EP0520908A1 - Linear antenna array - Google Patents

Abstract

The present invention relates to a linear array antenna composed of a radiating element formed by an alignment of annular slots (21) energised by at least one buried distributor element (14) enclosed within the radiating element and therefore invisible from the radiating face of the said antenna; so as to preclude any stray radiation arising from the said distributor. Application in particular to the field of space telecommunications. <IMAGE>

Description

The invention relates to a high efficiency linear array antenna. Such an antenna is particularly advantageous in the field of synthetic aperture radars (SAR) used in particular in the space field.

• un gain élevé pour minimiser la puissance émise par le radar ;
• un lobe très fin dans le plan d'azimut (parallèle à l'orbite), de largeur variable pour maintenir une faute constante au sol dans le plan d'élévation (perpendiculaire à la trace) ;
• un balayage très élevé dans le plan d'élévation pour pouvoir accéder très rapidement à n'importe quel site (mission de surveillance) et viser les mêmes sites-test à faible puis forte incidence à 24 heures d'intervalle (observation de la végétation, séparation des effets du sol et du feuillu) ;
• un balayage peu élevé dans le plan d'azimut pour suivre une même zone afin d'augmenter la longueur synthétique de l'antenne et d'obtenir un pouvoir de résolution de quelques mètres parallèlement à l'orbite ;
• deux polarisations linéaires orthogonales (horizontale H et verticale V) pour séparer la rétrodiffusion du sol et celle de la couverture végétale ;
• un niveau de polarisation croisée très faible sur le domaine de balayage de l'antenne ;
• des éléments rayonnants de coût et de masse les plus faibles possibles vu leur grand nombre sur l'antenne.

The antennas used for such radars must have the following properties:

• a high gain to minimize the power emitted by the radar;
• a very fine lobe in the azimuth plane (parallel to the orbit), of variable width to maintain a constant fault on the ground in the elevation plane (perpendicular to the trace);
• a very high sweep in the elevation plan to be able to access any site very quickly (monitoring mission) and target the same test sites at low then high incidence at 24 hour intervals (observation of vegetation, separation of soil and hardwood effects);
• a low scanning in the azimuth plane to follow the same area in order to increase the synthetic length of the antenna and to obtain a resolving power of a few meters parallel to the orbit;
• two orthogonal linear polarizations (horizontal H and vertical V) to separate the backscattering of the soil and that of the plant cover;
• a very low cross-polarization level over the antenna scanning range;
• the lowest possible cost and mass radiating elements given their large number on the antenna.

These specifications can only be held with a large direct radiation antenna made up of a very large number of planar radiating elements. Indeed, the lobe theory of networks shows that, as soon as the maximum antenna depointing angle (ϑ _max ) exceeds a few degrees, the distance d between the phase control points on the antenna must verify the relationship : $$\text{d ≦}\frac{\text{λ}}{{\text{1 + sin ϑ}}_{\text{max}}\text{- sin (Bw / 2)}}$$

With Bw width at the base of the main lobe of the array antenna when this is pointed at normal to the antenna.

If this condition is not observed, network lobes appear in the radiation pattern. Their level is higher than that required for the secondary lobes, which reduces the gain of the antenna and creates ambiguities.

• Pour le balayage en élévation, le pas doit être proche de λ/2.
• Pour le balayage en azimut, le faible dépointage est compatible avec un pas proche de λ .

Such a condition has the following consequences:

• For elevation scanning, the step must be close to λ / 2.
• For the azimuth scan, the weak deflection is compatible with a step close to λ.

However, the antenna having a long length, to generate a fine lobe in the plane this condition leads to a prohibitive number of sources.

