Quasi-optical beam shaper comprising two reflectors

Abstract

Quasi-optical beamformer (1) comprising: a plurality of feed sources (PF), each of the feed sources (PF) being adapted to emit a respective radio frequency beam denoted by a radio frequency transmission line (LT) fed at one end by the feed sources and comprising at a second end a plurality of network ports (PR) adapted to collect radio frequency radiation, the transmission line (LT) comprising a radio frequency waveguide (GO) comprising at least two stacked metal plates (PM) to guide the radio frequency beams to the network ports (PR), the transmission line (LT) extending along a central principal axis denoted by axis, a first reflector (M1) having a first center (C1) centered on the axis and a first effective radius of curvature and being adapted to reflect the guided radio frequency beams,a second reflector (M2) having a second effective radius of curvature and being adapted to reflect the radio frequency beams reflected by the first reflector by directing them towards the network ports so as to form output radio frequency beams, the first and second effective radii of curvature and an arrangement of the power supplies (PF) relative to the first being adapted so that: each of the output radio frequency beams is a plane wave, and a transverse amplitude distribution on the network ports (PR) of each output radio frequency beam is substantially identical.

Description

Title of the invention: Quasi-optical beamformer comprising two reflectors Technical field

[0001] The invention relates generally to the field of telecommunications, and in particular to quasi-optical beamformers (QOBs) for active multi-beam antennas. Prior art

[0002] Quasi-optical beamformers can be carried on satellites or in ground stations. Antennas using such formers can operate in transmission or reception, reciprocally.

[0003] A quasi-optical beamformer is a focusing (in reception) and collimating (in transmission) device. [Fig.l] represents a state-of-the-art quasi-optical beamformer that can be applied, for example, to pillbox beamformers, continuous delay lens beamformers or Rotman beamformers. A quasi-optical beamformer conventionally integrates a parallel plate guide 16, connecting beam ports 17 and network ports 18. The parallel plate waveguide 16 makes it possible to guide the waves in TEM mode (acronym for “Transverse Electric Magnetic”), in which the electric field E and the magnetic field H evolve in directions perpendicular to the propagation direction X.

[0004] The wavefronts are curved in the XY plane. In order to compensate for the curvature of the wavefront, a quasi-optical device 23 is introduced between the beam ports and the array ports. This quasi-optical device may be, for example, a lens such as used for continuous delay lenses or a reflector such as used for pillbox beamformers. Each array port 18 is connected to an amplifier 19 followed by a radiating element 20 via a delay line 21 and an amplifier port 22. It transforms the cylindrical waves emanating from the beam ports into plane waves radiated by the radiating panel of the active multibeam antenna.

[0005] FFQOs for active antennas are devices that are applied to low-level signals. An amplification is integrated between the FFQO and the radiating elements. Moderate losses are then acceptable.

[0006] Beamformers aim to generate the phase and amplitude laws required to form the beams in the desired directions with low sidelobe levels. Figure 2 shows examples of these laws for three beams numbered 1, z and N. It highlights the variations of the laws of desired phase to direct the beams in the desired directions. It also highlights an apodized amplitude law in order to obtain low levels of side lobes on the radiation patterns. It should be noted that in these three examples, the amplitude law is stable and centered at the center of the lens and all the amplifiers are powered in order to benefit from good radiation efficiency. These characteristics must be respected when designing an active antenna to avoid being forced to oversize the antenna.

[0007] Indeed, in the case of a dependence of the amplitude law on the pointing direction as described in [Fig. 3], the radiating panel of the antenna must be oversized in order to be able to follow the displacement of the amplitude law with the pointing direction. As a result, for each pointing direction, a part of the radiating amplifier network is not illuminated, thus leading to a degradation of the antenna efficiency.

[0008] Known quasi-optical beamformers operating over a wide angular sector unfortunately do not have a stable amplitude law. Use over a wide angular sector is necessary in many applications (especially for satellite applications). Their use in beamformers for active antennas thus leads to low antenna radiation efficiencies and additional costs due to the implementation of additional front-end circuits. These are major drawbacks for their use in active antennas.

[0009] Currently none of the known FFQOs can cover an angular sector greater than +-50° with an amplitude law at the output of the optical device stable over the angular sector covered.

