WO2018101386A1

WO2018101386A1 - Phase-modulated high frequency oscillator array

Info

Publication number: WO2018101386A1
Application number: PCT/JP2017/042972
Authority: WO
Inventors: 澄人常木; 章雄福島; 慎吾田丸; 久保田　均; 広太朗水沼; 欣河野; 磯部　良彦; 和也鈴木; 成美水上
Original assignee: 国立研究開発法人産業技術総合研究所; 株式会社デンソー; 国立大学法人東北大学
Priority date: 2016-12-02
Filing date: 2017-11-30
Publication date: 2018-06-07
Also published as: JP6784377B2; JPWO2018101386A1

Abstract

Proposed is a phase-modulated high frequency oscillator array capable of varying a phase difference between a plurality of output signals as desired. The phase-modulated high frequency oscillator array 10 is provided with: a reference signal source 20 which outputs a reference signal having a frequency that is n times or 1/n times an oscillation frequency f; a plurality of magnetoresistive elements 31 with oscillation frequencies aligned with f by injection-locking; a plurality of bias voltage application mechanisms 32 which, by varying a bias voltage applied to at least two magnetoresistive elements 31 that are selected from the plurality of magnetoresistive elements 31 as desired, varies a phase difference between output signals from the at least two magnetoresistive elements 31; and a plurality of removal means 40 which remove the reference signal entering a signal path for transmitting respective output signals from the plurality of magnetoresistive elements 31.

Description

Phase modulation high frequency oscillator array

The present invention relates to a phase modulation high frequency oscillator array.

Communication technology that uses high-frequency signals (about 100 MHz to 300 GHz) including microwave bands (about 1 GHz to 300 GHz) has become increasingly important in recent years, and the technology is widely used in mobile phones, wireless communications, satellite broadcasting, radar, etc. ing. In particular, radar is increasingly used to control the directivity of high-frequency signals. Against this background, a phase modulation high frequency oscillator array that controls the directivity of a high frequency signal using a plurality of oscillators has been studied. In this type of phase-modulated high-frequency oscillator array, the reference signal output from the reference signal source is used to synchronize the reference signal source and each oscillator while transmitting the output signal from each oscillator. There is a type that performs phase modulation by forming a phase difference between output signals from each oscillator by a phase shifter provided in a signal path. As the phase shifter, for example, a phase shifter using a delay line and a phase shifter using an electronic circuit are known. As an oscillator, a voltage controlled oscillator is known.

However, the phase shifter using the delay line has the disadvantage that the circuit becomes redundant and the phase difference is fixed and cannot be changed arbitrarily. In addition, a phase shifter using an electronic circuit has a drawback that a frequency band capable of forming a phase difference is generally narrow and signal loss is large. On the other hand, when a voltage controlled oscillator is used as the oscillator, there is a drawback that it is difficult to reduce the size in a frequency band of several GHz or more.

Therefore, an object of the present invention is to propose a phase-modulated high-frequency oscillator array that can eliminate the above-described drawbacks and can arbitrarily change the phase difference between a plurality of output signals.

In order to solve the above-described problem, a phase modulation high frequency oscillator array according to the present invention is a phase modulation high frequency oscillator array that synchronizes and oscillates at a frequency f, and (i) when n is an integer of 2 or more, the frequency a reference signal source that outputs a reference signal having a frequency n times or 1 / n times f; and (ii) a plurality of magnetoresistive elements, wherein the reference signal is injected into each of the plurality of magnetoresistive elements. And (iii) a plurality of biases for applying a bias voltage to each of the plurality of magnetoresistive elements, and the frequency of the output signal from each of the plurality of magnetoresistive elements matches the frequency f. A voltage application mechanism, wherein at least two magnetoresistive elements are changed by changing a bias voltage applied to at least two magnetoresistive elements arbitrarily selected from the plurality of magnetoresistive elements. A plurality of bias voltage applying mechanisms for changing the phase difference between the output signals; and (iv) a plurality of removing means for removing a reference signal mixed in a signal path for transmitting the output signals from the plurality of magnetoresistive elements. And comprising.

It is a block diagram which shows schematic structure of the phase modulation high frequency oscillator array in connection with embodiment of this invention. It is explanatory drawing which shows the structure of the magnetoresistive element concerning embodiment of this invention. It is a block diagram which shows schematic structure of the phase modulation high frequency oscillator array in connection with embodiment of this invention. It is a block diagram which shows schematic structure of the phase modulation high frequency oscillator array in connection with embodiment of this invention. It is a block diagram which shows schematic structure of the phase modulation high frequency oscillator array in connection with embodiment of this invention. It is a block diagram which shows schematic structure of the phase modulation high frequency oscillator array in connection with embodiment of this invention. It is a block diagram which shows schematic structure of the phase modulation high frequency oscillator array in connection with embodiment of this invention. It is a block diagram which shows the specific structure of the phase modulation high frequency oscillator array in connection with the Example of this invention. 6 is a graph showing the measurement result of the spectral intensity of the magnetoresistive element when the bias voltage applied to the magnetoresistive element according to the example of the present invention is changed in the range of 235 to 255 mV. 3 is a graph showing measurement results of spectral intensity of a magnetoresistive element when a bias voltage applied to the magnetoresistive element according to an example of the present invention is changed in a range of 200 to 220 mV. 6 is a graph showing an oscillation frequency of a magnetoresistive element when a bias voltage applied to the magnetoresistive element according to an example of the present invention is changed in a range of 235 to 255 mV. 6 is a graph showing an oscillation frequency of a magnetoresistive element when a bias voltage applied to the magnetoresistive element according to an example of the present invention is changed in a range of 200 to 220 mV. It is a graph which shows the relationship between the oscillation frequency when the magnetoresistive element concerning the Example of this invention is injection-locked, and a measurement output. It is a graph which shows the relationship between the oscillation frequency when the magnetoresistive element concerning the Example of this invention is injection-locked, and a measurement output. The oscillation frequency of the first magnetoresistive element when the bias voltage applied to the first magnetoresistive element according to the embodiment of the present invention is fixed and the bias voltage applied to the second magnetoresistive element is changed. It is a graph which shows the result of having measured _fMR1 and the oscillation frequency _fMR2 of the 2nd magnetoresistive element. The first magnetoresistive element output signal when the bias voltage applied to the first magnetoresistive element according to the embodiment of the present invention is fixed and the bias voltage applied to the second magnetoresistive element is changed, and It is a graph which shows the result of having measured the phase difference between the output signals of the 2nd magnetoresistive element.

