Control device, optical frequency comb generation system, control method, optical frequency comb generation method, and program

Abstract

The purpose of the present invention is to provide a technology for reducing the burden on a user required to set the spectral intensity distribution of an optical comb to a predetermined prescribed intensity distribution.　A control device (4) comprises a control unit (41) that executes first processing for estimating a plurality of candidates for a current true value on the basis of a detection result of an optical comb generated by a parameterized optical frequency comb generator (1), the current true value being a true value of a current parameter value that is a value of a parameter of a current generator that is the optical frequency comb generator (1) when the optical comb is generated.

Description

Control device, optical frequency comb generation system, control method, optical frequency comb generation method and program

[0001] The present invention relates to a control device, an optical frequency comb generation system, a control method, an optical frequency comb generation method, and a program.

[0002] In recent years, there has been growing interest in controlling the spectral intensity distribution of optical combs to a predetermined intensity distribution, for example, from the perspective of applications in optical communications and optical measurement.

[0003] T. Sakamoto et al, Opt. Lett., vol.32. p.1515, 2007

[0004] However, controlling the spectral intensity distribution of an optical comb to a predetermined intensity distribution is not always easy, and this can sometimes place a heavy burden on the user.

[0005] In view of the above circumstances, the present invention aims to provide a technology that reduces the burden on users required to make the spectral intensity distribution of an optical comb a predetermined intensity distribution.

[0006] One aspect of the present invention is a control device comprising a control unit that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator, at the time of optical comb generation, based on the detection result of an optical comb generated by a parameterized optical frequency comb generator.

[0007] One aspect of the present invention is an optical frequency comb generation system comprising: a parameterized optical frequency comb generator; and a control device that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator, at the time of optical comb generation, based on the detection result of the optical comb generated by the optical frequency comb generator.

[0008] One aspect of the present invention is a control method executed by a control device comprising a control unit that performs a first process of estimating a plurality of candidates for a current true value, which is the true value of a current parameter value, which is the parameter value of a current generator, which is the optical frequency comb generator, at the time of optical comb generation, based on the detection result of an optical comb generated by a parameterized optical frequency comb generator, the control method comprising a control step of performing the first process.

[0009] One aspect of the present invention is an optical frequency comb generation method performed by an optical frequency comb generation system comprising: a parameterized optical frequency comb generator; a control unit that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator, at the time of optical comb generation, based on the detection result of the optical frequency comb generated by the optical frequency comb generator, the method comprising a control step of performing the first process.

[0010] One aspect of the present invention is a program for causing a computer to function as the control device described above.

[0011] One aspect of the present invention is a program for causing a computer to function as the above-described optical frequency comb generation system.

[0012] This invention makes it possible to reduce the burden on users in order to set the spectral intensity distribution of an optical comb to a predetermined intensity distribution.

[0013] An explanatory diagram illustrating the overview of the optical frequency comb generation system of the embodiment. A diagram showing an example of an optical frequency comb generator that generates an optical comb using a Mach-Zehnder modulator in the embodiment. A diagram illustrating an example of the hardware configuration of the control device in the embodiment. A flowchart showing an example of the processing flow executed by the control device in the embodiment.

[0014] (Embodiment) Figure 1 is an explanatory diagram illustrating the outline of an optical frequency comb generation system 100 according to an embodiment. The optical frequency comb generation system 100 comprises an optical frequency comb generator 1, an optical demultiplexer 2, a photodetector 3, and a control device 4.

[0015] The optical frequency comb generator 1 is a parameterized optical frequency comb generator. Therefore, the operation of the optical frequency comb generator 1 is controlled by controlling the value of a parameter that controls the operation of the device itself. Since the optical frequency comb generator 1 is an optical frequency comb generator, it generates an optical comb. The optical demultiplexer 2 demultiplexes the optical comb in at least two directions. The optical demultiplexer 2 may be, for example, an optical coupler or a MUX (multiplexer). The photodetector 3 detects the optical comb demultiplexed by the optical demultiplexer 2. Therefore, the photodetector 3 detects the optical comb output by the optical frequency comb generator 1.

[0016] The photodetector 3 is, for example, a photodiode. The optical demultiplexer 2 may have a branching ratio and loss that depend on the optical frequency. The optical demultiplexer 2 may also have a branching ratio and loss that do not depend on the optical frequency. The photodetector 3 is used to detect light intensity, but it may also be used in combination with a device that detects optical phase and optical frequency.

[0017] The control device 4 includes a control unit 41 which comprises a processor 91 such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), or NPU (Neural Network Processing Unit) connected by a bus, and a memory 92. The control unit 41 controls the operation of, for example, the optical frequency comb generator 1. The control unit 41 executes, for example, a first process.

[0018] The first process involves estimating multiple candidates for the true current value based on the detection results of the optical comb generated by the optical frequency comb generator 1. The optical comb detection results are, for example, the results of optical comb detection by the photodetector 3.

[0019] The true current value is the true value of the current parameter value. The current parameter value is the parameter value of the current generator, which is the optical frequency comb generator 1, when the optical comb is generated.

[0020] <Specific Example of the First Process> The first process can be any process that can estimate a candidate for the current true value based on the above detection result. Therefore, the first process may be, for example, a process in which a theoretical value acquisition process is performed to obtain multiple candidates for the current true value.

[0021] The theoretical value acquisition process is the process of obtaining the optimal solution to the Sakamoto-Ishijima-Harada equation. The Sakamoto-Ishijima-Harada equation is a system of equations in which each of the following equations represents the result of the interference of light at each of the following frequencies, indicated by a predetermined k frequency peaks (k is 5 or greater) in the frequency spectrum of the optical comb generated by the current generator. The optimal solution can be obtained by any predetermined optimization technique, for example, by Newton's method.

[0022] Each of the optimal solutions obtained through the theoretical value acquisition process is a candidate for the current true value. In other words, multiple optimal solutions obtained through the theoretical value acquisition process are multiple candidates for the current true value. The reason for specifying "multiple optimal solutions" will now be explained.

[0023] In fact, multiple optimal solutions can always be obtained for the Sakamoto-Ishijima-Harada equation. This is because, as shown in the example equation (1) described later, in optical interference, some of the parameters of the optical frequency comb generator 1 appear in the equation that shows the result of optical interference in the form of a square. The reason it appears in the form of a square is because it is an equation for optical interference. More specifically, according to Maxwell's equations, the observed light intensity is expressed by the square of the light amplitude. Since it appears in the form of a square, when optimized, the optimal value of the square of the parameter becomes, for example, the optimal value Q. ２ Therefore, there are two optimal values ​​for this parameter: +Q and -Q, with different signs. Thus, since there are always multiple optimal solutions to the Sakamoto-Ishijima-Harada equation, multiple optimal solutions to the Sakamoto-Ishijima-Harada equation are obtained.

[0024] The reason why there are multiple optimal solutions can be explained even considering the following situations. When using the Mach-Zehnder modulator 11 (including during parallel and serial expansion), even if the positive and negative of Δφ and Δθ are swapped, an optical comb with the same spectral shape is generated. Also, even if each parameter can be correctly estimated, it cannot be specified which part of the parameters of the optical modulator the parameter to be estimated corresponds to. Even during parallel and serial expansion described later, there arises the problem that it cannot be specified which part of the parameters of the optical modulator the optimal solution is.

[0025] <<Specific Example of the Sakamoto-Ishijima-Harada Equation>> Here, a specific example of the Sakamoto-Ishijima-Harada equation will be explained by taking as an example the case where the optical frequency comb generator 1 generates an optical comb by the interference of light using the Mach-Zehnder modulator 11 shown in FIG. 2 (hereinafter referred to as "example comb generator").