[0010] The invention aims to overcome certain problems of the prior art. To this end, an object of the invention is a quasi-optical beamformer comprising a plurality of power sources, each adapted to emit a radiofrequency beam guided in a radiofrequency waveguide and comprising two reflectors coupled to the waveguide. The reflectors are adapted to form output radiofrequency beams in the form of a plane wave and to obtain a substantially identical transverse amplitude distribution at the output of the quasi-optical beamformer. The use of two particularly adapted reflectors allows great compactness of the beamformer while ensuring optimized operation over a wide angular range. In addition, the beamformer makes it easy to control the dimension of the output zone where the network ports adapted to collect radiofrequency radiation are located.These characteristics make the beamformer of the invention particularly suitable for being combined with a radiating panel to form a planar active antenna. Summary of the invention

[0011] For this purpose, an object of the invention is a quasi-optical beamformer comprising: - a plurality of N > 1 power sources, each of the i E [1; N] power sources being adapted to emit a respective radiofrequency beam denoted Ri - a parallel plate radiofrequency waveguide extending along a central main axis denoted x-axis and being supplied at a first end by the power sources, a plurality of network ports being arranged at a second end of the radiofrequency waveguide to collect radiofrequency radiation, the radiofrequency waveguide comprising at least two stacked metal plates to guide the radiofrequency beams R^ i E [1; N] towards the network ports, - a quasi-optical system having a focal plane and comprising a first reflector having a first center centered on the x axis and a first effective radius of curvature and being adapted to reflect the guided radiofrequency beams i E [1; N], and comprising a second reflector having a second effective radius of curvature and being adapted to reflect the radiofrequency beams reflected by the first reflector by directing it towards the network ports so as to form output radiofrequency beams the first and second effective radii of curvature and an arrangement of the power sources being adapted so that each of the output radiofrequency beams is a plane wave, and so that a transverse amplitude distribution A, on the network ports of each output radiofrequency beam, is substantially identical.

[0012] According to one embodiment of the invention, a distance between a focal center of each feed source and the first center is less than a useful dimension of the first reflector divided by two.

[0013] According to one embodiment of the invention, a direction of propagation of the radiofrequency beams Ri? i E [1; N] and the first and second effective radii of curvature are configured so that at least one of the output radiofrequency beams propagates with an angle greater than 45°, preferably 55° relative to the x axis.

[0014] According to one embodiment of the invention, the power sources are arranged so that each of the radiofrequency beams Rr i E [1; N] partially illuminates the first and second reflectors over a zone different from that of the other radiofrequency beams.

[0015] According to one embodiment of the invention, a transverse dimension of a zone formed by all of the network ports, called the sampling zone, is less than a maximum dimension of the first and second reflectors.

[0016] According to one embodiment of the invention, the power sources are superimposed on at least two stages, each of the at least two stages being separated by a conductive plane common to two adjacent stages, the quasi-optical beamformer comprising a resistive film arranged in the continuity of the conductive plane.

[0017] According to one embodiment of the invention, a focal center of each power source is arranged in the focal plane of the quasi-optical system.

[0018] According to a preferred embodiment of the invention, the beamformer of the invention is produced from a multilayer printed circuit. Preferably, the multilayer printed circuit comprises at least three stacked layers, the three stacked layers being metallized or partially metallized so as to form the radiofrequency waveguide, the waveguide comprising a first, a second and a third stacked portions, the first reflector being arranged so as to reflect the radiofrequency beams R._ i E [1; N] guided in the first portion towards the second portion and the second reflector being arranged so as to reflect the radiofrequency beams R-, i E [1; N] guided in the second portion towards the third portion.

[0019] Preferably, in the preferred embodiment, the first and second reflectors each comprise metal lamellae deposited on a metallized face of the multilayer circuit, a height and an arrangement of the lamellae being adapted so that the first and second reflectors respectively have the first and second effective radius of curvature. Alternatively, according to another embodiment, the reflectors are produced with contactless transitions between several layers of the radiofrequency waveguide.

[0020] According to one embodiment of the invention, the beamformer of the invention comprises a number of reflectors greater than two, the reflectors being arranged so as to successively reflect the radiofrequency beams Ri? i E [1; N] and collimate them.

