A method and system for suppressing signal jitter in a multimode optical module

Abstract

A method and system for suppressing jitter in multimode optical module transmission signals are disclosed. The optical module includes a hardware execution unit and a digital control unit. The method includes: an acquisition step; a temperature compensation step, in which drive current compensation data is obtained based on the acquired hardware temperature data; and a signal equalization compensation step, in which initial first tap coefficients and initial second tap coefficients are determined based on the signal characteristic data, standard temperature data, and first and second tap coefficients read from the electrically erasable programmable read-only memory of the optical module; and equalization compensation data is obtained based on the initial first tap coefficients, initial second tap coefficients, first tap coefficients, and second tap coefficients. This invention, by setting up an acquisition step, a temperature compensation step, and a signal equalization calculation step, effectively suppresses jitter during long-distance transmission between multimode optical module devices.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method and system for suppressing jitter in the transmission signal of a multimode optical module. Background Technology

[0002] In the signal transmission process of multimode optical modules, jitter affects system reliability. Multimode fiber (MMF), due to its large core diameter (e.g., 50 / 62.5 μm) and the ability to transmit multiple optical modes (e.g., LP01, LP11, etc.) simultaneously, suffers from mode competition and dispersion effects between different modes, leading to time shifts in optical pulses during transmission and consequently causing signal jitter. The light source characteristics of multimode optical modules (e.g., the large divergence angle of VCSEL lasers) and dynamic changes in the link (e.g., temperature fluctuations, mechanical stress) further exacerbate jitter. In short-distance transmission scenarios in data centers, the jitter problem of multimode fiber is particularly prominent, potentially causing increased bit error rate (BER) and even resulting in pixelation or black screen phenomena in video signals.

[0003] In the signal transmission process of multimode optical modules, jitter caused by compatibility issues arises from problems when crossing link types or device interfaces. The core cause lies in the differences in physical characteristics between multimode and single-mode optical fibers, as well as mode compatibility failures between the optical module and the link port.

[0004] A search revealed a Chinese patent with publication number CN120238195A, which discloses a method and system for implementing fiber optic interconnect transmission using a high-speed serial transceiver based on FPGA. This method includes extracting jitter characteristics of the received signal in the fiber optic link, including jitter standard deviation and jitter peak value, and dynamically adjusting the clock data recovery CDR bandwidth based on the jitter characteristics and the current BER (bit error rate) to balance jitter tracking capability and noise suppression performance. It also utilizes nonlinear mapping rules to precisely control the recovery of the optical signal. This patent determines the joint quantization interval based on jitter characteristics and BER data, dynamically adjusts the CDR bandwidth level through a lookup table (LUT), and dynamically switches the FEC (Flexible Encoder Control) mode through hysteresis threshold logic, thereby improving signal jitter suppression capability and effectively enhancing the quality of the fiber optic communication link.

[0005] However, the above technical solutions still have at least the following drawbacks: jitter cannot be effectively suppressed during long-distance transmission between multimode optical module devices. Summary of the Invention

[0006] To address the above issues and overcome the shortcomings of existing technologies, this invention provides a method for suppressing jitter in multimode optical module transmission signals. The multimode optical module includes a hardware execution unit and a digital control unit. The switch communicates with the multimode optical module via electrical signals, and the multimode optical module communicates with other multimode optical modules by transmitting and receiving optical signals. The method includes the following steps executed by the digital control unit:

[0007] The acquisition step involves acquiring hardware temperature data and signal characteristic data of the received optical signal from the hardware execution unit.

[0008] The temperature compensation step involves obtaining drive current compensation data based on the hardware temperature data and the pre-stored standard temperature data, and sending it to the hardware execution unit to suppress jitter in the transmitted optical signal.

[0009] The signal equalization compensation step involves determining initial first tap coefficients and initial second tap coefficients based on the standard temperature data. Using these initial first tap coefficients and initial second tap coefficients as the starting point for iteration, the first and second tap coefficients read from the electrically erasable programmable read-only memory of the multimode optical module, along with the signal characteristic data, are used to obtain equalization compensation data based on linear and nonlinear filtering compensation algorithms. This data is then sent to the hardware execution unit to suppress jitter in the received optical signal.

[0010] A second aspect of the present invention provides a system for suppressing jitter in the transmission signal of a multimode optical module, the system comprising at least one processor; and a memory storing instructions which, when executed by the at least one processor, implement the steps of the method described in the first aspect.

[0011] Compared with existing technologies, this invention stabilizes the laser's output power and drive current by compensating for the drive current based on temperature. This effectively suppresses amplitude and dispersion jitter caused by temperature drift at the source, providing a more stable input signal for the signal equalization calculation step. The temperature compensation step also determines the initial first tap coefficient and initial second tap coefficient based on a selected standard temperature. By setting up the signal equalization calculation step, based on the standard temperature selected in the temperature compensation step, and using the initial first and second tap coefficients obtained from the EEPROM, linear filtering compensation algorithms, and nonlinear filtering compensation algorithms, the equalization compensation data, the decided signal, and the first and second tap coefficients at different times can be calculated. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating the method for suppressing jitter in the transmission signal of the multimode optical module according to the present invention;

[0013] Figure 2 This is a structural diagram of the multimode optical module of the present invention;

[0014] Figure 3 This is a schematic diagram of data transmission of the multimode optical module of the present invention;

[0015] Figure 4 This is a schematic diagram of data transmission between the multimode optical modules of the present invention;

[0016] Figure 5 This is a schematic diagram of the jitter suppression system for multimode optical module transmission signals according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] To facilitate understanding of the methods and systems provided in the embodiments of this application, the background of the embodiments of this application will be introduced before introducing the embodiments of this application.

[0019] In the signal transmission process of multimode optical modules, jitter affects system reliability. Multimode fiber (MMF), due to its large core diameter (e.g., 50 / 62.5 μm) and the ability to transmit multiple optical modes (e.g., LP01, LP11, etc.) simultaneously, suffers from mode competition and dispersion effects between different modes, leading to time shifts in optical pulses during transmission and consequently causing signal jitter. The light source characteristics of multimode optical modules (e.g., the large divergence angle of VCSEL lasers) and dynamic changes in the link (e.g., temperature fluctuations, mechanical stress) further exacerbate jitter. In short-distance transmission scenarios in data centers, the jitter problem of multimode fiber is particularly prominent, potentially causing increased bit error rate (BER) and even resulting in pixelation or black screen phenomena in video signals.

[0020] In the signal transmission process of multimode optical modules, jitter caused by compatibility issues arises from problems when crossing link types or device interfaces. The core cause lies in the differences in physical characteristics between multimode and single-mode optical fibers, as well as mode compatibility failures between the optical module and the link port.

