network; each of the arrays being tiltably orientated about a predefined axis associated therewith; said arrays being orientated such as to form a pyramid configuration; and wherein the elements of each array are controllably energised, in use, by the beam forming network so that the outputs from the arrays controllably produce a radiation pattern characteristic over a predetermined range/coverage region. In accordance with an exemplary embodiment of the invention which will be described hereinafter in detail, there are three equal-sized, regular rhombic arrays and there are eighty-one elements in each of the arrays.

Preferably, the beam-forming network comprises an analogue beam- forming network. Optionally, a digital beam-forming network can be conveniently used in conjunction with the analogue beam-forming hardware, enabling various shaped radiation patterns to be produced.

As described more fully hereinafter, the antenna/antenna system conveniently comprises means for additively combining overlapping beams of radiation produced by adjacent arrays or by each of the arrays at predetermined regions, enabling the antenna gain performance to be enhanced.

The proposed antenna design is thus concerned with the objective technical problem of how to scan a beam out to a wide angle (50 degrees) in any direction (conical coverage zone), whilst maintaining a minimum level of beam gain. This is effectively achieved by using three rhombic phased array antenna facets, each illuminating a 120 degree sector of the coverage zone. Advantageously, behind each facet is a beam forming network which feeds a maximum of eight-one (81 ) array elements in order to generate a set of beams that span the relevant illumination sector. Control over the deployment of the beams (including beam shaping strategies) can also be carried out using a digital beam forming network (DBFN). The DBFN can also be made to combine the beams from different array facets in order to enhance the antenna performance in certain parts of the overall coverage region.

The inventors have realised that the inventive aspect of the proposed design resides not so much in the component parts, but in their combination so as to provide an elegant, effective engineering solution to a difficult technical problem.

In essence, the proposed antenna design confers the following key advantages:

(1 ) Producing a mechanically rigid pyramidal structure - note that the inherent structural rigidity helps to reduce mass for applications in space.

(2) Making use of a rhombic array aperture shape which has improved performance characteristics and reduced scanning requirements. This is compatible for use with various beam forming network hardware designs. This particular combination conveniently provides compactness and low mass with a high degree of functionality.

(3) Using a DBFN in conjunction with analogue beam forming hardware, to produce shaped radiation patterns. (4) Maintaining adequate antenna gain characteristics as the beam is scanned out to a wide angle (50 degrees) in any direction.

(5) Producing improved directivities from individual arrays greater than 25dbi over 30% bandwidth - this can be boosted in areas where beams overlap: helps to offset gain-loss due to production of grating lobes.

(6) Carrying out pattern synthesis in beam space.

(7) Low data-rate digital "beam - forming" made possible using selected beam data.

The proposed antenna design finds wide application in space-based systems (for example, in the ESA PARIS System) as well as in ground-based systems. The antenna design is compact, simplified and can be implemented at reasonable cost. The present invention further extends to a method of operating the above described antenna/antenna system.

The above and further features of the invention are set forth with particularity in the appended claims and will be described hereinafter with reference to the accompanying drawings. Brief Description of the Drawings

Figure 1 shows an antenna array structure embodying the present invention;

Figure 2 shows the coverage achievable using the antenna array structure of Figure 1 ; Figure 3 is a diagram showing how overlapping beams from the arrays of the antenna of Figure 1 can be combined to boost antenna gain;

Figure 4 shows a distribution of lobes associated with the antenna array structure of Figure 1 ;

Figures 5, 6 and 7 show various radiation pattern characteristics associated with the antenna array structure of Figure 1 ;

Figure 8 is a table of directivities associated with the antenna array structure of Figure 1 ; Figure 9 shows the typical aggregate pattern response associated with the antenna array structure of Figure 1, after combining beams from the arrays; and

Figure 10 shows an antenna passband response for specific application in the PARIS space system.

Detailed Description of an Exemplary Embodiment

Referring first to Figure 1 , there is schematically shown therein a preferred antenna array structure embodying the present invention. As shown, the antenna array structure comprises three ('trio') rhombic-shaped array facets 1 , 2, 3 which are arranged to form a pyramidal structure 10.

