WO2015019084A2

WO2015019084A2 - Antenna array

Info

Publication number: WO2015019084A2
Application number: PCT/GB2014/052407
Authority: WO
Inventors: Kristian Zarb ADAMI; Eman Omar FARHAT; Charles V. SAMMUT; Yongwei Zhang; Anthony Keith Brown
Original assignee: The University Of Malta; The University Of Manchester
Priority date: 2013-08-09
Filing date: 2014-08-06
Publication date: 2015-02-12
Also published as: GB2516980A; WO2015019084A3; GB201314339D0; GB2516980B

Abstract

An antenna array has a plurality of elements including at least one of a first type and at least four of a second type. The element of the first type (12) comprises part of two seeds (20, 21) with two elements (13,16) of the second type, and is capacitively coupled to two further elements (14,15) the second type. The elements may each comprise a plurality of hollow sub-elements, arranged one within the other, to define openings in each element between the sub-elements. Alternatively, or in addition, the antenna may include a ground plane separated from the planar element array and a frequency selective surface formed by a resistive component on a layer of dielectric material, between the planar element array and the ground plane.

Description

ANTENNA ARRAY

Background of the Invention

Field of the Invention

The present invention relates to antennas of the array type and in particular to such antennas which are designed to have a wide usable frequency bandwidth.

Summary of the Prior Art

There are a large variety of existing microwave antenna designs, including those consisting of an array of flat conductive elements which are spaced apart from a ground plane . Wide band dual -polarised phased arrays are increasingly desired for many applications. Such arrays which include elements that present a vertical conductor to the incoming fields, often suffer from high cross polarisation level. Many system functions have well defined polarisation requirements. Generally, low cross polarisation level is desired across the whole bandwidth.

Mutual coupling always occurs in aperture arrays especially in compact antenna arrays and it is related to the element type, the element separation in terms of wavelength and the array geometry. It is normally a particular problem in wide bandwidth arrays where grating lobe production must be avoided. For the conventional Vivaldi notch antennas, the spacing of elements in the arrays must be less than the maximum element separation allowed for grating lobe free scan. This is due to input impedance anomalies caused by the strong coupling induced between the elements for large scan angles. Potentially more elements are required to cover the same collecting area. As a result, the design seeks to minimise the coupling although this is problematic.

'Munk' antennas as disclosed in B. Munk, " A wide band, low profile array of end loaded dipoles with dielectric slab compensation," Antennas Applications Symp . , pp. 149-165, 2006, use a fundamentally different approach to design the wideband array. An example is shown in Fig. 1. Mutual coupling is intentionally utilised between the array elements, and controlled by introduction of capacitance. An element consists of a part of coupled dipoles (5,8) and (4,6) . The capacitance (7,9) between the ends of dipoles smoothes the radiated fields and achieves a broad bandwidth. The impedance stability over the frequency band and scan angles required is enhanced by placing dielectric layers on top of the dipole array.

The superimposed dielectric layers are important to the design of the Munk dipole array. Three or four layers of dielectric slabs are required in order to achieve a broad bandwidth. Cost becomes high for a large scale array.

One antenna type using the principles expounded by Munk is the Current Sheet Array (CSA) . A CSA formed by using closely spaced dipole elements is shown in Fig. 1. The configuration here consists of two layers of dielectric material (la, lb) on top of the dipole array (one part shown in Fig. 1) in addition to two thin sheets (both shown as layer 2) on both sides to embed the dipole elements (4,5,6,7,8,9) therebetween, and a further dielectric layer 3, Fig 2 shows a Munk Array , in which that the layers of dielectric slabs on the top are replaced by array of metal patches with predetermined shapes and a relative distance from the array elements as shown in Fig. 2. The scan performance for the dipole array of Fig. 1 is shown in Fig. 3a, and that for the array of Fig. 2 is show in Fig. 3b.

WO2010/112857 disclosed developments of the CSA arrangement discussed above, in which the antenna array included a

plurality of elements, the elements including at least one element of a first type and at least four elements of a second type. The elements of the first type comprised parts of two balanced fees with two elements of the second type and the element of the first type being capacititvely coupled to two further elements of the second type. Although the elements of the first and second types could have different physical structures, WO2010/112857 disclosed that the elements of both types may have the same physical structure. However, in either case, the elements were arranged so that they performed the functions of one or the other of the types referred to above.

