EP1690317A1

EP1690317A1 - Multiband dual-polarised array antenna

Info

Publication number: EP1690317A1
Application number: EP04805601A
Authority: EP
Inventors: Mostafa Jelloul
Original assignee: Arialcom
Current assignee: Itas Itas International Telecommunications And Ser
Priority date: 2003-12-01
Filing date: 2004-12-01
Publication date: 2006-08-16
Anticipated expiration: 2024-12-01
Also published as: EP1690317B1; WO2005055362A8; ES2400047T3; WO2005055362A1; FR2863110B1; EP1690317B8; PL1690317T3; FR2863110A1

Abstract

The invention relates to a multiband array antenna comprising a ground plane (4) and at least a first array of radiating elements (10) and a second array of radiating elements (20). The invention can also comprise a third array of radiating elements. Each of the aforementioned three arrays operates in a different frequency band and the arrays are arranged such as to form a set of elementary cells (51). Moreover, each elementary cell (51) comprises at least one radiating element from the second array (20C) and two adjacent radiating elements from the first array (10A, 10B) and, optionally, two radiating elements from the third array (30A, 30B). In each elementary cell, the radiating element from the second array (20) is arranged such as to face the two adjacent radiating elements from the first array (10) and the two adjacent radiating elements from the third array (30) essentially symmetrically and at a right angle.

Description

Multi-band network antenna with dual polarization

The present invention relates to multi-band network antennas used in particular in base stations of cellular radio networks.

Existing mobile communication systems include second-generation systems such as GSM900, GSM1800 and DCS 1800, and new third-generation systems such as UMTS systems. In order to exploit these new third generation systems, cellular networks compatible with second generation systems and new third generation systems are required. For this, operators typically migrate existing cellular networks that were intended only for second-generation systems to networks that are compatible with both second-generation and third-generation systems.

Base station antenna providers must then replace existing second-generation antennas, for example GSM and / or DCS, with next-generation multiband antennas, for example dual-band GSM / UMTS antennas and GSM tri-band antennas. / DCS / UMTS.

These antennas are constituted from a network antenna comprising several sets of radiating elements each operating in a distinct frequency band.

US 6,211,841 proposes such a multi-band network antenna. The network antenna comprises a first set of radiating elements operating in a first central wavelength frequency band λ 1, a second set of radiating elements operating in a second central wavelength frequency band λ 2 and a ground plane. The first set of radiating elements is arranged in two columns spaced from each other by a distance less than λ 1. The radiating elements of the first and the second set are intercalated, and the radiating elements of the second set are distant from each other by less than λ, the λ2 ratio over λ1 being between 0.25 and 0.75. The second set of radiating elements is arranged in two columns spaced from each other by a distance less than λ2, which are interposed between the two columns of the first set of radiating elements.

WO 02/084790 also proposes a network antenna capable of operating simultaneously in two different frequency bands, in double polarization. The two bands are respectively centered around a low frequency f1 and a high frequency f2, with a ratio f2 / f1 less than 1.5. The array antenna has a first row of bipolar antenna elements aligned along a first vertical axis and operating at the high frequency f2. The array antenna further comprises a second row of bipolar antenna elements aligned along a second vertical axis, and operating at the low frequency f1. The pitch between the elements of the second row is the same as that of the first row and the second vertical axis is substantially parallel to the first axis. The elements of the first row are offset from the elements of the second row, in the vertical direction and the two rows are spaced from each other.

For such network antennas to operate in tri-band, it is necessary to add duplexers to them to separate the different frequency bands. In addition, the decoupling between the different polarizations of the same frequency band or of different frequency bands is not optimized.

The invention improves the situation.

For this purpose, the invention proposes a network antenna, comprising a ground plane, or reflector, on which at least:

a first row of radiating elements capable of operating in a first frequency band and

a second row of radiating elements, adjacent and parallel to the first row, and capable of operating in a second frequency band, the first row and the second row being arranged to form a set of elementary cells. Advantageously, each elementary cell comprises a radiating element of the second row and two adjacent radiating elements of the first row, the radiating element of the second row being arranged to see the two adjacent radiating elements of the first row substantially symmetrically and at an angle. right, while each radiating element of the first and second row comprises two crossed dipoles arranged to operate in broadband and double polarization.

According to a complementary feature of the invention, a third row of radiating elements is further mounted on the ground plane, parallel to the first and second rows, the third row being capable of operating in a third frequency band and disposed so that the second row is interposed substantially equidistant between the first row and the third row, while each unit cell further comprises two radiating elements of the third row, the radiating element of the second row being arranged to see the two adjacent radiating elements of the third row substantially symmetrically and at a right angle.

In particular, each radiating element of the third row comprises two crossed dipoles arranged to operate in broadband and double polarization.

Other characteristics and advantages of the invention will appear in the detailed description below, made with reference to the appended drawings, in which:

FIGS. 1A to 1C show various antenna configurations in a multi-band network,

FIG. 2A is a schematic view from above of an elementary cell of the network antenna according to a first embodiment of the invention,

FIG. 2B is a schematic view from above of a radiating element suitable for use in the first, in the second and in the third row, according to the first embodiment of the invention,

FIG. 3 is a schematic view from above of two adjacent elementary cells of the network antenna according to a second embodiment of the invention, FIG. 4 is a schematic view from above of an elementary cell of the network antenna according to a third embodiment of the invention,

FIG. 5 is a cross-sectional view of an elementary cell according to the invention; FIG. 6 is a schematic view from above of the whole of the network antenna according to the second embodiment of FIG. invention,

FIG. 7 is a schematic view from above of a radiating element suitable for use in the first row and in the third row, according to the second or the third embodiment of the invention; FIG. 8 is a view; schematic plan of a radiating element suitable for use in the second row, according to the third embodiment of the invention,

FIG. 9 is a schematic top view of an elementary cell of a two-band network antenna according to the invention;

FIGS. 10 and 11 are diagrammatic front views of examples of main transverse partitions,

FIG. 12 is a schematic view from above of a radiating element suitable for use in the second row, according to another embodiment of the invention,

FIG. 13 is an alternative embodiment of the radiating element of FIG. 2,

FIG. 14 is a schematic representation of the radiation patterns in the horizontal plane of an antenna with double and orthogonal linear polarizations, and

- Figure 15 shows an alternative embodiment of an elementary cell of a network antenna according to the invention.

The drawings contain, for the most part, elements of a certain character. They can therefore not only serve to better understand the description, but also contribute to the definition of the invention, if any.

FIG. 1C represents a configuration in "side-by-side rows" of a multi-band network antenna, in particular of a tri-band network.