[0021] Another object of the invention is a planar active antenna comprising: - at least one beamformer according to the invention - delay lines, each connected to a respective one of the network ports at a proximal end and to a power amplifier device at a distal end, the power amplifier device being adapted to amplify a radio frequency signal collected by the respective beam port - radiating elements each connected to an amplifier device respective power and adapted to emit a radio frequency wave from the radio frequency signal amplified by the respective power amplifier device.

[0022] According to one embodiment of the antenna of the invention, the delay lines each have an electrical length different from the others. Brief description of the drawings

[0023] Other characteristics, details and advantages of the invention will emerge from reading the description given with reference to the appended drawings given by way of example and which represent, respectively:

[0024] [Fig-1] a schematic view of a quasi-optical beamformer of the at prior

[0025] [Fig.2], phase and amplitude laws required to form beams in several directions with low levels of side lobes,

[0026] [Fig.3], an example of dependence of the amplitude law on the pointing direction

[0027] [Fig.4], a schematic view of a quasi-optical beamformer according to the invention,

[0028] [Fig.5], a schematic view of a quasi-optical beamformer according to an embodiment of the invention

[0029] [Fig.6A] and [Fig.6B] the trajectory of the output radiofrequency beam propagating with an angle of 0° and 54° respectively in a beamformer according to an embodiment of the invention

[0030] [Fig.6C], a graph representing the distribution, over the sampling area, of the normalized field amplitude of the 0° beams (curve C1) and -54° beams (curve C2) in the beamformer according to the embodiment of Figures 6A and 6B,

[0031] [Fig.7], the radiation pattern of the beamformer of the embodiment of Figures 6A to 6C for output beams having a different angle 0 and for three different center frequencies generated by the PF sources: 17 GHz, 18 GHz and 20 GHz.

[0032] [Fig.8], the radiation pattern of the beamformer according to an embodiment of the invention for RS output beams, having a different angle 0 and for three different central frequencies generated by the PF sources: 17 GHz, 18 GHz and 20 GHz.

[0033] [Fig.9], a schematic view of an arrangement of the power sources in one embodiment of the invention,

[0034] [Fig. 10], a schematic view of an embodiment of the invention in which the beamformer is made from a multilayer printed circuit

[0035] [Fig. 11], a schematic view of an active antenna according to the invention comprising at least one beamformer according to the invention.

[0036] In the figures, unless otherwise indicated, the elements are not to scale. Description of the embodiments

[0037] Figure 4 schematically illustrates a quasi-optical beamformer 1 according to the invention which mainly comprises a plurality of N > 1 power sources PF, a parallel plate radiofrequency waveguide GO and a plurality of network ports PR.

[0038] In the quasi-optical beamformer 1 according to the invention, each of the ie [1; N] power sources PF is adapted to emit a respective radiofrequency beam denoted R.. According to one embodiment of the invention, each radiofrequency beam Rj is generated by a plurality of sources.

[0039] According to one embodiment of the invention, the sources are horns, for example H-plane horns, which are particularly suitable for carrying out beam reconfiguration because each power source defines a spot access. In addition, the horns can be easily designed and manufactured in PCB (Printed Circuit Board) technology.

[0040] According to another embodiment, the sources are other forms of sources well known to those skilled in the art, for example monopole networks, transitions between micro-strip lines and parallel plate guides, transitions between tri-plate lines and parallel plate guides, etc.

[0041] The parallel plate radiofrequency waveguide GO extends along a central main axis denoted x axis and is supplied at a first end by the power sources PF. As will be explained in the description of [Fig. 10], preferably, all or part of the parallel plate radiofrequency waveguide GO is designed and manufactured in PCB technology to improve the compactness of the beamformer 1.

[0042] The network ports PR are arranged at a second end of the waveguide GO to collect radiofrequency radiation and possibly allow the coupling of the beamformer 1 to a radiating panel in order to form an active antenna (see [Fig. 11]).

[0043] The waveguide GO comprises at least two stacked metal plates PM to guide the radiofrequency beams R.. ie [1; N] towards the network ports PR. Subsequently, the area formed by the spatial distribution of the network ports PR is called “sampling zone ZP”. As a non-limiting example, in FIG. 4, this sampling zone is a straight line segment in the direction perpendicular to the x axis and the parallel plates PM extend along the XY plane.

[0044] The PF power sources are adapted to generate a TEM wave or a TE wave or both in the GO waveguide.