Summary of the Invention

The present invention seeks to develop an antenna of the general type disclosed in WO2012/112857 , in order to further improve the performance of the antenna array.

A first aspect of the present invention is concerned with the configuration of the elements. In WO2012/112857 , the elements were simple solid shapes, or simple hollow shapes.

In a first aspect of the present invention at its most general, on the other hand, the elements are each formed by a plurality of sub-elements, arranged one within the other, so that the element has openings in it, between the sub-elements. It has been found elements composed of sub- elements in this way provide good bandwidth and in particular may provide broader bandwidth than completely solid or simple hollow elements, whilst enabling the amount of material used to form the elements to be kept low. Thus, according to the first aspect of the invention there may be provided an antenna array including a plurality of

elements, the elements including at least one element of a first type and at least four elements of a second type wherein the element of the first type comprises part of two feeds with two elements of the second type and the element of the first type is capacitively coupled to two further elements of the second type;

wherein the elements each comprise a plurality of hollow sub- elements, arranged one within the other, to define openings in each element between the sub-elements.

Whilst it is possible for the sub-elements within one element to have different shapes, they will normally have the same shape, but differ in size. This makes for ease of manufacture. Moreover, whilst it is possible for different elements to have differently shaped sub-elements, it is preferable for them to have the same shape, again for ease of manufacture. In such an arrangement, all the elements then have the same pattern of sub-elements. It is preferable that the sub-elements are symmetric with a common centre of symmetry. For example, the sub-elements may have a square, hexagonal or octagonal outer periphery, the last of these being preferred. Where there is such a centre of symmetry, it is preferable that the sub- elements of each element are concentric on a common centre of symmetry . Such an arrangement, in which the sub-elements of all the symmetric shape, and with a common centre of symmetry, so that the sub-elements within an element differ only in size and possibly in orientation, will be referred to as a "fractal" arrangement .

Normally, the inner periphery of the sub-elements may be the same geometric shape as the outer periphery, although again this is not essential. Where the inner and outer peripheries have the same shape, those shapes may be rotated relative to each other. This assists in the formation of the openings in the elements, between the sub-elements. It will normally be the case that the elements will be arranged in a planar array. In such a case, there will normally be a ground plane of the antenna array separated from the planar element array. An element of the array may then be connected to that ground plane by suitable leads. Usually, that is the element of the first type.

In a further development of the present invention, it is possible for there to be a frequency selective surface (FSS) structure between the planar element array and the ground plane. The frequency selective surface structure may be formed by a resistive component on a layer of dielectric material such as polyethylene. The component may be a loop of square, circular or other shape, although square is preferred. It is also possible for the component to be in the form of wires (possibly intersecting), dipoles, crossed dipoles or one or more crosses .

It has been found that the use of such a frequency selective surface structure improves the bandwidth of the antenna array. Whilst such a frequency selective surface structure may be used in an antenna array according to the first aspect, it has been found that it has benefit in antenna arrays with other element configurations, including those discussed in

WO2012/112857. Thus, the use of such frequency selective surface structure between the planar element array and the ground plane represents, at its most general, a second aspect of the present invention.

Thus, according to this second aspect, they may be provided an antenna array including a plurality of elements, the elements including at least one element of a first type and at least four elements of a second type wherein

the element of the first type comprises part of two feeds with two elements of the second type and

the element of the first type is capacitively coupled to two further elements of the second type; wherein the elements are arranged in a planar array; wherein the antenna array further includes :

a ground plane separated from the planar element array; and

a frequency selective surface formed by a resistive component on a layer of dielectric material, between the planar element array and the ground plane.

In the second aspect, as in the first, it is normally the element of the first type that is connected to the ground plane by suitable leads.

In both aspects, as mentioned above, the element of the first type comprises part of two feeds with two elements of the second type. Those feeds are normally balanced.

Preferably, in either aspect, the antenna array includes further elements of the first type and arranged such that each element of the second type is both capacitively coupled to an element of the first type and also forms part of a balanced feed with an element of the first type.

