JP6196879B2 - Signal analysis apparatus and signal analysis method - Google Patents

Description

  The present invention relates to a technique for analyzing the spectrum, modulation quality, and the like of an electric signal, and more particularly to a technique for stably and accurately analyzing a signal in a millimeter wave or higher frequency band.

  As the communication speed increases and the capacity increases, the frequency band used is increasing year by year. In recent years, the use of millimeter-wave or higher frequency bands has been proposed, and the signal quality is quantitatively evaluated. Therefore, it is necessary to convert the signal into a frequency band that can be digitally processed by heterodyne frequency conversion processing.

  In this frequency conversion process, the signal to be measured and the local signal are input to the mixer and the difference frequency component is extracted, but since the frequency of the signal to be measured is high, the frequency of the local signal is inevitably high, Conventional electric signal generators are not compatible.

  Conventionally, harmonic mixing has been used to measure signals in such a high frequency band. Harmonic mixing is to extract a difference frequency component between a harmonic of a local signal and a signal under measurement, and can measure a signal in a high frequency band near an integral multiple of the local frequency.

  However, harmonic mixing has a problem in that a large number of false signals are generated because a difference frequency component between a harmonic other than the desired order and the signal under measurement is also generated.

  As a technique for solving this, two coherent lights having different wavelengths are made incident on a light receiver, an electric signal having a frequency corresponding to the wavelength difference is output from the light receiver, and this is used as a local signal for frequency conversion. It has been proposed. Since this method can generate a high-frequency local signal, a high-frequency signal can be measured by fundamental wave mixing.

  This method includes a method using one light source and an optical modulator and a method using two light sources.

FIG. 26 shows an example of the former method. A continuous light (coherent light) P emitted from the light source 1 and a sine wave modulation signal Sm output from the modulation signal generator 2 are obtained from a LiNbO ₃ modulator or the like. The continuous light P is modulated by the modulation signal Sm (see Non-Patent Document 1).

Here the frequency of the continuous light P and f _0, and the frequency of the modulating signal Sm to f _m, the frequency components of f ₀ ± f _m is generated in the modulated light. Depending on the operating conditions of the optical modulator 3 is also possible to modulate at twice the frequency of the modulation signal, the modulated light in this case has a frequency component of f ₀ ± 2f _m. Removing the component of the frequency f ₀ by an optical filter (not shown) as necessary, when the incident modulated light to the light receiver 4, the beat signal between the respective frequency components (electrical signal) is generated. Frequency of the beat signal can be obtained 2f m in the former _case, next 4f _m In the latter case, twice using the modulation signal Sm of lower frequency, a local signal of a higher frequency, such as 4 times. In addition, since this local signal is generated by generating two continuous lights having different frequencies using continuous light emitted from one light source, and extracting the difference component thereof, continuous light emitted from the light source. Even if the frequency f ₀ changes, the frequency of the local signal is not affected.

By inputting the beat signal as a local signal S _{LO to} the mixer 5a of the frequency converter 5, the signal Sx to be measured is converted into a signal S _IF in a low frequency band (intermediate frequency band) that can be digitally processed. Can do. In addition, by performing a plurality of stages of frequency conversion processing in the frequency conversion unit 5, there is a case where conversion to a low frequency band capable of digital processing may be performed.

In this way, the low frequency band converted signal S _IF in is converted by the A / D converter 6 into a digital signal D _IF, is input to the analyzing section 7 for signal analysis receiving a predetermined signal processing. For example, in the case of spectrum analysis performs detection processing of the digital signal D _IF output in the frequency sweep of the other local signal of the frequency conversion of the modulation signal Sm or frequency converter 5, the data of the spectrum waveform Ask for. In the case of modulation analysis performs a baseband conversion process in the previous paragraph for the digital signal D _IF, performs modulation analysis processing of data to the baseband signal obtained by the conversion process, obtaining the modulation quality, etc. .

FIG. 27 shows an example of the latter method, in which the continuous light P ₁ emitted from the first light source 11 and the continuous light P ₂ emitted from the second light source 12 are combined by the optical multiplexer 13. 14 is incident.

In this case, assuming that the frequencies of the two continuous lights P ₁ and P ₂ are f ₁ and f ₂ , a beat signal of the difference frequency | f ₂ −f ₁ | is generated from the light receiver 14. By applying this beat signal as a local signal S _LO to the frequency conversion unit 5, the frequency conversion process can be performed as described above. In the case of this method, the frequency of the local signal S _LO can be varied by changing the frequency of at least one of the light sources 11 and 12.

Akihito Otani, "Current Status and Prospects of Research and Development of Ultra 100GHz Band Measuring Instruments," Proceedings of 2013 IEICE General Conference, CI-2-3, 2013.

M.Kourogi, K. Nakagawa, M. Ohtsu, “Wide-spanoptical frequency comb generator for accurate optical frequency differencemeasurement,” IEEE Journal of Quantum Electronics, vol.29, no.10, pp.2693-2701, Oct. 1993.

Satoshi Tsukamoto, Yuta Ishikawa, and Kazuro Kikuchi, “Optical Homodyne Receiver Comprising Phase and PolarizationDiversities with Digital Signal Processing,” ECOC 2006, Mo.4.2.1, 2006.

In the system using the optical modulator of FIG. 26, there is a problem that the variable range of the frequency of the local signal S _LO , that is, the measurement frequency range of the analyzer is limited by the frequency band of the optical modulator 3, and currently 100 GHz. The degree is the limit.

  In the method of FIG. 27, there is no frequency limitation due to the band of the optical modulator, but the two light sources 11 and 12 are independent of each other and their frequencies are not synchronized, so the frequency accuracy of the local signal is low. is there. In addition, the phase noise of two light sources that are independent of each other is included in the local signal, which causes phase noise in the frequency-converted signal to be analyzed, and the original signal quality cannot be measured correctly. was there.

  In order to solve this problem, a method of synchronizing the two light sources 11 and 12 using an optical phase lock loop (Non-Patent Document 2) has also been proposed, but instability, response delay time, etc. peculiar to feedback loop control As a result, there is a problem that unlocking is likely to occur and the stable operation is difficult particularly when the frequency is greatly varied.

  An object of the present invention is to solve the above-mentioned problems and to provide a signal analysis apparatus and a signal analysis method capable of stably and accurately analyzing a signal in a millimeter wave or higher frequency band.

In order to achieve the object, a signal analyzing apparatus according to claim 1 of the present invention is provided.
The first continuous light emitted from the first light source (22) and the second continuous light emitted from the second light source (23) at a wavelength different from that of the first continuous light are combined. A local signal generator (21) that enters a light receiver (25, 93, 54A) and outputs a local signal having a frequency equal to the frequency difference between the first continuous light and the second continuous light from the light receiver;
A heterodyne frequency converter (30) that converts a signal under measurement to be analyzed into a signal in an intermediate frequency band using the local signal;
An A / D converter (35) for converting the intermediate frequency band signal into a digital signal;
In a signal analysis apparatus having an analysis unit (40) for analyzing the signal under measurement based on a digital signal output from the A / D converter,
At least one of phase noise and frequency error included in the local signal due to phase fluctuations of the first continuous light and the second continuous light or a frequency error with respect to a set value, and an error in the analysis result of the analysis unit. An error factor detection unit (50) for detecting an error factor to be given;
Correction processing is performed in the signal processing path from the frequency conversion unit to the analysis unit based on the error factor detected by the error factor detection unit so that an error due to the error factor does not appear in the analysis result of the analysis unit. and correction means (70) to provided,
The error factor detector is
An optical sideband generator (52) that receives the first continuous light and a high-frequency signal of a predetermined frequency and generates a plurality of sidebands of the predetermined frequency interval around the first continuous light;
An optical bandpass filter (60) for receiving the output light of the optical sideband generator and extracting the sideband in a predetermined band including the frequency of the second continuous light from the output light;
Optical multiplexing / demultiplexing means (53, 53 ′, 61, 92) for receiving and outputting the output light of the optical bandpass filter and the second continuous light;
The output light of the optical multiplexing / demultiplexing means is received, and the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light among the output light of the optical sideband generator and the second A receiver (54, 54A, 54B, 54A to 54D, 93) that outputs a beat component with continuous light;
Based on the beat component of the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light of the optical sideband generator, the local continuous wave It is characterized by detecting at least one of a phase noise or a frequency error of a signal .