The multi-band network antenna 1 comprises three rows of independent single-band radiating elements 10, 20 and 30 arranged side-by-side, parallel, and oriented in the direction of the longitudinal axis AA ', which is generally vertical by ground ratio. The first row of radiating elements 10 operates in a first frequency band, in particular in the DCS frequency band ([1710 MHz, 1880 MHz]). The third row of radiating elements 30 operates in a third frequency band, particularly in the UMTS frequency band ([1920 MHz, 2170 MHz]). The second row of radiating elements 20 operates in a second frequency band, generally lower than the first and the third frequency band, in particular in the GSM frequency band ([870 MHz, 960 MHz]). It is interposed between the first and the third row.

These three rows are arranged on the same conductive ground plane or reflector 4.

FIG. 1A represents a "nested rows" configuration in which the first row 10 'and the third row 30' are mixed: in each of these rows, a DCS radiating element of the first frequency band is followed by an element radiating UTMS from the third frequency band. The second row 20 'of GSM radiating elements of the second frequency band is interposed between the first and the third row. This configuration has the disadvantage of requiring a reduced spacing between a radiating element of the first row 10'A and the element opposite the third row 30 'A and between a radiating element of the first row 10' A and the radiating element of the same nearest frequency band in the third row 30'B. This reduced spacing must indeed be typically less than 0.33 λ, where λ is the wavelength in the first or in the third frequency band, and therefore practically does not allow the interposition of bipolarized radiating elements, such as crossed dipoles of the half-wave type.

FIG. 1B represents a configuration in "superimposed rows" in which the three rows are superimposed in a longitudinal direction AA 'so that UMTS radiators of the third row 30 "and GSM radiators of the second row 20" are superimposed on a first part 100, for example the element 30 "A and the element 20" B, and DCS radiating elements of the first row 10 "and GSM radiating elements of the second row 20" are superimposed on a second part 200, for example the element 10 "A and the element 20" A. This configuration has the disadvantage of requiring an antenna height that is too great, especially for a tri-band antenna for which the number of DCS and UMTS radiating elements required is greater than the number of GSM radiating elements. For example, for a tri-band antenna having 9 GSM elements, 9 DCS elements and 9 UMTS elements, the antenna height would typically be around 2600 mm, which is the upper limit generally tolerated by mobile telecommunications operators. On the other hand, if the number of UMTS and DCS elements is to increase to 12 for greater directivity in these frequency bands, the height of the antenna would increase to 3600mm.

Thus the configuration in "side-by-side rows" of FIG. 1C is preferable to the "nested rows" or "superimposed rows" configurations because it is not subject to the constraint of reduced spacing between the radiating elements, nor to the significant antenna height constraint. It offers the possibility of easily modulating the number of radiating elements according to the need for antenna directivity in each of the frequency bands, independently of the other frequency bands. In addition, it offers the decisive advantage of allowing the independent misalignment of each beam formed in the vertical plane by each of the networks (or rows) composing the multiband antenna. This beam offset (or "tilt" in English) of a given network is indeed obtained by an electrical means which consists in creating a constant phase shift between the successive radiating elements of this network, thus avoiding a mechanical inclination of a set of the multi-band antenna.

The multi-band network antenna according to the invention is based on a configuration in "side-by-side rows". However, in classical embodiments based on this "side-by-side" configuration, squint effects are often observed in the horizontal plane (transverse plane to the reflector) associated with the horizontal plane. lack of symmetry in the network structure. This phenomenon is manifested by unsymmetrical horizontal radiation patterns. Furthermore, in conventional networked multi-band antennas which have a "side-by-side" configuration, it is also possible to encounter a strong mutual coupling problem between the orthogonal polarization channels of the same row of a multi-antenna. band, said intra-band coupling and / or between the different rows of a multi-band antenna, said inter-band coupling. These decoupling are commonly of the order of 20dB and generally less than 25dB, while in many applications, especially in the Telecommunications with mobiles, it is required to have at least 30dB.

The invention proposes to improve the decoupling between the first row (DCS), the second row (GSM) and the third row (UMTS) orthogonal polarization channels, as well as the symmetry of the radiation patterns of the first row (GSM) and the third row (UMTS). antenna.

The multi-band network antenna according to the invention comprises a set of elementary cells, aligned in the direction of the longitudinal axis AA ', and corresponding to a chosen arrangement of the three rows of radiating elements 10, 20 and 30 of Figure 1C. The respective axes of the first, second and third rows are substantially parallel to the longitudinal axis AA '.

FIG. 2A is a view from above of an elementary cell 5 of the network antenna 1, according to a first embodiment of the invention. Each elementary cell of a multi-band network antenna according to the invention comprises a radiating element of the second row 20C and two radiating elements 10A and 10B of the first row. The radiating element of the second row 20C thus sees the radiating elements 10A and 10B of the first row substantially symmetrically and at a right angle. The radiating elements of the first row 10 and the second row 20 are wideband dual polarized dipoles.

Each elementary cell of a multi-band network antenna according to the invention may further comprise two radiating elements 30A and 30B of the third row. The radiating element of the second row 20C also sees the radiating elements 30A and 30B of the third row substantially symmetrically and at a right angle.

The following description will be made firstly with reference to a tri-band network antenna.

As indicated above, the first row of radiating elements 10 operates in a first frequency band, in particular in the DCS frequency band ([1710 MHz, 1880MHz]), the third row of radiating elements 30 operates in a third frequency band, notably in the UMTS frequency band ([1920MHz, 2170MHz]) and the second row of radiating elements 20 operates in a second frequency band , generally lower than the first and the third frequency band, especially in the GSM frequency band ([870 MHz, 960 MHz]).

The frequency band of the first row 10 may be substantially greater than the frequency band of the second row 20.

The frequency band of the third row 30 may be substantially greater than the frequency band of the second row 20.

In particular, the ratio between the center frequency of the frequency band of the first row 10 and the center frequency of the frequency band of the second row 20 is substantially between 1.5 and 2.5.

Likewise, the ratio between the center frequency of the third row frequency band 30 and the center frequency of the second row frequency band 20 may be substantially between 1.5 and 2.5.

In the rest of the description, reference will be made to the DCS, UTMS and GSM frequency bands, by way of non-limiting example.

FIG. 2A shows a dotted "virtual" square outline 9 delimiting the elementary cell 5. At the vertices of the virtual square S1, S2, S3 and S4 are arranged the two radiating elements 10A and 10B of the first row and the two elements third row radiators 30A and 30B. Furthermore, the radiating element of the second row 20C is arranged in the center of the virtual square. Each radiating element of the third row also comprises two crossed dipoles arranged to operate in broadband and double polarization.

The second row 20 is interposed equidistant from the first row 10 and the third row 30. The radiating elements of the first row and the elements of the third row may be of the same type and in particular identical.