[0021] A search revealed a Chinese patent with publication number CN120238195A, which discloses a method and system for implementing fiber optic interconnect transmission using a high-speed serial transceiver based on an FPGA. This method includes extracting jitter characteristics of the received signal in the fiber optic link, including jitter standard deviation and jitter peak value, and dynamically adjusting the clock data to restore the CDR bandwidth based on the jitter characteristics and the current BER (bit error rate), thus balancing jitter tracking capability and noise suppression performance. It also utilizes nonlinear mapping rules to precisely control the recovery of the optical signal. This patent determines the joint quantization interval based on jitter characteristics and BER data, dynamically adjusts the CDR bandwidth level through a lookup table (LUT), and dynamically switches the FEC (Flexible Encoder Control) mode through hysteresis threshold logic, thereby improving signal jitter suppression capability and effectively enhancing the quality of fiber optic communication links.

[0022] like Figures 1 to 5 As shown, some embodiments of the present invention relate to a method for suppressing jitter in the transmission signal of a multimode optical module. The multimode optical module includes a hardware execution unit and a digital control unit. The switch communicates with the multimode optical module via electrical signals, and the multimode optical module communicates with other multimode optical modules via transmitting and receiving optical signals. The method includes the following steps performed by the digital control unit:

[0023] The acquisition step involves acquiring hardware temperature data and signal characteristic data of the received optical signal from the hardware execution unit.

[0024] The temperature compensation step involves obtaining drive current compensation data based on the hardware temperature data and the pre-stored standard temperature data, and sending it to the hardware execution unit to suppress jitter in the transmitted optical signal.

[0025] The signal equalization compensation step involves determining initial first tap coefficients and initial second tap coefficients based on the standard temperature data. Using these initial first tap coefficients and initial second tap coefficients as the starting point for iteration, the first and second tap coefficients read from the electrically erasable programmable read-only memory of the multimode optical module, along with the signal characteristic data, are used to obtain equalization compensation data based on linear and nonlinear filtering compensation algorithms. This data is then sent to the hardware execution unit to suppress jitter in the received optical signal.

[0026] In these implementations, by setting temperature acquisition steps, signal feature acquisition steps, and response timing acquisition steps, the hardware execution unit is equipped with real-time hardware temperature data, switch timing data, and signal feature data. A temperature compensation module is included to calculate equalization compensation data and establish a temperature-compensation mapping relationship based on the drive current compensation algorithm, target output power, hardware temperature data, and pre-stored standard temperature data.

[0027] By compensating for the drive current based on temperature, the laser's output power and drive current are stabilized, effectively suppressing amplitude and dispersion jitter caused by temperature drift at the source, and providing a more stable input signal for the equalization calculation module. The temperature compensation module can also determine the initial first tap coefficient and the initial second tap coefficient based on the selected standard temperature.

[0028] By setting up a signal equalization calculation module, it is possible to calculate equalization compensation data, as well as the first and second tap coefficients at different times, based on the standard temperature selected in the temperature compensation module, the initial first tap coefficient and initial second tap coefficient obtained from the EEPROM, the linear filtering compensation algorithm and the nonlinear filtering compensation algorithm.

[0029] The standard temperature data is a reference temperature artificially selected based on the nominal operating temperature of the laser in the multimode optical module and the actual operating environment. The determination method includes the following steps: First, during the production and debugging phase of the optical module, a reference temperature point is selected within the rated operating temperature range of the laser, based on the optical module's grade (e.g., commercial or industrial) and the laser's hardware parameters. Second, the laser's power-current characteristics are tested at this reference temperature point to obtain the threshold current and differential slope efficiency at the standard temperature. Then, the aforementioned reference temperature point and the corresponding threshold current and slope efficiency are pre-stored in the electrically erasable programmable read-only memory (EEPROM) of the multimode optical module; this reference temperature point is the standard temperature data. During actual operation of the optical module, the digital control unit reads the pre-stored standard temperature data from the EEPROM as a reference, compares it with the real-time acquired hardware temperature data to calculate the temperature change, and thus achieves temperature compensation of the drive current. Preferably, the standard temperature data is 25℃ (i.e., room temperature) by default.

[0030] By configuring a response timing adjustment module, the timing data of the multimode optical module is adjusted based on the timing data of different switch models, resulting in timing adjustment data. The hardware execution unit communicates with the switch based on the timing adjustment data, thereby suppressing jitter.

[0031] Furthermore, by compensating the drive current based on temperature, and performing linear and nonlinear compensation on the input signal, the timing data of the multimode optical module is adjusted to achieve the dual objectives of temperature drift suppression and channel impairment equalization, ensuring signal quality for long-distance transmission (800 meters). Specifically, the input signal is a digital domain signal converted from the received optical signal by the hardware execution unit.

[0032] Furthermore, the digital signal processor located in the transmission link acquires the drive current compensation data and sends it to the digital-to-analog converter (DAC). The DAC converts the digital compensation signal into an analog electrical signal. The analog signal output by the DAC controls the driver, causing the driver to provide the corresponding drive current to the laser based on the current value in the compensation data. The laser operates under the compensated drive current, suppressing output power fluctuations caused by temperature drift. By stabilizing the output optical power, amplitude jitter and dispersion jitter caused by temperature are suppressed at the source, thereby outputting a stable transmitted optical signal.

[0033] Furthermore, the digital signal processor (DSP) is located within the hardware execution unit. Internally, the DSP uses the received linear filter compensation signal to perform feedforward filtering on the digital domain signal (i.e., the input signal) converted from the received optical signal, to eliminate deterministic jitter caused by modal dispersion and other factors. Subsequently, a nonlinear filter compensation signal is used to perform decision feedback filtering on the output of the previous stage, further eliminating residual nonlinear jitter caused by modal noise and other factors, thereby completing the jitter suppression of the received optical signal.

[0034] like Figure 2 and Figure 3 As shown, in some embodiments of methods for suppressing jitter in multimode optical module transmission signals, drive current compensation data is obtained based on hardware temperature data acquired from the temperature acquisition module, including:

[0035] The temperature change is obtained based on the hardware temperature data and the standard temperature data.

[0036] Based on the temperature change, the pre-stored target output power, and the drive current compensation algorithm, drive current compensation data is obtained.

[0037] The drive current compensation algorithm is shown in equation (1). (1);

[0038] The The calculation algorithm is shown in equation (2).

[0039] (2);

[0040] The The calculation algorithm is shown in equation (3).