It is to be noted that the proposed non-planar array structure is of unique design in that it is derived from a regular planar hexagon geometry by dividing the area into three equal-sized regular rhombic portions. This is shown in Figure 1. Each of the three rhombic portions is tilted outwards by a predetermined amount about its associated longer geometric axis (namely, about the rhombic tilt axes aa, bb, cc as shown in Figure 1), so as to form a pyramidal configuration having three sub-array planar facets. By extending the facets to fill-in the gaps at the top of the pyramid 10, it is possible to create a robust mechanical structure - a desirable attribute for space applications. Advantageously, the rhombic sub-array facets 1 , 2, 3 of the embodiment of the invention are directly compatible with a new type of 2D analogue beam forming network prototype designed by the European Space Agency (ESA) and currently available to the Applicant as a technical solution option. Note that ESA beam former hardware design is arranged to feed a maximum of 81 array elements, which in turn imposes a limit on the size of the array aperture subject to the condition that grating lobes are not generated (such lobes cause unwanted gain loss in the radiation pattern and also restrict the beam scan range).

Conveniently, the proposed array configuration of the invention is able to use three 81 -element ESA beam formers to deploy one or more beams over an extended scan range, without suffering excessive gain loss due to the production of grating lobes. This is effected by using each sub-array facet to illuminate a 120 degree sector of a conical coverage zone. It is to be understood that the individual electrical boresights of the facets are orientated outwards by the tilt angle, so that the associated beam-scan angle is half of that of the complete antenna. It is also to be understood that the beam-scan angle is measured out from the pyramid axis.

Thus, having regard to the foregoing description, the inventors' idea is to partially compensate for the reduced array size, compatible with using only 81 elements, by restricting the scan angle to 25 degrees from the individual arrays. In this way, the expected scan loss can be typically be cut from 1.9dB to 0.4dB in the absence of grating lobes. Also, only 30 beams are required from each array beam forming network (BFN) - giving 90 in total over its 50 degree conical coverage zone.

From a mechanical aspect, the resultant pyramidal structure has high inherent rigidity, providing considerable scope for lightweight construction techniques.

Figure 2 shows the coverage achievable using the antenna array structure of the embodiment. More particularly, Figure 2 shows the proportion of the 50° coverage cone spanned by the three array facets of the embodiment when the individual scan angles are limited to 25° each. If the complete cone is covered, then this value has to increase to a maximum 41°.

Considering the array design, the maximum dimension, L, is 1.382m (from Figure 2). The number of elements per row is 9 for the 81 element BFN, hence the element spacing, S = 153.56 mm. Array area = L²Sin60 =1.654 sq.m.

Figure 3 shows how overlapping beams from adjacent arrays and from all three arrays of the structure of Figure 1 can be combined to increase antenna gain. More particularly as shown, along any scan plane separating two 120 degree coverage sectors 50, the relative positions of adjacent sub-array facet phase-centres are invariant with scan angle, which allows overlapping beams produced by each sub-array to be added, thereby typically yielding a maximum 3dB gain improvement along these directions. Furthermore, as shown, on and near to the complete antenna boresight 60 (pyramid axis), overlapping beams from all three array facets can be added to provide typically a maximum gain enhancement of 4.8dB. Checking next for grating lobes at 1591.7 MHz: the element spacing, S =

0.815λ giving the lobes distribution as depicted in Figure 4. Thus, in the worst case scan directions, the grating lobes should just about enter visible space at 25° array scan angle - which is about optimum for the proposed configuration. No such grating lobe problems exist at the lower frequencies. By way of example, computed radiation patterns for the proposed array facet, unscanned and scanned out to its 25° limit, are given in Figures 5, 6 and 7. Note that the combined 'trio' antenna boresight occurs at the 25° limit in the scan direction chosen in the first two of these figures. Figure 7 is for a cut in the orthogonal direction which is directed straight through the peak of the closest grating lobe.

A summary of the key 'trio' antenna pattern directivities is also tabulated in Figure 8.

It is to be appreciated that the antenna array configuration of the invention provides an excellent vehicle for applying digital beam-forming techniques in conjunction with proven 2D analogue beam forming techniques, if desired. This is significant because the pattern synthesis is performed in beam space and because only the information from a few selected beams needs to be digitised (i.e. rather than the signal from every array element). The data rate can thus be drastically cut and be within the scope of modest DSPs. A description of the function/role of the digital beam forming network

(DBFN), as applied to the invention, is provided hereinafter.

Conventionally, the function of the DBFN is to preserve all the spatially received information incident on the array aperture by down-converting each element output to baseband (or a low IF band) with the signal phase characterised through I and Q channel extraction. This allows beam forming operations encompassing: (1) Adaptive nulling,

(2) Formation/optimisation of shaped coverage patterns,

(3) Source tracking, and

(4) Array calibration , to be performed through the manipulation of information in a computer, rather than by means of hardware control. Such a distinction is important, because it means that the physical limitations normally encountered no longer apply. For example, it becomes possible to create any number of beams with arbitrary angular spacing without the gain loss and the matching problems that would otherwise inevitably feature in any beam forming hardware performing such a task. Furthermore, the functionality of the 'software' beam former can be modified simply by changing the control algorithm and may include real-time mathematical operations (possibly non-linear) on the data during beam generation - all of which is impractical in a hardware based system. The penalty for such flexibility is the increased antenna system complexity needed to handle the huge data rates generated, the latter being set by sampling requirements needed to characterise the information received by each of the array elements. As an example, a 100 element array with 5MHz bandwidth and 9 bit ADCs (IF sampling) generates serial data at the rate of: 100 x (2 x 5MHz) x 9 = 9GHz.