The signal analysis device according to claim 2 of the present invention is:
The first continuous light emitted from the first light source (22) and the second continuous light emitted from the second light source (23) at a wavelength different from that of the first continuous light are combined. A local signal generator (21) that enters a light receiver (25, 93, 54A) and outputs a local signal having a frequency equal to the frequency difference between the first continuous light and the second continuous light from the light receiver;
A heterodyne frequency converter (30) that converts a signal under measurement to be analyzed into a signal in an intermediate frequency band using the local signal;
An A / D converter (35) for converting the intermediate frequency band signal into a digital signal;
In a signal analysis apparatus having an analysis unit (40) for analyzing the signal under measurement based on a digital signal output from the A / D converter,
At least one of phase noise and frequency error included in the local signal due to phase fluctuations of the first continuous light and the second continuous light or a frequency error with respect to a set value, and an error in the analysis result of the analysis unit. An error factor detection unit (50) for detecting an error factor to be given;
Correction processing is performed in the signal processing path from the frequency conversion unit to the analysis unit based on the error factor detected by the error factor detection unit so that an error due to the error factor does not appear in the analysis result of the analysis unit. Correction means (70) for
The error factor detector is
An optical sideband generator (52) that receives the first continuous light and a high-frequency signal of a predetermined frequency and generates a plurality of sidebands of the predetermined frequency interval around the first continuous light;
Optical multiplexing / demultiplexing means (53, 53 ', 61, 92) that receives and combines the output light of the optical sideband generator and the second continuous light;
The output light of the optical multiplexing / demultiplexing means is received, and the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light among the output light of the optical sideband generator and the second A receiver (54, 54A, 54B, 54A to 54D, 93) that outputs a beat component with continuous light;
Based on the beat component of the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light of the optical sideband generator, the local continuous wave Configured to detect at least one of phase noise or frequency error in the signal;
Further, the first continuous light is included in one of the lights incident on the optical multiplexing / demultiplexing means of the error factor detecting unit, and the light receiver receiving the output light of the optical multiplexing / demultiplexing means is connected to the sideband. A function of outputting a beat component of a wave and the second continuous light and a function of the light receiver of the local signal generation unit are provided .

The signal analysis device according to claim 3 of the present invention is the signal analysis device according to claim 1 ,
The first continuous light is included in one light incident on the optical multiplexing / demultiplexing means of the error factor detecting unit, and the sideband and the receiver receiving the output light of the optical multiplexing / demultiplexing means A function of outputting a beat component with the second continuous light and a function of the light receiver of the local signal generator are provided.

A signal analysis device according to claim 4 of the present invention is the signal analysis device according to any one of claims 1 to 3 ,
The optical multiplexing / demultiplexing means of the error factor detection unit is a single polarization type 90 degree optical hybrid (53) or a dual polarization type 90 degree optical hybrid (53 ').

The signal analysis device according to claim 8 of the present invention is the signal analysis device according to claim 1,
The error factor detector is
A high frequency signal generator (102) for generating a high frequency signal of a predetermined frequency;
A harmonic mixer (104) that receives the local signal output from the local signal generator and the high-frequency signal output from the high-frequency signal generator and outputs a harmonic component of the high-frequency signal and a beat component of the local signal; And detecting at least one of phase noise or frequency error of the local signal based on a beat component output from the harmonic mixer.

Moreover, the signal analysis device according to claim 5 of the present invention is the signal analysis device according to any one of claims 1 to 4 ,
The correction means complex-multiplies the output signal of the A / D converter by a correction signal including information on at least one of phase noise and frequency error detected by the error factor detector. The signal input to the analysis unit is corrected so that at least one of phase noise or frequency error of the local signal output from the local signal generation unit does not appear.

Moreover, the signal analysis device according to claim 6 of the present invention is the signal analysis device according to any one of claims 1 to 4 ,
A baseband converter (80) for converting the signal converted into the intermediate frequency band into a baseband signal is provided between the A / D converter and the analyzer.
The correction means gives the local signal used by the baseband conversion unit for baseband conversion by providing at least one of phase noise or frequency error detected by the error factor detection unit, so that the output of the baseband conversion unit The signal is corrected so that at least one of phase noise and frequency error of the local signal output from the local signal generator does not appear in the signal.

Moreover, the signal analysis device according to claim 7 of the present invention is the signal analysis device according to any one of claims 1 to 4 ,
The frequency conversion unit includes a first-stage mixer (30a) that receives the local signal generated by the local signal generation unit as a first-stage local signal, at least one subsequent-stage mixer (30c) following the first-stage mixer, A post-stage local signal generator (30d) for supplying a post-stage local signal to the post-stage mixer,
The correction means provides the information on the frequency error detected by the error factor detection unit to the subsequent-stage local signal output by the subsequent-stage local signal generator of the frequency conversion unit, so that the output of the frequency conversion unit The signal is corrected so that a frequency error of the local signal output from the local signal generator does not appear in the signal.

Moreover, the signal analysis device according to claim 8 of the present invention is the signal analysis device according to any one of claims 1 to 4 ,
The analysis unit is provided with frequency output means (40a) for obtaining the frequency of the signal under measurement and outputting it as a part of the analysis result,
The correction means corrects the frequency output by the frequency output means of the analysis unit with information on the frequency error detected by the error factor detection unit.

The signal analysis method according to claim 9 of the present invention includes:
First continuous light and second continuous light having different wavelengths are combined and incident on a light receiver, and a local signal equal to the frequency difference between the first continuous light and the second continuous light is transmitted from the light receiver. A local signal generation stage to output,
A frequency conversion stage of a heterodyne system that converts a signal under measurement to be analyzed into a signal in an intermediate frequency band using the local signal;
An A / D conversion step of converting the intermediate frequency band signal into a digital signal;
In a signal analysis method including an analysis step of analyzing the signal under measurement based on the digital signal,
At least one of phase noise or frequency error included in the local signal due to a phase variation of the first continuous light and the second continuous light or a frequency error with respect to a set value is given to an analysis result in an analysis stage. An error factor detection stage to detect as an error factor;
A correction step for performing a correction process in the signal processing step from the frequency conversion step to the analysis step based on the detected error factor, so that an error due to the error factor does not appear in the analysis result of the analysis step; Provided ,
The error factor detection step includes
An optical sideband generation stage that receives the first continuous light and a high-frequency signal of a predetermined frequency and generates a plurality of sidebands of the predetermined frequency interval around the first continuous light;
An optical multiplexing / demultiplexing stage that receives and combines the output light of the optical sideband generation stage and the second continuous light; and
The output light in the optical multiplexing / demultiplexing stage is incident on a light receiver, and the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light in the optical sideband generation stage, Outputting a beat component with the second continuous light;
Based on the beat component of the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light in the optical sideband generation stage, the local continuous wave Detecting at least one of phase noise or frequency error of the signal,
Further, the first continuous light is included in one light incident on the optical multiplexing / demultiplexing stage of the error factor detecting stage, and the side band is supplied to the light receiver that receives the output light of the optical multiplexing / demultiplexing stage. A function of outputting a beat component of a wave and the second continuous light and a function of the light receiver in the local signal generation stage are provided .

  Thus, in the present invention, the first continuous light and the second continuous light having different wavelengths are combined and incident on the light receiver, and the frequency is equal to the frequency difference between the first continuous light and the second continuous light. Signal analysis is performed by outputting a local signal from the receiver, converting the signal under measurement into a signal in the intermediate frequency band using the heterodyne method, and converting the signal in the intermediate frequency band into a digital signal for analysis. In the apparatus and the signal analysis method, at least one of the phase noise or the frequency error included in the local signal due to a phase variation of the first continuous light and the second continuous light or a frequency error with respect to a set value is analyzed. Phase error or frequency error included in the local signal by detecting the error factor that gives an error to the signal and correcting the detection result along the path from frequency conversion processing to analysis processing. At least one of them so as not appear on the analysis results.

  For this reason, it is possible to stably and accurately analyze a signal under measurement without performing an unstable operation such as feedback loop control while maintaining the broadband characteristics by using two light sources.

Basic configuration diagram of the present invention Operation explanatory diagram of the present invention The figure which shows the structural example of the local signal generation part of this invention, and an error factor detection part. Operation explanatory diagram of the embodiment The figure which shows the structural example of the optical sideband generator of embodiment The figure which shows the structural example of the single polarization | polarized-light type 90 degree optical hybrid of embodiment. The figure which shows the structural example of the balance light receiver of embodiment. The figure which shows the structural example of the calculating part of embodiment. Diagram showing a configuration example of a numerically controlled oscillator The figure which shows the structural example of the correction | amendment means of embodiment. The figure which shows the structural example of the correction | amendment means of embodiment. The figure which shows the structural example of the correction | amendment means of embodiment. The figure which shows the structural example of the correction | amendment means of embodiment. The figure which shows the structural example of the correction | amendment means of embodiment. The figure which shows the structural example of the correction | amendment means of embodiment. The figure which shows the structural example of the local signal generation part of embodiment, and an error factor detection part. The figure which shows the structural example of the 90 degree optical hybrid of the both polarization type of embodiment The figure which shows the structural example of the error factor detection part of embodiment. Operation explanatory diagram of the embodiment The figure which shows the structural example of the local signal generation part of embodiment, and an error factor detection part. Operation explanatory diagram of the embodiment The figure which shows the structural example of the local signal generation part of embodiment, and an error factor detection part. The figure which shows the structural example of the local signal generation part of embodiment, and an error factor detection part. Operation explanatory diagram of the embodiment The figure which shows the structural example of the local signal generation part of embodiment, and an error factor detection part. Configuration diagram of conventional equipment Configuration diagram of conventional equipment

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a basic configuration diagram of a signal analysis apparatus 20 to which the present invention is applied.

