Resonant circuit with variable frequency and impedance

Abstract

A resonant circuit comprises an input terminal and an output terminal and at least: a group of N resonators, where N1, the resonators having the same resonance frequency and the same antiresonance frequency; a first and a second impedance matching element having a non-zero reactance, the first element being in series with the group of resonators, and the second element being in parallel with the group of resonators, the resonant circuit comprising: first means for controlling the group of resonators, enabling the static capacitance of the group to be fixed at a first value; second control means, enabling the impedance of the first impedance matching element and that of the second element to be fixed at second values; the first and second values being such that the triplet of values composed of the static capacitance of the group, the impedance of the first element, and the impedance of the second element can be used to determine the following triplet of parameters: the characteristic impedance Z.sub.c of the assembly formed by the group, the first impedance matching element and the second matching element; the resonance frequency .sub.r of the assembly; the antiresonance frequency .sub.a of the assembly, in order to stabilize the impedance of the circuit at a chosen characteristic impedance.

Claims

1. A resonant circuit with a characteristic impedance stabilized at a chosen value, comprising an input terminal and an output terminal, and at least: a group of N resonators, where N>1, said group of N resonators having a same resonance frequency and a same antiresonance frequency; the group of N resonators configured to provide a variable static capacitance; a first impedance matching element and a second impedance matching element having a non-zero reactance, the first impedance matching element being in series with said group of N resonators, and the second impedance matching element being in parallel with said group of N resonators, said resonant circuit comprising: first means for controlling said group of N resonators, enabling a static capacitance of said group of N resonators to be fixed at a first value; second control means, enabling the impedance of the first impedance matching element and that of the second impedance matching element to be fixed at second values; said first and second values being such that: a triplet of values composed of the static capacitance of said group of N resonators, the impedance of the first impedance matching element, and the impedance of the second impedance matching element being used to determine the following triplet of parameters: the characteristic impedance Z.sub.c of an assembly formed by said group of N resonators, said first impedance matching element and said second matching element; the resonance frequency .sub.r of said assembly formed by said group of N resonators, said first impedance matching element and said second matching element; the antiresonance frequency .sub.a of said assembly formed by said group of N resonators, said first impedance matching element and said second matching element, in order to stabilize the impedance of said circuit at a chosen characteristic impedance.

2. The resonant circuit according to claim 1, wherein: said group of N resonators consists of a plurality of resonators: said first control means comprising a switching circuit for selecting and connecting one or more resonators.

3. The resonant circuit according to claim 2, wherein the resonators are bulk acoustic wave resonators which may have different geometrical dimensions.

4. The resonant circuit according to claim 2, wherein the resonators are surface acoustic wave resonators which may have different arrangements of electrodes on the surface of a piezoelectric substrate.

5. The resonant circuit according to claim 1, wherein: said resonator has a variable capacitance whose value is a function of the value of an electrical control signal; said first means comprising means for causing said electrical control signal to vary.

6. The resonant circuit according to claim 5, wherein said resonator is an electrostrictive resonator based on BST material.

7. The resonant circuit according to claim 1, wherein at least one impedance matching element is a capacitor, an inductance or a set of passive elements.

8. The resonant circuit according to claim 1, wherein at least one impedance matching element is an active circuit.

9. The resonant circuit according to claim 8, wherein the active circuit comprises transistors.

10. The resonant circuit according to claim 1, wherein the second impedance matching element is connected, on the one hand, to one of the input/output terminals, and on the other hand to an intermediate node between the group of resonators and said first impedance matching element.

11. The resonant circuit according to claim 1, wherein the second impedance matching element is placed between the input and output terminals, fitted in parallel with the assembly composed of the group of resonators and the first impedance matching element, placed in series.

12. The resonant circuit according claim 1, comprising: a first chip comprising at least said first control means for fixing the static capacitance of said group of resonators at a first value; a second chip comprising said group of resonators; means for the interconnection of said first control means with said group of resonators.

13. The resonant circuit according to claim 12, wherein the first chip also comprises the first and second impedance matching elements.

