Delay control circuit, optical transmitter, and delay control method

The delay control circuit addresses delay fluctuations in optical transceivers by automatically adjusting electrode segment delays using a frequency-based monitoring system, ensuring consistent signal quality across environmental and manufacturing variations.

JP2025000277AActive Publication Date: 2025-01-071FINITY INC
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2023100050
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-06-19
Publication Date
2025-01-07
Estimated Expiration
2043-06-19

AI Technical Summary

Technical Problem

As baud rates increase, delay fluctuations due to process and temperature variations in electronic circuits become significant, necessitating a highly accurate delay control method for optical transceivers that can adjust delay amounts in varying environmental conditions.

Method used

A delay control circuit that includes a delay circuit, control circuit, frequency filter, and monitor to automatically adjust the delay between electrode segments of a Mach-Zehnder interferometer optical modulator by monitoring the power of extracted frequency components, allowing precise control of signal timing.

Benefits of technology

The circuit enables precise control of delay fluctuations, maintaining signal quality across varying temperatures and manufacturing processes, enhancing performance in high baud rate applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2025000277000001_ABST
    Figure 2025000277000001_ABST
Patent Text Reader

Abstract

To provide a delay control technique for automatically controlling a delay amount of a driving signal of an optical modulator having a plurality of electrode segments.SOLUTION: A delay control circuit includes: a delay circuit that gives a predetermined delay to signals input to a plurality of electrode segments provided in series along one or both of two waveguides of a Mach-Zehnder interferometer of an optical modulator; a control circuit that selects two electrode segments from the plurality of electrode segments and controls a delay amount between the selected two electrode segments; a frequency filter that extracts a predetermined frequency component from output light of the Mach-Zehnder interferometer in a state where identical signals that vary at random are input to the two electrode segments; and a monitor that monitors power of the extracted predetermined frequency component. The control circuit controls the delay amount between the two electrode segments on the basis of a monitored result of the extracted predetermined frequency component.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a delay control circuit, an optical transmission device, and a delay control method. [Background technology]

[0002] A configuration has been considered in which multiple signal electrodes are provided along one or both of the two waveguides that make up the Mach-Zehnder (MZ) interferometer of an optical modulator, and each electrode is driven independently. An optical modulator with such a configuration is sometimes called a "segmented modulator." By dividing a long signal electrode into multiple electrodes and applying a drive voltage to each divided electrode individually, the capacity of the element can be reduced. As a result, it becomes possible to alleviate the trade-off between the power consumption and the bandwidth of the optical modulator. In a segmented modulator, a delay circuit must be provided to match the timing at which light passes through each electrode and the timing at which a drive voltage is applied to that electrode. Until now, the delay amount that has been mainly assumed is the propagation delay of the optical signal (for example, a few picoseconds to about 100 picoseconds). The control circuit that controls the timing is realized by a general circuit that combines, for example, a phase interpolator circuit and a D-flip-flop circuit. In addition, as a method for monitoring and controlling the amount of delay, a configuration has been proposed in which the timing of applying signals to multiple electrodes arranged along the waveguide of an optical modulator is controlled in accordance with the average power of the output light from the optical modulator (see, for example, Patent Document 1). [Prior art documents] [Non-patent literature]

[0003] [Patent Document 1] International Publication No. 2014 / 103231 Summary of the Invention [Problem to be solved by the invention]

[0004] As baud rates increase in response to the recent rapid increase in communication traffic, delay variations occurring on the electronic circuit side due to process and temperature variations cannot be ignored. There is a high possibility that the environmental and local temperatures will differ between the factory where optical transceivers are manufactured and shipped and the site where the optical transceivers are installed and started up. Even if delay adjustment is completed at the time of shipment, readjustment may be required at start-up in the operating environment of the site. In this case, it is desirable to implement a delay circuit and a highly accurate method for monitoring and controlling the delay amount within the optical transceiver.

[0005] The inventors confirmed through electronic circuit simulations assuming an optical modulator drive circuit and an ideal delay circuit that a temperature change from -5°C to 95°C results in a delay of at least 6 picoseconds. This delay amount corresponds to 2 / 5 of 15.6 picoseconds, which is 1 unit interval (UI) of 64 Gbaud, and 3 / 5 of 10.4 picoseconds, which is 1 UI of 96 Gbaud, and is a non-negligible amount of delay for 1 UI. One object of the present disclosure is to provide a delay control technology that automatically controls the delay amount of a drive signal for an optical modulator having multiple electrode segments. [Means for solving the problem]

[0006] In an embodiment, the delay control circuit comprises: a delay circuit that applies a predetermined delay to a signal input to a plurality of electrode segments that are arranged in series along one or both of two waveguides of a Mach-Zehnder interferometer of the optical modulator; A control circuit that selects two electrode segments from the plurality of electrode segments and controls a delay amount between the two selected electrode segments; a frequency filter that extracts a predetermined frequency component from the output light of the Mach-Zehnder interferometer while the same randomly changing signal is input to the two electrode segments; a monitor for monitoring the power of the extracted predetermined frequency component; Equipped with The control circuit controls the amount of delay between the two electrode segments based on a monitoring result of the extracted predetermined frequency component. Effect of the Invention

[0007] A delay control technique is realized that automatically controls the amount of delay in a drive signal of an optical modulator having a plurality of electrode segments. [Brief description of the drawings]

