EP0545742A1 - Method and device for the rejection of harmonics emitted by an active antenna having electronic scanning - Google Patents

Abstract

The method relates to active antennas having electronic scanning and consisting of an array of radiating elements (Eo,...Em) excited on emission by individual power amplifiers (Ao,...Am) arranged in active modules (M'o,...M'm) placed as close as possible to the radiating elements (Eo,...Em). It consists in reducing the level of the harmonics emitted by dividing the power amplifiers (Ao,...Am) into at least two groups and by impressing upon the power amplifiers (Ao,...Am) a non-zero common value of phase shift phi which is added to the individual values of phase shift which are required by the electronic scanning and which is obtained, for one of the groups, at the input of the power amplifiers and for the other group at the output of the power amplifiers. The common value of phase shift phi is advantageously taken equal to pi in order to reduce the even harmonics or pi /2 in order to reduce the odd harmonics. The reduction in the level of the harmonics is due to the fact that the latter no longer add together in phase in the signal from the active antenna but with a phase difference equal to (n-1) phi , n being the order of the relevant harmonic. <IMAGE>

Description

The present invention relates to active antennas with electronic scanning consisting of a network of radiating elements excited, on emission, by individual power amplifiers arranged in active modules placed as close as possible to the radiating elements. It relates, more particularly, to the reduction in the level of the harmonics emitted which are due to the inevitable non-linearities of the power amplifiers. This reduction in the level of harmonics is useful in order to avoid parasitic interference between the various items of equipment of an aircraft. It is also known that the absence of harmonics makes it possible to better optimize the circuits at the frequency of use, therefore to improve the power and the electronic efficiency.

A known method of reducing the level of harmonics is filtering. However, it cannot be used for harmonics that fall within the bandwidth of the antenna. In addition, with microwave antennas, the small volume available for the active modules which are arranged as close as possible to the radiating elements makes it difficult to produce harmonic filters.

Another known method for limiting the harmonics consists in reducing the operating power of the power amplifiers supplying the excitation signals, but this results in a lower efficiency and in a loss of transmission power for the active antenna.

It is also known to eliminate the even harmonics generated by an amplifier by equipping it at the output of two push-pull stages which work in parallel and in phase opposition to symmetrize the positive and negative arches of the amplified signal, and whose signals from output are added in phase opposition so that the voltages and therefore the powers are added. However, the production of such a microwave amplifier requires the use, at the output of the push pull stages, of a combiner, for example of the Lange or Wilkinson type, or of a balun device having the necessary bandwidth which are devices bulky with microwave losses.

The present invention aims to reduce the harmonics emitted by an active antenna while fighting against the aforementioned drawbacks.

It relates to a process for rejection of harmonics emitted by an active antenna with electronic scanning comprising a network of radiating elements each excited, on emission, by an individual power amplifier. this method consists in dividing the individual power amplifiers into at least two groups and in imposing on the amplifiers of each group a common non-zero phase shift value which is added to the individual phase shift values required by the electronic scanning and which is obtained for the 'one of the two groups at the input of the power amplifiers and for the other group at the output of the power amplifiers. (This common value does not modify the operation of the antenna).

The harmonics of order n generated by the group of power amplifiers having been imposed the common phase shift φ at input are all phase shifted by nφ, while those generated by the group of power amplifiers having been imposed the common phase shift φ at output are all phase shifted by φ, phases φ and nφ being measured with respect to the phase of the harmonic of order n, generated by an amplifier having no phase shift at its input, nor at its output ( the output phase shifter is very wide band, and therefore has the same transmission phase at the frequency φ and at the harmonic frequency nφ). Thus the harmonics present in the output signals of these two groups of power amplifiers are no longer in phase but out of phase with each other by (n-1) φ. They can therefore be canceled by combining the radiated signals coming from two power amplifiers belonging to different groups provided that these amplifiers are imposed the same individual phase shift value by electronic scanning and that the value (n-1) φ worth ¶. Everything then happens as if these harmonics were filtered by being reflected on the antenna. The power amplifiers then "see" a filter rejecting harmonics.

The common value of phase shift can be chosen equal to ± ¶ so as to reduce even harmonics (n = 2, 4 ...) by push-pull effect. It can also be chosen equal to ± ¶ / 2 so as to reduce the harmonic 3 (n = 3).

