Method and apparatus for compensating dispersion of optical fiber communication signals

Abstract

To provide a dispersion compensation method and a dispersion compensation device of an optical fiber communication signal which can contain a short / middle-range high-capacity IM-DD optical fiber transmission system in a realistic circuit scale.SOLUTION: A dispersion compensation method of an optical fiber communication signal by a dispersion compensation device 1 of the optical fiber communication signal comprises: an optical pre-processing step in which an optical pre-processing circuit 3 branches an optical signal transmitted through an optical fiber 5 and causes the branched optical signals to interfere with each other to obtain the optical signals after pre-processing; and a post-processing step in which an electric dispersion compensation circuit 7 performs dispersion compensation processing to the optical signals after pre-processing.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] This invention relates to a method for compensating dispersion of an optical fiber communication signal and a dispersion compensation device.

Background Art

[0002] In recent years, the demand for medium- and short-distance (20 to 80 km) high-capacity optical fiber transmission systems for data center interconnection networks and mobile access network front-haul has been increasing. For this reason, the extension of the optical Ethernet (registered trademark) standard such as 400GBASE-ER / ZR has been actively discussed. In intensity modulation-direct detection (IM-DD) optical fiber transmission exceeding 20 km, it is desirable to use the wavelength band of 1550 nm with low fiber propagation loss. On the other hand, in such optical fiber transmission, signal distortion caused by fiber dispersion becomes a problem. Since the influence of dispersion is proportional to the square of the optical signal bandwidth, it becomes a main limiting factor for the transmission distance especially in broadband IM-DD transmission exceeding 50 Gbps / lane. For this reason, various dispersion compensation methods have been studied in the above-mentioned discussion on extension.

[0003] As a method for compensating dispersion of an optical fiber communication signal, an electrical dispersion compensation technique is known. For example, Japanese Patent Application Laid-Open No. 2019-514312 describes an electrical dispersion compensation system and a dispersion compensation method using the system. The electrical dispersion compensation technique equalizes a communication path by digital signal processing. However, in broadband IM-DD transmission of 400GBASE-ER or more, the circuit scale of the electrical dispersion compensation system tends to increase significantly. In reports at the paper level, an equalizer with nearly 100 taps is often adopted for the electrical dispersion compensation system.

[0004] Therefore, for example, a method for compensating dispersion of an optical fiber communication signal and a dispersion compensation device that can implement a medium- and short-distance high-capacity IM-DD optical fiber transmission system such as broadband transmission exceeding 50 Gbps / lane and 40 km (ER) / 80 km (ZR) with a realistic circuit scale are desired.

Prior Art Documents

[0005] [Patent Document 1] Special Publication No. 2019-514312 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] The purpose of this invention is to provide a dispersion compensation method and dispersion compensation device for optical fiber communication signals that can implement a short-to-medium distance high-capacity IM-DD optical fiber transmission system with a realistic circuit scale. [Means for solving the problem]

[0007] The above problems can be basically solved by using a simple optical preprocessing circuit to remove only the components of the signal distortion due to dispersion that are particularly contributing to the increased size of the electrical dispersion compensation circuit, thereby simplifying the electrical dispersion compensation circuit.

[0008] The first invention relates to a method for compensating for dispersion in optical fiber communication signals. This method includes an optical pre-processing step and a post-processing step. The optical preprocessing step is a process in which the optical preprocessing circuit 3 branches and interferes with the optical signal transmitted through the optical fiber 5 to obtain the preprocessed optical signal. The post-processing step is a step in which the electrical dispersion compensation circuit 7 performs dispersion compensation processing on the pre-processed optical signal.

[0009] The following invention relates to a dispersion compensation device 1 for optical fiber communication signals. The optical fiber communication signal dispersion compensation device 1 includes an optical preprocessing circuit 3, an electrical dispersion compensation circuit 7, and a photoelectric processing circuit optimization unit 9. The optical preprocessing circuit 3 is an element for obtaining a preprocessed optical signal by branching and interfering with the optical signal transmitted through the optical fiber 5. The electrical dispersion compensation circuit 7 is an element for performing dispersion compensation processing on the pre-processed optical signal. The photoelectric processing circuit optimization unit 9 is an element for optimizing the driving conditions of the optical preprocessing circuit 3 and the electrical dispersion compensation circuit 7 based on the quality of the output signal from the electrical dispersion compensation circuit 7 or a prediction of the quality of the output signal by an analysis model. [Effects of the Invention]

