Optical communication device, optical communication system, and frequency offset reduction method

Abstract

A power derivation unit derives the power difference or power ratio between a positive-side frequency component and a negative-side frequency component of a signal spectrum of a coherently detected reception signal by using light output from a light source unit. A digital signal processing unit estimates the frequency offset amount of the coherently detected reception signal, and compensates the reception signal for the estimated frequency offset amount. A control amount derivation unit derives a control amount on the basis of the power difference or the power ratio when the power difference or the power ratio is larger than a predetermined condition, and derives a control amount on the basis of the estimated frequency offset amount when the power difference or the power ratio is smaller than the predetermined condition. A frequency control unit changes, by the derived control amount, the frequency of the light output by the light source unit.

Description

Optical communication device, optical communication system, and frequency offset reduction method

[0001] The present invention relates to an optical communication device, an optical communication system, and a method for reducing frequency offset.

[0002] In recent years, with the increase in data center traffic, there has been a demand for higher speeds and larger capacities in optical communication devices that transmit and receive optical signals, such as optical transceivers that connect between data centers and within data centers. In response to this demand for higher speeds and larger capacities, digital coherent transmission methods, which have previously been applied to medium-distance applications such as between cities and long-distance applications such as between continents, are beginning to be applied to short-distance applications between data centers and within data centers. The standard "400ZR," established by the standardization organization "The Optical Internetworking Forum (OIF)," defines the specifications for digital coherent transmission methods for distances of 120 km or less.

[0003] Digital coherent transmission systems require a light source with high frequency accuracy and a high-precision digital signal processor (DSP). Compared to the Intensity Modulation - Direct Detection (IM-DD) transmission system used in conventional short-distance applications, the increased power consumption, cost, and size of optical communication equipment are problems. In particular, the light source uses a wavelength locker to improve frequency accuracy, which leads to increased power consumption and cost. Eliminating the wavelength locker is effective in reducing the power consumption, cost, and size of optical communication equipment.

[0004] When a light source unit without a wavelength locker is used, the frequency accuracy of the light source unit decreases, and problems such as an increased frequency offset arise. Light source units with wavelength lockers, which are commonly used in digital coherent transmission, require a frequency accuracy of several GHz. For example, the frequency offset value for "400ZR" is specified to be within a range of ±3.6 GHz. However, if a wavelength locker is not used, the frequency accuracy of the light source unit will be approximately 12.5 GHz, and the maximum frequency offset value is expected to be approximately 25 GHz.

[0005] When the frequency offset increases, the signal quality deteriorates due to the bandwidth limitation caused by the sampling rate of the analog-to-digital converter (ADC), making it impossible to detect the frequency offset, which may make it impossible to apply the frequency offset estimation and frequency offset compensation (see, for example, Non-Patent Document 1) that are performed in conventional digital signal processing.

[0006] FIG. 12 shows the relationship between the ADC bandwidth and the detected received signal. Symbol E1 represents the power of the received signal component, and symbol E2 represents the power of the noise component. In FIG. 12(a), the frequency f of the received signal is within the band limit of the ADC, so the optical communication device can detect the received signal. On the other hand, in FIG. 12(b), most of the frequency range of the received signal is outside the band limit of the ADC, so the optical communication device has difficulty detecting the received signal. In this case, the optical communication device cannot apply frequency offset estimation or frequency offset compensation to the received signal. Therefore, when using a light source unit without a wavelength locker, new estimation and compensation methods that can accommodate a wide range of frequency offsets are required.

[0007] Therefore, a method can be considered in which the power difference or power ratio between the positive frequency component and the negative frequency component of the received spectrum is used as a variable to derive a control amount for the light source unit so that the frequency offset approaches zero, and frequency control of the light source unit is repeated based on the derived control amount. Fig. 13 is a diagram showing an example configuration of an optical communication system that performs such processing. The optical communication system has multiple optical communication devices. In Fig. 13, the multiple optical communication devices are two devices, optical communication device #1 and optical communication device #2.

[0008] The optical communication device is an optical transceiver. In the optical communication device, the optical detection unit coherently detects an optical signal using local oscillator (LO) light output from the light source unit. The ADC converts the electrical signal output from the optical detection unit from an analog signal to a digital signal. The power derivation unit performs a fast Fourier transform (FFT) on the received signal output by the ADC to convert it into a frequency domain signal. The power derivation unit calculates a power difference or a power ratio between the power of a positive frequency component and the power of a negative frequency component of the received signal converted into the frequency domain. Specifically, the power derivation unit uses frequency "0" as a reference (boundary of the domain) and extracts frequency components on the positive side and negative side of the reference frequency from the received signal in the frequency domain. The power derivation unit derives a power difference or a power ratio between the power of the extracted positive frequency component and the power of the extracted negative frequency component. Hereinafter, the power difference between the power of the positive frequency component and the power of the negative frequency component in the received signal will be referred to as the "positive / negative power difference" or simply as the "power difference," and the power ratio between the power of the positive frequency component and the power of the negative frequency component will be referred to as the "positive / negative power ratio" or simply as the "power ratio."

[0009] FIG. 14 shows an example of a received signal spectrum derived by the power derivation unit of the conventional optical communication device shown in FIG. 13 . In FIG. 14 , the horizontal axis represents frequency with a reference value of 0, and the vertical axis represents power. The control amount derivation unit derives a control amount using a cost function with a power difference or power ratio as a variable so that the frequency offset approaches a cost value X (X is, for example, 0) corresponding to 0. That is, the control amount derivation unit calculates a control amount such that the positive and negative power difference approaches 0 or the positive and negative power ratio approaches 1. The frequency control unit controls the frequency of the light source unit to change based on the control amount derived by the control amount derivation unit. The control amount is, for example, the temperature of the light source unit or the amount of current injected into the light source unit. In the optical communication system shown in FIG. 13 , as shown in FIG. 14 , frequency offset reduction is possible even if the received signal is missing from the frequency range that the ADC can process.

[0010] I. Fatadin and SJ Savory, “Compensation of Frequency Offset for 16-QAM Optical Coherent Systems Using QPSK Partitioning,” in IEEE Photonics Technology Letters, Vol. 23, No. 17, pp. 1246-1248, Sept. 1, 2011.

[0011] Fig. 15 is a diagram showing an example of frequency control in a conventional optical communication system. Fig. 15 shows an extracted configuration related to frequency control in the optical communication system of Fig. 13. When the frequency offset is 0, ideally the power difference is 0 and the power ratio is 1. Therefore, if frequency control is performed so that the power difference approaches 0 or the power ratio approaches 1, the frequency offset approaches 0.

[0012] However, if there is a spectral power difference due to factors unrelated to the frequency offset, such as optical amplifier tilt, the spectral power difference will be 0 or the power ratio will be 1 when the frequency offset is not 0. In particular, if the spectral power difference unrelated to the frequency offset differs between opposing optical transceivers, there is a problem in that the laser frequency gradually increases or decreases after the frequency offset has converged to a certain extent, resulting in unstable operation. Furthermore, as a result, the temperature of the temperature regulator that controls the laser frequency and the current used to adjust the temperature also increase or decrease, causing an increase in power consumption in the light source unit.

[0013] For example, in optical communication device #1, when the frequency offset is 0, the power difference of the received spectrum is 0 (or the power ratio is 1), and in optical communication device #2, when the frequency offset is not 0, the power difference of the received spectrum is 0 (or the power ratio is 1). Also, assume that at a certain time, the frequency offset is reduced to a certain extent. At this time, the frequency in optical communication device #1 is f １１ , the frequency in the optical communication device #2 is f ２１ In this state, the optical communication device #2 adjusts the laser frequency to f so that the power difference of the received spectrum becomes 0 (or the power ratio becomes 1). ２２ (Step S901). As a result, a state in which a frequency offset exists is created. Then, the optical communication device #1 changes the laser frequency so that the power difference of the received spectrum becomes 0 (or the power ratio becomes 1). As a result, the optical communication device #1 changes the laser frequency of its own device to f so as to reduce the frequency offset. １２ and the laser frequency f ２２ (Step S902) By repeating these operations, the laser frequency gradually increases.

[0014] In view of the above circumstances, an object of the present invention is to provide an optical communication device, an optical communication system, and a frequency offset reduction method that can reduce the frequency offset while stabilizing the laser frequency, even when there is a spectral power difference between opposing optical transceivers that is unrelated to the frequency offset.

[0015] an optical communication device according to one embodiment of the present invention, comprising: a light source unit; an optical detection unit that performs coherent detection on a received optical signal using light output from the light source unit; a power derivation unit that derives a power difference or a power ratio between the power of a frequency component on the positive side with respect to a reference frequency and the power of a frequency component on the negative side with respect to the reference frequency in a signal spectrum of the received signal obtained by the coherent detection; a digital signal processing unit that performs an estimation process to estimate a frequency offset amount of the received signal obtained by the coherent detection and a compensation process to compensate for the estimated frequency offset amount in the received signal; a control quantity derivation unit that performs a control quantity derivation process to derive a control quantity for changing the frequency of light output by the light source unit based on the derived power difference or power ratio if the derived power difference or power ratio is greater than a predetermined condition, and to derive the control quantity based on the frequency offset amount estimated by the digital signal processing unit if the derived power difference or power ratio is smaller than the predetermined condition; and a frequency control unit that controls the light source unit using the derived control quantity to change the frequency of light output by the light source unit.