In the example of FIG. 2A, each of the radiating elements of the first, second and third rows 10A, 10B, 30A, 30B and 20C is of the conventional half wave crossover dipole type disposed above the reflector at a height of quarter wave order. Such a dual polarization radiating element is shown in FIG. 2B. The radiating element comprises two radiating dipoles 6 and 7 each constituted by two strands of collinear conductors 6a-6b and 7a-7b. The two strands 6a-6b (respectively 7a-7b) of each pair are aligned on the same alignment axis Δ (Δ ') and the alignment axes of the two pairs of strands intersect at right angles at a crossing point 0. The alignment axes of the two pairs of strands correspond to the two orthogonal polarization paths offset by an angle of ± 45 ° with respect to the longitudinal axis AA '. The radiating element further comprises a conventional device for feeding the dipoles 6 and 7 of the balun type.

In the cell shown in FIG. 2A, the radiating element of the second row 20C has physical dimensions greater than those of the radiating elements of the first row 10A and 10B. These dimensions are related to the wavelength in the operating frequency band of the second row 20. The radiating elements of the third row 30A and 30B have dimensions substantially equal to those of the radiating elements of the first row 10A and 10B . In addition, each radiating element is arranged so that the respective alignment axes Δ (Δ ') of the two pairs of strands are oriented by 45 ° with respect to the longitudinal axis AA'. In addition, the strands of the central radiating element of the second row 20C may extend above the other radiating elements of the cell 10A, 10B, 30A and 30B.

In particular, the pitch P 1 between two adjacent radiating elements of the first row 10, for example between the elements 10A and 10B, is substantially the same as the pitch P3 between two adjacent radiating elements of the third row 30, for example between the elements 30A and 30B.

Furthermore, the arrangement of the first row 10 (DCS) and the third row 30 (UMTS) with respect to the second central row 20 (GSM), as well as the spacing transverse Q between the first row and the third row 10 and 30 have a great influence on the decoupling between these two rows of radiating elements on the one hand and between the two orthogonal polarizations of the same row on the other hand, in particular between the orthogonal polarizations of the radiating elements of the second row 20.

The spacing Q is preferably of the order of one wavelength in the operating frequency bands of the two lateral rows 10 and 30 (rows DCS and UMTS), in order to promote the decoupling between these two rows. For example, the Q spacing may be 155mm if we consider the average wavelength of the full band [1710MHz, 2170MHz].

In particular, the spacing Q between the first row (DCS) and the third row (UMTS) may be substantially equal to the pitch PI between two adjacent radiating elements of the first row 10, which may itself be equal to the pitch P3 between two adjacent radiating elements of the third row 30.

This symmetrical arrangement, as well as the proximity of the radiating elements DCS of the first row 10 and / or UMTS of the third row 30 with respect to the GSM elements of the second row 20, have the effect of significantly improving the decoupling between the two orthogonal polarizations of the same row.

The elementary cell shown in Figure 2A, in fact has a symmetry with respect to the GSM radiating element of the second row 20C. The two diagonals of the virtual square 9 at the center of which is disposed the GSM radiating element indeed coincide with the two alignment axes Δ and Δ 'pairs of orthogonal strands 6 and 7 (orthogonal polarizations) of the GSM radiating element. It is the same with the two axes of alignment of the orthogonal polarizations of the DCS elements of the first row 10A, 10B and UMTS of the third row 30A, 30B.

The fact that the pitch PI of the first row 10 (DCS network) is substantially equal to the pitch P3 of the third row 30 (UMTS network) but also to the transverse spacing Q, further improves the symmetry. This symmetry of the elementary cell makes it possible to obtain a decoupling between the two orthogonal polarizations of the GSM radiating element of the second row which corresponds practically to the decoupling that the GSM radiating element would have if it were isolated. Decoupling greater than 30dB can be observed in the GSM frequency band. Similarly, a decoupling greater than 30 dB can be observed between the two orthogonal polarizations of the radiating elements of the first row in the DCS band, and a decoupling greater than 29 dB can be observed between the two orthogonal polarizations of the radiating elements of the third row. in the UMTS band. In addition, the decoupling between the DCS radiators of the first row and UMTS of the third row may be greater than 28 dB in the DCS band and greater than 30 dB in the UMTS band, all polarizations combined (parallel and orthogonal polarization channels). In addition to the structure of FIG. 2A, the half-power widths of the horizontal radiation patterns are observed around 90 ° in the GSM band and around 65 ° in the DCS and UMTS bands.

In addition, the ratio between the pitch P2 between two adjacent radiating elements of the second GSM row and the pitch PI between two adjacent radiating elements of the first row DCS is substantially between 1.5 and 2.5, especially when the band of frequency of the second row is substantially lower than that of the first and third row. In the remainder of the description, it will be considered that the pitch P2 between the radiating elements GSM of the second row 20 is substantially twice the pitch PI between two adjacent elements DCS of the first row 10, by way of non-limiting example.

The pitch P2 may be chosen in particular between 260mm and 310mm. For example, pitch P2 may be chosen equal to 310mm, and pitch PI and P3 to 155mm.

Furthermore, it is often required that the half-power aperture of the radiation patterns, in the plane that is transverse to the reflector and perpendicular to the longitudinal axis AA 'of the array antenna (hereinafter referred to as " horizontal plane "), is around 65 °. The Applicant has observed that the surface and the shape of the radiating strands of the DCS elements of the first row and the UMTS radiators of the third row have an effect on the directivity of the radiation patterns of the central GSM element of the second row, and in particular on the half-power width of the diagrams in the horizontal plane. Therefore, the DCS and UMTS radiating elements can be chosen according to the desired opening of the radiation patterns of the GSM central element in the horizontal plane.

FIG. 3 is a view from above of two successive elementary cells 51 and 52 of the multi-band network antenna, according to a second embodiment of the invention.

The DCS radiators of the first row 10 are identical to the UMTS radiators of the third row 30. The GSM radiators of the second row 20 are again of the half-wave cross-dipole type as in FIG. 2 A.

Each cell 51 (respectively 52) comprises a radiating element 20C (respectively 20E) of the second row 20 (GSM) arranged in the center of a virtual square with vertices of which two radiating elements 10A and 10B (respectively 10C and 10D) are placed. of the first row 10 (DCS) and two radiating elements 30 A and 30B (respectively 30C and 30D) of the third row (UMTS).

The pitch PI of the first row 10 is substantially equal to the pitch P3 of the third row 30 and to the spacing Q between the first and the third row. The pitch P2 of the second row 20 is substantially twice the pitch PI of the first row.

Figure 7 shows a radiating element of the first or third row. Such a radiating element has been proposed in French Patent Application No. 0206852.

Figure 7 shows a dashed virtual square outline 71, whose side length is "a". Inside this virtual square, the radiating element shown comprises four metal radiating plates 2a, 2b, 2c, 2d, of square shape, whose side length is "c". These four plates are juxtaposed in the same plane inside the virtual square 71.