[0008] [Figure 1] 1 is a schematic diagram of an optical transmission device according to an embodiment; [Diagram 2] FIG. 13 is a diagram showing an example of waveform degradation due to signal delay at 32 Gbaud. [Diagram 3] FIG. 2 is a schematic diagram of a delay control circuit according to an embodiment. [Figure 4] 11 is a diagram showing the relationship between the delay deviation amount and the monitor light intensity. FIG. [Diagram 5] FIG. 13 is a diagram illustrating the significance of using a frequency filter. [Figure 6] 1 is a diagram illustrating the center frequency and delay adjustment range of a frequency filter. [Figure 7] FIG. 13 is a diagram illustrating an adjustment range of a delay deviation amount according to a frequency. [Figure 8] FIG. 13 is a diagram showing the delay amount dependence of monitor light intensity when the filter center frequency is 50 GHz and 96 GHz. [Figure 9] FIG. 13 is a diagram showing the delay amount dependency of the frequency spectrum when the filter center frequency is 96 GHz. [Figure 10] 4 is a flowchart of a delay control method according to the first embodiment. [Figure 11] FIG. 13 shows the selection of two electrode segments to be subjected to delay adjustment. [Figure 12] 11 is a flowchart of a delay control method according to a second embodiment. [Figure 13] FIG. 13 is a diagram showing a modification of the delay control circuit. [Figure 14] FIG. 1 is a diagram showing the center frequency and delay adjustment range of a filter. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] In an embodiment, the delay control circuit includes: a delay circuit that gives a predetermined delay to a signal input to a plurality of electrode segments that are provided in series along one or both of two waveguides of a Mach-Zehnder interferometer of the optical modulator; A control circuit that selects two electrode segments from the plurality of electrode segments and controls a delay amount between the two selected electrode segments; a frequency filter that extracts a predetermined frequency component from the output light of the Mach-Zehnder interferometer when the same randomly changing signal (e.g., a PRBS pattern) is input to the two electrode segments; a monitor for monitoring the power of the extracted predetermined frequency component; The control circuit controls the delay amount between the two selected electrode segments based on the monitoring result of the extracted predetermined frequency component, thereby automatically controlling the delay amount of the drive signal of the segment type optical modulator with high accuracy.

[0010] The specific configuration and method of delay control of the embodiment will be described below with reference to the drawings. The embodiment shown below is an example for embodying the technical idea of ​​the present disclosure, and does not limit the disclosed contents. The size, positional relationship, etc. of the components shown in each drawing may be exaggerated to facilitate understanding of the invention. The same components or functions may be given the same names or symbols, and duplicate explanations may be omitted.

[0011] <Optical transmitter> FIG. 1 is a schematic diagram of an optical transmitter 1 according to an embodiment. The optical transmitter 1 has an electronic circuit and an optical circuit, and generates and outputs an optical signal by utilizing the interaction between light and electricity. The optical transmitter 1 has an optical modulator 130, a driver array 120 for driving the optical modulator 130, a digital signal processor (DSP) 5, and a delay control circuit 10. The DSP 5, the delay control circuit 10, and the driver array 120 are included in the electronic circuit, and the optical modulator 130 is included in the optical circuit. As an example of the explanation of the present disclosure, an electric digital-to-analog converter (DAC) is omitted, and a bit signal from the DSP is applied to a divided signal electrode of the optical modulator 130 to perform multi-level conversion in the optical modulator 130. However, the use of the segment type modulator according to the embodiment is not limited to the contents of the present disclosure.

[0012] The optical modulator 130 is a so-called nested type modulator, and two child Mach-Zehnder modulators (MZM) 131 and 132 are connected in parallel between a splitter 149 and a multiplexer 152 to form a parent MZM 133. A carrier wave emitted from a light source is split into two by the splitter 149 and enters each of the child MZMs 131 and 132. The light modulated by each of the MZMs 131 and 132 is multiplexed by the multiplexer 152, and a modulated optical signal is output. The optical modulator 130 is an IQ modulator in which a phase difference of 90° is given between the output lights of the two child MZMs 131 and 132 by the phase shifter 137, but is not limited to this example. A configuration using a single MZM as the optical modulator 130 or a configuration in which two parent MZMs 133 are connected in parallel may be used. In the example of FIG. 1, the waveguide portion including the child MZM 131 may be called an I arm, and the waveguide portion including the child MZM 132 may be called a Q arm.

[0013] The slave MZM 131 has two waveguides 141 and 142 connected in parallel between a splitter 143 and a multiplexer 144, forming an MZ interferometer. Along at least one of the waveguides 141 and 142, a plurality of divided signal electrodes are provided. For convenience, each of the divided signal electrodes is called an "electrode segment." Along the waveguides 141 and 142, electrode segments 138-1, 138-2, ..., 138-7, ... (hereinafter, sometimes collectively referred to as "electrode segments 138") are provided. A phase shifter 135 that maintains the operating point of the slave MZM 131 at an optimal point is provided in series with the electrode segment 138. The phase shifter 135 is used as a DC bias electrode.

[0014] A digital electrical signal representing the logical value of the input data signal is input to each of the electrode segments 138. For example, starting from the light incident side of the child MZM 131, a signal representing the logical value of bit 0 is input to electrode segment 138-1, and a signal representing the logical value of bit 1 is input to electrode segments 138-2 and 138-3. Subsequently, signals representing the value of the corresponding bit are input to electrode segments 138 in number according to the bit string number. The signal for bit (n-1), which is the most significant bit (MSB), is n-1 The input is input to electrode segments 138. For n-bit data (bit 0 to bit (n-1)), 2 n - One electrode segment is provided.

[0015] The slave MZM 132 has two waveguides 145 and 146 connected in parallel between a splitter 147 and a multiplexer 148, and constitutes an MZ interferometer. Along at least one of the waveguides 145 and 146, it has a plurality of electrode segments 139-1, 138-9, ... (hereinafter, sometimes collectively referred to as "electrode segments 139"). A phase shifter 136 that maintains the operating point of the slave MZM 132 at an optimal point is provided in series with the electrode segment 139. The phase shifter 136 is used as a DC bias electrode. The electrode segment 139 of the slave MZM 132 has the same configuration as the electrode segment 138 of the slave MZM 131, except that the input signal is different.

