An adaptive control method for wavelength division multiplexer

By performing the adaptive control method of cycling steps in the MZI unit of the wave division repeat device, the channels and light source wavelengths caused by process errors are solved, and the control circuits of adaptive correction and simplified channels are realized.

CN115524972BActive Publication Date: 2025-01-28NANO TECHNOLOGY (BEIJING) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202211246600.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-10-12
Publication Date
2025-01-28
Estimated Expiration
2042-10-12

AI Technical Summary

Technical Problem

During the process of manufacturing, existing wavelength division multiplexers are prone to channel offset and light source wavelength offset due to errors, and existing solutions such as using reverse thermal light or different polarized light to increase complexity and loss, and are not applicable. Due to the offset problem caused by process error.

Method used

Using adaptive control methods, through the implementation of cycles in the MZI unit, adjust the voltage steps with the change direction of the PS voltage and MPD current to achieve the adaptive correction of the wave -to -docratic channel channel.

Benefits of technology

No need to use disturbance signals and detectors, simplifying the control circuit, and realizing the full -time correction of the wavelength of the wavelength of the wave splitter, solving the problem of wavelength drift caused by process errors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115524972B_ABST
    Figure CN115524972B_ABST
Patent Text Reader

Abstract

The present invention discloses an adaptive control method for a wavelength division multiplexer, comprising: performing cyclic stepping on the MZI units in each level in order from low level to high level; wherein, the process of performing the stepping is that when the PS voltage and the MPD current change in the same direction, the current PS voltage is reduced by one step, otherwise it is increased by one step. It is not necessary to use a disturbance (Dither) signal, a detector and other components in the circuit. The feedback adjustment method of the present invention is simple and reliable, and can realize full-time automatic correction of the wavelength division multiplexer channel relative to the wavelength of the light source; it solves the problem of expected value of wavelength drift of the wavelength division multiplexer caused by process error and wavelength offset of the light source and wavelength offset of the wavelength division multiplexer channel during operation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of wavelength division multiplexer control, and in particular to an adaptive control method of a wavelength division multiplexer. Background Art

[0002] At present, the following structures are mainly used in the wavelength division multiplexer in the integrated optical circuit: cascaded Mach Zehnder (MZI) structure, etched diffraction grating structure, arrayed waveguide grating structure, sidewall grating structure, and micro-ring resonance structure. Among them, the cascaded MZI structure has the advantages of low loss, flat-top passband, strong scalability, and independent phase adjustment. It has attracted the attention and research of industry and academia in recent years. However, when the optical communication system is working, the wavelength of the light source and the wavelength of the wavelength division multiplexer channel will shift due to temperature changes. For the above technical problems, the existing solutions are as follows:

[0003] First, add reverse thermo-optical materials or structures to the phase shift arm: when the thermo-optical coefficient of the added material is opposite to that of the material used in the integrated optical circuit, a specific structural design can be used to reduce the temperature sensitivity of the wavelength division multiplexer with the reverse thermo-optical material. However, the reverse thermo-optical material is incompatible with the complementary metal oxide semiconductor process and is difficult to mass produce. When a special structure such as a slit waveguide is added, the thermo-optical coefficient of the structure is quite different from that of the integrated optical circuit waveguide structure, thereby constructing a wavelength division multiplexer that is insensitive to ambient temperature. However, the formation of the slit structure requires very fine processing accuracy (usually tens of nanometers), which makes manufacturing difficult.

[0004] Second, by utilizing light of different polarizations: light of different polarization states has different thermo-optical coefficients in the integrated optical circuit waveguide, a wavelength division multiplexer with low temperature sensitivity can also be constructed. However, the generation of multiple polarization states requires the addition of a polarization separation rotator, which makes the integrated optical circuit more complex, larger in size, and has greater losses.

[0005] The above two solutions can only solve the impact of temperature on the wavelength division multiplexer. However, during the process of manufacturing, manufacturing errors such as wafer thickness, waveguide width, waveguide etching depth, etc. will also cause the wavelength division multiplexer's own channel offset and light source wavelength offset. Therefore, the above two solutions are no longer applicable.

[0006] In order to solve the impact of process manufacturing on wavelength division multiplexers, technicians in this field introduced phase shifters and photodetectors to compensate for wavelength deviation, that is, adding a photodetector to one of the output ends of the MZI to monitor the channel deviation state of the MZI, and adding a phase shifter to the asymmetric arm of the MZI to adjust the channel state of the MZI. The commonly used control method is to load a relatively low-frequency disturbance (Dither) signal on the phase shifter, and process the signal detected by the photodetector with the original Dither signal by multiplication, integration, etc., and judge and adjust the working point of the MZI according to the size of the first and second harmonics of the Dither signal. However, the use of the Dither signal increases the complexity of the control circuit and introduces noise of the Dither signal frequency to the MZI, thereby affecting the communication quality.