• un réseau raréfié qui rompt la périodicité de la loi d'éclairement et par conséquent détruit les lobes de réseau mais provoque une perte de gain très élevée.
• un réseau à pas non régulier, obtenu en augmentant progressivement la distance entre commandes de phase lorsqu'on s'éloigne du centre de l'antenne. Cette solution, pour ne pas baisser le gain de l'antenne, nécessite de réaliser de nombreux types d'éléments rayonnants différents, ce qui interdit le découpage de l'antenne en sous-panneaux tous identiques et augmente considérablement le coût.
• un réseau formé de sous-réseaux dont la loi d'éclairement dans le plan d'azimut doit être la plus proche possible de celle d'une ouverture uniforme de même longueur. Dans ce cas, pour une visée dans l'axe, les lobes de sous-réseaux de l'antenne sont annulés par les premiers zéros du diagramme du sous-réseau. Cette condition n'est plue vérifiée lors d'un faible balayage en azimut. On détermine alors, pour l'angle de dépointage azimut maximum, la longueur maximale des sous-réseaux pour maintenir les lobes de réseaux à un niveau inférieur à celui des lobes latéraux.

Three solutions are then possible to reduce the number of orders:

• a rarefied network which breaks the periodicity of the law of illumination and consequently destroys the network lobes but causes a very high loss of gain.
• a non-regular pitch network, obtained by gradually increasing the distance between phase commands when moving away from the center of the antenna. This solution, in order not to lower the gain of the antenna, requires making many different types of radiating elements, which prohibits the cutting of the antenna into all identical sub-panels and considerably increases the cost.
• a network formed of sub-networks whose law of illumination in the azimuth plane must be as close as possible to that of a uniform aperture of the same length. In this case, for an aiming in the axis, the sub-array lobes of the antenna are canceled by the first zeros of the diagram of the sub-array. This condition is no longer verified during a weak azimuth scan. We then determine, for the maximum azimuth depointing angle, the maximum length of the sub-networks to keep the network lobes lower than that of the side lobes.

Similarly, it can be seen that, if no azimuth deflection is specified, a subarray of length equal to that of the antenna is sufficient. In practice, the length of the sub-networks is limited by the size of the antenna panels (1 to 3 meters) to be compatible with the space taken up by the launcher's cap.

• . Les guides à fentes, tels que décrits dans l'article intitulé "the planar array antennas for the european remote sensing satellite ERS-1" de Robert Peterson et Per Ingvarson publié dans "Proceedings of IGARSS 1988" (pages 289 à 294), utilisés notamment sur ERS-1 (1991), RADARSAT (1995) et SIR-C (X-SAR), malgré leurs faibles pertes ne permettent pas de réaliser aisément la bipolarisation car il faudrait intercaler deux guides différents rayonnant l'un en polarisation H et l'autre en polarisation V dans le pas d'un réseau très réduit à cause du fort balayage en élévation. De plue, si un balayage en azimut est spécifié, pour limiter le niveau des lobes de réseau, ces guides doivent être coupés en petits tronçons ce qui rend notablement plus complexe leur réalisation et diminue leur intérêt. Cette technologie n'est intéressante que pour les radars possédant une polarisation en bande C ou dans des fréquences supérieures, sans balayage dans le plan d'azimut.
• . Les éléments rayonnants imprimés sur nid d'abeilles en bandes L et S, tels que décrits dans l'article intitulé "SEASAT and SIR-A microstrip antennas" de L.R. Murphy paru dans "Proceedings of Workshop on printed antennas technology" (Las Cruces; 1979; pages 18-1 à 18-20), utilisés sur SEASAT (1978), SIR-A (1981), SIR B (1984) et J. ERS (japonais 1992) sont plus légers que les guides et permettent une bipolarisation. Cependant leurs pertes linéiques sont élévées. Ceci limite donc la longueur des sous-réseaux à quelques longueurs d'onde (10 λ au maximum); ces pertes ne permettant pas d'obtenir un éclairement uniforme sur le sous-réseau.