[0016] An optical communication system according to one aspect of the present invention includes a plurality of optical communication devices connected by optical transmission paths, each of the optical communication devices including a light source unit, an optical detection unit that performs coherent detection on a received optical signal using light output from the light source unit, a power derivation unit that derives a power difference or a power ratio between a power of a frequency component on the positive side with respect to a reference frequency and a power of a frequency component on the negative side with respect to the reference frequency in a signal spectrum of the received signal obtained by the coherent detection, an estimation process that estimates a frequency offset amount of the received signal obtained by the coherent detection, and a frequency offset amount estimated in the received signal. a control quantity derivation unit that performs control quantity derivation processing to derive a control quantity for changing the frequency of light output by the light source unit based on the derived power difference or power ratio if the derived power difference or power ratio is greater than a predetermined condition, and to derive the control quantity based on the frequency offset amount estimated by the digital signal processing unit if the derived power difference or power ratio is smaller than the predetermined condition; and a frequency control unit that controls the light source unit using the derived control quantity to change the frequency of light output by the light source unit.

[0017] a frequency offset reduction method according to one aspect of the present invention, the frequency offset reduction method comprising: an optical detection step of coherently detecting a received optical signal using light output from a light source unit; a power derivation step of deriving a power difference or a power ratio between a power of a frequency component on the positive side with respect to a reference frequency and a power of a frequency component on the negative side with respect to the reference frequency in a signal spectrum of the received signal obtained by the coherent detection; an estimation step of estimating a frequency offset of the received signal obtained by the coherent detection; a compensation step of compensating the received signal for the estimated frequency offset; a control variable derivation step of performing a control variable derivation process of deriving a control variable for changing a frequency of light output by the light source unit based on the derived power difference or power ratio if the derived power difference or power ratio is greater than a predetermined condition, and deriving the control variable based on the frequency offset estimated in the estimation step if the derived power difference or power ratio is smaller than the predetermined condition; and a frequency control step of controlling the light source unit using the derived control variable to change the frequency of light output by the light source unit.

[0018] According to the present invention, even when there is a power difference in the spectrum between opposing optical transceivers that is not related to the frequency offset, it is possible to reduce the frequency offset while stabilizing the laser frequency.

[0019] FIG. 1 is a configuration diagram of an optical communication system according to a first embodiment. FIG. 2 is a diagram showing signal power after coherent detection in the optical communication device according to the first embodiment. FIG. 3 is a flow diagram showing frequency control processing of the optical communication device according to the first embodiment. FIG. 4 is a flow diagram showing frequency control processing of the optical communication device according to the first embodiment. FIG. 5 is a configuration diagram of an optical communication system according to a second embodiment. FIG. 6 is a functional block diagram of an optical communication device according to the second embodiment. FIG. 7 is a flow diagram showing frequency control processing of the optical communication device according to the second embodiment. FIG. 8 is a flow diagram showing frequency control processing of the optical communication device according to the second embodiment. FIG. 9 is a diagram showing an example of change in frequency of a light source unit according to a third embodiment. FIG. 10 is a flow diagram showing frequency control processing of the optical communication device according to the third embodiment. FIG. 11 is a flow diagram showing frequency control processing of the optical communication device according to the third embodiment.

[0020] Embodiments of the present invention will be described in detail below with reference to the drawings. Conventional laser frequency control in optical communication devices using the power difference or power ratio of received spectra can reduce frequency offset even when it is difficult to estimate the frequency offset due to a large frequency offset, but it may cause frequency instability. Therefore, an optical communication device according to this embodiment switches between laser frequency control using the power difference or power ratio of received spectra and laser frequency control based on frequency offset estimation by a DSP depending on whether the power difference or power ratio exceeds a predetermined threshold. This makes it possible to reduce frequency offset while stabilizing the laser frequency even when a frequency offset that cannot be compensated for by a conventional DSP exists due to a missing received signal in the frequency domain of the ADC, and when a spectral power difference exists due to factors unrelated to the frequency offset exists. Furthermore, it is possible to reduce power consumption required for laser temperature control. Each embodiment will be described below.

[0021] (First Embodiment) FIG. 1 is a configuration diagram of an optical communication system 1 according to a first embodiment of the present invention. FIG. 1 shows only the configuration related to this embodiment. The optical communication system 1 has two optical communication devices 10. The two optical communication devices 10 are referred to as optical communication devices 10-1 and 10-2, respectively. The opposing optical communication devices 10-1 and 10-2 are connected by two optical transmission paths 20. The optical transmission paths 20 are, for example, SMF (single mode fiber) or MCF (multicore fiber). The optical transmission path 20 along which the optical signal output from the optical communication device 10-1 transmits is referred to as optical transmission path 20-1, and the optical transmission path 20 along which the optical signal output from the optical communication device 10-2 transmits is referred to as optical transmission path 20-2.

[0022] The optical communication device 10 is an optical transceiver and includes a light source unit 11, an electrical signal generation unit 12, an optical modulation unit 13, an optical detection unit 14, an ADC (analog-to-digital converter) 15, a digital signal processing unit 16, and a frequency control unit 17. The electrical signal generation unit 12 and the digital signal processing unit 16 are, for example, digital signal processors (DSPs).

[0023] The light source unit 11 outputs light. The light source unit 11 includes a laser such as a DFB (Distributed Feedback) laser or a VCSEL (Vertical Cavity Surface Emitting Laser), and a temperature control element for frequency control. Here, the light source unit 11 of the optical communication device 10-i (i=1, 2) emits light at a frequency f ｉ The light output from the light source unit 11 is branched, one of the branched lights is input to the optical modulation unit 13, and the other branched light is input to the optical detection unit 14.

[0024] The electrical signal generator 12 is realized by, for example, an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). The electrical signal generator 12 maps the transmission signal to symbols and samples the symbol-mapped transmission signal. The electrical signal generator 12 performs band limiting and pre-equalization on the sampled transmission signal, and then converts the digital signal into an analog signal. The electrical signal generator 12 outputs the I (in-phase) component and Q (quadrature) component of each of the X polarization and the Y polarization, which have been converted into analog signals, to the optical modulator 13.

[0025] The optical modulation unit 13 is an optical modulator that modulates the light output from the light source unit 11 with the electrical signal generated by the electrical signal generation unit 12 to generate an optical signal to be transmitted. The optical modulation unit 13 includes a modulator driver and a Mach-Zehnder modulator. The modulator driver amplifies the I component of the X polarization, the Q component of the X polarization, the I component of the Y polarization, and the Q component of the Y polarization of the transmission signal output from the electrical signal generation unit 12, and drives the Mach-Zehnder modulator with these amplified signal components. As a result, the optical modulation unit 13 generates an optical signal in which the X polarization optical signal and the Y polarization optical signal are polarization-multiplexed, and outputs the generated optical signal to the optical transmission path 20.

[0026] The optical detection unit 14 is composed of an optical 90-degree hybrid and a BPD (balanced photodiode). The optical detection unit 14 receives an optical signal output from the opposing optical communication device 10 and transmitted through the optical transmission path 20. The optical detection unit 14 coherently detects the optical signal using local oscillator (LO) light output from the light source unit 11. The optical detection unit 14 outputs a received electrical signal obtained by coherent detection.

[0027] The ADC 15 is a digital-to-analog converter that converts the electrical signal output from the optical detector 14 from an analog signal to a digital signal. The ADC 15 outputs the digitally converted received signal to the digital signal processor 16.

[0028] The digital signal processing unit 16 is realized by, for example, an FPGA or an ASIC, and includes a power derivation unit 161, a control amount derivation unit 162, a compensation unit 163, and a demodulation unit 164.

[0029] The power derivation unit 161 analyzes the spectrum of the received signal received by the optical detection unit 14. Specifically, the power derivation unit 161 performs a fast Fourier transform (FFT) on the received signal output by the ADC 15 to convert it into a frequency domain signal. The power derivation unit 161 calculates the power difference or power ratio between the power of the positive frequency component and the power of the negative frequency component of the received signal converted into the frequency domain. That is, the power derivation unit 161 uses frequency "0" as a reference (boundary of the domain) and extracts frequency components (spectrum) on the positive side and negative side of the reference frequency from the frequency domain received signal. For example, the power derivation unit 161 may split the frequency domain received signal and extract the positive frequency component from one of the split received signals using a bandpass filter with a passband for the positive frequency, and extract the negative frequency component from the other split received signal using a bandpass filter with a passband for the negative frequency. After acquiring the power of the positive frequency component and the power of the negative frequency component, the power derivation unit 161 derives a power difference or a power ratio therebetween. Before calculating the power difference or the power ratio, the power derivation unit 161 may normalize the spectrum of the received signal obtained by FFT.

[0030] When the frequency offset is larger than a predetermined value, the control amount derivation unit 162 derives the control amount based on the power difference or power ratio of the reception spectrum derived by the power derivation unit 161. On the other hand, when the frequency offset is smaller than a predetermined value, the control amount derivation unit 162 derives the control amount based on the frequency offset estimated value calculated by the DSP, i.e., one or both of the compensation unit 163 and demodulation unit 164 of the digital signal processing unit 16. The control amount is the temperature of the light source unit 11, the amount of current injected into the laser used as the light source unit 11, etc. The following mainly describes an example in which the control amount is the temperature of the light source unit 11.

[0031] The compensating unit 163 compensates for the frequency offset of the received signal output by the ADC 15. Note that the digital signal processing unit 16 may further compensate for the frequency characteristics in the optical detecting unit 14, the chromatic dispersion suffered in the optical transmission line 20, and the phase noise of the received signal before or after the frequency offset compensation by the compensating unit 163. The demodulating unit 164 performs predetermined demodulation and decoding processes on the received signal whose frequency offset has been compensated for by the compensating unit 163.