[0016] The optical transmitter 1 having such a segment type optical modulator 130 can be used for both digital coherent transmission and IM-DD (Intensity Modulation - Direct Detection) transmission. In the case of digital coherent transmission, the operating point or bias point of the child MZMs 131 and 132 may be set to a null point. In the case of IM-DD transmission, the operating point or bias point of the child MZMs 131 and 132 may be set to a quad point.

[0017] A portion of the output light from the child MZM 131 is branched before the multiplexer 152 of the parent MZM 133 and detected by a monitor PD 140-1, which is a monitor photodetector. Similarly, a portion of the output light from the child MZM 131 is branched before the multiplexer 152 and detected by a monitor PD 140-2. The output of the monitor PD 140-1 is supplied to a delay control circuit 10 for the I arm. The output of the monitor PD 140-2 is supplied to a delay control circuit for the Q arm. Monitor PDs 140-1 and 140-2 may be collectively referred to as "monitor PD 140."

[0018] 1, for the purpose of making the drawing easier to see, the delay control circuit for the Q arm is omitted, but the delay control circuit for the Q arm is connected between the driver array 120Q for the Q arm and the DSP 5 in the same connection relationship as the delay control circuit 10 for the I arm. The driver array 120Q for the Q arm and the delay control circuit are also included in the electronic circuit. However, the bit data input to the delay control circuit for the Q arm is different from the bit data input to the delay control circuit 10 for the I arm.

[0019] The DSP 5 outputs a digital signal representing the logical value of each bit of the transmission data. The digital signal representing the value of each bit is amplified by the driver 121 of the driver array 120 and input to the corresponding electrode segment 138. The driver 121 is, for example, an inverter driver that applies an inverted output and a non-inverted output to the corresponding electrode segments 138 of the two waveguides 141 and 142.

[0020] When inputting digital signals to the corresponding electrode segments 138 on the waveguides 141 and 142, it is necessary to match the timing at which the incident light passes through the waveguide portion directly below the electrode segment 138 with the timing of the signal input to that electrode segment 138. For this reason, a delay control circuit 10 is inserted between the DSP 5 and the driver array 120 to control the timing acting between the electrode segments 138. The amount of delay between the electrode segments 138 is controlled based on the output light of the slave MZM 131 that is the target of delay control. Similarly, in the slave MZM 132, the timing of the input digital electrical signal is adjusted to match the speed of light passing through the waveguides 145 and 146 based on the output light of the slave MZM 132.

[0021] Each electrode segment 138 is driven in a lumped manner with the same voltage by each driver 121 of the driver array 120. A digital electrical signal input to the electrode segments 138 provided in the waveguides 141 and 142 or the waveguides 145 and 146 of the optical modulator 130 is converted into an analog optical signal by the optical modulator 130. In this sense, the optical transmitting device 1 in FIG. 1 may be called an "optical DAC transmitter."

[0022] The delay control circuit 10 of the embodiment controls the delay during the actual operation of the optical DAC transmitter as described above, and also controls minute delay variations caused by temperature variations and process variations during on-site startup. In addition, it is possible to adjust the propagation delay of an optical signal and the delay caused by non-equal length wiring of an electronic circuit, for example, during inspection before shipment from a factory. The specific configuration and operation of the delay control circuit 10 that controls such delay variations will be described later with reference to FIG. 3 and subsequent figures.

[0023] Figure 2 shows an example of waveform degradation caused by signal delay. It shows the change in eye pattern when two bits, "bit 0" and "bit 1," are used in 32 Gbaud PAM4 (Pulse Amplitude Modulation 4) signal transmission. As shown in Figure 2(a), if the delay between electrode segments is adjusted correctly, an almost equal eye opening is obtained among the four levels. As shown in Figure 2(b), (c), and (d), if the delay from the correct signal input timing becomes large, such as 5 picoseconds, 10 picoseconds, and 15 picoseconds, the signal rise time is delayed, the eye opening is distorted, and the waveform quality is degraded.

[0024] The 5 picosecond delay shown in FIG. 2(b) can occur in electronic circuits due to variations in environmental temperature and process even after adjusting for delays due to variations in electrical wiring length and optical signal propagation delays. The 5 picosecond delay at 32 Gbaud narrows the eye opening between levels 0 and 1. When the baud rate increases to 64 Gbaud and 96 Gbaud, the effect of waveform degradation due to the 5 picosecond signal delay becomes even greater. By precisely controlling such minute delay variations with the delay control circuit 10, good signal quality can be maintained regardless of variations in the manufacturing process and installation environment.

[0025] <Delay control circuit> 3 is a schematic diagram of the delay control circuit 10 of the embodiment. The configuration and operation of the delay control circuit 10 will be described with a focus on a single MZM, for example, the child MZM 131. The operation of FIG. 3 is performed during inspection before shipment from the factory, when the optical transmitting device 1 is introduced into a network and started up, when it is restarted for repair or maintenance, etc.

[0026] The delay control circuit 10 includes a delay circuit 11, a control circuit 15 for controlling the delay circuit 11, a frequency filter 12 for extracting a predetermined frequency component from the detection result of the output light of the child MZM 131, and a monitor 13 for monitoring the power of the extracted frequency component. The delay circuit 11 includes delay adjustment circuits 111-1, 111-2, 111-3, ... (collectively referred to as "delay adjustment circuits 111" as appropriate) for adjusting the delay amount of a signal input to each of electrode segments 138-1, 138-2, 138-3, ... provided along the waveguide 141 or 142 of the child MZM 131. A delay adjustment clock CLK is supplied to each delay adjustment circuit 111, for example, from the DSP 5.