[0010] This invention provides a dispersion compensation method and dispersion compensation device for optical fiber communication signals that enable the implementation of a short-to-medium distance high-capacity optical fiber transmission system with a realistic circuit scale. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 is a block diagram of a dispersion compensation device for optical fiber communication signals. [Figure 2] Figure 2 is a flowchart illustrating a dispersion compensation method for optical fiber communication signals. [Figure 3] Figure 3(a) is a schematic diagram of a photoelectric-integrated dispersion compensation system. Figure 3(b) shows a single-tap delay line, which is an example of an optical preprocessing circuit. φ represents the optical phase shift and D represents the optical delay. [Figure 4] Figure 4 is a conceptual diagram showing the experimental setup in the embodiment. In the embodiment, a single-tap delay line is used as the optical preprocessing circuit, and a feedforward equalizer is used as the electrical dispersion compensation circuit. In the figure, PPG represents a pulse pattern generator, MZM represents a Mach-Zehnder modulator, SMF represents a single-mode optical fiber, VOA represents a variable optical attenuator, EDFA represents an optical amplifier, FSR represents a free spectral region, OBPF represents an optical bandpass filter, PD represents a photodetector, ADC represents an analog-to-digital converter, and FFE represents a feedforward equalizer. [Figure 5]Figure 5(a) is a graph that replaces a diagram showing the frequency response of a 50km IM-DD optical fiber transmission system with and without a single-tap delay line. Figure 5(b) is a graph that replaces a diagram showing the pole-zero plot of a 50km IM-DD optical fiber transmission system without a single-tap delay line. Figure 5(c) is a graph that replaces a diagram showing the pole-zero plot of a 50km IM-DD optical fiber transmission system with a single-tap delay line. [Figure 6] Figure 6 shows the experimental results. Figure 6(a) shows the bit error rate (BER) versus the number of electrical compensation circuit taps after 50km SMF transmission of a 112Gb / sPAM4 signal using an optoelectronic dispersion compensation circuit. Figure 6(b) shows the BER versus PD input optical power after 50km SMF transmission of a 112Gb / sPAM4 signal with and without an optoelectronic dispersion compensation circuit. Figure 6(c) shows the BER versus the number of electrical compensation circuit taps for 50km and 80km 60Gb / sPAM2 transmissions using an optoelectronic dispersion compensation circuit. [Modes for carrying out the invention]

[0012] The embodiments for carrying out the present invention will be described below with reference to the drawings. The present invention is not limited to the embodiments described below, but also includes modifications made to the embodiments described below to the extent that is obvious to those skilled in the art.

[0013] Figure 1 is a block diagram of a dispersion compensation device for optical fiber communication signals. As shown in Figure 1, the dispersion compensation device 1 for optical fiber communication signals includes an optical preprocessing circuit 3, an electrical dispersion compensation circuit 7, and a photoelectric processing circuit optimization unit 9.

[0014] Dispersion compensation device 1 for optical fiber communication signals The optical fiber communication signal dispersion compensation device 1 is a device for compensating for signal distortion caused by fiber color dispersion. The optical fiber communication signal dispersion compensation device 1 can preferably be used for optical fiber transmission systems of short-to-medium distance high capacity or greater. Examples of the transmission distance by optical fiber are 20 km or more and 1000 km or less, and may also be 40 km or more and 500 km or less, or 40 km or more and 100 km or less, or 40 km or more and 80 km or less. Examples of the bandwidth of optical fiber communication are 10 Gbps / lane or more and 1000 Gbps / lane or less, and may also be 20 Gbps / lane or more and 500 Gbps / lane or less, or may be 40 Gbps / lane or more and 500 Gbps / lane or less, or may be 50 Gbps / lane or more and 1000 Gbps / lane or less, or may be 50 Gbps / lane or more and 500 Gbps / lane or less, or may be 100 Gbps / lane or more and 500 Gbps / lane or less. Examples of the wavelength of light used in optical fiber communication are 256 nm or more and 3200 nm or less, and may also be 500 nm or more and 2500 nm or less, or may be 1000 nm or more and 2000 nm or less. A specific example of an optical signal is an optical signal that has transmitted an optical fiber of 20 km or more by wideband transmission of 50 Gbps / lane or more based on the IM-DD method.

