JP2018165685A - Phase characteristic calibration system and phase characteristic calibration method for millimeter wave band signal measurement circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable calibration of a phase characteristic of a millimeter wave band signal measurement circuit to be accurately performed.SOLUTION: A calibration signal generating unit 40 generates a calibration signal Er which is obtained by frequency-synchronizing an electric harmonic signal of a frequency at an integer multiple of a repetition frequency of short pulse light P0 and combining a plurality of continuous waves of frequencies different within a frequency domain of a millimeter wave band. A calibration signal waveform acquisition unit 60 samples the calibration signal Er while changing a delay time of sampling light Ps by the sampling light to acquire a waveform of the calibration signal in a time domain. Phase information calculation means 70 calculates a phase of the plurality of continuous waves on the basis of the waveform of the calibration signal obtained for each change of the frequency of the continuous waves to determine a phase characteristic Δφ(f) of a calibration target within the frequency domain. A calibration processing unit 90 determines a phase characteristic ΔΦ(f) of a millimeter wave band signal measurement circuit 1 itself from a difference between the determined phase characteristic Δφ(f) and a phase characteristic ΔΦ(f)' measured at the millimeter wave band signal measurement circuit 1 with respect to the calibration signal Er.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for enabling calibration of phase characteristics of a millimeter wave band signal measurement circuit used in a millimeter wave band or a submillimeter wave band with high accuracy.

  In recent years, mobile data traffic has increased dramatically and is expected to increase further in the future, and securing communication capacity is an urgent issue.

  In order to solve these problems and meet the increasing demand for mobile data traffic, the use of millimeter wave bands and submillimeter wave bands (including terahertz bands) capable of realizing transmission speeds of several tens of Gbps is strongly demanded. . In the following description, the millimeter wave band and the submillimeter wave band are collectively referred to as the millimeter wave band.

  Further, although it is expected that a modulation scheme having high frequency efficiency such as a QPSK modulation signal is used in communication in the millimeter wave band, a measurement method corresponding to phase modulation such as QPSK in these frequency bands has not been studied yet. The measurement environment for evaluating is not realized.

  In order to measure the modulation signal in the millimeter wave band, a down converter is required as a measurement circuit for converting the signal under measurement in these frequency bands into a frequency band capable of digital processing (A / D conversion processing). However, the measurement value obtained from the output of the down converter does not include the true characteristics of the signal under measurement and the characteristics of the measurement system including the down converter, and does not compensate for the frequency characteristics of the measurement system. As long as the correct measurement results cannot be obtained.

  In particular, in the case of a broadband modulation signal exceeding 10 Gbs, it is necessary to compensate not only the conventional amplitude characteristic but also the phase characteristic.

  As a method for calibrating the phase characteristics of circuits such as down converters, a calibration signal whose phase characteristics (frequency characteristics related to the phase) are known in advance is input to the calibration target circuit, and the calibration signal is output from the output of the calibration target circuit. A method is known in which the phase characteristic of the circuit to be calibrated is acquired by removing the influence of the phase characteristic.

  Among these methods, as a method for calibrating the phase characteristics of a measurement circuit including a down converter in a high frequency band such as a millimeter wave, an electro-optic sampling method, which is a kind of time-domain spectroscopy measurement, is used. Measures the time domain waveform of the electrical pulse signal output from the excited O / E conversion device (for example, a photoelectric conversion element such as a photodiode or photoconductor), and converts it to the frequency domain by signal processing such as Fourier transform However, there are known ones that obtain phase characteristics.

  FIG. 8 shows a schematic configuration of the calibration system 10 using the electro-optic sampling method. The calibration system 10 includes a short pulse light source 11, a branching unit 12, an O / E conversion device 13, an electro-optic crystal 14, an optical variable delay device 15, and a polarization measuring device 16.

  As shown in FIG. 9A, the calibration system 10 uses a branching means 12 to pump light from a short pulse light (laser light) P0 emitted from a short pulse light source 11 at a predetermined cycle Ts (repetition frequency fs). The pump light P1 is divided into P1 and probe light P2, and the O / E conversion device 13 is excited by inputting the pump light P1 to the O / E conversion device 13.

  The O / E conversion device 13 excited by the pump light P1 generates an electric pulse signal E1 as shown in FIG. 9B. The electric pulse signal E1 is applied to a coplanar waveguide, a microstrip line, or the like formed in the electro-optic crystal 14 via a probe head or the like (not shown), and generates a signal electric field in the electro-optic crystal 14.

  On the other hand, the probe light P2 branched by the branching unit 12 enters the optical variable delay device 15, and the probe light P2 'delayed by the variable delay device 15 enters the electro-optical crystal 14 and is transmitted through the electro-optical crystal 14. To do. Here, the polarization angle θ of the probe light P2 ″ transmitted through the electro-optic crystal 14 changes in proportion to the electric field strength in the electro-optic crystal 14 at the timing when the probe light P2 ′ is incident. The polarization angle θ of the probe light P <b> 2 ″ is measured by the polarimeter 16.