It is to be appreciated that there is considerable scope for parallelisation in the processors and a trade-off between processor speed and quantity of hardware is required. Note that to generate beam patterns is equivalent to spatial filtering the data and amounts to performing (in this case) 100 complex multiplications and additions at the element data rate.

Turning now to the specific application of DBFN to the antenna of the invention, there are two important factors:-

(1 ) The antenna(s) is already equipped with analogue beam formers, and (2) The required operational tasks are limited to tracking multiple sources and S/N optimisation.

The presence of a beam former does not necessarily mean that a simplified form of digital processing network can be used - the number of available beam ports is equal to the number of array elements, so the number of channels can be the same. However, for the proposed antenna of the invention, only a small subset of beams is actually needed to span the coverage zone, so there is significant scope for simplification. Having what is in fact a multi-beam antenna, rather than a phased array, provides reduced pattern synthesis since the aperture distribution can now only be derived from N (= No. beams) constant amplitude functions with discrete linear phase tapers. Nevertheless, if the beams form a spatially orthogonal set, then the synthesis 'solutions' are easy to generate and tend to be stable with frequency. The range of possible patterns is also useful and adequate for space-based applications (for example, it finds direct application in the PARIS system).

Thus, by digitising the beam data channels rather than operating at element level, the processor tasks (and therefore required degree of speed/parallisation) are advantageously reduced according to the following:-

(a) Forming the element weight vector inner product to create a 'radiation pattern' is now replaced by a similar process applied to the beam-weighting. Since the number of beams under consideration can be very small (sometimes just one), the simplification is substantial.

(b) For simple tracking + search/zoom tasks and beam shape optimisation, pattern synthesis is very simple in beam space and can be conducted at a 'leisurely pace', as dictated by the rate of change of the coverage zone scenario. Once 'locked-on', the system may need to digitise just a few of the available beams, with appropriate reductions in data rates. Digital beam forming can also be conveniently used to extend the capability of the proposed antenna by combining selected beams from the adjacent array facets. The combined gain along, and near, the coverage sector boundaries shown in Figure 3 can typically be increased up to 3dBi by this technique, which in turn allows the full 50° coverage cone to be accessed. Furthermore, the combined gain at the centre of the coverage sectors shown in Figure 3 can typically be increased up to 4.8dBi by this technique.

Because the phase centres of the array facets are substantially separated, the digital processing needs to apply true time delays (TBC) to the beam data to avoid frequency 'coloration' of the resultant patterns. True time delay can be accomplished using a digital filter (or digital filter algorithm). Note that the patterns generated by combining beams from separate arrays are not the same shape as the component beams, but are modified as typically shown in Figure 9.

As shown in Figure 9, the angular distance between nulls = Ω = 2 ArcSin (λ/2d) where d= separation between array facet phase centres. For the geometry considered, d = L sqrt (3)/2 = 1.382 x 0.866 = 1.2m. Hence, Ω = 9 degree ® 1591.7 MHz.

Paris-Gamma Space System: Antenna Design Considerations

Reguirements

The antenna is primarily intended to access signals from GPS satellites, but it needs to show compatibility with the proposed Galileo system so that opportunities to acquire ranging data are fully exploited. An outline of the technical requirements is given below:-

Antenna Gain: 22-25 dB (TBC)

Receive Noise figure: 2 dB (TBC) Satellite Altitude: - 500 km

Antenna coverage: up to 50 degree half cone

Simultaneously track up to 12 targets within this coverage. Minimum of 8 to be tracked.

LHCP Multiple frequency operation, as shown below :-

Maximum antenna physical dimension = 2.7m (Re. Launcher accommodation)

Assuming that printed-circuit array elements, such as patches or annular- slot radiators, are chosen to populate the array aperture(s), then striving for dual frequency operation is preferable to attempting the full range bandwidth of 30% (most wide band patch designs, are typically limited to ~ 20%). In practice, this amounts to having an 'idealised' pass band response, as depicted in Figure 10. The inventors have found that their proposed antenna design fulfils the above mentioned requirements and that their design finds wide application in space-based systems such as the PARIS space-based system. For the sake of completeness, a brief description of the PARIS System is provided hereinafter.