In such a case, it is preferable that each element of the second type is only capacitively coupled to one element of the first type and also forms part of only one balanced feed with an element of the first type. In either aspect, it is preferable that, for each element of the first type the four elements of the second type associated with it are spaced equally around it. In the above discussion of the aspects of the present

invention, we have referred to elements of the first and second type respectively. As noted above, it is possible for the elements used to have different configurations. However, this is not usual, and normally the elements of the first and second type will have the same physical structure. However, they will be arranged so that they perform the functions of one or the other of the types.

The present invention enables an increased bandwidth to be achieved, as compared to the antenna arrays shown in

WO2012/112857. A large scanning volume may also be achieved.

Brief Description of the Drawings

Embodiments of the present invention will now be described by way of example, with reference to the accompanying drawings in which:

Fig. 1 shows an example of a prior art "Munk" dipole antenna; Fig 2 shows another example of a "Munk" dipole antenna; Figs. 3a and 3b show the performances responses of the antennas of Figs . 1 and 2 ;

Fig. 4 shows a unit cell of an antenna array according to a first embodiment of the present invention;

Fig. 5 shows a detail of the unit cell of Fig. 4, illustrating the radiating elements in more detail; Fig. 6 is a graph showing the scanning performance for four directions of the antenna array of the first embodiment;

Fig. 7 is a graph showing the broadband scan performance of the first embodiment;

Fig. 8 is a graph showing the cross polarisation level in dB for an element in the first embodiment;

Fig. 9 shows a unit cell of an antenna array according to a second embodiment of the present invention;

Fig. 10 is a graph similar to that of Fig. 6, but for the second embodiment; Fig. 11 is a graph similar to that of Fig. 7, but for the second embodiment; Fig. 12 is a graph similar to Fig. 8, but for the second embodiment ;

Fig 13 is a graph comparing the gain of the first and second embodiments; and

Fig. 14 is a graph showing measurements and simulations carried out using an antenna according to the first

embodiment .

Detailed Description

A first embodiment of the present invention will now be described, with reference to Figs. 4 to 8.

This first embodiment is a tightly coupled phased array (TCPA) which is formed by a multiplicity of unit cells, each as illustrated in Fig. 4.

Each cell thus comprises a layer 10 of radiating elements comprising a central element 12 surrounded by four (preferably equi-spaced) elements 13 to 16. Only half of each of those elements 13 to 16 is shown in Fig. 4, with the other half forming, in each case, an element of an adjacent unit cell. The central element 12 is fed via feed lines 20, 21 extending to a ground plane 22. These feed lines 20, 21 provide two orthogonal feeding points for the radiating elements, so that they are dual polarised. The feed is balanced, with line 20 acting as a balun, and has a low insertion loss. An

additional layer 30, acting as an impedance transformer layer (also known as a matching layer) is provided parallel to, and spaced from, the layer 10. That layer 30 comprises similar (but in this case scaled-down) conductive elements of a shape corresponding to those of the elements of the layer 10. The matching layer 30 enhances the bandwidth of the array.

The configuration of the elements 12 to 16 is shown in more detail in Fig. 5. The central element 12 is connected to elements 14 and 15 by capacitors (not visible in Fig. 5) .

These lower the operating frequency of the array. In

addition, the central element 12 forms half of two balanced fed element pairs, one pair being with element 16 and the other being with element 13. Again, only half the elements 13 to 16 are shown in Fig. 5. Note that the capacitors which connect the central element 12 to the elements 14, 15 may be bulk capacitors or printed capacitors.

Thus, the antenna array of this embodiment is formed by a plurality of cells, with planar elements with strong mutual coupling. The array acts as an aperture array, rather than having discrete elements. The capacity of the coupling between the elements discussed above is used to counteract the ground plane reactive impedance, and maintain a real input impedance over a wide bandwidth. The array has closely spaced elements, so that there is a strong mutual coupling between the elements resulting in a continuous current distribution array.