  The electric field strength at the incident timing of the probe light P2 ′ can be specified from the polarization angle θ measured by the polarimeter 16. In other words, the electric field strength of the electric pulse signal E1 is sampled at the incident timing of the probe light P2 ′, and as shown in FIG. 9C, the delay time of the optical variable delay device 15 is obtained. Are sequentially changed to Δt1, Δt2,... And the incident timing of the probe light P2 ′ is shifted, and the polarization angles θ1, θ2,. , And the electric field strengths e1, e2,... Of the electric pulse signal E1 when the probe light P2 ′ is incident are obtained from the polarization angle, the time domain waveform of the electric pulse signal E1 can be obtained.

  Then, the time domain waveform of the electric pulse signal E1 is subjected to signal processing such as Fourier transform, so that it can be converted into the frequency domain and the amplitude / phase characteristics can be calculated. From the calculated amplitude / phase characteristics, the O / E The output characteristics of the conversion device 13 can be obtained (note that means for performing these operations are omitted).

  At the time of calibration, the output of the O / E conversion device 13 is connected to the millimeter waveband signal measurement circuit 1 that is the calibration target. In this example, the millimeter-wave band signal measurement circuit 1 to be calibrated includes a down-converter 2 that heterodyne-converts an input signal to the mixer 2a to a lower frequency band by a local signal output from the local signal generator 2b, and the down-converter 2 The signal processing unit (digitizer) 3 executes A / D conversion processing and various arithmetic processing necessary for measurement on the output signal, and the substantial calibration target is the down converter 2 of the analog circuit configuration. Amplitude characteristics and phase characteristics.

  In the case of this configuration, when the output of the O / E conversion device 13 is given to the millimeter waveband signal measurement circuit 1, the signal processing unit 3 performs a Fourier transform process or the like to convert it to the frequency domain, By obtaining the phase characteristic and comparing it with the amplitude / phase characteristic of the output signal E1 of the O / E conversion device 13 obtained in advance, the frequency characteristic relating to the amplitude / phase of the down converter 2 of the analog circuit configuration can be grasped, and its frequency By using the characteristics to compensate the measurement result obtained when the signal under measurement is actually input to the millimeter waveband signal measurement circuit 1, an accurate measurement result for the signal under measurement can be obtained.

  A basic configuration of a calibration system using the above-described electro-optic sampling method is disclosed in Non-Patent Document 1, for example.

D.F.Williams, P.D.Hales, T.S.Clement, and J.M.Morgan, "Calibrating electro-optical sampling systems," in IEEE MTT-S Int. Microwave Symp.Dig., Vol.3, pp.1527-1530, May 2001

  As described above, in the calibration system using the electro-optic sampling method, the time-domain waveform of the electric pulse signal output from the O / E conversion device is obtained, converted into the frequency domain, and the amplitude and phase characteristics are obtained. However, since the time domain waveform of the electric pulse signal is converted to the frequency domain, the signal level per wave in the frequency domain is lowered, and there is a problem that high-precision measurement is difficult.

  For example, the electric pulse signal in the millimeter wave band has a band of several hundreds GHz, but if the bandwidth of the frequency band to be calibrated is 100 GHz, the signal level per wave in this frequency region is the same as that of the original electric pulse signal. The level is reduced by 110 dB to a level close to the noise level, and the measurement accuracy is significantly reduced. As for the frequency characteristics of the amplitude, an accurate characteristic can be obtained by power measurement or the like even in the above-described millimeter wave band region, and therefore a system capable of performing phase characteristic calibration with high accuracy is required.

  An object of the present invention is to solve this problem and provide a phase characteristic calibration system and a phase characteristic calibration method for a millimeter wave band signal measurement circuit that can calibrate the phase characteristic of the millimeter wave band signal measurement circuit with high accuracy. Yes.

In order to achieve the above object, a phase characteristic calibration system for a millimeter waveband signal measurement circuit according to claim 1 of the present invention comprises:
A millimeter wave band signal measuring circuit (1) including a down converter (2) that receives a signal under measurement in a predetermined frequency region of a millimeter wave band or a submillimeter wave band and converts the signal under measurement into a frequency band that can be digitally processed. A phase characteristic calibration system to be calibrated,
A short pulse light source (21) for outputting short pulse light at a predetermined repetition frequency;
Branching means (22) for branching the short pulse light into a first short pulse light and a second short pulse light;
A calibration signal generator (40) for generating a calibration signal in which a plurality of continuous waves having different frequencies within the predetermined frequency region are combined;
Receiving the first short pulse light, extracting an electrical harmonic signal that is an integral multiple of the predetermined repetition frequency, and using the extracted harmonic signal as a reference, a calibration signal output by the calibration signal generator A synchronization processing unit (30) for performing synchronization processing of
An optical variable delay device (50) that receives the second short pulse light, gives a predetermined delay time to the second short pulse light, and outputs it as sampling light;
The calibration signal and the sampling light are received, the calibration signal is sampled at a timing synchronized with the sampling light while changing the delay time of the optical variable delay device, and the time domain waveform of the calibration signal is obtained. A calibration signal waveform acquisition unit (60) to acquire;
Continuous wave frequency changing means (80) for sequentially changing the frequencies of a plurality of continuous waves included in the calibration signal generated by the calibration signal generating unit within the predetermined frequency region;
From the time domain waveform of each calibration signal obtained by the calibration signal waveform acquisition unit for each frequency change by the continuous wave frequency changing means, obtain the phase of a plurality of continuous waves included in each calibration signal, Phase information calculation means (70) for obtaining the characteristic of the phase difference between the frequencies of the continuous wave from the phase over the entire predetermined frequency region,
The measurement result when the calibration signal is supplied to the calibration target millimeter-wave band signal measurement circuit is compared with the phase difference characteristic obtained in advance for the calibration signal, and the millimeter-wave band signal measurement is performed. The phase characteristic of the circuit including the down converter is calibrated.