14. The resonant circuit according to claim 12, wherein the second chip also comprises the first and second impedance matching elements.

15. A filter comprising a set of resonant circuits according to claim 1.

16. A duplexer comprising a set of resonant circuits according to claim 1.

17. A device comprising a set of at least two resonant circuits according to claim 1 and having an input impedance and an output impedance, contained between an input port and an output port, comprising first means for controlling the group of resonators and second means for controlling the first and second impedance matching elements, for adjusting: the characteristic impedance of each of said circuits; the resonance and antiresonance frequencies of each of said circuits.

18. The device comprising a set of at least two resonant circuits according to claim 1, wherein said first means for controlling the group of resonators and said second means for controlling the first and second impedance matching elements cause variations of the triplets of values composed of the static capacitance of said group, the impedance of the first impedance matching element, and the impedance of the second impedance matching element, for the purpose of: adjusting the values of characteristic impedance of the two resonant circuits to fixed values; causing the resonance and antiresonance frequencies of the two resonant circuits to vary.

19. The device comprising a set of at least two resonant circuits according to claim 1, wherein said first means for controlling the group of resonators and said second means for controlling the first and second impedance matching elements cause variations of the triplets of values composed of the static capacitance of said group, the impedance of the first impedance matching element, and the impedance of the second impedance matching element, for the purpose of: causing the values of characteristic impedance of the two resonant circuits to vary; adjusting the resonance and antiresonance frequencies of the two resonant circuits to fixed values.

20. A resonant circuit with a characteristic impedance stabilized at a chosen value, comprising an input terminal and an output terminal, and at least: a group of N resonators, where N said group of N resonators having a same resonance frequency and a same antiresonance frequency; a first impedance matching element and a second impedance matching element having a non-zero reactance, the first impedance matching element being in series with said group of N resonators, and the second impedance matching element being in parallel with said group of N resonators, said resonant circuit comprising: first means for controlling said group of N resonators, enabling a static capacitance of said group of N resonators to be fixed at a first value; second control means, enabling the impedance of the first impedance matching element and that of the second impedance matching element to be fixed at second values; said first and second values being such that: a triplet of values composed of the static capacitance of said group of N resonators, the impedance of the first impedance matching element, and the impedance of the second impedance matching element being used to determine the following triplet of parameters: the characteristic impedance Z.sub.c of an assembly formed by said group of N resonators, said first impedance matching element and said second matching element; the resonance frequency .sub.r of said assembly formed by said group of N resonators, said first impedance matching element and said second matching element; the antiresonance frequency .sub.a of said assembly formed by said group of N resonators, said first impedance matching element and said second matching element, in order to stabilize the impedance of said circuit at a chosen characteristic impedance, wherein: said group of N resonators consists of a plurality of resonators: said first control means comprising a switching circuit for selecting and connecting one or more resonators.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be more clearly understood and other advantages will be apparent from the following description provided in a non-limiting way, and with the aid of the attached drawings, in which:

(2) FIGS. 1a, 1b and 1c show, respectively, a conventional BAW resonator stack, the equivalent circuit and the resonance and antiresonance frequencies of this resonator;

(3) FIG. 2 shows the response of a filter composed of acoustic resonators;

(4) FIG. 3 shows an example of topology for a UMTS duplexer based on BAW resonators;

(5) FIG. 4 shows a prior art topology intended to associate resonators and variable capacitors;

(6) FIGS. 5a and 5b show the effect of a capacitor connected in parallel and that of a capacitor connected in series on the electrical response of a resonator;

(7) FIGS. 6a and 6b show examples of prior art topology associating variable capacitors in series and in parallel with resonators;

(8) FIGS. 7a and 7b show the variation of the losses as a function of frequency for the two examples shown in FIGS. 6a and 6b;

(9) FIG. 8 shows the response in terms of mismatching of filters when they are centered on 0% and 11% of the resonance of the resonators in series, according to the topology shown in FIG. 7b;

(10) FIG. 9 shows an example of a ladder filter architecture;