[0015] Optical preprocessing circuit 3 The optical preprocessing circuit 3 is an element for branching and interfering an optical signal transmitted through the optical fiber 5 to obtain an optical signal after preprocessing. To realize the above functions, the optical preprocessing circuit 3 has, for example, an intensity separation unit 11, a phase adjustment unit 13, a delay adjustment unit 15, and a multiplexing unit 17. Examples of such an optical preprocessing circuit 3 are an optical delay interferometer and a Mach-Zehnder type optical modulator.

[0016] The intensity separation unit 11 is an element for separating an optical signal transmitted through the optical fiber 5 at a desired intensity ratio. For example, a coupler functions as the intensity separation unit 11. An example of the number of waveguide branches is 2. However, the number of waveguide branches may be 2 or more. Methods for adjusting the intensity ratio of the light separated into each waveguide are well known.

[0017] The phase adjustment unit 13 is an element for adjusting the phase of the optical signal separated by the intensity separation unit 11. Usually, a phase modulator provided in one of the waveguides functions as the phase adjustment unit 13. The phase adjustment unit 13 adjusts the relative phase of the split optical signals. The phase modulator can adjust the phase of the optical signal by a drive signal from a signal source, and thereby adjust the relative phase of the two split optical signals.

[0018] The delay adjustment unit 15 is an element for adjusting the delay of the optical signal separated by the intensity separation unit 11. Usually, a path length difference adjustment circuit of the waveguide functions as the delay adjustment unit 15. The delay adjustment unit 15 adjusts the relative time difference of the optical signals separated by the intensity separation unit 11. The delay adjustment unit 15 can adjust the delay of the optical signal by a drive signal from a signal source, and thereby adjust the relative delay difference of the two split optical signals.

[0019] The multiplexing unit 17 is an element for multiplexing the optical signals separated by the intensity separation unit 11 and whose phase and delay are adjusted by the phase adjustment unit 13 and the delay adjustment unit 15. Most importantly, the optical signals in the case of not giving a phase difference and a delay difference are also included in the optical signals whose phase and delay are adjusted.

[0020] Since the optical preprocessing circuit 3 has the above configuration, components that become unstable zeros and poles after the optical - electrical conversion can be removed in advance. Also, by optimizing the drive conditions of the optical preprocessing circuit 3 and the drive conditions of the electrical dispersion compensation circuit 7 described later, the electrical dispersion compensation circuit can be simplified, and a medium - distance large - capacity optical fiber transmission system can be implemented with a realistic circuit scale.

[0021] Electrical dispersion compensation circuit 7 The electrical dispersion compensation circuit 7 is an element for performing dispersion compensation processing on the pre-processed optical signal. The basic configuration of the electrical dispersion compensation circuit 7 is publicly known. Examples of the electrical dispersion compensation circuit 7 include the electrical dispersion compensator described in Japanese Patent No. 5658991 (an electrical dispersion compensator composed of reflective microstrip lines having group delay characteristics for electrical signals) and the electrical dispersion compensation circuit described in Japanese Patent Application Publication No. 2010-278528. As the electrical dispersion compensation circuit 7, any of the following may be used: a feedback equalizer, a feedforward equalizer, and a signal determination circuit. An example of a feedback equalizer is a determination feedback equalizer. An example of a feedforward equalizer is, Examples include FIR filters, frequency-domain equalizers, and Volterra filters. An example of a signal decision circuit is a maximum likelihood estimation circuit.

[0022] Photoelectric processing circuit optimization unit 9 The photoelectric processing circuit optimization unit 9 is an element for optimizing the driving conditions of the optical preprocessing circuit 3 and the electrical dispersion compensation circuit 7 based on the quality of the output signal from the electrical dispersion compensation circuit 7 or a prediction of the output signal quality by an analysis model. Optimization may be performed adaptively based on the actual output signal, or it may be performed in advance using an analytical solution for the output signal quality. For example, the photoelectric processing circuit optimization unit 9 adaptively optimizes the driving signals of the phase adjustment unit 13 and the delay adjustment unit 15, as well as the tap coefficients and number of taps of the electrical dispersion compensation circuit 7, based on the bit error rate characteristics of the output signal from the electrical dispersion compensation circuit 7 and its predicted values ​​by an analysis model. An analysis model for predicting the quality of the output signal (e.g., bit error rate characteristics) is publicly known. Therefore, it is sufficient to obtain predicted values ​​using a known analysis model. The photoelectric processing circuit optimization unit 9 receives measured values ​​from a sensor that measures the parameters of the output signal, for example. Then, the photoelectric processing circuit optimization unit 9 reads a program stored in the memory unit, uses the measured values ​​of the output signal to have the calculation unit perform various calculations, and optimizes the driving conditions of the optical preprocessing circuit 3 and the electrical dispersion compensation circuit 7. The program may include, for example, publicly known machine learning algorithms.

