Optical modulation apparatus and method using thin-film lithium niobate modulation

Abstract

The present invention relates to an optical modulation device and method using thin-film lithium niobate modulation, wherein the optical output ratio is calculated by measuring the optical intensity at the input and output terminals of a thin-film lithium niobate Mach-Zehnder modulator (LIFN MZM), and the DC bias is controlled without a pilot tone based on the calculated optical output ratio, and the predicted bias voltage is calculated by utilizing a lookup table of multiple bins classified according to drift characteristics, and the device automatically switches to a suitable bin according to changes in drift characteristics during operation. By measuring the optical intensity at the input and output terminals of the optical modulator to calculate the optical output ratio and controlling the DC bias without a pilot tone based on this, the device has the effect of fundamentally preventing jitter caused by a pilot tone and thereby improving signal quality. In addition, the present invention has the effect of enabling stable bias control by accurately identifying the operating point without being affected by fluctuations in the intensity of the input light source by utilizing the ratio of input light to output light, and by providing a lookup table composed of multiple bins classified according to drift characteristics, it is possible to immediately obtain a predicted bias voltage for fast driving and to respond quickly in a feedforward manner when disturbances such as temperature changes occur.

Description

Technology Field

[0001] The present invention relates to an optical modulation technology using thin-film lithium niobate modulation, and more specifically, to an optical modulation device and method using thin-film lithium niobate modulation that controls the DC bias without a pilot tone based on the optical output ratio calculated by measuring the optical intensity at the input and output terminals of a thin-film lithium niobate Mach-Zehnder modulator (TFLN MZM), calculates a predicted bias voltage using a lookup table of multiple bins classified according to drift characteristics, and automatically switches to a suitable bin according to changes in drift characteristics during operation. Background Technology

[0002] Recently, Linear Drive Pluggable Optics (LPO) technology is gaining attention in optical communication systems used in AI data centers and high-performance computing environments. This technology eliminates the DSP inside the optical module and allows the Serializer / Deserializer (SerDes) of the Application-Specific Integrated Circuit (ASIC) to directly drive optical devices in order to minimize data transmission delay and power consumption. In such environments, an optical modulator capable of ultra-high-speed modulation of over 50 Gbps with high linearity is essential.

[0003] An optical modulator in an optical communication system receives continuous wave (CW) light output from a laser diode and transmits information by changing the intensity, phase, or frequency of the light according to the transmission data (RF data) to be transmitted.

[0004] The Mach-Zehnder Modulator (MZM) is known as an optical modulator used in ultra-high-speed communication utilizing LPO technology. This Mach-Zehnder modulator evenly distributes the input light into two waveguide paths and then applies different voltages to each path to create a difference in refractive index, thereby forming a phase difference between the light passing through the two paths. Subsequently, when the light from the two paths is combined again, constructive interference (brightness) or destructive interference (darkness) occurs depending on the phase difference, thereby modulating the intensity of the light.

[0005] The output optical intensity of a Mach-Zehnder modulator is the cosine square (cos 2 It represents a transmission characteristic curve of the form ). In this transmission characteristic curve, at least one of the following is used as an operating point to control the phase difference: the peak point where the output is maximum, the null point where the output is minimum, and the quadrature point (Q-Point) where the rate of change of the output is most linear. In particular, in NRZ (Non-Return-to-Zero) modulation schemes, it is common to operate at the quadrature point, which has excellent linearity.

[0006] Meanwhile, conventional Mach-Zehnder modulators control the refractive index of the optical path using materials such as Bulk LiNbO3, InP, and SiPh, but these applied devices have limitations such as relatively large size, high power consumption, and low bandwidth.

[0007] Thin-film lithium niobate (LiNbO3, TFLN), which has recently been attracting increasing interest, is a promising material capable of meeting these requirements. Compared to conventional bulk lithium niobate, TFLN has the advantages of miniaturizing device size to less than a few millimeters, lowering driving voltage to less than a few volts, and enabling ultra-high-speed modulation of over 100 GHz. However, due to the characteristics of its thin-film structure, TFLN has a problem in that it is sensitive to DC drift caused by temperature changes and charge accumulation.

[0008] DC drift refers to the phenomenon in which the operating point of a modulator shifts depending on time, temperature, or applied voltage. For example, even if the operating point is initially set to a quadrature point, if the operating point shifts toward a peak point or null point due to drift, modulation efficiency is reduced and signal distortion occurs. Therefore, DC bias control technology capable of compensating for drift is essential for the stable operation of a TFLN modulator.

[0009] Among the known DC bias control methods using a TFLN modulator, one is a method of checking drift using a pilot tone. The pilot tone method superimposes a low-frequency dithering signal onto the DC bias voltage and applies it to the modulator, and tracks the operating point by lock-in detection of the component of the dithering signal in the output light.

[0010] However, the method using pilot tones has a fatal problem in that the pilot tones themselves act as jitter in the signal. In other words, a low-frequency dithering signal superimposed on the DC bias causes the modulator's operating point to vibrate slightly, resulting in unwanted fluctuations in the output optical signal and degrading signal quality.

[0011] In addition, when checking drift using only pilot tones, there is a limitation in that it is difficult to distinguish between fluctuations in the intensity of the input light source and DC drift. That is, since the pilot tone method monitors only the intensity of the output light, it is not possible to distinguish whether the change in light intensity is due to fluctuations in the output of the laser diode or a change in modulation efficiency is due to drift.

[0012] The method of applying a dither signal such as a pilot tone and performing bias voltage control considering drift based on it has a high probability of errors and inevitably has a lot of noise, so there is a limitation in that the quality decreases as the communication speed increases.

[0013] In addition, because it is a feedback control method, the bias voltage is controlled subsequently by synchronizing the pilot tone detected from the output light, so it is not easy to maintain a stable control loop.