[0026] FIG. 2 is a diagram showing an example of an optical frequency comb generator 1 that generates an optical comb using the Mach-Zehnder modulator 11 in the embodiment. The Mach-Zehnder modulator 11 is a Mach-Zehnder modulator that receives an input of light with intensity P ｉｎ The input light with intensity P ｉｎ passes through the Mach-Zehnder modulator 11 and changes, for example, into an optical comb and is output from the Mach-Zehnder modulator 11.

[0027] In the Mach-Zehnder modulator 11, an electrical signal of A １ sin(ωt + φ1) is applied to one waveguide 111, and a first predetermined phase change θ １ is given to the light propagating through the waveguide 111. The phase change of θ １ is caused, for example, by a phase shifter that changes the phase of the light by θ １ when the phase shifter is located on the waveguide 111. Here, A １ represents the amplitude of the electrical signal applied to the waveguide 111, ω represents the angular velocity of the electrical signal, t represents time, and φ １ represents the phase of the electrical signal applied to the waveguide 111.

[0028] The Mach-Zehnder modulator 11 may include a mechanism that performs optical modulation by applying an electrical signal using the electro-optic effect or the like. １ This is the value obtained by converting the amplitude of an electrical signal into the amount of optical phase change induced. Also, θ １ This can be understood as the optical phase that has been altered by applying a bias voltage or the like.

[0029] In the Mach-Zehnder modulator 11, one waveguide 112 has A ２ When an electrical signal sin(ωt + φ2) is applied, a second predetermined phase change θ is applied to the light propagating through the waveguide 112. ２ The process of giving θ is carried out. ２ The phase change is, for example, the phase of light θ ２ This occurs when a phase shifter that only changes the phase is located on the waveguide 112, and is caused by that phase shifter. ２ φ represents the amplitude of the electrical signal applied to the waveguide 112. ２ This represents the phase of the electrical signal applied to the waveguide 112.

[0030] Strength P ｉｎ , amplitude A １ , amplitude A ２ Δφ and Δθ are examples of parameters for the optical frequency comb generator 1.

[0031] However, A １ A ２ This is the value obtained by converting the amplitude of an electrical signal into the amount of optical phase change induced. Also, θ １ θ2 can be understood as the optical phase that is changed by applying a bias voltage or the like.

[0032] In the frequency spectrum of the optical comb output by such a Mach-Zehnder modulator 11, the k-th spectral intensity, counting from the lowest frequency, is expressed by the following equation (1).

[0033]

[0034] Here, J ｋ This represents the k-th order Bessel function. In equation (1), Δθ = (θ １ -θ ２ ) And, Δφ = (φ１ -φ ２ ) Note that equation (1) is the equation for light interference itself.

[0035] Using this formula, for example, the Sakamoto-Ishijima-Harada equation in the case of k=5 is a system of equations that can be expressed, for example, by the following equation (2).

[0036]

[0037] In equation (2), i ０ i １ i ２ i ３ and i ４ i is a predetermined integer that is distinct from each other. ０ i １ i ２ i ３ and i ４ i can be any predetermined value that does not overlap with each other. ｍｉｎ < | i ０ | <i ｍａｘ i ｍｉｎ < | i １ | <i ｍａｘ i ｍｉｎ < | i ２ | <i ｍａｘ i ｍｉｎ < | i ３ | <i ｍａｘ i ｍｉｎ < | i ４ | <i ｍａｘ The range may be restricted as shown below. However, i ｍｉｎ is i ｍａｘ i is a non-negative integer less than a given value. ｍａｘ is a predetermined non-negative integer.

[0038] In the theoretical value acquisition process, the optimization of equation (2) is obtained using optimization techniques such as Newton's method. As can be seen from equation (2), A １ The terms relating to the amplitude of electrical signals such as these are squared. Therefore, by optimization, A １ Even if the value obtained is -Q, (-Q) is also a solution. Thus, A １From the terms relating to the amplitude of electrical signals, we obtain two solutions that differ only in sign. However, it is obviously unusual to have two optimal solutions, so one of them is a ghost solution.

[0039] Even if the optimal solution is obtained through the theoretical value acquisition process, multiple candidate optimal solutions may still exist. For example, when using the Mach-Zehnder modulator 11 (including when extended in parallel or series), the same spectral shape of the optical comb is generated even if the positive and negative values ​​of Δφ and Δθ are swapped. Therefore, multiple optimal solutions exist.

[0040] <Effects of the First Processing> As seen above, the first processing yields candidates for the current true value. In other words, the candidates for the current true value are narrowed down. When the candidates are narrowed down, the burden on the user to adjust the distribution of the spectral intensity of the optical comb to a predetermined intensity distribution is reduced compared to when the candidates are not narrowed down. Therefore, the control device 4 can reduce the burden on the user to adjust the distribution of the spectral intensity of the optical comb to a predetermined intensity distribution.

[0041] <Example of Hardware Configuration of Control Device 4> Figure 3 shows an example of the hardware configuration of the control device 4 in the embodiment. The control device 4 includes a control unit 41 which comprises a processor 91 and a memory 92, and executes a program. The control device 4 functions as a device comprising a control unit 41, an interface unit 42, and a storage unit 43 through the execution of the program.

[0042] More specifically, the processor 91 reads the program stored in the storage unit 43 and stores the read program in the memory 92. By executing the program stored in the memory 92, the processor 91 functions as a device comprising a control unit 41, an interface unit 42, and a storage unit 43.

[0043] The control unit 41 controls the operation of each functional unit of the control device 4. The control unit 41 is not limited to controlling the operation of the optical frequency comb generator 1 described above or executing the first process, but may also acquire information stored in the storage unit 43. Specifically, the process of acquiring information stored in the storage unit 43 is read. The control unit 41 acquires information acquired by the interface unit 42, for example.

[0044] The interface unit 42 is configured to include a communication interface for connecting the control device 4 to an external device. The interface unit 42 communicates with the external device via wired or wireless connection. The external device is, for example, an optical frequency comb generator 1. The interface unit 42 controls the operation of the optical frequency comb generator 1 by communicating with the optical frequency comb generator 1. Controlling the operation of the optical frequency comb generator 1 is, for example, updating the parameters of the optical frequency comb generator 1. The external device is, for example, a photodetector 3. The interface unit 42 obtains the detection result of the optical comb by communicating with the photodetector 3.

[0045] The interface unit 42 may include input devices such as a mouse, keyboard, or touch panel. The interface unit 42 may also be configured as an interface connecting these input devices to the control device 4. In this way, the input devices of the interface unit 42 receive various types of information to the control device 4 via wired or wireless connections. Note that the information does not necessarily have to be input to the communication interface of the interface unit 42; it may also be input to the input devices of the interface unit 42.

[0046] The interface unit 42 outputs various types of information, for example. The interface unit 42 is comprised of a display device such as a CRT (Cathode Ray Tube) display, a liquid crystal display, or an organic EL (Electro-Luminescence) display, and a sound output device such as a speaker. The sound is, for example, speech. The interface unit 42 may be configured as an interface to connect these display devices or sound output devices to the control device 4. Therefore, the interface unit 42 outputs information input to its input device as an image or sound.

[0047] The storage unit 43 is configured using a computer-readable storage medium (non-transitory computer-readable recording medium) such as a magnetic hard disk drive or a semiconductor storage device. The storage unit 43 stores various information related to the control device 4. The storage unit 43 stores various information generated by the operation of the control unit 41, for example. The storage unit 43 may reside, for example, on the cloud.

[0048] Figure 4 is a flowchart showing an example of the processing flow executed by the control device 4 in the embodiment. The control unit 41 acquires the detection result of the optical comb acquired by the photodetector 3 via the interface unit 42 (step S101). Next, the control unit 41 executes the first process (step S102). Next, the control unit 41 outputs the result of the first process to a predetermined output destination (step S103).