[0041] (3);

[0042] Among them, the The temperature is Drive current compensation data at that time The temperature is Threshold current at that time For standard temperature data, The difference between the hardware temperature data and the standard temperature data. The temperature is The efficiency of the differential slope at that time. For the target output power, The standard temperature threshold current, The first temperature coefficient, The second temperature coefficient, For standard temperature slope efficiency, The third temperature coefficient, It is the fourth temperature coefficient.

[0043] In these implementations, the temperature compensation module calculates the compensated drive current using polynomial interpolation based on the power-current characteristic curves of the laser at different temperatures, as shown in Equation (1). The temperature-power-current characteristic curves are pre-stored in the electrically erasable programmable read-only memory (EEPROM) of the multimode optical module.

[0044] Furthermore, the standard temperature data can be manually selected based on the actual operating environment, with 25℃ being the preferred default setting. The temperature change is obtained by subtracting the hardware temperature data from the standard temperature data. The target output power value is pre-stored in the EEPROM of the multimode optical module, and this value is determined based on the laser's hardware parameters.

[0045] Furthermore, by calculating the driving current compensation algorithm, the temperature characteristics of the threshold current and slope efficiency are integrated to offset the effect of temperature on the laser's output power. For example, the power decrease caused by the increase in threshold current at high temperatures is mitigated, thus keeping the output power fluctuation of the multimode optical module within a minimal range across the entire temperature spectrum. This also avoids power over- or under-compensation caused by blind compensation. Moreover, the driving current compensation algorithm has a simple structure, is suitable for real-time hardware processing, and can meet the timing requirements of high-speed optical modules.

[0046] Furthermore, equation (2) calculates the temperature using a quadratic polynomial. The threshold current at that time can capture the nonlinear temperature change of the threshold current, avoiding insufficient or excessive compensation caused by the linear model.

[0047] Furthermore, equation (3) calculates the temperature using a quadratic polynomial. The differential slope efficiency of the power-current characteristic curves at different temperatures can capture the temperature nonlinear decay of the slope efficiency, avoiding power degradation caused by a decrease in slope efficiency. By compensating for the decrease in slope efficiency, it ensures that an increase in drive current can be effectively converted into an increase in output power, maintaining the stability of the target power.

[0048] Furthermore, , and The unit is mA, and the unit of standard temperature data is °C. The difference between the hardware temperature data and the standard temperature data, expressed in °C.

[0049] Furthermore, the specific steps for determining the first and second temperature coefficients are as follows: First, set the parameters for the temperature scan test, performing a temperature scan test within the actual operating temperature range of the target laser. Preferably, the operating temperature range is selected as -10℃ to 80℃; set the temperature test points T with a fixed step size of 10℃, for example: T={-10℃,0℃,10℃,...,80℃}. Second, obtain the power-current curves at each temperature point. At each of the above temperature points, perform a power-current characteristic test on the laser to obtain the corresponding power-current curve. Specifically, place the target laser in a constant temperature environment (such as a temperature control station) and stabilize it to a specific temperature point; apply an increasing driving current to the laser (covering the range from "no laser output" to "rated output power"), simultaneously measuring the corresponding output optical power; and plot the power-current curve at that temperature point. The third step involves performing the following operations on the power-current curve for each temperature point: Extracting the threshold current, which is the critical current at which the laser transitions from spontaneous emission to stimulated emission, corresponding to the inflection point in the power-current curve where power changes from nonlinear to linear growth with increasing current. This can be determined through curve fitting (e.g., quadratic curve fitting) or automatic instrument identification (e.g., the slope abrupt change point of an optical power meter). Extracting the differential slope efficiency: Selecting the linear region of the power-current curve (i.e., the region where power and current have a strictly linear relationship, usually the segment "after exceeding the threshold current"); performing least-squares linear fitting on the power-current data for this region, the slope of the fitted line is the differential slope efficiency (unit: mW / mA). The fourth step involves constructing a temperature-threshold current dataset, selecting a standard temperature (e.g., the nominal operating temperature of the target laser, usually 25℃); calculating the temperature change of each temperature test point T relative to the standard temperature, collecting data from all temperature test points, and constructing the dataset. The fifth step is to solve for the coefficients by quadratic polynomial fitting. The least squares method is used to fit the above data set with a quadratic polynomial, as shown in equation (2). By fitting calculation (such as the polyfit function in MATLAB or numpy.polyfit in Python), the specific values ​​of the first and second temperature coefficients can be obtained.

[0050] Furthermore, the specific steps for determining the third and fourth temperature coefficients are as follows: First, set the parameters for the temperature scan test, performing a temperature scan test within the actual operating temperature range of the target laser. Preferably, the operating temperature range is selected as -10℃ to 80℃; set the temperature test points T with a fixed step size of 10℃, for example: T={-10℃,0℃,10℃,...,80℃}. Second, obtain the power-current curves at each temperature point. At each of the above temperature points, perform a power-current characteristic test on the laser to obtain the corresponding power-current curve. Specifically, place the target laser in a constant temperature environment (such as a temperature control station) and stabilize it to a specific temperature point; apply an increasing driving current to the laser (covering the range from "no laser output" to "rated output power"), simultaneously measuring the corresponding output optical power; and plot the power-current curve at that temperature point. The third step involves performing the following operations on the power-current curve for each temperature point: Extracting the threshold current, which is the critical current at which the laser transitions from spontaneous emission to stimulated emission, corresponding to the inflection point in the power-current curve where power changes from nonlinear to linear growth with increasing current. This can be determined through curve fitting (e.g., quadratic curve fitting) or automatic instrument identification (e.g., the slope abrupt change point of an optical power meter). Extracting the differential slope efficiency: Selecting the linear region of the power-current curve (i.e., the region where power and current have a strictly linear relationship, usually the segment "after exceeding the threshold current"); performing least-squares linear fitting on the power-current data for this region, the slope of the fitted line is the differential slope efficiency (unit: mW / mA). The fourth step involves constructing a temperature-threshold current dataset, selecting a standard temperature (e.g., the nominal operating temperature of the target laser, usually 25℃); calculating the temperature change of each temperature test point T relative to the standard temperature, collecting data from all temperature test points, and constructing the dataset. The fifth step is to solve for the coefficients by fitting a quadratic polynomial. The least squares method is used to fit the above data set with a quadratic polynomial, as shown in equation (3). By fitting the data (such as the polyfit function in MATLAB or numpy.polyfit in Python), the specific values ​​of the third and fourth temperature coefficients can be obtained.