[0032] Note that one or both of the compensating unit 163 and the demodulating unit 164 estimate the frequency offset amount using any existing technology. For example, the compensating unit 163 estimates the frequency offset amount using a phase error of the received signal. The phase error can also be obtained based on the position of a symbol estimated for the received signal. As an example, the technology described in Non-Patent Document 1 can be used to estimate the frequency offset amount. Specifically, the compensating unit 163 uses the fourth power method, which utilizes the distribution characteristics of symbols in a Quadrature Phase Shift Keying (QPSK) signal, to estimate the frequency offset amount based on the position between adjacent symbols obtained from the received signal by the demodulating unit 164. The compensating unit 163 compensates the received signal for the frequency offset of the estimated frequency offset amount.

[0033] The frequency control unit 17 controls the light source unit 11 to change the frequency of the light based on the control amount derived by the control amount derivation unit 162. The digital signal processing unit 16 checks the power difference or power ratio of the received spectrum after the frequency control by the frequency control unit 17, and performs the next frequency control depending on the check result.

[0034] 2A and 2B are diagrams showing the received power obtained by the power derivation unit 161 of the optical communication device 10. FIG. 2A shows the received power when the frequency offset is large, and FIG. 2B shows the received power when the frequency offset is small. T ｓis the sampling period of the ADC 15. Symbol E1 is the power of the received signal component, and symbol E2 is the power of the noise component. With frequency "0" as the reference (boundary of the region), components of frequencies greater than the reference frequency are positive frequency components, and components of frequencies smaller than the reference frequency are negative frequency components.

[0035] As shown in FIG. 2(a), when the frequency offset is large, the sampling period T ｓ Due to the limitation of the ADC band caused by the above, part of the spectrum of the received signal cannot be detected. In such a case, demodulation by the DSP, i.e., the compensation unit 163 and demodulation unit 164 of the digital signal processing unit 16, becomes impossible. When demodulation by the DSP is impossible, the power derivation unit 161 calculates the power difference or power ratio of the received spectrum, and the control amount derivation unit 162 repeatedly performs frequency control to change the frequency of the light output by the light source unit 11 so that the power difference becomes 0 or the power ratio becomes 1. As a result, as shown in FIG. 2( b), the spectrum of the received signal becomes included within the ADC band, and the frequency offset is reduced to a level that can be demodulated by the DSP.

[0036] When the frequency offset is reduced to the point where the DSP can demodulate the received signal, the control amount derivation unit 162 switches to frequency control based on the frequency offset estimated by the DSP. The control amount derivation unit 162 repeats this frequency control to adjust the laser temperature, thereby reducing the frequency offset. In this way, the optical communication device 10 switches from laser frequency control using a power difference or power ratio, which may cause frequency instability, to laser frequency control based on the frequency offset estimated by the DSP. This switching eliminates frequency instability that accompanies frequency offset reduction.

[0037] 3 is a flow diagram showing the frequency control process in the optical communication device 10. Here, the case where the transmitting side is the optical communication device 10-j and the receiving side is the optical communication device 10-i (i, j = 1, 2, and j ≠ i) is explained. The optical modulation unit 13 of the optical communication device 10-j modulates the frequency f output from the light source unit 11. ｊThe optical communication device 10-j modulates the light with the electrical signal output by the electrical signal generator 12 to generate an optical signal to be transmitted. The optical modulator 13 of the optical communication device 10-j outputs the generated optical signal to the optical transmission path 20-j.

[0038] The optical communication device 10-i on the receiving side performs the process shown in Fig. 3. The optical detection unit 14 of the optical communication device 10-i receives the optical signal transmitted through the optical transmission line 20-j. The optical detection unit 14 detects the frequency f output from the light source unit 11. ｉ The optical detector 14 performs coherent detection of the received optical signal using the LO light. The optical detector 14 outputs the received signal obtained by coherent detection. The ADC 15 converts the electrical received signal output by the optical detector 14 from an analog signal to a digital signal. The digital received signal output by the ADC 15 is branched into two. One branch is output to the power derivation unit 161, and the other branch is output to the compensation unit 163. The power derivation unit 161 calculates the power of the positive frequency component and the negative frequency component in the spectrum of the received signal output by the ADC 15, and derives the power difference or power ratio between them (step S101). Meanwhile, the compensation unit 163 performs frequency offset compensation on the received signal, and the demodulation unit 164 demodulates and decodes the frequency offset-compensated received signal. Furthermore, one or both of the compensation unit 163 and the demodulation unit 164 calculates an estimate of the frequency offset in the received signal.

[0039] The control amount derivation unit 162 determines whether at least one of a first condition that the power difference derived in step S101 is within a predetermined threshold A and a second condition that the power ratio is within a predetermined threshold A' is satisfied (step S102). If the control amount derivation unit 162 determines that both the first condition and the second condition are not satisfied (step S102: NO), the control amount derivation unit 162 performs processing in step S103. Note that in step S102, the control amount derivation unit 162 may determine that both the first condition and the second condition are not satisfied when the received signal spectrum is missing due to ADC band limitation caused by the sampling rate and demodulation cannot be performed in the DSP (the compensation unit 163 and the demodulation unit 164).

[0040] The control amount derivation unit 162 derives a control amount for the light source unit 11 corresponding to the power difference or power ratio calculated in step S101 (step S103). The derivation of the control amount will be described in detail later.

[0041] If the control amount derivation unit 162 determines that one or both of the first condition and the second condition are satisfied (step S102: YES), the control amount derivation unit 162 performs the process of step S104. Note that in step S102, the control amount derivation unit 162 may determine that the frequency offset of the received signal spectrum is small and that at least one of the first condition and the second condition is satisfied if DSP demodulation of the received signal is possible. The control amount derivation unit 162 derives the control amount of the light source unit 11 based on the frequency offset estimated by the DSP, i.e., the compensator 163 and the demodulator 164 (step S104). The derivation of the control amount will be described in detail later.

[0042] The frequency control unit 17 controls the light source unit 11 using the control amount derived in step S103 or step S104, and changes the frequency of the light output from the light source unit 11 (step S105). The optical communication device 10-i receives the next new signal from the optical communication device 10-j, and repeats the process from step S101.

[0043] 4 is a flow diagram showing another frequency control process in the optical communication device 10. In FIG. 4, processes that are the same as those in the frequency control process shown in FIG. 3 are assigned the same reference numerals, and detailed descriptions thereof will be omitted. The receiving optical communication device 10-i receives an optical signal that is output by the transmitting optical communication device 10-j and transmitted through the optical transmission path 20-j. The power derivation unit 161 of the optical communication device 10-i derives a power difference or a power ratio between the power of the positive-side frequency component and the power of the negative-side frequency component in the spectrum of the received signal (step S101).

[0044] If the control amount derivation unit 162 determines that neither the first condition that the power difference derived in step S101 is within a predetermined threshold A nor the second condition that the power ratio is within a predetermined threshold A' is satisfied (step S102: NO), the control amount derivation unit 162 performs the process of step S103. That is, the control amount derivation unit 162 derives a control amount for the light source unit 11 corresponding to the power difference or power ratio calculated in step S101 (step S103). The frequency control unit 17 controls the light source unit 11 using the control amount derived in step S103, and changes the frequency of the light output from the light source unit 11 (step S105). The optical communication device 10-i receives the next new signal from the optical communication device 10-j and repeats the process from step S101.

[0045] If the control amount derivation unit 162 determines that one or both of the first condition and the second condition are satisfied (step S102: YES), it determines whether at least one of a third condition that the power difference derived in step S101 is within a predetermined threshold B and a fourth condition that the power ratio is within a predetermined threshold B' is satisfied (step S201), where threshold B<threshold A and threshold B'<threshold A'.

[0046] If the control amount derivation unit 162 determines that neither the third condition nor the fourth condition is satisfied (step S201: NO), the control amount derivation unit 162 performs the process of step S104. That is, the control amount derivation unit 162 derives the control amount of the light source unit 11 based on the frequency offset estimation amount of the DSP (step S104). The frequency control unit 17 controls the light source unit 11 using the control amount derived in step S104, and changes the frequency of the light output from the light source unit 11 (step S105). The optical communication device 10-i receives the next new signal from the optical communication device 10-j and repeats the process from step S101.

[0047] If the control amount derivation unit 162 determines that one or both of the third condition and the fourth condition are satisfied (step S201: YES), it determines not to perform frequency control of the light source unit 11. The optical communication device 10-i receives the next new signal from the optical communication device 10-j and repeats the processing from step S101.

[0048] Next, a method for deriving the control amount in step S103 will be described. First, a method for deriving the control amount based on a cost function using the power ratio will be described.

[0049] f ｉ is the center frequency of the light source unit 11 of the optical communication device 10-i, a ｉ is the control amount for the light source unit 11 of the optical communication device 10-i. １ and f ２ is the center frequency of the light source unit 11 of each of the opposing optical communication devices 10, a １ and a ２ is the frequency f １ The control amount of frequency f ２ The control amount is the temperature of the light source unit 11 or the amount of current injected into the laser. １ -f ２ ) is the power ratio between the power of the positive frequency component and the power of the negative frequency component in the received signal spectrum. In this case, the cost function is given by the following equation (1).

[0050]

[0051] Control amount a １ The partial derivative of the cost function E with respect to １ ) is expressed as the following equation (2), and the control amount a ２ The partial derivative of the cost function E with respect to ２ ) is expressed as the following equation (3).