In the case where the active antenna has a network of radiating elements organized in horizontal rows and vertical columns, it is advantageous to assign the power amplifiers exciting the radiating elements of a row in two to one of the groups, and the remaining power amplifiers to the other group. This makes it possible to obtain a reduction rate of the harmonics independent of the sweep in bearing. One can also assign the power amplifiers exciting the radiating elements of a column on two to one of the groups and the remaining power amplifiers to the other group. This gives a reduction rate of harmonics independent of the elevation sweep. A combination of the two distributions is possible, for example, to have a reduction rate of even harmonics independent of the scanning in bearing and a reduction rate of harmonics 3 independent of the scanning in elevation.

The present invention also relates to a device for implementing the above method.

• la figure 1 représente, de manière schématique, la structure habituelle d'une antenne active d'émission ;
• la figure 2 représente, de manière schématique, la structure d'une antenne active d'émission selon l'invention ;
• les figures 3A et 3B illustrent une répartition possible, entre les entrées et les sorties des amplificateurs de puissance d'une antenne active selon l'invention, d'une valeur commune de déphasage ¶ imposée à tous les amplificateurs de puissance pour réduire les harmoniques pairs ;
• les figures 4A et 4B, illustrent une répartition possible, entre les entrées et les sorties des amplificateurs de puissance d'une antenne active selon l'invention, d'une valeur commune de déphasage ¶/2 imposée à tous les amplificateurs de puissance pour réduire l'harmonique 3 ;
• les figures 5A et 5B, illustrent une répartition possible, entre les entrées et les sorties des amplificateurs de puissance d'une antenne active selon l'invention, de deux valeurs communes additionnelles de déphasage ¶ et ¶/2 imposées à tous les amplificateurs de puissance pour réduire les harmoniques pairs et l'harmonique 3 et
• les figures 6 et 7, représentent un type d'éléments rayonnants hyperfréquences permettant de simuler une inversion de phase de leur signal d'excitation par simple retournement de leur structure.

Other advantages and characteristics of the invention will emerge from the description of an embodiment given by way of example. This description will be made below with reference to the drawing in which:

• Figure 1 shows, schematically, the usual structure of an active transmitting antenna;
• FIG. 2 schematically represents the structure of an active transmitting antenna according to the invention;
• FIGS. 3A and 3B illustrate a possible distribution, between the inputs and outputs of the power amplifiers of an active antenna according to the invention, of a common value of phase shift ¶ imposed on all the power amplifiers to reduce the even harmonics ;
• FIGS. 4A and 4B illustrate a possible distribution, between the inputs and outputs of the power amplifiers of an active antenna according to the invention, of a common value of phase shift ¶ / 2 imposed on all the power amplifiers to reduce harmonic 3;
• FIGS. 5A and 5B illustrate a possible distribution, between the inputs and outputs of the power amplifiers of an active antenna according to the invention, of two additional common values of phase shift ¶ and ¶ / 2 imposed on all the power amplifiers to reduce even harmonics and harmonic 3 and
• FIGS. 6 and 7 represent a type of radiating microwave elements making it possible to simulate a phase inversion of their excitation signal by simple reversal of their structure.

Figure 1 shows the usual structure of an active electronic scanning antenna. This comprises a linear network of m + 1 identical radiating elements Eo, ... Em arranged at the output of m + 1 identical active modules Mo, ... Mm. Each active module Mo, ... Mm contains an amplifier of power Ao, ... Am which excites the radiating element placed at the output of the act module considered and a phase shifter α o, ... α m adjustable between 0 and 360 °, placed at the input of the power amplifier.

(λ étant la longueur d'onde émise et d la distance d'espacement entre deux éléments rayonnants consécutifs) qui permet de construire un plan d'onde dans la direction souhaitée.A pilot circuit 1 delivers an excitation signal common to all the power amplifiers Ao, ... Am active modules Mo, ... Mm via adjustable phase shifters αo, ... α m to ensure electronic scanning. Electronic scanning consists in giving the main emission lobe of the active antenna the desired direction making an angle ϑ with respect to the plane of the network of radiating elements Ao, ... Am by imposing on each adjustable phase shifter α i, a phase shift value such as: $$\frac{\text{2¶}}{\text{λ}}{\text{d sin ϑ = (α}}_{\text{i + 1}}\text{-αi)}$$

(λ being the wavelength emitted and d the spacing distance between two consecutive radiating elements) which makes it possible to construct a wave plane in the desired direction.

In the case of an active microwave antenna, the power amplifiers of the active modules are integrated circuits AsGa employed at the limits of their characteristics generating a non-negligible amount of harmonics which monopolize part of the transmitted power decreasing the efficiency of the antenna and which can interfere with other radio equipment placed in the vicinity.