[0045] According to the embodiment illustrated in [Fig. 4], the metal plates PM extend in the XZ plane and are therefore superimposed in the Y direction. As will be illustrated in [Fig. 10] for example, the waveguide GO may comprise several pairs of plates, the pairs being stacked over all or part of their extent, so as to form stacked portions of waveguide GO.

[0046] Furthermore, the beamformer comprises a quasi-optical system SQO adapted to collimate the radiofrequency beams R,, i G [1; N]. For this, the quasi-optical system SQO comprises a first reflector M1 and a second reflector M2. The first reflector M1 has a first center Cl centered on the x axis and a first effective radius of curvature and is adapted to reflect the radiofrequency beams Rp î G [ h N] guided in the parallel plate waveguide GO. The second reflector M2 has a second effective radius of curvature and is adapted to reflect the radiofrequency beams reflected RF], i G [1; N] by the first reflector by directing it towards the network ports so as to form output radiofrequency beams RS], Îg[1; N]-

[0047] In the beamformer 1, the reflectors M1 and M2 (via in particular the first and second effective radii of curvature) and the arrangement of the power sources PF are adapted so that: - each of the output radiofrequency beams RS], i G [1; N] is a plane wave, and - so that the transverse amplitude distribution Aâ of each output radiofrequency beam RS], i G [1; N] on the network ports PR is substantially identical.

[0048] By “transverse amplitude distribution A, substantially identical”, we mean here that the individual excitation of each of the sources PF leads to amplitudes of the signals received by each of the network ports PR of the sampling zone ZP varying over a dynamic range less than or equal to 3dB between each of these sources.

[0049] The profile of the reflectors M1 and M2 is optimized by ray tracing and physical optics software that was developed by the inventors. In this optimization, each reflector is defined by a Spline, itself defined by "nodes". By "Spline" is meant a function defined piecewise by polynomials. During the iterative optimization process, it is these "nodes" that are moved and which consequently modify the contour of the reflectors. The displacement of these "nodes" is dictated by the local phase difference between the desired phase law on the sampling area and that current of the system during the iterative process.

[0050] Other parameters are taken into account for the global optimization of the quasi-optical system: - the position of the PF sources - the orientation of PF sources - their orientation around their phase centers - the width of the sources. - the width of the quasi-optical system - the distance between the two reflectors.

[0051] The use of two particularly suitable reflectors M1, M2 allows a high compactness of the beamformer while ensuring optimized operation over a wide angular range. The compactness of the device of the invention is made possible by placing the sources PF close to the center Cl via the use of a quasi-optical system SQO comprising at least two reflectors. Indeed, the use of two (or more) reflectors makes it possible to reduce the distance between the focal plane FPL and the first center Cl compared to the use of a single reflector with the same center CL

[0052] Furthermore, to the first order, the reflectors of the SQO optical system are adapted so that an angle between the x axis and the propagation direction of the output beams RSj is equal to an angle between the focal center of the PF sources and the first center CL. Also, it is possible to control the angular coverage of the beamformer 1 via the arrangement of the PF sources.

[0053] According to a preferred embodiment, the direction of propagation of the radiofrequency beams at the output of each source PF and the first and second effective radii of curvature are configured so that the beamformer 1 has an angular coverage greater than 45°, preferably 55°. By “angular coverage of 8°”, it is meant here that the beamformer 1 is adapted to emit output beams RSj propagating with an angle ranging from + 8° to - 9° relative to the x axis. The use of an angular coverage greater than 50° for example is desirable in many applications, in particular for satellites in LEO orbit.

[0054] Preferably, as illustrated in [Fig.4], the focal center of each power source PF is arranged in the focal plane FPL of the quasi-optical system SQO in order to guarantee optimal collimation by the SQO system.

[0055] Preferably, the reflectors and the arrangement of the power sources PF are adapted so that the transverse amplitude distribution of an output beam RSj on the sampling zone ZP follows an apodized law, which allows a widening of the main lobe of this beam and a reduction of the secondary lobes associated with this beam.

[0056] Figure 5 schematically illustrates a preferred embodiment of the invention. in which, in order to facilitate the optimization of the reflector profile, the PF feed sources are arranged so that each of the R-, i E [1; N] beams partially illuminates the reflectors over a different area from that of the other beams. This makes it possible to increase the degrees of freedom when designing the reflectors for the optimization of the parameters of the different angular directions. It is then possible to obtain better control of the sidelobe level over a wide angular sector while ensuring that the condition on the amplitude distribution over the sampling area is respected. In Figure 5, only the central propagation axis of the R( beams has been represented. It is understood that in reality, there may be a partial overlap between some of the areas of the reflectors illuminated by the Rb ie [1; N] beams.