(11) FIGS. 10a, 11a, 12a, 13a and 14a relate to the transmission response of the filter (S.sub.21 or S.sub.12) shown in FIG. 9;

(12) FIGS. 10b, 11b, 12b, 13b and 14b show, for the filter illustrated in FIG. 9: the curves C.sub.10b1, C.sub.11b1, C.sub.12b1, C.sub.13b1 and C.sub.14b1 relate to the reflection response S(4,4), expressed in dB, of the filter shown in FIG. 9; the curves C.sub.10b2, C.sub.11b2, C.sub.12b2, C.sub.13b2 and C.sub.14b2 relate to the reflection response S(5,5), expressed in dB, of the filter shown in FIG. 9;

(13) FIGS. 10c, 11c, 12c, 13c and 14c relate to the standing wave ratio (VSWR) calculated at the node Num4 shown in FIG. 9;

(14) FIGS. 10d, 11d, 12d, 13d and 14d show, for the filter illustrated in FIG. 9: the curves C.sub.10d1, C.sub.11d1, C.sub.12d1, C.sub.13d1 and C.sub.14d1 relating to the impedance of a resonator known as a relaxed resonator, that is to say one having no variable capacitor, this impedance being determined by its dimensions and the technological characteristics of the piezoelectric layer; the curves C.sub.10d2, C.sub.11d2, C.sub.12d2, C.sub.13d2 and C.sub.14d2 relating to the impedance response of the parallel component, composed of the relaxed resonator and the associated capacitors; the curves C.sub.10d3, C.sub.11d3, C.sub.12d3, C.sub.13d3 and C.sub.14d3 relating to the response of the series component, composed of a resonator identical to that of the parallel component, and associated capacitors which differ from those of the parallel component;

(15) FIGS. 15a, 15b, 15c and 15d relate, respectively, to the insertion losses, to the matching, to the impedance (without series capacitor and without parallel capacitor) and to the impedance of the pairs of resonant circuits (with variable values of series capacitance and parallel capacitance):

(16) FIGS. 16a, 16b, 16c and 16d relate, respectively, to superimposed responses of the same filter in which the surface of the resonators varies, and, respectively, to the insertion losses, to the matching, to the impedance (without series capacitor and without parallel capacitor) and to the impedance of the pairs of resonant circuits (with fixed values of series capacitance and parallel capacitance);

(17) FIG. 17 shows an example of filter topology that can be used in different frequency bands;

(18) FIG. 18 shows the frequency response of the topology shown in FIG. 17;

(19) FIG. 19 shows schematically a first example of a resonant circuit according to the invention;

(20) FIG. 20 shows the impedance of a resonator and the characteristic impedance of the resonant circuit according to the invention using said resonator;

(21) FIG. 21 shows schematically a second example of a resonant circuit according to the invention;

(22) FIG. 22 shows the equivalence of a large resonator and a set of small resonators arranged in parallel;

(23) FIG. 23 shows a first embodiment comprising means for selectively controlling a subset of resonators in the series of resonators shown schematically in FIG. 22 and means for controlling the reactive elements, and associating two chips comprising the resonators and the control means;

(24) FIG. 24 shows in detail an example of means of interconnection between two chips shown in FIG. 23;

(25) FIG. 25 shows the variation of impedance of a resonator based on perovskite material, as a function of the frequency and the applied voltage, which can be used in a resonant circuit according to the invention;

(26) FIG. 26 shows the variation of the resonance and antiresonance frequencies as a function of the applied voltage for a resonator based on perovskite material;

(27) FIG. 27 shows an embodiment comprising an electrostrictive resonator associated with ferroelectric capacitors;

(28) FIGS. 28a and 28b show examples of reactive components that can be used in a resonant circuit according to the invention.

DETAILED DESCRIPTION

(29) In a general way, the resonant circuit according to the present invention comprises:

(30) a group of N resonators, where N1, said resonators having the same resonance frequency and the same antiresonance frequency, and where the group may advantageously have a high intrinsic coupling coefficient (operating within the range of frequency agility) which ideally exceeds 10%, or even 50%,

(31) associated with a first impedance matching element which may be a variable capacitor in series with said group, and a second impedance matching element which may be a variable capacitor in parallel (operating within the bandwidth of the filter and the impedance matching, in association with said group).