A method for calibrating the phase characteristics of the millimeter waveband signal measuring circuit according to claim 2 of the present invention includes:
A millimeter wave band signal measuring circuit (1) including a down converter (2) that receives a signal under measurement in a predetermined frequency region of a millimeter wave band or a submillimeter wave band and converts the signal under measurement into a frequency band that can be digitally processed. A phase characteristic calibration method to be calibrated,
The short pulse light output at a predetermined repetition frequency is branched into a first short pulse light and a second short pulse light, and based on the first short pulse light, an electrical harmonic that is an integral multiple of the predetermined repetition frequency. Extracting a wave signal;
Generating a calibration signal in which a plurality of continuous waves having different frequencies within the predetermined frequency region are synthesized in a frequency-synchronized state with reference to the extracted harmonic signal;
Providing the second short pulse light with a predetermined delay time and outputting it as sampling light;
Sampling the calibration signal at a timing synchronized with the sampling light while changing the delay time of the sampling light, obtaining a time domain waveform of the calibration signal;
The frequency of a plurality of continuous waves included in the calibration signal is sequentially changed within the predetermined frequency domain, and is included in each calibration signal from the time domain waveform of each calibration signal obtained for each frequency change. Obtaining a phase of a plurality of continuous waves, and obtaining a characteristic of a phase difference between the frequencies of the continuous waves from the phase over the entire predetermined frequency region;
The measurement result when the calibration signal is supplied to the calibration target millimeter wave measurement circuit is compared with the characteristics of the phase difference obtained in advance for the calibration signal, and the millimeter wave band signal measurement circuit Calibrating phase characteristics including the down converter.

  Thus, in the present invention, based on the short pulse light, an electrical harmonic signal having a frequency that is an integral multiple of the repetition frequency is extracted, the frequency is synchronized with the harmonic signal as a reference, and the frequency to be calibrated Generate a calibration signal in which multiple continuous waves of different frequencies in the region are combined, sample the calibration signal at the timing synchronized with the sampling light while changing the delay time of the sampling light obtained from the short pulse light, The process of obtaining the time domain waveform of the calibration signal is performed while sequentially changing the frequency of the continuous wave included in the calibration signal, and the calibration signal is included in the calibration signal from the obtained waveform of each calibration signal. The phase of a plurality of continuous waves is calculated, and the characteristics of the phase difference in the frequency domain to be calibrated are obtained.

  In other words, the calibration signal is a signal that is frequency-synchronized with an integral multiple of the repetition frequency of the short pulse light, and a combination of multiple continuous waves with different frequencies in the millimeter waveband frequency range. Since the phase characteristics of the entire frequency region of the millimeter wave band to be calibrated are calculated, the phase information can be obtained with a much higher level of signal than the conventional method using a pulse signal as the calibration signal. The calibration accuracy can be remarkably increased.

Overall configuration diagram of an embodiment of the present invention The figure which shows the structural example of the principal part of embodiment of this invention. Operation explanatory diagram of the embodiment of the present invention The figure which shows another structural example of the principal part of embodiment of this invention. Operation explanatory diagram of the embodiment of the present invention Operation explanatory diagram of the embodiment of the present invention Operation explanatory diagram of the embodiment of the present invention Schematic configuration diagram of a conventional calibration system Operation explanation diagram of conventional calibration system

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 shows an overall configuration of a phase characteristic calibration system (hereinafter simply referred to as a calibration system) 20 of a millimeter waveband signal measurement circuit to which the present invention is applied.

  As described above, the calibration system 20 includes the mixer 2a and the local signal generator 2b, receives a signal under measurement in a predetermined frequency region in the millimeter wave band, and converts it into a frequency band that can be digitally processed. And a millimeter-wave band signal measuring circuit 1 including a signal processing unit 3 that performs A / D conversion processing and digital arithmetic processing on the output signal of the down-converter 2, the short pulse light source 21, branching means 22, a synchronization processing unit 30, a calibration signal generation unit 40, an optical variable delay device 50, a calibration signal waveform acquisition unit 60, a phase information calculation unit 70, a continuous wave frequency change unit 80, and a calibration processing unit 90.

  The short pulse light source 21 performs pulse modulation on a semiconductor laser, for example, and emits short pulse light P0 at a predetermined repetition period Ts (= 1 / fs: fs is, for example, 100 MHz). The branching means 22 receives the short pulse light P0 and branches it into the first short pulse light P1 and the second short pulse light P2.

  The synchronization processing unit 30 that has received the first short pulse light P1 includes a plurality of calibration signals Er that are output from the calibration signal generation unit 40 (here, although described as three, two or four or more may be used). The processing for synchronizing the frequency of the continuous wave with the harmonic of the repetition frequency of the short pulse light P0 is performed, and its configuration corresponds to the configuration of the calibration signal generation unit 40.