  The signal analysis device 20 includes a local signal generation unit 21, a frequency conversion unit 30, an A / D converter 35, an analysis unit 40, an error factor detection unit 50, and correction means 70.

As shown in FIG. 2A, the local signal generation unit 21 includes first continuous light (coherent light: the same applies hereinafter) P having different wavelengths emitted from the first light source 22 and the second light source 23, respectively. _{The first} continuous light P ₁ and the second continuous light P ₂ are combined by the optical multiplexer 24 and incident on the light receiver 25, and as shown in FIG. 2B, the first continuous light P ₁ and the second continuous light P 2. A local signal S _LO having a frequency equal to the frequency difference of P ₂ is output from the light receiver 25 to the frequency converter 30.

In this local signal _LO , the phase noise Δφ due to the respective phase fluctuations Δφ ₁ (t) and Δφ ₂ (t) of the first continuous light P ₁ and the second continuous light P ₂ , and the respective frequency settings. A frequency error Δf caused by the frequency errors Δf ₁ and Δf ₂ with respect to the values f ₁ and f ₂ is included.

The frequency converter 30 has at least one mixer 30a therein, and as shown in FIG. 2C, the signal under test Sx and the local signal S _LO are fed to the first stage mixer 30a and mixed, is of the heterodyne method for extracting a frequency component, as in FIG. 2 (d), it converts the measured signal Sx to the signal S _IF in the intermediate frequency band which can be digitally processed. FIG. 2 illustrates a case where the frequency conversion unit 30 performs one frequency conversion. The frequency converter 30 may perform a plurality of stages of frequency conversion. In this case, the output of the frequency converter 30 is a signal in a frequency band that can be digitally processed.

The converted signal S _IF to the intermediate frequency band, then a frequency error and phase noise contained in the local signal S _LO is added, the signal S _IF of an intermediate frequency band, by the A / D converter 35 It is converted into a digital signal _DIF and input to the correction means 70, and at least one (both in this embodiment) of the frequency error and phase noise added by the local signal S _LO is corrected and input to the analysis unit 40. . The analysis unit 40 analyzes the spectrum of the signal under measurement, the modulation quality of the demodulated data, and the like based on the input digital signal D _IF ′.

On the other hand, the error factor detection unit 50 _generates the phase noise Δφ of the local signal S _LO caused by the phase fluctuations Δφ ₁ (t) and Δφ ₂ (t) of the first continuous light P ₁ and the second continuous light P _2. , At least one of the frequency errors Δf of the local signal S _LO resulting from the frequency errors Δf ₁ and Δf ₂ with respect to the respective frequency setting values f ₁ and f ₂ (both in this embodiment) is an error factor that gives an error to the analysis result The correction means 70 corrects the error signal in the signal processing path from the frequency conversion unit 30 to the analysis unit 40 so that the error factor does not finally appear as the analysis result of the analysis unit 40. . In FIG. 1, the correction means 70 is provided between the A / D converter 35 and the analysis unit 40. However, as will be described later, the position for correcting the frequency error and the phase noise is limited to the example of FIG. Not.

Here, since it is difficult to perform direct frequency measurement and phase detection on the local signal S _LO having an extremely high frequency, the error factor detection unit 50 uses the first continuous light P ₁ and the second continuous light P ₂ . Receiving the light including the frequency and phase information of both lights to the light receiver to obtain the beat component, and the electric local signal _SLO as a high-frequency signal with very little frequency error and phase noise or its multiplied output Is used, and an electric processing method for obtaining the beat component is used.

FIG. 3 shows a configuration example of the local signal generation unit 21 and the error factor detection unit 50 of the signal analysis device 20 using the optical processing method. The local signal generation unit 21 includes the first continuous light P. the _first and second continuous light optical demultiplexer 26 and 27 P ₂ the branches respectively provided, first branched by the first continuous beam P _1a and the optical demultiplexer 27 which is branched by the optical demultiplexer 26 The second continuous light P _2a is incident on the optical multiplexer 24, and the output light is incident on the light receiver 25 to generate the local signal S _LO . Here, the optical demultiplexers 26 and 27 and the optical multiplexer 24 may be realized by an optical fiber coupler, or may be realized by a free space optical system such as a half mirror.

  The error factor detection unit 50 includes a reference signal generator 51, an optical sideband generator 52, a 90-degree optical hybrid 53, balance light receivers 54A and 54B, A / D converters 55A and 55B, and a calculation unit 56. ing.

Reference signal generator 51, the frequency error and phase noise providing a reference signal Sm extremely small predetermined frequency f _m to the optical sideband generator 52. Optical sideband generator 52, as shown in FIG. 4 (a), around the first continuous light P _1b are incident to generate a plurality of sideband frequencies f _m intervals on both sides Then, the light enters the 90-degree optical hybrid 53.

The 90-degree optical hybrid 53 constitutes the optical multiplexing / demultiplexing means of this embodiment. The 90-degree optical hybrid 53 receives the output light Ps from the optical sideband generator 52 and the branched second continuous light _P2b . as in (b), it performs quadrature mixing of the second continuous light P _2b and sideband combining component (phase component = real part) of the second continuous light P _2b and sideband I (+) , I (-), and the combined component (orthogonal component = imaginary part) Q (+) and Q (-) of the light obtained by shifting the phase of the second continuous light _P2b by 90 degrees and the sideband waves are balanced. vessel 54A, by respectively incident to 54B, the balance photodetector 54A, the 54B, as shown in (c) of FIG. 4, to output the beat component of the second continuous light _{P 2b} and sidebands. Here, since the frequency of the local signal is usually higher than the band of the balance light receivers 54A and 54B, the beat component corresponding to the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light. And beat components having a frequency within the band of the balance light receivers 54A and 54B are output.

  Note that the signs (+) and (−) of the output light I (+), I (−), Q (+), and Q (−) of the 90-degree optical hybrid 53 are equal in intensity and are optical phases at the time of multiplexing. Indicates the components inverted from each other, and the phases of the beat components are inverted from each other. Therefore, when the light enters the balance light receivers 54A and 54B, the amplitude of the beat component is doubled and the amplitude of the noise component is √2. Since the amplitude is doubled and the amplitude noise of each incident light is canceled, a high S / N output can be obtained.

  These light receiver outputs I and Q are converted into digital signals Di and Dq by A / D converters 55A and 55B and input to the arithmetic unit 56.

The arithmetic unit 56, the digital signal Di input, based on Dq, Ya phase noise Δφ of the local signal S _LO due to the first continuous light P ₁ and the second phase shift and the frequency error of the continuous light P ₂ A frequency error Δf is obtained.

As shown in FIG. 1, the phase noise and frequency error information obtained by the calculation unit 56 is supplied to a correction means 70 provided between the A / D converter 35 and the analysis unit 40, and the digital signal correction processing for the D _IF is sent to conducted by analyzer 40. The analysis unit 40 performs processing such as spectrum analysis, modulation analysis, and data demodulation on the digital signal D _IF ′ corrected by the correction unit 70.

Next, the operation of the signal analyzing apparatus 20 having the above configuration will be described using mathematical expressions. The first continuous light _wave (setting value) lambda of P _1, the wavelength of the second continuous light (set value) and lambda _2, the first continuous light P ₁ of the optical frequency (set value) f ₁ a second continuous light _{P 2} optical frequency (set value) _{f 2,} the frequency (set _{value) f LO} of the local signal _{S LO} output from the photodetector 25 is expressed as follows.

f ₁ = c / λ ₁ (1)
f ₂ = c / λ ₂ (2)
f _LO = f ₁ −f ₂ (3)
Here, c is the speed of light, and f ₁ > f ₂ . For example, when λ ₁ = 1550 nm and λ ₂ = 1555 nm, f _LO ≈622 GHz.

Further, the optical frequency error with respect to the set value of the first continuous light P ₁ is Δf ₁ , the phase noise is φ ₁ (t), the optical frequency error with respect to the set value of the second continuous light P ₂ is Δf ₂ , and the phase noise is When phi _{2 (t),} the actual first continuous light P ₁ optical electric field e _{1 (t)} and the optical electric field e ₂ of the second continuous light P _{2 (t)} is
e ₁ (t) = E ₁ · exp [j {2π (f ₁ + Δf ₁ ) t + φ ₁ (t)}] (4)
e ₂ (t) = E ₂ · exp [j {2π (f ₂ + Δf ₂ ) t + φ ₂ (t)}] (5)
It is represented by Here, E ₁ and E ₂ are electric field amplitudes (constants) of the first and second continuous lights, respectively.