[0049] The predetermined output destination may be, for example, a display device or sound output device provided by the interface unit 42, or it may be the storage unit 43. The display device or sound output device to which the result of the first processing is output outputs the result of the first processing by display or sound. Output to the storage unit 43 means writing to the storage unit 43. The information written to the storage unit 43 can be retrieved by the user.

[0050] The control device 4 in this configured embodiment performs the first process. Therefore, for the reasons explained in <Effects of the First Process>, the control device 4 can reduce the burden on the user in order to make the distribution of the spectral intensity of the optical comb a predetermined intensity distribution.

[0051] (Modification) <Second Processing> The control unit 41 may further perform a second processing. The second processing is a process of estimating the optimal comb generation value based on a plurality of candidates (i.e., candidates for the current true value) estimated in the first processing. The optimal comb generation value is a parameter value of the optical frequency comb generator 1 that generates the optimal comb. The optimal comb is an optical comb with a spectral intensity distribution whose difference from a predetermined intensity distribution (hereinafter referred to as the "target distribution") is smaller than a predetermined difference. Such an optimal comb may be, for example, an optical comb with spectral intensity whose variation is smaller than a predetermined variation. An optical comb with spectral intensity whose variation is smaller than a predetermined variation is an optimal comb whose target distribution is a uniform intensity distribution.

[0052] <<Type 1 Processing>> Here, if the optical frequency comb generator 1 is an optical frequency comb generator 1 that generates an optical comb by modulation and interference of light, then in the second processing, for example, Type 1 processing may be performed. Type 1 processing is a process that performs a conversion process, a post-conversion spectrum estimation process, a real spectrum acquisition process, and a first optimal comb estimation process.

[0053] <<<Transformation process and post-transformation spectrum estimation process>>> The transformation process is the process of performing the transformation indicated by the spectral equalization condition (hereinafter referred to as "spectral equalization transformation") on each of the multiple candidates estimated in the first process.

[0054] The post-conversion spectrum estimation process estimates the spectrum of the optical comb generated by the optical frequency comb generator 1, for each converted value obtained through the conversion process, where the converted value is set as a parameter value. More specifically, the converted value is the result of converting candidate current true values ​​by the conversion indicated by the spectral flattening condition.

[0055] The spectral flattening transformation is a well-known transformation, for example, as described in Non-Patent Document 1. The spectral flattening conditions are expressed by the following equations (3) and (4), for example, when the optical frequency comb generator 1 is an example comb generator.

[0056]

[0057]

[0058] However, in equation (3), Δθ = (θ １ -θ ２ ) / 2. Also, in equation (4), Δφ = (φ １ -φ ２ ) / 2.

[0059] Spectral flattening transformation is a transformation in which, if the value to be transformed is the true value of the current parameter, the variation in the spectral intensity of the optical comb output by an optical frequency comb generator that satisfies the first condition is further suppressed. The first condition is that the transformed value obtained by spectral flattening transformation is set as the parameter value.

[0060] If the value to be converted is not the true value of the current parameter value, setting the converted value obtained by spectral flattening conversion to the parameter value of the optical frequency comb generator will not suppress the variation in the spectral intensity of the optical comb output by that optical frequency comb generator.

[0061] The estimation of the spectrum based on the transformed values ​​is performed, for example, by a predetermined mathematical model. The predetermined mathematical model is, for example, a mathematical model that executes equation (1) described above. The predetermined mathematical model may be, for example, equation (5) in the serial type m-th parameter estimation process described later. Alternatively, it may be, for example, equation (6) in the parallel type m-th parameter estimation process described later.

[0062] <<<Actual Spectrum Acquisition Process>>> The actual spectrum acquisition process is the process of acquiring the actual spectrum, which is the spectrum of the actual light, for each conversion value. The actual light is the light generated by the optical frequency comb generator 1, which is in a changing state. Depending on the parameters of the optical frequency comb generator 1, the actual light may be from an optical comb or not from an optical comb.

[0063] The changed state is the state in which the parameter value has changed from the current parameter value by the difference between the value to be converted by the conversion process and the corresponding converted value (hereinafter referred to as the "first difference"). Since the first difference is used to indicate the change from the current parameter value, the first difference is more specifically the value obtained by subtracting the corresponding converted value from the value to be converted by the conversion process.

[0064] As mentioned above, the transformation process involves multiple objects, and therefore, there are multiple transformation values ​​corresponding to each object. In the actual spectrum acquisition process, an actual spectrum is acquired for each of the multiple transformation values.

[0065] In the actual spectrum acquisition process, it is necessary to update the current generator to the optical frequency comb generator 1 which is in a changed state. This update process is performed, for example, by the control unit 41 controlling the optical frequency comb generator 1 so that the parameter values ​​of the optical frequency comb generator 1 are changed by the first difference from the current parameter values. However, this control does not necessarily have to be performed by the control unit 41; for example, it may be performed by another device other than the control device 4, or it may be controlled manually by a person.

[0066] If spectral flattening transformation is performed on the true values ​​of the current parameter values, then the optical frequency comb generator 1, whose parameters have been updated based on the results, should output an optical comb with more suppressed spectral intensity variation. Therefore, the process of acquiring the spectrum of the optical comb output by the optical frequency comb generator 1, whose parameter values ​​have actually been updated, is called the actual spectrum acquisition process.

[0067] <<<First Optimal Comb Estimation Process>>> The first optimal comb estimation process is a process that estimates the converted value from which the first true spectrum was estimated to be the optimal comb generation value. Note that the converted value from which the first true spectrum was estimated is the converted value from which the first true spectrum was estimated based on that converted value. In other words, the converted value from which the first true spectrum was estimated is the converted value from which the first true spectrum was estimated.

[0068] The first true spectrum is the spectrum that differs the smallest from the spectrum obtained in the actual spectrum acquisition process among the multiple spectra estimated in the post-transformation spectrum estimation process.

[0069] As described above, the post-conversion spectrum estimation process is a process that estimates the spectrum for each conversion value using, for example, a predetermined mathematical model, while the actual spectrum acquisition process is a process that acquires the spectrum of the optical comb actually obtained from the optical frequency comb generator 1. Therefore, the first optimal comb estimation process is a process that looks at the difference between the spectrum estimated by calculation and the observed spectrum and estimates the conversion value that gives the spectrum with the smallest difference.

[0070] The transformed values ​​are the result of spectral flattening transformation assuming that the target of the transformation is the current true value. Therefore, the spectrum estimated by the calculation corresponding to the transformed values ​​when this assumption is correct and the observed spectrum both show little variation and little difference from each other. On the other hand, the spectrum estimated by the calculation corresponding to the transformed values ​​when the assumption is incorrect shows a large variation, and the observed spectrum is not even the spectrum of the optical comb.

[0071] This is because the transformed values ​​when the assumption is incorrect are those obtained from ghost solutions and do not satisfy the constraint of optical comb generation. As a result, the difference between spectra corresponding to the transformed values ​​when the assumption is correct is smaller than the difference between spectra corresponding to the transformed values ​​when the assumption is incorrect.

[0072] Thus, when the second process involves the execution of the post-transformation spectrum estimation process, the actual spectrum acquisition process, and the first optimal comb estimation process, it becomes possible to estimate the candidate with a higher probability of being the true value among the candidates for the current true value.

[0073] <<<First explanation using mathematical formulas>>> For clarity, each term will be explained using a mathematical formula with an example. The value of the parameter set in the current generator (i.e., the true current value) is W ０ Let's assume there are two candidates for the estimated current true value, and we'll call them W ＋ , W －Let H be the spectral equalization transformation. The argument of the spectral equalization transformation H is a candidate for the current true value. Therefore, the value to be transformed by the transformation process is W ＋ or W － Therefore, candidate W ＋ The corresponding conversion value is H(W) ＋ ) and candidate W － The corresponding conversion value is H(W) － )

[0074] Therefore, the first difference, which is the difference between the value to be converted by the conversion process and the corresponding converted value, is that the value to be converted by the conversion process is W ＋ In that case, H(W ＋ )-W ＋ = dW ＋ The value to be converted by the conversion process is W － In that case, H(W － )-W － = dW － That is the case.