  In this embodiment, the calibration signal generation unit 40 generates and outputs a combined signal SUM (if) in which a plurality of continuous waves having different frequencies are combined in an intermediate frequency region lower than the predetermined frequency region. The local signal generator 42 for outputting the local signal LOC, the synthesized signal SUM (if) and the local signal LOC are received, and the synthesized signal SUM (if) in the intermediate frequency domain is used for calibration in the predetermined frequency domain of the millimeter wave band. The up-converter configuration (sum heterodyne) includes a mixer 43 that converts the signal Er into an amplifier 44 and an amplifier 44 that amplifies the output of the mixer 43.

  The synthesized signal generator 41 can be constituted by an arbitrary waveform generator that performs an internal signal synthesizing process in synchronization with an externally input clock signal. The local signal generator 42 needs to generate a local signal LOC having a frequency close to the millimeter wave band (several hundred GHz) in order to convert the frequency of the synthesized signal SUM (if) into the millimeter wave band. For example, a signal generator 42a in the 10 to several tens GHz band capable of synchronous control of the frequency by a reference signal (reference signal) REF (generally 10 MHz or 100 MHz) input from the outside, and its signal generation The output signal SGout of the multiplier 42b is multiplied by N (for example, N = 18), and can be constituted by a multiplier 42b that converts it into a local signal LOC having a frequency close to the millimeter wave band.

  As a specific numerical example, the synthesized signal generator 41 receives a 2.6 GHz clock signal CLK and sets, for example, a start frequency f0 = 24 GHz and a frequency interval Δf = 100 MHz, and f1 = f0, f2 = f1 + Δf, f3. A combined signal SUM (if) in which three continuous waves of = f1 + 2Δf are combined is generated. The continuous wave frequencies f1 to f3 are sequentially shifted to higher ones by, for example, a 200 MHz step change of the start frequency f0 under the control of the continuous wave frequency changing means 80 described later. By changing the start frequency f0 in 200 MHz steps with respect to the interval Δf = 100 MHz, the higher frequency f3 of a certain set of three frequencies f1 to f3 and the next set of three frequencies f1 ′ to f3 ′ The continuity of the phase difference between the frequencies of the continuous wave included in the calibration signal Er is maintained by overlapping the lower frequency f1 ′.

  The signal generator 42a of the local signal generator 42 outputs a signal SGout in the range of 15 ± 2.5 GHz, for example, at 100 MHz intervals, and the multiplier 42b multiplies the output signal SGout of the signal generator 42a by 18 times. This is output as a local signal LOC.

  Therefore, when the frequency of the output signal SGout of the signal generator 42a is changed from 12.5 GHz to 17.5 GHz in a 100 MHz step, the frequency of the local signal LOC is changed in a 1.8 GHz step between 225 GHz and 315 GHz. Become. In order to fill this 1.8 GHz band at 100 MHz intervals, as described above, the frequencies f1 to f3 of the three continuous waves included in the synthesized signal SUM (if) are changed eight times in 200 MHz steps from the initial value. . As a result, the frequency of the continuous wave included in the calibration signal Er is 100 MHz step within the range of (225 + 24) GHz to (315 + 24 + 1.7) GHz while allowing overlap between the previous set f3 and the next set f1. Will be buried.

  For the calibration signal generation unit 40 having the above configuration, the synchronization processing unit 30 receives the first short pulse light P1, extracts a harmonic signal that is an integral multiple of the repetition frequency of the first short pulse light P1, and extracts the harmonic signal. The clock signal CLK having a predetermined frequency (for example, 2.6 GHz) necessary for synchronizing the frequencies of a plurality of continuous waves included in the synthesized signal output from the synthesized signal generator (arbitrary waveform generator) 41 on the basis of the generated harmonic signal. A reference signal REF for synchronously controlling the frequency of the output signal SGout of the signal generator 42a of the local signal generator 42, and a plurality of calibration signals Er included in the calibration signal generator 40 output by the calibration signal generator 40. Synchronization processing for the continuous wave is performed so that the waveform of the calibration signal does not shift with respect to the first short pulse light P1.

  FIG. 2 shows a configuration example of the synchronization processing unit 30. The first short pulse light P1 is received by a photoelectric conversion element 31 such as a photodiode, and the first short pulse light P1 is received as shown in FIG. An electric pulse signal E1 synchronized with the short pulse light P1 is output and input to two band pass filters (hereinafter referred to as BPF) 32 and 33.

  The BPF 32 uses a harmonic signal E2 of 2.6 GHz as a reference signal for frequency-synchronizing the clock signal CLK from the harmonics included in the electric pulse signal E1 shown in FIG. 3B, as shown in FIG. Extract as follows. Further, the BPF 33 generates a harmonic signal E3 in the vicinity of 15 GHz, which is a reference for synchronizing the frequency of the output signal SGout of the signal generator 42a of the local signal generator 42 among the harmonics included in the electric pulse signal E1. Extract as shown in (d).

  Here, since the frequency of the output signal SGout of the signal generator 42a of the local signal generator 42 is changed, for example, in 100 MHz steps, the frequency of the harmonic signal E3 extracted from the electric pulse signal E1 is also extracted in 100 MHz steps. There is a need. Therefore, as the BPF 33, for example, a variable passing center frequency type such as YTF (YIG frequency variable filter) is used.