Hereinafter, the mechanism of action of the phase-modulated high-frequency oscillator array according to the embodiment of the present invention, which is estimated by the inventors, will be described.
As reported in Japanese Patent Laid-Open No. 2006-295908, it is known that a magnetoresistive element (hereinafter referred to as “MR element”) functions as a high-frequency oscillation element. By microfabrication using a semiconductor process, an MR element that functions as a small high-frequency oscillation element (for example, a diameter of about 30 nm to 300 nm and a height of about 30 nm to 100 nm) can be manufactured. The MR element includes a magnetization free layer (hereinafter simply referred to as “free layer”), a nonmagnetic layer, and a magnetization fixed layer (hereinafter simply referred to as “fixed layer”). When a direct current is passed from the free layer to the fixed layer or from the fixed layer to the free layer, the electron spin of the free layer is excited by the spin torque of the direct current, and the magnetization between the free layer and the fixed layer changes. The relative angle vibrates over time. This vibration is converted into a high-frequency electric signal through the tunnel magnetoresistance effect or the giant magnetoresistance effect, and as a result, a high-frequency signal is oscillated from the MR element. Since the pre-synchronization oscillation frequency f _MR of the high-frequency signal oscillated from the MR element depends on the resonance condition of the MR element (for example, a direct current (or DC voltage) applied to the MR element, a magnetic field, or a temperature), By changing the resonance condition of the MR element, the pre-synchronization oscillation frequency f _MR can be changed.

On the other hand, the high-frequency signal oscillated from the MR element has a problem that the peak width is wide and the pre-synchronization oscillation frequency f _MR is unstable. In order to solve this problem, a technique called injection locking is known. Injection locking refers to the difference between the frequency f _{REF of the} reference signal injected (supplied) from the reference signal source to the MR element and the pre-locking oscillation frequency f _MR of the MR element with a certain bandwidth W (hereinafter “ This is a phenomenon in which the pre-synchronization oscillation frequency f _MR of the MR element coincides with the frequency f _{REF of the} reference signal when it is smaller than the “synchronizable bandwidth”, that is, when the relationship of Expression (1) is satisfied.
(| f _REF −f _MR | <W) (1)

In injection locking, the pre-synchronization oscillation frequency f _MR of the MR element not only matches the frequency f _REF of the reference signal, but also accurately follows and synchronizes with the frequency f _{REF of the} reference signal. For this reason, by using injection locking, the peak width of the high-frequency signal oscillated from the MR element can be narrowed, and the pre-synchronization oscillation frequency f _MR can be stabilized. If injection locking is used, not only can the oscillation frequency f _MR of the MR element be synchronized but also the phase difference Δ, which is the above-mentioned problem, can be arbitrarily controlled. The phase difference Δφ between the frequency f _{REF of the} reference signal and the oscillation frequency f _MR before synchronization of the MR element satisfies the relationship of equation (2).
Δφ = arcsin {(f _MR −f _REF ) / W} (2)

Here, arcsin is an inverse sine function. When the pre-synchronization oscillation frequency f _MR of the MR element coincides with the frequency f _{REF of the} reference signal by injection locking, f _MR is fixed while being coincident with f _REF . Here, it is assumed that the oscillation frequency of the MR element after injection locking is assumed to change due to a change in resonance conditions (for example, a direct current (or DC voltage) applied to the MR element, a magnetic field, or a temperature). When the oscillation frequency f _MR of the MR element is substituted into the equation (2), the phase difference Δφ between the high frequency signal oscillated from the MR element after injection locking and the reference signal can be obtained from the equation (2). That is, f _MR in the equation (2) originally means the oscillation frequency before synchronization of the MR element, but the oscillation frequency of the MR element after injection locking changes depending on the change of the resonance condition. It may be interpreted that it also means the oscillation frequency of the MR element assumed when assumed. For this reason, by changing the resonance condition of the MR element after injection locking, the oscillation frequency of the MR element after injection locking is matched with the frequency f _{REF of the} reference signal from the MR element after injection locking. The phase difference Δφ between the oscillated high frequency signal and the reference signal can be arbitrarily changed.

Next, consider that two MR elements are synchronized using one reference signal. For convenience of explanation, one of the two MR elements is called a first MR element, and the other is called a second MR element. When the pre-synchronization oscillation frequency f _MR1 of the first MR element is different from the pre-synchronization oscillation frequency f _MR2 of the second MR element with respect to the frequency f _{REF of the} reference signal, oscillation is performed from each of the two MR elements. The phase difference Δφ _MR of the high-frequency signal satisfies the relationships of equations (3) to (5).
Δφ _MR = Δφ _MR1 −Δφ _MR2 (3)
Δφ _MR1 = arcsin {(f _MR1 −f _REF ) / W _MR1 } (4)
Δφ _MR2 = arcsin {(f _MR2 −f _REF ) / W _MR2 } (5)

Here, W _MR1 and W _MR2 are bandwidths with which the first MR element and the second MR element can be synchronized, respectively. For example, by fixing the DC voltage applied to the first MR element and changing the DC voltage applied to the second MR element while the two MR elements are synchronized by injection locking, f _MR1 While _maintaining = f _MR2 = f _REF , the phase difference Δφ _MR of the high-frequency signal oscillated from each of the two MR elements can be arbitrarily changed.

Since the phase of the high-frequency signal oscillated from the MR element is shifted in the process of propagating through the signal path, the equations (3) to (5) are expressed by the high-frequency signal oscillated from each of the two MR elements through the signal path. It should be noted that the phase difference is not necessarily the same.