[0052]

[0053]

[0054] center frequency f １ The control amount a for the light source unit 11 of the optical communication device 10-1 that outputs １ is the partial differential coefficient (∂E / ∂a １ ) and the step size parameter μ, it is derived as shown in the following equation (4). ２ The control amount a for the light source unit 11 of the optical communication device 10-2 that outputs ２is the partial differential coefficient (∂E / ∂a ２ ) and the step size parameter μ, it is derived as shown in the following equation (5).

[0055]

[0056]

[0057] The control amount derivation unit 162 of the optical communication device 10-1 calculates the control amount a of the light source unit 11 of the device itself using equation (4). １ The control amount derivation unit 162 of the optical communication device 10-2 derives the control amount a of the light source unit 11 of the device itself using equation (5). ２ is derived.

[0058] Next, a specific example of calculation of the partial differential coefficient of the cost function will be described. Here, the control amount a of the cost function using the power ratio is １ The explanation will be given using the partial differential coefficient with respect to the control variable a ２ The same is true for partial derivatives with respect to

[0059] (Calculation example 1) Control amount a １ It may be assumed that the relationship between the frequency offset and the frequency offset, and the relationship between the frequency offset and the positive and negative power ratio in the received signal spectrum, are proportional to each other. In this case, the control amount a １ The partial derivative of the frequency offset with respect to １ -f ２ ) / ∂a １ ), and the partial differential coefficient of the positive / negative power ratio of the received signal spectrum with respect to the frequency offset shown in equation (7) (∂Pr(f １ -f ２ ) / ∂(f １ -f ２ )) will be constant.

[0060]

[0061]

[0062] Therefore, the partial differential coefficient (∂(f １ -f ２ ) / ∂a １ ) and the partial differential coefficient (∂Pr(f １-f ２ ) / ∂(f １ -f ２ )) in advance, the (∂E / ∂a １ ) is Pr(f １ -f ２ ) shown in equation (2). Therefore, the control amount derivation unit 162 uses the value of the positive / negative power ratio of the received signal spectrum detected by the power derivation unit 161 to calculate the partial differential coefficient (E / ∂a １ The control amount deriving unit 162 of the optical communication device 10-1 can uniquely determine the value of the partial differential coefficient (∂E / ∂a １ ) is used to calculate the control variable a １ Calculate.

[0063] Similarly, the partial differential coefficient (∂(f １ -f ２ ) / ∂a ２ ) and the partial differential coefficient (∂Pr(f １ -f ２ ) / ∂(f １ -f ２ )) in advance, the (∂E / ∂a ２ ) is Pr(f １ -f ２ The control amount derivation unit 162 of the optical communication device 10-2 calculates the positive / negative power ratio Pr(f １ -f ２ ) is used to calculate the partial differential coefficient (∂E / ∂a ２ ) and calculate the partial differential coefficient (∂E / ∂a ２ ) is used to calculate the control variable a ２ Calculate.

[0064] (Calculation Example 2) Control amount a １ The relationship between the control variable a and the frequency offset, and the relationship between the frequency offset and the positive / negative power ratio in the received signal spectrum may be assumed to be, for example, a hyperbolic tangent relationship. １and frequency offset, and the relationship between the frequency offset and the positive and negative power ratio in the received signal spectrum are calculated in advance. １ -f ２ ) and the positive / negative power ratio Pr(f １ -f ２ ) is the partial differential coefficient (∂Pr(f １ -f ２ ) / ∂(f １ -f ２ )) and the current control amount a １ The partial differential coefficient (∂(f １ -f ２ ) / ∂a １ ) based on the value of the partial differential coefficient (∂E / ∂a １ The control amount deriving unit 162 of the optical communication device 10-1 can uniquely determine the value of the partial differential coefficient (∂E / ∂a １ ) is used to calculate the control variable a １ Calculate.

[0065] Similarly, the control amount derivation unit 162 of the optical communication device 10-2 calculates the control amount a ２ and the frequency offset, and the relationship between the frequency offset and the positive and negative power ratio in the received signal spectrum are stored in advance. １ -f ２ ) and the positive / negative power ratio Pr(f １ -f ２ ) the partial differential coefficient (∂Pr(f １ -f ２ ) / ∂(f １ -f ２ )) and the current control amount a ２ The partial differential coefficient (∂(f １ -f ２ ) / ∂a ２ ) based on the value of the partial differential coefficient (∂E / ∂a ２ ) and calculate the partial differential coefficient (∂E / ∂a ２ ) is used to calculate the control variable a ２ Calculate.

[0066] Next, a method for deriving a control amount based on a cost function using a power difference will be described.

[0067] As in the case of using the power ratio, f １ and f ２ is the center frequency of the light source unit 11 of each of the opposing optical communication devices 10, a １ and a ２ , frequency f １ The control amount of frequency f ２ The control amount is Pd(f １ -f ２ ) is the power difference between the power of the positive frequency component and the power of the negative frequency component in the received signal spectrum. In this case, the cost function is given by the following equation (8).

[0068]

[0069] Control amount a １ The partial derivative of the cost function E with respect to １ ) is expressed as the following equation (9), and the control amount a ２ The partial derivative of the cost function E with respect to ２ ) is expressed as the following equation (10).

[0070]

[0071]

[0072] center frequency f １ The control amount a for the light source unit 11 of the optical communication device 10-1 that outputs １ is the partial differential coefficient (∂E / ∂a １ ) and the step size parameter μ, it is derived as shown in the following equation (11). ２ The control amount a for the light source unit 11 of the optical communication device 10-2 that outputs ２ is the partial differential coefficient (∂E / ∂a ２ ) and the step size parameter μ, it is derived as shown in the following equation (12).

[0073]

[0074]

[0075] The control amount derivation unit 162 of the optical communication device 10-1 calculates the control amount a of the light source unit 11 of the device itself using equation (11). １ The control amount derivation unit 162 of the optical communication device 10-2 derives the control amount a of the light source unit 11 of the device itself using equation (12). ２ is derived.

[0076] The partial differential coefficient of the cost function using the power difference is calculated by multiplying the control amount a １ That is, in the case of calculation example 1, the control amount derivation unit 162 of the optical communication device 10-1 calculates the partial differential coefficient (∂(f １ -f ２ ) / ∂a １ ) and partial differential coefficient (∂Pd(f １ -f ２ ) / ∂(f １ -f ２ )) and the positive and negative power difference Pd(f １ -f ２ ) is used to calculate the partial differential coefficient (∂E / ∂a １ ) and calculate the partial differential coefficient (∂E / ∂a １ ) is used to calculate the control variable a １ Similarly, the control amount derivation unit 162 of the optical communication device 10-2 calculates the partial differential coefficient (∂(f １ -f ２ ) / ∂a ２ ) and partial differential coefficient (∂Pd(f １ -f ２ ) / ∂(f １ -f ２ )) and the positive and negative power difference Pd(f １ -f ２ ) and the partial differential coefficient (∂E / ∂a ２ ) and calculate the partial differential coefficient (∂E / ∂a ２ ) is used to calculate the control variable a ２ Calculate.

[0077] In the case of calculation example 2, the control amount derivation unit 162 of the optical communication device 10-1 calculates the detected positive and negative power difference Pd(f １ -f ２ ) and the positive and negative power difference Pd(f １ -f ２ ) the partial differential coefficient (∂Pd(f １ -f ２ ) / ∂(f １ -f ２ )) and the current control amount a １ The partial differential coefficient (∂(f １ -f ２ ) / ∂a １ ) and the partial differential coefficient (∂E / ∂a １ ) and calculate the partial differential coefficient (∂E / ∂a １ ) is used to calculate the control variable a １ Similarly, the control amount derivation unit 162 of the optical communication device 10-2 calculates the detected positive and negative power difference Pd(f １ -f ２ ) and the positive and negative power difference Pd(f １ -f ２ ) the partial differential coefficient (∂Pd(f １ -f ２ ) / ∂(f １ -f ２ )) and the current control amount a ２ The partial differential coefficient (∂(f １ -f ２ ) / ∂a ２ ) and the partial differential coefficient (∂E / ∂a ２ ) and calculate the partial differential coefficient (∂E / ∂a ２ ) is used to calculate the control variable a ２ Calculate.

[0078] Next, a method of deriving a control amount so that the value of the cost function approaches 0 will be described based on whether the power of the positive frequency component or the power of the negative frequency component is greater.

[0079] As above, the control amount a ｉ is the control amount for the light source unit 11 of the optical communication device 10-i, and f ｉis the center frequency of the light source unit 11 of the optical communication device 10-i. ｉ is the temperature of the light source unit 11 or the amount of current injected into the laser used as the light source unit 11. １ Increasing the laser frequency f １ becomes larger, and the control amount a １ Decreasing the laser frequency f １ Similarly, the control variable a ２ Increasing the laser frequency f ２ becomes larger, and the control amount a ２ Decreasing the laser frequency f ２ In the following, it is assumed that the optical communication device 10-1 is controlled by the control amount a １ The following description will be given using an example in which the optical communication device 10-2 derives the control amount a ２ The derivation is similar.

[0080] In the power ratio derivation process in step S101 in Fig. 3 or 4, the power derivation unit 161 checks the power of the positive frequency component and the power of the negative frequency component. If the power of the positive frequency component is greater, in step S103 in Fig. 3 or 4, the control amount derivation unit 162 calculates the control amount a １ The change amount Δa １ This increases the frequency f １ On the other hand, if the power of the negative frequency component is large, in step S103 of FIG. 3 or FIG. 4, the control amount derivation unit 162 increases the control amount a １ The change amount Δa １ This reduces the frequency f １ By repeating this procedure in the processing flow shown in FIG. 3 or 4, the cost function E in equation (1) that represents the frequency offset can be made closer to 0.