To reduce the amplitude of these harmonics at the global level of the signal emitted by the active antenna, it is proposed to create a push pull effect between pairs of power amplifiers. To do this, we share, as shown in Figure 2, the active modules (M′o, ... M′m) into two groups, for example those of even indices and those of odd indices and we impose, to the power amplifiers of the active modules, a non-zero common phase shift value φ which is added to the individual phase shift values required by the electronic scanning and which is obtained, for one of the groups of active modules, those of even indices , by a fixed phase shifter 3 placed at the input of the power amplifier and, for the other group of active modules, those of odd indices, by a fixed phase shifter 4 placed at the output of the power amplifier.

The fundamental signal emitted by each radiating element of the active antenna has thus been shifted by a value φ ce which does not change the pointing of the antenna or any of its characteristics. On the other hand, the harmonics are phase shifted by the value n φ (n being the rank of the harmonic considered), for the group of active modules of even indices, having the fixed phase shifter 3 placed at the input of the power amplifier, and of the value φ, for the group of active modules of odd indices having the fixed phase shifter 4 placed at the output of the power amplifier. The harmonics present in the signals emitted by the radiating elements excited by the two groups of active modules are no longer in phase, but out of phase with (n-1) φ, and can therefore be canceled by combination in the radiated signals coming from two power amplifiers of two active modules, neighboring or not, belonging to two different groups as long as these power amplifiers are imposed the same value or values close to individual phase shift by the pointing computer and that the value (n-1) φ worth ¶ + 2K¶.

The fixed phase shifter 3 placed at the input of the power amplifier of the active modules of even indices is only mentioned in FIG. 2 for explanatory purposes. In reality, it is confused with the adjustable phase shifter α i, the calibration of which is modified accordingly. It can also, for the particular value ¶ and certain types of power amplifier with differential inputs, be removed and replaced by an attack of the power amplifier on its inverting input instead of its non-inverting input or vice versa.

Note that the phase shifter 4 placed at the output of the power amplifier of the active modules of odd index must operate over a very wide band. In particular, it must have an identical transmission phase φ for the signal and the harmonic frequencies. This phase shifter can also, for the particular value ¶ and certain types of radiating elements, be removed and replaced by an inversion of the orientation of the radiating elements.

It is therefore possible, for certain types of active antennas and a common value of phase shift equal to ¶, to a structure which does not have a greater complexity than the conventional structure.

The common value of phase shift equal to ¶ has the advantage, in addition to the structural simplifications that it entails, of reducing the emission of even 2nf type harmonics.

ces multiples solutions ϑ_p se produisant pour : $${\text{α}}_{\text{i+1}}{\text{- α}}_{\text{i}}\text{= Δφ+ 2p¶}$$

avec p = indice entier de 1 à n et Δφ = différence de phase c'est à dire pour les incréments de déphasage de 360° des déphaseurs des modules.
On sait aussi que cela est impossible lorsque : $$\frac{\text{2¶d}}{\text{λ}}\text{< 0,5}$$

(le cas limite où $$\frac{\text{2¶d}}{\text{λ}}\text{= 0,5}$$

correspond au cas où un lobe de réseau évanescent à ϑ = 90° est produit par un déphasage α_i+1 - α_i = 180°).
En conséquence, tant que la condition $$\frac{\text{2¶d}}{\text{λ}}\text{< 0,5}$$

est remplie à la fréquence harmonique, il y a rejection de cet harmonique pour tous les sites.
Dans le cas contraire, la rejection de l'harmonique ne se produit que dans les angles de sites ϑ faibles, tels que -toujours à la fréquence harmonique- on ait : $${\text{sinϑ}}_{\text{p}}\text{<}\frac{\text{λ}}{\text{2d}}$$

FIGS. 3A and 3B illustrate a possible distribution, in two groups, of the active modules of an active antenna having a network of radiating elements organized in horizontal rows and vertical columns. Each active module is represented by a box containing, in FIG. 3A, the value of the phase shift produced at the input of its power amplifier independently of the individual value of phase shift imposed by the electronic scanning and, in FIG. 3B, the value of the phase shift produced at the output of its power amplifier, the common value of phase shift being chosen equal to ¶. In this distribution, the active modules in every second row are assigned to one of the groups while the remaining active modules are assigned to the other group. A reduction in even harmonics is thus obtained which is independent of the bearing angle taken by the beam of the active antenna and which is all the better as the angle of elevation taken by the beam of the active antenna is small (but sufficient in the usual coverage area of the antenna). More precisely when the antenna moves in elevation, the reduction of the even harmonics occurs as long as no lobe of network (or diffraction lobe) exists for these harmonics. We know that these lobes of networks are linked to the existence of multiple solutions in ϑ _p for the equation: $$\frac{\text{2¶}}{\text{λ}}{\text{d sin ϑ}}_{\text{p}}{\text{= (α}}_{\text{i + 1}}{\text{- α}}_{\text{i}}\text{)}$$

these multiple solutions ϑ _p occurring for: $${\text{α}}_{\text{i + 1}}{\text{- α}}_{\text{i}}\text{= Δφ + 2p¶}$$

with p = integer index from 1 to n and Δφ = phase difference, ie for the 360 ° phase shift increments of the phase shifters of the modules.
We also know that this is impossible when: $$\frac{\text{2¶d}}{\text{λ}}\text{<0.5}$$