When f ₁ + Δf ₁ > f ₂ + Δf ₂ , the voltage e _LO (t) of the local signal is
e _LO (t) = E _LO · exp [j {2π (f ₁ −f ₂ + Δf ₁ −Δf ₂ ) t
+ Φ ₁ (t) −φ ₂ (t)}] (6)
= E _LO · exp [j {2π (f _LO + Δf ₁ −Δf ₂ ) t
+ Φ ₁ (t) −φ ₂ (t)}] (7)
Thus, a frequency error of Δf ₁ −Δf ₂ and a phase noise of φ ₁ (t) −φ ₂ (t) are generated in the local signal. Here, E _LO is the voltage amplitude (constant) of the local signal.

  Since the phase noise of the two light sources is usually uncorrelated with each other even if they are the same magnitude, the magnitude of the phase noise of the local signal is usually √2 times the phase noise of one light source.

When the measured signal Sx input to the first stage mixer 30a of the frequency converter 30 is e _RF (t), the frequency of the measured signal Sx is f _RF , and the phase is φ _RF (t), f _RF > f _LO In this case, the intermediate frequency signal e _IF (t) output from the mixer is as follows.

e _RF (t) = E _RF exp [j {2πf _RF t + φ _RF (t)}] (8)
e _IF (t) = ηE _RF exp [j {2π (f _RF −f _LO −Δf ₁ + Δf ₂ ) t
+ Φ _RF (t) −φ ₁ (t) + φ ₂ (t)}] (9)
= ΗE _RF exp [j {2π (f _IF −Δf ₁ + Δf ₂ ) t
+ Φ _RF (t) −φ ₁ (t) + φ ₂ (t)}] (10)
Here, the intermediate frequency is f _IF = f _RF −f _LO , η is the conversion efficiency of the mixer, and E _RF is the voltage amplitude of the signal under measurement.

Therefore, the intermediate frequency signal (finally the analysis result of the analysis device) includes Δf ₂ −Δf ₁ (= −Δf) frequency error and φ ₂ (t) −φ ₁ (t) (= −Δφ). Phase noise is generated.

On the other hand, the optical field of the output of the optical sideband generator 52 e _{m (t),} the modulation frequency of the optical sideband generator 52 (frequency interval sideband) f _m, the I side and the Q side of the balance photodetector If the output (beat component) is e _bi (t) and e _bq (t),
e _m (t)
^{_{= I = -n Σ n E mi}} exp [j {2π (f 1 + Δf 1 + if m) t + φ 1 (t) + φ mi}]
...... (11)
e _b (t)
= E _bi (t) + je _bq (t) (12)
^{_{= I = -n Σ n η b}} E mi E 2 exp [j {2π (f 1 + Δf 1 + if m -f 2 -Δf 2) t
+ Φ ₁ (t) + φ _mi −φ ₂ (t)}] (13)
Here, the symbol _{i = −n} Σ ⁿ is the sum of _{i = −n to} ⁿ , i is the order (integer) of the sideband wave, n is the maximum order (natural number) of the sideband wave, and E _mi is each sideband wave The electric field amplitude, φ _mi is the phase (fixed value) of each sideband, and η _b is the conversion efficiency. The modulation frequency f _m of the optical sideband generator 52 is synchronized with the reference frequency, the frequency error and phase noise is neglected for sufficiently smaller than the frequency error and phase noise of the light source.

The output of the optical sideband generator 52 there are a number of sidebands in the frequency f _m intervals, but caused many also beat component thereof sideband and the second continuous light P _2, balanced photodetector 54A, 54B, Only beat components having frequencies within the bands of the A / D converters 55A and 55B are converted into digital signals. Here, since the frequency of the local signal is usually higher than the band of the balance light receivers 54A and 54B and the A / D converters 55A and 55B, the frequency of the second continuous light is higher than the frequency of the first continuous light. Are beat components corresponding to sidebands close to each other, and beat components having frequencies within the bands of the balance light receivers 54A and 54B and the A / D converters 55A and 55B are converted into digital signals. A low-pass filter that limits the frequency band may be added as necessary.

Further, by performing quadrature mixing using the 90-degree optical hybrid 53 as the optical multiplexing / demultiplexing means, the in-phase component I and the quadrature component Q for the second continuous light P ₂ are obtained, and the positive frequency and the negative frequency are distinguished. Can be detected.

For example, the beat component corresponding to the sideband whose frequency is closest to the second continuous light is selected. If the selected beat component is e _b1 (t),
e _b1 (t)
= Η _b E _mi1 E ₂ exp [j {2π (f ₁ + Δf ₁ + i ₁ f _m −f ₂ −Δf ₂ ) t
+ Φ ₁ (t) + φ _mi1 −φ ₂ (t)}] (14)
It becomes. Here, i ₁ is the order (integer) of the selected sideband. f _m is exactly known for modulation frequency synchronized with the reference frequency. Therefore, for example, by counting the number of beat components appearing in the outputs of the balance light receivers 54A and 54B while sweeping the optical frequency of the light source from the vicinity of f ₁ ≈f ₂ , the order i ₁ of the sideband can be known. . Alternatively, by applying frequency modulation or phase modulation to the reference signal given from the reference signal generator 51 to the optical sideband generator 52, and measuring the modulation degree of the frequency modulation or phase modulation of the beat component e _b1 (t), The order i ₁ of the sideband can be known. Since f ₁ and f ₂ are optical frequencies that do not include an error, that is, set values of optical frequencies, they are naturally known.

Therefore, when the frequency f _b1 of the beat component e _b1 (t) is obtained,
f _b1 = f ₁ + Δf ₁ + i ₁ f _m −f ₂ −Δf ₂ (15)
The frequency error Δf ₁ −Δf ₂ is
Δf ₁ −Δf ₂ = f _b1 −f ₁ + f ₂ −i ₁ f _m (16)
Can be obtained.

That is, if the frequency f _{b1 of the} beat component is accurately measured, the other values f ₁ , f ₂ , i ₁ f _m are known as described above, and therefore the frequency error Δf ₁ −Δf of the local signal S _{LO. 2} is understood.

Further, when the beat component e _b1 (t) represented by the equation (14) is multiplied by an equation [exp [−j {2πf _b1 t}] using the frequency f _{b1 of the} beat component, the low frequency component is obtained. ,
e _b1 (t) exp [−j {2πf _b1 t}]
= Η _b E _mi1 E ₂ exp [j {φ ₁ (t) + φ _mi1 −φ ₂ (t)}] (17)
Thus, by detecting the phase of this output, φ ₁ (t) + φ _mi1 −φ ₂ (t) can be obtained as phase information.

Information on the frequency error and phase noise necessary for the correction obtained in this way is given to the correction means 70. In the correction means 70, for example, a correction signal H including information on frequency error and phase noise with respect to the intermediate frequency signal e _IF (t),
H = exp [j {2π (Δf ₁ −Δf ₂ ) t + φ ₁ (t) + φ _mi1 −φ ₂ (t)}] (18)
Multiply

The intermediate frequency signal e _IF (t) ′ obtained by this multiplication is
e _IF (t) ′
= E _IF (t) exp [j {2π (Δf ₁ −Δf ₂ ) t + φ ₁ (t) + φ _mi1 −φ ₂ (t)}
...... (19)
= ΗE _RF exp [j {2πf _IF t + φ _RF (t) + φ _mi1 } (20)
Thus, the frequency error and phase noise of the two light sources are canceled. Incidentally, actual correction is performed by complex multiplication for digital signal sequence D _IF.

In the analysis unit 40 that has received the digital signal sequence D _IF ′ corrected by the correction means 70 in this way, for example, the frequency spectrum is obtained and the spectrum waveform is displayed on a display (not shown). Also, baseband conversion processing is performed on the corrected digital signal sequence in the intermediate frequency band in the previous stage, modulation analysis processing is performed from the signal converted into the baseband signal, and the modulation quality is measured. In any case, since the influence of the phase noise and frequency error of the local signal is canceled in the signal to be analyzed, the quality of the signal under measurement itself is accurately represented, and a highly accurate signal analysis can be performed.

Although the fixed phase φ _mi1 is added to the calculation result relating to the phase noise, since the absolute phase of the signal under measurement cannot be measured in the first place by spectrum analysis, modulation analysis, or the like, there is no problem with the addition of the fixed phase.

  In the above description, the sideband whose frequency is closest to the second continuous light is selected. However, the present invention is not limited to this, and the beat components for the second continuous light are balanced light receivers 54A, 54B, A / D converters. Arbitrary sidebands falling within the 55A and 55B bands can be selected.

  Even if the magnitude relationship between the optical frequencies of the two light sources and the magnitude relationship between the signal under measurement and the frequency of the local signal are different from those in this description, the frequency error and phase of the light source are the same as described above only by the difference in the negative sign of the equation. An effect of canceling noise can be obtained.

  When performing broadband spectrum analysis, the frequency of the output light of at least one of the first light source 22 or the second light source 23 is continuously varied under the control of the analysis unit 40, and the local frequency is swept. A detection process is performed on the signal corrected by the correction means 70 during the sweep. During the sweep, the sideband is selected as needed so that the beat frequency corresponding to the sideband is within the bands of the balance light receivers 54A and 54B and the A / D converters 55A and 55B.