The present inventors diligently examined the conditions for causing injection locking. When the oscillation frequency of the MR element after the synchronization is f, the frequency f _{REF of the} reference signal is equal to f, and the oscillation frequency of the MR element Rather than injection locking only when f _MR is equal to f, the frequency f _{REF of the} reference signal is equal to n times or 1 / n times f or approximately equal to n times or 1 / n times f. It turns out that sometimes happens. Here, n is an integer of 2 or more. “Approximately equal” means that synchronization is possible if the difference between f _REF and n times or 1 / n times f is within 40%, preferably within 30% of f. For this reason, in the present specification, “almost equal frequency” that is n times or 1 / n times f is referred to as “synchronizable frequency”.

The present inventor pays attention to such technical knowledge and uses a reference signal having a frequency n times or 1 / n times f when f is an oscillation frequency of a plurality of MR elements after synchronization. The phase-modulated high-frequency oscillator array that injects and locks the MR elements was invented. Due to injection locking, the oscillation frequencies of the plurality of MR elements coincide with f. By changing the resonance condition of at least two MR elements arbitrarily selected from among the plurality of injection-locked MR elements, the oscillation frequency of the high-frequency signal oscillated from each of the plurality of MR elements is kept matched with f. The phase difference between the high-frequency signals output from the selected at least two MR elements can be arbitrarily controlled. In this way, phase modulation can be performed by controlling the phase difference between the high-frequency signals. The phase modulation high frequency oscillator array can be applied as, for example, a phased array radar. As a mechanism for changing the resonance condition of the MR element, for example, a bias voltage applying mechanism for changing a bias voltage, which is a DC voltage applied to the MR element, or a magnetic field applying mechanism for changing the magnetic field strength applied to the MR element is suitable. It is. By changing the bias voltage or the magnetic field strength applied to each of the plurality of MR elements, the phase difference between the high-frequency signals can be changed while keeping the oscillation frequency of the high-frequency signal oscillated from each of the plurality of MR elements at f. It can be controlled arbitrarily.

In the phase modulation high-frequency oscillator array configured as described above, the reference signal is mixed in a signal path for transmitting a high-frequency signal oscillated from each of the plurality of MR elements. If the oscillation frequency f of the plurality of MR elements after synchronization and the frequency f _{REF of the} reference signal are equal, the phase-modulated high-frequency signal from the MR element and the reference signal are distinguished and output from the phase-modulated high-frequency oscillator array. It becomes difficult to do. When the frequency f _{REF of the} reference signal is n times or 1 / n times the oscillation frequency f of the plurality of MR elements after synchronization, the removing means for selectively removing the reference signal is used to remove the reference signal from the MR element. The phase-modulated high-frequency signal can be output from the phase-modulated high-frequency oscillator array. As such a removing means, for example, a filter having a frequency characteristic in which the frequency f _{REF of the} reference signal is set as an attenuation band (or a stop band) and the oscillation frequency f of the MR element is set as a pass band can be used. When the frequency f _{REF of the} reference signal is n times the oscillation frequency f of the plurality of MR elements after synchronization, a low-pass filter can be used as the removing means. When the frequency f _{REF of the} reference signal is 1 / n times the oscillation frequency f of the plurality of MR elements after synchronization, a high-pass filter can be used as the removing means. The removing means is not limited to the above-described filter, and has, for example, a frequency characteristic in which the frequency f _{REF of the} reference signal is an attenuation region (or a stop region), and a high frequency signal oscillated from the MR element is used as a radio wave. An output antenna may be used.

According to the experiment by the present inventor, the injection locking is performed by multiplying f by n times or 1 / n times, such as when the frequency f _{REF of the} reference signal is 2/3 times the oscillation frequency f of the MR element after synchronization. It has been confirmed that it occurs at frequencies other than frequencies. However, a half-frequency such as 2/3 times f is ½ times f and the energy required for injection locking is high because the locking width is narrow. Therefore, a lot of power from the reference signal source must be injected into the MR element, which is not practical.

Next, the configuration of the phase-modulated high-frequency oscillator array according to the embodiment of the present invention will be described with reference to FIGS. Here, the same reference numerals indicate the same elements, and duplicate descriptions are omitted.
FIG. 1 is a block diagram showing a schematic configuration of a phase modulation high frequency oscillator array 10 according to an embodiment of the present invention. The phase modulation high-frequency oscillator array 10 includes a reference signal source 20, N high-frequency oscillators 30-1, 30-2,..., 30-N and N removal means 40-1, 40-2,. -N. Here, N is an integer of 2 or more. Each of the N high-frequency oscillators 30-1, 30-2,..., 30-N includes an MR element 31 and a bias voltage applying mechanism 32. The phase modulation high-frequency oscillator array 10 is configured to synchronize with the oscillation frequency f, and the reference signal source 20 outputs a reference signal having a frequency n times or 1 / n times the oscillation frequency f. n is an integer of 2 or more, and is preferably an integer of 2 to 12, for example. As the reference signal source 20, for example, a voltage controlled oscillator, a phase locked loop oscillator, or the like can be used. The reference signal source 20 is composed of a combination of a primitive signal source that outputs a signal having an oscillation frequency f and a multiplier that multiplies the oscillation frequency f of the signal output from the atomic signal source by n times or 1 / n times. May be.

The reference signal source 20 supplies a reference signal to each of the N high-frequency oscillators 30-1, 30-2, ..., 30-N. Each of the high-frequency oscillators 30-1, 30-2,..., 30-N injects a reference signal supplied from the reference signal source 20 into the MR element 31, so that N high-frequency oscillators 30-1, 30- The oscillation frequency of the high frequency signal oscillated from each of the MR elements 31 of 2,. In this embodiment, one reference signal source 20 is shared by N high frequency oscillators 30-1, 30-2,..., 30-N, but N high frequency oscillators 30-1, 30-2, ..., 30-N may be provided with a dedicated reference signal source.