[0014] Meanwhile, due to the device characteristics of the TFLN, the bias point changes rapidly when voltage is first applied after manufacturing and then stabilizes as it converges to a constant level; therefore, a long-term burn-in test is performed, and control parameters are set based on information regarding drift characteristics measured during the burn-in test. The TFLN MZM, which utilizes a pilot tone, controls the bias voltage in accordance with the drift characteristics by changing the values ​​of the control parameters based on the change in modulation efficiency obtained by analyzing the pilot tone included in the output.

[0015] Therefore, long-term burn-in testing is essential, which leads to a problem of reduced mass production efficiency. Furthermore, if such drift characteristics change, long-term stability cannot be guaranteed through control parameter settings determined based on initial burn-in test results; consequently, communication quality deteriorates after a certain period of use, resulting in a shortened lifespan and incurring the burden of time, cost, and communication interruptions due to recalibration.

[0016] Therefore, there is a need for a new technology that enables accurate DC bias control without using dither signals, which are the cause of jitter, effectively responds to changes in drift characteristics, and allows for feedforward control; furthermore, it can reduce burn-in test time while controlling according to drift characteristics, or automatically change control settings to suit drift characteristics resulting from aging even if aging occurs due to use. Prior art literature

[0017] Korean Registered Patent No. 10-1070409 [Title: Mach-Zehnder Optical Modulator] Korean Registered Patent No. 10-2892845 [Title: Mach-Zehnder Optical Modulator Including Multimode Interferometer] Korean Registered Patent No. 10-2162833 [Title: Tunable U-Laser Transmitter Equipped with Integrated Mach-Zehnder Modulator] The problem to be solved

[0018] The objective of the present invention, which is to solve the above-mentioned problems, is to provide an optical modulation device and method using thin-film lithium niobate modulation that prevents jitter generation by measuring optical intensity at the input and output terminals of an optical modulator to calculate the optical output ratio, and controlling the DC bias without a pilot tone based on this, thereby enabling accurate identification of the operating point regardless of fluctuations in the intensity of the input light source.

[0019] Another objective of the present invention is to provide an optical modulation device and method using thin-film lithium niobate modulation that can shorten the initial driving time and improve the response speed to disturbances through feedforward control by having a lookup table in which a DC bias voltage according to temperature, voltage application time, and optical output ratio is recorded and a predicted bias voltage is obtained immediately.

[0020] Another objective of the present invention is to provide an optical modulation device and method using thin-film lithium niobate modulation that shortens the calibration time in the mass production stage by configuring a lookup table with a plurality of bins grouping elements with similar drift characteristics and selecting a suitable bin based on the results of an initial test over a short period of time.

[0021] Another objective of the present invention is to provide an optical modulation device and method using thin-film lithium niobate modulation that supports stable operation throughout its lifespan without recalibration by monitoring the error between the predicted bias voltage and the actual bias voltage during operation to detect changes in drift characteristics and providing a self-revining function that automatically switches to a suitable bin. means of solving the problem

[0022] An optical modulation device using thin-film lithium niobate modulation according to one embodiment of the present invention may include: a laser diode that outputs an optical signal; a modulation unit comprising a TFLN modulator that modulates and outputs light input from the laser diode according to a DC bias voltage and a transmission data signal, and a temperature sensor that measures the temperature of the TFLN modulator; an input tap that branches off a portion of the input light in an optical path between the laser diode and the modulation unit; an output tap that branches off a portion of the output light in an optical path of the output end of the modulation unit; an input monitoring photodiode that measures the intensity of the light branched from the input tap; an output monitoring photodiode that measures the intensity of the light branched from the output tap; and a composite modulation control unit that calculates an optical output ratio based on the measurement values ​​of the input monitoring photodiode and the output monitoring photodiode, obtains a predicted bias voltage by referring to a lookup table in which a DC bias voltage according to the optical output ratio, temperature, and voltage application time is stored, and calculates a DC bias voltage based on the predicted bias voltage and the optical output ratio and applies it to the modulation unit.

[0023] As an example related to the present invention, the composite modulation control unit may include a dual-tap diagnostic unit that calculates in real time an optical output ratio, which is the ratio of the measured value of the output monitoring photodiode to the measured value of the input monitoring photodiode.

[0024] As an example related to the present invention, the lookup table is composed of a plurality of bins classified into groups having similar characteristics based on drift characteristics measured from a plurality of TFLN modulator samples, and each bin may store a DC bias voltage value in a three-dimensional grid structure of temperature, voltage application time, and optical output ratio.

[0025] Furthermore, the above-described composite modulation control unit may include a rebining unit that reanalyzes the drift slope and switches from the current bin to another bin when the error between the predicted bias voltage of the lookup table and the actual calculated bias voltage exceeds a reference value.

[0026] At this time, the above-mentioned rebining unit can gradually transition by mixing the value of the previous bin and the value of the new bin according to weights during bin transition.

[0027] As an example related to the present invention, the composite modulation control unit can perform feed-forward control that preemptively adjusts the DC bias voltage without feedback delay when detecting a temperature change based on the lookup table.

[0028] As an example related to the present invention, the composite modulation control unit can analyze the switching pattern of the bin and generate an alarm signal indicating that the time for replacing the device is imminent.

[0029] A method for optical modulation using thin-film lithium niobate modulation according to an embodiment of the present invention may include: a step of configuring a lookup table in an optical modulation device, wherein drift characteristics are measured in a plurality of TFLN modulator samples and classified into groups having similar characteristics, and a DC bias voltage is mapped for each group according to temperature, voltage application time, and optical output ratio, and a lookup table is configured with a plurality of bins; a step of recording information in the optical modulation device that selects one of the plurality of bins of the lookup table according to the measured drift characteristics by performing an initial test on the TFLN modulator of the optical modulation device; a step of calculating an optical output ratio by measuring light intensity at the input and output terminals of a modulation unit including a TFLN modulator and a temperature sensor configured inside the optical modulation device; and a step of the optical modulation device obtaining a predicted bias voltage by referring to a bin of the lookup table according to the recorded selection information based on the optical output ratio, the temperature of the temperature sensor, and the voltage application time, and calculating a DC bias voltage based on the predicted bias voltage and the optical output ratio and applying it to the TFLN modulator of the modulation unit.