Currently, all of these specifications are not held for the radiating elements of synthetic aperture radars. Indeed :

• . The slotted guides, as described in the article entitled "the planar array antennas for the european remote sensing satellite ERS-1" by Robert Peterson and Per Ingvarson published in "Proceedings of IGARSS 1988" (pages 289 to 294), used notably on ERS-1 (1991), RADARSAT (1995) and SIR-C (X-SAR), despite their low losses do not allow bipolarization to be easily achieved because two different guides should be inserted, radiating one in H polarization and the other in V polarization in the pitch of a very small network due to the strong sweep in elevation. In addition, if an azimuth scan is specified, to limit the level of the network lobes, these guides must be cut into small sections, which makes their construction considerably more complex and reduces their interest. This technology is only interesting for radars having a polarization in C band or in higher frequencies, without scanning in the azimuth plane.
• . The radiating elements printed on honeycomb in L and S bands, as described in the article entitled "SEASAT and SIR-A microstrip antennas" by LR Murphy published in "Proceedings of Workshop on printed antennas technology" (Las Cruces; 1979; pages 18-1 to 18-20), used on SEASAT (1978), SIR-A (1981), SIR B (1984) and J. ERS (Japanese 1992) are lighter than the guides and allow bipolarization. However their linear losses are high. This therefore limits the length of the sub-networks to a few wavelengths (10 λ maximum); these losses do not make it possible to obtain uniform illumination on the sub-network.

The object of the linear array antenna of the invention is to satisfy all of these specifications.

The invention proposes, in fact, a linear array antenna composed of a radiating element formed by an alignment of annular slots excited by at least one distributor element, characterized in that this distributor element is a buried distributor, included in the radiating element, and therefore invisible from the radiating face of said antenna; so as to avoid any stray radiation from said distributor. Such an antenna is intended to radiate (or receive) simultaneously in two linear polarizations, or to radiate (or receive) simultaneously with two circular polarizations (right and left).

In an advantageous embodiment using only three levels of conductors, said antenna comprises an integrated radome and a distributor stage. Advantageously, said antenna only comprises substrates of small thickness, and therefore of low mass, spaced apart by conductive spacers without mechanical continuity in the plane of the network, so as to avoid differential expansions with the substrates. Advantageously, said antenna comprises shields, between polarizations of a sub-network and between adjacent sub-networks, radiating elements and distributors so as to avoid any coupling between them.

In a particular embodiment of the invention, the antenna is a broadband antenna using resonant discs etched on the distributor stage, resonating at a frequency close to that of the annular slots.

In another particular embodiment of the invention; the antenna is a dual-band antenna using engraved resonant disks, on the distributor stage, resonating at a frequency distant from that of the annular slots.

• de faibles pertes,
• un faible niveau de polarisation croisée sur le domaine de balayage,
• une utilisation en bipolarisation : l'antenne réseau selon l'invention pouvant rayonner (ou recevoir) simultanément dans deux polarisations linéaires ou circulaires (droite et gauche);
• un large domaine de balayage en élévation;
• une longueur importante des sous-réseaux;
• un gain élevé;
• des pertes par dépointage similaires dans les deux polarisations.

Such an antenna has many advantages; In particular, it provides:

• low losses,
• a low level of cross polarization on the scanning domain,
• use in bipolarization: the array antenna according to the invention can radiate (or receive) simultaneously in two linear or circular polarizations (right and left);
• a wide sweep range in elevation;
• a significant length of the sub-networks;
• high gain;
• similar depointing losses in the two polarizations.

• la figure 1 illustre une vue éclatée d'un tronçon de sous-réseau d'une réalisation de l'antenne selon l'invention;
• les figures 2 et 3 illustrent deux vues éclatées de tronçons de sous-réseaux de réalisations de deux variantes de l'antenne selon l'invention;
• les figures 4 et 5 illustrent respectivement les diagrammes de rayonnement pour les polarisations horizontale et verticale,
• les figures 7 et 8 illustrent deux autres variantes de l'antenne selon l'invention.

The characteristics and advantages of the invention will become apparent from the description which follows, by way of nonlimiting example, with reference to the appended figures in which:

• Figure 1 illustrates an exploded view of a subarray section of an embodiment of the antenna according to the invention;
• Figures 2 and 3 illustrate two exploded views of sections of sub-arrays of embodiments of two variants of the antenna according to the invention;
• FIGS. 4 and 5 respectively illustrate the radiation diagrams for the horizontal and vertical polarizations,
• Figures 7 and 8 illustrate two other variants of the antenna according to the invention.