Next, a specific configuration example of the main part of the embodiment of the configuration example shown in FIG. 3 will be described.
5A and 5B show a configuration example of the optical sideband generator 52. FIG.

Configuration example of FIG. 5 (a), the first continuous beam P _1b the receiving optical phase modulator 52a half mirror 52b on the entrance side and the exit side of the 52c arranged at predetermined intervals, synchronized with the reference frequency fr by the high-frequency signal Sm of the frequency f _m to generate a high-frequency signal generator 52d is inputted to the optical phase modulator 52a, as shown in (a) of FIG. 4, the frequency f of the first continuous light P _{1b 1} centered and generates a sideband of the frequency f _m intervals on both sides. Here, twice of the half mirror 52b, 52c of the spacing for a number of sideband generation is set to an integral multiple of the wavelength of the frequency f _m. In other words, the half mirror 52 b, is an integer multiple frequency f _m of the free spectral range of the configured Fabry-Perot resonator 52c is set to be equal.

The configuration example of FIG. 5B is a supercontinuum type, and the synchronized frequency f _m synchronized with the reference frequency fr with respect to the light intensity modulator 52e that receives the first continuous light P _1b. The pulse signal Pm is generated by a pulse generator 52f and input to apply pulse modulation, the optical pulse is amplified by an optical amplifier 52g, compressed by a pulse compressor 52h, and then a highly nonlinear fiber or nonlinear optical crystal. by entering into the optical nonlinear medium 52i such as to generate a sideband wave of the frequency f _m intervals.

FIG. 6 is a principle configuration diagram of the 90-degree optical hybrid 53 that receives and orthogonally demodulates the output light Ps of the optical sideband generator 52 and the second continuous light _P2b . The 90-degree optical hybrid 53 splits the signal light S (output light Ps from the optical sideband generator 52) and the local light LO (second continuous light P _2b ) into two branches by optical demultiplexers 53a and 53b, respectively. Each one of the branched lights Sa and LOa is input to the optical multiplexer / demultiplexer 53c, the phase of the other branched light LOb is shifted by 90 degrees by the 90-degree phase shifter 53d, and branched with the phase-shifted light LOb '. The light Sb is input to the optical multiplexer / demultiplexer 53e, and the in-phase components (real part) I (+), I (-) of the signal light S with respect to the local light LO and the quadrature components (imaginary part) Q (+), Q (-) Is output. This configuration example is based on the assumption that the input light has a single polarization (single polarization type).

  7A and 7B show configuration examples of the balance light receivers 54A and 54B. In the configuration of FIG. 7A, the (+) side input light and the (−) side input light that make a pair are received by the photodiodes 155a and 155b, respectively, and the output current is converted into a current-voltage converter (for example, transimpedance). Amplifier) 156a and 156b convert it into a voltage signal and subtract the converted voltage with a subtractor 157 to obtain a high S / N in-phase component signal and quadrature component signal from which amplitude noise has been removed. Further, as shown in FIG. 7B, a configuration in which two photodiodes 155a and 155b are connected in series and a current-voltage converter 156 is connected to the connection point is equivalent to the configuration in FIG. Output can be obtained.

FIG. 8A shows an example of the configuration of the arithmetic unit 56. The beat component signals Di and Dq are supplied to the discrete Fourier transformer 56a to perform complex Fourier transform, and the frequency of the beat component is detected by the frequency detection means 56b. f _b1 is obtained, and the frequency error Δf of the local signal S _LO is obtained from the frequency information by the equation (16).

This frequency error Δf is indirectly the difference between the first continuous light P ₁ and the second continuous light P ₂ because the frequency error of the modulation signal Sm for generating the sideband is much smaller than that of the continuous light. This represents the frequency error Δf ₁ −Δf ₂ . Then, the beat frequency f _b1 is given to the numerically controlled oscillator 56c to generate a sine wave whose phase is shifted by 90 degrees at a frequency equal to the negative frequency, and given to the complex multiplication means 56d together with the input signals Di and Dq. The phase noise Δφ can be obtained by performing the complex multiplication of (17), extracting the low frequency component from the calculation result by the low-pass filter 56e, and detecting the phase by the phase detection means 56f.

  Then, the correction process is performed using the correction signal of the equation (18) including the frequency error Δf and the phase noise Δφ thus obtained.

8B is a configuration example in the case where the phase detector 56f of FIG. 8A is omitted and phase noise is output as complex numbers I _Δφ and Q _Δφ .

FIG. 8C shows a calculation unit 56 having a digital PLL configuration, which is a coefficient multiplier for the initial frequency of the beat frequency f _b1 detected by the initial frequency detection means 56b ′ and the detected phase noise. The result obtained by multiplying the feedback coefficient by 56 g is added by the adder 56 h, and the sideband order determination unit 56 i that receives the addition result determines the order of the sideband and outputs frequency error information, corresponding to the frequency error. The two-phase local signal is output from the numerically controlled oscillator 56c, and the error and phase noise can be detected while following the frequency fluctuation.

  The numerically controlled oscillator 56c used in (a) to (c) of FIG. 8 and numerically controlled oscillators 71 and 81 used in the correcting means 70 and the baseband converting unit 80 described later are configured as shown in FIG.

  That is, the adder 71a adds the output data of the register 71b and the frequency data fnco input from the outside, and inputs the addition result to the register 71b. The register 71b takes in data input in synchronization with a clock signal having a predetermined frequency and holds the data for a clock cycle. Accordingly, the output of the register 71b accumulates frequency data every clock input, and the accumulated value becomes data corresponding to the phase. If scaling is performed so that the value of the register 71b makes a round at the phase 2π, continuous operation is possible. The adder 71c adds the phase data φnco input from the outside and the output data of the register 71b, and inputs the addition result to the trigonometric function generator 71d. The trigonometric function generator 71d outputs cos θ and sin θ values for the input phase θ. With the above configuration, a complex sine wave signal of exp [j {2πfncot + φnco}] is obtained. Here, when the frequency data fnco input from the outside changes with time, a complex signal of exp [j {2πΣfnco (t) Δt + φnco}] is obtained (Δt is a clock cycle).

FIG. 10 shows a configuration example of the correction means 70. The correction means 70 generates a complex two-phase local signal (correction signal H) having a frequency equal to the detected frequency error Δf and including the detected phase noise Δφ and having a phase difference of 90 degrees from each other by the numerically controlled oscillator 71. By giving to the multiplier 72, complex multiplication of the correction signal H with respect to the input signal D _IF in the intermediate frequency band is performed, and a complex signal D _IF ′ in the intermediate frequency band in which the frequency error and the phase noise are canceled is generated. It is also possible to input the complex signal D _IF ′ to a band pass filter (not shown) and extract only the positive frequency component and convert it to a real signal D _IF ″ in the intermediate frequency band.

  In the case of an analysis apparatus having a baseband conversion unit that converts an intermediate frequency band signal into a baseband signal between the A / D converter 35 and the analysis unit 40, the baseband conversion unit Correction processing can be performed by including the detected frequency error and phase noise for the local signal.

That is, like the baseband converter 80 shown in FIG. 11, a numerically controlled oscillator 81 for outputting two-phase local signals S _LOa and S _LOb that are 90 degrees out of phase with each other at a predetermined intermediate frequency (−f _IF ). When, the local signal S _LOa, a complex multiplier 82 for performing complex multiplication between the input signal D _IF and S _LOb, a low-pass filter 83 for extracting a base band component of the low band from the output of the complex multiplier 82 In this case, an adder 73 for shifting the frequencies of the local signals S _LOa and S _LOb by a detected frequency error Δf with respect to a predetermined intermediate frequency (−f _IF ) is used as the correction means 70. In addition, the detected base error Δφ is applied to the phase control terminal of the numerically controlled oscillator 81, so that the baseband signal B that has been corrected in the same manner as described above. , It is possible to obtain a Bq.

  Further, as shown in FIG. 12, the baseband conversion unit 80 does not perform the correction process, and the correction process similar to the correction means 70 shown in FIG. 10 (however, the signals Bi and Bq converted into the baseband (however, By performing the correction in the baseband instead of the intermediate frequency band, it is possible to obtain baseband signals Bi ′ and Bq ′ in which the frequency error and the phase noise are corrected in the same manner as described above.

  In the above embodiment, correction is performed by complex multiplication using a correction signal including information on frequency error Δf and phase noise Δφ. However, frequency error correction and phase noise correction may be performed by separate signal processing units. it can.

  For example, the phase noise may be corrected for the numerically controlled oscillator 81 of the baseband converter 80 as indicated by a dotted line in the configuration of FIG. Conversely, the frequency error may be corrected for the numerically controlled oscillator 81 of the baseband converter 80, and the phase noise may be corrected for the baseband signals Bi and Bq. In this case, the phase noise correction means can be composed of a complex multiplication means and a trigonometric function generator.

  Further, as in the correction means 70 shown in FIG. 13, trigonometric functions (for example, sinΔφ and cosΔφ) that are orthogonal to each other with the phase noise Δφ as a phase angle with respect to the outputs Bi and Bq of the baseband conversion unit 80 are used as a trigonometric function generator Baseband signals Bi ′ and Bq ′ in which only the phase error is corrected may be obtained by being generated at 75 and multiplied by the complex multiplier 72 and provided to the analysis unit 40.