[0027] In adjusting delay variations before actual operation, the control circuit 15 selects two electrode segments from among the multiple electrode segments 138. The same signal having a sufficiently wide continuous spectrum in the frequency domain extracted by the frequency filter 12 is input to the two selected electrode segments 138. In the example of FIG. 3, electrode segments 138-1 and 138-2 are selected. As a signal having a sufficiently wide continuous spectrum in the frequency domain extracted by the frequency filter 12, a random signal that changes randomly or a pseudo-random binary bit sequence (PRBS) is used. The same signal is a signal that has a random change pattern and has the same amplitude and phase. The same signal is input from the DSP 5 to the delay control circuit 10 for delay adjustment, and is input to the delay adjustment circuits 111-1 and 111-2 that adjust the delay of the selected electrode segments 138-1 and 138-2. No signal is input to the unselected electrode segments 138, and they are in the off state.

[0028] A delay adjustment clock causes delay adjustment circuits 111-1 and 111-2 to give a predetermined delay to the same randomly varying signal. The delayed signal is amplified by drivers 121-1 and 121-2 and input to electrode segments 138-1 and 138-2. Child MZM 131 is set, for example, at a quad point, and an inverted signal and a non-inverted signal are input to electrode segment 138-1 provided on waveguide 141 and the corresponding electrode segment provided on waveguide 142. In this configuration, the term "electrode segment 138-1" includes both electrode segment 138-1 on waveguide 141 and the corresponding electrode segment on waveguide 142.

[0029] The optical output of the child MZM 131 is maximum when the timing at which light passes through the waveguide 141 directly below electrode segment 138-1 coincides with the timing at which a signal is input to electrode segment 138-1, and when the timing at which light passes through the waveguide 141 directly below electrode segment 138-2 coincides with the timing at which a signal is input to electrode segment 138-2.

[0030] A portion of the optical output of the child MZM 131 is detected by the monitor PD 140 (corresponding to the monitor PD 140-1 in FIG. 1), and an electrical signal indicating the detection result is input to the frequency filter 12. The frequency filter 12 is, for example, an electrical filter that extracts a predetermined frequency component from the detection result of the monitor PD 140. The monitor 13 monitors the power of the extracted frequency component and supplies the monitoring result to the control circuit 15. Because a randomly changing signal is used as the input signal, the monitor 13 can monitor the power of any frequency component extracted from a spectrum that spreads continuously over the extracted frequency range.

[0031] The control circuit 15 sweeps the delay amount of either one of the delay adjustment circuits 111-1 or 111-2 to control the delay amount to the maximum power monitored by the monitor 13. One of the two delay adjustment circuits 111-1 or 111-2 is used as a reference circuit, and the delay amount of the other delay adjustment circuit is swept. The relative delay amount that maximizes the monitor power of the extracted frequency component is set in the delay adjustment circuit 111 to be controlled.

[0032] After the relative delay amount between the two selected electrode segments 138-1 and 138-2 is adjusted, another two electrode segments, for example, electrode segments 138-1 and 138-3, or 138-2 and 138-3, are then selected. After the relative delay amount between the newly selected two electrode segments 138 is adjusted in the above-mentioned procedure, two more electrode segments 138 are selected. The relative delay amount between the two electrode segments 138 is adjusted in sequence until the delay between the electrode segments 138 in the entire child MZM 131 is adjusted.

[0033] This delay adjustment can be applied not only when starting up the optical transmission device 1 on-site, but also in inspection before shipment from a factory. Normally, in inspection before shipment from a factory, the delay amount of each delay adjustment circuit 111 is adjusted while measuring the eye pattern, but by using the delay control circuit 10 of the embodiment, the total delay amount can be automatically controlled, including adjustment of minute delay fluctuations.

[0034] FIG. 4 shows the relationship between the delay deviation between two electrode segments 138 and the monitor light intensity. The same pseudo random binary bit sequence (PRBS) is input to the two selected electrode segments 138, and the output light of the slave MAM 131 is detected by the monitor PD 140-1. The frequency filter 12 extracts a frequency component of 20±5 GHz from the output of the monitor PD 140, and the delay amount of the delay adjustment circuit 111 to be adjusted is swept in the range of ±25 picoseconds. When the delay amount is 0 picoseconds, the optical output of the slave MZM 131 is maximized, and when the delay amount is 25 picoseconds, the optical output is extinguished. The delay amount of the delay adjustment circuit 111 corresponding to the electrode segment 138 to be controlled is swept, and the delay amount at which the optical output is maximized in the extracted frequency range is determined, thereby minimizing the timing deviation between the two selected delay adjustment circuits 111.

[0035] Fig. 5 is a diagram showing the significance of using the frequency filter 12. It shows an example of actual measurement of power frequency characteristics at different delay amounts when generating a PAM4 waveform using 2-bit data. The thickest line is the actual measurement result with a delay amount of 0 picoseconds, the medium thickness line is a delay of 40 picoseconds, and the thin line is a delay of 120 picoseconds. The horizontal axis is frequency (Hz), and the vertical axis is monitored power (dBm). When the delay amount is 0 picoseconds (NRZ), the monitored power is almost constant up to a certain frequency (20 GHz), and the monitored power decreases once the frequency exceeds 20 GHz.

[0036] In contrast, when the delay is 40 picoseconds or 120 picoseconds, the power is quenched at a certain frequency interval according to the delay. In other words, it can be seen that the frequency at which the power is quenched changes depending on the delay between the segments. In the example of Figure 5, as shown in the shaded area, when the delay is 120 picoseconds, the power monitored near 4 GHz decreases. When the delay is 40 picoseconds, the power monitored near 12 GHz decreases. Therefore, by extracting frequency components near 4 GHz, 12 GHz, 20 GHz, etc. and monitoring the power, it is possible to extract a steep change in power according to the delay and detect minute delay differences with high accuracy.