(the limiting case where $$\frac{\text{2¶d}}{\text{λ}}\text{= 0.5}$$

corresponds to the case where an evanescent lobe at ϑ = 90 ° is produced by a phase shift α _{i + 1} - α _i = 180 °).
Consequently, as long as the condition $$\frac{\text{2¶d}}{\text{λ}}\text{<0.5}$$

is filled at the harmonic frequency, there is rejection of this harmonic for all the sites.
Otherwise, the rejection of the harmonic occurs only in the angles of weak sites sites, such that -always at the harmonic frequency- we have: $${\text{sinϑ}}_{\text{p}}\text{<}\frac{\text{λ}}{\text{2d}}$$

When this condition is not fulfilled, the rejection of the harmonics is only partial.

FIGS. 4A and 4B illustrate another possible distribution, in two groups, of the active modules of an active antenna having a network of radiating elements organized in horizontal rows and vertical columns, each active module is represented by a box containing, in the FIG. 4A, the value of the phase shift produced at the input of its power amplifier independently of the individual value of phase shift imposed by the electronic scanning and, in FIG. 4B, the value of the phase shift produced at the output of its power amplifier, the common value of phase shift φ being chosen equal to ¶ / 2 to reduce the emission of third order harmonics. In this distribution, the active modules of one column out of two are assigned to one group while the remaining modules are assigned to the other group. A reduction of order 3 harmonics is then obtained which is independent of the angle of elevation taken by the beam of the active antenna and which is all the better as the angle of bearing taken by the beam of the antenna active is weak. The evolution of this reduction in harmonic 3 as a function of the elevation scanning angle occurs under the same conditions as those described above.

FIGS. 5A and 5B illustrate a combination of the two distributions illustrated in FIGS. 3A, 3B and 4A, 4B to reduce both the even harmonics and the harmonic of order 3 emitted by an active antenna having radiating elements organized in horizontal rows and vertical columns. In this combination, the active modules successively undergo two different partitions in two groups.

During a first partition intended to reduce even harmonics, the active modules of one row out of two are assigned to a group and the remaining active modules to the other group, and a first value is applied to their power amplifiers phase shift common equal to ¶ which is added to the individual phase shift values required by the electronic scanning and which is imposed, for one group, at the input of the power amplifiers and, for the other group, at the output of the power amplifiers .

During the second partition intended to reduce third order harmonics, the active modules of one column out of two are assigned to one group and the active modules remaining to the other group, and a power amplifier is applied to their power amplifiers. second common phase shift value equal to ¶ / 2 which is added to the individual phase shift values necessitated by electronic scanning and at the first common phase shift value, which is imposed, for one group, at the input of the power amplifiers and, for the other group, at the output of the power amplifiers.

Each active module is represented by a box containing, in FIG. 5A, the value of the phase shift produced at the input of its power amplifier independently of the individual value of phase shift imposed by the electronic scanning and, in FIG. 5B, the value of the phase shift produced at the output of its power amplifier.

In this combination of FIGS. 5A and 5B, all of the power amplifiers of the active antenna have their signals generally phase-shifted by 3 ¶ / 2 which does not change the pointing of the antenna. It can also be seen that the power amplifiers belonging to two consecutive rows have phase-shifters having a phase difference of ¶ at the output allowing a reduction by recombination of even harmonics and that the power amplifiers belonging to two consecutive columns have phase-shifters at the output. having a phase difference of ¶ / 2 allowing a reduction by recombination of the harmonics of order 3.

FIGS. 6 and 7 illustrate a type of radiating element which can be used for an active microwave antenna and which has the advantage of facilitating the production of a phase shift of ¶ at the output of the power amplifiers.

This type of radiating element has a triplate structure. It consists of a pair of flared slots 10, 11 and an excitation line element 14. The flared slots 10, 11 of the Vivaldi type are hollow facing one another in the external metal walls 12, 13 of the triplate structure with their flared opening opening onto an edge of the triplate structure in the direction of propagation of the waves in free space. The excitation line element 14 is arranged in the median plane of the triplate structure. It runs on one of the sides of the slots 10, 11 parallel to these and has a bent end at a right angle which passes through slots 10, 11 perpendicular to their direction before ending in an enlarged part 15 carrying out an impedance matching.