[0023] Design of the optical preprocessing circuit 3 and the electrical dispersion compensation circuit 7 (design optimization process) The photoelectric processing circuit optimization unit 9 may determine the number of optical signal branches and interference methods in the optical preprocessing circuit 3, as well as the number of addition / subtraction circuits and multiplication / division circuits in the electrical dispersion compensation circuit 7, based on the quality of the output signal of the post-processing step or predictions based on an analysis model of the output signal. In this way, the photoelectric processing circuit optimization unit 9 can design the optical preprocessing circuit 3 and the electrical dispersion compensation circuit 7. This embodiment can also be described as a method for constructing (manufacturing) the dispersion compensation device 1 for optical fiber communication signals. The photoelectric processing circuit optimization unit 9 receives measured values ​​from a sensor that measures the parameters of the output signal, for example. The photoelectric processing circuit optimization unit 9 then reads a program stored in the memory unit and uses the measured values ​​of the output signal to have the calculation unit perform various calculations to determine the number of optical signal branches and interference methods in the optical preprocessing circuit 3, as well as the number of addition / subtraction circuits and multiplication / division circuits in the electrical dispersion compensation circuit 7. The program may include, for example, a known machine learning algorithm. An example of determining the interference method is to determine the driving conditions for the phase modulator 13 and delay adjustment unit 15 in each waveguide.

[0024] Dispersion compensation method for optical fiber communication signals Next, a dispersion compensation method for optical fiber communication signals using the above-described apparatus will be explained. Figure 2 is a flowchart illustrating the dispersion compensation method for optical fiber communication signals. As shown in Figure 2, this method includes an optical pre-processing step (S101) and a post-processing step (S102). This method may further include a drive condition optimization step (S103).

[0025] The optical preprocessing step (S101) is a process in which the optical preprocessing circuit 3 branches and interferes with the optical signal transmitted through the optical fiber 5 to obtain the preprocessed optical signal.

[0026] The post-processing step (S102) is a step in which the electrical dispersion compensation circuit 7 performs dispersion compensation processing on the pre-processed optical signal.

[0027] The drive condition optimization step (S103) is a step in which the photoelectric processing circuit optimization unit 9 optimizes the drive conditions of the optical pre-processing circuit 3 and the electrical dispersion compensation circuit 7 based on the quality of the output signal of the post-processing step. Since this step simultaneously optimizes the drive conditions of the optical pre-processing circuit 3 and the electrical dispersion compensation circuit 7, it is possible to provide a dispersion compensation method for optical fiber communication signals that enables the implementation of a short-to-medium distance high-capacity optical fiber transmission system with a realistic circuit scale. After optimizing the drive conditions, the optical pre-processing step (S101) and the post-processing step (S102) may be performed using the optimized conditions. [Examples]

[0028] Implementation-related concepts Figure 3(a) shows a conceptual diagram of a photoelectric-integrated dispersion compensation circuit for short-to-medium distance IM-DD optical fiber transmission. The purpose of the optical pre-processing circuit in the photoelectric-integrated dispersion compensation circuit is to remove, in the optical domain, the signal distortion components due to fiber dispersion that would lead to an increase in the scale of the electrical dispersion compensation circuit. The optical circuit is preferably simple and low-cost, and if possible, composed only of passive optical elements. By adaptively or pre-optimizing the parameters of the optical pre-processing circuit and the electrical dispersion compensation circuit based on the signal quality at the output of the photoelectric-integrated dispersion compensation circuit and predictions from its analysis model, it is possible to significantly reduce the overall complexity and power consumption of the compensation circuit. Figure 3(b) shows a single-tap delay line, one of the simplest implementations of an optical preprocessing circuit. This optical component consists of a 1x2 optical demultiplexer, a phase modulator, an optical delay, and a 2x1 optical multiplexer, and can be integrated as a waveguide optical circuit if necessary.