[0037] FIG. 6 shows the relationship between the center frequency of the frequency filter 12 and the delay adjustment range. As described above, the frequency filter 12 is, for example, an electrical bandpass filter, and can arbitrarily select the frequency components to be extracted. In this case, the relationship between the frequency components removed due to the timing shift and the delay amount can be expressed as 1 / 2τ, where τ is the timing shift amount (delay amount). Therefore, when the filter center frequency is 5 GHz, the delay can be adjusted in the range of ±100 picoseconds. When the filter center frequency is increased to 10 GHz, 20 GHz, 50 GHz, and 100 GHz, the delay adjustment range becomes narrower, to ±50 picoseconds, ±25 picoseconds, ±10 picoseconds, and ±5 picoseconds.

[0038] FIG. 7 shows the adjustment range of the delay deviation amount according to the frequency. When the filter central frequency is low, as shown in range B, the delay deviation amount can be controlled over a wide range. When the filter central frequency is high, as shown in range A, the control range of the delay deviation amount is narrow, but minute delay deviations can be adjusted with high precision. The center frequency of the frequency filter 12 may be set low to adjust a relatively large delay amount caused by non-equal length wiring on the electronic circuit side or optical propagation delay, and then the center frequency of the frequency filter 12 may be set high to adjust minute delay fluctuations caused by temperature variations and process variations. For example, the center frequency of the frequency filter 12 may be set to about 5 to 10 GHz at the time of factory shipment to adjust the delay caused by optical propagation delay or wiring length variations, and when the optical transmitter 1 is started up at the site, the filter center frequency may be set to the higher frequency side to perform minute delay adjustment.

[0039] FIG. 8 shows the delay dependency of the monitor light intensity when the filter center frequency is 50 GHz and 96 GHz (±5 GHz). The delay dependency in FIG. 8 is an example corresponding to the case where the control range of the delay deviation is narrow as shown in range A in FIG. 7. In FIG. 8(A), when the center frequency of the frequency filter 12 is set to 50 GHz±5 GHz, if the timing deviation between the electrode segments 138 increases from 0 picoseconds to 7 picoseconds, the power (monitor intensity) measured by the monitor 13 simply decreases. In particular, from 1 picosecond to 7 picoseconds, the monitor intensity changes with a sufficient gradient. When an ideal power monitor with high resolution is used, even a slight change in the monitor intensity caused by a minute delay of 0.5 picoseconds or less can be detected.

[0040] In FIG. 8B, when the center frequency of the frequency filter 12 is set to 96 GHz±5 GHz, the timing shift between the electrode segments 138 increases, and the monitor intensity decreases with a larger gradient than in FIG. 8A, and is extinguished in approximately 5.2 picoseconds. In particular, the gradient of the monitor intensity change is large in the range of 0.5 picoseconds to 5 picoseconds, and minute delay amounts can be detected with high accuracy. In a region where the delay variation amount of the controlled object is large, the center frequency of the frequency filter 12 may be set to, for example, 50 GHz or less, and in a region where the delay amount of the controlled object is minute, the center frequency of the frequency filter 12 may be set to a frequency higher than 50 GHz. This enables flexible delay control according to the cause of the timing shift.

[0041] Fig. 9 is a diagram showing the delay dependence of the frequency spectrum when the center frequency of the frequency filter 12 is set to 96 GHz ± 5 GHz, i.e., in the case corresponding to Fig. 8 (B). When there is no delay difference, i.e., no timing shift, between the electrode segments 138 selected for delay adjustment, the electrical frequency spectrum of the monitor output is as shown in Fig. 9 (A). In contrast, when there is a delay difference of 5 picoseconds between the electrode segments 138 selected for delay adjustment, as shown in Fig. 9 (B), it can be seen that the power of the frequency response in the high frequency range is greatly reduced due to the effect of the high frequency filter. In this way, the phenomenon of monitor light intensity at the segment timing shift as shown in Fig. 8 (B) is observed.

[0042] As described above, the effect of the present disclosure is that the range of the monitored delay amount can be changed from the order of 100 picoseconds to about sub-picoseconds by adjusting the center frequency of the frequency filter 12. That is, it also contributes to improving the accuracy of delay control. For example, Patent Document 1 indicates that the dependency of the optical output intensity depends on the baud rate. Assuming 64 Gbaud, 1 UI is 15.625 picoseconds, and there is almost no difference in the output intensity from the graph of the optical output intensity and the delay time, making it difficult to detect a change of about 2 picoseconds. On the other hand, in the present disclosure, by narrowing down the high frequency region to be monitored using the frequency filter 12, it is possible to accurately detect a timing deviation of 0.5 picoseconds to 2 picoseconds. It can be said that the delay control circuit 10 of this embodiment has a great effect on Patent Document 1 in terms of the delay adjustment range and the accuracy of the delay adjustment.

[0043] The delay control circuit 10 of the embodiment can detect delays from 0.5 picoseconds to 100 picoseconds regardless of the baud rate by setting the extraction frequency band of the frequency filter 12 to the gigahertz band. It can be used for delay adjustment before shipping from the factory and for fine adjustment of the delay variation amount at the time of start-up in the field.

[0044] <Delay control method> 10 is a flowchart of the delay control method of the embodiment 1. The main part of this delay control is executed by the delay control circuit 10. In the embodiment 1, the delay control is performed so that the power of the electrical spectrum monitored by the monitor 13 of the delay control circuit 10 is maximized.