  The harmonic signal E2 having a frequency of 2.6 GHz extracted by the BPF 32 is input to the synchronization circuit 34 as a reference signal for performing PLL synchronization processing. The synchronization circuit 34 includes an oscillator 34a, a phase comparator 34b, and a loop filter 34c. The phase comparator 34b performs phase comparison between the harmonic signal E2 and the 2.6 GHz clock signal CLK output from the oscillator 34a. A control signal corresponding to the difference between the phase difference and a predetermined phase reference value (for example, 0 or π / 2) is extracted by the loop filter 34c and applied to the oscillator 34a so that the phase difference matches the phase reference value. The oscillator 34a is controlled.

  By the action of the synchronizing circuit 34, the frequency of the clock signal CLK output from the oscillator 34a is maintained in a state synchronized with the repetition frequency (100 MHz) of the electric pulse signal E1. In the 2.6 GHz frequency band, a general-purpose VCO (including a YIG oscillator and the like, which is a voltage-controlled oscillator) can be used as the oscillator 34a.

  On the other hand, the harmonic signal E3 extracted by the BPF 33 is also input to the synchronization circuit 35 as a reference signal for performing PLL synchronization processing. The synchronization circuit 35 includes an oscillator 35a, a phase comparator 35b, and a loop filter 35c. The phase comparator 35b compares the phase of the harmonic signal E3 with the output signal SGout of the signal generator 42a of the calibration signal generator 40. The control signal corresponding to the difference between the phase difference and a predetermined phase reference value is extracted by the loop filter 35c and applied to the oscillator 35a, and the reference signal REF corresponding to the phase difference is generated to generate the signal generator 42a. The oscillator 35a is controlled so that the phase difference matches the phase reference value. As a result, the frequency of the output signal SGout of the signal generator 42a is synchronously controlled with the harmonic signal E2. As the oscillator 35a, a variable frequency crystal oscillator having a very high frequency stability at 10 MHz or 100 MHz can be used.

  Due to the action of the synchronizing circuit 35, the output signal SGout of the signal generator 42a frequency-synchronized with the harmonic signal E3 is multiplied by, for example, 18 by the multiplier 42b to become the local signal LOC and input to the mixer 43. The frequency change of the output signal SGout of the signal generator 42a (for example, the frequency change of 100 MHz step) is performed by the frequency setting signal from the continuous wave frequency changing means 80.

  Therefore, the calibration signal generator 40 generates a calibration signal Er in which a plurality of continuous waves synchronized with a harmonic signal having a frequency that is an integral multiple of the repetition frequency (100 MHz) of the electric pulse signal E1 is synthesized. As a result, the waveform of the calibration signal Er is synchronized with the short pulse light P1, and the waveform sampling by the sampling light Ps described later can be performed stably.

  It should be noted that the calibration signal Er generated as described above is a horn so that the signal can be supplied to the millimeter waveband signal measuring circuit 1 to be calibrated and an electromagnetic wave detector (described later) necessary for calibration in an equivalent environment. It shall be radiated | emitted via the antenna 45. FIG.

  Here, the phase fluctuation of the calibration signal Er is proportional to the frequency ratio of the phase fluctuation of the signal serving as a reference for generating the calibration signal Er. For example, when the phase fluctuation is within 10 degrees at 300 GHz, if the reference signal is 100 MHz, the phase fluctuation is 1/3000, that is, 0.0033 degrees or less, and when adjusting with a VCO or the like, the voltage becomes minute. Although it is difficult to realize, the signal that becomes the reference for generating the calibration signal Er as in the present embodiment can be easily realized by using a harmonic signal that is an integral multiple of the repetition frequency of the short pulse light. Yes.

  The configurations of the synchronization processing unit 30 and the calibration signal generation unit 40 described above are merely examples, and various other configuration examples are conceivable. For example, in the synchronization processing unit 30, two BPFs 32 and 33 are used to extract a harmonic signal that is an integral multiple of the repetition frequency of the first short pulse light P1, but the clock signal CLK and the output signal SGout are When it can be set with an accuracy less than the repetition frequency (100 MHz), it is possible to extract a specific harmonic signal by using the detection action of the phase comparators 34b and 35b of the synchronization circuits 34 and 35. In this case, the BPF 32, 33 can be omitted. Of course, the frequency of the continuous wave included in the calibration signal and the variable method thereof are not limited to the above example.

  On the other hand, the second short pulse light P2 branched by the branching means 22 is input to the optical variable delay device 50, given a predetermined delay time Δt, and output as sampling light Ps. The optical variable delay device 50 can be configured, for example, by using a mechanism that varies the delay time Δt by moving the mirror or the like in part of the length of the optical path formed using a mirror.

  The sampling light Ps output from the optical variable delay device 50 is incident on the calibration signal waveform acquisition unit 60. The calibration signal waveform acquisition unit 60 receives the calibration signal Er output from the calibration signal generation unit 40 while changing the delay time Δt of the optical variable delay device 50, and samples the received calibration signal Er ′. Sampling is performed at a timing synchronized with the light Ps, and a time-domain waveform of the calibration signal Er ′ is acquired. This sampling method is basically the same as that of the conventional system described above, and various conventionally proposed configurations can be adopted.