(32) The resonant circuit according to the present invention further comprises:

(33) first means for fixing the static capacitance of said group at a first value;

(34) second control means for fixing the impedance of the first impedance matching element and that of the second impedance matching element at second values.

(35) The first and second control means for causing the variation of, respectively, the characteristic impedance Zc of said first group and the impedances of the first and second impedance matching elements are typically selective addressing means which may typically be switches with associated controls.

(36) FIG. 19 shows a first variant of a resonant circuit according to the invention, the non-zero reactance elements being capacitors.

(37) This figure shows, between an input terminal or port Pe and an output terminal or port Ps, the group represented schematically by a resonator R.sub.C0 with a variable static capacitance C.sub.0, associated with the non-zero reactance impedance matching element represented by a variable capacitor in series Cs and associated with the second non-zero reactance impedance matching element represented by a variable capacitor in parallel Cp, the arrows indicating a control for causing, notably, the impedance values to vary.

(38) It should be noted that the variable capacitors may also be replaced by variable inductances in series and in parallel, or by a combination of inductances and capacitors in series and in parallel. The properties of the assembly are then substantially different, and potentially open the way to new applications of the invention. In fact, the invention assumes the association of variable reactive elements of any kinds with a group of resonators forming a resonator with variable characteristic impedance.

(39) As a general rule, the impedance Z of a resonator is characterized, in a first approach, by the impedance of its equivalent Butterworth-Van Dyke (BVD) model, described above and illustrated in FIG. 1, that is to say without allowance for any losses:

(40) $Z=\frac{1}{j}\mathrm{where}=\frac{L_m{C}_m{C}_0{}^2-C_0-C_m}{L_m{C}_m{}^2-1}=\frac{N}{D}$

(41) All the elements of this model are correlated. This means that Lm, Cm and Co are interdependent variables, determined by the technological characteristics of the resonator. The capacitance Co is determined by the geometrical dimensions of the resonator, and Lm and Cm are calculated in such a way that they allow for the resonance frequency .sub.r and antiresonance frequency .sub.a of said resonator, according to the following equations:

(42) ${}_r^2=\frac{1}{L_m{C}_m}$${}_a^2=\frac{C_0+C_m}{L_m{C}_m{C}_0}={}_r^2\left(1+\frac{1}{}\right)$$=\frac{C_0}{C_m}$

(43) where

(44) $N=C_0.Math.\left(D-\frac{1}{}\right)$

(45) Additionally, the characteristic impedance Z.sub.c of the resonator outside the distinctive frequencies can be defined when C.sub.m.fwdarw.0 (and therefore, as a result, .sub.r.fwdarw. and .fwdarw.). This is equivalent to eliminating the piezoelectric effect. It can then be seen that Z.sub.c corresponds to the impedance of a capacitor of a size Co.

(46) $Z_c=\frac{1}{j{C}_0}$

(47) When components are associated in series and in parallel with this resonator, a new impedance Z, characterized by new resonance and antiresonance frequencies, and a new characteristic impedance Z.sub.c are obtained.

(48) More precisely, in the case of a resonator associated with a capacitor in series Cs and a capacitor in parallel Cp, according to the arrangement shown in FIG. 19, it can then be demonstrated that the impedance Z has the new resonance and antiresonance frequencies, determined, respectively, by the cancellation of the denominator or the numerator of , defined as follows:

(49) $\stackrel{\_}{Z}=\frac{1}{j\stackrel{\_}{}}$
where

(50) $\stackrel{\_}{}=\frac{\left(N+C_{p}D\right).Math.C_s}{N+\left(C_p+C_s\right).Math.D}$

(51) And

(52) ${\stackrel{\_}{}}_r={}_r\sqrt{1+\frac{1}{}.Math.\frac{C_0}{C_0+C_p+C_s}}$${\stackrel{\_}{}}_a={}_r\sqrt{1+\frac{1}{}.Math.\frac{C_0}{C_0+C_p}}$

(53) The new characteristic impedance Z.sub.c is defined as that of a capacitor of the following size:

(54) ${\stackrel{\_}{C}}_0=\frac{\left(C_0+C_p\right).Math.C_s}{C_0+C_p+C_s}$

(55) Therefore

(56) 0$\stackrel{\_}{Z_c}=\frac{1}{j\stackrel{\_}{C_0}}$

(57) FIG. 20 shows the frequency response of the impedance Z and that of Z, as well as that of C.sub.0.