[0045] First, the bias of the MZ interferometer (MZI) is adjusted to an optimal point (S11). The bias control is performed by, for example, an automatic bias control (ABC) circuit provided in the optical transmission device 1. The ABC circuit may be realized by a logic device such as a field programmable gate array (FPGA) or a microprocessor. When the optical modulator 130 is used in the optical transmission device 1 for IM-DD, the bias may be set to a quad point. When the optical modulator 130 is used in the optical transmission device 1 for digital coherent, the bias may be set to a null point. With the bias set to the optimal point, delay control is performed.

[0046] The control circuit 15 of the delay control circuit 10 selects two electrode segments to be delayed, for example, electrode segments 138-1 and 138-2, from among a plurality of electrode segments 138 provided along the waveguide of the MZ interferometer. The same signal that changes randomly is input to the selected electrode segments 138-1 and 138-2 (S12). The unselected electrode segments 138 are turned off and no signal is input. The output light of the MZ interferometer is received by the monitor PD 140, and a desired frequency component is extracted from the electrical signal representing the output of the monitor PD 140 to monitor the power (S13).

[0047] The control circuit 15 sweeps the delay amount of the delay adjustment circuit 111 corresponding to the electrode segment 138 to be controlled, and detects the delay amount when the power monitored by the monitor 13 is maximized (S14). Using one of the two selected electrode segments 138 as a reference, the delay amount of the delay adjustment circuit 111 that adjusts the delay of the other electrode segment 138 is swept to relatively adjust the delay between the two electrode segments 138. When the maximum power point is detected, the delay adjustment circuit 111 of the electrode segment 138 to be adjusted is fixed to the delay amount that maximizes the monitor power (S15). At this stage, the relative delay adjustment process I between the two currently selected electrode segments 138 is completed.

[0048] Next, two other electrode segments 138 are selected, and the relative delay adjustment process II is performed in the same procedure as the delay adjustment process I. That is, two other electrode segments to be adjusted, for example, electrode segments 138-2 and 138-3, are selected, and the same signal that changes randomly is input to the selected electrode segments 138 (S16). Thereafter, steps S13 to S15 of the delay adjustment process I are repeated to relatively adjust the delay between the two electrode segments 138. Different pairs of electrode segments 138 are sequentially selected to adjust the relative delay amount. When the relative delay amount has been controlled for all pairs of electrode segments 138 to be selected, the delay adjustment of the entire MZ interferometer is completed.

[0049] Fig. 11 is a diagram showing selection of two electrode segments 138 to be subjected to delay adjustment. As shown in selection method (A) of Fig. 11, one electrode segment (for example, electrode segment 138-1) may be fixed as a reference segment, and this reference segment and another electrode segment 138 may be selected in sequence.

[0050] As shown in the selection method (B) of FIG. 11, one electrode segment (for example, electrode segment 138-1) may be fixed as a reference segment, and one electrode segment 138 may be selected from an electrode segment group to which another bit data is input, to form an electrode segment pair. For example, the electrode segment 138-1 to which bit 0 data is input is set as a reference segment, and one of the electrode segments 138- and 138-2 to which bit 1 data is input is selected. Next, one of the four electrode segments to which bit 2 data is input is selected based on the electrode segment 138-1, to form a segment pair. The delay amount of an electrode segment not selected in the same electrode segment group can be interpolated, for example, by linear interpolation. However, the selection method (B) of FIG. 11 is effective only when the relationship between the delay amounts of the electrode segments is linear.

[0051] As another method, the electrode segment pairs may be selected by sequentially shifting the reference electrode segments themselves. After performing delay adjustment between electrode segments 138-1 and 138-2, electrode segments 138-2 and 138-3 may be selected, and the electrode segment pairs may be shifted one by one. Regardless of the selection method, the relative delay between the two electrode segments is determined in the extracted frequency domain, and the signal input timing of the entire MZ interferometer is correctly adjusted.

[0052] If the MZ interferometer to be delay-controlled is the MZM of the I arm of the IQ optical modulator, the same delay adjustment process is also performed for the MZ interferometer forming the MZM of the Q arm. The delay adjustment of the MZM of the Q arm is performed by a delay control circuit provided for the Q arm. The delay adjustment of the I arm and Q arm may be performed simultaneously using the same random signal for the delay adjustment of the I arm and the Q arm, or may be performed alternately in a time-division manner.

[0053] FIG. 12 is a flowchart of a delay control method according to the second embodiment. In the second embodiment, the delay amount is determined by utilizing the attenuation of the frequency component due to the timing shift. Steps S11 to S13 are the same as those in the first embodiment. The bias of the MZ interferometer (MZI) is adjusted to an optimum point (S11) to perform delay adjustment. Two electrode segments to be delayed are selected from a plurality of electrode segments 138 provided along the waveguide of the MZ interferometer, and the same signal that changes randomly is input to the two selected electrode segments (S12). The output light of the MZ interferometer is received by the monitor PD 140, and the desired frequency component is extracted from the electrical signal representing the output of the monitor PD 140 to monitor the power (S13).

[0054] The control circuit 15 sweeps the delay amount of the delay adjustment circuit 111 corresponding to the electrode segment 138 to be controlled, and calculates the delay amount when the monitor power is minimized in the extracted frequency domain (S24). Here, the relationship between the frequency component removed due to the timing shift and the delay amount can be expressed as 1 / 2τ, where the timing shift amount (delay amount) is τ. Therefore, the delay amount at which the power is extinguished is known in advance depending on the center frequency of the filter set in the frequency filter 12. By setting an arbitrary frequency component extraction in the frequency filter 12 and monitoring the extinction point of the power of the arbitrary extracted frequency component, the corresponding delay amount can be added. The delay amount corresponding to the time when the monitor power is minimized is fed back to the delay adjustment circuit 111 to be controlled, and set to this delay amount (S25). At this stage, the relative delay adjustment process I' between the two electrode segments 138 currently selected is completed.