[0075] Let S be a function that represents the process of estimating the spectrum based on the transformed value. The argument of function S is the transformed value. Using function S, S is the spectrum estimated by the transformed spectrum estimation process of the optical comb generated by the optical frequency comb generator 1, where the transformed value is set as the parameter value.

[0076] Therefore, the converted value H(W) ＋ The spectrum estimated by the post-conversion spectrum estimation process of the optical comb generated by the optical frequency comb generator 1, where ) is set as the parameter value, is S(H(W) ＋ )) is the converted value H(W － The spectrum estimated by the post-conversion spectrum estimation process of the optical comb generated by the optical frequency comb generator 1, where ) is set as the parameter value, is S(H(W) － ))

[0077] Let R(x) be the actual spectrum of light output by the optical frequency comb generator 1 with parameter value x. In this case, the actual spectrum acquisition process will be R(W ０ +dW ＋ ) and the real spectrum R(W０ +dW － ) is obtained.

[0078] Using the notation up to here, the first true spectrum is either S(H(W ＋ )) and S(H(W － )), or R(W ０ +dW ＋ )) and S(H(W ＋ )), and is the one with a smaller difference when compared with the difference between R(W ０ +dW － )) and S(H(W － )). Therefore, when W ０ =W ＋ holds, the first true spectrum is S(H(W ＋ )), and when W ０ =W － holds, the first true spectrum is S(H(W － )).

[0079] The difference between spectra is represented by a predetermined measure. The predetermined measure may be, for example, Σ(measured value of the k-th component - predetermined spectral intensity distribution of the k-th component) ２ ). The predetermined measure may also be, for example, Σ(estimated value of the k-th component - predetermined spectral intensity distribution of the k-th component) ２ . The predetermined light intensity of the k-th component is the light intensity of the k-th component of the optical comb of the predetermined spectral intensity distribution. For example, when generating a flat optical comb, the intensities are equal for different k. That is, when generating a flat optical comb, the light intensity of the k-th component is the same regardless of k.

[0080] As another example, the predetermined measure may be, for example, the light intensity of the k-th component of an optical comb from which unnecessary optical comb components have been removed and whose optical frequency interval is an integer multiple of ω / 2π. The predetermined measure may also be, for example, the light intensity of the k-th component of an optical comb from which unnecessary optical comb components have been removed, whose optical frequency interval is an integer multiple of ω / 2π, and whose necessary optical comb components have equal intensities.

[0081] Therefore, the converted value for which the first true spectrum is estimated, or the converted value from which the first true spectrum is estimated, is W０ = W ＋ In some cases H(W) ＋ ) and W ０ = W － In some cases H(W) － )

[0082] <<Type 2 Processing>> Here, if the optical frequency comb generator 1 is an optical frequency comb generator 1 that generates an optical comb by modulation and interference of light, then in the second processing, for example, Type 2 processing may be performed. Type 2 processing is a process that performs a conversion process, a real spectrum acquisition process, and a second optimal comb estimation process. The conversion process and the real spectrum acquisition process are as described above.

[0083] <<<Second Optimal Comb Estimation Process>>> The second optimal comb estimation process estimates the converted value from which the second true spectrum was estimated to be the optimal comb generation value. Note that the converted value from which the second true spectrum was estimated is the converted value from which the second true spectrum was estimated based on that converted value. In other words, the converted value from which the second true spectrum was estimated is the converted value from which the second true spectrum was estimated.

[0084] The second true spectrum is the spectrum obtained in the actual spectrum acquisition process that has the smallest difference from the spectrum whose spectral intensity distribution is the target distribution (hereinafter referred to as the "target spectrum").

[0085] The target spectrum is a spectrum whose spectral intensity distribution matches the target distribution, and can therefore be described as a predetermined spectrum. The target spectrum can be any predetermined spectrum, for example, the spectral shape of the optical comb that you ultimately want to generate (i.e., the target). The target spectrum can be any shape as long as it is a predetermined spectrum. For example, if you want spectral flattening of the generated optical comb component, the target spectrum can be a spectrum with optical frequency components at equal intensity.

[0086] As another example, the target spectrum may be, for instance, the components of the generated optical comb, with unwanted components removed or attenuated.

[0087] The target spectrum may be a spectrum in which the light intensity is set according to the desired shape. For example, if it is a spectrum from which unwanted components have been removed, the target value of the unwanted components may be zero. The definition of the target value is the light intensity of each component in an optical comb having the shape of the desired target spectrum. As a result, the parameters are optimized to generate an optical comb with an optical frequency interval such that the frequency interval is an integer multiple of 2ω / π.

[0088] Thus, when the second process involves performing a real spectrum acquisition process and a second optimal comb estimation process after the transformation, it becomes possible to estimate the current true value using a process that places less burden on the estimation process.

[0089] <<Third-Type Processing>> If the optical frequency comb generator 1 is an optical frequency comb generator 1 that generates an optical comb by modulation and interference of light, then in the second processing, for example, the third-type processing may be performed. The third-type processing is a process that performs a conversion process, a post-conversion spectrum estimation process, and a third-optimal comb estimation process. The conversion process and the post-conversion spectrum estimation process are as described above.

[0090] <<<Third Optimal Comb Estimation Process>>> The third optimal comb estimation process estimates the converted value estimated as the third true spectrum as the optimal comb generation value. The third true spectrum is the spectrum with the smallest variation in spectral intensity among the multiple spectra obtained in the post-converted spectrum estimation process.

[0091] Furthermore, as a derivative method of the third optimal comb estimation process, a process may be performed in which it is determined whether all candidate current true values ​​can sufficiently suppress the intensity variation of the optical comb, and if not, the original current value is modified.

[0092] This makes it possible to modify the current true value candidates so that they include current true values ​​that sufficiently suppress the intensity variation of the optical comb. Furthermore, by using this in the high-precision processing described later, it becomes possible to gradually improve the precision of the current true value candidates.

[0093] In the third optimal comb estimation process and its derived methods, physical quantities other than the optical spectrum, such as light intensity, may be measured and used to determine the true value. Specifically, the light intensity of the optical comb, which is driven by the current true value and obtained after the conversion process, may be measured, and this value may be compared with the sum of the light intensities obtained from equation (2). The difference between these values ​​may be determined to be less than or equal to the expected value, thereby determining the current true value from among the candidates for the current true value. However, here, the sum of the light intensities is different from the P of equation (2) for different i. ｉ This represents the sum of the values ​​of |i|, where |i| < i ｍａｘ i is all integer values. ｍａｘ This value is large enough to approximate the total light intensity with a predetermined accuracy.

[0094] Thus, the results of measurements of physical quantities other than the optical spectrum, such as light intensity, may be used to estimate candidate current true values.

[0095] The third type of processing can be described as a process that estimates a more favorable conversion value for the spectral flattening of the optical comb, based solely on the results of the post-conversion spectral estimation process, without performing the actual spectral acquisition process in the first type of processing. The fact that the actual spectral acquisition process is not performed in the first type of processing means that the estimation of the favorable conversion value is performed based on the results of the post-conversion spectral estimation process without comparing it with the spectral output of the optical frequency comb generator 1. Spectral flattening refers to a process that reduces the variation in spectral intensity.

[0096] Because the actual spectrum acquisition process is not performed in this way, the third type of processing requires less computation than the first type of processing. Therefore, by performing the third type of processing, the candidate with a higher probability of being the true value among the current true value candidates is estimated with less computation.