[0030] As an example related to the present invention, the optical modulation device may further include the step of reanalyzing the drift slope and switching to another bin when the error between the predicted bias voltage and the actual calculated bias voltage exceeds a reference value.

[0031] Meanwhile, the step of switching the above bin may also be performed gradually by mixing the value of the previous bin and the value of the new bin according to weights. Effects of the invention

[0032] An optical modulation device and method using thin-film lithium niobate modulation according to an embodiment of the present invention can improve signal quality by measuring optical intensity at the input and output terminals of an optical modulator, respectively, to calculate the optical output ratio, and controlling the DC bias without a pilot tone based on this, thereby fundamentally preventing the occurrence of jitter caused by a pilot tone.

[0033] In addition, the optical modulation device and method using thin-film lithium niobate modulation according to an embodiment of the present invention can accurately identify the operating point without being affected even if the intensity of the input light source fluctuates by using the ratio of input light to output light, thereby having the effect of enabling stable bias control.

[0034] In addition, the optical modulation device and method using thin-film lithium niobate modulation according to an embodiment of the present invention have the effect of enabling rapid operation by immediately obtaining a predicted bias voltage by providing a lookup table composed of a plurality of bins classified according to drift characteristics, and enabling rapid response in a feedforward manner when disturbances such as temperature changes occur.

[0035] In addition, the optical modulation device and method using thin-film lithium niobate modulation according to an embodiment of the present invention have the effect of reducing mass production costs by enabling control that matches the drift characteristics of individual devices without a long-term burn-in test, by allowing a suitable bin to be selected through a short-term initial test.

[0036] In addition, the optical modulation device and method using thin-film lithium niobate modulation according to an embodiment of the present invention provide a self-revining function that detects changes in drift characteristics during operation and automatically switches to a suitable bin, thereby enabling stable operation throughout the lifespan without device stabilization or recalibration due to aging. Brief explanation of the drawing

[0037] FIG. 1 is a configuration diagram of a pilot tone-based optical modulation device according to the prior art. FIG. 2 is a configuration diagram of an optical modulation device using thin-film lithium niobate modulation according to an embodiment of the present invention. FIG. 3 is a flowchart of a light modulation method using thin-film lithium niobate modulation according to an embodiment of the present invention. Specific details for implementing the invention

[0038] It should be noted that the technical terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. Furthermore, unless specifically defined otherwise in this invention, the technical terms used in this invention should be interpreted in the sense generally understood by those skilled in the art to which this invention pertains, and should not be interpreted in an overly broad or overly narrow sense. Additionally, if a technical term used in this invention is an incorrect technical term that fails to accurately express the concept of the invention, it should be replaced with a technical term that can be correctly understood by those skilled in the art. Moreover, general terms used in this invention should be interpreted according to their prior definitions or the context, and should not be interpreted in an overly narrow sense.

[0039] Furthermore, singular expressions used in the present invention include plural expressions unless the context clearly indicates otherwise. Terms such as "composed of" or "comprising" in the present invention should not be interpreted as necessarily including all of the various components or steps described in the invention, and should be interpreted as meaning that some of the components or steps may not be included, or that additional components or steps may be included.

[0040] Additionally, terms including ordinal numbers, such as first, second, etc., used in the present invention may be used to describe components, but the components should not be limited by the terms. The terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component.

[0041] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. Identical or similar components are given the same reference number regardless of the drawing symbols, and redundant descriptions thereof will be omitted.

[0042] Furthermore, in describing the present invention, detailed descriptions of related prior art are omitted if it is determined that such descriptions could obscure the essence of the invention. Additionally, it should be noted that the attached drawings are intended only to facilitate an understanding of the concept of the present invention and should not be interpreted as limiting the concept of the present invention.

[0043] FIG. 1 is a configuration diagram of a pilot tone-based optical modulation device (10) according to the prior art.

[0044] As described above, a conventional optical modulation device (10) comprises a laser diode (11), an optical modulator (12), a tap (13), a monitoring photodiode (14), a synchronization detector (15), a control unit (16), and a pilot tone generator (17).

[0045] The laser diode (11) generates a continuous wave (CW) optical signal used in optical communication and outputs it to the optical modulator (12). The laser diode (11) may be a distributed feedback laser diode (DFB-LD), a Fabry-Perot laser diode (FP-LD), etc.

[0046] The optical modulator (12) modulates the light input from the laser diode (11) according to a transmission data signal (RF signal data for transmission) and outputs it. The optical modulator (12) may be configured as a Mach-Zehnder modulator (MZM) structure, distributes the input light into two waveguide paths, forms a phase difference between the two paths due to a change in refractive index according to the applied voltage, and then combines the light from the two paths to modulate the intensity of the light through an interference phenomenon according to the phase difference.

[0047] The pilot tone generator (17) generates a low-frequency dithering signal in the range of several kHz to tens of kHz. The low-frequency signal generated by the pilot tone generator (17) is superimposed on a DC bias voltage and applied to the optical modulator (12). This low-frequency signal serves to cause the operating point of the optical modulator (12) to vibrate slightly.

[0048] The control unit (16) outputs a DC bias voltage to maintain the operating point of the optical modulator (12) at an optimal position (e.g., a quadrature point). The DC bias voltage output from the control unit (16) is combined with the low-frequency dithering signal generated by the pilot tone generator (17) and applied to the optical modulator (12). That is, the DC bias voltage, the low-frequency dithering signal, and the transmission data signal are applied together to the optical modulator (12).