The antenna according to the invention is a linear array antenna composed of an alignment of annular slots excited by at least one buried distributor element, included in the radiating element, and therefore invisible from the radiating face of said antenna; so as to avoid any stray radiation from said distributor.

Such an antenna is designed to radiate (or receive) simultaneously in two linear polarizations; or to radiate (or receive) simultaneously in two circular polarizations (right and left).

• une demi-coquille inférieure 10 en matériau conducteur (aluminium, fibre de carbone métallisée, ou plastique conducteur par exemple) dans laquelle est réalisée la cavité 11 des éléments rayonnants et le conduit 12 des répartiteurs des deux polarisations;
• un premier substrat 13 très mince (par exemple d'environ 0,1 mm d'épaisseur), possédant un coefficient de dilatation proche de celui du matériau des demi-coquilles, sur lequel sont gravés, en face supérieure, les répartiteurs H et V 14;
• une demi-coquille supérieure 15 dans laquelle est réalisée le conduit 16 des répartiteurs des deux polarisations et percée (17) la partie supérieure de la cavité.
• un second substrat 18 identique au premier substrat dont une face supérieure 19 est entièrement décuivrée et l'autre 20 contient les fentes annulaires gravées 21. Ce substrat a pour but de centrer les fentes annulaires au-dessus des cavités et d'assurer une fonction de radôme.

FIG. 1 illustrates a sub-network section of such an antenna which comprises:

• a lower half-shell 10 of conductive material (aluminum, metallized carbon fiber, or conductive plastic by example) in which is produced the cavity 11 of the radiating elements and the conduit 12 of the distributors of the two polarizations;
• a first very thin substrate 13 (for example about 0.1 mm thick), having a coefficient of expansion close to that of the material of the half-shells, on which are etched, on the upper face, the distributors H and V 14;
• an upper half-shell 15 in which is made the conduit 16 of the distributors of the two polarizations and pierced (17) the upper part of the cavity.
• a second substrate 18 identical to the first substrate, one upper face 19 of which is completely uncoppered and the other 20 of which contains the etched annular slots 21. The purpose of this substrate is to center the annular slots above the cavities and to provide a function of radome.

For an operation in "suspended triplate" the two half-shells 10 and 15 must be connected to the same potential. This is ensured by the first substrate thanks to the metallization of the facing surfaces with the two half-shells and to the presence of a large number of metallized holes located on the periphery of the cavities, connecting these two surfaces.

• de faibles pertes : le répartiteur en technologie triplaque suspendu sur substrat très mince et de faible tangente de perte avec des parois argentées assure parfaitement cette fonction. L'élément rayonnant est réalisée dans la même technologie.
• un faible niveau de polarisation croisée sur le domaine de balayage : Grâce à l'utilisation d'une technologie triplaque, on évite le rayonnement parasite des lignes et des coudes d'un répartiteur gravé sur le plan rayonnant. On génère un diagramme différence sur la polarisation croisée, grâce à des répartiteurs symétriques par rapport à l'axe du réseau. Ceci permet d'annuler le niveau, déjà faible, de polarisation croisée des fentes annulaires sur tout le domaine de balayage.
• une utilisation en bipolarisation : l'excitation des fentes annulaires est obtenue par deux polarisations linéaires orthogonales ou par deux polarisations circulaires en insérant un coupleur 3 dB à deux sorties déphasées de 90° au centre du sous-réseau rayonnant.
• un large domaine de balayage en élévation : L'utilisation de canaux pour les répartiteurs permet une très grande compacité sur le sous-réseau sans risque de couplage entre polarisations. Le pas dans le plan d'élévation est ainsi considérablement réduit et autorise un large dépointage en élévation.
• une longueur importante des sous-réseaux : Les faibles pertes du répartiteur assurent un éclairement quasi-uniforme en amplitude et en phase du sous-réseau. Sa longueur n'est limitée que par la modification de la loi d'éclairement sur la bande passante du sous-réseau.
• un gain élevé : une adaptation parfaite de la directivité de l'élément rayonnant à celle de la maille du réseau est obtenue grâce à l'utilisation d'une cavité concentrique dont le diamètre permet d'ajuster très précisément cette grandeur. Le domaine de réglage est compris entre 6 dBi pour des fentes quasi-découplées de la cavité dont le périmètre, à la résonance, correspond à la largeur d'onde guidée λ_g, et 9 dBi pour des fentes résonant à
  