In this case, the analysis unit 40 performs an analysis process on the baseband signal in which the phase noise is corrected, so that the analysis result regarding the phase is accurate. Here, for example, the analysis unit 40, the inputted baseband signal Bi ', Bq' and frequency output unit 40a for outputting as part of the analysis results it seeking frequency f _RF of the signal to be measured is provided on the basis of If so, the adder 76 of the correction means 70 corrects (adds and corrects) the output frequency f _RF of the frequency output means 40a by the frequency error Δf to obtain f _RF ′. Further, when measuring the spectrum of the signal under measurement, the spectrum may be shifted by Δf on the frequency axis.

Further, as shown in FIG. 14, when the frequency conversion unit 30 performs a heterodyne conversion process of a plurality of stages (two stages are shown in the figure), the latter stage (any number of stages other than the first stage) may be used. As frequency setting information for the post-stage local signal generator 30d for generating the post-stage local signal supplied to the mixer 30c, the adder 77 subtracts the detected frequency error Δf from the specified local frequency f _LO2 (in the case of the lower heterodyne). Thus, the frequency error can be corrected (addition correction is performed in the case of the upper heterodyne). Then, with respect to the intermediate frequency band S _IF signal frequency error is corrected, for example, if there is a base-band conversion unit 80, the same correction of the phase noise Δφ for numerically controlled oscillator 81 in the baseband converting unit 80 To do. In this case, the configuration of FIG. 10 in which the correction process of the phase noise Δφ is performed on the signal in the intermediate frequency band and the configuration of FIG. 12 in which the baseband converted signal is performed may be employed.

As shown in FIG. 8B, when the phase noise is obtained by the complex data I _Δφ and Q _Δφ in the error factor detection unit 50, the complex data of the phase noise is obtained as shown in FIG. By applying I _Δφ and Q _Δφ to the complex multiplier 78, the phase noise can be corrected.

  In the error factor detection unit 50 of the above-described embodiment, the single polarization type 90-degree optical hybrid 53 that performs orthogonal mixing with respect to single polarization input light is used as optical multiplexing / demultiplexing means. When the frequency error and phase noise of the local signal can be detected regardless of the state, both polarizations that output the in-phase component and the quadrature component for each of the x-polarized wave and the y-polarized wave as shown in FIG. This can be achieved by using four sets of balanced light receivers 54A to 54D that respectively receive the four sets of output light output from the 90-degree optical hybrid 53 'of the type and the 90-degree optical hybrid 53' of both polarization types.

As shown in FIG. 17, the dual-polarization type 90-degree optical hybrid 53 'includes a signal light S (output light Ps from the optical sideband generator 52) and a local light LO (second continuous light P _2b). ) incident respectively polarization separator 53f, in 53 g, were divided into x polarization component and the y polarization component, respectively, the x polarization component Xs of the signal light S and the local light _LO, the _{X LO} between 90 ° light Input to the hybrid 53h, the y polarization components Ys and Y _{LO of} the signal light S and the local light LO are input to the 90-degree optical hybrid 53i, and the in-phase component (real part) and quadrature component (imaginary number) for each polarization Part).

  The actual structures of the single-polarization-type 90-degree optical hybrid 53 and the dual-polarization-type 90-degree optical hybrid 53 'are well known, and although details thereof are omitted here, For example, the 90-degree optical hybrid 53 ′ can be realized with the configuration described in Non-Patent Document 3, and the single-polarization-type 90-degree optical hybrid 53 is realized with half the configuration of Non-Patent Document 3. be able to.

When this dual-polarization 90-degree optical hybrid 53 'is used, the polarization of light from the optical sideband generator 52 is indefinite by using an adaptive filter based on the maximum ratio synthesis method or CMA (Constant Modulus Algorithm). Even in this case, the frequency f _b1 can be obtained in the same manner as described above.

For example, when using a maximum ratio combining, balanced photodetector output I and Q components of the x polarization _{e bxi (t),} e bxq _{(t), the} balance photodetector of I and Q components of the y polarization From the output e _byi (t) and e _byq (t),
_{e bx (t) = e bxi} (t) + je bxq (t) ...... (21)
e _by (t) = e _byi (t) + je _byq (t) (22)
r = (1 / T) ∫ {e _bx (t) / e _by (t)} dt (23)
e _b (t) = r ^* e _bx (t) + e _by (t) (24)
Thus, e _b (t) independent of polarization can be obtained. Here, the symbol の in the equation (23) represents a definite integral at a predetermined time T, and r ^{* in the} equation (24) represents a complex conjugate of r.

  In the above-described embodiment, the output of the optical sideband generator 52 is directly incident on the single polarization type 90-degree optical hybrid 53 or the dual-polarization type 90-degree optical hybrid 53 ′ as optical multiplexing / demultiplexing means. However, the error factor detection process requires beat components of the second continuous light and the sideband near the second continuous light among the sidebands including the frequency error and phase noise of the first continuous light. Other sidebands and the first continuous light component are unnecessary.

  Therefore, in the embodiment shown in FIGS. 3 and 16, sidebands in a predetermined band including the frequency of the second continuous light are generated between the optical sideband generator 52 and the 90-degree optical hybrids 53 and 53 ′. By providing an optical bandpass filter that selectively extracts, beat components due to unnecessary sidebands can be prevented from appearing.

  In the above embodiment, the paired light output in pairs from the 90-degree optical hybrids 53 and 53 ′ is received by the balance light receiver to obtain a beat component from which amplitude noise has been removed. When the influence of the amplitude noise can be ignored, only one of the paired lights can be received by a light receiver including a single light receiving element (photodiode) to obtain a beat component.

  Further, as the optical multiplexing / demultiplexing means of the error factor detection unit 50, a simple optical multiplexer / demultiplexer such as an optical fiber coupler or a half mirror is used without using the 90-degree optical hybrids 53 and 53 'as described above. You can also.

FIG. 18 shows an example of the configuration, and the output light Ps of the optical sideband generator 52 shown in FIG. 19A is input to the optical bandpass filter (optical BPF) 60, and FIG. As shown in b), the sideband component Ps ′ within a predetermined band including the frequency of the second continuous light P _2b is selectively extracted, and the extracted sideband component Ps ′ and the second continuous light P _2b are extracted. Are incident on an optical multiplexer / demultiplexer 61 such as an optical fiber coupler or a half mirror, and the demultiplexed components of both lights are combined and input to the balance light receiver 54.

In this configuration, for example, as shown in (c) of FIG. 19, the multiplexing component sideband components S1~S4 extracted by the optical bandpass filter 60 and the second continuous light P _2b is to balance photodetector 54 is input, the beat component S3 corresponding to the second continuous beam _P sideband component S3 is higher frequency than _2b, S4 as (d) in FIG. 19 ', S4' appears in the frequency position of the difference, second the beat component S1, a frequency from continuous light P _2b correspond to low sideband components S1, S2 ', S2' appears at the folded frequency position in the frequency 0. Of these beat components, for example, by selecting the beat component having the lowest frequency and obtaining the frequency information and phase information, the same correction processing as described above can be performed.

In this configuration example, an optical multiplexer / demultiplexer 61 such as an optical fiber coupler or a half mirror is used as the optical multiplexing / demultiplexing means, and since the beat component is not obtained with complex elements of I and Q, two sidebands are supported. Frequency relations such as when beat components (eg, S2 ′ and S3 ′) to be overlapped cannot be separated or when the frequency of beat components is very close to 0 and phase noise components are folded back at frequency 0 Depending on the case, frequency error and phase noise may not be detected. Moreover, since the magnitude relationship between the sideband wave (for example, S3) and the frequency of the second continuous light _P2b cannot be identified, it is necessary to identify the magnitude relationship between the frequencies by another method. The methods for determining the magnitude of this frequency, and a method for observing the change in the beat frequency frequency f _m to be input to the continuous light P ₁ or frequency or optical sideband generator 52 of P ₂ by slightly changing, continuous There is a method of determining the polarity of the modulation component appearing in the beat frequency by modulating the frequency P _m of the light P ₁ or P ₂ or the frequency fm input to the optical sideband generator 52.

  On the other hand, when 90-degree optical hybrids 53 and 53 'which are orthogonal mixers of light are used as the optical multiplexing / demultiplexing means as shown in FIGS. 3 and 5 together with the optical bandpass filter 60, As shown in e), since the beat component on the negative frequency side can be recognized, the beat components corresponding to the two sidebands are not overlapped or folded back at zero frequency, regardless of the frequency relationship. It is always possible to detect frequency error and phase noise.

Further, as shown in FIG. 19E, the magnitude relationship between the sideband wave (for example, S3) and the frequency of the second continuous light _P2b can be identified on the frequency axis, so that frequency error and phase noise can be easily detected. I can do it.