[0029] Experimental System Figure 4 shows the experimental setup for 1550nm band high-speed optical PAM transmission. On the transmitter side, a 56GBd electrical PAM4 signal was generated from a pulse pattern generator (PPG, Anritsu), and a 1547nm optical carrier (NKT Coheras Basik) was modulated using the PAM method with a Mach-Zehnder modulator (MZM, 3dB bandwidth approximately 25GHz). The modulated signal was transmitted over a 50km single-mode fiber (SMF), amplified by an optical amplifier (EDFA), and then pre-processed by a single-tap delay line with a delay (D) of approximately 8ps. The optical signal output from the optical pre-processing circuit was detected by a 50GHz photodetector (PD) and measured by a 160GSa / s analog-to-digital converter (ADC, Agilent real-time oscilloscope). Electrical dispersion compensation and PAM demodulation were performed offline. A particularly simple feedforward equalizer (FFE) was used for electrical dispersion compensation. Characterization was performed using the post-demodulation bit error rate (BER).

[0030] result Figure 5(a) is a graph that replaces a diagram showing the overall system frequency response in a 50km optical fiber transmission using intensity modulation-direct detection (IM-DD) with and without a single-tap delay line. Figure 5(b) is a graph that replaces a diagram showing the pole-zero plot of a 50km IM-DD optical fiber transmission system without a single-tap delay line. Figure 5(c) is a graph that replaces a diagram showing the pole-zero plot of a 50km IM-DD optical fiber transmission system with a single-tap delay line.

[0031] Figure 5(a) shows that the overall system frequency response has five frequency notches due to fiber chromatic dispersion. Figure 5(b) shows that five unstable pole-zeros, corresponding to these five notches, exist on the upper half-plane unit circle. This indicates that the IM-DD system is a non-minimum phase system with unstable zeros due to the effects of fiber dispersion. In non-minimum phase systems, conventional electrical dispersion compensation circuits such as FFEs generate noise amplification, making efficient dispersion compensation impossible. On the other hand, as shown in Figure 5(c), using a single-tap delay line removes (part of) the unstable zeros, making it possible to efficiently compensate for signal distortion caused by dispersion using FFEs, etc.

[0032] Figure 6(a) shows the relationship between the BER characteristics and the number of FFE taps for 56 Gbaud (112 Gb / s) PAM4 transmission when the PD input optical power is 3 dBm. By using an optical preprocessing circuit, 31 FFE taps achieve the threshold BER required for error-free transmission using a 6.7% redundancy hard-decision error correction code (HD-FEC).

[0033] Figure 6(b) shows the BER versus received optical power for a 50km PAM4 transmission employing a photoelectric integrated dispersion compensation circuit (single-tap delay line and 31-tap FFE). For comparison, Figure 6(b) also shows the BER characteristics for the case with only electrical dispersion compensation, without a single-tap delay line. When using photoelectric integrated dispersion compensation, a threshold BER of 6.7% HD-FEC can be achieved for input powers of 3dBm or higher. On the other hand, when optical preprocessing is not used, the threshold BER cannot be achieved even with a DFE that has three times the circuit size.

[0034] Figure 6(c) shows the BER versus FFE tap length for 80km (ZR) 60Gb / s PAM2 transmission using photoelectric dispersion compensation. By combining single-tap delay lines and electrical dispersion compensation, the threshold BER of the KP4 FEC is achieved with only 23 FFE taps. Figure 6(c) also shows the BER characteristics for 50km 60Gb / s PAM2 transmission using the same optical preprocessing circuit as for 80km. In particular, the threshold BER is achieved with only 13 FFE taps without any changes to the optical system.

[0035] Consideration Experiments demonstrated that the generation of unstable zero-pole points caused by fiber dispersion, a factor contributing to the complexity of electrical dispersion compensation circuits, can be efficiently avoided by a simple single-tap delay line. Furthermore, it was shown that the desired BER characteristics can be achieved with a lightweight FFE of approximately 31 taps, demonstrating the effectiveness of the photoelectric-integrated dispersion compensation technology of the present invention. Since the analytical solutions for the output signal characteristics of the optical delay line and FFE used in photoelectric-integrated dispersion compensation can be derived relatively easily, the optimal design of the photoelectric-integrated dispersion compensation circuit can be performed in advance based on these analytical solutions. [Industrial applicability]