[0097] <<Type 4 Processing>> If the optical frequency comb generator 1 is an optical frequency comb generator 1 that generates an optical comb by modulation and interference of light, then in the second processing, for example, Type 4 processing may be performed. Type 4 processing is a process that performs a conversion process, a real spectrum acquisition process, a real conversion estimated value estimation process, and a fourth optimal comb estimation process. The conversion process and the real spectrum acquisition process are as described above.

[0098] <<<Actual Transformation Estimated Value Estimation Process and Fourth Optimal Comb Estimation Process>>> The actual transformation estimated value estimation process is a process for estimating the actual transformation estimated value. The actual transformation estimated value is a value estimated for each actual spectrum obtained in the actual spectrum acquisition process, and is the parameter value of the optical frequency comb generator 1 when generating actual light whose spectrum is the actual spectrum described above. The actual transformation estimated value may be estimated, for example, by executing the theoretical value acquisition process. Therefore, in the actual transformation estimated value estimation process, the actual transformation estimated value may be estimated, for example, by the theoretical value acquisition process.

[0099] The fourth optimal comb estimation process is a process that estimates the conversion value with the smallest third difference among the conversion values ​​obtained in the conversion process as the optimal comb generation value, based on the difference between the actual conversion estimate obtained in the actual conversion estimate estimation process and the conversion value obtained in the conversion process (hereinafter referred to as the "third difference").

[0100] The third difference is, for example, the absolute value of the difference between the actual transformation estimate obtained in the actual transformation estimate estimation process and the transformed value obtained in the transformation process. The third difference can be expressed in any way, as long as it is expressed using a predetermined measure that represents the actual transformation estimate obtained in the actual transformation estimate estimation process and the transformed value obtained in the transformation process.

[0101] As described above, the first type of processing can be described as a technique for estimating the optimal comb generation value by comparing the estimated spectrum with the actual spectrum. In contrast, the fourth type of processing can be described as a technique for estimating the parameter values ​​of the optical frequency comb generator 1 from the actual spectrum and comparing these with the parameter values ​​estimated using a conversion process to estimate the optimal comb generation value.

[0102] <<<Second explanation using mathematical formulas>>> For clarity, each term will be explained using a mathematical formula with an example. The symbols used below that are the same as those used in <<<First explanation using mathematical formulas>>> have the same definition as those defined in <<<First explanation using mathematical formulas>>>. Also, for the sake of simplicity, the explanation below will use the example where the third difference is the absolute value of the difference between the actual transformation estimate obtained in the actual transformation estimation process and the transformation value obtained in the transformation process.

[0103] The parameter value of optical frequency comb generator 1 is W ０ +dW ＋ In that case, the actual spectrum is R(W ０ +dW ＋ Therefore, the actual spectrum obtained in the actual spectrum acquisition process is R(W). ０ +dW ＋ If this is the case, the actual transformation estimate is W ０ +dW ＋ In this case, the third difference is W ０ +dW ＋ -H(W) ＋ It is the size of ).

[0104] The parameter value of optical frequency comb generator 1 is W ０ +dW － In that case, the actual spectrum is R(W ０ +dW － Therefore, the actual spectrum obtained in the actual spectrum acquisition process is R(W). ０ +dW － If this is the case, the actual transformation estimate is W ０ +dW － In this case, the third difference is W ０ +dW － -H(W) － It is the size of ).

[0105] In this case, the fourth optimal comb estimation process uses the converted value H(W). ＋ ) and the converted value H(W － The conversion value with the smallest third difference among these is estimated to be the optimal comb generation value.

[0106] <High-Precision Processing> Incidentally, the optical frequency comb generator 1, in which the optimal comb generation value estimated by the first and second processes is set as the parameter value, is highly likely to output an optical comb with a spectrum closer to the target spectrum than before the execution of the first and second processes. However, it is thought that by performing the first and second processes again on the output optical comb, the difference from the target spectrum may be further reduced.

[0107] Furthermore, it is believed that the more times these first and second processes are repeated while updating the parameters of the optical frequency comb generator 1, the more likely it is that an optical comb with a spectrum that differs even more from the target spectrum can be obtained.

[0108] Therefore, the control unit 41 may perform such repetitive processing. Specifically, the control unit 41 may perform high-precision processing.

[0109] The high-precision processing is a process that includes the execution of a difference determination process. The difference determination process is a process that determines, based on the detection result of the confirmation comb, whether the difference between the spectral intensity of the confirmation comb and the target spectrum (hereinafter referred to as the "target difference") is smaller than a predetermined difference (hereinafter referred to as the "reference difference"). The confirmation comb is an optical comb generated by the confirmation comb generator. The confirmation comb generator is an optical frequency comb generator 1 in which the parameter values ​​are set to the optimal comb generation values ​​estimated by the second processing.

[0110] More specifically, the high-precision processing is a process that repeatedly executes the unit processing until the difference determination process determines that the target difference is smaller than the reference difference, provided that the difference determination process determines that the target difference is not smaller than the reference difference. The unit processing is a process that sequentially executes the first processing, the second processing, the difference determination processing, and the update processing, with the confirmation comb generator as the current generator. The update processing is a process that updates the parameter values ​​of the optical frequency comb generator 1 to the optimal comb generation values ​​estimated in the second processing.

[0111] Therefore, the parameter values ​​of the current generator in the first process during each unit process are the optimal comb generation values ​​estimated in the immediately preceding second process. The current generator in the first process during the unit process is the verification comb generator. The current generator in the first process refers to the optical frequency comb generator 1 that generates an optical comb whose spectrum is a detection result used to estimate multiple candidate current true values.

[0112] <Series-type m-th parameter estimation processing> The optical frequency comb generator 1 may be an optical frequency comb generator (hereinafter referred to as "series comb generator") that generates an optical comb using M modulators (where M is an integer of 2 or more) connected in series. In such a case, the control unit 41 may, for example, sequentially perform the series-type m-th parameter estimation processing for each m from m=1 to m=M.

[0113] The serial-type m-th parameter estimation process is an estimation based on the optical comb output of the serial-type m-th modulator, and is a process that estimates the optimal comb generation value for the serial-type m-th Mach-Zehnder modulator by first and second processes. The serial-type m-th modulator is the m-th modulator of a series comb generator that is connected in series with the 1st to (m-1)th modulators (where m is an integer between 1 and M) of the series comb generator.

[0114] This process treats the m modulators connected in series from the 1st to the mth as a single modulator (hereinafter referred to as the "series-type mth group modulator"), and obtains the optimal comb generation value for that series-type mth group modulator through the first and second processes. However, in this case, the optimization variables are only the parameters of the series-type mth modulator included in the series-type mth group modulator.

[0115] The output of the optical frequency comb generator 1, which consists of M modulators connected in series, is the output of the Mth modulator, and no light is emitted from modulators other than the Mth modulator. However, the series-type mth parameter estimation process also requires the output from modulators other than the Mth modulator. Therefore, the outputs of modulators other than the Mth modulator of the optical frequency comb generator 1 can be obtained by techniques that branch and partially acquire light, such as optical couplers or optical branching circuits.

[0116] Furthermore, by setting the drive signal amplitude of optical modulators other than the Mth to zero, parameter estimation processing can also be performed for the Mth optical modulator.

[0117] By performing this serial-type m-th parameter estimation process, it becomes possible to flatten the spectrum of the optical comb in an optical frequency comb generator 1 consisting of multiple modulators connected in series. Alternatively, it becomes possible to generate an optical comb with an optical spectrum close to the target spectrum.

[0118] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using M (where M is an integer of 2 or more) Mahatzehnder modulators connected in series (hereinafter referred to as the "series Mahatzehnder modulator type").