  Note that the characteristics of the calibration signal Er output from the calibration signal generator 40 and the characteristics of the calibration signal Er ′ received by the calibration signal waveform acquisition unit 60 are the same as those obtained from the calibration signal generator 40. Although not matched due to the influence of the measurement system such as the propagation path of the electromagnetic wave to the waveform acquisition unit 60, when the calibration signal Er is given to the millimeter waveband signal measurement circuit 1 to be calibrated, from the calibration signal generation unit 40 If the propagation path of the electromagnetic wave reaching the millimeter waveband signal measurement circuit 1 to be calibrated is matched with the measurement system, the influence can be canceled out, so here, explanation will be made assuming that Er = Er ′.

  The calibration signal waveform acquisition unit 60 receives the calibration signal Er with an electromagnetic wave detector. As the electromagnetic wave detector, for example, as shown in FIG. 1, an electro-optic crystal 61 (E / 0 crystal) is used. be able to. As described above, the electro-optic crystal 61 has an action of changing the polarization angle of light traveling along the optical path in proportion to the electric field intensity when an electric field is generated in the predetermined optical path. In this case, the sampling light Ps output from the optical variable delay device 50 passes through a polarization beam splitter (hereinafter referred to as PBS) 62 and is incident on the electro-optic crystal 61, is reflected by the reflection film on the crystal end face, and is reflected on the PBS 62. It is configured to return. Here, if an electric field is generated in the electro-optic crystal 61 by the calibration signal Er, the polarization angle θ of the sampling light Ps ′ returned from the electro-optic crystal 61 changes in proportion to the electric field strength. The light component whose polarization direction has changed is separated from the sampling light Ps ′ returned to, and is incident on the polarization measuring device 63, and its polarization angle θ is detected.

  Then, the waveform acquisition means 64 obtains the electric field strength e generated in the electro-optic crystal 61 from the polarization angle θ measured by the polarization measuring device 63 by the calibration signal Er, and uses this to determine the delay of the optical variable delay device 50. A calibration signal waveform is obtained by continuously performing the process while changing the time Δt.

  In addition to the above configuration example, the calibration signal waveform acquisition unit 60 can also use a dipole photoconductive antenna 61 ′ as an electromagnetic wave detector, as shown in FIG. In the photoconductive antenna 61 ', parallel transmission paths 66 and 67 are formed on the surface of a semiconductor substrate 65 such as GaAs, and projecting portions (dipole antenna elements) 66a and 67a are formed so as to approach each other from an intermediate portion thereof. When an electric field is generated in the photoconductive gap G, which is a gap between the tips of the protrusions 66a and 67a, a current proportional to the electric field strength when light enters the photoconductive gap G is parallel transmission lines 66 and 67. It has the effect | action of flowing into.

  As shown in FIG. 4, the current I flowing through the parallel transmission paths 66 and 67 by irradiating the photoconductive gap G of the photoconductive antenna 61 'with the sampling light Ps while receiving the calibration signal Er is obtained. The current is measured by the current measuring device 68 and given to the waveform acquisition means 64 ′, the electric field of the calibration signal Er at the timing when the sampling light Ps is incident is obtained from the current I, and the waveform information is acquired. As a means for realizing the photoconductive antenna, various conventionally proposed configurations can be adopted in addition to the dipole type.

  The phase information calculation means 70 calculates phases φ (fa) to φ (fc) for a plurality of continuous waves included in the calibration signal from the calibration signal waveform acquired by the calibration signal waveform acquisition unit 60. Based on this, the characteristics of the phase difference between the frequencies are obtained. The calculation processing of the phases φ (fa) to φ (fc) is to calculate phases for the frequencies of the plurality of continuous waves by Fourier transform processing for the calibration signal waveform in the time domain, as described above. Since the Fourier transform processing is performed on a signal waveform obtained by synthesizing a plurality of continuous waves having different frequencies, the phases φ (fa) to φ (fc) can be accurately obtained.

  Further, the phase difference calculation process is performed, for example, assuming that a set of phases obtained from the waveform of one calibration signal is φ (fa1) to φ (fc1), a phase difference Δφ (fa1) at frequencies fa1 and fb1, Δφ (fb1) is obtained by the following calculation.

Δφ (fa1) = φ (fb1) −φ (fa1)
Δφ (fb1) = φ (fc1) −φ (fb1)

  If the phase of the next set is φ (fa2) to φ (fc2), since fa2 = fc1, the phase differences Δφ (fa2) and Δφ (fb2) at frequencies fa2 (= fc1) and fb2 are Calculate by

Δφ (fa2) = Δφ (fc1) = φ (fb2) −φ (fa2)
Δφ (fb2) = φ (fc2) −φ (fb2)

  Thereafter, by performing the same processing, it is possible to obtain the characteristics of the phase difference that maintains the continuity in frequency over the entire calibration target region.

  In the following description, the calculation process of the phase and the phase difference is described every time the waveform of the calibration signal Er is acquired. However, all calibrations covering the entire predetermined frequency region of the millimeter wave band are described. A method of performing the calculation processing of the phase and the phase difference after the waveform of the calibration signal is acquired, or obtaining the phase of each continuous wave every time the waveform of the calibration signal Er is acquired, After all the phases have been obtained for all the calibration signals covering the entire predetermined frequency region, the phase difference calculation process for the entire region may be performed collectively.