The square plates 2a and 2c have a common diagonal, that is to say located substantially on the same alignment axis Δ ₃ ; likewise, the plates 2b and 2d have a common diagonal, that is to say situated substantially on the same alignment axis Δ ₄ . The term "diagonal" is used here with reference to the square in which each plate is inscribed. These alignment axes Δ ₃ , Δ ₄ , which are diagonals common to the two pairs of respective plates, intersect at right angles to a cross point "O" located between the plates of each pair or dipole. In FIG. 7, the alignment axes Δ ₃ , Δ ₄ also form the diagonals of the dashed virtual square 71.

The two orthogonal pairs of plates thus generate two electric fields orthogonal to each other. The pair 2a, 2c generates an electric field parallel to the axis Δ ₃ and the pair 2b, 2d generates an electric field parallel to the axis Δ ₄ . The polarization planes are at an angle of +/- 45 ° with respect to the longitudinal axis VV of FIG. 7, which passes in the space between the plates 2a, 2b on the one hand, and 2c, 2d on the other hand share.

In particular, the plates 2a, 2b, 2c, 2d may be recessed, and each comprise a hole 79, substantially of the same shape, for example a circular hole centered at the point of intersection of the diagonals of the square defined by each plate. This helps to lighten their weight.

In addition, the four outer corners of the plates 2a, 2b, 2c, 2d located at the ends of the two alignment axes Δ ₃ and Δ ₄ , can also be cut along sectional planes perpendicular to the alignment axes; this cut is substantially identical on the four corners to maintain the geometric symmetry of the two polarization paths.

Such radiating elements used in the first row and in the third row of FIG. 3 are arranged so that the alignment axes Δ ₃ and Δ ₄ form an angle of 45 ^° with respect to the longitudinal axis AA '. The strands of the central radiating element of the second row 20C (respectively 20E) may further extend above the other radiating elements of the cell 10A, 10B, 30A and 30B (respectively 10C, 10D, 30C and 30D).

For this configuration, the half-power aperture of the GSM radiating element diagrams of the second row in the horizontal plane is around 65 ^° throughout the GSM frequency band (from 870 MHz to 960 MHz). The diagrams of the DCS radiators of the first row and UMTS radiators of the third row 30 also have a half-power aperture around 65 ° in the plane. horizontal and in their respective frequency bands.

FIG. 4 represents a cell 53 according to a third embodiment of the invention.

The DCS radiators of the first row 10 and the UMTS radiators of the third row 30 are of the same type as in FIG.

The cell 53 comprises a radiating element 20C of the second row 20 (GSM), arranged in the center of a virtual square 9 at the vertices of which are placed two radiating elements 10A0 and 10B of the first row 10 and two radiating elements 30A and 30B of the third row 30.

The pitch PI of the first row 10 is substantially equal to the pitch P3 of the third row 30 and to the spacing Q between the first and the third row. FIG. 8 is a view from above of the central radiating element 20C. the second row. Figure 8 shows a dashed virtual square outline 71 ', the side length of which is "a". Within this virtual square, the radiating element shown comprises four metal radiating plates 2a ', 2b', 2c ', 2d', of the same geometric shape and of the same o dimensions. These four plates are juxtaposed in the same plane inside the virtual square 71 '.

The plates 2a 'and 2c' have a common diagonal, that is to say located substantially on the same alignment axis Δ ₃ '; likewise, the plates 2b 'and 2d' have a common diagonal, that is to say located substantially on the same alignment axis Δ ₄ ,.

These alignment axes Δ ₃ , ', Δ ₄ ', which are diagonals common to both pairs of plates, intersect at right angles to a crossing point O 'located between the plates of each pair or dipole. In FIG. 8, the alignment axes Δ ₃ ', Δ ₄ ' form o also the diagonals of the virtual dashed square 71 '.

Each plate comprises a deep recess inward from the outer corners which are located on the alignment axes Δ ₃ , ', Δ ₄ '. So each plate has a shape general triangle with a recess from the base of the triangle so that the radiating element has a general shape of cross, whose branches have substantially a length c '.

The two pairs of plates thus generate two electric fields orthogonal to each other. The pair 2a, 2c generates an electric field parallel to the axis Δ ₃ and the pair 2b, 2d generates an electric field parallel to the axis Δ ₄ . The polarization planes are at an angle of +/- 45 ° with respect to the longitudinal axis VV of FIG. 8, which passes in the space between the plates 2a ', 2b' on the one hand, and 2c ', 2d On the other hand, in particular, the plates 2a ', 2b', 2c ', 2d' may be recessed, and each have a perforation 79 ', substantially of the same shape, to lighten their weight.

This radiating element, used in the second row 20 of FIG. 4, is disposed in the array so that its alignment axes Δ ₃ ', Δ ₄ ' which define the polarization planes form an angle of +/- 45 ° with respect to the longitudinal axis AA 'of the network.

In other embodiments according to the invention, the radiating element of FIG. 8 can also be used in the first row 10 as a DCS radiating element o and / or in the third row 30 as a UMTS radiating element.

According to a complementary feature of the invention, transverse metal partitions 8 may be provided. They comprise in particular elementary transverse partitions 80 between two adjacent elements of the first row, for example between the element 10A and 10B, and between two adjacent elements of the third row, for example between the element 30A and 30B . They are placed substantially equidistant from adjacent elements they separate.

Furthermore, longitudinal metal partitions 90 may also be provided, in each elementary cell, between an element of the first row and the element opposite the third row, for example between the elements 10A and 30A, and between the elements 10B and 30B. They are placed substantially equidistant from the elements they separate and therefore along the axis of the second row 20. Figure 5 is a cross-sectional view of the cell 53 of Figure 4 along the axis BB '. The elementary transverse partitions 80 have a height H4 less than the height Z1 of the radiating elements 10B of the first row (DCS) and the height Z2 of the radiating elements 30B of the third row (UMTS). Likewise, the longitudinal partitions 90 have a height H3 less than the height Z1 of the radiating elements 10B of the first row (DCS) and the height Z2 of the radiating elements 30B of the third row (UMTS). The height Z1 of the radiating elements 10B of the first row (DCS) and the height Z2 of the radiating elements 30B of the third row (UMTS) may be of the order of a quarter of a wave in the highest frequency band. The height Z3 of the radiating elements (GSM) of the third row may be substantially greater than the height Z1 of the radiating elements 10B of the first row (DCS) and the height Z2 of the radiating elements 30B of the third row (UMTS).

The reflector 4 further comprises walls 41 and 42 on these edges. These walls may have a height H 1 and H2 substantially less than the height of the radiating elements of the first and third row.