[0007] Although the control method of the IQ modulator is simpler and has a faster response speed than other commonly used modulator control methods, due to the design requirements of the IQ modulator itself, the control circuit not only requires the detector photocurrent to be kept at a minimum value, but also requires the RF signal to be kept at a minimum value. Therefore, a mirror current source, RF detector, etc. have to be introduced into the control circuit. In addition, the control sequence of the MZI in the IQ modulator is very different from the control sequence of the MZI in the cascaded MZI wavelength division multiplexer. Therefore, the above control method is not suitable for wavelength division multiplexers. Summary of the invention

[0008] To solve the above technical problems, the present invention provides an adaptive control method for a wavelength division multiplexer. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0009] The present invention adopts the following technical solution:

[0010] The present invention provides an adaptive control method for a wavelength division multiplexer, comprising: performing cyclic stepping on the MZI units in each level in order from low level to high level; wherein the stepping process is that when the PS voltage and the MPD current change in the same direction, the current PS voltage is reduced by one step, otherwise it is increased by one step.

[0011] Furthermore, the process of performing cyclic stepping on the MZI units in each level in order from low level to high level includes: after initialization, performing the first stepping on the MZI units in the first level; after completing the first stepping on all the MZI units from low level to high level, restarting the second stepping on the MZI units in the first level, and repeating the cycle.

[0012] Furthermore, the MZI units in the same order perform stepping simultaneously.

[0013] Furthermore, the process of executing the cyclic stepping includes:

[0014] S11: Let x=1, set the PS voltage of the initial cycle V(0)=V(x-1), the PS voltage of the first cycle V(1)=V(x)=V(x-1)+Vstep, where V(x) is the PS voltage of the xth cycle, I(x) is the MPD current of the xth cycle, Vstep is the voltage change step and Vstep>0;

[0015] S12: Read I(x) under V(x), read I(x-1) under V(x-1), calculate ΔV=V(x)-V(x-1), ΔI=I(x)-I(x-1);

[0016] S13: If ΔV×ΔI>0, set V(x+1)=V(x)-Vstep;

[0017] S14: If ΔV×ΔI≤0, let V(x+1)=V(x)+Vstep;

[0018] S15: After waiting for all the MZI units in the current cycle to complete stepping, set x=x+1 and return to step S12.

[0019] Furthermore, the process of executing the cyclic stepping includes:

[0020] S21: Let x=1, set the PS voltage of the initial cycle V(0)=V(x-1), the PS voltage of the first cycle V(1)=V(x)=V(x-1)+Vstep, where V(x) is the PS voltage of the xth cycle, I(x) is the MPD current of the xth cycle, Vstep is the voltage change step and Vstep>0;

[0021] S22: read I(x) under V(x), read I(x-1) under V(x-1);

[0022] S23:

[0023] If V(x)>V(x-1) and I(x)>I(x-1), let V(x+1)=V(x)-Vstep;

[0024] If V(x)≤V(x-1) and I(x)≤I(x-1), let V(x+1)=V(x)-Vstep;

[0025] If V(x)>V(x-1) and I(x)≤I(x-1), let V(x+1)=V(x)+Vstep;

[0026] If V(x)≤V(x-1) and I(x)>I(x-1), let V(x+1)=V(x)+Vstep;

[0027] S24: After waiting for all the MZI units in the current cycle to complete stepping, set x=x+1 and return to step S22.

[0028] Furthermore, in the process of executing the cyclic stepping, the first voltage change step is first used to execute the stepping, and when the MPD current approaches the target minimum point and jumps over the false minimum point, the second voltage change step is used to execute the stepping, and the value of the first voltage change step is greater than the value of the second voltage change step.

[0029] Furthermore, when the PS voltage continuously steps 10 to 50 Vsteps, if the difference between the maximum and minimum PS voltages is less than a preset threshold, it is determined that the MPD current is close to the target minimum point; wherein the preset threshold is 0.04 to 0.2 times of V 周期 , V 周期 is the distance between the PS voltages corresponding to two adjacent target minimum points of the MPD current.

[0030] Furthermore, the first voltage change step is a fixed value, and the value range of the first voltage change step is 0.01 to 0.05 times of V 周期 The second voltage change step range is 0.001 to 0.02 times V 周期 .

[0031] Furthermore, the value of the first voltage change step is a non-fixed value that changes cyclically within a limited interval, and the limited interval is located at 0.01 times V 周期 ~0.05 times V 周期 The second voltage change step range is 0.001 to 0.02 times V 周期 .