[0051] Furthermore, the first, second, third, and fourth temperature coefficients are all inherent parameters of the laser, ensuring the specificity of the compensation. Specifically, when an 850nm VCSEL is used as the laser, The range of values ​​is 1× Up to 5× , The range of values ​​is 1× Up to 1× , The value range is -5× to -1× , The range of values ​​is -1× to -1× Preferred, =0.0005、 =0.000001、 =-0.0001 and =-0.00001.

[0052] like Figure 2 and Figure 3 As shown, in some implementations of methods for suppressing jitter in multimode optical module transmission signals, the equalization compensation data includes linear filter compensation signals and nonlinear filter compensation signals;

[0053] The initial first tap coefficient, the first tap coefficient, and the signal feature data are used to obtain a linearly filtered and compensated signal based on a linear filtering and compensation algorithm.

[0054] The feedback filter output signal is obtained, and the initial second tap coefficient, the second tap coefficient, the linear filter compensation signal, and the feedback filter output signal are combined to obtain the nonlinear filter compensation signal based on the nonlinear filter compensation algorithm.

[0055] In some implementations of methods for suppressing jitter in multimode optical module transmission signals, the linear filtering compensation algorithm is shown in equation (4).

[0056] (4);

[0057] The nonlinear filtering compensation algorithm is shown in equation (5).

[0058] (5);

[0059] in, For the first Linear filtering compensation signal at each time point For the first The feedforward filter at the sampling time is... The first tap coefficient. For the first The input signal at each moment, The nonlinear filtering compensation signal is at time n. For the first The feedback filter at the sampling time is... The second tap coefficient For the first The signal has already been determined at this moment. This is the sequence number of the current sampling time. This is the tap index for the feedforward filter. For the feedback filter tap index, This refers to the advance payment amount. To delay the number of taps, For the feedback filter length, The output signal of the feedback filter;

[0060] The initial first tap coefficient is the value of the first tap coefficient at the start of the iteration update, and the initial second tap coefficient is the value of the second tap coefficient at the start of the iteration update.

[0061] In these embodiments, the present invention employs a phased optimization approach to compensate the received optical signal. Linear filtering compensation quickly eliminates major linear distortion and combats deterministic jitter caused by mode dispersion, while nonlinear filtering compensation further addresses residual nonlinear impairments, combats random jitter caused by mode noise, and improves compensation accuracy. The nonlinear filtering compensation, combined with a feedback mechanism, can respond in real-time to channel changes (such as temperature fluctuations and fiber bending) and adapt to different transmission scenarios.

[0062] Furthermore, the initial first tap coefficient is the value of the first tap coefficient at the start of the iteration update, and the initial second tap coefficient is the value of the second tap coefficient at the start of the iteration update. The initial first and second tap coefficients are a set of optimal solutions that have been factory calibrated at standard temperatures. Obtaining the initial first and second tap coefficients helps prevent the equalizer from converging from a random state, accelerating the convergence speed and reducing real-time computation latency. Both the first and second tap coefficients are pre-stored in the electrically erasable programmable read-only memory of the multimode optical module. The signal characteristic data includes the input signal at different times, such as the first... Input signal at each moment The signal characteristic data specifically consists of digital domain sampled values ​​obtained after the received optical signal is converted by the hardware execution unit, which characterize the instantaneous amplitude (or intensity) characteristics of the received optical signal at discrete sampling times.

[0063] Furthermore, the sequence number of the current sampling time. Essentially a dimensionless integer (sampling point number). Feedforward filter tap index. Essentially, it is a time offset in units of the sampling period, a dimensionless integer, and also serves as the tap index of the feedforward filter, with a value of [value missing]. Feedback filter tap index Essentially, it is a time offset in units of the sampling period, a dimensionless integer, i.e., the tap index of the feedback filter, with a value of... .

[0064] like Figure 3As shown, in some embodiments of methods for suppressing jitter in multimode optical module transmission signals, the method further includes:

[0065] The first tap coefficient calibration step involves acquiring the decided signal, and then dynamically adjusting the first tap coefficient based on the first tap coefficient adjustment algorithm using the decided signal, the linear filter compensation signal, the first tap coefficient, and the signal feature data to obtain the first calibration tap coefficient.

[0066] The first tap coefficient adjustment algorithm is shown in equation (6).

[0067] (6);

[0068] The The calculation algorithm is shown in equation (7). (7);

[0069] in, For the first The sampling time of the first sampling moment The first calibration tap coefficient. For the first The feedforward filter at the sampling time is... The first tap coefficient. For the first step, long parameters, For the first The first error signal at time 1 For the first The input signal at each moment, This is the sequence number of the current sampling time. This is the tap index for the feedforward filter. For the first The signal has already been determined at this moment. For the first Linear filtering compensation signal at each time point.

[0070] In these implementations, the problem of traditional fixed tap coefficients being unable to adapt to changes in channel characteristics is solved by adjusting the tap coefficients in real time. (For example, in high-temperature environments, when mode dispersion intensifies, dynamic adjustment can compensate for channel distortion.) Simultaneously, the first tap coefficient calibration module, combined with the calculated first error signal, forms a closed-loop feedback, improving compensation accuracy and enabling dynamic adjustment of the first tap coefficient. This, in turn, dynamically compensates the input signal, meeting the requirements of high-speed optical modules.

[0071] Furthermore, the decided signal is generated by the decision unit in the digital signal processor (DSP) performing a hard decision operation on the signal after nonlinear filtering compensation (i.e., the final equalized output). The digital control unit obtains the decided signal from the digital signal processor in the hardware execution unit.

[0072] Furthermore, the first error signal can reflect the degree of signal distortion at the current moment. Since delay is common in multimode fiber transmission, the introduction of a time offset of the input signal relative to the current sampling moment compensates for the signal transmission delay and ensures synchronization.

[0073] Furthermore, by introducing a first-step length parameter into equation (6), a balance is achieved between convergence speed and stability. Specifically, during sudden disturbance recovery, the first-step length parameter has a larger value, accelerating convergence. During steady-state jitter suppression, the first-step length parameter has a smaller value, preventing overshoot. Specifically, the value range of the first-step length parameter is 1× Up to 1× .