[0081] Control amount a in one procedure １ The change in Δa １For example, in the case of controlling the range of the assumed maximum frequency offset +25 GHz in the optical communication device 10-1, the control amount for operating the offset of +25 GHz in the design value for controlling the laser is set as a １ max, the control amount for operating the -25 GHz offset is a １ In this case, the amount of change Δa to be increased or decreased in step S103 of one control cycle shown in FIG. １ (absolute value), as an example, |a １ max-a １ min| / 100. １ is the amount of current injected into the laser, the design control amount a of the laser fine tuning step (0.001 GHz) specified by the CMIS (Common Management Interface Specification) is １ The change in the amount of change Δa １ It may also be possible to use the following.

[0082] Next, a method for deriving the control amount in step S104 will be described. In step S104, the control amount for the light source unit 11 is derived using the frequency offset estimated by the DSP. ｉ is the center frequency of the light source unit 11 of the optical communication device 10-i, a ｉ is the control amount for the light source unit 11 of the optical communication device 10-i. １ and f ２ is the center frequency of the light source unit 11 of each of the opposing optical communication devices 10, and a １ and a ２ is the frequency f １ The control amount of frequency f ２ is the control amount.

[0083] For example, the control amount derivation unit 162 calculates the frequency offset (f １ -f ２ ) and the value (∂(f２ -f １ ) / ∂a １ ), (∂(f １ -f ２ ) / ∂a ２ ) and a step size parameter μ, the control amount is calculated by the following equations (13) and (14). The control amount is the temperature of the light source unit 11 or the amount of current injected into the laser.

[0084]

[0085]

[0086] According to the above-described embodiment, when the power difference and power ratio between the power of the positive-side frequency component and the power of the negative-side frequency component of the received signal are larger than a predetermined value, the optical communication device derives a control amount so that the value of a cost function using the power difference or the power ratio as a variable approaches a predetermined cost value corresponding to a frequency offset of 0, and performs frequency control of the light source unit based on the derived control amount. Furthermore, when at least one of the power difference or the power ratio is smaller than a predetermined value, the optical communication device derives a control amount using a frequency offset estimate calculated by the DSP, and performs frequency control of the light source unit based on the derived control amount. Note that the optical communication device can avoid performing frequency control of the light source unit when at least one of the power difference or the power ratio is sufficiently small.

[0087] Furthermore, the optical communication device of this embodiment uses a step size parameter μ to perform frequency control of the light source unit multiple times while checking the power difference or power ratio in the power derivation unit, thereby enabling stable laser frequency control even if the opposing optical communication device performs frequency offset compensation independently.

[0088] Second Embodiment One optical communication device may communicate with multiple optical communication devices. The optical communication device includes an optical communication unit that transmits and receives optical signals to and from each of the multiple optical communication devices with which it communicates. If the optical communication device has only one light source, the multiple optical communication units share a common LO. In this case, the optical communication device must compensate for the frequency offset of received signals from multiple different optical communication devices, which have frequency offsets, so as to ensure transmission performance with the multiple optical communication devices.

[0089] 5 is a diagram showing the configuration of an optical communication system 3 according to the second embodiment. The optical communication system 3 includes M (M is an integer equal to or greater than 3) optical communication devices 30. Each optical communication device 30 transmits and receives optical signals to K (K is an integer equal to or greater than 1 and equal to or less than M-1) other optical communication devices 30. Here, an example will be described in which M=3 and K=2. The M optical communication devices 30 will be referred to as optical communication devices 30-1 to 30-M, respectively.

[0090] The optical communication device 30 includes a light source unit 11 and K optical communication units 32. The optical communication device 30 is a transmission device, and the optical communication units 32 are optical transceivers. The light source unit 11 of the optical communication device 30-m (m is an integer between 1 and M) emits light at a frequency f ｍ 6, one of the two optical communication units 32 of the optical communication device 30-1 faces the optical communication unit 32 of the optical communication device 30-2, and the other faces the optical communication unit 32 of the optical communication device 30-3. Also, one of the two optical communication units 32 of the optical communication device 30-2 faces the optical communication unit 32 of the optical communication device 30-3, and the other faces the optical communication unit 32 of the optical communication device 30-1. Similarly, one of the two optical communication units 32 of the optical communication device 30-3 faces the optical communication unit 32 of the optical communication device 30-1, and the other faces the optical communication unit 32 of the optical communication device 30-2.

[0091] FIG. 6 is a functional block diagram showing an example configuration of an optical communication device 30. In FIG. 6, only functional blocks related to this embodiment are extracted and shown. The optical communication device 30 includes a light source unit 11, K optical communication units 32, a digital signal processing unit 36, and a frequency control unit 37. The K optical communication units 32 are respectively referred to as optical communication units 32-1 to 32-K. The opposing optical communication device 30 connected via the kth (k is an integer between 1 and K) optical communication unit 32-k is referred to as opposing optical communication device 30-k. Furthermore, the opposing optical communication devices 30-1 to 30-K are collectively referred to as opposing optical communication device 30, or when no particular optical communication device is specified. In the following, we will explain an example in which each optical communication unit 32 of the optical communication device 30-m is connected to an optical communication unit 32 of a different opposing optical communication device 30, but each of the multiple optical communication units 32 of the optical communication device 30-m may also be connected to multiple different optical communication units 32 of the same opposing optical communication device 30.

[0092] The light source unit 11 is similar to the light source unit 11 included in the optical communication device 10 of the first embodiment. The local light output from the light source unit 11 is branched into K beams, which are input to the optical communication units 32-1 to 32-K, respectively.

[0093] The optical communication unit 32 includes the electrical signal generation unit 12, optical modulation unit 13, optical detection unit 14, and ADC 15 of the optical communication device 10 of the first embodiment. The electrical signal generation unit 12, optical modulation unit 13, optical detection unit 14, and ADC 15 included in an optical communication unit 32-k (k is an integer equal to or greater than K) will be referred to as the electrical signal generation unit 12-k, the optical modulation unit 13-k, the optical detection unit 14-k, and the ADC 15-k, respectively. Light output from the light source unit 11 to the optical communication unit 32-k is branched, and one branched light is input to the optical modulation unit 13-k, and the other branched light is input to the optical detection unit 14-k. The electrical signal generation unit 12-k, the optical modulation unit 13-k, the optical detection unit 14-k, and the ADC 15-k each perform the same processing as the electrical signal generation unit 12, the optical modulation unit 13, the optical detection unit 14, and the ADC 15 of the first embodiment on the optical signal transmitted between the opposing optical communication device 30-k.

[0094] The digital signal processing unit 36 ​​is a DSP. The digital signal processing unit 36 ​​is realized by, for example, an FPGA or an ASIC. The digital signal processing unit 36 ​​has a power derivation unit 361, a control amount derivation unit 362, K compensation units 163, and K demodulation units 164. The K compensation units 163 are respectively referred to as compensation units 163-1 to 163-K, and the K demodulation units 164 are respectively referred to as demodulation units 164-1 to 164-K.

[0095] The power derivation unit 361 performs the same processing as the power derivation unit 161 of the first embodiment for each of the reception signal spectra received by the optical detection units 14-1 to 14-K. As a result, the power derivation unit 361 calculates the power difference or power ratio between the power of the positive-side frequency component and the power of the negative-side frequency component in the reception signal from each of the opposing optical communication devices 30-1 to 30-K.

[0096] The control amount derivation unit 362 calculates a control amount element for calculating the control amount of the light source unit 11 for each of the opposing optical communication devices 30-1 to 30-K. That is, when the frequency offset with the opposing optical communication device 30-k is larger than a predetermined value, the control amount derivation unit 362 calculates the kth control amount element based on the power difference or power ratio obtained based on the received signal from the opposing optical communication device 30-k. On the other hand, when the frequency offset with the opposing optical communication device 30-k is smaller than a predetermined value, the control amount derivation unit 362 derives the kth control amount element based on an estimated value of the frequency offset calculated by the DSP, that is, by one or both of the compensation unit 163-k and the demodulation unit 164-k of the digital signal processing unit 36. The control amount derivation unit 362 combines the control amount elements calculated for each of the opposing optical communication devices 30-1 to 30-K to derive the control amount of the light source unit 11. An example of a combination of control amount elements is weighted addition.

[0097] The compensator 163-k performs frequency offset compensation on the received signal output by the ADC 15-k. The demodulator 164-k performs predetermined demodulation and decoding processes on the received signal whose frequency offset has been compensated for by the compensator 163-k. In addition, one or both of the compensator 163-k and the demodulator 164-k calculate an estimate of the frequency offset between the optical communication device 30 and the opposing optical communication device 30-k by processing similar to that in the first embodiment. Note that the digital signal processor 36 may not include the compensator 163-k and the demodulator 164-k, and the optical communication unit 32-k may include a DSP (digital signal processor) that includes the compensator 163-k and the demodulator 164-k.

[0098] The frequency control unit 37 controls the light source unit 11 to change the frequency of the light output based on the control amount derived by the control amount derivation unit 362 .

[0099] Next, the operation of the optical communication system 3 will be described. Fig. 7 is a flow diagram showing frequency control processing in the optical communication device 30. Here, an example will be described in which the optical communication device 30-m performs frequency control. In each of the K opposing optical communication devices 30-1 to 30-K opposing the optical communication device 30-m, the optical modulation unit 33 modulates the light output from the light source unit 11 with the electrical signal output from the electrical signal generation unit 12 to generate an optical signal to be transmitted. The generated light is output to the optical transmission path 20.