[0032] The beneficial effects brought by the present invention are as follows: there is no need to use disturbance (Dither) signals, detectors and other components in the circuit, the feedback adjustment method of the present invention is simple and reliable, and can realize full-time automatic correction of the wavelength division multiplexer channel relative to the wavelength of the light source; it solves the problem of expected value of wavelength drift of the wavelength division multiplexer caused by process errors and wavelength deviation of the light source and wavelength deviation of the wavelength division multiplexer channel during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 It is a schematic diagram of the structure of the passive MZI filter;

[0035] Figure 2 It is a schematic diagram of the structure of a controllable MZI filter;

[0036] Figure 3 It is a simplified schematic diagram of the MZI unit;

[0037] Figure 4 It is the structural diagram of cascaded MZI wavelength division multiplexer;

[0038] Figure 5 It is a step sequence diagram of the control circuit of the present invention;

[0039] Figure 6 This is a flowchart of one of the minimum value tracking methods;

[0040] Figure 7 This is another flow chart of the minimum value tracking method;

[0041] Figure 8 It is the photocurrent minimum tracking curve;

[0042] Fig. 9 It is a structural diagram of a 4×1 wavelength division multiplexer;

[0043] Fig.10 This is the MZI filter spectrum of the 4×1 wavelength division multiplexer in the best state;

[0044] Fig.11 This is the MZI filter spectrum of the 4×1 wavelength division multiplexer in the offset state;

[0045] Fig.12 It is the curve diagram of PS voltage changing with period after the control circuit is turned on;

[0046] Fig.13 It is the curve diagram of MPD current changing with period after the control circuit is turned on;

[0047] Fig.14 This is the filter spectrum before the control circuit is turned on;

[0048] Fig.15 is a filtered spectrum diagram after executing the control method of the present invention;

[0049] Fig.16 It is a comparison chart of the working spectrum before and after the control circuit is turned on. DETAILED DESCRIPTION

[0050] The embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. It should be clear that the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The MPD mentioned in the following text and the accompanying drawings of the present invention refers to a photodetector, PS refers to a phase shifter, the first stage refers to the first stage, the second stage refers to the second stage, the third stage refers to the third stage, and so on.

[0052] The present invention provides an adaptive control method for a wavelength division multiplexer, in particular to an adaptive control scheme for a cascaded MZI wavelength division multiplexer.

[0053] The following first introduces the specific structure and working principle of the cascaded MZI wavelength division multiplexer:

[0054] like Figure 1 The figure shows a schematic diagram of a passive MZI filter, which includes two input ports 101 and two output ports 102, two phase-shift waveguides 103, one optical splitter 104, and one optical combiner 105. Figure 1 A phase shifter 106 is added to one of the phase shift waveguides, and a photodetector 107 is connected to one of the output ports. The photodetector 107 and the phase shifter 106 can be connected through a control circuit 108 to form a Figure 2 The controllable MZI filter shown in Figure 1 is Figure 2 One output port of the controllable MZI filter is occupied by the photodetector 107, so it can be expressed as follows Figure 3 The MZI unit shown includes two input ports 101, one output port 102, one photodetector 107, one phase shifter 106 and a control circuit 108. The MZI unit can be an MZI unit composed of a single MZI or an MZI unit composed of multiple MZIs. The purpose of using multiple MZIs is to make the filtering spectrum of the MZI unit form a flat top or increase the optical bandwidth.

[0055] like Figure 4 and Figure 5 As shown, the cascaded MZI wavelength division multiplexer based on the MZI unit includes: n Input ports can be connected to 2 n One-to-one correspondence between the path wavelength; 1 output port; 2 n -1 MZI unit, each MZI unit contains 1 MPD and 1 PS.

[0056] Each MZI unit contains 2 input ports and 2 output ports. n The input ports are connected to the MZI units in the first stage in pairs. There are 2 MZI units in the first stage. n / 2, the second-level MZI unit has 2 n / 4...There is one MZI unit in the last level, with a total of n levels, that is, the number of MZI units in each adjacent higher level is half the number of MZI units in the adjacent lower level. One output port of each MZI unit is connected to the input port of the MZI unit in the next higher level, until one output port of an MZI unit in the nth level is used as the sum wave output of the entire wavelength division multiplexer, forming a reverse binary tree structure based on the MZI unit.

[0057] 2 n The wavelength interval of the input ports is △λ, so the free spectral range of the MZI unit in the first stage should be designed to be 2 n ×△λ, the second stage should be designed to be 2 n-1 ×△λ, the last stage should be designed to be 2 1 ×△λ=2△λ.

[0058] 2 n -1 MPD photocurrent signal 109 and 2 n -1 PS voltage signal 110 is used as the input and output of the control circuit 108 respectively, that is, the PS voltage of the present invention refers to the voltage applied by the control circuit 108 to the phase shifter in the MZI unit, and the MPD current refers to the photocurrent signal output from one of the output ports in the MZI unit detected by the control circuit through the photodetector.