[0074] Furthermore, the first-step length parameter is used to control the update amplitude of the first tap coefficient. Its value is determined jointly by the statistical characteristics of the input signal and the stability constraints of the algorithm. The calculation steps are as follows: First, during operation, a sufficiently long continuous sampling data segment of the input signal is collected (the number of sampling points must meet statistical sufficiency, preferably greater than 1000, to eliminate the influence of random noise). The collected sampling data is statistically processed to calculate the average power of the input signal. Second, the length of the first tap coefficient (i.e., the filter order) is determined. The value of the first tap coefficient length must cover the main energy part of the channel impulse response (such as the channel memory length), or be preset according to system performance requirements (such as bit error rate and convergence speed). Third, the boundary conditions of the first-step length parameter are determined based on stability theory. According to the convergence theory of adaptive filtering algorithms (such as the LMS algorithm), to ensure that the weight update process converges and does not diverge, the first-step length parameter must satisfy the stability boundary condition: i.e., 0. ;in, The length of the first tap coefficient. Let be the average power of the input signal. Under the premise of satisfying the above stability boundary, and to balance convergence speed and steady-state error, the first step's long parameter is optimized through engineering: ,in, For the first step, long parameters, The length of the first tap coefficient. This represents the average power of the input signal. Specifically, it represents the length of the first tap coefficient. With advance tapping and delayed taps The relationship is: .

[0075] Furthermore, when the multimode optical module detects a sudden change in link state (such as a rapid temperature change, fiber bending, or hot-plugging of a switch) causing a sudden increase in the bit error rate, it needs to converge quickly to a new optimal tap coefficient. In this case, a larger first-step length parameter is used to accelerate convergence. Specifically, the first-step length parameter is set to 0.001. When in a stable operating state, with only slow temperature drift or mode noise, a smaller first-step length parameter should be used to reduce steady-state error and avoid overshoot. Specifically, the first-step length parameter is set to 0.0005.

[0076] Furthermore, in equation (6), the first... The signal has been decided at this moment. With the Linear filter compensation signal at time 1 The difference serves to quantify the degree of signal distortion.

[0077] Furthermore, as a first implementation scheme, the independent error signal of the feedforward filter is used for updating. That is, the first error signal shown in equation (7) is used. The first error signal at time 1 At this time, the first The feedforward filter at the nth sampling time The first tap coefficient Adjustments are made according to equation (6). This scheme decouples the error calculations of feedforward equalization (FFE) and decision feedback equalization (DFE), allowing the feedforward filter to converge independently to prioritize the elimination of inter-symbol interference. This scheme effectively reduces the coupling of the computation loop and the feedback delay, making it suitable for equalizer initialization or applications with extremely high real-time processing requirements. It can also promote rapid initial convergence of the feedforward taps.

[0078] Furthermore, as a second implementation scheme, the final decision error signal of the equalization system is used for updating. That is, the first... The first error signal at time 1 Replace with the first The second error signal at time 1 The specific formula for adjusting the first tap coefficient at this point is: ,in, This scheme jointly optimizes feedforward equalization and decision feedback equalization as a whole. This scheme effectively avoids functional overlap and conflict between the feedforward and feedback parts when dealing with post-inter-symbol interference, enabling them to converge to the global optimal solution. It is suitable for stable tracking stages with high requirements for channel impairment compensation accuracy and jitter suppression.

[0079] like Figure 3 As shown, in some embodiments of methods for suppressing jitter in multimode optical module transmission signals, the method further includes:

[0080] The second tap coefficient calibration step involves dynamically adjusting the second tap coefficient based on the second tap coefficient adjustment algorithm, taking the decided signal, the nonlinear filter compensation signal, the second tap coefficient, and the signal feature data, to obtain the second calibration tap coefficient.

[0081] The second tap coefficient adjustment algorithm is shown in equation (8).

[0082] (8);

[0083] in, For the first The sampling time of the first sampling moment The second calibration tap coefficient. For the first The feedback filter at the sampling time is... The second tap coefficient This is the second step size parameter. For the first The second error signal at time 1 For the first The signal has already been determined at this moment. This is the sequence number of the current sampling time. This is the tap index for the feedback filter.

[0084] In some implementations of methods for suppressing jitter in multimode optical module transmission signals, the... The calculation algorithm is shown in equation (9). (9);

[0085] in, For the first The second error signal at time 1 For the first The signal has already been determined at this moment. The nonlinear filtering compensation signal is at time n. This is the sequence number of the current sampling time.

[0086] In these implementations, by setting a second tap coefficient calibration module, the second tap coefficient can be adjusted according to the second tap coefficient adjustment algorithm, solving the problem that traditional fixed tap coefficients cannot adapt to changes in channel characteristics. (For example, in high-temperature environments, when mode dispersion intensifies, dynamic adjustment can compensate for channel distortion.) Simultaneously, the second tap coefficient calibration module, combined with the calculated second error signal, forms a closed-loop feedback, improving compensation accuracy and enabling dynamic adjustment of the second tap coefficient to meet the requirements of high-speed optical modules.

[0087] Furthermore, the first in equation (8) The second error signal at time 1 It serves to reflect the degree of signal distortion at the current moment. Since delays are common in multimode fiber transmission, the introduction of a time offset between the decided signal and the current sampling time compensates for signal transmission delays (e.g., mode delay differences in multimode fiber can reach 10 ps / km), ensuring compensation and synchronization.

[0088] Furthermore, by introducing a second step size parameter into equation (8), a balance is achieved between convergence speed and stability. In equation (9), the step size is calculated by... The signal has been decided at this moment. With the nonlinear filter compensation signal at time n The difference serves to quantify the degree of signal distortion. Specifically, the value range of the second step size parameter is 1× Up to 1× .

[0089] Furthermore, the calculation steps for the second step length parameter are as follows: First, acquire the decision feedback signal and calculate the average power. Acquire a continuous sampling data segment of the decided signal. Calculate the average power for the acquired data. Second, determine the length of the second tap coefficient. The value of the second tap coefficient length is preset based on system performance requirements (such as bit error rate and convergence speed). Third, calculate the theoretical boundary of the second step length parameter based on stability theory. According to the convergence theory of LMS-type adaptive algorithms (such as the LMS algorithm commonly used in decision feedback equalizers), to ensure the convergence of the weight update process and prevent divergence, the second step length parameter must satisfy the stability constraint condition, i.e., 0. ;in, The length of the second tap coefficient. The average power of the decided signal. Fourth step: Calculate the specific value of the second step size parameter using a scaling factor. Under the premise of satisfying the above stability boundary, introduce a scaling factor and determine the second step size parameter using the following algorithm: ;in, This is the proportionality coefficient. The length of the second tap coefficient. This represents the average power of the decided signal. The proportional gain is used to adjust the specific position of the second step size parameter within the stability boundary, balancing convergence speed and steady-state error. Engineering practice has verified that the range of values ​​for the proportional gain is: The fifth step is to determine the final range of values ​​for the second step length parameter. ;in, The length of the second tap coefficient. The average power of the decided signal, This is the second step size parameter. Preferably, it is the length of the second tap coefficient. With feedback filter length The values ​​are the same.