Here, the MR element 31 will be described with reference to FIG. The MR element 31 includes a free layer 31A, a nonmagnetic layer 31B, and a fixed layer 31C. When a bias voltage is applied, the MR element 31 functions as a high-frequency oscillation element that resonates at the oscillation frequency f _MR and outputs a high-frequency signal. At this time, the MR element 31 oscillates not only the fundamental wave component of the high frequency signal but also its harmonic component. Therefore, the fundamental component of the high frequency signal oscillated from the MR element 31 may be handled as an output signal output from the MR element 31, or the harmonic component of the high frequency signal oscillated from the MR element 31 may be treated as the MR element. You may handle as an output signal output from 31. For example, when the harmonic component of the high frequency signal oscillated from the MR element 31 is handled as an output signal output from the MR element 31, the fundamental component of the high frequency signal oscillated from the MR element 31 is removed by a filter. May be. The MR element 31 is classified into a horizontal type, a vertical type, and a magnetic vortex type according to the direction of magnetization, and any of them can be used.

The material of the free layer 31A is a ferromagnetic material. Typical examples of the ferromagnetic material include iron-based or iron-based alloys (for example, CoFeB) such as Fe, Co, and Ni. In addition, for example, a single-layer film of FeB, a two-layer film of CoFeB and NiFe, a two-layer film of CoFe and NiFe, a three-layer film of CoFeB, Ru and CoFeB, a three-layer film of CoFeB, Ru and NiFe, a CoFe and Ru and A three-layer film of NiFe, a four-layer film of CoFeB, CoFe, Ru, and CoFe, and a four-layer film of CoFeB, CoFe, Ru, and NiFe can also be used. The crystal structure of the free layer 31A is preferably a BCC structure, for example. The thickness of the free layer 31A is, for example, about 1.5 to 20 nm.

The material of the nonmagnetic layer 31B is classified into a nonmagnetic metal and an insulator. A material using a nonmagnetic metal as the material of the nonmagnetic layer 31B is called a GMR element, and a material using an insulator is called a TMR element. As the nonmagnetic metal that functions as the nonmagnetic layer 31B, for example, Cu, Ag, Cr, or the like can be used. The thickness of the nonmagnetic metal is, for example, about 0.3 nm to 10 nm. In particular, when Cu or Ag that realizes a large MR ratio is used, the thickness is, for example, 2 nm to 10 nm. As the insulator functioning as the nonmagnetic layer 31B, for example, various dielectrics such as oxides, nitrides, halides such as Mg, Al, Si, Ca, and Li can be used. In particular, magnesium oxide having both a large MR ratio and a small sheet resistance is preferable. When the above-described oxide or nitride is used as the nonmagnetic layer 31B, some oxygen or nitrogen deficiency may exist in the oxide or nitride. The thickness of the insulator is, for example, about 0.3 nm to 2 nm. In the case of a TMR element, the nonmagnetic layer 31B is also called a tunnel barrier layer.

The material of the fixed layer 31C is a ferromagnetic material. Typical examples of the ferromagnetic material include iron-based or iron-based alloys such as Fe, Co, and Ni (for example, FeB and CoFeB). When an amorphous state is desired as an intermediate state for the convenience of the manufacturing method, alloys CoFeBSi, CoFeBTi, CoFeBCr, CoFeBV, etc. to which B, Si, Ti, Cr, V, etc. are added can be used. In particular, in the case of perpendicular magnetization, alloys such as CoPt, CoPd, FePt, and FePd, multilayer films of these alloy thin films, or alloys obtained by adding B, Cr, or the like to these alloys can be used. The crystal structure of the fixed layer 31C is preferably a BCC structure, for example. In order to crystallize the amorphous film, for example, heat treatment (annealing) may be performed. Regarding the film thickness of the fixed layer 31C, when the film thickness is reduced, there are problems that the stability of the magnetization direction with respect to current and heat is lowered, and that it is difficult to form a continuous film. On the contrary, when the film thickness of the fixed layer 31C is increased, there arises a problem that a leakage magnetic field from the fixed layer 31C to the free layer 31A is increased, and a problem that microfabrication becomes difficult. In view of these circumstances, the film thickness of the fixed layer 31C is preferably about 2 to 30 nm, for example.

Although the free layer 31A, the nonmagnetic layer 31B, and the fixed layer 31C, which are the basic components of the MR element 31, have been described, for example, the magnetization direction of the extraction electrode layer and the fixed layer 31C unless the object of the present invention is contrary. A first support layer that assists in maintaining the magnetization, a second support layer that assists in adjusting the magnetization easy direction of the free layer 31A, and a third that assists in increasing the read signal when reading the magnetization direction of the free layer 31A. Layers such as a support layer (read-only layer) and a capping layer may be provided. Here, when the magnetization direction of the free layer 31A is a horizontal type or a magnetic vortex type, the material of the second support layer is an iron-based or iron-based alloy such as Fe, Co, Ni (for example, FeB). And CoFeB) are typical. When the magnetization direction of the free layer 31A is vertical, the second support layer is made of an alloy such as CoPt, CoPd, FePt, or FePd, or a multilayer film of these alloy thin films, or B, Cr An alloy to which etc. are added can be used. By providing the MR element 31 with the second support layer, the oscillation frequency f _MR may be easily changed when a bias voltage is applied to the MR element 31.

Here, we return to the explanation of FIG. The bias voltage application mechanism 32 applies a bias voltage to the MR element 31. The bias voltage application mechanism 32 changes the bias voltage applied to the MR element 31 in a state where the MR element 31 is injection-locked, thereby changing the phase difference between the output signal output from the MR element 31 and the reference signal. Can be controlled. Such phase modulation can be applied to control of a phase difference between an output signal output from the i-th high-frequency oscillator 30-i and an output signal output from the j-th high-frequency oscillator 30-j, for example. It is. However, i and j are mutually different arbitrary integers of 1 or more and N or less. The bias voltage application mechanism 32 includes a bias voltage source and a signal path for applying the bias voltage. The MR element 31 generally has a resistance of about 10 to 1 kΩ, and requires about 100 to 700 mV as a bias voltage for satisfying the resonance condition of the MR element 31. Is preferably a DC power supply with a maximum output of about 1 V and 100 mA.