[0059] like Figure 4 As shown, the direction of the top arrow in the figure is the step determination sequence of the control circuit 108 in a single cycle; Figure 5 As shown in the figure, the direction of the arrow on the black solid line is the control order of the cycle step. Figure 4-5 As shown, level 1 of 2 n-1 MZI units, 2 in total at level 2 n-2 MZI units... n-2th level, 2 in total 2 MZI units, 2 in total at level n-1 1 MZI units, 2 in total for the nth level 0 MZI units. Among them, the first level is the lowest level, the nth level is the highest level, and the order from the low level to the high level mentioned in the present invention refers to the arrangement order from the first level, the second level... the n-2th level, the n-1th level until reaching the nth level. Figure 4-5In the example, the MZI units in each level are numbered. MZIk-m refers to the mth MZI unit in the kth level, k is 1, 2, 3...n, and m is 1, 2, 3...2 n-1 , wherein the arrangement order of the MZI units in each stage can be any arrangement order.

[0060] The adaptive control method of the present invention comprises: performing cyclic stepping on the MZI units in each level in order from low level to high level.

[0061] Specifically, the process of performing cyclic stepping on the MZI units in each level in order from low level to high level includes: after initialization, first performing the first stepping on the MZI units in the first level, that is, stepping 1 step, and then performing the first stepping on the MZI units in the second level in order; after completing the first stepping on all the MZI units from low to high levels, restarting the second stepping on the MZI units in the first level, and repeating the cycle.

[0062] It is understandable that the present invention can perform stepping on MZI units in the same order in any order, such as performing stepping operations sequentially. Preferably, MZI units in the same order perform stepping at the same time, which can save stepping time and improve control efficiency.

[0063] For the cascaded MZI wavelength division multiplexer, in the process of multiplexing multiple wavelengths from multiple waveguides to a single waveguide, it is hoped that as much signal light as possible passes through the multiplexer (coherent enhancement at one output end of the MZI); when the MPD is set at the optical destructive end of the MZI, it is hoped that as little signal light as possible is output from the destructive end (coherent destructive at another output end of the MZI), so for the MZI in the cascaded wavelength division multiplexer, when the optical signal received by the MPD reaches a minimum value, it is the optimal working point of the MZI. Therefore, the present invention adjusts the voltage on the PS by using a minimum value tracking and holding circuit, so that the MPD photocurrent always maintains a minimum value, and processes the MPD and PS in all MZI units according to the control method and control sequence of the present invention, so as to realize the adaptive control of the cascaded MZI wavelength division multiplexer.

[0064] Among them, the process of executing the step is that when the PS voltage and the MPD current change in the same direction, the current PS voltage is reduced by one step, otherwise it is increased by one step. The PS voltage and the MPD current change in the same direction means that when the PS voltage value increases, the MPD current value also increases, and when the PS voltage value decreases, the MPD current value also decreases. Correspondingly, the PS voltage and the MPD current change in different directions means that when the PS voltage value increases, the MPD current value decreases, and when the PS voltage value decreases, the MPD current value increases.

[0065] like Figure 6 The flowchart shown is a control method used by each MZI unit to track the minimum value of the MPD current in a cycle, that is, the process of executing the cycle stepping of the present invention includes:

[0066] S11: Let x=1, set the PS voltage of the initial cycle V(0)=V(x-1), the PS voltage of the first cycle V(1)=V(x)=V(x-1)+Vstep, where V(x) is the PS voltage of the xth cycle, I(x) is the MPD current of the xth cycle, Vstep is the voltage change step and Vstep>0;

[0067] S12: Read I(x) under V(x), read I(x-1) under V(x-1), calculate ΔV=V(x)-V(x-1), ΔI=I(x)-I(x-1);

[0068] S13: If △V×△I>0, it means that the PS voltage and the MPD current change in the same direction. In this case, V(x+1)=V(x)-Vstep, that is, reduce the step size by one.

[0069] S14: If △V×△I≤0, it means that the PS voltage and the MPD current change in different directions. In this case, V(x+1)=V(x)+Vstep, that is, increase the step size by one.

[0070] S15: After waiting for all the MZI units in the current cycle to complete stepping, set x=x+1, return to step S12, and start the next cycle of cyclic control.

[0071] like Figure 7 The flowchart shown is another control method used by each MZI unit to track the minimum value of the MPD current in a cycle period, that is, another process of performing cyclic stepping of the present invention includes:

[0072] S21: Let x=1, set the PS voltage of the initial cycle V(0)=V(x-1), the PS voltage of the first cycle V(1)=V(x)=V(x-1)+Vstep, where V(x) is the PS voltage of the xth cycle, I(x) is the MPD current of the xth cycle, Vstep is the voltage change step and Vstep>0;

[0073] S22: read I(x) under V(x), read I(x-1) under V(x-1);

[0074] S23: Judgment process:

[0075] S231: First determine whether I(x)>I(x-1) is satisfied;

[0076] S232: Determine whether V(x)>V(x-1) is satisfied;

[0077] S233: If V(x)>V(x-1) and I(x)>I(x-1), it indicates that the PS voltage and the MPD current change in the same direction. In this case, V(x+1)=V(x)-Vstep, that is, reduce the step size by one.