In particular, the height H4 of the elementary transverse partitions 80 may be between about 18 mm and about 25 mm, the height Z1 of the radiating elements 10B of the first row (DCS) may be equal to the height Z2 of the radiating elements 30B of the third row (UMTS) and be of the order of 37 mm. The longitudinal partitions 90 may have a height H3 of between about 18 mm and about 25 mm. The height Z3 of the radiating elements of the third row may be between 55 and 82 mm.

The longitudinal partitions 90 make it possible to further improve the decoupling between the UMTS network (third row) and the DCS network (first row), in particular in the DCS frequency band. The elementary transverse partitions 80 make it possible to further improve the decoupling between the two orthogonal polarizations of the same network, for example the UMTS network (third row) or the DCS network (first row).

The longitudinal partitions 90 also make it possible to better mirror the radiation patterns in the plane of the reflector, on either side of the main axis of radiation, which is perpendicular to the plane of the reflector. As shown in FIG. 4, each radiating element of the first and third rows 10A, 10B, 30A and 30B is thus surrounded by low walls which comprise two elementary transverse partitions 80, a longitudinal partition 90 and a portion of the wall 41 or 42 of the reflector. These walls form a square having a side equal to half the pitch P 1 of the first row, the pitch PI being in particular equal to the pitch P3 of the third row.

The partitions 80 and 90 may or may not be in contact with each other, except at the level of the central radiating element 20C of the second row and they may or may not be in contact with the walls of the reflector, without the operation of the antenna substantially modified.

In addition, the transverse partitions 8 may comprise main metal partitions 800 arranged between two adjacent elementary cells of the array antenna, for example between the cells 51 and 52 of FIG. 3. These main partitions 800 may extend over any the width of the reflector 4. They may be provided, for example, to mechanically stiffen the reflector, if necessary, or further improve the polarization decoupling between the adjacent radiating elements of the second row (GSM). Such main partitions 800 may have a regular polygonal shape, as shown in FIGS. 10 and 11. With reference to FIG. 11, each main transverse partition 800 may comprise a central portion 801 and two peripheral portions 802 and 803 of height less than the central part 801.

The width Le of the central portion 801 may be of the order of a quarter wave in the GSM band, for example 80mm. The height Zc of the central portion 801 may be substantially less than the height of the central element GSM, but substantially greater than the height of the walls of the reflector 4. For example, it may be equal to 50mm. The height Zp of the edges of the peripheral portions 802 and 803 may be substantially equal to the height of the walls of the reflector.

Figure 6 is a top view of a tri-band network antenna 1 according to the invention. The first row 10 comprises 9 DCS radiating elements, the third row 30 comprises 9 UMTS radiating elements and the second row 20 comprises 9 GSM radiating elements. The elementary cell 5 is of the type shown in FIG. 3, and the network comprises transverse partitions 8 and longitudinal partitions 90. The transverse partitions 8 comprise the elementary partitions 80 and the main partitions 800.

The width of the reflector, which is generally of the order of the wavelength in the lowest frequency band (GSM), may be about 260mm. The length of the reflector, which is proportional to the number of radiating elements used and the pitch of the rows, in particular of the GSM array, may be of the order of 2600 mm. The half-power aperture of the horizontal radiation patterns is substantially of the order of 65 ^° in the three GSM, DCS, UMTS frequency bands. The gains with respect to the isotropic are substantially of the order of 17dBi in the GSM band and of the order of 18dBi in the DCS and UMTS frequency bands.

The fact of using radiating elements of the same type and substantially identical in the first row (DCS) and the third row (UMTS) has the advantage of providing horizontal radiation diagrams in the symmetrical GSM band on either side of the main axis of radiation of the antenna. The particular structure of the radiating elements of the first row (DCS) and the third row (UMTS), described with reference to FIG. 7, allows these two rows to operate indifferently in one or the other of the two bands of DCS and UMTS frequencies, or in other words, to operate in an extended band (DCS-UMTS) covering both the DCS band and the UMTS band. The tri-band antenna thus has 6 accesses, including 2 GSM accesses, one for each polarization (+ 457-45 °), 2 DCS-UMTS accesses and 2 other DCS-UMTS accesses.

The invention further provides a dual-band network antenna, for example operating in GSM / UMTS or GSM / DCS.

Each elementary cell of a bi-band network according to the invention comprises a radiating element of the second row and two radiating elements of the first row. The radiating element of the second row sees the radiating elements of the first row substantially symmetrically and at a right angle. The radiating elements of the first row and the second row are broadband and double polarization dipoles of the dipole type described above. In particular, the first row may comprise UMTS radiating elements and the second row may comprise GSM radiating elements for a GSM / UMTS antenna. In a variant, the first row may comprise DCS radiating elements and the second row may comprise GSM radiating elements for a GSM / DCS antenna.

The Applicant has also observed that the architecture of the tri-band network antenna described above makes it possible to obtain such a dual-band antenna. It observed that all the radio properties of the antenna, the decoupling between the accesses, the radiation patterns, the adaptation of the impedances, etc. are substantially conserved when the radiating elements of the first row DCS (for a dual-band GSM / UMTS network) or the third row UMTS (for a dual-band GSM / UMTS network) are disconnected but physically present, that is, ie not powered. Although these observations are not fully explained to date, it is arguable that they are due to the weak mutual coupling between the radiating elements of one row and those of another row on the one hand and, on the other hand, on the other hand, the strong decoupling between the orthogonal polarizations of the radiating elements of the same row, in particular the central elements GSM, obtained thanks to the symmetrical arrangement of the elementary cells of the tri-band antenna. As a result, the load impedance between the supply terminals of the radiating elements of one of the networks of the tri-band antenna, for example the UMTS array, has only a small effect on the electrical properties of the antennas. two other networks (the GSM and DCS rows in this example) if it is zero (radiating element short-circuited at its terminals), if it is infinitely large (radiating element open at its terminals) or if it has an intermediate value.

It further observed that these electrical properties can be substantially preserved by replacing the elements of the first row (for a dual-band GSM / UMTS network) or the third row (for a dual-band GSM / DCS network) with conductive plates substantially identical peripheral shape, same size and same layout.

FIG. 9 is a view from above of an elementary cell 54 of such a dual-band antenna, in particular of a dual-band GSM / DCS antenna (or GSM / UMTS or GSM / DCS-UMTS in enlarged band). The third row (UMTS network) of the tri-band antenna according to the invention is replaced by a row of plates 300 of substantially identical shape to the radiating elements of the third row 30 described above. The first row 10 has DCS radiating elements and the second row 20 has GSM radiating elements. Thus each elementary cell 54 comprises two radiating elements DCS, 10A and 10B, two plates, 300A and 300B, and a GSM element, 20C.