  Moreover, in the said embodiment, the light receiver 25 of the local signal generation part 21 and the light receiver (balance light receiver 54, 54A-54D) used in the error factor detection part 50 for detecting a frequency difference and phase noise are independent. However, it can be shared.

In that case, the component of the first continuous light P ₁ is included in _one light incident on the optical multiplexing / demultiplexing means (the 90-degree optical hybrid 53 or the optical multiplexer / demultiplexer 61) of the error factor detection unit 50, A sideband near the frequency of the continuous light P ₁ is not included, and a beat component of the sideband and the second continuous light P ₂ is output to a receiver that receives the output light of the optical multiplexing / demultiplexing means. And a function of the light receiver 25 of the local signal generator 21 for outputting a local signal equal to the frequency difference between the first continuous light P ₁ and the second continuous light P ₂ .

An example is shown in FIG. In the configuration example of FIG. 20, the first continuous light P ₁ output from the first light source 22 is branched into two by the optical demultiplexer 26, and one of the P _{1 a} is input to the optical multiplexer 91 and the other P _1b. the enter the optical sideband generator 52 generates the sideband frequency f _m intervals around the first continuous light P ₁ as shown in FIG. 21 (a).

The output light Ps of the optical sideband generator 52 is input to the optical bandpass filter 60 and includes the optical frequency of the _second continuous light P2 among the sideband components as shown in FIG. sideband components in the predetermined band Ps' is input is extracted to the optical multiplexer 91 are multiplexed with the first continuous light P _1a, the similarly demultiplexer such as an optical fiber coupler or a half mirror The data is input to one of the input terminals 92. The other input terminal of the optical multiplexer / demultiplexer 92 is input with the _second continuous light P2 output from the second light source 23 as shown in FIG. Are combined and input to the balance light receiver 93.

Therefore, the output of the balanced photodetector 93, as shown in (d) of FIG. 21, the first continuous light _{P 1a} and the second beat component of the continuous light _{P 2 (f LO + Δf 1} -Δf 2 ) together are included as a high frequency component HF, beat component of the second continuous light P ₂ each sideband component is included as a low frequency component LF, they are separated by a filter (not shown), the first continuous light P _1a and beat component of the second continuous light P ₂ is output to the frequency conversion unit 30 from the high-frequency output terminal of the balanced photodetector 93 as a local signal S _LO, and the second continuous light P ₂ each sideband component The beat component is output from the low frequency output terminal of the balance light receiver 93 to the arithmetic unit 56 via the A / D converter 55 as a signal for obtaining a frequency error and phase noise. In this case, as in the configuration of FIG. 18, the spectrum of the beat component on the low frequency side is folded at a frequency of 0, so that there is a restriction on the frequency relationship as in the case of FIG.

  In the calculation unit 56, the frequency error and the phase noise are detected as described above, and correction processing is performed based on the detected values.

  In the case of this configuration, the optical multiplexer 91, the optical multiplexer / demultiplexer 92, and the balance light receiver 93 are shared by the local signal generation unit 21 and the error factor detection unit 50.

FIG. 22 shows another configuration example. In this configuration example, instead of the optical multiplexer / demultiplexer 92 and the balance light receiver 93 in the configuration example of FIG. 20, the 90-degree optical hybrid 53 and the balance light receivers 54A and 54B are used. 21 (e), the beat component (f _LO + Δf ₁ −Δf ₂ ) of the first continuous light P _1a and the second continuous light P ₂ from the high frequency output terminal (HF) is obtained. causes outputs as a local signal S _LO, from the low-frequency output terminal (LF) to output the in-phase component I of the beat between the second continuous light P ₂ and sideband components, from the other balance photodetector 54B, low frequency An orthogonal component Q of the beat between the _second continuous light P2 and the sideband component is output from the output terminal.

In this configuration example, a beat having a negative frequency can be recognized, and therefore, the frequency error and the phase noise can be detected regardless of the frequency relationship as in the case of FIG. In the case of the configuration of FIG. 22, the local signal _SLO may be output from the high frequency output terminal of the balance light receiver 54B.

Each of the above embodiments is an example of an optical processing method that receives the first continuous light and the second continuous light of the local signal generation unit 21 and obtains the frequency error and phase noise of the local signal. An example of the scheme is shown. FIG. 23 shows an example of the configuration of the electrical processing method. The local signal S _LO output from the light receiver 25 is demultiplexed by the demultiplexer 29, one of which is given to the frequency converter 30, and the other is the error factor detector. 50 local input terminals of the mixer 101.

On the other hand, given the frequency multiplier 103 by the frequency signal of a frequency f _m which is locked to the output reference signal from the reference signal generator 51 generates a high frequency signal generator 102, the frequency f as in (a) of FIG. 24 A harmonic component that is an integral multiple of _m is generated. Multiplying the degree of frequency _{f m} and the frequency multiplier 103 of the high frequency signal, a frequency an if _m of the multiplier output is close to the local signal frequency _{(f LO + Δf 1 -Δf 2} ) ( the difference is the A / D converter 55 Of the input signal band). The output of the frequency multiplier 103 is input to the RF input terminal of the mixer 101. Therefore, from the IF output terminal of the mixer 101, a beat component (difference frequency component) between the multiplied output if _m having a stable phase and frequency and the local signal S _LO is obtained as shown in FIG. and input to the arithmetic unit 56 output signal is converted into a digital signal by the a / D converter 55, the same way so that the frequency error and phase noise of the local signal S _LO is detected.

In this configuration, the output signal e _m (t) of the frequency multiplier 103 is
_{e m (t) = i =} 1 Σ n E mi exp [j {2πif m t + φ mi}] ...... (25)
It becomes. Here, the symbol _{i = 1} Σ ⁿ is the sum of _{i = 1 to} ⁿ , i is the order of the harmonics, n is the maximum order of the harmonics, E _mi is the amplitude of each harmonic, and φ _mi is each harmonic. Is the phase.

When the multiplication order close to the frequency of the local signal S _LO is i ₁ , the beat component e _IF (t) ′ output from the mixer 101 in the case of the lower heterodyne is
e _IF (t) ′
_{= Η'E mi1 exp [j {2π} (i 1 f m -f LO -Δf 1 + Δf 2) t
+ Φ _mi1 −φ ₁ (t) + φ ₂ (t)}] (26)
It becomes. Here, η ′ is the conversion efficiency of the mixer 101.

Here, it is assumed that i ₁ f _m > f _LO + Δf ₁ −Δf ₂ . The f _m increased to | Δf ₁ -Δf ₂ | _<When the _f m / _2, it is possible to estimate the _{i 1} from f LO / _{f m.} Then, Δf ₁ −Δf ₂ and φ ₁ (t) −φ ₂ (t) −φ _mi1 can be obtained by _obtaining the frequency and phase of the beat component e _IF (t) ′. Therefore, the frequency error and phase noise caused by the two light sources used for generating the local signal can be canceled as described above.

  FIG. 25 shows another configuration example of the electrical processing method. In this configuration example, a harmonic mixer 104 is used instead of the mixer 101 having the configuration shown in FIG. 23 and the frequency multiplier 103 is omitted.

That is, the local signal S _LO output from the light receiver 25 is demultiplexed by the demultiplexer 29, one of which is given to the frequency converter 30, and the other is the RF input terminal of the harmonic mixer 104 of the error factor detector 50. To give. On the other hand, a high-frequency signal of a frequency f _m which is locked to the output reference signal from the reference signal generator 51 generates a high frequency signal generator 102 is input to the local input terminal of the harmonic mixer 104. Therefore, the IF output terminal of the harmonic mixer 104, the frequency components of the beat (difference between the harmonics of the high frequency signal of the frequency _{f m} (frequency i · _{f m)} and the local signal (frequency _{f LO} + _{Δf 1 -Δf 2)} The output signal is converted into a digital signal by the A / D converter 55 and input to the arithmetic unit 56.

Again, phase and frequency is the frequency component of the difference between the harmonic and the local signal of a high frequency signal of a stable frequency f _m can be obtained, detecting a frequency error and phase noise of the local signal from the frequency components of the difference Can do.

  In the configuration examples of FIGS. 23 and 25, the spectrum of the signal output from the mixer 101 or the harmonic mixer 104 is folded back at a frequency of 0, so that there is a limitation on the frequency relationship as in the case of FIG.

  In addition, the A / D converter 35 and the A / D converters 55 and 55A to 55D of the respective embodiments described above are A / D converted based on the clock signal synchronized with the reference signal output from the reference signal generator 51. As a result, the frequency error and phase noise of the local signal can be detected more accurately, and the signal under measurement can be analyzed more accurately.

  In the above description, the configuration example of the local signal generation unit 21 and the error factor detection unit 50 and the configuration example of the correction unit 70 are shown, but the combination thereof is arbitrary.

  In the above embodiment, the case where both the phase noise and the frequency error are obtained as error factors has been described. However, in the analysis where the stationary frequency error does not become a problem, only the phase noise is detected and corrected. Conversely, only the frequency error may be detected and corrected.