[0078] S234: If V(x)≤V(x-1) and I(x)≤I(x-1), it indicates that the PS voltage and the MPD current change in the same direction. In this case, V(x+1)=V(x)-Vstep, that is, reduce the step size by one.

[0079] S235: If V(x)>V(x-1) and I(x)≤I(x-1), it indicates that the PS voltage and the MPD current change in different directions. In this case, V(x+1)=V(x)+Vstep, that is, increase the step size by one.

[0080] S236: If V(x)≤V(x-1) and I(x)>I(x-1), it indicates that the PS voltage and the MPD current change in different directions. In this case, V(x+1)=V(x)+Vstep, that is, increase the step size by one.

[0081] S24: After waiting for all the MZI units in the current cycle to complete stepping, set x=x+1, return to step S22, and start the next cycle of cyclic control.

[0082] like Figure 6 and Figure 7 The process shown is explained by taking an MZI unit as an example. When executing the cyclic stepping, the MZI units in each level are stepped according to the above process. For some extreme cases, such as when △V>0 and △I=0, Figure 6 and Figure 7 The method shown uses an increase in step size; when △V<0 and △I=0, Figure 6 The method shown uses an additional step size, Figure 7 The method shown uses a step size reduction; when △V=0 and △I=0, Figure 6 The method shown uses an additional step size, Figure 7 The method shown adopts one step reduction.

[0083] like Figure 8 As shown, V MZI is the PS voltage, I MPD is the MPD current.

[0084] When initialization is completed and the cycle control begins, the voltage V on PS MZI As the number of cycles increases (the value of cycle x gradually increases), it gradually moves toward IMPD The minimum value is close. When Vx+1 crosses the minimum value, I MPD The change is reversed, the control circuit will make V MZI The stepping direction is reversed, and finally, V MZI The MZI unit will cycle between Vx-1, Vx, and Vx+1 in the figure. When the voltage change step VStep is small enough, the MZI unit will always remain at I MPD Near the minimum value, the function of tracking and holding the minimum value is realized.

[0085] One advantage of the control method of the present invention is that it can achieve real-time adjustment of the central wavelength of the MUX channel during the operation of the system. Because the target minimum point of the MPD current is locked by real-time monitoring and comparison of the MPD current size during the PS voltage stepping process, the method is not affected by the change of the input signal power, and the central wavelength of each MUX channel can be automatically aligned with the input light wavelength when the input light wavelength drifts.

[0086] In the process of executing cyclic stepping, the first voltage change step is first used to execute stepping. When the MPD current approaches the target minimum point and jumps over the false minimum point, the second voltage change step is used to execute stepping, and the value of the first voltage change step is greater than the value of the second voltage change step.

[0087] Optionally, the first voltage change step is a fixed value, and the value range of the first voltage change step is 0.01 to 0.05 times of V 周期 , that is, 0.01 to 0.05 times V 周期 A fixed value is selected within the range as the first voltage change step; the second voltage change step range is 0.001 to 0.02 times V 周期 The first voltage change step size and the second voltage change step size have the advantage of higher control efficiency within the above numerical range.

[0088] Because the power of the input light will change in different types of communication systems and different working hours, causing the target minimum point of the MPD current to change accordingly, so the absolute value of the MPD current cannot be used to determine whether the MPD current is at a false minimum point or close to the target minimum point. Device process errors, temperature changes, and input light wavelength drift will cause the PS voltage corresponding to the target minimum point of the MPD current to change accordingly, so the absolute value of the PS voltage cannot be used to determine whether the MPD current is at a false minimum point or close to the target minimum point.

[0089] Therefore, the first voltage change step of the present invention can be a non-fixed value that cycles in a certain interval, so as to automatically skip the false extreme value point. Specifically, the value of the first voltage change step is a non-fixed value that cycles in a limited interval, and the limited interval is between 0.01 times V 周期 ~0.05 times V 周期 The range of the second voltage change step is 0.001 to 0.02 times of V 周期 .

[0090] When there is a need to skip a false extreme point, the first voltage change step size can be cyclically changed with the number of cycles, for example: Vstep[n] = [0.05, 0.06, 0.07, 0.08, 0.09, 0.10, 0.05, 0.06, 0.07, 0.08, 0.09, 0.10…], in V; for example: Vstep[n] = [0.05, 0.06, 0.07, 0.08, 0.09, 0.10, 0.09, 0.08, 0.07, 0.06, 0.05, 0.06, 0.07, 0.08, 0.09, 0.10, 0.09, 0.08, 0.07, 0.06…], in V.