The support of the plates 300A and 300B can be simply a metal or insulating column fixed to the geometric center of the plates and to the reflector, the height of the plates relative to the reflector remaining the same as the height of the radiating elements replaced. A dual-band antenna thus constituted has 2 GSM accesses and two UMTS or DCS or DCS-UMTS accesses (broadband).

The mufti-band antennas according to the invention have the advantage of providing the same quality of service as second-generation single-band antennas.

The multi-band antennas in accordance with the invention make it possible, in particular, to have a radiation along two strongly decoupled orthogonal polarizations in each frequency band, inclined by +/- 45 ° with respect to the longitudinal direction, thus favoring the reception of the signals by polarization diversity. Indeed, the mobile signals received at an antenna are altered by a multipath propagation phenomenon. Their reception following two strongly decoupled orthogonal polarizations makes it possible to have two statistically decorrelated signals whose additive processing by a diversity receiver makes it possible to significantly improve the signal-to-noise ratio on reception, which is the main measure of the quality of the communications.

Moreover, they make it possible to obtain electrically and independently depointable beams in each of the frequency bands, which enables the telecommunication operators to optimize the radio coverage of the cellular networks in each of the frequency bands (GSM, DCS and UMTS). Indeed, the inclination of the beam ("tilt") of one of the networks composing the multi-band antenna can be achieved by an electrical phase shift means, which consists in creating a constant phase shift between the successive radiating elements of this network. , thus avoiding a mechanical inclination of the antenna as a whole. Such an electrical means is proposed for example in the French patent application No. 0307483.

The multi-band antennas according to the invention also have the advantage of providing good isolation between the orthogonal accesses of the same frequency band and between the inter-band access, which limits the interference between the high power signals. transmitted by one of the accesses and signals of very low power received by another access of the antenna, interference harmful to the quality of the communications.

Finally, they can have a height and width similar to those of second-generation single-band antennas operating in the GSM band already deployed, which minimizes the visual impact but also the impact on the environment. Indeed, an antenna according to the invention integrates two or three elementary antennas operating in two or three different communication systems (for example GSM / DCS / UMTS or GSM / DCS). The number of antennas required is thus minimized for a given base station. Different operators can also share the same antenna and consequently the number of sites of implantation of such stations is also reduced.

Figures 12 and 13 are top views of an alternative embodiment of the radiating element of the second row 20C. The Applicant has found that such a radiating element has properties that improve the overall performance of the network antenna according to the invention.

FIG. 12 shows a radiating element inscribed in a square represented by a dashed line 71. The radiating element represented comprises two dipoles D1 and D2 Each dipole D1 and D2 comprises a pair of coplanar conductive plates, of the same geometry respectively (DU, D12} and {D21, D22} Each plate DU, D12, D21 and D22 has an axis of symmetry The two plates D1 and D12 of the dipole D1 are positioned with their axes of symmetry substantially aligned along the same alignment axis Δ Likewise, the two plates D21 and D22 of the dipole D2 are positioned with their axes of symmetry substantially aligned along the same alignment axis Δ '.

The alignment axes Δ and Δ 'of the dipoles D1 and D2 intersect at right angles at a point of crossing O "situated between the four plates of the radiating element As shown in FIGS. 12 and 13, the alignment axes Δ and Δ 'substantially coincide with the diagonals of the square 71" and the cross point O "substantially coincides with the center of the square 71 ".

Each plate of a dipole, for example DU, has two lateral branches p 11 and p 11 'arranged to form a V, substantially open at 90 °, the point of which is situated in the vicinity of the point of intersection O "of the The lateral branch pl i extends in particular along the axis VV while the lateral branch pl i 'is substantially perpendicular to the axis V. Each plate of a dipole further comprises an intermediate branch designated hereinafter after "strand", for example Bl, which extends between the two lateral branches, along the axis of alignment of the dipole, for example Δ.

Similarly, the plate D12 has two lateral branches pi 2 and pi 2 'and a strand b21, the plate D22 has two lateral branches p12 and p2' and a strand b21 and the plate D22 has two lateral branches p22 and p22 'and a strand b22.

In addition, the lateral branches of each plate, for example the branches pi 1 and pi 1 'of the plate DU, can be hollowed out. Thus, they may comprise perforations 79 'for reducing their weight In the embodiment of FIG. 12, the perforations of the two branches intersect at the corresponding alignment axis Δ or Δ'.

The two pairs of dipoles D1 and D2 thus generate two orthogonal electric fields one to the other. The pair D1 1, D12 generates an electric field parallel to the axis Δ and the pair D2 1, D22 generates an electric field parallel to the axis Δ '. The polarization planes are at an angle of +/- 45 ° with respect to the longitudinal axis VV of FIGS. 12 and 13, which passes in the space between plates D1 and D21 on the one hand, and D12 and D22 d 'somewhere else.

The radiating element of FIGS. 12 and 13, when used in the second row 20 (GSM) of FIG. 4, is arranged in the network so that its alignment axes Δ and Δ 'which define the planes of polarization form an angle of +/- 45 ° with respect to the longitudinal axis AA 'of the network.

The use of such a radiating element in the second row GSM improves the horizontal radiation patterns of the first row DCS and the third row UMTS, with respect to embodiments that use the radiating element of Figure 2B or Figure 8.

In particular, the radiating element of FIGS. 12 and 13 makes it possible to have a low dispersion of the half-power aperture of the horizontal diagrams, as a function of frequency, in the UMTS and DCS bands. It also generates a relatively small strabismus in its GSM frequency band and improves the symmetry of the horizontal diagrams with respect to the main axis, in the UMTS and DCS frequency bands and for electrical inclination angles of the beam. overall antenna approximately between 0 ° and 10 ° to the ground.

The radiating element of FIG. 13 is a variant of the radiating element of FIG.

According to this variant, each lateral branch of a plate, for example the lateral branch pl i of the plate DU, comprises a recess 79 "(1) disjoined from the recess 79" (2) of the other lateral branch pl i from the same plate. The end of the recess 79 "(1) and the end of the recess 79" (2), in the connection zone of the lateral branches, are separated from each other by at least a distance equal to the width d of strand b 11 of the plate.

This separation forms a conductive connection which makes it possible to reinforce the mechanical strength of the strands. The general structure of the radiating element of FIG. 13 then corresponds substantially to the superposition of the radiating strands of the radiating element of FIG. 8 and the radiating element of FIG. 2B.

The radiating elements of FIG. 2B, FIG. 8, FIG. 12 and FIG. 13 may be powered or excited at their center in a similar manner to create the two orthogonal polarizations in the two diagonal directions of the radiating elements. The power supply may comprise electrical connections 60, as shown in FIG. 7, and a balun which also serves as a support leg for the radiating element, as shown in FIG. 5. Such a power supply device is described by FIG. example in the French patent application FR 2 840 455, filed in the name of the Applicant.