As mentioned above, the elements 12 and 16 are fed by the balun formed by line 20, and the elements 12 and 13 are fed by line 21 which makes the radiator dual polarised due to the orthogonal feeding points in the element. The layer 30, which acts as a matching layer for impedance transformation, is a scaled down version of the radiating elements, and is used to enhance the bandwidth of the array. In this embodiment, the scaling factor used was 0.9. As can be seen from Figs. 4 and 5, the elements 12 to 16 are not solid geometrical shapes . Each element comprises a plurality of hollow octagonal sub-elements, arranged

concentrically one within the other. They thus form a fractal structure, as mentioned above. In this embodiment, the sub- elements have an octagonal periphery with the vertices of the plurality of elements aligned along radial lines. In

addition, the inner periphery of each sub-element is also octagonal, but with the vertices of the inner periphery shifted relative to the vertices of the outer edge of each octagon. Each sub-element is a reduced size copy of the sub- element immediately outside it. As a result, triangular openings are formed in the element, between adjacent sub- elements. Two such openings are identified by references 40, 41 in Fig. 5. It can be seen that, between each adjacent sub- elements, there are eight such openings.

In a detailed configuration of the embodiment of Fig. 1, the space between the adjacent elements is 800 mm, which is slightly greater than half wavelength at the highest frequency (250 MHz for this unit cell) . The unit cell covers frequency range from 60 MHz to 250 MHz. The distance from the radiator elements to the matching layer is 235 mm while the ground plane is placed at 390 mm from the

radiator elements. To lower the array's operating

frequency, capacitors are used in the gap between the neighbouring rings at the tip ends. The gap between the sub elements is 5 mm. These capacitors are bulk

capacitors of 3 pF. The array design parameters are jointly optimized together, using a genetic algorithm implemented in Matlab to validate a perfect match via commercial software Ansoft HFSS vl2. This optimisation reduces the VSWR values over the operating frequency band of the antenna array. Nine scaled and subdivided octagonal rings forming the sub-elements are provided to create an aperture array of fractal geometry patches. The bandwidth performance of the periodic array may be sensitive to the outer and inner diameter of the sub-elements of sub- elements 12 to 16. Therefore, the optimised values of both the outermost and the innermost sub -element radius may be 197.5 mm and 170 mm respectively. As mentioned above, the central element 12 is fed by feed lines 20, 21. Those lines 20, 21 are formed by striplines, and to match antennas input impedance with a 50 Ω input S A connector, an impedance transformer is implemented with the stripline design.

The feeding lines have the significant feature of

providing a low insertion loss. For an ultra wideband performance, the single-ended feeding lines were optimised with the unit cell. As a result, the stripline length is 503.5 mm. Thus, an extra part of the body is reaching out the ground plane. This array offers a convenient integration with the feeding lines.

To test the array of this first embodiment, a test was made in which the unit cell was modelled using periodic boundary conditions. The broadside, as well as the scanning active voltage standing wave ratios (VSWR) of an infinite dual- polarised array was calculated over a unit cell. The VSWR performance shown in Figs. 6 and 7 indicates that satisfactory scanning properties and wideband performance can be achieved over a 4.4:1 frequency band. Fig. 8 also shows that the array antenna of the first embodiment may have a wideband performance with a stable lowcross - polarisation level across a bandwidth from 60MHz to 266MHz. Measurements of the reflection co-efficient of a single antenna in a 4x4 array have been taken, and compared with values predicted by electromagnet simulations. The results are illustrated in Fig. 14. The reflection co-efficient of an antenna provides a way of measuring the antenna's performance and its response to electromagnetic radiation. In general, it has been the view in the art that an antenna with a reflection co-efficient of less than -lOdB represents a very good receiver. Fig. 14

illustrates that the antenna of the first embodiment achieve broadband performance from 90 to 450 Mhz which is less than -lOdB. There is also a good agreement between measurements and simulations . Thus, this embodiment enables an array of a conformal

performance to be achieved, while maintaining a low profile of roughly 0.05X, where λ is the wavelength of the lowest operational frequency of the operating bandwidth. This contrasts with e.g. a Vivaldi antenna, which is widely used as an ultra-wide antenna, but exhibits a profile of 0.5Λ, which may be too thick, especially in the low frequency range. The modelling described demonstrates that the first embodiment may provide an antenna array with a bandwidth of about 5:1, which compares favourably with the bandwidth of about 2.5:1 achievable by the antenna arrays disclosed in WO 2010/112857. It has a large scanning volume of about ± 45 degrees, and has low cross polarisation performance.

A second embodiment of the present invention will now be described with reference to Figs . 9 to 13. In fact the second embodiment is generally similar to the first embodiment, and the same reference numerals are used to indicate corresponding part, and those parts will not be described in further detail now. It should also be noted that although Fig. 9 does not illustrate the impedance transformer layer 30 in Fig. 4, such a layer will normally be present.