  1,5 λ_g On couvre ainsi le domaine des mailles carrées de 0,56 à 0,8 λ requises pour de tels réseaux.
• des pertes par dépointage similaires dans les deux polarisations : Les fentes annulaires sur cavité ont un diagramme à quasi-symétrie de révolution. Dans le plan d'élévation, le diagramme est donc invariant avec la polarisation, ce qui garantit les mêmes pertes sur le domaine de balayage.
• des sous-réseaux de faibles coût et masse.

Such an antenna has been produced and tested on sub-arrays of 10, 12 and 24 annular slots. The adequacy between the specifications of the sub-network and its various constituent elements is given below; such a sub-network makes it possible to obtain:

• low losses: the distributor in triplate technology suspended on very thin substrate and low loss tangent with silver walls perfectly performs this function. The radiating element is made in the same technology.
• a low level of crossed polarization on the scanning domain: Thanks to the use of a triplate technology, the parasitic radiation of the lines and the elbows of a distributor engraved on the radiating plane is avoided. A difference diagram on the cross polarization is generated, by means of distributors symmetrical with respect to the axis of the network. This makes it possible to cancel the already low level of cross-polarization of the annular slots over the entire range of scanning.
• use in bipolarization: the excitation of the annular slots is obtained by two orthogonal linear polarizations or by two circular polarizations by inserting a 3 dB coupler with two outputs phase shifted by 90 ° in the center of the radiating sub-network.
• a large sweep range in elevation: The use of channels for the distributors allows a very great compactness on the sub-network without risk of coupling between polarizations. The pitch in the elevation plane is thus considerably reduced and allows a large depointing in elevation.
• a significant length of the sub-networks: The low losses of the distributor ensure an almost uniform illumination in amplitude and phase of the sub-network. Its length is only limited by the modification of the lighting law on the bandwidth of the subnetwork.
• a high gain: a perfect adaptation of the directivity of the radiating element to that of the mesh of the network is obtained thanks to the use of a concentric cavity whose diameter makes it possible to very precisely adjust this quantity. The adjustment range is between 6 dBi for quasi-decoupled slots of the cavity whose perimeter, at resonance, corresponds to the guided wavelength λ _g , and 9 dBi for slots resonating at
  
  1.5 λ _g One thus covers the field of the square meshes of 0.56 to 0.8 λ required for such networks.
• similar depointing losses in the two polarizations: The annular slots on the cavity have a quasi-symmetry of revolution diagram. In the elevation plane, the diagram is therefore invariant with the polarization, which guarantees the same losses on the scanning domain.
• low cost and mass subnets.

• . un substrat inférieur 25 ayant :
  • * une face inférieure 26 localement cuivrée (zone 27) autour des connecteurs d'alimentation RF 28 et 29, collée au panneau structural;
  • * une face supérieure 30 entièrement cuivrée (sauf épargne d'alimentation RF), utilisée comme plan de masse;
  • * des trous métallisés 31 reliant les deux surfaces.
• . un substrat intermédiaire 32 ayant :
  • * une face inférieure 33 cuivrée sur les surfaces en regard avec les cloisons formées par des entretoises conductrices 41;
  • * une face supérieure 34 cuivrée sur les surfaces 42 en regard avec les cloisons et sur les circuits 35 de répartition H et V;
  • * des trous métallisés 44, assurant la liaison électrique entre les deux plans de masse, qui sont disposés à la verticale des cloisons formées par les entretoises 41; (les troue 36 servent pour la soudure des âmes des connecteurs 28 et 29 alimentant les répartiteurs H et V 35);
• . un substrat supérieur 37 ayant :
  • * une face inférieure 38 entièrement cuivrée sauf aux endroits des fentes annulaires 39 et servant de plan de masse au triplaque;
  • * une face supérieure 40 entièrement décuivrée et faisant fonction de radôme.