The radiating elements of FIG. 2B, FIG. 8, FIG. 12 and FIG. 13 considered in isolation have close radio properties, namely an opening of substantially equal diagrams (around 65 °), a low cross-polarization rate, a strong decoupling of the two orthogonal polarization channels in a wide frequency band, and a good impedance matching in a wide frequency band.

It has been found that the radiating elements of FIG. 12 and FIG. 13, when used in the central row of the tri-band grating according to the invention, provide more satisfactory performance than the radiating elements of FIG. FIG. 2B and FIG. 8. In fact, it has been found that the radiating elements of FIG. 12 and FIG. 13 do not have certain disadvantages of the radiating elements of FIG. 2B and FIG.

More precisely, it has been observed that the radiating element of FIG. 2B "weakly disturbs the radiation of the UMTS and DCS rows. This seems to be due to the fact that the radiating strands of such an element are oriented at 45 ° with respect to the UMTS and DCS row axis, as shown in Figure 3. In addition, it has been observed that the radiation patterns present, at least in certain situations, undesirable asymmetry phenomena in the horizontal plane and in the GSM frequency band, for significant angles of inclination of the beam radiated by the antenna.

Furthermore, it has been found that the radiating element of FIG. 8 generates radiation diagrams which exhibit relatively low asymmetry phenomena in the horizontal plane, in its GSM frequency band.5 It has appeared that such an element radiating "disrupts" the radiation of the DCS and UMTS arrays, which results in a relatively large dispersion of the aperture of radiation patterns, in the horizontal plane, in the DCS and UMTS.0 frequency bands. Dissymmetry phenomena are now to be schematically described with reference to Figure 14, which illustrates an example of radiation patterns in the horizontal plane of a dual linear polarization antenna and orthogonal. In FIG. 14, the curve C1 corresponds to the radiation pattern in the horizontal plane for the + 45 ° polarization path and the curve C2 corresponds to the radiation pattern in the horizontal plane for the polarization path - 45 °.

The dissymmetry phenomena may comprise a "tracking" phenomenon which results in a difference in the power levels transmitted (or received) by the antenna, in two directions of the horizontal plane, symmetrical with respect to the main axis of radiation XX '.

For example, on the curve C1 for the + 45 ° polarization channel:

- the relative power level is approximately - 2.2 dB, in the azimuth direction - 30 °, while the power level is - 0.9 dB, in the direction of symmetrical azimuth + 30 °, which corresponds to a "tracking" T1 of about 1.3 dB;

- the power level is approximately - 6.7 dB, in the azimuth direction - 60 °, while the power level is - 4 dB, in the symmetrical azimuth direction + 60 °, which corresponds to a "tracking" Tl of about 2.7 dB.

In particular, it has been observed that the phenomena of asymmetry increase with the angle of electrical inclination of the antenna beam.

The phenomena of dissymmetry may further include a phenomenon of "strabismus" ("squint" in English). Strabismus appears when the maximum radiation is not in the main axis. It results in the difference in azimuth S between the maximum power point of the radiation pattern and the main axis XX '.

Conventionally, for a linear polarization antenna, the dissymmetry phenomena mentioned above are only found in the planes other than the main polarization planes of the antenna. The main plane of polarization of the antenna comprises the plane containing the electric field E, said "plane E", and the plane containing the magnetic field H, said "plane H". The plane E and the plane H are orthogonal.

For example, in FIG. 3, one of the main polarization planes contains the Δ axis while the other plane contains the Δ 'axis. The main polarization planes are therefore inclined +45 ^° and -45 ° relative to the vertical plane containing the axis AA '. The phenomena of dissymmetry can therefore be observed in the horizontal plane or in the vertical plane containing the axis AA '. In particular, it has become apparent that asymmetry phenomena are generally more troublesome in the horizontal plane than in the vertical plane.

Furthermore, it has been observed that the radiation pattern C1 of the +45 ° polarization channel is generally symmetrical with the C2 radiation pattern of the -45 ° polarization path with respect to the main axis. Under such conditions, the tracking T1 on the radiation pattern C1 of the +45 ° polarization channel, for a given direction, is generally substantially equal to the T2 tracking on the C2 radiation pattern of the other polarization channel. -45 °, for the symmetrical direction.

This has the effect that the difference in power level between two points A1 and A of the radiation pattern of the same polarization path, for example C1, located in respective directions α and a 'symmetrical with respect to the main axis is substantially equal to the difference in power level, between a point A1 of the radiation pattern C1 of one of the polarization channels and a point A2 of the radiation pattern C2 of the other polarization channel, all located two in one of the directions a and a '. For example: - in symmetrical directions + 30 ° and - 30 °, the difference in power level of the polarization channel at + 45 ° is about 1.3 dB, and in the +30 ° direction, the difference of the polarization channel power level at + 45 ° and the power level of the polarization channel at -45 ° is also 1.3 dB.

As a result, the difference in power level, between a point A1 of the radiation pattern C1 of one of the polarization channels and a point A2 of the radiation pattern C2 of the other polarization channel, both located in the same direction. direction is generally also referred to as a "pursuit" phenomenon.

Some radiating elements have, according to their structure, a phenomenon of "pursuit" more marked than others, in the frequency band where they operate. In the tri-band embodiment of the invention, it has been found that the dissymmetry phenomena in the GSM band depend on the type of GSM radiating element used in the central row. In the UMTS and / or DCS bands, it appears that the asymmetry phenomena result from the general structure of the antenna itself, and in particular from the network configuration. It has been observed that it is possible to mitigate these asymmetry phenomena by using the radiating elements of FIG. 12 or FIG. 13 in the central row of the grating.

The radiating elements of FIGS. 12 and 13 may also be used as DCS radiating elements in the first row and as UMTS radiators in the third row 30. For this purpose, these radiating elements must be sized to operate in the corresponding frequency band. (DCS and / or UMTS). In the embodiment of FIG. 15 which shows an exemplary elementary cell 55, the radiating elements of FIGS. 12 and 13 are used in the three rows 10, 20 and 30. It has been found that such an embodiment provides satisfactory performance.

Certain elements described in the context of the present invention may be of particular interest when considered separately. This is the case in particular of the generally cross-shaped radiating element described with reference to FIG. 8 which has broadband electrical properties in terms of impedance and decoupling between the two orthogonal and radiation polarizations. This is also the case of the radiating elements of FIGS. 12 and 13, which not only possess broadband electrical properties in terms of impedance and decoupling between the two orthogonal and radiation polarizations but also generate a "tracking" phenomenon. 5 acceptable.

Furthermore, the invention is not limited to the embodiments described above. It encompasses all the embodiments that may be envisaged by those skilled in the art. In particular, the radiating elements of the first, second and third rows are not limited to the types of broadband and double polarization crossed dipoles described.