[0091] The principle is that the PS voltage range corresponding to the pseudo-extreme point of the MPD current is much smaller than the PS voltage range corresponding to the target minimum point of the MPD current, that is, the range of monotonic changes of the MPD current with the PS voltage on both sides of the pseudo-extreme point of the MPD current is much smaller than the voltage range of monotonic changes of the MPD current with the PS voltage on both sides of the target minimum point. Therefore, the cyclic change of the first voltage change step will always cause the PS voltage to deviate from the pseudo-extreme point at a certain time, thereby skipping the pseudo-extreme point through the correct voltage step direction determination. Specifically, the range and step size of the cyclic change of the first voltage change step can be debugged according to the performance of the MUX device actually manufactured.

[0092] If the cyclic change of the first voltage change step is reasonably chosen, the PS voltage will not be repeatedly cycled near the pseudo extreme point of the MPD photocurrent, but will cross the pseudo extreme point, and finally the PS voltage will be repeatedly cycled near the target minimum point of the MPD photocurrent.

[0093] When the PS voltage continuously steps 10 to 50 Vsteps, if the difference between the maximum and minimum PS voltages is less than a preset threshold, it is determined that the MPD current is close to the target minimum point; the preset threshold is 0.04 to 0.2 times V 周期 , V 周期 is the distance between the two adjacent target minimum points of the MPD current corresponding to the PS voltage, that is, the periodic voltage of the MPD current changing with the PS voltage is V 周期 . It can ensure the accuracy of the judgment results and thus ensure the accuracy of the control.

[0094] The determination of the Vstep size depends on the control accuracy of the MZI unit. The smaller the VStep, the higher the control accuracy, but more cycles and longer time are required to reach a stable state. When a false extreme point appears in the transmission spectrum of the MZI unit, the false extreme point can be skipped by temporarily increasing VStep. You can try a larger VStep first during the stepping process to let V MZI Fast Approaching I MPD The minimum point is selected and the false extreme point is skipped; then a smaller VStep is used to improve the control accuracy after stabilization.

[0095] PS is set on each MZI unit of the cascaded MZI structure, and MPD is set at the optical phase cancellation outlet. The present invention uses a minimum value tracking and holding circuit to adjust the voltage on the PS so that the MPD current always maintains a minimum value. The minimum value tracking and holding circuit uses the "read → determine → step" method to control the filter unit. The MZI units of each level are controlled in order of level within a single cycle. The high-level MZI unit is controlled after the low-level MZI unit, and the MZI units of the same level can be controlled at the same time. VStep in the control circuit can change according to the control requirements to achieve faster stabilization, skipping false extreme points, and higher precision. Moreover, the minimum value tracking and holding circuit of the present invention does not require a disturbance signal, an integration circuit, or a harmonic monitoring function, and the circuit is simple and effective to implement.

[0096] like Fig. 9 As shown, the present invention is explained by taking a 4×1 wavelength division multiplexer as an example, which has 4 input ports, namely n=2. The 4×1 wavelength division multiplexer based on the present invention contains a 2-stage MZI cascade structure, the first stage is MZI1-1 and MZI1-2, and the second stage is MZI2-1; 3 phase shifters (PS); and 3 photodetectors (MPD).

[0097] The first input port 201 and the second input port 202 are connected to the two input ports of MZI1-1, and the third input port 203 and the fourth input port 204 are connected to the two input ports of MZI1-2. The lower output ports of MZI1-1 and MZI1-2 are respectively connected to the two input ports of MZI2-1, and the upper output ports of MZI1-1 and MZI1-2 are respectively connected to the two MPDs and connected to the phase shifters of MZI1-1 and MZI1-2 through the control circuit. The lower output port of MZI2-1 serves as the combined output port of the 4×1 wavelength division multiplexer, and the upper output port of MZI2-1 is connected to the MPD and connected to the phase shifter of MZI2-1 through the control circuit.

[0098] like Fig.10As shown, when the 4×1 wavelength division multiplexer works in the optimal state, the intervals of the Cross and Bar path transmission spectra of its three MZI units, wavelength 1, wavelength 2, wavelength 3, and wavelength 4 are all △λ, then the free spectral range (peak spacing of the same transmission spectrum) of MZI1-1 and MZI1-2 is 4 times △λ, and the phase difference is π / 2. The peaks of the Cross / Bar path transmission spectrum of MZI1-1 are aligned with wavelength 1 / wavelength 3, and the peaks of the Cross / Bar path transmission spectrum of MZI1-2 are aligned with wavelength 2 / wavelength 4. The free spectral range of MZI2-1 is 2 times △λ, and the peaks of the Cross / Bar path transmission spectrum are aligned with (wavelength 1+wavelength 3) / (wavelength 2+wavelength 4).