A particularly advantageous embodiment of the antenna according to the invention is shown in Figure 2. To obtain a low mass, the antenna consists of three substrates, about 0.1 mm thick, whose functions are following:

• . a lower substrate 25 having:
  • * a lower face 26 locally coppery (zone 27) around the RF power connectors 28 and 29, glued to the structural panel;
  • * a fully copper-coated upper face 30 (except for RF power savings), used as a ground plane;
  • * metallized holes 31 connecting the two surfaces.
• . an intermediate substrate 32 having:
  • * a lower face 33 coppery on the facing surfaces with the partitions formed by conductive spacers 41;
  • * an upper face 34 coppery on the surfaces 42 facing the partitions and on the distribution circuits 35 H and V;
  • * metallized holes 44, ensuring the electrical connection between the two ground planes, which are arranged vertically in the partitions formed by the spacers 41; (the holes 36 are used for soldering the cores of the connectors 28 and 29 supplying the distributors H and V 35);
• . an upper substrate 37 having:
  • * a lower face 38 entirely coppered except at the locations of the annular slots 39 and serving as a ground plane for the triplate;
  • * an upper face 40 fully uncoped and acting as a radome.

These three substrates 25, 32 and 37 are spaced apart by the conductive spacers 41, for example made of aluminum, which delimit the cavities of the annular slots and the channels of the distributors.

The assembly can be carried out in a single operation by brazing in the oven after depositing a solder preform at the interfaces between the spacers and the copper-plated substrates, or by tinning the spacers. The materials and process used in such an embodiment guarantee very good performance in terms of cost and mass of the sub-networks.

At the thermal level, the spacers of small dimensions and without continuity in the plane of the network make it possible to use materials with a low coefficient of expansion, compatible for example with a structural carbon panel ensuring high rigidity at the antenna.

If an azimuth offset is specified, the subnets are much shorter, which allows to provide an antenna with slightly higher ohmic losses.

• une partie inférieure 45 en matériau conducteur : le plan de masse est réalisé par un substrat 43 de faible épaisseur sur lequel sont brasées les entretoises 46;
• une partie supérieure 47 en matériau diélectrique; C'est un substrat qui contient sur la face supérieure 48 les fentes annulaires 49 et sur la face inférieure 50 les répartiteurs de polarisation 51 et 52 et les excitateurs 53 de fentes. La continuité de la cavité des fentes annulaires dans le substrat est assurée par des trous métallisés 54 qui relient les entretoises au plan de masse des fentes. L'alimentation des répartiteurs se fait par sonde coaxiale à travers deux trous métallisés 55 et 56 pour la continuité de masse entre la sonde et le triplaque.

A variant of the previous antenna is shown in Figure 3. The antenna then consists of only two parts:

• a lower part 45 of conductive material: the ground plane is produced by a thin substrate 43 on which the spacers 46 are brazed;
• an upper part 47 of dielectric material; It is a substrate which contains on the upper face 48 the annular slots 49 and on the lower face 50 the polarization distributors 51 and 52 and the slot exciters 53. The continuity of the cavity of the annular slots in the substrate is ensured by metallized holes 54 which connect the spacers to the ground plane of the slots. The distributors are supplied by coaxial probe through two metallized holes 55 and 56 for mass continuity between the probe and the triplate.

• d'un répartiteur "triplaque suspendu" où l'énergie micro-onde se propage essentiellement dans le vide car les lignes sont supportées par une substrat très fin (environ 0,1 mm);
• d'un répartiteur "micro-ruban inversé" où la propagation se fait en grande partie dans le substrat diélectrique d'épaisseur plue forte (environ 1 mm).