[0119] The optical frequency comb generator 1 may be an optical frequency comb generator (hereinafter referred to as "series phase modulator type") that generates an optical comb using M optical phase modulators (where M is an integer of 2 or more) connected in series.

[0120] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using M (where M is an integer of 2 or more) optical orthogonal amplitude modulators connected in series (hereinafter referred to as the "series orthogonal amplitude modulator type").

[0121] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using a series-connected configuration of M modulators (where M is an integer of 2 or more) in which optical phase modulators and Mahatzehnder modulators are mixed. (Hereinafter referred to as the "series optical phase modulator and Mahatzehnder modulator type").

[0122] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using a series-connected configuration of M modulators (where M is an integer of 2 or more) in which an optical phase modulator, a Mahatzehnder modulator, and an optical quadrature amplitude modulator, or a part thereof, are mixed. (Hereinafter referred to as the "mixed series modulator type")

[0123] Some or all of the M modulators may be composed of multiple modulators. Each of these M modulators may be optimized by the serial-type m-th parameter estimation process described above.

[0124] For example, when the MZM and the optical phase modulator are connected in series, the theoretical acquisition process may be performed using equation (5) below.

[0125]

[0126] In equation (3), Δθ = (θ １ -θ ２ ) / 2. However, in equation (5), A' ｕｐ＿ｎ This represents one arm of the Mahatzehnder modulator (called the upper arm) and the equivalent drive amplitude of the phase shift induced by n-1 phase modulators. Also, A' ｌｏｗ＿ｎ This represents the equivalent drive amplitude of the phase shift induced by one arm of the Mahatzehnder modulator (called the lower arm) and n-1 phase modulators. ｕｐ＿ｎ This represents the phase of one arm of the Mahatzehnder modulator (called the upper arm) and the equivalent drive signal phase of the phase shift induced by n-1 phase modulators. The drive signal phase φ ｕｐ＿ｎ This represents one arm of the Mahatzehnder modulator (called the upper arm) and the equivalent drive signal phase of the phase shift induced by n-1 phase modulators.

[0127] The equivalent drive amplitude is the amplitude of the drive signal that drives the equivalent modulator. An equivalent modulator is an equivalent circuit configuration of multiple modulators, which is equivalently replaced by a single modulator with equivalent modulation capability. The equivalent drive amplitude value may also be converted into the amount of phase shift induced within the modulator.

[0128] Furthermore, the equivalent drive phase is the phase of the drive signal that drives the equivalent modulator.

[0129] Equation (5) is a recurrence relation, and by using this equation recursively, for any number of modulator stages n, P ｋ This can be obtained.

[0130] In the example above, k is a sufficiently large value, for example, k = L ｓ A system of equations of the form ×M' + 1 may be used, where M' is the number of modulators whose parameters are estimated at one time. ｓThis is the number of parameters that drive the mth modulator, which constitutes the parallel comb generator.

[0131] <Parallel-type m-th parameter estimation processing> The optical frequency comb generator 1 may be an optical frequency comb generator (hereinafter referred to as "parallel comb generator") that generates an optical comb using M (where M is an integer of 2 or more) Mach-Zehnder modulators connected in parallel. In such a case, the control unit 41 may, for example, sequentially perform parallel-type m-th parameter estimation processing for each m from m=1 to m=M.

[0132] The parallel m-th parameter estimation process is an estimation based on the optical comb output of the parallel m-th Mach-Zehnder modulator, and is a process that estimates the optimal comb generation value for the parallel m-th Mach-Zehnder modulator by first and second processes. The parallel m-th Mach-Zehnder modulator is the m-th Mach-Zehnder modulator of a series comb generator that is connected in parallel to the 1st to (m-1)th Mach-Zehnder modulators (where m is an integer between 1 and M) of the parallel comb generator.

[0133] This process involves treating the first to mth m Mach-Zehnder modulators connected in parallel as a single Mach-Zehnder modulator (hereinafter referred to as the "parallel m-th group modulator"), and obtaining the optimal comb generation value for that parallel m-th group modulator through the first and second processes. In this case, however, the optimization variables are only the parameters of the parallel m-th Mach-Zehnder modulator included in the parallel m-th group modulator.

[0134] In this optical frequency comb generator 1, which consists of M Mach-Zehnder modulators connected in parallel, the output is the output of the M-th Mach-Zehnder modulator, and no light is emitted from the Mach-Zehnder modulators other than the M-th. However, in parallel-type m-th parameter estimation processing, the outputs from the Mach-Zehnder modulators other than the M-th are also required. Therefore, the outputs of the Mach-Zehnder modulators other than the M-th in the optical frequency comb generator 1 may be obtained by techniques that branch and partially acquire light, such as optical couplers or optical branching circuits.

[0135] Furthermore, by setting the drive signal amplitude of optical modulators other than the Mth to zero, parameter estimation processing can also be performed for the Mth optical modulator.

[0136] Furthermore, by providing a light blocking mechanism to optical modulators other than the Mth modulator and setting the light intensity to 0, parameter estimation processing can also be performed for the Mth optical modulator.

[0137] In this case, the theoretical value acquisition process may be performed using an equation other than k=0 in equation (1).

[0138] By performing this parallel type m-parameter estimation process, it becomes possible to flatten the spectrum of the optical comb even for an optical frequency comb generator 1 consisting of multiple Mach-Zehnder modulators connected in parallel. Alternatively, it becomes possible to generate an optical comb with an optical spectrum close to the target spectrum.

[0139] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using M (where M is an integer of 2 or more) optical phase modulators connected in parallel (hereinafter referred to as the "parallel phase modulator type").

[0140] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using M (where M is an integer of 2 or more) optical orthogonal amplitude modulators connected in parallel (hereinafter referred to as the "parallel orthogonal amplitude modulator type").

[0141] The optical frequency comb generator 1 may be an optical frequency comb generator that generates an optical comb using M (where M is an integer of 2 or more) optical phase modulators and Mahtzehnder modulators connected in parallel. (Hereinafter referred to as the "parallel optical phase modulator and Mahtzehnder modulator type").

[0142] The optical frequency comb generator 1 may be an optical frequency comb generator (hereinafter referred to as the "mixed parallel modulator type") that generates an optical comb using M units (where M is an integer of 2 or more) of optical phase modulators, Mahatzehnder modulators, and optical quadrature amplitude modulators, or a combination of parts thereof, connected in parallel.

[0143] The optical frequency comb generator 1 may also be a mixed-type optical modulator combining the series and parallel types described above (hereinafter referred to as "parallel-series mixed modulator type").

[0144] As a derivative of the parallel m-th parameter estimation process, some or all of the processing may be performed by combining the processes without dividing them into stages. In that case, the parameter estimation process described above should be performed on all parameters involved in the combined section. For example, when optical phase modulators are connected in parallel, the theoretical acquisition process may be performed using equation (6) below.

[0145]

[0146] However, ξ ｋ is the optical offset given by the Mth modulator. The optical offset may be a real number, a purely imaginary number, or a complex number. Re(ξ ｋ ) represents the amount of light loss. Im(ξ ｋ ) represents the amount of optical phase shift. Also, ξ ｋ The value of k may be a value that depends on the value of k.

[0147] In the example above, k is a sufficiently large value, for example, k = L ｐ A system of equations of the form ×M' + 1 may be used, where M' is the number of modulators whose parameters are estimated at one time. ｐ This is the number of parameters that drive the mth modulator, which constitutes the parallel comb generator.

[0148] <Other> The k frequency peaks used in the theoretical value acquisition process may be frequency peaks extracted using a frequency filter. The frequency filter may be provided, for example, in the photodetector 3.

[0149] The k frequency peaks used in the theoretical value acquisition process may be frequency peaks obtained by a predetermined signal processing method.

[0150] The following describes the case where the optical frequency comb generator 1 is a Mahatzehnder type comb generator. It is clear that this can also be applied to the parallel and series expansions already described.