The removing means 40-i selectively removes the reference signal mixed in the signal path for transmitting the output signal output from the high frequency oscillator 30-i. Here, i is an arbitrary integer of 1 or more and N or less. As such a removing means 40-i, for example, a filter having a frequency characteristic in which the frequency f _{REF of the} reference signal is set as an attenuation band (or stop band) and the oscillation frequency f of the MR element 31 after injection locking is set as a pass band. Can be used. When the frequency f _{REF of the} reference signal is n times the oscillation frequency f of the plurality of MR elements 31 after injection locking, a low-pass filter can be used as the removing means 40-i. When the frequency f _{REF of the} reference signal is 1 / n times the oscillation frequency f of the plurality of MR elements 31 after injection locking, a high-pass filter can be used as the removing means 40-i. The removing unit 40-i is not limited to the above-described filter, and has, for example, a frequency characteristic in which the band of the reference signal frequency f _REF is an attenuation region, and an output signal output from the MR element 31 is a radio wave. May be used as an antenna. When an antenna is used as the removing unit 40-i, a filter for selectively removing the reference signal is not necessary, and the antenna may be connected to the output of the high frequency oscillator 30-i. An amplifier having a frequency characteristic with a low gain at the frequency f _REF of the reference signal and a high gain at the oscillation frequency f of the MR element 31 after injection locking may be used as the removing means 40-i.

As shown in FIG. 3, the high frequency oscillator 30-i superimposes a bias voltage, which is a DC signal output from the bias voltage application mechanism 32, on a reference signal, which is an AC signal output from the reference signal source 20. May be provided to the MR element 31. The bias tee 33 includes a capacitor element and an inductor element. When a DC signal may be superimposed on a signal path through which the reference signal from the reference signal source 20 propagates or a signal path through which the output signal from the MR element 31 propagates, the bias voltage is not used and the bias voltage is not used. A bias voltage supplied from the application mechanism 32 may be directly applied to the MR element 31. 3 shows only the high-frequency oscillator 30-i among the high-frequency vibrators 30-1, 30-2,..., 30-N for convenience. -2, ..., 30-N may all include a bias tee 33.

As shown in FIG. 4, the high-frequency oscillator 30-i may include a magnetic field application mechanism 34 that applies a magnetic field to the MR element 31. The magnetic field application mechanism 34 controls the phase difference between the output signal output from the MR element 31 and the reference signal by changing the magnetic field strength applied to the MR element 31 while the MR element 31 is injection-locked. can do. Such phase modulation can be applied to control of a phase difference between an output signal output from the i-th high-frequency oscillator 30-i and an output signal output from the j-th high-frequency oscillator 30-j, for example. It is. However, i and j are arbitrary different integers of 2 or more and N or less. For example, a permanent magnet or an electromagnet can be used as the magnetic field application mechanism 34. According to the electromagnet, it is easy to change the magnetic field strength. An electromagnet generates a magnetic force by energizing a coil wound around a magnetic body. However, the magnetic body is not always essential, and a magnetic field generated by energizing the coil is applied to the MR element 31. Also good. 4 shows only the high-frequency oscillator 30-i among the high-frequency vibrators 30-1, 30-2,..., 30-N for the sake of convenience. -2, ..., 30-N may all include the magnetic field application mechanism 34.

As shown in FIG. 5, the phase-modulated high-frequency oscillator array 10 may include an amplifier 50-i that amplifies an output signal output from the high-frequency oscillator 30-i. Since the intensity of the output signal output from the high frequency oscillator 30-i is as low as about 10 mV (rms), for example, it is preferable to amplify the output signal using the amplifier 50-i. Here, rms means the root mean square. FIG. 5 shows only the high-frequency oscillator 30-i among the high-frequency vibrators 30-1, 30-2,..., 30-N for convenience. , 50-N includes N amplifiers 50-1, 50-2,..., 50-N that amplify output signals output from the N high-frequency vibrators 30-1, 30-2,. Also good.

In the configuration shown in FIG. 1, the bias voltage from the bias voltage application mechanism 32 and the reference signal from the reference signal source 20 are supplied to the MR element 31 through the same signal path, but as shown in FIG. The bias voltage from the bias voltage application mechanism 32 and the reference signal from the reference signal source 20 may be supplied to the MR element 31 through different signal paths.

FIG. 7 is a block diagram showing a schematic configuration of the two-output type phase modulation high-frequency oscillator array 10. The configuration shown in the figure is a configuration when N = 2 in FIG. The phase modulation high-frequency oscillator array 10 includes a reference signal source 20, two high-frequency oscillators 30-1 and 30-2, and two removal means 40-1 and 40-2. Each of the high frequency oscillators 30-1 and 30-2 includes an MR element 31, a bias voltage applying mechanism 32, a bias tee 33, and a magnetic field applying mechanism 34, and an output signal and a high frequency output from the high frequency oscillator 30-1. The phase difference between the output signal output from the oscillator 30-2 can be controlled. The removing means 40-1 and 40-2 selectively remove the reference signals mixed in the signal path for transmitting the output signals output from the high frequency oscillators 30-1 and 30-2, respectively.

FIG. 8 is a block diagram showing a specific configuration of the phase modulation high frequency oscillator array 10 according to the present embodiment. The phase modulation high-frequency oscillator array 10 includes a reference signal source 20, two high-frequency oscillators 30-1 and 30-2, a distributor 60, two attenuators 70-1 and 70-2, and two directional couplings. Devices 80-1, 80-2, two amplifiers 50-1, 50-2, and two filters 90-1, 90-2. Each of the high frequency oscillators 30-1 and 30-2 includes an MR element 31, a bias voltage application mechanism 32, a bias tee 33, and a magnetic field application mechanism. The amplifiers 50-1 and 50-2 amplify output signals output from the high frequency oscillators 30-1 and 30-2, respectively. Filters 90-1 and 90-2 selectively remove a reference signal mixed in a signal path for transmitting output signals output from high-frequency oscillators 30-1 and 30-2, respectively.