The second embodiment differs from the first in that there is a resistive frequency selective surface (FSS) structure 50 between the layer 10 of elements and the ground plane 22. The structure 50 comprises a square loop of resistive material arranged on a dielectric material, such as polyethylene, it should be noted that although the loop in Fig. 9 is square, other shapes are possible. The presence of a frequency selective surface structure 50 has the effect of enhancing the bandwidth of the antenna array. For example, it may increase the operating range from 50 to 250MHz to 50 to 400MHz.

In general, a tightly coupled array bandwidth is limited by the ground plane. More precisely, an array with a ground plane spacing of h is short circuited at the upper bond of the operational band at fupper - c/2h. In other words, the upper bound bandwidth of any tightly coupled phase array is limited due to ground plane ¾p given by:

where η₀ is the substrate impedance (in this array it is free space), β is the substrate propagation constant, and h is the array spacing above the ground plane. The array depicts a resonance peak at h = c/2 f_upper , because the ground plane impedance becomes Z_GP = 0.

A conformal array is short circuited when h = ίί_/2 (where X_H is the wavelength at the highest/upper operational frequency) , and thus a method to alleviate ground plane effect is desirable.

Though electromagnetic band gap structures (EBG) have been proposed to overcome the ground plane effect, they operate only over limited bandwidth. Ferrites were also used to improve bandwidth, but their weight limits their applications. In order to avoid boresight radiation cancellation from the ground plane image current, the second embodiment thus has the resistive frequency selective surface structure so between the radiating elements 12 to 16 and the ground plane 22. The structure 50 suppresses the interfering of the ground plane reflection, increasing the bandwidth performance of the capacitively coupled array.

In a detailed implementation of this embodiment, the resistive loop of structure 50 is printed on a polyethylene dielectric substrate of 0.2 mm thickness, this square ring loop is an ohmic sheet of 75 Ω = sq. However, inserting the structure 50 in the array may lead to losses. A

superstrate layer of a polyethylene dielectric substrate (60 mm thickness) may therefore be mounted over the

radiating elements 12Wb. Both the structure 50 and the superstrate are preferably designed together in tandem.

The purpose of a dielectric superstrate layer is to

alleviate losses. In addition, this layer behaves as a matching impedance transformer. In an infinite array configuration, using a superstrate in conjunction with structure 50 leads to a 8 : 1 bandwidth with VSWR < 2. Hence, this unique aspect of the array maximizes the bandwidth by a factor of 2, resulting in a very low profile antenna array.

Similarly to the first embodiment, the elements 12 to 16 are linearly distributed pairs of fractal octagon rings, perpendicular to each other. The element unit cell spacing is 730 mm, and the array is dual polarised. The array is capacitively coupled by inserting bulk capacitors at the end tips of adjacent rings, the value of these capacitors is 3 pF. The array is backed by a ground plane at 390 mm. The overall profile of the array with structure 50 in this second embodiment is 450 mm (element spacing 730 mm) , while it is 625 mm for the first embodiment (element spacing 800 mm) . Thus, the advantage of this embodiment is that it has a very wide bandwidth with a lower profile. This is a considerable feature for large scale arrays applications, and it facilitates low cost for mass production.

The performance of the second embodiment is then illustrated in the graphs of Figs. 10 to 12. These graphs correspond respectively to the Figs. 6 to 8 discussed with reference to the first embodiment. It can be seen that an antenna array corresponding to the second embodiment exhibits an improved wideband performance as compared with that of the first embodiment. It can also be seen that the second embodiment is shown to be polarisation intensive. Thus, the bandwidth of this second embodiment may be 7.5:1, as compared to the bandwidth of 5:1 of the first embodiment. As in the first embodiment, a large scanning volume of + 45 degrees may be achievable.

It should be noted that the presence of the structure 50 causes some reduction in the gain (efficiency) of the antenna array, as illustrated in Fig. 13. The antenna array of the second embodiment will structure 50 is that corresponding to the line TCFORA with FSS . The antenna array of the first embodiment without the structure 50 is illustrated by the line ' TCFORA' . It is believed that this reduction in gain may be due to power dissipated by the resistance of the structure 50.