(58) It is particularly interesting to note the constancy of the parameter C.sub.0 as a function of the frequency.

(59) Thus there are three independent equations for defining three parameters (Z.sub.c, .sub.r and .sub.a) on the basis of the intrinsic nature of the resonator (Z.sub.c, .sub.r and .sub.a) and the two associated capacitors. In other words, there is always one, and only one, solution for establishing the values Z.sub.c, .sub.r and .sub.a on the basis of Co, Cp and Cs. Thus, by varying the geometrical size of the resonator and the values of Cs and Cp, a pole and a zero associated with a chosen characteristic impedance can be freely positioned.

(60) This characteristic is a fundamental characteristic of the present invention.

(61) However, the range of possibilities for positive values of the capacitors Cp and Cs is limited in frequency by the lower boundary represented by the series resonance du resonator (.sub.r) and the upper boundary represented by its antiresonance (.sub.a), which is itself determined by the intrinsic coupling coefficient (characterized by the parameter in the equivalent model), which is also seen to appear in the form of constants in the equations of .sub.r and .sub.a.

(62) These limits are encountered when C.sub.s.fwdarw. and C.sub.p.fwdarw.0.

(63) FIG. 21 shows a second variant of the invention in which the capacitor Cp, connected in parallel, is connected differently from that of the preceding variant. It is placed in parallel with the assembly formed by the resonator and the series element Cs. In this case, is defined as follows:

(64) $\stackrel{\_}{}=\frac{.Math.\left(C_s+C_p\right)+C_s.Math.C_p}{+C_s}$${\stackrel{\_}{}}_r={}_r\sqrt{1+\frac{1}{}.Math.\frac{C_0}{C_0+C_s}}$${\stackrel{\_}{}}_a={}_r\sqrt{1+\frac{1}{}.Math.\frac{C_0}{C_0+C_{\mathrm{sp}}}}$

(65) Where

(66) $C_{\mathrm{sp}}=\frac{C_s.Math.C_p}{C_s+C_p}$

(67) Additionally, the characteristic impedance Z.sub.c is then defined as that of a capacitor of the following size:

(68) $\stackrel{\_}{C_0}=\frac{C_s+C_p}{C_s+C_0}.Math.\left(C_0+C_{\mathrm{sp}}\right)$

(69) As a general rule, in order to make the characteristic impedance of a resonator variable, it is necessary to cause its equivalent static capacitance to vary. A first way of doing this is to cause the geometrical dimensions of the resonator to vary. Although there are methods in the field of microwave cavity filters for modifying the physical dimensions of the resonators by mechanical means (using servomotors or simply adjusting screws), these means are not feasible in the case of integrated resonators.

(70) For this reason, the present applicants propose, by way of example, the solution described below in a first exemplary embodiment.

First Exemplary Embodiment

(71) As shown in FIG. 22, a single large resonator R.sub.C0 with a variable static capacitance C.sub.0 can be equivalent to a group of elementary resonators Ri arranged in parallel, the different resonators having identical resonance and antiresonance frequencies, but having a static capacitance such that the sum of the elementary static capacitances is equal to the static capacitance C.sub.0.

(72) Thus, in the context of a BAW resonator, with a size of 10N10N, it is possible to construct a checkerboard array of 100 BAW resonators with a size of NN, connected electrically in parallel with one another (the square shape suggested by the term checkerboard array is not limiting in itself: in fact, any geometrical surface may be placed in parallel with another).