Claims

claims

An array antenna comprising a ground plane (4) on which is mounted at least: a first array of radiating elements (10) capable of operating in a first frequency band and a second array of elements radiating (20), adjacent and parallel to the first row, and capable of operating in a second frequency band, the first and second rows being arranged to form a set of elementary cells (5, 51, 52, 53), characterized in that each elementary cell (5) comprises a radiating element of the second row (20C) and two adjacent radiating elements of the first row (10A, 10B), the radiating element of the second row being arranged to see the two radially adjacent elements of the first row substantially symmetrically and at a right angle, and in that each radiating element of the first and second rows5 comprises two dipoles s arranged to operate in broadband and dual polarization.

An array antenna according to claim 1, characterized in that a third row (30) of radiating elements is further mounted on the ground plane, parallel to the first row (10) and the second row (20). ), the third row being capable of operating in a third frequency band and arranged so that the second row is interposed substantially equidistant between the first row and the third row, and in that each elementary cell further comprises two elements in the third row (30A, 30B), the radiating element of the second row being arranged to see the two adjacent radiating elements of the third row substantially symmetrically and at a right angle.

3. Network antenna according to claim 2, characterized in that each radiating element of the third row comprises two crossed dipoles arranged to operate in broadband and double polarization.

4. Antenna network according to one of claims 1 to 3, characterized in that the ratio between the pitch between two adjacent radiating elements of the second row (P2) and the pitch between two adjacent radiating elements of the first row (PI) is substantially between 1.5 and 2.5.

5. Antenna network according to one of claims 2 to 4, characterized in that the pitch 5 between two adjacent radiating elements of the third row (P3) is substantially equal to the pitch between two adjacent radiating elements of the first row ( PI).

6. An array antenna according to one of claims 2 to 5, characterized in that the spacing between the first row and the third row (Q) is substantially equal to the step of the first row (PI).

7. Antenna network according to one of claims 1 to 6, characterized in that each dipole of a radiating element of the first row comprises a pair of coplanar conductive plates (2a, 2b, 2c, 2d), same geometry and in generally square form, the two plates of each pair being positioned with their diagonals substantially aligned on the same alignment axis for each pair (Δ ₃ , Δ ₄ ), the respective alignment axes of the two pairs of plates intersecting at right angles to a crossing point (O) located between the plates of each pair, and in that the radiating elements of the first row are arranged so that the respective alignment axes of the two pairs of plates o are oriented at 45 ° to the axis defined by the first row.

8. An array antenna according to claim 7, characterized in that the plates (2a, 2b, 2c, 2d) of a radiating element of the first row are internally recessed. 5

9. An array antenna according to one of claims 2 to 8, characterized in that the radiating elements of the first row (10) and the third row (30) are of the same type.

10. An array antenna according to claim 9, characterized in that the radiating elements of the first row (10) and the third row (30) are substantially identical.

11. Antenna network according to one of the preceding claims, characterized in that each dipole of a radiating element of at least one of the rows (10, 20, 30) comprises a pair of coplanar conductive plates, of the same geometry, each plate having an axis of symmetry and the two plates of each dipole being positioned with their axes of symmetry substantially aligned along the same alignment axis (Δ, Δ ', Δ ₃ ', Δ ₄ '), the alignment axes of the two dipoles intersecting at right angles at a point of intersection (O, O ', O ") situated between the plates of each pair, and in that the radiating elements of said row are arranged so that the respective alignment axes (Δ, Δ') of the two pairs of plates are substantially oriented from 45 ° with respect to the axis defined by the row.

12. An array antenna according to claim 11, characterized in that each plate (2a ', 2b', 2c ', 2d') of a dipole of the radiating element has a general shape of triangle and comprises a recess towards the inside from the base of the triangle.

Antenna array antenna according to claim 12, characterized in that each plate (2a ', 2b', 2c ', 2d') of a dipole of the radiating element is internally recessed.

14. An array antenna according to claim 11, characterized in that each plate of a dipole of the radiating element comprises a strand extending from the point of intersection (O ') along the axis of alignment of the dipole (Δ, Δ ').

15. An array antenna according to claim 11, characterized in that each plate of a dipole of the radiating element has two lateral branches (pi 1, pi 1, pi 2, pi 2 ', p21, p21', p22, p22 ') arranged to form a V, substantially open at 90 °, the tip of which is situated in the vicinity of the crossing point (0 ") of the radiating element, and an intermediate branch (b1, M2, b21 , b22) extending between the two lateral branches, along the axis of alignment of the dipole.

16. An array antenna according to claim 15, characterized in that each of the lateral branches (fold, pl ', pl2, pl2', p21, p21 ', p22, p22') of a plate of the radiating element comprises an inner recess (79 ", 79" (1), 79 "(2)).

17. An array antenna according to claim 16, characterized in that the recesses (79 "(1), 79" (2)) of the two lateral branches of each plate are disjoint, the distance (d) separating the respective recesses of the two lateral branches being at least equal to the width of the intermediate branch of the plate, at the tip of the V formed by said lateral branches.

18. An array antenna according to one of the preceding claims, characterized in that each radiating element of the first row is separated from an adjacent radiating element of the same row by an elementary transverse partition (80). 0

19. An array antenna according to one of claims 2 to 18, characterized in that each radiating element of the third row is separated from an adjacent radiating element of the same row by an elementary transverse partition (80).

20. Network antenna according to one of claims 18 and 19, characterized in that each elementary transverse wall is placed substantially equidistant from the radiating elements that it separates.

21. An array antenna according to one of claims 18 to 20, characterized in that the height of each elementary transverse wall is less than the height of the radiating elements o it separates.

22. An array antenna according to one of claims 2 to 21, characterized in that each radiating element of the first row is separated from the radiating element vis-à-vis the third row by a longitudinal partition (90) .5

23. An array antenna according to claim 22, characterized in that each longitudinal partition is placed substantially equidistant from the radiating elements which it separates. 0

24. An array antenna according to claim 23, characterized in that the height of each longitudinal partition is less than the height of the radiating element of the first row and the height of the radiating element vis-à-vis the the third row.

25. Antenna network according to one of the preceding claims, characterized in that the frequency band of the first row (10) is substantially greater than the frequency band of the second row (20).

26. An array antenna according to one of claims 2 to 25, characterized in that the frequency band of the third row (30) is substantially greater than the frequency band of the second row (20).

Network antenna according to one of the preceding claims, characterized in that the ratio between the center frequency of the frequency band of the first row (10) and the center frequency of the frequency band of the second row (20). ) is substantially between 1.5 and 2.5.

28. An array antenna according to one of claims 2 to 27, characterized in that the ratio between the center frequency of the third row frequency band (30) and the center frequency of the second row frequency band (20) is substantially between 1.5 and 2.5.