As mentioned above, the graphs of Figs. 6 to 8 , and also the graphs of Figs. 10 to 12, were produced by full-wave

simulation data for a unit cell using periodic boundary conditions, based on an infinite array. In practice, although the array will contain many unit cells, there will be edge effects at the boundary of any finite array. This generates edge and corner diffraction, affecting the outer peripheral elements, resulting in some mismatching. As a result, the impedance of the edge and corner unit cells within a finite array may differ significantly from those of an infinite array. This may affect the bandwidth of the finite array, and optimisation may be necessary to minimize this.

As mentioned above, the antenna array of the first and second embodiments has a low profile, i.e. in a direction

perpendicular to the ground plane. This is a benefit of the element configuration used.

In the first and second embodiments, the elements 12 to 16 are formed by sub-elements with octagonal inner and outer

peripheries to produce a fractal structure. However, although the elements are formed as sub-elements, the specific

structure of the sub-elements first and second embodiments is not essential. Other geometric shape for the sub-element are possible. For example, they may be square or even cross- shaped. It may even be possible to obtain the advantages of the invention in which the sub-elements are not

circumferentially symmetric. Furthermore, although the sub- elements of the first and second embodiments are concentric, in that their sizes decrease on a common centre of symmetry, again this is not essential, even with symmetric elements. The centre of symmetry of the elements may be off-set relative to each other. In the discussion of the first and second embodiments above it is assumed that the antenna will transmit signals from the elements . However, it is also possible to use the present invention in an antenna which receives signals.

An antenna array of the present invention may be of particular applicability for a radio telescope in a space base station, because of its low profile but having a wideband performance. However, it may be applicable to other applications such as communication systems. The antenna array of the present invention may be produced at a lower cost than known antennas used for such purposes.

In the above discussion, it has been assumed that the array of unit cells is planar. However, it may be possible to make the array flexible, so that it can be mounted on a curved surface, such as the surface of a cylinder. In such a case, the ground plane 22, and the layers 10, 30, 50 are themselves curved. Folding arrangements may also be possible, e.g. for deployment in space. Moreover, although the present invention has been described with reference to preferred embodiments, modifications of those embodiments, further embodiments and modification thereof will be apparent to a skilled person and as such are within the scope of the present invention.

Claims

CLAIMS :

1. An antenna array including a plurality of elements, the elements including at least one element of a first type and at least four elements of a second type wherein

the element of the first type is capacitively coupled to two further elements of the second type;

wherein the elements each comprise a plurality of hollow sub-elements, arranged one within the other, to define openings in each element between the sub-elements.

2. An antenna array according to claim 1, wherein the sub- elements of each element have the same shape but differ in size .

3. An antenna array according to claim 2, wherein the sub- elements of all the elements have the same shape.

4. An antenna array according to claim 2 or claim 3, wherein the sub-elements of each element are symmetric.

5. An antenna array according to claim 4, wherein the sub- elements of each element are concentric on the centre of symmetry of the elements.

6. An antenna array according to any one of the preceding claims wherein the periphery of each of the sub-elements of each element is octagonal.

7. An antenna array according to any one of the preceding claims, wherein the elements are arranged in a planar array.

8. An antenna array according to any one of the preceding claims, further including a ground plane separated from the planar element array.

9. An antenna array according to claim 8, further including a frequency selective surface structure formed by a resistive component on a layer of dielectric material, between the planar element array and the ground plane.

10. An antenna array including a plurality of elements, the elements including at least one element of a first type and at least four elements of a second type wherein

the element of the first type is capacitively coupled to two further elements of the second type; wherein the elements are arranged in a planar array; wherein

the antenna array further includes: a ground plane separated from the planar element array; and

11. An antenna array according to any one of claims 8 to 10, wherein the at least one element of the first type is fed through a balun.

12. An antenna array according to any one of the preceding claims, wherein the two feeds are balanced.

13. An antenna array according to any one of the preceding claims, including further elements of the first type and arranged such that each element of the second type is both capacitively coupled to an element of the first type and also forms part of a balanced feed with an element of the first type.

14. An antenna array according to claim 13, wherein each element of the second type is only capacitively coupled to one element of the first type and also forms part of only one balanced feed with an element of the first type.

15. An antenna array according to any one of the preceding claims, wherein for each element of the first type the four elements of the second type associated with it are spaced equally around it.