This variant of the invention retains all the properties of the antenna shown in FIG. 1. Only the ohmic losses are slightly increased in favor of a simplification of construction. This perfectly acceptable increase in losses for short subnetworks (<10 λ) is due to the use:

• a "suspended triplate" distributor where the microwave energy propagates essentially in a vacuum because the lines are supported by a very thin substrate (about 0.1 mm);
• an "inverted micro-ribbon" distributor where the propagation takes place largely in the dielectric substrate of greater thickness (about 1 mm).

The invention thus makes it possible, in any of these embodiments, to form a space radar antenna with synthetic aperture comprising a network of 1.33 x 10 m² composed of sub-panels of 0.66 x 1m².

The elevation scanning requires a step between the sub-networks of about 0.7 λ at 5.3 GHz, no azimuth scanning being specified, to avoid an additional distributor stage supplying short radiating elements, Bipolar sub-networks of the length of the panel (about 1 m) are therefore best suited. The invention makes it possible to satisfy all these specifications with sub-networks of 24 annular slots.

The antenna therefore performs the functions of radiating element and distributor on a single level. The lower and upper spaces are made by aluminum half-shells assembled by screws. The two polarizations are excited by miniature coaxial probes placed near the center of the array.

= 22, 1dB; S étant la surface de l'antenne.The radiation diagrams for the horizontal and vertical polarizations are given in FIGS. 4 and 5. The normal polarization diagrams 60 and 61 in the azimuth planes are very close to those of an equi-amplitude equi-phase opening like the show the indicators at "*" at ± 3.22 ° which correspond to the first dummies of such an opening. The cross polarization diagrams 62 and 63 in the axis are less than 30 dB / max over the entire bandwidth. To judge the quality of the antenna, the gains of the two polarizations must be compared to the theoretical directivity D of the mesh of the network given by the formula:
D = 10 log $$\frac{\text{4πS}}{\text{λ2}}$$

= 22, 1dB; S being the surface of the antenna.

• Les erreurs d'excitation des 24 fentes vis-à-vis d'une loi équi-amplitude et équi-phase;
• Les pertes dues à la polarisation croisée;
• Les pertes ohmiques;
• Les pertes de désadaptation en entrée.

We then obtain respectively, in horizontal and vertical polarizations, losses of 1.5 dB and 2.5dB compared to this quantity. These losses include:

• The excitation errors of the 24 slits with respect to an equi-amplitude and equi-phase law;
• Losses due to cross polarization;
• Ohmic losses;
• Input mismatch losses.

These results make it possible to judge the interest of the invention for the production of linear sub-arrays of antennas.

• Si F2 est proche de F1, on peut ainsi faire fonctionner l'antenne sur une large bande.
• Si F2 est éloignée de F1, on obtient une antenne bi-bande; chacun des deux répartiteurs étant adapté à une bande particulière et excitant une des deux fentes annulaires.

The invention makes it possible to obtain a double resonator antenna, as shown in FIGS. 6 and 7: In order to increase the bandwidth, or to operate the antenna at two spaced frequencies, it is possible to place at the end of the distribution lines, as shown in Figures 2 and 3, in the same plane as these, discs 65 whose diameter is adjusted so that they resonate at a frequency F2 different from the resonance frequency F1 of the annular slot located above.

• If F2 is close to F1, it is thus possible to operate the antenna over a wide band.
• If F2 is far from F1, we obtain a dual-band antenna; each of the two distributors being adapted to a particular strip and exciting one of the two annular slots.

Such an embodiment is compatible with the two variants of the invention shown above: In the suspended triplate variant the additional annular slots are etched either on the upper face or on the lower face of the upper substrate. In FIGS. 6 and 7, the variants taken into account are respectively those illustrated in FIGS. 2 and 3; the upper substrate 37 ′ shown in FIG. 6 being provided, on its upper face 67, with slots 69; metallized holes 66 being provided to allow the slit-mass-triplate mass connection.