[0100] The optical communication device 30-m, which receives an optical signal from the opposing optical communication device 30-k (k=1 to K), performs the process shown in Fig. 7. The optical communication device 30-m splits the light output from the light source unit 11 into K parts and outputs them to the optical communication units 32-1 to 32-K. The optical detection unit 14-k detects the frequency f ｍ The optical detection unit 14-k performs coherent detection of the optical signal received from the opposing optical communication device 30-k using the LO light of the optical detection unit 14-k. The optical detection unit 14-k outputs the received signal obtained by coherent detection. The ADC 15-k converts the received electrical signal output by the optical detection unit 14-k from an analog signal to a digital signal. The digital received signal output by the ADC 15-k is branched into two, one of which is output to the power derivation unit 361 and the other of which is output to the compensation unit 163-k.

[0101] The compensators 163-k, k=1 to N, perform frequency offset compensation on the received signal from the opposing optical communication device 30-k, and the demodulators 164-k demodulate and decode the received signal that has been frequency offset compensated by the compensators 163-k. Furthermore, one or both of the compensators 163-k and the demodulators 164-k calculate an estimate of the frequency offset between the optical communication device 30-m and the opposing optical communication device 30-k.

[0102] The power derivation unit 361 calculates the power of the positive frequency components and the negative frequency components in the spectrum of the received signal output by the ADC 15-k for each of k=1 to K, and derives the power difference or power ratio between them (step S301).

[0103] The control amount derivation unit 362 performs the processes of steps S303 to S305 for each of k = 1 to K (step S302). Note that the control amount derivation unit 362 may perform the processes of steps S303 to S305 for each of k = 1 to K sequentially, or may perform some or all of them in parallel.

[0104] The control amount derivation unit 362 determines whether either a first condition, that is, the power difference derived for the reception spectrum of the optical communication unit 32-k in step S301 is within a predetermined threshold A, or a second condition, that is, the power ratio derived for the reception spectrum of the optical communication unit 32-k is within a predetermined threshold A', is satisfied (step S303). If the control amount derivation unit 362 determines that both the first and second conditions are not satisfied (step S303: NO), it performs processing of step S304. Note that in step S303, the control amount derivation unit 362 may determine that both the first and second conditions are not satisfied if the reception signal spectrum is missing due to ADC band limiting caused by the sampling rate and demodulation cannot be performed in the DSP (compensation unit 163-k and demodulation unit 164-k). The control amount derivation unit 362 derives the kth control amount element for the light source unit 11 corresponding to the power difference or power ratio calculated in step S301 (step S304). The derivation of the controlled variable elements will be described in detail later.

[0105] On the other hand, if the control amount derivation unit 362 determines that at least one of the first condition and the second condition is satisfied (step S303: YES), it performs the process of step S305. Note that in step S303, the control amount derivation unit 362 may determine that the frequency offset of the received signal spectrum is small and that at least one of the first condition and the second condition is satisfied if the received signal can be demodulated in the DSP (compensation unit 163-k and demodulation unit 164-k). The control amount derivation unit 362 derives the kth control amount element of the light source unit 11 based on the frequency offset estimate calculated by the compensation unit 163-k and the demodulation unit 164-k (step S305). Details of the derivation of the control amount will be described later.

[0106] After completing the calculation of the control amount elements for all k=1 to K, the control amount derivation unit 362 combines the K control amount elements calculated for each of the opposing optical communication devices 30-1 to 30-K to derive the control amount for the light source unit 11 (step S306). The frequency control unit 37 controls the light source unit 11 using the control amount derived in step S306, and changes the frequency of the light output from the light source unit 11 (step S307). The optical communication device 30-m receives the next new signal from the opposing optical communication device 30-1 to 30-K, and repeats the process from step S301.

[0107] Fig. 8 is a flow diagram showing another frequency control process in the optical communication device 30. In Fig. 8, the same processes as those in the frequency control process shown in Fig. 7 are assigned the same reference numerals, and detailed description thereof will be omitted. As in Fig. 7, an example will be described in which the optical communication device 30-m performs frequency control. The optical communication device 30-m performs the process of step S301 in Fig. 7, and derives, for each of k = 1 to K, the power difference or power ratio between the power of the positive-side frequency component and the power of the negative-side frequency component in the spectrum of the received signal from the opposing optical communication device 30-k (step S301).

[0108] The control quantity derivation unit 362 determines whether a fifth condition, that all of the K power differences derived in step S301 are within a predetermined threshold A, or a sixth condition, that all of the K power ratios derived in step S301 are within a predetermined threshold A', is satisfied (step S401). If the control quantity derivation unit 362 determines that at least one of the fifth condition and the sixth condition is satisfied (step S401: NO), the control quantity derivation unit 362 performs the determination process of step S402. That is, the control quantity derivation unit 362 determines whether a seventh condition, that all of the K power differences derived in step S301 are within a predetermined threshold B, or an eighth condition, that all of the K power ratios derived in step S301 are within a predetermined threshold B', is satisfied (step S402). Here, threshold B<threshold A, and threshold B'<threshold A'. If the control amount derivation unit 362 determines that at least one of the seventh condition and the eighth condition is satisfied (step S402: YES), it determines not to control the light source unit 11. The optical communication device 30-m receives the next new signal from the opposing optical communication devices 30-1 to 30-K and repeats the processing from step S301.

[0109] On the other hand, if the control amount derivation unit 362 determines that both the fifth condition and the sixth condition are not satisfied (step S401: YES), or if it determines that both the seventh condition and the eighth condition are not satisfied (step S402: NO), it performs processing similar to the processing from step S302 onwards in Figure 7.

[0110] Next, the control amount element derivation method in step S304 and step S305 and the control amount derivation method in step S306 will be described. The K optical communication units 32-1 to 32-K of the optical communication device 30-m are each connected to K opposing optical communication units 32, and share the light source unit 11. By the processing in FIG. 7 or FIG. 8, the derivation of the control amount element of the light source unit 11 based on the power difference or power ratio and the derivation of the control amount element of the light source unit 11 based on the frequency offset estimated by the DSP are mixed among the optical communication units 32-1 to 32-K of the optical communication device 30-m. For the reception spectrum received by each optical communication unit 32-k of the optical communication device 30-m from the opposing optical communication device 30-k, the control amount element derived in step S304 or step S305 in FIG. 7 or FIG. 8 is calculated as Δc ｋ Let's say.

[0111] First, in step S304, the optical communication device 30-m calculates the control amount element Δc using the power ratio. ｋ A method for deriving Δf will be described. ｋ is the frequency offset of the spectrum of the signal received by the optical communication unit 32-k (the frequency offset between the optical communication device 30-m and the opposing optical communication device 30-k), and a ｍ is the control amount of the light source unit 11 of the optical communication device 30-m, and Pr ｋ (Δf ｋ ) is the power ratio between the positive and negative sides of the spectrum of the signal received by the optical communication unit 32-k of the optical communication device 30-m. ｋ is a cost function for the opposing optical communication device 30-k. ｋ is expressed as in equation (15).

[0112]

[0113] Control amount element Δc in the optical communication device 30-m ｋ is calculated by the following equations (16) and (17): μ is a step size parameter of a predetermined value.

[0114]

[0115]

[0116] The partial differential coefficient can be calculated in the same manner as in the calculation example of the partial differential coefficient in the first embodiment. For example, ｍ and frequency offset Δf ｋ and the frequency offset Δf ｋ and the positive and negative power ratio Pr in the received signal spectrum ｋ (Δf ｋ In this case, the control amount a in the formulas (6) and (7) in the calculation example 1 of the partial differential coefficient is assumed to be proportional to １ , (f １ -f ２ ), Pr(f １ -f ２ ) is the control amount a ｍ , Δf ｋ , Pr ｋ (Δf ｋ ) can be replaced with the partial differential coefficient (∂Δf ｋ / ∂a ｍ ), and the partial differential coefficient (∂Pr ｋ (Δf ｋ ) / ∂(Δf ｋ )) is a constant known value. Therefore, the control amount derivation unit 362 calculates the positive / negative power ratio Pr ｋ (Δf ｋ ) is used to calculate the partial differential coefficient (∂e ｋ / ∂a ｍ ) can be uniquely determined.

[0117] Similarly, in the calculation example 2 of the partial differential coefficient, the control amount a １ , (f １ -f ２ ), Pr(f １ -f ２ ) is the control amount a ｍ , Δf ｋ , Pr ｋ (Δf ｋ ) can be replaced with the control variable a ｍ The relationship between the frequency offset and the positive / negative power ratio in the received signal spectrum is calculated in advance. ｋ (Δf ｋ) and the positive / negative power ratio Pr ｋ (Δf ｋ ) from the partial differential coefficient (∂Pr ｋ (Δf ｋ ) / ∂(Δf ｋ )) and the current control amount a ｍ The partial differential coefficient (∂(Δf ｋ ) / ∂a ｍ ) based on the value of the partial differential coefficient (∂e ｋ / ∂a ｍ ) can be uniquely determined.

[0118] Next, in step S304, the control amount element Δc is calculated using the power difference. ｋ The method for deriving Pd will be explained. ｋ (Δf ｋ ) is the power difference between the positive and negative sides of the spectrum of the signal received by the optical communication unit 32-k of the optical communication device 30. ｋ is a cost function for the opposing optical communication device 30-k. ｋ is expressed as in equation (18).