[0099] When light of wavelength 1 is incident on the first input port 201 and passes through MZI1-1, the peak of the cross path transmission spectrum of MZI1-1 is aligned with wavelength 1, so that all the light of wavelength 1 passes through the lower output port of MZI1-1. When light of wavelength 1 continues to pass through MZI2-1, the peak of the cross path transmission spectrum of MZI2-1 is aligned with wavelength 1, so that all the light of wavelength 1 passes through the lower output port of MZI2-1.

[0100] When light of wavelength 3 is incident on the second input port 202 and passes through MZI1-1, the peak of the Bar path transmission spectrum of MZI1-1 is aligned with wavelength 3, so that all the light of wavelength 3 passes through the lower output port of MZI1-1. When continuing to pass through MZI2-1, the peak of the Cross path transmission spectrum of MZI2-1 is aligned with wavelength 3, so that all the light of wavelength 3 passes through the lower output port of MZI2-1.

[0101] When light of wavelength 2 is incident on the third input port 203 and passes through MZI1-2, the peak of the Cross path transmission spectrum of MZI1-2 is aligned with wavelength 2, so that all the light of wavelength 2 passes through the lower output port of MZI1-2. When it continues to pass through MZI2-1, the peak of the Bar path transmission spectrum of MZI2-1 is aligned with wavelength 2, so that all the light of wavelength 2 passes through the lower output port of MZI2-1.

[0102] When light of wavelength 4 is incident on the fourth input port 204 and passes through MZI1-2, the peak of the Bar path transmission spectrum of MZI1-2 is aligned with wavelength 4, so that all the light of wavelength 4 passes through the lower output port of MZI1-2. When continuing to pass through MZI2-1, the peak of the Bar path transmission spectrum of MZI2-1 is aligned with wavelength 4, so that all the light of wavelength 4 passes through the lower output port of MZI2-1.

[0103] like Fig.11As shown in the figure, when the 4×1 wavelength division multiplexer channel is offset or the wavelength of the light source is offset, take wavelength 1 and wavelength 3 passing through MZI1-1 as an example: the peak of the Cross path transmission spectrum of MZI1-1 is no longer aligned with wavelength 1 and wavelength 3. Therefore, when wavelength 1 and wavelength 3 pass through MZI1-1, not all of them pass through the Cross path, but some of them pass through the Bar path and are detected by the MPD connected to the Bar port.

[0104] The automatic control method of the present invention adjusts the PS voltage on MZI1-1 to keep the MPD current at a minimum value, thereby achieving the purpose of keeping MZI1-1 in the best working state in real time. When the multi-stage MZI wavelength division multiplexer is automatically controlled, the simultaneous change of the working state of the multi-stage MZI wavelength division multiplexer may cause an error in the control judgment of a certain high-level MZI unit. Therefore, in a single control cycle, all MZI units are read, judged, and step-controlled step by step, and the MZI units in the same level can be adjusted at the same time.

[0105] Effects of the present invention examples:

[0106] Take 4×1 wavelength division multiplexer as an example. Fig.12 and Fig.13 As shown, after the control program is started, the PS voltage and MPD current changes on the three MZI units of the 4×1 wavelength division multiplexer are changed. The initial voltages of the PS voltages on the three MZI units are all set to 2.00V and Vstep=0.01V.

[0107] As the number of control cycles increases, the PS voltages on the three MZI units stabilize at V MZI1-1 =2.28V, V MZI1-2 =2.72V, V MZI2-1 =1.99V. Where, V MZI1-1 is the PS voltage on MZI1-1, V MZI1-2 is the PS voltage on MZI1-2, V MZI2-1 It refers to the PS voltage on MZI2-1.

[0108] I MZI1-1 and I MZI1-2 First follow V MZI1-1 and V MZI1-2 After the PS voltage stabilizes, it also stabilizes at the minimum value. MZI2-1 It first increases with the change of PS voltage on the three MZI units because the light intensity entering MZI2-1 is constantly increasing during the process of MZI1-1 and MZI1-2 reaching the optimal state. In the non-simultaneous "read → judge → step" process with MZI1-1 and MZI1-2, MZI2-1 will eventually reach the optimal state. MZI1-1is the MPD current on MZI1-1, I MZI1-2 refers to the MPD current on MZI1-2, I MZI2-1 It refers to the MPD current on MZI2-1.

[0109] like Fig.14 and Fig.15 The figure shows the 4-channel filtering spectrum of the 4×1 wavelength division multiplexer without any voltage (offset state) and after the control circuit is turned on and stabilized (optimal state). After reaching the optimal state, the loss spectrum peaks of the 4 channels are aligned with the wavelengths allowed to pass.