[0049] The above tap (13) is placed at the output end of the optical modulator (12) to divert a portion of the output light (e.g., 1 to 5%). The above tap (13) may be composed of an optical fiber coupler, a beam splitter, etc., and the diverted light is transmitted to the monitoring photodiode (14), and the remaining light is transmitted to the receiving end through an optical path.

[0050] The monitoring photodiode (14) receives light branched from the tap (13) and converts it into an electrical signal proportional to the light intensity. The electrical signal output from the monitoring photodiode (14) includes a modulated data component, a component due to DC bias, and a low-frequency component due to a pilot tone.

[0051] The synchronization detection unit (15) extracts and analyzes a pilot tone component from an electrical signal received from the monitoring photodiode (14). The synchronization detection unit (15) may be configured to include a lock-in amplifier and a low-frequency filter, and separates and detects the fundamental (first harmonic) component and the second harmonic component of the pilot tone.

[0052] When the operating point of the above optical modulator (12) is precisely located at the quadrature point, it corresponds to the symmetric point of the transmission characteristic curve, so the fundamental component of the pilot tone is minimized (ideally 0) and the second harmonic component is maximized. On the other hand, when the operating point deviates from the quadrature point, the fundamental component increases, and depending on the phase, it is possible to determine which direction the operating point has moved. The above synchronization detection unit (15) uses this principle to provide a feedback signal to the control unit (16) to make the fundamental component 0, and the control unit (16) adjusts the DC bias voltage based on this to maintain the optimal operating point.

[0053] However, this conventional pilot tone-based optical modulation device (10) has the following problems.

[0054] First, the low-frequency dithering signal generated by the pilot tone generator (17) and superimposed on the DC bias continuously vibrates the operating point of the optical modulator (12). As a result, minute fluctuations corresponding to the pilot tone frequency occur in the output optical signal, which acts as jitter in the signal and degrades signal quality. In particular, in an ultra-high-speed communication environment of 50 Gbps or higher, such jitter becomes a factor that increases the bit error rate (BER).

[0055] In addition, since the above-mentioned tap (13) and monitoring photodiode (14) only measure the intensity of the output light, it is difficult to distinguish from a change in modulation efficiency caused by drift when the output of the laser diode (11) is unstable and the intensity of the input light fluctuates. That is, it is not possible to distinguish whether the decrease in output light intensity is due to a decrease in input light intensity or a decrease in modulation efficiency due to a shift in the operating point, so incorrect bias control may occur.

[0056] Meanwhile, due to significant variations in the manufacturing process of TFLN devices, drift characteristics (drift speed, temperature sensitivity, etc.) differ for each individual device. Therefore, conventional methods require performing burn-in tests for extended periods (e.g., several hours to tens of hours) to determine the drift characteristics of each device, which increases mass production costs.

[0057] Furthermore, during long-term operation, the drift characteristics themselves may change due to the aging of the device, but it is difficult to respond to this. TFLN modulators (122) generally exhibit saturation characteristics in which the drift speed is fast at the time of shipment, but decreases and stabilizes as time passes. However, since the conventional method does not have the function to adapt to such changes in drift characteristics, the control parameters set at the beginning may not be valid after long-term operation.

[0058] To solve the problems of the prior art, the present invention proposes an optical modulation device (100) that enables accurate DC bias control without using a dither signal that causes jitter, effectively responds to changes in drift characteristics, and allows for feedforward control, as well as reducing burn-in test time while controlling according to drift characteristics, and automatically changes control settings to suit drift characteristics due to aging even if aging occurs due to use.

[0059] FIG. 2 shows the configuration of an optical modulation device (100) using thin-film lithium niobate modulation according to an embodiment of the present invention.

[0060] An optical modulation device (100) which is an embodiment of the present invention is configured to include a laser diode (110), a modulation unit (120), an input tap (131), an output tap (132), an input monitoring photodiode (141), an output monitoring photodiode (142), a composite modulation control unit (150), and a digital-to-analog converter (160) as illustrated.

[0061] The laser diode (110) is a light source that generates and outputs an optical signal used in optical communication. The laser diode (110) may be composed of a distributed feedback laser diode (DFB-LD), a Fabry-Perot laser diode (FP-LD), a vertical resonant plane emitting laser (VCSEL), etc., and may output continuous wave light from one of several optical communication wavelength bands including 1310 nm and 1550 nm. The optical signal output from the laser diode (110) is transmitted to the modulation unit (120) via the input tap (131).

[0062] The above modulation unit (120) includes a TFLN modulator (122) and a temperature sensor (121).

[0063] The above TFLN modulator (122) modulates the light input from the laser diode (110) according to the DC bias voltage and transmission data (RF data signal) and outputs it. The TFLN modulator (122) according to an embodiment of the present invention may be configured as a Mach-Zehnder modulator structure fabricated based on a thin-film lithium niobate material.

[0064] The above TFLN modulator (122) may include an optical splitter that evenly distributes input light into two waveguide paths, a phase modulation arm placed in each path that changes the refractive index according to the applied voltage, and an optical coupler that combines the light of the two paths. The TFLN modulator (122) uses the electro-optical effect (Pockels Effect) to change the refractive index of each path according to the applied voltage, and modulates the intensity of the light by generating constructive or destructive interference through the resulting phase difference.

[0065] The above TFLN modulator (122) has the advantage of being smaller in size than a few millimeters compared to a conventional bulk lithium niobate modulator, having a low driving voltage of a few volts or less, and being capable of ultra-high-speed modulation of 100 GHz or more. However, due to the characteristics of the thin film structure, DC drift caused by temperature changes and charge accumulation occurs sensitively, so precise bias control is required to compensate for this, and in the present invention, this is performed by the composite modulation control unit (150).