The electric field radiated by this type of radiating element is perpendicular to the direction of the flared slots 10, 11 and parallel to the planes of the walls of the triplate structure. It reverses when the three-ply structure is turned over, as is clear from the comparison of FIGS. 6 and 7, which amounts to reversing the orientation of the end of the excitation line passing through the slots. This reversal of the electric field corresponds to a phase shift of ¶ of the excitation signal which is therefore achieved without the addition of a specific structure of phase shifter, and independently of the frequency.

This type of radiating element therefore makes it possible to impose on the power amplifiers which excite them, a common value of phase shift of ¶ realized for some at their input and for others at their output so as to reduce even harmonics without necessarily make the antenna structure more complex.

Claims (10)

Method for rejection of harmonics emitted by an active antenna with electronic scanning comprising a network of radiating elements (Eo, ... Em) each excited on emission by an individual power amplifier (Ao, ... Am) characterized in what it consists in dividing the power amplifiers (Ao, ... Am), of the active antenna in at least two groups and in imposing on the power amplifiers a common value of non-zero phase shift which is added to the values individual phase shift required by electronic scanning and which is obtained, for one of the groups, at the input of the power amplifiers and for the other group at the output of the power amplifiers. Method according to Claim 1, characterized in that the common value of phase shift imposed on the two groups of power amplifiers is equal to ± ¶. Method according to claim 1, characterized in that the common value of phase shift imposed on the two groups of power amplifiers is equal to ± ¶ / 2. Method according to claim 1, applied to an active antenna having its array of radiating elements organized in horizontal rows and vertical columns, characterized in that the power amplifiers exciting the radiating elements in every second row are assigned to one of the groups, the remaining power amplifiers being assigned to the other group. Method according to claim 1, applied to an active antenna having its array of radiating elements organized in horizontal rows and vertical columns characterized in that the power amplifiers exciting the radiating elements of one column out of two are assigned to one of the groups, the remaining amplifiers being assigned to the other group. Method according to Claim 1, applied to an active antenna having an array of radiating elements organized in horizontal rows and vertical columns, characterized in that even harmonics are reduced by first sharing the power amplifiers of the active antenna in first two sets, each first set containing the power amplifiers exciting the radiating elements of every other row, and imposing on the power amplifiers a first common phase shift of ± ¶ which is added to the individual phase shift values required by the scanning which is obtained for one of the first sets at the input of the power amplifiers and for the other at the output of the power amplifiers, and in that the harmonic 3 is reduced by sharing the power amplifiers again of the active antenna in two second sets, each second set containing the amplifiers of power exciting the radiating elements of a column on two, and by imposing on the power amplifiers a second common phase shift ± ¶ / 2 which is added to the individual values of phase shift required by the electronic scanning and at the first common phase shift, and which is obtained for one of the second sets at the input of the power amplifiers and for the other at the output of the power amplifiers. Rejection device of the harmonics emitted by an active antenna with electronic scanning comprising a network of radiating elements (Eo, ... Em) each excited on emission by an individual power amplifier (Ao, ... Am) characterized in what it includes: - first phase shift means (α i, 3) which are placed at the input of a certain number of power amplifiers forming a first group and which generate phase shift values equal to the individual phase shift values required by the increased electronic scanning of a non-zero common phase shift value φ; - And second phase shift means (4) which are placed at the output of a number of other power amplifiers forming a second group and which generates the common value of phase shift φ. Device according to claim 7 characterized in that the common value of phase shift φ is equal to ± ¶. Device according to claim 8 for rejection of the harmonics emitted by an active antenna with electronic scanning having an array of radiating elements each consisting of a pair of flared slots (10, 11) of Vivaldi type dug one above the other in the external metal walls (12, 13) of a triplate structure and excited by means of a line element (14) which is arranged on the median element of the triplate structure and whose end comes opposite slits (10, 11) perpendicular to their direction characterized in that the second phase shifting means consist of a change of orientation of ¶ of the end (15) of the excitation line (14) whatever the frequency . Device according to claim 7 for the rejection of harmonics emitted by an active antenna with electronic scanning having a network of radiating elements organized in horizontal rows and vertical columns characterized in that the first phase-shifting means are placed at the input of the exciting power amplifiers the radiating elements belonging to a row out of two and in that the second phase-shifting means are placed at the output of the remaining power amplifiers.

Images