[0090] Furthermore, for multimode optical modules with speeds of 25Gbps and above, the channel memory effect is significant, and the feedback filter needs to quickly track high-frequency jitter. The value of the second step size parameter is 0.002. For applications with speeds of 10Gbps and below, channel changes are relatively slow, and a smaller step size can be selected to reduce power consumption and computational noise. The value of the second step size parameter is 0.0001.

[0091] like Figure 3 As shown, in some embodiments of methods for suppressing jitter in multimode optical module transmission signals, the method further includes:

[0092] The temperature compensation calibration step determines whether calibration is required based on the second error signal obtained from the second tap coefficient calibration step and a preset error threshold.

[0093] If calibration is required, the threshold current is calibrated based on the calibration step size and the sign of the average value of the second error signal within a preset judgment window, using a temperature compensation calibration algorithm, to obtain the calibration threshold current.

[0094] In these embodiments, by setting up a temperature compensation calibration module, it is used to determine whether temperature compensation calibration is needed based on the second error signal calculated by the second tap coefficient calibration module and the preset error threshold inside the temperature compensation calibration module.

[0095] Furthermore, the temperature compensation calibration judgment algorithm is shown in equation (10).

[0096] (10);

[0097] in, For the first The second error signal at time 1 To determine the number of sampling points in the window, This is the sequence number of the current sampling time. Before the trigger time A judgment window consisting of sampling points. This is the error threshold. The error threshold is preset within the multimode optical module, and different models of multimode optical modules have different error thresholds. Preferably, the recommended range for the error threshold is the [missing value]. The second error signal noise standard deviation at each time point is 0.01 to 0.1 times.

[0098] Furthermore, (11); among which Before the trigger time A judgment window consisting of sampling points. To determine the number of sampling points in the window, The signed average of the second error signal within the judgment window. For the first The second error signal at time 1 This is the sequence number of the current sampling time.

[0099] Furthermore, the temperature compensation calibration algorithm is shown in equation (12).

[0100] (12);

[0101] in, The temperature is The calibration threshold current at that time The temperature is Threshold current at that time To calibrate the step size, The sign is the error symbol. The signed average of the second error signal within the judgment window. At the current sampling time, For standard temperature data, The difference between the hardware temperature data and the standard temperature data.

[0102] Furthermore, in equation (12), by combining the difference between the hardware temperature data and the standard temperature data with the calibration step size, the threshold current drift caused by temperature changes can be compensated. By setting the error sign, nonlinear calibration of temperature compensation is achieved, adapting to the nonlinear relationship between temperature and error.

[0103] Furthermore, as shown in equation (12), the calibration direction can be adjusted according to the sign of the error (e.g., increasing the threshold current for positive errors and decreasing the threshold current for negative errors) to avoid overcompensation or undercompensation. The value of the calibration step size can be manually adjusted according to the usage environment (e.g., increasing the step size at high temperatures to quickly track temperature changes) to improve the response speed.

[0104] Furthermore, a second error signal is first obtained from the second tap coefficient calibration step. This signal reflects the degree of deviation between the decided signal and the nonlinear filter compensation signal, and can quantify the current signal distortion level. Then, this second error signal is compared with the error threshold preset in the multimode optical module. When the average absolute value of the second error signal within the judgment window exceeds the error threshold (i.e., the calibration condition is met), temperature compensation calibration is initiated. When calibration is required, the threshold current is adjusted according to the preset calibration step size and error sign. That is, the error sign determines the calibration direction (increase the threshold current for positive errors and decrease the threshold current for negative errors), and the calibration step size controls the magnitude of each adjustment, thereby avoiding overcompensation or undercompensation. Finally, the calibration threshold current is obtained and updated in the temperature compensation module, realizing dynamic correction of temperature compensation, ensuring that the optical module maintains stable output optical power across the entire temperature range, and thus ensuring continuous suppression of signal jitter.

[0105] like Figure 3 and Figure 4 As shown, in some implementations of methods for suppressing jitter in multimode optical module transmission signals, the method further includes:

[0106] The response timing adjustment step involves comparing the switch timing data with the pre-stored timing data in the response timing adjustment module. If there is timing data in the pre-stored timing data that is completely identical to the switch timing data, then the timing data is the timing adjustment data.

[0107] If there is no timing data in the pre-stored timing data that is completely identical to the timing data of the switch, the timing adjustment data includes a general frequency, a general return duration, and a general output duration. The SCL clock frequency of the multimode optical module is adjusted to the general frequency, and the ACK signal return time delay of the multimode optical module is adjusted to the general return duration.

[0108] In these implementations, the timing data of the switch is compared with the pre-stored timing data in the signal characteristic acquisition module. If the pre-stored timing data is the same as the switch timing data, the pre-stored timing data is set as the timing adjustment data for communication between the multimode optical module and the switch. This avoids the additional jitter introduced by dynamic adjustment, which could affect the jitter suppression effect of the temperature compensation and signal equalization compensation steps. Specifically, the pre-stored timing data can be stored in the EEPROM within the multimode optical module.

[0109] Furthermore, due to the differences in timing data between the switch and the multimode optical module, data transmission in the bus frequently fails or generates communication errors due to timing mismatch. Under extreme conditions of long-distance transmission of multimode fiber, this indirectly affects system-level jitter, causing delays or interruptions in ongoing temperature compensation and signal equalization compensation steps. By responding to timing adjustment steps, the timing data of the multimode optical module can be adjusted, thereby indirectly suppressing jitter.

[0110] Furthermore, SCL clock adjustment, ACK signal return time adjustment, and SCL clock falling edge insertion delay are not directly applied to the optical signal itself, but rather to the I²C management interface between the multimode optical module and the switch. Through the management interface, the debugging parameters are configured on the signal transmission link to improve signal quality and indirectly suppress signal jitter.

[0111] Furthermore, if no timing data matching the switch timing data is found in the pre-stored timing data, the SCL clock frequency of the multimode optical module is adjusted to the universal frequency, and the ACK signal return time delay is adjusted to the universal return duration, ensuring signal integrity while effectively suppressing link noise.

[0112] Furthermore, the general frequency is 100kHz, the general return duration is after the 8th falling edge of SCL (either pulled low or transmitted), and the general output duration is 50ns. Furthermore, the pre-stored timing data in the multimode optical module includes the SCL clock frequency, the ACK signal return time, and the falling edge of the SCL clock.

[0113] In some implementations of methods for suppressing jitter in multimode optical module transmission signals, the hardware execution unit includes a receiving link consisting of a photodetector, a transimpedance amplifier, an analog-to-digital converter, and a digital signal processor, and a transmitting link consisting of a digital signal processor, a digital-to-analog converter, a driver, and a laser. The hardware execution unit is electrically connected to a digital control unit.