  As described above, the continuous wave frequency changing unit 80 sequentially sets the frequencies fa to fc of the plurality of continuous waves included in the calibration signal Er generated by the calibration signal generation unit 40 within a predetermined frequency region of the millimeter wave band. The phase information calculation means 70 calculates the phase difference characteristic Δφ (f) of the phase for each frequency of the continuous wave and the phase difference of the entire predetermined frequency region. As described above, this frequency change is performed by switching the frequency setting information of the combined signal generator 41 and the signal generator 42a of the calibration signal generator 40.

Next, an example of the frequency change process and the phase calculation process will be described.
First, in the state where the frequency of the output signal SGout of the signal generator 42a is set to 12.5 GHz and the passing center frequency of the BPF 33 of the synchronization processing unit 30 is set to 12.5 GHz accordingly, the synthesized signal generator 41 The start frequency f0 of the frequencies f1 to f3 of the continuous wave to be synthesized is set to 24 GHz.

  In this state, as shown in FIG. 5A, the calibration signal Er1 obtained by synthesizing continuous waves of fa1 = (225 + 24) GHz, fb1 = (225 + 24 + 0.1) GHz, and fc1 = (225 + 24 + 0.2) GHz is obtained. The waveform of the calibration signal Er is obtained by changing the delay time Δt of the sampling light Ps generated and output by ΔT, and each continuous wave is obtained by arithmetic processing on the waveform, for example, as shown in FIG. The phases φ (fa1) to φ (fc1) for the frequencies fa1 to fc1 are calculated. Further, as shown in FIG. 5C, phase differences Δφ (fa1) and Δφ (fb1) between frequencies are obtained from this phase.

  Then, the start frequency f0 of the continuous wave synthesized by the synthesized signal generator 41 is set higher by 200 MHz, and the frequencies are fa2 = (225 + 24 + 0.2) GHz and fb2 = (225 + 24 + 0) as shown in FIG. .3) Calibration signal Er2 in which continuous waves of GHz, fc2 = (225 + 24 + 0.4) GHz are synthesized is generated, and the waveform is acquired. As shown in FIG. Phases φ (fa2) to φ (fc2) with respect to the frequency are calculated, and the phase difference between the frequencies Δφ (fa2) = Δφ (fc1), Δφ (fb2) as shown in FIG. Is required.

  Similarly, the continuous wave that fills the range of (225 + 24) to (225 + 24 + 1.7) GHz in 100 MHz steps by increasing the start frequency f0 of the continuous wave synthesized by the synthesized signal generator 41 8 times in 200 MHz steps. And the phase change between these frequencies, the frequency change of the synthesized signal generator 41, the frequency change of the output signal SGout of the signal generator 42a and the frequency change of 100 MHz steps of the passing center frequency of the BPF 33 are obtained. By combining, as shown in FIG. 7, phase characteristics Δφ (f) at a frequency interval of 100 MHz can be acquired in a millimeter wave band range of approximately 245 to 340 GHz. FIG. 7 shows the final M sets of frequencies of three continuous waves represented by faM to fcM.

  The phase characteristic Δφ (f) for the calibration signal Er obtained in this way is given to the calibration processing unit 90.

  The calibration processing unit 90 performs processing necessary for calibration of the phase characteristics of the millimeter waveband signal measurement circuit 1, and has an environment equivalent to the electromagnetic wave detector of the waveform acquisition unit 60 with respect to the calibration signal generation unit 40. The calibration signal Er is transmitted to the millimeter waveband signal measurement circuit 1 arranged at (1) (specifically, replaced with an electromagnetic wave detector), and the calibration mode is designated to the millimeter waveband signal measurement circuit 1. .

  The millimeter waveband signal measurement circuit 1 for which the calibration mode is designated converts the received calibration signal Er into a signal Er (if) in an intermediate frequency band that can be digitally processed by the down converter 2 and performs processing such as Fourier transform. The phase information of a plurality of continuous waves included in the calibration signal Er is acquired. Here, the down converter 2 is configured to receive the calibration signal Er by the horn antenna 2c.

  Then, the calibration processing unit 90 receives the calibration signal Er by changing the frequency of the continuous wave included in the calibration signal Er within the frequency region of the millimeter wave band to be calibrated by the continuous wave frequency changing means 80. Further, the phase characteristic Δφ (f) ′ of the millimeter waveband signal measuring circuit 1 is acquired.

  The calibration processing unit 90 compares the phase characteristic Δφ (f) ′ detected by the millimeter waveband signal measurement circuit 1 with the phase characteristic Δφ (f) obtained in advance for the calibration signal Er, and, for example, calculates the difference. Thus, the phase characteristic ΔΦ (f) of only the millimeter wave band signal measurement circuit 1 is obtained, and information on the phase characteristic ΔΦ (f) is set in the signal processing unit 3 of the millimeter wave band signal measurement circuit 1.

  Note that the phase characteristic ΔΦ (f) of the millimeter waveband signal measurement circuit 1 can be regarded substantially as the phase characteristic of the down converter 2 because the phase characteristic of the down converter 2 portion of the analog circuit is dominant.