[0119]

[0120] Control amount element Δc in the optical communication device 30-m ｋ is calculated by the following equations (19) and (20).

[0121]

[0122]

[0123] As in the case of the power ratio, the partial differential coefficient is calculated by the control amount a １ , (f １ -f ２ ), Pd(f １ -f ２ ) is the control amount a ｍ , Δf ｋ , Pd ｋ (Δf ｋ That is, in the case of calculation example 1, the control amount derivation unit 362 can calculate the control amount by replacing the partial differential coefficient (∂Δf ｋ / ∂aｍ ) and partial differential coefficient (∂Pd ｋ (Δf ｋ ) / ∂(Δf ｋ )) and the positive and negative power difference Pd of the received signal spectrum from the opposing optical communication device 30-k ｋ (Δf ｋ ) is used to calculate the partial differential coefficient (∂e ｋ / ∂a ｍ ) is calculated, and the calculation result is used to calculate Δc ｋ In the case of calculation example 2, the control amount derivation unit 362 calculates the detected positive and negative power difference Pd ｋ (Δf ｋ ) and the positive and negative power difference Pd ｋ (Δf ｋ ) from the partial differential coefficient (∂Pd ｋ (Δf ｋ ) / ∂(Δf ｋ )) and the current control amount a ｍ The partial differential coefficient (∂(Δf ｋ ) / ∂a ｍ ) based on the value of the partial differential coefficient (∂e ｋ / ∂a ｍ ) is calculated, and the calculation result is used to calculate Δc ｋ Calculate.

[0124] Next, in step S305, the control variable element Δc is calculated using the frequency offset estimate of the DSP. ｋ A method for deriving the control amount element Δc in the optical communication device 30-m will be described. ｋ is the frequency offset amount Δf estimated by the compensator 163-k and the demodulator 164-k. ｋ is used to calculate the following equation (21).

[0125]

[0126] In step S306, the control amount derivation unit 362 of the optical communication device 30-m calculates the control amount element Δc ｋ Finally, the control amount a of the light source unit 11 is calculated by the following equation (22): ｍ Calculate.

[0127]

[0128] In addition, h ｋ is the weighting value. ｋ may be determined in consideration of the control amount element derivation method, the line bandwidth, the penalty amount (chromatic dispersion amount, etc.) according to the transmission distance, or may be constant regardless of k.

[0129] As described above, in the second embodiment, the optical communication device calculates the same control amount as in the first embodiment for each transceiver (optical communication unit 32) that shares the light source, sets the control amount element, and calculates the weighted sum ΣΔc ｋ h ｋ The optical communication device calculates a new control amount by adding the calculated sum to the control amount currently being used. The optical communication device repeats this process to update the control amount.

[0130] (Third Embodiment) In the third embodiment, the temperature control for changing the laser frequency or the control range of the laser current amount is limited. This allows the laser frequency to fall within a certain frequency range, thereby stabilizing the frequency. The following description of the third embodiment will focus on the differences from the first embodiment, but these differences may also be applied to the second embodiment.

[0131] The configuration of the optical communication system of the third embodiment is similar to that of the optical communication system 1 of the first embodiment shown in Fig. 1. The control amount derivation unit 162 of the optical communication device 10 of the third embodiment differs from the control amount derivation unit 162 of the optical communication device 10 of the first embodiment in the following points regarding the temperature control method or the laser current amount control method.

[0132] (Control Method 1) First, the control amount derivation unit 162 calculates the initial laser controlled temperature or the laser controlled temperature after the frequency offset has converged as a ０ Store it as a ０may be the initial laser current amount or the laser current amount after frequency offset convergence. As in the first embodiment, the control amount derivation unit 162 derives the control amount of the controlled temperature based on the positive and negative power difference or the positive and negative power ratio of the spectrum, or derives the control amount of the controlled temperature based on the frequency offset estimation by the DSP. Alternatively, as in the first embodiment, the control amount derivation unit 162 derives the control amount of the laser current amount based on the positive and negative power difference or the positive and negative power ratio of the spectrum, or derives the control amount of the laser current amount based on the frequency offset estimation by the DSP. ０ The change from a is defined as Δa. ０ +Δa is a ０ a is a predetermined temperature range from ０ ±a ｌｉｍｉｔ If it is within the range, the laser frequency control is performed, and if it is not within the range, the control is not performed.

[0133] (Control Method 2) The power derivation unit 161 calculates the initial laser controlled temperature or the laser controlled temperature after the frequency offset convergence by a ０ Store it as a ０ may be the initial laser current amount or the laser current amount after frequency offset convergence. The control amount derivation unit 162 derives the control amount of the controlled temperature based on the positive and negative power difference of the spectrum using an equation obtained by adding a regularization term to equations (11) and (12), or derives the control amount of the controlled temperature or the laser current amount based on the positive and negative power ratio using an equation obtained by adding a regularization term to equations (4) and (5). Alternatively, the control amount derivation unit 162 derives the control amount of the controlled temperature or the laser current amount based on the frequency offset estimation by the DSP using an equation obtained by adding a regularization term to equation (13). The regularization term includes a ０ By adding the regularization term, a ０ a within a predetermined temperature range or a predetermined current amount range ０ ±a ｌｉｍｉｔ Derive the control amount that fits within

[0134] 9 is a diagram showing an example of a change in frequency of the light source unit 11 in the opposing optical communication device 10. The opposing optical communication devices 10-1 and 10-2 each control the laser frequency using a control amount calculated by the above-mentioned control method 1 or control method 2. Symbol L1 indicates a change in frequency of the light source unit 11 of the optical communication device 10-1, and symbol L2 indicates a change in frequency of the light source unit 11 of the optical communication device 10-2. The initial laser controlled temperature at the initial frequency of the optical communication device 10-1 is a ０ The initial laser temperature control temperature at the initial frequency of the optical communication device 10-2 is a ０ The temperature control range of the laser of the optical communication device 10-1 is a ０ -a ｌｉｍｉｔ From a ０ +a ｌｉｍｉｔ The temperature control range of the laser of the optical communication device 10-2 is a ０ '-a ｌｉｍｉｔ From a ０ '+a ｌｉｍｉｔ is.

[0135] Initially, the frequency offsets of the optical communication device 10-1 and the optical communication device 10-2 converge once. That is, the laser frequency of the optical communication device 10-1 and the laser frequency of the optical communication device 10-2 approach the same value, although their initial values ​​are different. However, for one of the optical communication device 10-1 and the optical communication device 10-2, when the frequency offset is 0, the power difference is 0 or the power ratio is 1, and for the other, when the frequency offset is a value other than 0, the power difference is 0 or the power ratio is 1. In such a case, the laser frequency of the optical communication device 10-1 indicated by the symbol L1 and the laser frequency of the optical communication device 10-2 indicated by the symbol L2 rise from the value when the frequency offset has converged. However, when the laser frequency is controlled by the control amount a ０ -a ｌｉｍｉｔWhen the laser frequency reaches a frequency corresponding to the control amount a, the optical communication device 10-1 stops the laser frequency control. The optical communication device 10-2 may continue the laser frequency control, but the laser frequency of the optical communication device 10-2 is controlled to be close to the laser frequency of the optical communication device 10-1. Alternatively, the laser frequency of the optical communication device 10-1 indicated by the symbol L1 and the laser frequency of the optical communication device 10-2 indicated by the symbol L2 may decrease from the value when the frequency offset has converged, and the laser frequency may decrease to the value corresponding to the control amount a. ０ '+a ｌｉｍｉｔ When the laser frequency of the optical communication device 10-1 reaches a frequency corresponding to the laser frequency of the optical communication device 10-2, the optical communication device 10-2 stops the laser frequency control. The optical communication device 10-1 may continue the laser frequency control, but the laser frequency of the optical communication device 10-1 is controlled to be close to the laser frequency of the optical communication device 10-2.

[0136] Fig. 10 is a flow diagram showing frequency control processing in the optical communication device 10 using control method 1. In Fig. 10, the same processes as those in the first embodiment shown in Fig. 3 are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0137] The control amount derivation unit 162 of the receiving-side optical communication device 10-i calculates the initial value a ０ and stores the power difference or power ratio between the power of the positive-side frequency component and the power of the negative-side frequency component in the spectrum of the received signal (step S101). If the control variable derivation unit 162 determines that neither the first condition, that the power difference is within a predetermined threshold A, nor the second condition, that the power ratio is within a predetermined threshold A', is satisfied (step S102: NO), the control variable derivation unit 162 derives a control variable for the light source unit 11 corresponding to the power difference or power ratio (step S103). If the control variable derivation unit 162 determines that at least one of the first condition and the second condition is satisfied (step S102: NO), the control variable derivation unit 162 derives a control variable for the light source unit 11 based on the frequency offset estimated by the compensator 163 and the demodulator 164 (step S104).

[0138] The optical communication device 10-i performs the process of step S502 after the process of step S103 or step S104. That is, the control amount derivation unit 162 calculates the control amount a calculated in step S103 or the control amount a calculated in step S104. ０ +Δa is a ０ ±a ｌｉｍｉｔ The control amount derivation unit 162 determines whether the value is within a range of a ０ +Δa is a ０ ±a ｌｉｍｉｔ If it is determined that the value is within (step S502: YES), the calculated control amount is output to the frequency control unit 17. The frequency control unit 17 controls the light source unit 11 based on the control amount acquired from the control amount derivation unit 162 (step S105). The optical communication device 10-i receives the next new signal from the optical communication device 10-j and repeats the processing from step S101.