[0114] Furthermore, the laser within the multimode optical module of this invention can be an 850nm VCSEL laser such as Lumentum V1183-001 or II-VI41A08; the photodetector can be a Broadcom AFBR-2698A or Hamamatsu S10784; and the DSP can be a Macom MATA-03840 or Semtech GN1086. The analog-to-digital converter can be a Broadcom BCM87400 series analog-to-digital converter.

[0115] Furthermore, the pre-stored timing data in the multimode optical module includes timing data from various common switch models, including: Huawei CE6865-48S6CQ, with an SCL clock frequency of 400kHz, an ACK signal return time of pulling low (or sending) after the 8th falling edge of SCL, and an SDA setup time ≥250ns; Cisco Nexus 93180YC-EX, with an SCL clock frequency of 400kHz, an ACK signal return time of pulling low (or sending) after the 8th falling edge of SCL, and an SDA setup time ≥250ns; and Juniper QFX5120-48T, with an SCL clock frequency of 400kHz, an ACK signal return time of pulling low (or sending) after the 8th falling edge of SCL, and an SDA setup time ≥250ns.

[0116] Table 1. Optimization Results Data Table

[0117] As shown in Table 1, the test conditions were as follows: Multimode fiber specifications and length: OM4 multimode fiber (core diameter 50μm, effective mode bandwidth ≥4700 MHz·km) was used, and the transmission distance was set to 550m (before optimization) and 800m (after optimization). OM4 fiber has a higher mode bandwidth and can support longer transmission distances at high speeds. It is the standard test fiber for 25G / 100G multimode optical module performance verification. Signal rate: The test signal rate was 25.78 Gb / s (corresponding to 25G Ethernet applications), and the PRBS31 (pseudo-random binary sequence, order 31) test code pattern was used. PRBS31 is the standard code pattern for high-speed optical module bit error rate testing, which can fully simulate the randomness of real service data and effectively excite mode dispersion and jitter effects in the channel. Ambient temperature control method: The test was conducted in a constant temperature environment (25℃±2℃), and the optical module was kept at a stable temperature by a thermal chuck or a constant temperature chamber. Performance data for full-temperature range validation (-10℃ to 80℃) was obtained based on temperature cycling tests, using the Telcordia GR-468-CORE reliability testing standard. BER measurement cumulative bits / measurement duration: Bit Error Rate (BER) testing employed a Bit Error Analyzer (BERT), with a cumulative measurement duration of no less than 24 hours, or a cumulative number of bits of measurement of no less than [amount missing]. bit. For For BER tests of this magnitude, a sufficiently long testing time needs to be accumulated to ensure statistical confidence (e.g., at a rate of 25.78 Gb / s). (Approximately 10.8 hours per bit) This test uses 24 hours to ensure the reliability of the results. Sample size: At least 10 optical module samples are selected for each test condition. The test results are the average of the measurements of each sample to ensure statistical significance and eliminate the influence of individual differences on performance evaluation. The sample size and test method comply with the sample size requirements for device performance verification in commercial optical module reliability standards (such as GR-468-CORE).

[0118] As shown in Table 1, the method for suppressing jitter in multimode optical module transmission signals according to the present invention can increase the transmission distance of the optical module from 550 meters to 800 meters while meeting performance requirements, reduce the power consumption of the optical module from 0.8W to 0.75W, and reduce the bit error rate from... Reduce to This expands the operating temperature range of the multimode optical module from 0~70℃ to -10~80℃.

[0119] like Figure 5 As shown, the present invention also provides a jitter suppression system for multimode optical module transmission signals. The jitter suppression system for multimode optical module transmission signals includes at least one processor, a memory, an input device, and a display device. The input device is used to obtain input from the outside. The memory stores instructions. When the instructions are executed by at least one processor, the steps of the method described in the method embodiment are implemented, and the running results are displayed on the display device, thus implementing the steps of the method described in the method embodiment.

[0120] The embodiments and functional operations of the subject matter described in this specification can be implemented in the following ways: digital electronic circuits, tangibly implemented computer software or firmware, computer hardware, including the structures disclosed in this specification and their equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, that is, one or more modules of computer program instructions encoded on one or more tangible non-transitory program carriers, for execution by a data processing device or to control the operation of the data processing device.

[0121] Alternatively or additionally, program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are then generated as coded information to be transmitted to an appropriate receiver device executed by data processing equipment. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or one or more combinations of the above.

[0122] The processing and logic flows described in this specification can be executed by one or more programmable computers, which execute one or more computer programs by processing input data and generating output to run functions. The processing and logic flows can also be executed by special-purpose logic circuitry, such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits), and the device can also be implemented as special-purpose logic circuitry.

[0123] To transmit interactions with a user, embodiments of the subject matter described in this specification can be implemented on a computer having: a display device, such as a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user; and a keyboard and a positioning device, such as a mouse or trackball, which the user can use to send input to the computer. Other types of devices can also be used to transmit interactions with the user; for example, feedback provided to the user can be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including sound input, voice input, or tactile input. Additionally, the computer can interact with the user by sending documents to and receiving documents from a device used by the user; for example, by sending a webpage to a web browser on the user's client device in response to a received request from a web browser.

[0124] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather as descriptions of features that can embody specific embodiments of a particular invention. Specific features described in this specification within the context of an independent embodiment may also be implemented in combination with a single embodiment. Conversely, various features described within the context of a single embodiment may also be implemented independently in multiple embodiments, or in any suitable sub-combination. Furthermore, while features may be described for combination and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and the claimed combination may be redirected to a sub-combination or a variation thereof.

[0125] Similarly, although operations are described in the accompanying drawings in a specific order, it should not be construed as requiring that such operations be performed in the specific order shown or in sequential order, or that all illustrated operations be performed, in order to achieve the desired result. In certain cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that program components and systems can generally be integrated into a single software product or packaged into multiple software products.

[0126] Specific implementations of the subject matter have been described. Other implementations are within the scope of the following claims. For example, the activities described in the claims can be performed in a different order and still achieve the desired result. As an example, the processes described in the drawings do not necessarily require a specific order or sequence to be shown in order to achieve the desired result. In certain implementations, multitasking and parallel processing may be advantageous.