[0057] The reference LU in Figure 5 corresponds to the maximum dimension of the region formed by the sum of the zones of the reflector M1 illuminated by the beams R,, i E [1, ; N]. This maximum dimension is subsequently called the useful dimension LU. According to one embodiment, in order to ensure satisfactory compactness while maintaining optimal operation of the quasi-optical system SQO, in the beamformer 1 the distance between a focal center of each z E [1, ; jV] power source PF and the first center C| is less than a useful dimension LU of the first reflector divided by two.

[0058] According to the embodiment illustrated in [Fig.4] and [Fig.5], the quasi-optical system SQO comprises only two reflectors M1, M2. This makes it possible to obtain satisfactory angular coverage while maintaining an interesting compactness.

[0059] Alternatively, according to another embodiment, the quasi-optical system SQO comprises a number of reflectors greater than two, the reflectors being arranged so as to successively reflect the radiofrequency beams Ri? i E [1; N] and collimate them. Compared to only two reflectors, the use of a number of reflectors greater than two makes it possible to increase the angular coverage of the beamformer 1. In addition, this makes it possible to increase the degrees of freedom for optimizing the effective radius of curvature of each reflector, thus ensuring equal transverse amplitude distributions on the network ports for all z E [1; N]. It is understood that this complicates the design and manufacture of the beamformer 1.

[0060] Preferably, by the arrangement of the sources PF relative to the reflectors and the optimization of the reflectors, the phase aberrations in the output beams RS; are less than λ / 10, preferably less than λ / 8. Here, 2 is the central wavelength for which the beamformer 1 is optimized. A phase aberration less than λ / 10 is desirable to reduce the de- pointing and level of the secondary lobes.

[0061] Preferably, the first and second reflectors are adapted so that the transverse amplitude distribution A, of each output radiofrequency beam N] has a maximum amplitude variation over the sampling zone equal to 6 dB at ± 1 dB. A variation of 6 dB makes it possible to reject the secondary lobes at a satisfactory value (17 dB or less of the maximum amplitude) without oversizing the active antenna comprising the beamformer of the invention. Indeed, assuming that the transverse amplitude distribution A, is adapted, a variation greater than 6 dB would make it possible to further reduce the secondary lobes but would have the consequence of having to enlarge the size of the antenna at a given antenna gain.

[0062] According to a preferred embodiment of the invention, the sampling zone ZP formed by the network ports PR has a transverse dimension Dp less than a maximum dimension of the reflectors. In the embodiment of FIG. 5, the network ports PR are aligned along the direction 3' with an alignment length Dp and the reflectors M1, M2 have a length (maximum dimension) of dimension L > Dp. This possibility of being able to undersize the sampling zone is particularly interesting for active antenna applications in order to avoid oversizing the radiating elements and therefore the antenna gain. In the invention, it is possible to undersize the sampling zone without in any way degrading the operation of the beamformer over the entire angular coverage.

[0063] As a non-limiting example, according to one embodiment, the reflectors M1, M2 have a length of dimension L = 21 2 and the beamformer 1 comprises 26 network ports PR aligned side by side in the direction, each network port PR having a width of 0.6 2. Thus, the alignment formed by the network ports has an alignment length Dp = 16 2. Network ports having a width of 0.6 2 or less than 0.6 2 are preferred in order to reduce parasitic reflections at high incidence.

[0064] Figures 6A and 6B schematically illustrate the trajectory of the output radiofrequency beam propagating with an angle of 0° (“0° beam” hereinafter) and 54° (“54° beam” hereinafter) respectively in a beamformer 1 according to an embodiment of the invention comprising only a first and a second reflector M1, M2. By way of non-limiting example, the network ports PR are aligned in the sampling zone ZP, side by side, in the direction 3'. In Figures 6A and 6B, it can be seen that the 0° beam does not illuminate the same area of ​​the reflectors M1, M2, nor the same region as the sampling zone ZP as those illuminated by the 54° beam. As explained previously, this makes it possible to optimize separately and more easily the parameters of the 0° and 54° beams (for example reduction of the lobes secondary or amplitude distribution over the sampling area).