It is understood that the present invention has only been described and shown as a preferred example and that its constituent elements can be replaced by equivalent elements without, however, departing from the scope of the invention.

Claims (10)

1) Linear array antenna composed of a radiating element formed by an alignment of annular slots (21) excited by at least one distributor element (14), this distributor element being a buried distributor included in the radiating element, and therefore invisible of the radiating face of said antenna, so as to avoid any parasitic radiation coming from said distributor, said antenna characterized in that it only comprises thin substrates spaced apart by conductive spacers without mechanical continuity in the plane of the network so to avoid differential expansions with the substrates. 2) network antenna according to claim 1, characterized in that said antenna is designed to radiate (or receive) simultaneously in two linear polarizations. 3) network antenna according to claim 1, characterized in that said antenna is designed to radiate (or receive) simultaneously in two circular polarizations (right and left). 4) network antenna according to claim 1, characterized in that it comprises an integrated radome and a distributor using only three levels of conductors. 5) network antenna according to any one of the preceding claims, characterized in that it comprises shields, between polarizations of a sub-network and between adjacent sub-networks, radiating elements and distributors so as to avoid any coupling between these. 6) network antenna according to claim 1, characterized in that it comprises: - A lower half-shell (10) of conductive material in which is formed the cavity (11) of the radiating elements and the conduit (12) of the distributors of the two polarizations; - A first very thin substrate (13) having a coefficient of expansion close to that of the material of the half-shells, on which the horizontal and vertical distributors (14) are etched on the upper face. - an upper half-shell (15) in which the conduit (16) of the distributors of the two polarizations and piercing (17) the upper part (17) of the cavity; - A second substrate (18) identical to the first substrate, an upper face (19) of which is completely uncoppered and the other (20) 5 contains the etched annular slots (21); the two half-shells (10 and 15) being connected to the same potential. This is ensured by the first substrate thanks to the metallization of the facing surfaces with the two half-shells and to the presence of a large number of metallized holes located on the periphery of the cavities, connecting these two surfaces. 7) network antenna according to claim 1, characterized in that it comprises: - a lower substrate (25) having: . a bottom face (26) locally coppery (zone 27) around connectors (28 and 29) of RF power supply and glued to the structural panel; . an upper face (30) entirely coppery (except RF power saving) and used as a ground plane; . metallized holes (44) connecting the two surfaces; - an intermediate substrate (32) having: . a lower face (33) coppery on the surfaces opposite with the partitions formed by conductive spacers (41); . an upper face (34) coppery on the surfaces (35) facing these partitions and the distribution circuits H and V; . metallized holes (44) ensuring the electrical connection between the two ground planes; - an upper substrate (37) having: . A bottom face (38) entirely coppery except at the locations of the annular slots (39) and serving as a ground plane for the triplate; . An upper face (40) fully uncoped and acting as a radome;
These three substrates (25, 32 and 37) being spaced apart by conductive spacers (41) which delimit the cavities of the annular slots and the channels of the distributors. 8) network antenna according to claim 1, characterized in that it includes: - A lower part (45) of conductive material; The ground plane being produced by a thin substrate on which are brazed spacers (46); - An upper part (47) of dielectric material which is a substrate which contains on the upper face (48) the annular slots (49) and on the lower face (59) the polarization distributors (51 and 52) and the exciters ( 53) of slots. The continuity of the cavity of the annular slots in the substrate being ensured by metallized holes (54) which connect the spacers to the ground plane of the slots; The distributors are supplied by coaxial probe through metallized holes (55 and 56) for mass continuity between the probe and the triplate. 9) Network antenna according to any one of the preceding claims, characterized in that said antenna is a broadband antenna using resonant discs (65) etched on the resonant distributor stage at a frequency close to that of the annular slots. 10) network antenna according to any one of claims 1 to 5, characterized in that said antenna is a dual-band antenna using resonant discs (65) etched on the distributor stage, resonating at a frequency distant from that of annular slots; one distributor exciting the resonant disc and the other the annular slot.

Images