[0055] Next, two other electrode segments 138 are selected, and the relative delay adjustment process II is performed in the same procedure as the delay adjustment process I'. Two other electrode segments to be adjusted, for example, electrode segments 138-2 and 138-3, are selected, and the same signal that changes randomly is input to the selected electrode segments 138 (S16). Thereafter, steps S13, S24, and S25 of the delay adjustment process I' are repeated to relatively adjust the delay between the two electrode segments 138. Different pairs of electrode segments 138 are sequentially selected to adjust the relative delay amount. When the relative delay amount has been controlled for all pairs of electrode segments 138 to be selected, the delay adjustment of the entire MZ interferometer is completed.

[0056] Depending on the purpose of the delay adjustment, the method of the first embodiment and the method of the second embodiment may be used separately. For example, in delay adjustment at the time of shipping from the factory, the method of the first embodiment adjusts delays caused by variations in wiring length and delays caused by optical propagation to match the timing of optical propagation and the timing of signal input. When starting up or restarting the optical transmitter 1 on-site, the method of the second embodiment adjusts minute delay variations caused by temperature variations and environmental differences. When the baud rate is high, a delay deviation of several picoseconds may affect the signal waveform, so by performing delay adjustment with high accuracy using the method of the second embodiment, the timing of signal input to the electrode segments can be automatically controlled with high accuracy.

[0057] <Modification> FIG. 13 is a schematic diagram of a delay control circuit 10A which is a modified example of the delay control circuit 10. As shown in FIG. The delay control circuit 10A has two types of frequency filters 12-1 and 12-2. The frequency filters 12-1 and 12-2 are switchably connected between the monitor PD 140 and the monitor 13 of the delay control circuit 10A. The frequency filters 12-1 and 12-2 extract different frequency components. The frequency f1 extracted by the frequency filter 12-1 is in a lower frequency band than the frequency f2 extracted by the frequency filter 12-2. The frequency filter 12-1 extracts low frequency components to detect the delay deviation amount over a wide range. The frequency filter 12-2 extracts high frequency components to detect the delay deviation amount in a narrow range with high accuracy.

[0058] FIG. 14 shows the relationship between the center frequency and the delay adjustment range of the frequency filters 12-1 and 12-2. For example, after roughly adjusting the delay deviation over a wide range with the frequency filter 12-1, the frequency filter 12-2 may be switched to adjust the delay deviation of 2 picoseconds or less with high precision. Instead of using the two frequency filters 12-1 and 12-2, one variable frequency filter may be used. In this case, the frequency to be extracted can be changed to suit the purpose or mode of delay control. The other configurations of the circuit 10A are the same as those of the delay control circuit 10, and therefore a duplicated description will be omitted.

[0059] Although the optical node device and the optical communication system have been described above based on specific configuration examples, the present disclosure is not limited to the above configuration. In the embodiment, an electric bandpass filter or an electric filter with a variable center frequency and frequency range is used as a means for extracting a predetermined frequency component, but the use of an optical filter is not excluded. An optical filter may be provided at the output stage of the optical modulator 130, and the filtered light may be detected by the monitor PD 140. When the optical modulator 130 is formed using silicon photonics technology, an optical filter may be integrated on the same substrate. A random training sequence generated by the DSP 5 may be used as the same signal to be input to the electrode segment selected during delay adjustment. Alternatively, a configuration may be adopted in which the control circuit 15 of the delay control circuit 10 holds a random pattern for delay adjustment and inputs it to the delay adjustment circuit corresponding to the selected electrode segment.

[0060] In response to the above disclosure, the following addendum is presented: (Appendix 1) a delay circuit that gives a predetermined delay to a signal input to a plurality of electrode segments that are provided in series along one or both of two waveguides of a Mach-Zehnder interferometer of the optical modulator; A control circuit that selects two electrode segments from the plurality of electrode segments and controls a delay amount between the two selected electrode segments; a frequency filter that extracts a predetermined frequency component from the output light of the Mach-Zehnder interferometer while the same randomly changing signal is input to the two electrode segments; a monitor for monitoring the power of the extracted predetermined frequency component; Equipped with The control circuit controls the delay between the two electrode segments based on a monitoring result of the extracted predetermined frequency component. Delay control circuit. (Appendix 2) The control circuit sweeps the delay given to the same signal input to the two electrode segments by the delay circuit, and sets the delay amount in the delay circuit at which the power of the extracted predetermined frequency component is maximized. 2. The delay control circuit of claim 1. (Appendix 3) the control circuit sweeps the delay given to the same signal input to the two electrode segments by the delay circuit, and sets in the delay circuit a delay amount corresponding to a frequency at which the power is minimized among the extracted predetermined frequency components as a relative delay amount between the two electrode segments. 2. The delay control circuit of claim 1. (Appendix 4) a photodetector for detecting the output light of the Mach-Zehnder interferometer; the frequency filter is an electric filter that extracts the predetermined frequency component from an electric signal indicating a detection result of the photodetector. 2. The delay control circuit of claim 1. (Appendix 5) 5. The delay control circuit according to claim 4, wherein the frequency filter includes a plurality of frequency filters having different center frequencies and frequency ranges. (Appendix 6) The frequency filter is a variable filter having a variable center frequency and a variable frequency range. 5. The delay control circuit of claim 4. (Appendix 7) the optical modulator is an IQ modulator in which a first Mach-Zehnder interferometer and a second Mach-Zehnder interferometer are connected in parallel to provide an optical phase difference of 90 degrees between the first Mach-Zehnder interferometer and the second Mach-Zehnder interferometer; a power of a first frequency component included in the output light of the first Mach-Zehnder interferometer and a power of a second frequency component included in the output light of the second Mach-Zehnder interferometer are monitored, and a delay of a first signal input to a first electrode segment set provided in the first Mach-Zehnder interferometer and a delay of a second signal input to a second electrode segment set provided in the second Mach-Zehnder interferometer are independently controlled. 2. The delay control circuit of claim 1. (Appendix 8) the delay circuit includes a plurality of delay adjustment circuits corresponding to the plurality of electrode segments; The control circuit sweeps the delay amount of one of the two delay adjustment circuits corresponding to the two electrode segments. 2. The delay control circuit of claim 1. (Appendix 9) The two electrode segments are two adjacent electrode segments. 2. The delay control circuit of claim 1. (Appendix 10) The two electrode segments are selected one by one from two electrode segment groups among a plurality of electrode segment groups provided corresponding to each bit of the bit data. 2. The delay control circuit of claim 1. (Appendix 11) A delay control circuit according to any one of claims 1 to 10; A digital signal processor; the delay control circuit is provided between the digital signal processor and the optical modulator; Optical transmitting device. (Appendix 12) a driver array for amplifying the signals input to the plurality of electrode segments; having the delay control circuit is connected between the digital signal processor and the driver array; The driver array amplifies the digital signal output from the digital signal processor and delayed by the delay control circuit, and inputs the delay-adjusted digital signal to the plurality of electrode segments. 12. The optical transmitting device according to claim 11. (Appendix 13) selecting two electrode segments from a plurality of electrode segments provided in series along one or both of the two waveguides of the Mach-Zehnder interferometer of the optical modulator; inputting the same randomly varying signal to the two selected electrode segments; extracting a predetermined frequency component from the output light of the Mach-Zehnder interferometer; monitoring the power of the extracted predetermined frequency component; controlling an amount of delay between the two electrode segments based on the power monitoring result; Delay control method. (Appendix 14) a delay amount given to the same signal input to the two electrode segments is swept, and a delay amount at which the power of the predetermined frequency component is maximized is set in the delay circuit; 14. The delay control method according to claim 13. (Appendix 15) a delay amount given to the same signal input to the two electrode segments is swept, and a delay amount corresponding to a frequency at which the power of the extracted predetermined frequency components is minimized is set in the delay circuit as a relative delay amount between the two electrode segments; 14. The delay control method according to claim 13. [Explanation of symbols]