[0065] Figure 6C is a graph representing the distribution, over the sampling zone, of the normalized field amplitude of the 0° beams (curve C1) and -54° beams (curve C2) in the beamformer 1 according to the embodiment of Figures 6A and 6B. Given the geometry of Figures 6A and 6B, this amplitude distribution corresponds to a profile along the Y direction over the sampling zone ZP. It can be seen in [Fig.6C] that the amplitude distribution of the 0° and -54° beams is substantially identical over the sampling zone ZP, the 0° and -54° beams perfectly illuminating the network ports, with an apodized law, which allows a broadening of the main lobe of each beam and a reduction of the secondary lobes associated with these beams. In addition, it can be seen that the two output beams each have a maximum amplitude variation over the sampling zone equal to 6 dB.

[0066] Figure 7 illustrates the radiation pattern of the beamformer 1 of the embodiment of Figures 6A to 6C for output beams RS; having an angle S with respect to the X axis of ±55°, ±35°, ±22.5°, ±10° and 0° (and each associated with a respective PF source) and for three different center frequencies generated by the PF sources: 17 GHz, 18GHz and 20 GHz.

[0067] First of all, we observe that the levels of the secondary lobes are low (approximately -20 dB relative to the maximum gain) and this for the three central frequencies. However, we note that the beamformer 1 produces RS output beams, which intersect at a low gain level (approximately -15 dB relative to the maximum gain).

[0068] Figure 8 illustrates the radiation pattern of the beamformer 1 according to an embodiment with two reflectors different from that of Figures 6A to 7. Due to an optimization of the profile of the reflectors M1, M2 different from that of the reflectors of Figures 6A to 7, the beamformer of Figure 8 has an angular coverage of ± 70 °. The radiation pattern is provided for the same three central frequencies as those of Figure 7 and for beams having an angle 9 relative to the X axis of ± 70 °, ± 50 °, ± 32.5 °, ± 17.5 ° and 0 °. As in Figure 7, in Figure 8 we observe the levels of the secondary lobes are low (approximately -16 dB relative to the maximum gain) and this for the three central frequencies. Just as in Figure 7, we notice that beamformer 1 produces RS output beams, which intersect at a low gain level (about -15 dB relative to maximum gain).

[0069] As illustrated in Figures 7 and 8, the beamformer 1 produces RS output beams, which intersect at a low gain level. This level of beam overlapping comes from a compromise on the size of the sources: on the one hand, they must be large enough to properly illuminate the system quasi-optical SQO, and thus avoid overflow losses, and on the other hand, they must be close enough so that the output beams RS, overlap.

[0070] To overcome this problem, according to an embodiment illustrated in FIG. 9, the power sources PF are superimposed on at least two stages E1 and E2, each of the at least two stages being separated by a conductive plane PC common to two adjacent stages, the quasi-optical beamformer comprising a resistive film FR arranged in the continuity of the conductive plane. As illustrated in FIG. 9, the PF sources of the upper stage E1 and the PF sources of the stage E2 can be offset relative to each other along the focal plane FPL of the quasi-optical system SQO. As a non-limiting example, in the illustration of FIG. 9, the PF sources of the stage E1 and the PF sources of the stage E2 are offset by a distance (with d2 the width of a PF source), and the focal center of a source of the upper stage E1 coincides with the junction between two sources of the lower stage E2, and vice versa.

[0071] The presence of the resistive film in the continuity of the conductive plane makes it possible to minimize the losses linked to the crossing of the beams while increasing the density of PF sources. This therefore makes it possible to increase the angular resolution of the beams R, and therefore makes it possible to avoid obtaining a beamformer generating angularly separated beams and crossing at very low levels. The resistive film allows, among other things, the use of larger PF sources, which also makes it possible to limit the amplitude of the field on the edges of the beamformer 1 and therefore to limit the parasitic reflections on these edges.

[0072] The resistive film FR makes it possible to isolate the sources of the lower stage El from the sources of the stage E2, and to obtain, at the output port common to the two stages and located in the waveguide GO, the lossless summation of the signals coming from the input beam ports when they are in phase and of the same amplitude.