[0066] In the above TFLN modulator (122), a DC bias voltage for setting the operating point and transmission data corresponding to the data to be transmitted are applied as voltages. The DC bias voltage serves to set the operating point of the TFLN modulator (122) to an optimal position on the transmission characteristic curve (e.g., a quadrature point), and the transmission data serves to modulate the intensity of light around the set operating point and transmit the data by carrying it on an optical signal.

[0067] The temperature sensor (121) of the modulation unit (120) measures the temperature of the TFLN modulator (122) in real time and provides it to the composite modulation control unit (150). The temperature sensor (121) may be composed of an NTC thermistor, a PTC thermistor, a thermocouple, an RTD, a semiconductor temperature sensor, etc. The temperature sensor (121) is placed in close proximity to the chip surface or inside the package of the TFLN modulator (122) to accurately measure the actual operating temperature of the TFLN modulator (122).

[0068] The input tap (131) is placed in the optical path between the laser diode (110) and the modulation unit (120) to divert a portion of the input light. The input tap (131) may be composed of a fiber optic coupler, a beam splitter, a directional coupler, etc. The input tap (131) diverts 1 to 3 percent of the input light and transmits it to the input monitoring photodiode (141), and transmits the remaining light to the TFLN modulator (122) of the modulation unit (120).

[0069] The output tap (132) is placed in the output optical path of the modulation unit (120) to divert a portion of the output light. The output tap (132), like the input tap (131), may be composed of a fiber optic coupler, a beam splitter, a directional coupler, etc. The output tap (132) diverts 1 to 3 percent of the output light and transmits it to the output monitoring photodiode (142), and the remaining light is transmitted to the receiving end through the optical path.

[0070] The input monitoring photodiode (141) receives light branched from the input tap (131) and measures the intensity of the input light. The input monitoring photodiode (141) may be composed of a PIN photodiode, an APD (Avalanche Photodiode), etc., and outputs a current or voltage signal proportional to the intensity of the incident light.

[0071] The output monitoring photodiode (142) receives light branched from the output tap (132) and measures the intensity of the output light. The output monitoring photodiode (142) may also be composed of a PIN photodiode, an APD, etc., and outputs a current or voltage signal proportional to the intensity of the modulated output light.

[0072] The above composite modulation control unit (150) calculates an optical output ratio based on the ratio of the measured values ​​of the input monitoring photodiode (141) and the output monitoring photodiode (142), obtains a predicted bias voltage by referring to the lookup table (152), calculates a DC bias voltage based on the predicted bias voltage and the optical output ratio, and applies it to the modulation unit (120) through the digital-to-analog converter (160).

[0073] The above-mentioned composite modulation control unit (150) can be implemented as a combination of hardware such as an MCU (Micro Controller Unit), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit), DSP (Digital Signal Processor), and firmware or software running on said hardware. The above-mentioned composite modulation control unit (150) stores a lookup table (152) and setting parameters in non-volatile memory and processes the analog signals of the input monitoring photodiode (141), output monitoring photodiode (142), and temperature sensor (121) by converting them into digital data through an analog-to-digital converter.

[0074] In an embodiment of the present invention, the composite modulation control unit (150) includes a dual tap diagnostic unit (151), a lookup table (152), and a bias determination unit (153).

[0075] The dual tap diagnostic unit (151) receives the input light intensity (Pin) measured by the input monitoring photodiode (141) and the output light intensity (Pout) measured by the output monitoring photodiode (142), and calculates the light output ratio (R) in real time.

[0076] The above optical output ratio (R) can be calculated by the formula 'R = Pout / Pin' and may be modified depending on the optical ratio of the optical path or other conditions.

[0077] The core principle of bias control using the above optical output ratio (R) is as follows. When the operating point of the TFLN modulator (122) is fixed at a specific position (e.g., a quadrature point), the optical output ratio is maintained constant because the input light intensity and the output light intensity change at the same rate even if the output of the laser diode (110) changes. However, if the operating point moves due to DC drift, the modulation efficiency changes and the optical output ratio changes; therefore, by monitoring the optical output ratio, the movement of the operating point can be accurately detected regardless of the change in the input light source.

[0078] Meanwhile, there is a possibility that error situations may frequently occur where the DC bias cannot be properly generated in response to the movement direction of the drift determined solely by the optical output ratio, the movement of the operating point due to the change in drift characteristics, or the rapid change in modulation efficiency due to changes in the external environment, and there are limitations such as errors accumulating or the convergence of the operating point due to the bias voltage slowing down as it operates only in a feedback manner. Therefore, in the present invention, in addition to the optical output ratio, a lookup table (152) in which drift characteristic information of the TFLN modulator (122) is stored is used to verify the drift characteristics confirmed for the TFLN modulator (122) and the expected DC bias value considering these drift characteristics, and to use this as a standard for generating the DC bias control voltage to be applied to the TFLN modulator (122) in a complex manner so that the DC bias to be applied to the TFLN modulator (122) can be generated stably without error.

[0079] The above lookup table (152) is a data table in which DC bias voltages according to temperature, voltage application time and optical output ratio are stored, and is composed of multiple bins that group devices with similar drift characteristics based on test data for multiple sample TFLN modulators.

[0080] The process of creating the above lookup table (152) is explained as follows.

[0081] First, prepare multiple TFLN modulator samples produced from the same process. It is desirable to use multiple samples to obtain statistically significant results.

[0082] Next, drift characteristics are measured for each sample while varying the temperature and voltage application time. For example, the temperature of the operating range is varied stepwise, and the change in bias voltage according to the voltage application time at each temperature is recorded. In addition, the DC bias voltage required to maintain the target optical output ratio under each condition is measured and mapped.

[0083] Subsequently, the collected data is analyzed to group samples with similar drift characteristics. Clustering algorithms such as k-means clustering, hierarchical clustering, and Gaussian mixture models can be applied for grouping. For example, samples can be classified into groups based on drift speed or groups with similar drift change patterns.