[0061] 1 Optical transmitter 5 DSP 10, 10A Delay control circuit 11 Delay Circuit 111, 111-1, 111-2, 111-3 Delay adjustment circuit 12, 12-1, 12-2 Frequency Filter 13 Monitor 15 Control circuit 120, 120Q driver array 121 Driver 130 Optical Modulator 131, 132 Child MZM 133 Parent MZM 138, 131-1 to 131-7, 139 Electrode segments 140, 140-1, 140-2 Monitor PD (photodetector)

Claims

1. a delay circuit for applying a predetermined delay to a signal input to a plurality of electrode segments provided in series along one or both of two waveguides of a Mach-Zehnder interferometer of the optical modulator; A control circuit that selects two electrode segments from the plurality of electrode segments and controls a delay between the two selected electrode segments; a frequency filter that extracts a predetermined frequency component from the output light of the Mach-Zehnder interferometer while the same randomly changing signal is input to the two electrode segments; a monitor for monitoring the power of the extracted predetermined frequency component; Equipped with The control circuit controls the delay between the two electrode segments based on a monitoring result of the extracted predetermined frequency component. Delay control circuit.

2. the control circuit sweeps a delay given to the same signal input to the two electrode segments by the delay circuit, and sets the delay amount in the delay circuit such that the power of the extracted predetermined frequency component is maximized.

2. The delay control circuit of claim 1.

3. the control circuit sweeps the delay given to the same signal input to the two electrode segments by the delay circuit, and sets a delay amount corresponding to a frequency at which the power is minimized among the extracted predetermined frequency components in the delay circuit as a relative delay amount between the two electrode segments.

2. The delay control circuit of claim 1.

4. a photodetector for detecting the output light of the Mach-Zehnder interferometer; the frequency filter is an electric filter that extracts the predetermined frequency component from an electric signal indicating a detection result of the photodetector.

2. The delay control circuit of claim 1.

5. 5. The delay control circuit according to claim 4, wherein the frequency filter includes a plurality of frequency filters having different center frequencies and frequency ranges.

6. The frequency filter is a variable filter having a variable center frequency and a variable frequency range.

5. The delay control circuit according to claim 4.

7. the delay circuit includes a plurality of delay adjustment circuits corresponding to the plurality of electrode segments; The control circuit sweeps the delay amount of one of the two delay adjustment circuits corresponding to the two electrode segments.

2. The delay control circuit of claim 1.

8. A delay control circuit according to any one of claims 1 to 7; A digital signal processor; the delay control circuit is provided between the digital signal processor and the optical modulator; Optical transmitting device.

9. a driver array for amplifying the signals input to the plurality of electrode segments; having the delay control circuit is connected between the digital signal processor and the driver array; The driver array amplifies the digital signal output from the digital signal processor and delayed by the delay control circuit, and inputs the delay-adjusted digital signal to the plurality of electrode segments.

9. The optical transmitter according to claim 8.

10. selecting two electrode segments from a plurality of electrode segments provided in series along one or both of the two waveguides of the Mach-Zehnder interferometer of the optical modulator; inputting the same randomly varying signal to the two selected electrode segments; extracting a predetermined frequency component from the output light of the Mach-Zehnder interferometer; monitoring the power of the extracted predetermined frequency component; controlling an amount of delay between the two electrode segments based on the power monitoring result; Delay control method.

Citation Information

Patent Citations

  • Light modulator and light modulation method

    JP2022024347A

  • Optical transmitter

    JP2023012575A

  • Optical waveguide modulator

    US20180341164A1

  • Electrical-optical modulator

    US20210080798A1

  • Optical waveguide modulator

    US20210373409A1