[0073] Figure 10 illustrates an embodiment of the invention in which the beamformer is produced from a multilayer printed circuit. This embodiment has the advantage of having greatly improved compactness. The multilayer printed circuit comprises at least three stacked layers C11, C12, C13 along the Z direction. The three stacked layers are metallized or partially metallized so as to form the parallel plate radiofrequency waveguide GO. More specifically, the waveguide comprises a first, a second and a third portions P1, P2 and P3 stacked along the Z direction and arranged so that the main direction of the waveguide GO is along the x axis.

[0074] In the embodiment of Figure 10, the first reflector is arranged to reflect the radiofrequency beams Rj guided in the first portion PI towards the second portion P2. In addition, the second reflector is arranged to reflect the radiofrequency beams Rj guided in the second portion towards the third portion.

[0075] The multilayer printed circuit of [Fig. 10] further comprises TPF transitions adapted to connect the PF sources to the CPF source connectors and TPR transitions adapted to connect the network ports to the CPR network port connectors.

[0076] Optionally, as illustrated in Figure 10, the PF sources are superimposed on at least two stages E1 and E2. In addition, the multilayer printed circuit comprises two parallel plate waveguides GO1, GO2 superimposed along the Z direction, each of the superimposed parallel plate waveguides GO1, GO2 being arranged facing the PF sources of a respective stage in order to guide the radiation emitted by these sources. The GO waveguide is then arranged in the continuity of the superimposed parallel plate waveguides GO1, GO2. Finally, the resistive film FR is arranged at the junction between each superimposed parallel plate waveguide and the GO waveguide in order to minimize the losses linked to the crossing of the beams while increasing the density of PF sources.

[0077] In the embodiment of [Fig. 10], the reflectors M1, M2 each comprise metal lamellae L deposited on a metallized face of the multilayer circuit (referenced FMI and FM2 respectively in [Fig.10]). In this embodiment, the height of each lamella and its arrangement on its respective metallized face are adapted to present respectively the first and second effective radius of curvature. The position and height of each lamella is preferably determined by software.

[0078] Alternatively, according to a variant of the embodiment of [Fig. 10], the reflectors M1, M2 are produced with contactless transitions between several layers of the radiofrequency waveguide. The creation of a reflector with contactless transitions in a multilayer circuit is described in document FR 3 057 999.

[0079] [Fig. 11] schematically illustrates an active antenna 2 according to the invention. The antenna 2 comprises at least one beamformer according to the invention and delay lines LR. Each of the delay lines LR is connected to a respective one of the network ports at a proximal end and to a power amplifier device AP at a distal end via an amplifier port PA. The power amplifier device AP is adapted to amplify a radiofrequency signal collected by the respective beam port. Finally, radiating elements ER are each connected to a respective power amplifier device and are adapted to emit a radiofrequency wave from the radiofrequency signal amplified by the respective power amplifier device.

[0080] As explained previously, the transverse dimension of the sampling zone directly impacts the number of radiating elements that can be supplied by the beamformer 1. This number of radiating elements is typically equal to the width in 2 of the sampling zone divided by approximately 0.6 (the network ports being separated by approximately 0.6 / to minimize parasitic reflections at the edges of beamformer 1).

[0081] It is therefore important to control the transverse dimension of this sampling zone so that it is in line with the number of radiating elements of the antenna. A sampling zone that is too large in relation to the desired number of radiating elements would lead to underutilizing the energy at the output of the beamformer and thus generating additional losses.

[0082] According to one embodiment of the active antenna 2 of the invention, the delay lines each have an electrical length different from the others. Thus, the delay lines introduce an additional degree of freedom when optimizing the reflectors in order to obtain a predetermined phase at the output of the active antenna.

[0083] Although the operation of the devices of the invention has been described in transmission, the operation is also transposed to reception. In this case, a parallel radiofrequency incident wave is collected then transmitted to the network ports and is focused by the quasi-optical system SQO in a direction determined by the propagation angle of the incident wave respectively to the x axis. Finally, each of the power sources PF is also adapted to collect the focused radiation of the quasi-optical system SQO in a respective direction.