The high frequency oscillators 30-1 and 30-2 were set to synchronize at an oscillation frequency f = 6.129 GHz. As the reference signal source 20, an analog signal generator E8257D manufactured by Keysight Technology Inc. having an output power of −135 dBm to +21 dBm and a band of 250 kHz to 40 GHz was used. In this embodiment, a reference signal having an output power of 16 dB and a frequency f _REF = 2f = 12.2258 GHz was output from the reference signal source 20.

An FeB layer (thickness 2 nm) was used as the free layer 31A constituting the MR element 31. As the nonmagnetic layer 31B constituting the MR element 31, an MgO layer (thickness 1 nm) was used. As the fixed layer 31C constituting the MR element 31, a multilayer film of CoFeB layer (thickness 3 nm) / Ru layer (thickness 0.86 nm) / CoFe layer (thickness 2.5 nm) / PtMn layer (thickness 15 nm) is used. Using. The CoFeB layer of the fixed layer 31C is in contact with the MgO layer that is the nonmagnetic layer 31B. The MR element 31 was processed into a cylinder having a diameter of 300 nm by fine processing. The MR element 31 had a resistance value of 27Ω and a magnetoresistance ratio (MR ratio) of about 100% at room temperature.

As the bias voltage application mechanism 32, a precision source measure unit B2912A manufactured by Keysight Technology Co., Ltd. having a minimum power source resolution of 100 nV was used. As the bias tee 33, a bias tee 11612A manufactured by Keysight Technology Inc. having an insertion loss of about 0.8 dB and a band of 45 MHz to 26.5 GHz was used. An electromagnet was used as the magnetic field application mechanism 34.

The distributor 60 includes three

ports

61, 62, and 63. The reference signal input to the distributor 60 from the reference signal source 20 through the port 61 is divided into two equal parts and the

ports

62 and 63 are respectively divided. Output. The intensity of the reference signal output from the

ports

62 and 63 is the same. As the distributor 60, a wide-band resistor type distributor ZFRSC-183 + manufactured by Minicircuit Corporation having an insertion loss of about 7 dB and a band of DC 18 GHz was used.

The attenuators 70-1 and 70-2 have

ports

71 and 72, respectively. The degree of attenuation of the signal input from the port 71 and output from the port 72 is the same as the degree of attenuation of the signal input from the port 72 and output from the port 71. By providing the attenuators 70-1 and 70-2, the output signal from the high frequency oscillator 30-1 is restricted from being input to the amplifier 50-2 and the filter 90-2, and from the high frequency oscillator 30-2. The output signal can be restricted from being input to the amplifier 50-1 and the filter 90-1. As the attenuators 70-1 and 70-2, a broadband attenuator BW-K20-2W44 ++ manufactured by Minicircuit having an insertion loss of about 20 dB and a band of DC 40 GHz was used.

Each of the directional couplers 80-1 and 80-2 includes an input port 81, a coupling port 82, and an output port 83. A signal input from the input port 81 is output from the output port 83 with almost no attenuation, and a signal with a certain ratio (for example, about −14 dB) is also output from the coupling port 82. A signal input from the output port 83 is output from the input port 81 with almost no attenuation, and a signal with a certain ratio (for example, about −28 dB, that is, almost zero) is also output from the coupling port 82. By providing the directional couplers 80-1 and 80-2, the output signal from the high frequency oscillator 30-1 is restricted from being input to the amplifier 50-2 and the filter 90-2, and the high frequency oscillator 30-2 is provided. Can be restricted from being input to the amplifier 50-1 and the filter 90-1. As the directional couplers 80-1 and 80-2, a coaxial directional coupler 87301D manufactured by Keysight Technology was used. In this embodiment, since the intensity of the reference signal from the reference signal source 20 is about 8 dB larger than the intensity of the output signal from the MR element 31, in order to avoid saturation of the amplifiers 50-1 and 50-2, Directional couplers 80-1 and 80-2 were used to attenuate the intensity of the reference signal from the reference signal source 20. However, when there is no possibility that the widths 50-1 and 50-2 are saturated (for example, the maximum allowable value of the input voltage when the output signal of the MR element 31 is saturated with the amplifiers 50-1 and 50-2). The directional couplers 80-1 and 80-2 are not essential, and the directional couplers 80-1 and 80-2 are not necessary. A distributor or a simple electrical contact may be used.

As the amplifier 50-1, the MR element 31 has an oscillation frequency f _MR = 6.129 GHz, an amplification factor of about 27 dB, a noise figure of 2.4 dB, and a band of 100 MHz to 18 GHz. S + was used. As the amplifier 50-2, an amplifier ZVA-213- manufactured by Minicircuit Corporation having an amplification factor of about 26 dB, a noise figure of 3.0 dB, and a band of 800 MHz to 21 GHz in the vicinity of the oscillation frequency f _MR = 6.129 GHz of the MR element 31. S + was used.

The filters 90-1 and 90-2 are virtual devices that have no real device and are artificially inserted into the subsequent stage of the amplifiers 50-1 and 50-2 by signal analysis processing. In this embodiment, a Barterworth low-pass filter whose cutoff frequency is set to 8 GHz is used as the filters 90-1 and 90-2.