[0084] For each group, information is generated by mapping a DC bias voltage (predicted bias voltage) to maintain an operating point under corresponding conditions to each grid using a three-dimensional grid structure of temperature, voltage application time, and optical output ratio, and this group-specific mapping information is divided into multiple bins and stored in a lookup table (152) of a composite modulation control unit (150). For example, the operating temperature range can be set at intervals of 1°C, the voltage application time at intervals according to a logarithmic scale, and the optical output ratio can be gridded at intervals of 0.05 in the range of 0 to 1, and then the predicted bias voltage at each grid position (e.g., the average value of the DC bias voltage measured at the corresponding grid position in the samples of the group) can be mapped. Some missing values ​​among the predicted bias voltages mapped for each grid can be calculated by a mathematical formula modeling the grid-specific measured DC bias voltage information of multiple samples divided into the corresponding bin, or determined through interpolation.

[0085] The lookup table information composed of multiple bins generated in this way is stored in a non-volatile memory configured in the lookup table (152) of the composite modulation control unit (150).

[0086] The lookup table (152) receives information about the light output ratio, temperature and time and provides a predicted bias value recorded in a designated bin, or provides relevant information so that interpolation is possible if the received information is a value between grids.

[0087] Meanwhile, when an individual TFLN modulator (122) is produced, a burn-in test must be performed. In an embodiment of the present invention, only a short burn-in process (e.g., 1 minute) is performed after the production of the TFLN modulator (122).

[0088] While performing such a short-time burn-in process, the drift slope (rate of change of bias voltage per hour) is measured during this period.

[0089] That is, in the embodiment of the present invention, long-term drift characteristics are predicted based on the drift slope measured through a short-time burn-in process, and a bin suitable for the device is selected, thereby significantly reducing the burn-in time.

[0090] Based on the observation that the drift of TFLN devices is caused by charge accumulation and redistribution phenomena, and that the time constant of this phenomenon is determined by the physical characteristics of the device, long-term drift characteristics are predicted through short-term initial tests. That is, there is a correlation between the initial drift gradient and the long-term drift characteristics, and this correlation can be modeled through regression analysis, neural network models (MLP, RNN, LSTM, etc.).

[0091] Based on the drift slope measured in the initial test, the bin that best matches the drift characteristics of the corresponding element among the multiple bins of the lookup table (152) is selected, and the identification information of the selected bin is stored in non-volatile memory.

[0092] The bias determination unit (153) determines the DC bias voltage by referring to the lookup table (152) based on the optical output ratio calculated by the dual tap diagnosis unit (151), the temperature information measured by the temperature sensor (121), and the voltage application time information accumulated and recorded by itself.

[0093] The above bias determination unit (153) includes a LUT Bin selection unit (153a), a bias calculation unit (153b), and a rebining unit (153c).

[0094] The LUT Bin selection unit (153a) manages initial selection bin information stored in non-volatile memory. When power is applied to the optical modulation device (100), the LUT Bin selection unit (153a) reads the stored bin information and sets the bin to be used currently. In addition, when a bin change request is received from the rebining unit (153c) described later, the current bin is changed to a new bin, and the changed bin information is updated and stored in non-volatile memory.

[0095] The bias calculation unit (153b) obtains a predicted bias voltage by referring to a lookup table of a bin selected in the LUT Bin selection unit (153a) based on the optical output ratio (R) calculated by the dual tap diagnosis unit (151), the current temperature (T) measured by the temperature sensor (121), and the voltage application time (t) accumulated and recorded by itself.

[0096] The bias calculation unit (153b) obtains a predicted bias voltage (V_predicted) at coordinates corresponding to the current temperature (T), voltage application time (t), and target light output ratio (R_target) by referring to the lookup table (152). If the coordinate values ​​do not exactly match the grid points, the predicted bias voltage can be calculated by obtaining information for interpolation and applying linear interpolation, trilinear interpolation, etc. If necessary, such interpolation may be performed in the lookup table (152).

[0097] The above bias calculation unit (153b) calculates the final bias voltage (V_final) by calculating a feedback correction value based on the difference between the current actual light output ratio (R_actual) and the target light output ratio (R_target), based on the predicted bias voltage.

[0098] V_final = V_predicted + K × (R_target - R_actual)

[0099] Here, K is the feedback gain, a parameter that determines the response speed and stability of the control loop.

[0100] The above bias calculation unit (153b) transmits the calculated final bias voltage in digital form to the digital-to-analog converter (160).

[0101] In this way, when the predicted bias voltage using the lookup table (152) is checked and used as a reference, the bias voltage to be applied at the current temperature can be obtained immediately without a voltage scan process during initial operation, enabling fast operation, and when disturbances such as sudden temperature changes occur, feed-forward control is possible to preemptively adjust the bias voltage based on the lookup table (152) without feedback delay, thereby improving the response speed.

[0102] In addition, although it is difficult to determine the drift direction based solely on the light output ratio, the drift direction can be identified by comparing the predicted value of the lookup table (152) with the actual value, thereby preventing malfunction.

[0103] Meanwhile, in the case of the TFLN modulator (122), the initial drift characteristics are stabilized by performing a long burn-in process, and since the drift characteristics change due to degradation from use, the selected bin of the lookup table (152) based on the drift characteristics measured immediately after mass production cannot be continuously used. Therefore, in the present invention, the rebining unit (153c) enables automatic switching to a bin that matches the current drift characteristics among the various bins configured in the lookup table (152).

[0104] The rebining unit (153c) monitors the error between the predicted bias voltage of the lookup table (152) and the actual calculated bias voltage, and if the error exceeds a reference value, it reanalyzes the drift slope and performs a self-rebining function of switching from the current bin to another bin.

[0105] Specifically, the rebining unit (153c) continuously monitors the error between the predicted bias voltage (V_predicted) and the actual calculated final bias voltage (V_actual), but if the error exceeds a preset threshold for a certain period of time or if the error accumulates and exceeds a certain standard, it determines that the drift characteristics according to the selected bin of the current lookup table (152) do not match the characteristics of the actual device.