(73) According to the present invention, it is then useful for the resonant circuit to comprise:

(74) at least said group formed by an assembly of elementary resonators Ri;

(75) of reactive components; and

(76) of selective control addressing means, which may typically be switches with associated controls.

(77) This configuration is illustrated in FIG. 23, showing this first embodiment comprising switches and associated controls, for switching elementary resonators and unitary reactive components.

(78) At the present time, it is possible to design BAW resonators with a wide range of variation on the basis of LNO material, using a known manufacturing process described in the paper by Pijolat, M. Deguet, C. Billard, C. Mercier, D., Reinhardt, A., Bias controlled electrostrictive longitudinal resonance in X-cut lithium niobate thin films resonator, Applied Physics Letters, Volume: 98, Issue: 23.

(79) However, this method is not currently compatible with the production of both switches and controls on the same substrate.

(80) For this reason, in order to produce a resonator R.sub.C0 made of LNO, comprising a set of elementary resonators Ri controllable from first control means, associated with a first series matching element Cs and a second parallel matching element Cp that can be controlled via control means, a configuration proposed in the present example provides, in combination:

(81) on the one hand, a chip Pu.sub.20 containing at least the resonator R.sub.C0 comprising at least two elementary resonators Ri, made of LNO;

(82) and, on the other hand, a chip Pu.sub.10 containing the switches and the reactive components Cp and Cs. More precisely, the chip Pu.sub.10 comprises the resonator control means 10, comprising a set of switches and I.sub.1i for I.sub.1j controlling the elementary resonators, and the control means 11+12, comprising a set of switches I.sub.2k and I.sub.2l.

(83) These two chips are then assembled facing one another (in a flip-chip configuration), via interconnections (using balls, copper pillars, etc., for example), using the copper pillar method in which copper contacts are formed in the shape of pillars on each of the chips, are aligned facing one another, and are then welded together by metal bonding methods (mechanical compression and heating) to reduce the footprint and resistance of these interconnections to a minimum.

(84) According to this approach, the switches can be designed using transistors whose sizing is based on a compromise between their internal parasitic resistance in the closed state and their internal parasitic capacitance in the open state. The elementary capacitors are formed in a conventional manner by sandwiching a thin dielectric layer between two metal electrodes. The quality of the dielectric material and the electrodes has an effect on the operation of the assembly.

(85) In order to reduce the resistive losses and parasitic capacitances of the assembly as far as possible, while withstanding any local overvoltages (typically 50 V for the power levels of the order of several watts which are used), these switches and these capacitors may advantageously be formed on a substrate of the high resistivity SOI type.

(86) FIG. 24 shows in greater detail an example of interconnections between the two chips in a case where a first chip is made of LNO and a second chip is made of HR-SOI (by an integrated circuit manufacturing technology using a thin layer of high-resistivity (HR) silicon placed on a layer of electrical insulation, which itself covers a massive silicon substrate).

(87) The chip Pu.sub.10 comprises:

(88) a substrate 100 which may be made of silicon;

(89) an electrical insulator 101;

(90) an electrical circuit 102 comprising the capacitors, the switches and control circuits;

(91) a passivation layer 103.

(92) The chip Pu.sub.20 comprises:

(93) a substrate 200 made of silicon;

(94) an acoustic insulator 201;

(95) an electrical insulator 202;

(96) a lower electrode 203;

(97) a layer of LNO material 204;

(98) an upper electrode 205.

(99) The chips Pu.sub.10 and Pu.sub.20 are interconnected by pillars I.sub.Pu10-Pu20 comprising an intermediate metallization part 300 and a contact metallization part 301.

Second Exemplary Embodiment

(100) According to a second embodiment, the resonant circuit comprises surface acoustic wave resonators. These resonators are advantageously formed on highly piezoelectric substrates such as lithium niobate, having a crystal orientation allowing the excitation of strongly coupled surface waves, for example those of the pseudo-surface wave type. In the case of SAW resonators, the characteristic impedance does not depend on the surface of a flat capacitor, but on a capacitance produced by the arrangement of interdigitated electrodes. A SAW resonator with 10 N electrodes can be produced by the parallel placing of 10 SAW resonators, each having N electrodes. The implementation in respect of the controls and interconnections may be equivalent to that described for the first exemplary embodiment.