Next, a signal flow and its intensity in the phase modulation high frequency oscillator array 10 will be described. A reference signal (frequency f _REF = 2f = 12.2258 GHz) having a signal intensity of 9 dBm was output from the reference signal source 20. The reference signal input to the port 61 of the distributor 60 was equally divided into signals having a signal strength of 3 dBm and output from the

ports

62 and 63. The reference signal output from the port 62 of the distributor 60 includes the attenuator 70-1, the output port 83 of the directional coupler 80-1, the input port 81 of the directional coupler 80-1, and the high frequency oscillator 30-1. The signal intensity was attenuated to about −7 dBm through the bias tee 33 and then injected into the MR element 31 of the high frequency oscillator 30-1. Similarly, the reference signal output from the port 63 of the distributor 60 includes an attenuator 70-2, an output port 83 of the directional coupler 80-2, an input port 81 of the directional coupler 80-2, and a high frequency oscillator. The signal intensity was attenuated to about −7 dBm through the bias tee 33 of 30-2, and then injected into the MR element 31 of the high frequency oscillator 30-2. As a result, the two MR elements 31 of the high-frequency oscillators 30-1 and 30-2 are injection-locked and synchronized at the oscillation frequency f = 6.129 GHz. At this time, the reference signals passing from the output ports 83 of the directional couplers 80-1 and 80-2 through the coupling port 82 are attenuated to about -35 dBm, and then the amplifiers 50-1, 50- 2 amplified the signal intensity to −8 dBm, and the filters 90-1 and 90-2 attenuated the signal intensity to −46 dBm (ie, almost zero).

On the other hand, the intensity of the output signal (oscillation frequency f = 6.129 GHz) output from the MR element 31 of the high-frequency oscillator 30-1 was about −30 dBm. The output signal from the MR element 31 of the high-frequency oscillator 30-1 passes through the coupling port 82 from the input port 81 of the directional coupler 80-1, and the signal intensity is attenuated to -43 dBm. The signal intensity was amplified to -16 dBm, and the signal intensity was attenuated to -17 dBm by the low-pass filter 90-1. Similarly, the intensity of the output signal (oscillation frequency f = 6.129 GHz) output from the MR element 31 of the high frequency oscillator 30-2 was about −30 dBm. The output signal from the MR element 31 of the high frequency oscillator 30-2 passes through the coupling port 82 from the input port 81 of the directional coupler 80-2, and the signal intensity is attenuated to -43 dBm. The signal intensity was amplified to -16 dBm, and the signal intensity was attenuated to -17 dBm by the low-pass filter 90-2.

The output signal output from the MR element 31 of the high-frequency oscillator 30-2 is input to the input port 81 of the directional coupler 80-2, the output port 83 of the directional coupler 80-2, the attenuator 70-2, and the distribution. The signal strength is finally passed through the amplifier 60, the attenuator 70-1, the output port 83 of the directional coupler 80-1, the coupling port 82 of the directional coupler 80-1, the amplifier 50-1, and the filter 90-1. It was confirmed that the signal was attenuated to −46 dBm and superimposed on the output signal (signal intensity −17 Bm) output from the MR element 31 of the high frequency oscillator 30-1. However, the output signal of −46 dBm is lower by about three digits than the output signal of −17 dBm, and can be ignored in practice. Similarly, the output signal output from the MR element 31 of the high-frequency oscillator 30-1 is finally attenuated to -46 dBm, and the output signal output from the MR element 31 of the high-frequency oscillator 30-2 ( The signal strength is -17 Bm), but can be ignored in practice.

FIG. 9 is a graph showing the measurement results of the spectral intensity of the MR element 31 when the bias voltage applied to the MR element 31 of the high-frequency oscillator 30-1 is changed in the range of 235 to 255 mV. FIG. 10 is a graph showing the measurement results of the spectral intensity of the MR element 31 when the bias voltage applied to the MR element 31 of the high-frequency oscillator 30-2 is changed in the range of 200 to 220 mV. When measuring the spectral intensity, a magnetic field application mechanism 34 is used to apply a magnetic field of 3 kOe at an angle of 120 ° with respect to the fixed layer 31C of each MR element 31 of the high-frequency oscillators 30-1 and 30-2. did.

FIG. 11 is a graph showing the oscillation frequency f _MR1 of the MR element 31 when the bias voltage applied to the MR element 31 of the high frequency oscillator 30-1 is changed in the range of 235 to 255 mV, and is obtained from the measurement results of FIG. It is a thing. From the graph of FIG. 11, the oscillation frequency f _MR1 of the MR element 31 of the high-frequency oscillator 30-1 is in the range of about 6.10 GHz to about 6.18 GHz, centering on 6.16 GHz, in the bias voltage range of 235 to 255 mV. It can be confirmed that the change is almost monotonous. FIG. 12 is a graph showing the oscillation frequency f _MR2 of the MR element 31 when the bias voltage applied to the MR element 31 of the high-frequency oscillator 30-2 is changed in the range of 200 to 220 mV, which is obtained from the measurement results of FIG. It is a thing. From the graph of FIG. 12, the oscillation frequency f _MR2 of the MR element 31 of the high-frequency oscillator 30-2 is in the range of about 6.07 GHz to about 6.16 GHz centering on 6.12 GHz in the bias voltage range of 200 to 220 mV. It can be confirmed that the change is almost monotonous.

FIG. 13 is a graph showing the relationship between the oscillation frequency f _MR1 and the measurement output when the MR element 31 of the high frequency oscillator 30-1 is injection-locked at the oscillation frequency f = 6.129 GHz. FIG. 14 is a graph showing the relationship between the oscillation frequency f _MR2 and the measurement output when the MR element 31 of the high frequency oscillator 30-2 is injection-locked at the oscillation frequency f = 6.129 GHz. In this injection locking, a reference signal (frequency f _REF = 2f = 12.2258 GHz) was injected into each MR element 31 of the high frequency oscillators 30-1 and 30-2. From the measurement results shown in FIGS. 13 and 14, the peak width of the output signal output from each MR element 31 of the high-frequency oscillators 30-1 and 30-2 is as low as about 3 kHz, and the oscillation frequencies f _MR1 and f _MR2 are It can be seen that it can be stabilized. When the oscillation frequency was measured without injecting the reference signal into the MR element (that is, without synchronizing the MR element), the peak width of the output was about 3 MHz and was very high and unstable.