  When the phase characteristic ΔΦ (f) of the millimeter waveband signal measurement circuit 1 is obtained in this way, the signal under measurement is input to the millimeter waveband signal measurement circuit 1 instead of the calibration signal and the measurement is performed. In addition, the signal processing unit 3 compensates the information regarding the phase with the phase characteristic ΔΦ (f), so that an accurate measurement result for the signal under measurement can be obtained.

  In the above description, the calibration target is limited to the phase. However, the amplitude of the continuous wave included in the calibration signal can be accurately grasped from the time-domain waveform acquired by the calibration signal waveform acquisition unit 60. Therefore, calibration related to amplitude can be performed in the same manner.

  DESCRIPTION OF SYMBOLS 1 ... Millimeter-wave signal measurement circuit, 2 ... Down converter, 20 ... Phase characteristic calibration system, 21 ... Short pulse light source, 22 ... Branch means, 30 ... Synchronization processing part, 31 ... Photoelectric conversion element 32, 33... BPF, 34, 35... Synchronous circuit, 40... Calibration signal generation unit, 41... Composite signal generator, 42... Local signal generator, 42 a. ... multiplier, 43 ... mixer, 50 ... optical variable delay device, 60 ... calibration signal waveform acquisition unit, 61 ... electro-optic crystal, 61 '... photoconductive antenna, 62 ... PBS, 63 ... Polarization measuring device, 64, 64 '... waveform acquisition means, 68 ... current measuring device, 70 ... phase information calculation means, 80 ... continuous wave frequency changing means, 90 ... calibration processing section

Claims (2)

A millimeter wave band signal measuring circuit (1) including a down converter (2) that receives a signal under measurement in a predetermined frequency region of a millimeter wave band or a submillimeter wave band and converts the signal under measurement into a frequency band that can be digitally processed. A phase characteristic calibration system to be calibrated,
A short pulse light source (21) for outputting short pulse light at a predetermined repetition frequency;
Branching means (22) for branching the short pulse light into a first short pulse light and a second short pulse light;
A calibration signal generator (40) for generating a calibration signal in which a plurality of continuous waves having different frequencies within the predetermined frequency region are combined;
Receiving the first short pulse light, extracting an electrical harmonic signal that is an integral multiple of the predetermined repetition frequency, and using the extracted harmonic signal as a reference, a calibration signal output by the calibration signal generator A synchronization processing unit (30) for performing synchronization processing of
An optical variable delay device (50) that receives the second short pulse light, gives a predetermined delay time to the second short pulse light, and outputs it as sampling light;
The calibration signal and the sampling light are received, the calibration signal is sampled at a timing synchronized with the sampling light while changing the delay time of the optical variable delay device, and the time domain waveform of the calibration signal is obtained. A calibration signal waveform acquisition unit (60) to acquire;
Continuous wave frequency changing means (80) for sequentially changing the frequencies of a plurality of continuous waves included in the calibration signal generated by the calibration signal generating unit within the predetermined frequency region;
From the time domain waveform of each calibration signal obtained by the calibration signal waveform acquisition unit for each frequency change by the continuous wave frequency changing means, obtain the phase of a plurality of continuous waves included in each calibration signal, Phase information calculation means (70) for obtaining the characteristic of the phase difference between the frequencies of the continuous wave from the phase over the entire predetermined frequency region,
The measurement result when the calibration signal is supplied to the calibration target millimeter-wave band signal measurement circuit is compared with the phase difference characteristic obtained in advance for the calibration signal, and the millimeter-wave band signal measurement is performed. A phase characteristic calibration system for a millimeter waveband signal measurement circuit, wherein the phase characteristic of the circuit including the down converter is calibrated. A millimeter wave band signal measuring circuit (1) including a down converter (2) that receives a signal under measurement in a predetermined frequency region of a millimeter wave band or a submillimeter wave band and converts the signal under measurement into a frequency band that can be digitally processed. A phase characteristic calibration method to be calibrated,
The short pulse light output at a predetermined repetition frequency is branched into a first short pulse light and a second short pulse light, and based on the first short pulse light, an electrical harmonic that is an integral multiple of the predetermined repetition frequency. Extracting a wave signal;
Generating a calibration signal in which a plurality of continuous waves having different frequencies within the predetermined frequency region are synthesized in a frequency-synchronized state with reference to the extracted harmonic signal;
Providing the second short pulse light with a predetermined delay time and outputting it as sampling light;
Sampling the calibration signal at a timing synchronized with the sampling light while changing the delay time of the sampling light, obtaining a time domain waveform of the calibration signal;
The frequency of a plurality of continuous waves included in the calibration signal is sequentially changed within the predetermined frequency domain, and is included in each calibration signal from the time domain waveform of each calibration signal obtained for each frequency change. Obtaining a phase of a plurality of continuous waves, and obtaining a characteristic of a phase difference between the frequencies of the continuous waves from the phase over the entire predetermined frequency region;
The measurement result when the calibration signal is supplied to the calibration target millimeter wave measurement circuit is compared with the characteristics of the phase difference obtained in advance for the calibration signal, and the millimeter wave band signal measurement circuit Calibrating phase characteristics including the down converter, and a phase characteristic calibration method for a millimeter waveband signal measurement circuit.

Images