  DESCRIPTION OF SYMBOLS 20 ... Signal analysis device, 21 ... Local signal generation part, 22 ... 1st light source, 23 ... 2nd light source, 24 ... Optical multiplexer, 25 ... Light receiver, 30 ... Frequency conversion part , 35... A / D converter, 40... Analysis unit, 50... Error factor detection unit, 52 .. optical sideband generator, 53, 53 '.. 90-degree optical hybrid, 54, 54A to 54D, 93... Balance light receiver, 55, 55A to 55D... A / D converter, 56... Arithmetic unit, 60 .. optical bandpass filter, 61, 92. 80 …… Baseband converter, 92 …… Optical multiplexer / demultiplexer

Claims (9)

The first continuous light emitted from the first light source (22) and the second continuous light emitted from the second light source (23) at a wavelength different from that of the first continuous light are combined. A local signal generator (21) that enters a light receiver (25, 93, 54A) and outputs a local signal having a frequency equal to the frequency difference between the first continuous light and the second continuous light from the light receiver;
A heterodyne frequency converter (30) that converts a signal under measurement to be analyzed into a signal in an intermediate frequency band using the local signal;
An A / D converter (35) for converting the intermediate frequency band signal into a digital signal;
In a signal analysis apparatus having an analysis unit (40) for analyzing the signal under measurement based on a digital signal output from the A / D converter,
At least one of phase noise and frequency error included in the local signal due to phase fluctuations of the first continuous light and the second continuous light or a frequency error with respect to a set value, and an error in the analysis result of the analysis unit. An error factor detection unit (50) for detecting an error factor to be given;
Correction processing is performed in the signal processing path from the frequency conversion unit to the analysis unit based on the error factor detected by the error factor detection unit so that an error due to the error factor does not appear in the analysis result of the analysis unit. and correction means (70) to provided,
The error factor detector is
An optical sideband generator (52) that receives the first continuous light and a high-frequency signal of a predetermined frequency and generates a plurality of sidebands of the predetermined frequency interval around the first continuous light;
An optical bandpass filter (60) for receiving the output light of the optical sideband generator and extracting the sideband in a predetermined band including the frequency of the second continuous light from the output light;
Optical multiplexing / demultiplexing means (53, 53 ′, 61, 92) for receiving and outputting the output light of the optical bandpass filter and the second continuous light;
The output light of the optical multiplexing / demultiplexing means is received, and the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light among the output light of the optical sideband generator and the second A receiver (54, 54A, 54B, 54A to 54D, 93) that outputs a beat component with continuous light;
Based on the beat component of the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light of the optical sideband generator, the local continuous wave A signal analysis apparatus for detecting at least one of a phase noise or a frequency error of a signal. The first continuous light emitted from the first light source (22) and the second continuous light emitted from the second light source (23) at a wavelength different from that of the first continuous light are combined. A local signal generator (21) that enters a light receiver (25, 93, 54A) and outputs a local signal having a frequency equal to the frequency difference between the first continuous light and the second continuous light from the light receiver;
A heterodyne frequency converter (30) that converts a signal under measurement to be analyzed into a signal in an intermediate frequency band using the local signal;
An A / D converter (35) for converting the intermediate frequency band signal into a digital signal;
In a signal analysis apparatus having an analysis unit (40) for analyzing the signal under measurement based on a digital signal output from the A / D converter,
At least one of phase noise and frequency error included in the local signal due to phase fluctuations of the first continuous light and the second continuous light or a frequency error with respect to a set value, and an error in the analysis result of the analysis unit. An error factor detection unit (50) for detecting an error factor to be given;
Correction processing is performed in the signal processing path from the frequency conversion unit to the analysis unit based on the error factor detected by the error factor detection unit so that an error due to the error factor does not appear in the analysis result of the analysis unit. Correction means (70) for
The error factor detector is
An optical sideband generator (52) that receives the first continuous light and a high-frequency signal of a predetermined frequency and generates a plurality of sidebands of the predetermined frequency interval around the first continuous light;
Optical multiplexing / demultiplexing means (53, 53 ', 61, 92) that receives and combines the output light of the optical sideband generator and the second continuous light;
The output light of the optical multiplexing / demultiplexing means is received, and the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light among the output light of the optical sideband generator and the second A receiver (54, 54A, 54B, 54A to 54D, 93) that outputs a beat component with continuous light;
Based on the beat component of the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light of the optical sideband generator, the local continuous wave Configured to detect at least one of phase noise or frequency error in the signal;
Further, the first continuous light is included in one of the lights incident on the optical multiplexing / demultiplexing means of the error factor detecting unit, and the light receiver receiving the output light of the optical multiplexing / demultiplexing means is connected to the sideband. waves and the the function of outputting a beat component of the second continuous light, signal analyzer you characterized by having the function of the light receiver of the local signal generator. The first continuous light is included in one light incident on the optical multiplexing / demultiplexing means of the error factor detecting unit , and the sideband and the receiver receiving the output light of the optical multiplexing / demultiplexing means the second a function of outputting a beat component of a continuous light, the local signal generation portion of the signal analyzer according to claim 1, characterized in that have a function of the light receiver. The optical multiplexing / demultiplexing means of the error factor detection unit is a single polarization type 90 degree optical hybrid (53) or a dual polarization type 90 degree optical hybrid (53 ') . 4. The signal analyzing apparatus according to any one of 3 . The correction means complex-multiplies the output signal of the A / D converter by a correction signal including information on at least one of phase noise and frequency error detected by the error factor detector. the signal input to the analysis unit, in any one of claims 1 to 4, wherein said correcting so that at least one of the phase noise or frequency error of the local signal does not appear that the local signal generation portion outputs The signal analysis apparatus described. A baseband converter (80) for converting the signal converted into the intermediate frequency band into a baseband signal is provided between the A / D converter and the analyzer.
The correction means gives the local signal used by the baseband conversion unit for baseband conversion by providing at least one of phase noise or frequency error detected by the error factor detection unit, so that the output of the baseband conversion unit the signal, signal analysis apparatus according to claim 1, characterized in that the corrected so that at least one of the phase noise or frequency error of the local signal does not appear that the local signal generation portion is output. The frequency conversion unit includes a first-stage mixer (30a) that receives the local signal generated by the local signal generation unit as a first-stage local signal, at least one subsequent-stage mixer (30c) following the first-stage mixer, A post-stage local signal generator (30d) for supplying a post-stage local signal to the post-stage mixer,
The correction means provides the information on the frequency error detected by the error factor detection unit to the subsequent-stage local signal output by the subsequent-stage local signal generator of the frequency conversion unit, so that the output of the frequency conversion unit the signal, signal analysis apparatus according to any one of claims 1 to 4, wherein said correcting so that the frequency error of the local signal does not appear that the local signal generation portion is output. The analysis unit is provided with frequency output means (40a) for obtaining the frequency of the signal under measurement and outputting it as a part of the analysis result,
Said correcting means, said the information of the detected frequency error by the error factor detection unit, according to any one of the preceding claims, characterized in that to correct the frequency of the frequency output means outputs the analysis unit Signal analysis equipment. First continuous light and second continuous light having different wavelengths are combined and incident on a light receiver, and a local signal equal to the frequency difference between the first continuous light and the second continuous light is transmitted from the light receiver. A local signal generation stage to output,
A frequency conversion stage of a heterodyne system that converts a signal under measurement to be analyzed into a signal in an intermediate frequency band using the local signal;
An A / D conversion step of converting the intermediate frequency band signal into a digital signal;
In a signal analysis method including an analysis step of analyzing the signal under measurement based on the digital signal,
At least one of phase noise or frequency error included in the local signal due to a phase variation of the first continuous light and the second continuous light or a frequency error with respect to a set value is given to an analysis result in an analysis stage. An error factor detection stage to detect as an error factor;
A correction step for performing a correction process in the signal processing step from the frequency conversion step to the analysis step based on the detected error factor, so that an error due to the error factor does not appear in the analysis result of the analysis step; Provided,
The error factor detection step includes
An optical sideband generation stage that receives the first continuous light and a high-frequency signal of a predetermined frequency and generates a plurality of sidebands of the predetermined frequency interval around the first continuous light;
An optical multiplexing / demultiplexing stage that receives and combines the output light of the optical sideband generation stage and the second continuous light; and
The output light in the optical multiplexing / demultiplexing stage is incident on a light receiver, and the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light in the optical sideband generation stage, Outputting a beat component with the second continuous light;
Based on the beat component of the sideband wave whose frequency is closer to the second continuous light than the frequency of the first continuous light in the output light in the optical sideband generation stage, the local continuous wave Detecting at least one of phase noise or frequency error of the signal,
Further, the first continuous light is included in one light incident on the optical multiplexing / demultiplexing stage of the error factor detecting stage, and the side band is supplied to the light receiver that receives the output light of the optical multiplexing / demultiplexing stage. A signal analysis method comprising: a function of outputting a beat component of a wave and the second continuous light; and a function of the light receiver in the local signal generation stage.

Images