[0036] This invention can be used in the information and communication industry. The electrical dispersion compensation circuit in the dispersion compensation device of this invention shares many common parts with the signal processing circuits of existing Ethernet® PAM4 transceivers. Furthermore, the optical pre-processing circuit 3 in the dispersion compensation device can be implemented using a passive optical waveguide of only a few millimeters. For this reason, the dispersion compensation device of this invention is expected to be used in applications that enable extension without significant modification of existing transceivers, simply by attaching a delay interferometer module to the fiber connector portion of the existing transceiver. [Explanation of symbols]

[0037] 1. Dispersion compensation device for optical fiber communication signals 3. Optical pre-processing circuit 5 Optical Fiber 7. Electrical Dispersion Compensation Circuit 9. Photoelectric Processing Circuit Optimization Unit 11 Strength separation part 13 Phase adjustment section 15. Delay Adjustment Section 17 Wave section

Claims

1. A method for compensating for the dispersion of optical fiber communication signals, The optical preprocessing circuit (3) splits and interferes with the optical signal transmitted by the optical fiber (5) to obtain the preprocessed optical signal, and The electrical dispersion compensation circuit (7) performs a post-processing step in which it performs dispersion compensation processing on the signal obtained by converting the pre-processed optical signal into an electrical signal, The photoelectric processing circuit optimization unit (9) performs a drive condition optimization step in which it selects the number of taps that make the electrical dispersion compensation circuit (7) the lightest among the number of taps that achieve the threshold BER required for error-free transmission, based on the bit error rate characteristics of the output signal from the electrical dispersion compensation circuit (7), Includes, The electrical dispersion compensation circuit (7) includes one or more addition / subtraction circuits and multiplication / division circuits. The photoelectric processing circuit optimization unit (9) further includes a design optimization step that determines the number of optical signal branches and interference method in the optical preprocessing circuit (3), and the number of addition / subtraction circuits and multiplication / division circuits in the electrical dispersion compensation circuit (7), based on predictions by an analysis model of the quality of the output signal of the post-processing step. The aforementioned pre-processing step is a step to remove components that become unstable zeros or poles after the optical pre-processing circuit (3) has converted the optical signal into an electrical signal. method.

2. A method according to claim 1, wherein the optical signal transmitted by the optical fiber (5) is an optical signal transmitted by a wideband transmission of 50 Gbps / lane or more based on an optical intensity modulation scheme, and the optical fiber (5) is 20 km or longer in length.

3. A dispersion compensation device (1) for optical fiber communication signals, An optical preprocessing circuit (3) for branching and interfering with the optical signal transmitted by the optical fiber (5) to obtain a preprocessed optical signal, The system includes an electrical dispersion compensation circuit (7) for performing dispersion compensation processing on a signal obtained by converting the pre-processed optical signal into an electrical signal, and a photoelectric processing circuit optimization unit (9) that selects the number of taps that makes the electrical dispersion compensation circuit (7) the lightest among the number of taps required to achieve the threshold BER necessary for error-free transmission, based on the bit error rate characteristics of the output signal from the electrical dispersion compensation circuit (7). The electrical dispersion compensation circuit (7) includes one or more addition / subtraction circuits and multiplication / division circuits. The optical preprocessing circuit (3) and the electrical dispersion compensation circuit (7) were designed by the photoelectric processing circuit optimization unit (9) based on predictions from an analysis model of the quality of the output signal from the electrical dispersion compensation circuit (7), determining the number of optical signal branches and interference methods in the optical preprocessing circuit (3), and the number of addition / subtraction circuits and multiplication / division circuits in the electrical dispersion compensation circuit (7). The aforementioned optical preprocessing circuit (3) is a circuit that performs processing to remove components that become unstable zeros or poles after the optical signal has been converted into an electrical signal. Dispersion compensator (1).

4. A dispersion compensation device (1) according to claim 3, The aforementioned optical preprocessing circuit (3) is An intensity separation unit (11) separates the optical signal transmitted by the optical fiber (5) at a desired intensity ratio, The intensity separation unit (11) has a phase adjustment unit (13) that adjusts the phase of the separated optical signals, The intensity separation unit (11) has a delay adjustment unit (15) that adjusts the delay of the separated optical signals, The intensity separation unit (11) separates the optical signals, and the combined wave unit (17) combines the optical signals whose phase and delay have been adjusted by the phase adjustment unit (13) and the delay adjustment unit (15). Dispersion compensator (1).

Images