[0110] like Fig.16 The figure shows the output optical power spectrum of the 4×1 wavelength division multiplexer under the input of 4 narrow linewidth light sources without any voltage (offset state) and after the control circuit is turned on and stabilized (optimal state). The loss of the wavelength division multiplexer in the optimal state is significantly reduced compared with that in the offset state.

[0111] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. An adaptive control method for a wavelength division multiplexer, characterized in that: The adaptive control method comprises: performing cyclic stepping on the MZI units in each level in order from a low level to a high level; The process of executing the step is that when the PS voltage and the MPD current change in the same direction, the current PS voltage is reduced by one step, otherwise it is increased by one step; In the process of executing the cyclic stepping, the stepping is first performed using the first voltage change step size, and when the MPD current approaches the target minimum point and jumps over the false maximum point, the stepping is then performed using the second voltage change step size, and the value of the first voltage change step size is greater than the value of the second voltage change step size; When the PS voltage continuously steps 10 to 50 Vsteps, if the difference between the maximum and minimum PS voltages is less than a preset threshold, it is determined that the MPD current is close to the target minimum point, where the preset threshold is 0.04 to 0.2 times of V 周期 , V 周期 is the distance between the PS voltages corresponding to two adjacent target minimum points of the MPD current.

2. The adaptive control method of a wavelength division multiplexer according to claim 1, characterized in that: The process of performing cyclic stepping on the MZI units in each level in order from the lower level to the higher level includes: After initialization, the first step is performed on the MZI unit in the first level; after the first step is completed for all the MZI units from low to high levels, the second step is restarted for the MZI unit in the first level, and this cycle is repeated.

3. The adaptive control method of a wavelength division multiplexer according to claim 2, characterized in that: The MZI units in the same order perform stepping simultaneously.

4. The adaptive control method of a wavelength division multiplexer according to claim 3, characterized in that: The process of executing the loop stepping includes: S11: Let x=1, set the PS voltage of the initial cycle V(0)=V(x-1), the PS voltage of the first cycle V(1)=V(x)=V(x-1)+Vstep, where V(x) is the PS voltage of the xth cycle, I(x) is the MPD current of the xth cycle, Vstep is the voltage change step and Vstep>0; S12: Read I(x) under V(x), read I(x-1) under V(x-1), calculate ΔV=V(x)-V(x-1), ΔI=I(x)-I(x-1); S13: If ΔV×ΔI>0, set V(x+1)=V(x)-Vstep; S14: If ΔV×ΔI≤0, let V(x+1)=V(x)+Vstep; S15: After waiting for all the MZI units in the current cycle to complete stepping, set x=x+1 and return to step S12.

5. The adaptive control method of a wavelength division multiplexer according to claim 3, characterized in that: The process of executing the loop stepping includes: S21: Let x=1, set the PS voltage of the initial cycle V(0)=V(x-1), the PS voltage of the first cycle V(1)=V(x)=V(x-1)+Vstep, where V(x) is the PS voltage of the xth cycle, I(x) is the MPD current of the xth cycle, Vstep is the voltage change step and Vstep>0; S22: read I(x) under V(x), read I(x-1) under V(x-1); S23: If V(x)>V(x-1) and I(x)>I(x-1), let V(x+1)=V(x)-Vstep; If V(x)≤V(x-1) and I(x)≤I(x-1), let V(x+1)=V(x)-Vstep; If V(x)>V(x-1) and I(x)≤I(x-1), let V(x+1)=V(x)+Vstep; If V(x)≤V(x-1) and I(x)>I(x-1), let V(x+1)=V(x)+Vstep; S24: After waiting for all the MZI units in the current cycle to complete stepping, set x=x+1 and return to step S22.

6. The adaptive control method of a wavelength division multiplexer according to claim 1, characterized in that: The first voltage variation step is a fixed value.

7. The adaptive control method of a wavelength division multiplexer according to claim 6, characterized in that: The value range of the first voltage change step is 0.01 to 0.05 times of V 周期 .

8. The adaptive control method of a wavelength division multiplexer according to claim 7, characterized in that: The second voltage change step range is 0.001 to 0.02 times V 周期 .

9. The adaptive control method of a wavelength division multiplexer according to claim 1, characterized in that: The value of the first voltage change step is a non-fixed value that changes cyclically within a limited interval, and the limited interval is located at 0.01 times V 周期 ~0.05 times V 周期 within the range.

10. The adaptive control method of a wavelength division multiplexer according to claim 9, characterized in that: The value range of the second voltage change step is: 0.001 to 0.02 times V 周期 .

Citation Information

Patent Citations

  • Maximum power point implementation method and device base on three-level step length

    CN105159387A

  • Electro-optical modulator optical pulse shaping apparatus and shaping method introducing reference light

    CN105572915A