The 'PARIS' concept offers a practical means of monitoring sea-states from space by receiving GPS (and ultimately GALILEO) L-band satellite signals, which have been reflected from the Earth's surface. Using the direct-path received signal as a reference, the coded modulation is extracted and processed to yield accurate ranging information on a continuous basis as the platform moves in a LEO trajectory. Although a passive system, PARIS does require a directional antenna of at least 23dBi gain, which can actively track one or more specular reflection points as they traverse the instrument's angular acquisition window (typically a circular zone, subtending an angle of 100°). This can be most easily accomplished by deploying a set of contiguous beams from a phased array via a BFN, each of which can be selected through a switching matrix, or by combining them using digital 'beam forming' techniques at baseband. The latter is particularly innovative and allows the application of dynamic pattern synthesis algorithms to optimise received signal to noise ratio in a way that would be difficult and costly, if carried out using hardware directly. This is regarded as a key performance-enhancing factor and it is proposed that the scope of this work include an initial assessment of the current technical possibilities.

To realise the full potential of the PARIS system, it should be capable of accessing all suitable satellite signals as the opportunities occur - implying a frequency bandwidth of 30% for the antenna. The design of a radiating element to satisfy this requirement is expected to be challenging, especially since it has to be lightweight and (ideally) a low profile printed circuit configuration. Completion of this phase is essential before the detailed design of the antenna can proceed, hence radiator development up to breadboard level is envisaged within the scope of this study.

Conclusions

The inventors have identified a new antenna configuration for space- based systems such as the proposed PARIS-GAMMA space-based system. It can make use of the new 'Coromina' 2D BFN, which does appear to combine compactness and low mass with a high degree of functionality.

Preferably the proposed design involves three 81 -element rhombic arrays configured on pyramidal lines. Each array illuminates a 120deg sector with reduced scanning requirements, which allows greater element spacing to be used. Total coverage is typically less than the 50 degree cone, but can be adequate for practical use. The possible role of a 'digital beam forming network' has been considered and found to be attractive provided that processing operations are carried out in 'beamspace' (Fourier Transform domain) rather than on the array element outputs. This dramatically reduces the data rate generated and allows essential pattern manipulation/spatial filtering tasks to be carried out easily in the digital domain. A particularly useful facility is the ability to combine the beams from adjacent arrays so as to extend the overall scan range of the antenna.

Having thus described the present invention by reference to a preferred embodiment, it is to be appreciated that the embodiment is in all respects exemplary and that modifications and variations are possible without departure from the spirit and scope of the invention. For example, various groups of substantially rhombic-shaped array facets could be used in the invention and the number of elements in each array could be readily varied (up to a maximum of 81 elements) so as to provide the same technical inventive effect. Further, it is to be appreciated that the analogue/digital beam forming network for use with the inventive array structure could comprise a single network feeding three sub networks which in turn feed the array structure.

Claims

1. An antenna comprising: three rhombic-shaped multi-element arrays for use with a beam forming network; each of the arrays being tiltably orientated about a predefined axis associated therewith; said arrays being orientated such as to form a pyramid configuration; and wherein the elements of each array are controllably energised, in use, by the beam forming network so that the outputs from the arrays controllably produce a radiation pattern characteristic over a predetermined range/coverage region.

2. An antenna system comprising: three rhombic-shaped multi-element arrays; a beam forming network; each of the arrays being tiltably orientated about a predefined axis associated therewith; said arrays being orientated such as to form a pyramid configuration; and wherein the elements of each array are controllably energised by operation of the beam forming network so that the outputs from the arrays controllably produce a radiation pattern characteristic over a predetermined range/coverage region.

3. An antenna/antenna system as claimed in Claim 1 or 2 wherein there are three equal-sized, regular rhombic arrays and there are eighty-one elements in each of the arrays.

4. An antenna/antenna system as claimed in Claim 1 , 2 or 3 wherein the beam forming network comprises an analogue beam forming network.

5. An antenna/antenna system as claimed in Claim 1 , 2, 3 or 4 wherein the beam forming network comprises a digital beam forming network.

6. An antenna/antenna system as claimed in any of the preceding claims further comprising means for additively combining overlapping beams of radiation produced by adjacent arrays or by each of the arrays at predetermined regions, enabling the antenna gain performance to be enhanced.

7. An antenna/antenna system substantially as herein described with reference to the accompanying drawings.

8. A spacecraft incorporating an antenna/antenna system as claimed in any of the preceding claims.