Third Exemplary Embodiment

(101) According to a third embodiment, the resonators are electrostrictive resonators, that is to say those having piezoelectric properties that can be activated by the effect of a voltage. This is the case, for example, with BST material, a material having a crystallographic organization known as perovskite. This material was used in the case of patent application EP2405574.

(102) In this case, a single resonator can be used, the application of a d.c. voltage to the terminals of said resonator causing a variation of the static capacitance of the resonator, as illustrated, for example, in FIG. 25 which shows figures extracted from patent application EP 2405574, showing, respectively, the variation of impedance of a resonator based on perovskite material as a function of the frequency and the applied voltage. In this patent application, it is disclosed that the resonators are used in an operating region where the resonance and antiresonance frequencies are relatively constant. The production of a filter with a fixed central frequency and bandwidth, but a variable characteristic impedance, is proposed here. This type of filter can be ranked among dynamic adaptation systems, for example those described in the patents of Emeric De Foucauld and others (EP2509227 or EP2509222).

(103) It should also be noted that, in patent application EP 2405574, there is no question of causing the frequencies of the filter to vary. On the contrary, the resonators operate in an area where the filter thus formed remains stable in terms of frequency.

(104) In the area where a frequency variation is present, a large disparity is seen in the relative positioning of the resonance and antiresonance frequencies, as indicated in FIG. 26 which reproduces FIG. 12 of patent application EP 2405574.

(105) Thus the antiresonance frequency (Fp) decreases slightly (1.5%) when the applied voltage increases (up to 10 V), while the resonance frequency (Fs) decreases by 5% over the same voltage range. In this operating region, the characteristic impedance remains at 50+/10. This type of variation cannot be used to provide the solution according to the present invention, since the two distinctive frequencies and the characteristic impedance vary in an interdependent manner.

(106) However, on the basis of the approach developed in this patent application EP2405574, and with the aim of controlling the resonance and antiresonance frequencies according to the present invention, these electrostrictive resonators can be associated (in their operating region where the distinctive frequencies are approximately stable) with variable capacitors in series and in parallel. By means of this approach it is possible to design a new assembly (Co, Cs, Cp) which can be used for the independent determination of the characteristic impedance, the resonance frequency, and the antiresonance frequency.

(107) It should be noted that the use of BST for producing the resonators makes it possible to co-integrate variable capacitors within the same chip, according to the method described in patent EP2713508. In fact, by preventing the elastic movements of the layer (by means of a mechanical overload, for example), an acoustic resonator can be converted into a simple variable capacitor.

(108) This chip is then connected to an active circuit responsible for supplying a bias voltage for each resonator and capacitor, so that the values of the static capacitances of the resonators and of the capacitors added to the resonators can be adjusted dynamically.

(109) FIG. 27 shows schematically this third exemplary circuit according to the invention, in which two chips are interconnected. One chip combines the functions of resonators R.sub.C0 and the capacitors Cp and Cs, while another chip combines the control means for the ferroelectric capacitors with those for the electrostrictive resonators. More precisely, the first chip Pu.sub.11 comprises the control means 14 of the electrostrictive resonator and the control means 15, 16 of the ferroelectric capacitors. The second chip Pu.sub.21 comprises the electrostrictive resonator R.sub.C0 and the ferroelectric capacitors Cp and Cs.

Fourth Exemplary Embodiment

(110) According to a fourth embodiment, the reactive components are made with active circuits.

(111) In fact, there is a known way of making these reactances by using electronic circuits based on transistors, instead of using passive elements. The variable capacitors are formed in an ordinary manner with diodes connected inversely, the capacitance of which depends on the d.c. voltage applied to their terminals. The reactive components used may also be those described, for example, in U.S. Pat. No. 7,187,240, referring to the association of a BAW resonator with reactive components, the latter being made as indicated on the figures extracted from U.S. Pat. No. 7,187,240 and shown in FIGS. 28a and 28b.