[0139] On the other hand, the control amount derivation unit 162 calculates a ０ +Δa is a ０ ±a ｌｉｍｉｔ If it is determined that the value is not within the range (step S502: NO), it determines not to perform frequency control of the light source unit 11. The optical communication device 10-i receives the next new signal from the optical communication device 10-j and repeats the process from step S101.

[0140] Fig. 11 is a flow diagram showing frequency control processing in the optical communication device 10 using control method 2. In Fig. 11, the same processes as those in the first embodiment shown in Fig. 3 and those using control method 1 shown in Fig. 10 are denoted by the same reference numerals, and detailed descriptions thereof will be omitted.

[0141] The control amount derivation unit 162 of the receiving-side optical communication device 10-i calculates the initial value a ０ The optical communication device 10-i on the receiving side receives the optical signal output from the optical communication device 10-j on the transmitting side and transmitted through the optical transmission path 20-j. The power derivation unit 161 of the optical communication device 10-i derives the power difference or power ratio between the power of the positive frequency component and the power of the negative frequency component in the spectrum of the received signal (step S101).

[0142] When the control amount derivation unit 162 determines that neither the first condition that the power difference is within a predetermined threshold A nor the second condition that the power ratio is within a predetermined threshold A' is satisfied (step S102: NO), the control amount derivation unit 162 performs the process of step S601. That is, the control amount derivation unit 162 derives the control amount of the light source unit 11 corresponding to the power difference or power ratio calculated in step S101 using an equation obtained by adding a regularization term to the equation used to derive the control amount in the first embodiment (step S601). For example, the control amount derivation unit 162 uses the following equation (23) instead of equations (4) and (11). Note that r is a regularization term coefficient and is set arbitrarily.

[0143]

[0144] On the other hand, when the control amount derivation unit 162 determines that at least one of the first condition and the second condition is satisfied (step S102: YES), the control amount derivation unit 162 performs the process of step S602. That is, the control amount derivation unit 162 derives the control amount of the light source unit 11 by the following equation (24) based on the frequency offset estimation amount of the DSP (step S602).

[0145]

[0146] The frequency control unit 17 controls the light source unit 11 using the control amount derived in step S601 or step S602, and changes the frequency of the light output from the light source unit 11 (step S105). The optical communication device 10-i receives the next new signal from the optical communication device 10-j, and repeats the processing from step S101.

[0147] According to the above-described embodiment, the optical communication system includes a plurality of optical communication devices connected by optical transmission paths. Each optical communication device includes a light source unit, an optical detector unit, a power derivation unit, a digital signal processing unit, a control variable derivation unit, and a frequency control unit. The optical detector unit coherently detects a received optical signal using light output from the light source unit. The power derivation unit derives a power difference or a power ratio between the power of a frequency component on the positive side of a reference frequency and the power of a frequency component on the negative side of the reference frequency in a signal spectrum of the received signal obtained by coherent detection. The digital signal processing unit performs an estimation process to estimate a frequency offset of the received signal obtained by coherent detection and a compensation process to compensate the received signal for the estimated frequency offset. If the derived power difference or power ratio is greater than a predetermined condition, the control variable derivation unit derives a first control variable for changing the frequency of light output from the light source unit based on the derived power difference or power ratio. That is, if the derived power difference or power ratio is greater than a predetermined condition, the control quantity derivation unit sets the derived power difference or power ratio as a parameter value and derives a first control quantity so that the value of the cost function, which would be a predetermined value in the absence of a frequency offset, approaches the predetermined value. Also, if the derived power difference or power ratio is smaller than the predetermined condition, the control quantity derivation unit performs control quantity derivation processing to derive the first control quantity based on the frequency offset amount estimated by the digital signal processing unit. The frequency control unit controls the light source unit using the derived first control quantity to change the frequency of light output by the light source unit.

[0148] The optical communication device may further include an optical modulation section that modulates the light output by the light source section to generate an optical signal and transmits the generated optical signal to another opposing optical communication device.

[0149] The control amount derivation unit may derive the first control amount so that the frequency of the light output from the light source unit falls within a predetermined range.

[0150] The optical communication device may include a plurality of optical detection units that each receive a different optical signal. The power derivation unit derives a power difference or a power ratio for each of the plurality of optical detection units regarding the received signals obtained by the detection units through coherent detection. The digital signal processing unit performs estimation processing and compensation processing for each of the plurality of optical detection units. The control amount derivation unit derives a second control amount for changing the frequency of light output by the light source unit based on a weighted sum of first control amounts derived by performing a control amount derivation processing for each of the plurality of optical detection units. The frequency control unit controls the light source unit using the derived second control amount to change the frequency of light output by the light source unit.

[0151] Although the embodiments of the present invention have been described in detail above with reference to the drawings, the specific configurations are not limited to these embodiments, and include designs within the scope of the present invention that do not deviate from the gist of the present invention.

[0152] The present invention is applicable to optical communication systems, especially digital coherent communication systems.

[0153] 1, 3 Optical communication system 10-1, 10-2, 30, 30-1, 30-2, 30-3 Optical communication device 11 Light source unit 12 Electrical signal generation unit 13 Optical modulation unit 14 Optical detection unit 15 Analog-to-digital conversion unit 16 Digital signal processing unit 17 Frequency control unit 20, 20-1, 20-2 Optical transmission path 32, 32-1, 32-2 Optical communication unit 36 ​​Digital signal processing unit 37 Frequency control unit 161 Power derivation unit 162 Control amount derivation unit 163, 163-1, 163-2 Compensation unit 164, 164-1, 164-2 Demodulation unit 361 Power derivation unit 362 Control amount derivation unit

Claims

a light source unit; an optical detection unit that performs coherent detection on a received optical signal using light output from the light source unit; a power derivation unit that derives a power difference or a power ratio between the power of a frequency component on the positive side with respect to a reference frequency and the power of a frequency component on the negative side with respect to the reference frequency in the signal spectrum of the received signal obtained by the coherent detection; a digital signal processing unit that performs an estimation process to estimate a frequency offset amount of the received signal obtained by the coherent detection and a compensation process to compensate for the estimated frequency offset amount of the received signal; a control variable derivation unit that performs a control variable derivation process to derive a first control variable for changing the frequency of light output by the light source unit based on the derived power difference or power ratio if the derived power difference or power ratio is greater than a predetermined condition, and to derive the first control variable based on the frequency offset amount estimated by the digital signal processing unit if the derived power difference or power ratio is smaller than the predetermined condition; and a frequency control unit that controls the light source unit using the derived first control variable to change the frequency of light output by the light source unit. An optical communication device comprising:

2. The optical communication device according to claim 1, wherein the control amount derivation unit derives the first control amount so that the frequency of the light output from the light source unit falls within a predetermined range.

3. The optical communication device according to claim 1, wherein, when the derived power difference or the power ratio is greater than a predetermined condition, the control variable derivation unit sets the derived power difference or the power ratio as a parameter value, and derives the first control variable so that the value of a cost function that takes a predetermined value when there is no frequency offset approaches the predetermined value.

4. The optical communication device according to claim 1, wherein the optical communication device comprises a plurality of optical detection units each receiving a different optical signal, the power derivation unit derives the power difference or the power ratio for the received signal obtained by the optical detection unit through coherent detection for each of the plurality of optical detection units, the digital signal processing unit performs the estimation process and the compensation process for each of the plurality of optical detection units, the control quantity derivation unit derives a second control quantity for changing the frequency of the light output by the light source unit based on a weighted sum of the first control quantities derived by performing a control quantity derivation process for each of the plurality of optical detection units, and the frequency control unit controls the light source unit using the derived second control quantity to change the frequency of the light output by the light source unit.

5. A device having a plurality of optical communication devices connected by optical transmission paths, the optical communication devices comprising: a light source unit; an optical detection unit that performs coherent detection on a received optical signal using light output from the light source unit; a power derivation unit that derives a power difference or a power ratio between the power of a frequency component on the positive side with respect to a reference frequency and the power of a frequency component on the negative side with respect to the reference frequency in the signal spectrum of the received signal obtained by the coherent detection; a digital signal processing unit that performs an estimation process to estimate a frequency offset amount of the received signal obtained by the coherent detection and a compensation process to compensate for the estimated frequency offset amount in the received signal; and a control quantity derivation unit that, if the derived power difference or power ratio is greater than a predetermined condition, derives a control quantity for changing the frequency of light output by the light source unit based on the derived power difference or power ratio, and, if the derived power difference or power ratio is smaller than the predetermined condition, performs a control quantity derivation process to derive the control quantity based on the frequency offset amount estimated by the digital signal processing unit. a frequency control unit that controls the light source unit using the derived control amount to change the frequency of light output from the light source unit.

6. A frequency offset reduction method comprising: an optical detection step of coherently detecting a received optical signal using light output from a light source unit; a power derivation step of deriving a power difference or a power ratio between the power of a frequency component on the positive side with respect to a reference frequency and the power of a frequency component on the negative side with respect to the reference frequency in the signal spectrum of the received signal obtained by the coherent detection; an estimation step of estimating a frequency offset amount of the received signal obtained by the coherent detection; a compensation step of compensating for the estimated frequency offset amount in the received signal; a control quantity derivation step of performing a control quantity derivation process of deriving a control quantity for changing the frequency of light output by the light source unit based on the derived power difference or power ratio if the derived power difference or power ratio is greater than a predetermined condition, and deriving the control quantity based on the frequency offset amount estimated in the estimation step if the derived power difference or power ratio is smaller than the predetermined condition; and a frequency control step of controlling the light source unit using the derived control quantity to change the frequency of light output by the light source unit.

Images