[0151] Instead of the measured values ​​obtained through the following steps <1> to <5> and equation (2), k theoretically obtained equations may be used. The following processing is performed by the control unit 41. In addition, in the explanation of the following steps <1> to <5>, Δθ = (θ １ -θ ２ ) / 2. Also, Δφ = (φ １ -φ ２ ) / 2.

[0152] Procedure <1> The control unit 41 sets the bias voltage of the Mahatzehnder modulator (= optical phase difference between waveguides) to a value that satisfies the following equation (7).

[0153]

[0154] However, in equation (7), the following equation (8) is satisfied.

[0155]

[0156] Procedure <2> Following Procedure <1>, the control unit 41 performs a theoretical value acquisition process using either equation (9) or equation (10) as predetermined. In addition, equation (8) is satisfied in equations (9) and (10) below.

[0157]

[0158]

[0159]

[0160]

[0161] P １ and P ２ This represents the intensity of light passing through each arm of the modulator. Note that P １ and P ２ This does not include optical branching or optical losses associated with optical propagation in waveguides.

[0162] However, c １ (t) and c ２ (t) is the value of θ in equations (13) and (14) below, where θ = θ １ θ = θ ２The settings are configured in two patterns, and for each, the value of t' is scanned and measured, and the result is taken. That is, in the following equations (13) and (14), the control unit 41 sets the value of θ to the predetermined value shown in <Procedure 1>, θ = θ １ θ = θ ２ After setting it to one of two patterns, the value of t' is scanned. Then, based on the measurement results performed for each value of t', the control unit 41 controls c １ (t) and c ２ Obtain the value of (t).

[0163] The control unit 41 controls θ = θ １ Set to and use the following formula (13) c １ Get (t).

[0164]

[0165] The control unit 41 sets θ = θ2 and obtains c2(t) using the following equation (14).

[0166]

[0167] The scan of t' may be achieved by scanning the phase of the signal driving the Mahatzehnder modulator, or by scanning the delay between signals. The integration range in the above equation may be a finite interval. For example, it may be an integer multiple of 2π / ω. Alternatively, experimental values ​​may be obtained at discrete sample points within the interval.

[0168] Procedure <3> Using one of the predetermined formulas (9) or (10) above, A １ and A ２ The calculation may be performed by the techniques described in (Technology 1) or (Technology 2) below.

[0169] (Technology 1) Based on the parameter estimation method described above, the control unit 41 determines solution A １ and A ２ This is calculated. In this process, as mentioned above, optimization algorithms such as Newton's method may be used. Note that the parameter estimation method mentioned above refers to the serial type m-th parameter estimation process.

[0170] (Technical 2) The control unit 41 performs the following steps (a1) to (a4) A １ and A ２ The following can be calculated. (a1) First, the control unit 41 will A １ = A ２ The amplitude of one of the modulators is temporarily adjusted so that this occurs. Here, for simplicity of explanation, A ２ (a2) Under these conditions (A １ = A ２ ) For one of the predetermined equations (9) or (10) obtained from the measured values, the control unit 41 calculates the 1 / 2 power of the right-hand side. As a result, the control unit 41 calculates J ｋ (A １ (a3) Next the control unit 41 obtains the value of the Bessel function J for order k. ｋ Based on (x), A １ We obtain the following. Furthermore, as is well known, the Bessel function J ｋ The relationship between (x) and the independent variable x is well known. １ This may be obtained using an optimization algorithm such as Newton's method. (a4) A that was temporarily adjusted ２ The control unit 41 returns to its original value and uses one of the predetermined formulas (9) or (10) to J ｋ (A ２ Calculate the value of ).

[0171] Furthermore, in (a1), A ２ A after adjustment １ = A ２ The process of determining whether the condition is met is performed using the Bessel function J estimated in (a3). ｋ This could also be a process to determine whether the value of (x) matches the theoretical value. １ and A ２ If it is determined that they are not equal, the control unit 41 returns to process (a1) and repeats processes (a1) to (a4).

[0172] Step 4: Following Step 3, the control unit 41 calculates the value of Δθ. In calculating Δθ, the control unit 41 first considers the drive amplitude as A １ = 0 and A ２Set to = 0. Next, the control unit 41 scans the value of θ. Then the MZM outputs light for each θ. From the measurement results of the light output by the MZM for each θ, the control unit 41 calculates the period of the output light intensity for each θ. The measurement is performed by the photodetector 3. Then the control unit 41 converts that period to 2π. In this way, the control unit 41 may calculate the value of θ.

[0173] Step <5> Following Step <4>, the control unit 41 calculates the value of Δφ. In calculating Δφ, the control unit 41 considers, for example, the phase f of the drive signal. １ or f ２ One of the predetermined options is scanned. The MZM then outputs light according to the phase of the drive signal. The intensity of the light output by the MZM is then measured according to the phase of the drive signal. The photodetector 3 performs the measurement. Based on the above change in intensity corresponding to the change in the phase of the drive signal, the control unit 41 calculates the period and converts that period to 2π. The control unit 41 calculates the period by setting Δφ = 0 when the light intensity is at its maximum. In this way, the control unit 41 calculates the value of Δφ.

[0174] As can be seen from the explanation so far, by performing steps <1> to <5>, the detection result of the optical comb by the photodetector 3 indicates the power of each light generated by current generators corresponding to the values ​​of multiple different parameters. In this case, the cost required for optical comb detection is reduced. Specifically, for optical comb detection, only light demultiplexing and intensity measurement are required, and optical filters such as MUX are unnecessary. Generally, optical filters are complex circuits and have limitations in resolution. Therefore, not needing to use optical filters has the effect of reducing costs. Thus, the cost required for optical comb detection is reduced when the detection result of the optical comb by the photodetector 3 indicates the power of each light generated by current generators corresponding to the values ​​of multiple different parameters.

[0175] The control device 4 may be implemented using multiple information processing devices connected to each other via a network. In this case, each functional unit of the control device 4 may be distributed and implemented across multiple information processing devices. The variation determination process, which determines whether the variation in the spectral intensity of the confirmation comb is smaller than a predetermined variation called a reference variation, is an example of a difference determination process. The reason is as follows: As described above, the difference determination process was a process that determined whether the difference between the spectral intensity of the confirmation comb and the target spectrum was smaller than a predetermined difference. The target spectrum is a spectrum whose spectral intensity distribution is the target distribution, and the target distribution could also be a distribution with uniform intensity. Therefore, the difference determination process when the target distribution is a distribution with uniform intensity is an example of a difference determination process, and it is a process that determines whether the variation in the spectral intensity of the confirmation comb is smaller than a predetermined variation called a reference variation.

[0176] Furthermore, all or part of the functions of the control device 4 may be implemented using hardware such as ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device), or FPGA (Field Programmable Gate Array). The program may be recorded on a computer-readable recording medium. Computer-readable recording media include, for example, portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into computer systems. The program may also be transmitted via a telecommunications line.

[0177] While embodiments of this invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments and includes designs and the like that do not depart from the spirit of this invention.

[0178] 100...Optical frequency comb generation system, 1...Optical frequency comb generator, 2...Optical demultiplexer, 3...Photodetector, 4...Control device, 11...Mach-Zehnder modulator, 111...Waveguide, 112...Waveguide, 41...Control unit, 42...Interface unit, 43...Storage unit, 91...Processor, 92...Memory

Claims

1. A control device comprising: a control unit that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator, at the time the optical comb is generated, based on the detection result of an optical comb generated by a parameterized optical frequency comb generator.