FIG. 15 shows a high frequency when the bias voltage applied to the MR element 31 of the high frequency oscillator 30-2 is changed around 210 mV while the bias voltage applied to the MR element 31 of the high frequency oscillator 30-1 is fixed at 250 mV. It is a graph which shows the result of having measured the oscillation frequency f _MR1 of the MR element 31 of the oscillator 30-1 and the oscillation frequency f _MR2 of the MR element 31 of the high frequency oscillator 30-2. Even if the bias voltage applied to the MR element 31 of the high-frequency oscillator 30-2 is changed, it can be confirmed that both the oscillation frequencies f _MR1 and f _MR2 remain unchanged at f = 6.129 GHz by injection locking. .

FIG. 16 shows a high frequency when the bias voltage applied to the MR element 31 of the high frequency oscillator 30-2 is changed around 210 mV while the bias voltage applied to the MR element 31 of the high frequency oscillator 30-1 is fixed at 250 mV. It is a graph which shows the result of having measured the phase difference between the output signal of the oscillator 30-1, and the output signal of the high frequency oscillator 30-2. The phase difference changed from 3π / 4 to 0 by changing the bias voltage applied to the MR element 31 of the high-frequency oscillator 30-2. From this measurement result, it is possible to control the phase difference by changing the bias voltage applied to the MR element 31 while the MR elements 31 of the high-frequency oscillators 30-1 and 30-2 are injection-locked, that is, the phase difference. It can be confirmed that modulation is possible.

In the present embodiment, the phase modulation by the phase modulation high-frequency oscillator array 10 when N = 2 has been described, but the phase modulation high-frequency oscillator array 10 performs phase modulation according to the same principle even when N = 3 or more. be able to. Depending on the application of the phase modulation high-frequency oscillator array 10, N is preferably an integer of 2 to 40.

Table 1 shows specific examples of phase modulation by the phase modulation high-frequency oscillator array 10.

Table 1 shows respective phase differences of output 2, output 3,..., Output N with respect to output 1. Here, “output 1”, “output 2”,..., “Output N” are respectively “output signal of high-frequency oscillator 30-1,” “output signal of high-frequency oscillator 30-2”,. 30-N output signal ". For example, in Example 1, the phase differences of output 2, output 3,..., Output N with respect to output 1 are all “0”. In Example 2, the phase differences of output 2, output 3, output 4, output 5,..., Output N with respect to output 1 are “0 to π”, “0”, “0 to π”, “0”, ... "0 to π". In example 3, the phase differences of output 2, output 3, output 4, output 5,..., Output N with respect to output 1 are “0”, “0 to π”, “0”, “0 to π”, ..., "0". Here, “0 to π” represents an arbitrary phase within the range of 0 to π. In example 4, the output 2, output 3, output 4, output 5,..., Output N with respect to the output 1 have phase differences of “0”, “π”, “π”, “π / 2”,. “Π / 3”. In Example 5, the phase differences of output 2, output 3, output 4, output 5,..., Output N with respect to output 1 are “π / 2”, “π / 3”, “π / 4”, “π / 5 ", ...," π / N ".

Each phase difference of output 2, output 3,..., Output N with respect to output 1 may be fixed to a specific phase difference, or may be changed over time. For example, the phase modulation high-frequency oscillator array 10 performs phase modulation as shown in Example 1 in a certain period, phase modulation as shown in Example 2 in a subsequent period, and further in Example 3 in a subsequent period. Phase modulation may be performed as shown.

The phase modulation high-frequency oscillator array 10 of this embodiment is useful for a phased array radar, a communication device, and the like. In particular, since the MR element 31 functions as a high-frequency oscillation element that outputs a high-frequency signal, a small and inexpensive phase-modulated high-frequency oscillator array 10 can be provided.

DESCRIPTION OF SYMBOLS 10 ... Phase modulation | alteration high frequency oscillator array 20 ... Reference signal source 30-1, 30-2, ..., 30-N ... High frequency oscillator 31 ... MR element 32 ... Bias voltage application mechanism 33 ... Bias tee 34 ... Magnetic field application mechanism 40-1 , 40-2, ..., 40-N ... removal means

Claims

A phase-modulated high-frequency oscillator array synchronized and oscillated at a frequency f,
a reference signal source that outputs a reference signal having a frequency that is n times or 1 / n times the frequency f when n is an integer of 2 or more;
A plurality of magnetoresistive elements, wherein the reference signal is injected into each of the plurality of magnetoresistive elements, so that the frequency of the output signal from each of the plurality of magnetoresistive elements matches the frequency f; A plurality of magnetoresistive elements; and
A plurality of bias voltage applying mechanisms for applying a bias voltage to each of the plurality of magnetoresistive elements, wherein bias voltages applied to at least two magnetoresistive elements arbitrarily selected from the plurality of magnetoresistive elements are provided. A plurality of bias voltage applying mechanisms that change a phase difference between output signals from the at least two magnetoresistive elements by changing;
A plurality of removing means for removing the reference signal mixed in a signal path for transmitting an output signal from each of the plurality of magnetoresistive elements;
A phase modulation high frequency oscillator array comprising:
The phase-modulated high-frequency oscillator array according to claim 1,
A plurality of magnetic field applying mechanisms for applying a magnetic field to each of the plurality of magnetoresistive elements, wherein a magnetic field applied to at least two magnetoresistive elements arbitrarily selected from the plurality of magnetoresistive elements is changed; A phase-modulated high-frequency oscillator array further comprising a plurality of magnetic field applying mechanisms for changing a phase difference between output signals from the at least two magnetoresistive elements.
The phase-modulated high-frequency oscillator array according to claim 1 or 2,
The phase modulation high-frequency oscillator array, wherein n is 2.
The phase-modulated high-frequency oscillator array according to any one of claims 1 to 3,
Each of the plurality of removing means is a phase-modulated high-frequency oscillator array, which is a filter having a frequency characteristic in which the frequency band of the reference signal is an attenuation band.
The phase-modulated high-frequency oscillator array according to any one of claims 1 to 3,
Each of the plurality of removing means is an antenna that has a frequency characteristic in which the frequency band of the reference signal is an attenuation region, and is an antenna that outputs the output signal from the magnetoresistive element as a radio wave. .