[0106] In this case, the rebining unit (153c) calculates the currently measured drift slope (rate of change of bias voltage per hour) and compares it with the representative drift slope of each bin to select a candidate bin with the most similar drift characteristics and switches the bin in use.

[0107] Meanwhile, the rebining unit (153c) may apply a hysteresis technique to prevent a sudden change in bias voltage during bin switching. That is, a method of gradually switching by mixing the value of the previous bin and the value of the new bin according to weights may be used.

[0108] In addition, the rebining unit (153c) can analyze bin switching patterns (switching cycle, switching direction, etc.) and generate an alarm signal when a pattern indicating that the device lifespan is imminent is detected, such as reaching the final bin or the switching cycle becoming shorter.

[0109] Through this rebinding section (153c), stable operation is possible for the lifetime of the TFLN modulator (122) without recalibration.

[0110] The digital-to-analog converter (160) converts the DC bias voltage information (digital value) determined by the bias determination unit (153) into an analog voltage and applies it to the TFLN modulator (122) of the modulation unit (120). That is, in the optical modulation device (100) of the present invention, unlike conventional methods, a pure DC bias voltage in which pilot tone signals are not superimposed is applied to the TFLN modulator (122), so no jitter is generated.

[0111] FIG. 3 is a flowchart of a light modulation method using thin-film lithium niobate modulation according to an embodiment of the present invention.

[0112] The optical modulation method of the present invention is broadly divided into a lookup table generation step, a bin selection step, and an operation step.

[0113] First, in the lookup table generation step, drift characteristics are measured from multiple TFLN modulator samples to generate a lookup table consisting of multiple bins.

[0114] Specifically, multiple TFLN modulator samples produced in the same process are prepared, and drift characteristics are measured for each sample while varying the temperature and voltage application time. For example, the temperature is varied from -40°C to 80°C in 5°C increments, and the amount of change in bias voltage according to the voltage application time and the DC bias voltage required to maintain the target optical output ratio are recorded at each temperature. Alternatively, the amount of change in bias voltage and the DC bias voltage required to maintain the target optical output ratio may be recorded while varying the temperature in accordance with the progression of the voltage application time.

[0115] Collected drift data is analyzed to group samples with similar drift characteristics. Clustering algorithms such as k-means clustering, hierarchical clustering, and Gaussian mixture models can be applied for grouping. For example, based on drift speed, samples can be classified into very slow group (Bin 1), slow group (Bin 2), normal group (Bin 3), fast group (Bin 4), and very fast group (Bin 5).

[0116] For each group, individual bins are created by mapping DC bias voltage values ​​to a three-dimensional grid structure of temperature, voltage application time, and optical output ratio. A lookup table composed of multiple created bins is stored in the non-volatile memory of the composite modulation control unit (150).

[0117] Next, in the bin selection step, an initial test is performed on individual TFLN modulators to select a suitable bin, and the selection information of the bin is stored in the composite modulation control unit (150).

[0118] Specifically, a short-term (e.g., 1 minute) burn-in process is performed on each individual TFLN modulator (122) produced, and the drift slope (rate of change of bias voltage per hour) is measured. Based on the measured drift slope, the bin that best matches the drift characteristics of the corresponding device is selected from among a plurality of bins in a lookup table. The identification information of the selected bin is stored in the non-volatile memory of the composite modulation control unit (150) and shipped.

[0119] Subsequently, during the operation phase, the DC bias is controlled based on the optical output ratio, and self-revining is performed if necessary.

[0120] When power is applied to the optical modulation device (100), the LUT Bin selection unit (153a) of the composite modulation control unit (150) reads the initial selection bin information stored in the non-volatile memory and sets the bin to be used currently.

[0121] When the laser diode (110) starts operating, light branched from the input tap (131) is transmitted to the input monitoring photodiode (141) to measure the input light intensity (Pin). At the same time, light output from the modulation unit (120) and branched from the output tap (132) is transmitted to the output monitoring photodiode (142) to measure the output light intensity (Pout).

[0122] The dual tap diagnostic unit (151) receives the input light intensity (Pin) and the output light intensity (Pout) and calculates the light output ratio (R = Pout / Pin).

[0123] The bias calculation unit (153b) obtains a predicted bias voltage (V_predicted) by referring to a lookup table of a selected bin based on the calculated optical output ratio (R), the current temperature (T) measured by the temperature sensor (121), and the voltage application time (t) accumulated and recorded by itself.

[0124] The bias calculation unit (153b) calculates the final bias voltage (V_final) by performing feedback correction based on the difference between the current measured optical output ratio and the target optical output ratio based on the predicted bias voltage.

[0125] The calculated final bias voltage is converted into an analog voltage through a digital-to-analog converter (160) and applied to a TFLN modulator (122). In the present invention, a pure DC bias voltage is applied without superimposing pilot tones.

[0126] The rebining unit (153c) continuously monitors the error between the predicted bias voltage and the actual calculated bias voltage. If the error continues to exceed a reference value, the rebining unit (153c) reanalyzes the drift slope and switches from the current bin to another bin. When switching bins, to prevent a sudden change in bias voltage, the value of the previous bin and the value of the new bin are mixed according to weights and switched gradually.

[0127] The above-described processes of calculating the optical output ratio, obtaining the predicted bias voltage, calculating the final bias voltage, and monitoring the error are continuously repeated while the optical modulation device (100) is in operation to maintain the optimal operating point of the TFLN modulator (122).

[0128] As described above, the optical modulation device according to the embodiment of the present invention controls the DC bias without a pilot tone based on the optical output ratio calculated by measuring the optical intensity at the input and output terminals of the TFLN MZM, calculates the predicted bias voltage by utilizing a lookup table of multiple bins classified according to drift characteristics, and automatically switches to a suitable bin according to changes in drift characteristics during operation, thereby increasing the yield of the optical modulation device and significantly improving its quality.