2. The control device according to claim 1, further comprising: a control unit that performs a second process of estimating an optimal comb generation value, which is a value for the parameters of the optical frequency comb generator that generates an optimal comb, having a spectral intensity distribution such that the difference from a target distribution, which is a predetermined intensity distribution, is smaller than a predetermined difference; and a control unit that further performs a second process of estimating an optimal comb generation value, which is a value for the parameters of the optical frequency comb generator, based on a plurality of candidates estimated in the first process, wherein the optimal comb generation value is a value for an optical comb having a spectral intensity distribution such that the difference from the target distribution, which is a predetermined intensity distribution, is smaller than a predetermined difference.

3. The control device according to claim 1, wherein the detection result indicates the power of each light generated by the current generator corresponding to the values ​​of a plurality of different parameters.

4. The optical frequency comb generator is an optical frequency comb generator that generates an optical comb by modulation and interference of light, and the control device according to claim 2, wherein the second process is performed by: a conversion process that performs a conversion indicated by the spectral flattening condition for each of the plurality of candidates estimated in the first process; a post-conversion spectrum estimation process that estimates the spectrum of an optical comb generated by the optical frequency comb generator, in which the conversion value is set as the parameter value for each conversion value obtained by the conversion process; a real spectrum acquisition process that acquires a real spectrum for each conversion value, which is the spectrum of real light generated by the optical frequency comb generator, in which the parameter value has changed from the current parameter value by the difference between the value to be converted by the conversion process and the conversion value corresponding to the value; and a first optimal comb estimation process that estimates the conversion value from which the first true spectrum, which is the spectrum with the smallest difference from the spectrum obtained in the real spectrum acquisition process among the plurality of spectra estimated in the post-conversion spectrum estimation process, is the optimal comb generation value.

5. The optical frequency comb generator is an optical frequency comb generator that generates an optical comb by modulation and interference of light, wherein the second process includes: a conversion process that performs a conversion indicated by the spectral flattening condition for each of the plurality of candidates estimated in the first process; a real spectrum acquisition process that uses the value obtained by the conversion process as the converted value and acquires a real spectrum for each converted value, which is the spectrum of real light generated by the optical frequency comb generator in a state where the parameter value has changed from the current parameter value by the difference between the value to be converted by the conversion process and the converted value corresponding to the value; and a second optimal comb estimation process that estimates the converted value from which the second true spectrum, which is the spectrum in which the distribution of spectral intensity differs the smallest from the target spectrum which is the target distribution among the spectra acquired in the real spectrum acquisition process, is the optimal comb generation value.

6. The optical frequency comb generator is an optical frequency comb generator that generates an optical comb by modulation and interference of light, and the control device according to claim 2, wherein the second process is performed by: a conversion process that performs a conversion indicated by the spectral flattening condition for each of the plurality of candidates estimated in the first process; a post-conversion spectrum estimation process that estimates the spectrum of an optical comb generated by the optical frequency comb generator, in which the conversion value is set as a parameter value for each of the conversion values ​​obtained by the conversion process; and a third optimal comb estimation process that estimates the conversion value, which is the third true spectrum that has the smallest variation in spectral intensity among the plurality of spectra obtained in the post-conversion spectrum estimation process, as the optimal comb generation value.

7. The optical frequency comb generator is an optical frequency comb generator that generates an optical comb by modulation and interference of light, wherein the second process includes: a conversion process that performs a conversion indicated by the spectral flattening condition for each of the plurality of candidates estimated in the first process; a real spectrum acquisition process that acquires a real spectrum which is the spectrum of real light generated by each optical frequency comb generator, where the parameter value has changed from the value of the current generator by the difference between the value to be converted by the conversion process and the converted value which is a value corresponding to the value obtained by the conversion process; a real conversion estimated value estimation process that estimates a real conversion estimated value which is a value estimated for each real spectrum acquired in the real spectrum acquisition process and is the parameter value of the optical frequency comb generator at the time of generation of the real light whose spectrum is the real spectrum; and a fourth optimal comb estimation process that estimates the conversion value with the smallest difference among the conversion values ​​obtained in the conversion process as the optimal comb generation value, based on the difference between the real conversion estimated value estimated in the real conversion estimated value estimation process and the converted value obtained in the conversion process.

8. The control device according to any one of claims 2 to 7, wherein the control unit performs a variation determination process, which includes a process of determining whether the variation in the spectral intensity of a confirmation comb is smaller than a predetermined variation, based on the detection result of a confirmation comb, which is an optical comb generated by a confirmation comb generator, which is an optical frequency comb generator, in which the parameter value is set to the optimal comb generation value estimated by the second process, and if the variation determination process determines that the variation is not smaller than the reference variation, the control device performs a high-precision processing process, which includes a first process of setting the confirmation comb generator as a current generator, a second process, a variation determination process, and an update process of updating the parameter value to the optimal comb generation value estimated by the second process, in order and repeatedly until the variation determination process determines that the variation is smaller than the reference variation.

9. The control device according to claim 1, wherein the first process involves performing a theoretical value acquisition process to obtain the optimal solution to the Sakamoto-Ishijima-Harada equation, which is a system of equations for each frequency represented by k frequency peaks (k is 5 or more) in the frequency spectrum of the optical comb generated by the current generator, and which is an equation representing the result of light interference at each of the aforementioned frequencies, and the plurality of optimal solutions obtained in the theoretical value acquisition process are a plurality of candidates for the true value of the current.

10. The control device according to claim 9, wherein the k frequency peaks used in the theoretical value acquisition process are frequency peaks extracted using a frequency filter.

11. The control device according to claim 9, wherein the k frequency peaks used in the theoretical value acquisition process are frequency peaks acquired by predetermined signal processing.

12. The control device according to claim 1, wherein the optical frequency comb generator generates an optical comb using a Mach-Zehnder modulator.

13. The control device according to claim 2, wherein the optical frequency comb generator generates an optical comb using M (M is an integer of 2 or more) Mach-Zehnder modulators connected in series, and the control unit performs a series-type m-th parameter estimation process for each m from m=1 to m=M, in which it estimates the optimal comb generation value for the m-th Mach-Zehnder modulator based on the optical comb output of the m-th Mach-Zehnder modulator of the optical frequency comb generator, which is connected in series with the 1st to (m-1)th (m is an integer of 1 or more and less than or equal to M) Mach-Zehnder modulators of the optical frequency comb generator, by first and second processes.

14. The control device according to claim 2, wherein the optical frequency comb generator generates an optical comb using M (M is an integer of 2 or more) Mach-Zehnder modulators connected in parallel, and the control unit performs a parallel type m-th parameter estimation process for each m from m=1 to m=M, in which it estimates the optimal comb generation value for the m-th Mach-Zehnder modulator based on the optical comb output of the m-th Mach-Zehnder modulator of the optical frequency comb generator, which is connected in parallel to the 1st to (m-1)th (m is an integer of 1 or more and less than or equal to M) Mach-Zehnder modulators of the optical frequency comb generator, by first and second processes.

15. An optical frequency comb generation system comprising: a parameterized optical frequency comb generator; a control unit that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator at the time of optical comb generation, based on the detection result of the optical comb generated by the optical frequency comb generator; and a control device.

16. A control method performed by a control device comprising: a control unit that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator, at the time of optical comb generation, based on the detection result of an optical comb generated by a parameterized optical frequency comb generator, the control method comprising: a control step of performing the first process.

17. An optical frequency comb generation method that an optical frequency comb generation system performs, comprising: a parameterized optical frequency comb generator; a control unit that performs a first process of estimating a plurality of candidates for the current true value, which is the true value of the current parameter value, which is the parameter value of the current generator, which is the optical frequency comb generator, at the time of optical comb generation, based on the detection result of the optical comb generated by the optical frequency comb generator, the method comprising a control step of performing the first process.

18. A program for causing a computer to function as the control device described in claim 1.

19. A program for causing a computer to function as the optical frequency comb generation system described in claim 15.

Images