Claims

1. A method for suppressing transmission signal jitter of a multimode optical module, the multimode optical module comprising a hardware execution unit and a digital control unit, communication between a switch and the multimode optical module being performed through electrical signals, and communication between the multimode optical module and other multimode optical modules being performed through transmission and reception of optical signals, the method comprising: The method comprises the following steps performed by the digital control unit: ​ an acquisition step of acquiring hardware temperature data and signal characteristic data of the received optical signal from the hardware execution unit; a temperature compensation step of obtaining driving current compensation data according to the hardware temperature data and pre-stored standard temperature data, and sending the driving current compensation data to the hardware execution unit to perform jitter suppression on the transmitted optical signal; a signal equalization compensation step of determining initial first tap coefficients and initial second tap coefficients according to the standard temperature data, taking the initial first tap coefficients and the initial second tap coefficients as an iteration starting point, and obtaining equalization compensation data based on a linear filter compensation algorithm and a nonlinear filter compensation algorithm by using the first tap coefficients and the second tap coefficients read from an electrically erasable programmable read-only memory of the multimode optical module and the signal characteristic data, and sending the equalization compensation data to the hardware execution unit to perform jitter suppression on the received optical signal.

2. The method of claim 1, wherein the step of suppressing the transmission signal jitter of the multimode optical module further comprises the step of: The driving current compensation data is obtained according to the hardware temperature data and pre-stored standard temperature data, and comprises: ​ obtaining a temperature variation according to the hardware temperature data and the standard temperature data; obtaining driving current compensation data according to the temperature variation, pre-stored target output power and a driving current compensation algorithm; the driving current compensation algorithm is shown in formula (1), （1）； The The calculation algorithm of the above formula (2) is shown as follows. （2）； The The computational algorithm of the above equation is given by equation (3), （3）； Wherein, the is driving current compensation data when the temperature is is threshold current when the temperature is is standard temperature data, is the difference between the hardware temperature data and the standard temperature data, is differential slope efficiency when the temperature is is target output power, is standard temperature threshold current, is a first temperature coefficient, is a second temperature coefficient, is standard temperature slope efficiency, is a third temperature coefficient, is a fourth temperature coefficient.​​​ 3. The method according to claim 1, characterized in that, the equalization compensation data comprises linear filter compensation signals and nonlinear filter compensation signals; the initial first tap coefficients, the first tap coefficients and the signal characteristic data are used to obtain the linear filter compensation signals based on the linear filter compensation algorithm; the initial second tap coefficients, the second tap coefficients, the linear filter compensation signals and the feedback filter output signals are used to obtain the nonlinear filter compensation signals based on the nonlinear filter compensation algorithm.

4. The method according to claim 3, characterized in that, the linear filter compensation algorithm is shown in formula (4), （4）； the nonlinear filter compensation algorithm is shown in formula (5), （5）； in, For the first Linear filtering compensation signal at each time point For the first The feedforward filter at the sampling time is... The first tap coefficient. For the first The input signal at each moment, The nonlinear filtering compensation signal is at time n. For the first The feedback filter at the sampling time is... The second tap coefficient For the first The signal has already been determined at this moment. This is the sequence number of the current sampling time. This is the tap index for the feedforward filter. For the feedback filter tap index, This refers to the advance payment amount. To delay the number of taps, For the feedback filter length, The output signal of the feedback filter; the initial first tap coefficients are values of the first tap coefficients at an iteration update starting time, and the initial second tap coefficients are values of the second tap coefficients at the iteration update starting time.

5. The method for suppressing jitter in multimode optical module transmission signals according to claim 4, characterized in that, The method further comprises: a first tap coefficient calibration step of obtaining a decided signal, and dynamically adjusting the first tap coefficients based on a first tap coefficient adjustment algorithm by using the decided signal, the linear filter compensation signals, the first tap coefficients and the signal characteristic data to obtain first calibrated tap coefficients; the first tap coefficient adjustment algorithm is shown in formula (6), （6）； The The computational algorithm is given by equation (7), (7); wherein is a first calibration tap coefficient for the first adaptive filter at the first sampling instant, is a first calibration tap coefficient for the first adaptive filter at the first sampling instant, is a first tap coefficient of the feedforward filter at the first sampling instant, is a first step size parameter, is a first error signal at the first sampling instant, is an input signal at the first sampling instant, is a sequence number of the current sampling instant, is a feedforward filter tap index, is a decided signal at the first sampling instant, is a linear filter compensation signal at the first sampling instant.

6. The method of claim 5, wherein the step of applying a signal to the first and second modulators is performed by applying a signal to the first and second modulators such that the first and second modulators are driven to a first state and a second state, respectively, and wherein the first and second states are different. The method further comprises: a second tap coefficient calibration step of dynamically adjusting the second tap coefficients based on a second tap coefficient adjustment algorithm by using the decided signal, the nonlinear filter compensation signals, the second tap coefficients and the signal characteristic data to obtain second calibrated tap coefficients; the second tap coefficient adjustment algorithm is shown in formula (8), （8）； wherein is a first calibration tap coefficient at a first sampling instant, is a second calibration tap coefficient at a first sampling instant, is a second calibration tap coefficient at a first sampling instant, is a first calibration tap coefficient at a first sampling instant, is a second tap coefficient of the feedback filter at a first sampling instant, is a second tap coefficient of the feedback filter at a first sampling instant, is a second step parameter, is a second error signal at a first instant, is a second error signal at a first instant, is a decided signal at a first instant, is a decided signal at a first instant, is a sequence number of a current sampling instant, is a feedback filter tap index.

7. The method of claim 6, wherein the method further comprises: a temperature compensation calibration step of determining whether calibration is needed according to the second error signal obtained from the second tap coefficient calibration step and a preset error threshold; and if calibration is needed, calibrating the threshold current based on a temperature compensation calibration algorithm according to a calibration step size and a sign of an average value of the second error signal within a preset determination window, to obtain a calibrated threshold current. The The computational algorithm is given by equation (9), (9); in, For the first The second error signal at time 1 For the first The signal has already been determined at this moment. The nonlinear filtering compensation signal is at time n. This is the sequence number of the current sampling time.

8. The method of claim 7, wherein the step of applying a signal to the first and second modulators is performed by applying a signal to the first and second modulators such that the first and second modulators are driven to a first state and a second state, respectively, and wherein the first and second states are different.

9. The method of claim 1, wherein the hardware execution unit comprises a receiving link composed of a photodetector, a transimpedance amplifier, an analog-to-digital converter, and a digital signal processor, and a transmitting link composed of a digital signal processor, a digital-to-analog converter, a driver, and a laser, and the hardware execution unit is electrically connected to the digital control unit. The system comprises at least one processor; and a memory storing instructions which, when executed by the at least one processor, implement the steps of the method according to any one of claims 1 to 9. ​ ​ ​ 10. A system for suppressing the jitter of a multimode optical module transmission signal, characterized by ​

Images