[0129] A person skilled in the art to which the present invention pertains will be able to make modifications and variations to the foregoing without departing from the essential characteristics of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical concept of the present invention, and the scope of the technical concept of the present invention is not limited by these embodiments. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention.

[0130] Furthermore, the optical modulation device may be implemented as a hardware component, a software component, and / or a combination of the hardware component and the software component.

[0131] Additionally, the components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions.

[0132] Software may include a computer program, code, instructions, or a combination of one or more of these, and may command the optical modulation device to operate as desired, either independently or collectively.

[0133] The various devices and components described in this specification may be implemented by hardware circuits (e.g., semiconductor-based logic circuits), optical systems (e.g., lenses, optical fibers, optical benches, semiconductor-based optical paths, etc.), firmware, software, or a combination thereof. For example, they may be implemented using transistors, logic gates, and electronic circuits in the form of various electrical structures, and lenses, optical paths, optical interfaces, filters, etc. in the form of optical structures. Explanation of the symbols

[0134] 10: Conventional optical modulation device 11: Laser diode 12: Optical modulator 13: Tap 14: Monitoring photodiode 15: Synchronization detector 16: Control unit 17: Pilot tone generator 100: Optical modulator 110: Laser diode 120: Modulation unit 121: Temperature sensor 122: TFLN Modulator 131: Input Tab 132: Output tab 141: Input monitoring photodiode 142: Output monitoring photodiode 150: Complex modulation control unit 151: Dual Tab Diagnostic Unit 152: Lookup Table 153: Bias determination unit 153a: LUT Bin selection unit 153b: Bias calculation unit 153c: Revining unit 160: Digital-to-Analog Converter

Claims

Claim 1 A laser diode that outputs an optical signal; a modulation unit comprising a TFLN modulator that modulates and outputs light input from the laser diode according to a DC bias voltage and a transmission data signal, and a temperature sensor that measures the temperature of the TFLN modulator; an input tap that branches off a portion of the input light in the optical path between the laser diode and the modulation unit; an output tap that branches off a portion of the output light in the optical path of the output end of the modulation unit; an input monitoring photodiode that measures the intensity of the light branched from the input tap; and an output monitoring photodiode that measures the intensity of the light branched from the output tap. An optical modulation device using thin-film lithium niobate modulation, comprising a composite modulation control unit that calculates an optical output ratio based on the measured values ​​of the input monitoring photodiode and the output monitoring photodiode, obtains a predicted bias voltage by referring to a lookup table in which a DC bias voltage according to the optical output ratio, temperature, and voltage application time is stored, and calculates a DC bias voltage based on the predicted bias voltage and the optical output ratio and applies it to the modulation unit, wherein the lookup table is composed of a plurality of bins classified by groups having similar characteristics based on drift characteristics measured from a plurality of TFLN modulator samples, and each bin stores a DC bias voltage value in a three-dimensional grid structure of temperature, voltage application time, and optical output ratio. Claim 2 An optical modulation device using thin-film lithium niobate modulation according to claim 1, wherein the composite modulation control unit includes a dual-tap diagnostic unit that calculates in real time an optical output ratio, which is the ratio of the measured value of the output monitoring photodiode to the measured value of the input monitoring photodiode. Claim 3 delete Claim 4 An optical modulation device using thin-film lithium niobate modulation according to claim 1, wherein the composite modulation control unit includes a rebining unit that reanalyzes the drift slope and switches from the current bin to another bin when the error between the predicted bias voltage of the lookup table and the actual calculated bias voltage exceeds a reference value. Claim 5 An optical modulation device using thin-film lithium niobate modulation according to claim 4, wherein the rebining unit gradually switches by mixing the value of the previous bin and the value of the new bin according to a weighting factor during bin switching. Claim 6 An optical modulation device using thin-film lithium niobate modulation according to claim 1, wherein the composite modulation control unit performs feed-forward control that preemptively adjusts the DC bias voltage without feedback delay upon detecting a temperature change based on the lookup table. Claim 7 An optical modulation device using thin-film lithium niobate modulation according to claim 4, wherein the composite modulation control unit analyzes the bin switching pattern and generates an alarm signal indicating that the time for replacing the device is imminent. Claim 8 A method for optical modulation using thin-film lithium niobate modulation, comprising: a step of configuring a lookup table in an optical modulation device, wherein drift characteristics are measured in a plurality of TFLN modulator samples and classified into groups having similar characteristics, and a DC bias voltage is mapped for each group according to temperature, voltage application time, and optical output ratio, and a lookup table composed of a plurality of bins; a step of recording information in the optical modulation device that selects one of the plurality of bins of the lookup table in accordance with the measured drift characteristics by performing an initial test on the TFLN modulator of the optical modulation device; a step of calculating an optical output ratio by measuring light intensity at the input and output terminals of a modulation unit including a TFLN modulator and a temperature sensor configured inside the optical modulation device; and a step of the optical modulation device obtaining a predicted bias voltage by referring to a bin of the lookup table according to the recorded selection information based on the optical output ratio, the temperature of the temperature sensor, and the voltage application time, and calculating a DC bias voltage based on the predicted bias voltage and the optical output ratio and applying it to the TFLN modulator of the modulation unit. Claim 9 A method for optical modulation using thin-film lithium niobate modulation according to claim 8, wherein the optical modulation device further includes the step of reanalyzing the drift slope and switching to another bin when the error between the predicted bias voltage and the actual calculated bias voltage exceeds a reference value. Claim 10 A light modulation method using thin-film lithium niobate modulation according to claim 9, wherein the step of switching the bin is characterized by gradually switching by mixing the value of the previous bin and the value of the new bin according to weights.

Images