Claims

Claims

1. A quasi-optical beamformer (1) comprising: - a plurality of N > 1 power sources (PF), each of the i G [1; N] power sources (PF) being adapted to emit a respective radiofrequency beam denoted Ri - a parallel plate radiofrequency waveguide (GO) extending along a central main axis denoted x-axis and being supplied at a first end by the power sources, a plurality of network ports (PR) being arranged at a second end of the radiofrequency waveguide (GO) to collect radiofrequency radiation, the radiofrequency waveguide (GO) comprising at least two stacked metal plates (PM) to guide the radiofrequency beams i G [1; N] towards the network ports (PR), - a quasi-optical system (QOS) having a focal plane (FPL) and comprising a first reflector (Ml) having a first center (Cl) centered on the x axis and a first effective radius of curvature and being adapted to reflect the guided radiofrequency beams Rr ie [1; N], and comprising a second reflector (M2) having a second effective radius of curvature and being adapted to reflect the radiofrequency beams reflected (RF?) by the first reflector by directing it towards the network ports so as to form output radiofrequency beams (RS;) the first and second effective radii of curvature and an arrangement of the power sources (PF) being adapted so that each of the output radiofrequency beams (RS^ is a plane wave, and so that the individual excitation of each of the sources leads to a transverse amplitude distribution Aj on the network ports (PR) of each output radiofrequency beam (RS / ) varying over a dynamic range less than or equal to 3dB between each of these sources.

2. A quasi-optical beamformer according to claim 1, wherein a distance between a focal center of each feed source (PF) and the first center is less than one dimension useful light (LU) of the first reflector divided by two.

3. Quasi-optical beamformer according to any one of the preceding claims, wherein a propagation direction of the radiofrequency beams R. i G [1; N] and the first and second effective radii of curvature are configured so that at least one of the output radiofrequency beams (RS;) propagates with an angle greater than 45°, preferably 55° relative to the x axis.

4. A quasi-optical beamformer according to any preceding claim, wherein the power sources (PF) are arranged such that each of the radio frequency beams Rp i G [1; N] partially illuminates the first and second reflectors over a different area than the other radio frequency beams.

5. Quasi-optical beamformer according to any one of the preceding claims, a transverse dimension (Dp) of an area formed by all the network ports, called the sampling area, is less than a maximum dimension of the first and second reflectors.

6. A quasi-optical beamformer according to any preceding claim, wherein the power sources are superimposed on at least two stages, each of the at least two stages being separated by a conductive plane common to two adjacent stages, the quasi-optical beamformer comprising a resistive film arranged in continuity with the conductive plane.

7. A quasi-optical beamformer according to any preceding claim, wherein a focal center of each feed source (PF) is arranged in the focal plane of the quasi-optical system (QOS).

8. A quasi-optical beamformer according to any preceding claim made from a multi-layer printed circuit.

9. Quasi-optical beamformer according to the preceding claim, wherein the multilayer printed circuit comprises at least three stacked layers (Cil, CI2, CI3), the three stacked layers being metallized or partially metallized so as to form the radiofrequency waveguide, the waveguide comprising a first, a second and a third stacked portion (pl, P2, P3), the first reflector being arranged so as to reflect the radiofrequency beams Rj, i G [1; N] guided in the first portion towards the second portion and the second reflector being arranged so as to reflect the radiofrequency beams R{, i G [1; N] guided in the second portion towards the third portion.

10. Quasi-optical beamformer according to claim 8 or 9, wherein the first and second reflectors each comprise metal lamellae (L) deposited on a metallized face (FMI, FM2) of the multilayer circuit, a height and an arrangement of the lamellae being adapted so that the first and second reflectors respectively have the first and second effective radius of curvature.

11. A quasi-optical beamformer according to claim 8 or 9, wherein the reflectors are made with contactless transitions between several layers of the radio frequency waveguide.

12. A quasi-optical beamformer according to any preceding claim, comprising a number of reflectors greater than two, the reflectors being arranged to successively reflect the radiofrequency beams i G [1; N] and collimate them.

13. Planar active antenna (2) comprising: - at least one beamformer (1) according to one of the preceding claims - delay lines (LR), each connected to a respective one of the network ports at a proximal end and to a power amplifier device (AP) at a distal end, the power amplifier device being adapted to amplify a radiofrequency signal collected by the respective beam port - radiating elements (ER) each connected to a respective power amplifier device and adapted to emit a radiofrequency wave from the radiofrequency signal amplified by the respective power amplifier device.

14. Planar active antenna according to the preceding claim, in which the delay lines each have an electrical length different from the others.

Images