An optical receiving end bandwidth test method, system and related device

Abstract

The application discloses an optical receiving end bandwidth test method, system and related equipment, which first performs coarse scanning by a first sweep frequency step and divides the amplitude-frequency response area, then determines a smaller second sweep frequency step according to the transition core area attenuation slope, adopts differential sampling in different areas, effectively reduces frequency point redundancy, shortens test time consumption, accurately captures the amplitude-frequency rapid attenuation characteristics, balances test efficiency and accuracy, and adapts to production line batch test requirements. By combining real-time signal-to-noise ratio adjustment sampling time and Fourier transform average number, the influence of various interference factors can be inhibited; by peak detection, roll-off characteristic determination, matching of a suitable low-pass model for global fitting, random sampling errors can be eliminated, the standard amplitude-frequency response curve can be fitted to the actual characteristics, and the stability and repeatability of bandwidth determination are improved. The method realizes accurate, efficient and stable test of the optical receiving end bandwidth through the organic combination of adaptive non-uniform sweep frequency and model fitting, and improves the test performance.

Description

Technical Field

[0001] This application relates to the field of bandwidth testing technology, and more specifically, to a method, system and related equipment for testing the bandwidth of an optical receiver. Background Technology

[0002] With the rapid development of optical communication technology, optical modules, as the core component of optical communication systems, directly determine the signal transmission rate and quality through their receiver bandwidth. This is one of the key indicators for measuring the performance of optical modules. Therefore, it is extremely necessary to achieve accurate and efficient testing of optical receiver bandwidth, which can not only ensure the quality of optical modules leaving the factory, but also provide support for the stable operation of optical communication systems.

[0003] In existing optical receiver bandwidth testing, the traditional frequency sweep method is the most widely used approach. This method often employs a fixed-step uniform frequency sweep mode, which has several inherent technical drawbacks: First, testing accuracy and efficiency are contradictory. Small-step sweeps offer high accuracy but suffer from frequency redundancy and long testing times, making them unsuitable for mass production testing. Large-step sweeps improve efficiency but lead to sparse sampling near the -3dB cutoff frequency, resulting in significant bandwidth positioning errors. Second, uniform sampling across the entire frequency band violates the amplitude-frequency response characteristics of optical receivers. Low-frequency flat regions do not require dense sampling but consume significant testing resources, while insufficient sampling in high-frequency transition regions makes it difficult to capture the rapid amplitude-frequency attenuation characteristics. Third, traditional bandwidth determination relies solely on a single -3dB frequency point, making it susceptible to interference from circuit noise, optical power fluctuations, and signal jitter, resulting in poor stability and repeatability of test results. Furthermore, existing testing methods do not fully consider the low-pass system characteristics of optical receivers and lack effective error elimination methods, making it difficult to eliminate random errors during the sampling process, further reducing the accuracy of bandwidth testing and failing to meet the testing requirements of practical applications.

[0004] Therefore, there is an urgent need for a new method for testing the bandwidth of optical receivers to overcome the shortcomings of existing technologies, achieve accurate, efficient, and stable testing of optical receiver bandwidth, and meet the actual needs of optical communication technology development and production line batch testing. Summary of the Invention

[0005] This application provides a method, system, and related equipment for testing the bandwidth of optical receivers, enabling accurate, efficient, and stable testing of optical receiver bandwidth, and adapting to the actual needs of optical communication technology development and production line batch testing.

[0006] A method for testing the bandwidth of an optical receiver includes:

[0007] The microwave signal source is controlled to perform a full-band sparse coarse scan of the optical receiver under test with the first sweep frequency step. An initial amplitude-frequency response curve is generated based on the amplitude-frequency response data corresponding to each coarse scan frequency point. The initial amplitude-frequency response curve is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0008] Calculate the local amplitude-frequency attenuation slope within the transition core region, and determine the second sweep step based on the local amplitude-frequency attenuation slope, wherein the second sweep step is smaller than the first sweep step and is negatively correlated with the local amplitude-frequency attenuation slope;

[0009] The microwave signal source is controlled to perform sampling using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region;

[0010] The real-time signal-to-noise ratio of each frequency point is obtained, and the sampling duration and the number of Fourier transforms are adjusted according to the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset.

[0011] Peaking detection and roll-off characteristic determination are performed on the fine scan amplitude-frequency dataset. Based on the detection results, a single-pole low-pass model or a multi-pole low-pass model is matched and selected from the preset model library for global curve fitting to obtain the standard amplitude-frequency response curve.

[0012] Based on the standard amplitude-frequency response curve, the frequency value corresponding to the attenuation to -3dB is calculated and used as the bandwidth test result of the optical receiver under test.

[0013] Optionally, the attenuation distribution of the initial amplitude-frequency response curve is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region, including:

[0014] The frequency band with attenuation less than the first preset threshold is divided into the low-frequency flat region;

[0015] The frequency band with attenuation greater than or equal to the first preset threshold and less than or equal to the second preset threshold is divided into the transition core region;

[0016] The frequency band with attenuation greater than the second preset threshold is divided into the high-frequency roll-off region;

[0017] Wherein, the first preset threshold corresponds to the starting point when the amplitude-frequency response begins to enter a significant decay, and the second preset threshold is greater than the first preset threshold and corresponds to the critical point when the amplitude-frequency response enters the rapid roll-off stage.

[0018] Optionally, the process of calculating the local amplitude-frequency attenuation slope within the transition core region includes:

[0019] Within the transition core region, differential operations are performed on adjacent coarse sweep frequency points to obtain the amplitude-frequency attenuation slope of each sub-interval;

[0020] The negative correlation between the second sweep step and the local amplitude-frequency attenuation slope is expressed as follows:

[0021] ;

[0022] in, For the second sweep frequency step, The slope of the local amplitude-frequency attenuation. This is a preset proportional coefficient. To prevent division by zero, a regularization constant is used.

[0023] Optionally, the step of obtaining the real-time signal-to-noise ratio of each frequency point and adjusting the sampling duration and Fourier transform averaging times of each frequency point based on the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset includes:

[0024] Obtain a preset first mapping function between signal-to-noise ratio and sampling duration, and a second mapping function between signal-to-noise ratio and the average number of Fourier transforms, wherein both the first mapping function and the second mapping function are monotonically increasing nonlinear functions;

[0025] Obtain the real-time signal-to-noise ratio of the current frequency point, and substitute the real-time signal-to-noise ratio into the first mapping function and the second mapping function respectively to calculate the target sampling duration and target average number of times corresponding to the current frequency point;

[0026] The time-domain signal of the current frequency point is collected according to the target sampling duration, and the collected time-domain signal is subjected to Fourier transform averaging processing according to the target averaging number to obtain the amplitude-frequency response data of the current frequency point. After traversing all fine-scan frequency points, the fine-scan amplitude-frequency dataset is generated.

[0027] Specifically, for frequency points where the real-time signal-to-noise ratio is lower than the preset low-noise threshold, the calculated target sampling duration is greater than the preset baseline sampling duration, and the target average number of times is greater than the preset baseline average number of times. For frequency points where the real-time signal-to-noise ratio is higher than the preset high-noise threshold, the calculated target sampling duration is less than the preset baseline sampling duration, and the target average number of times is less than the preset baseline average number of times.

[0028] Optionally, the process of performing peaking detection on the fine-scan amplitude-frequency dataset includes:

[0029] Perform second-order difference operations on the fine-scan amplitude-frequency dataset to identify local maxima.

[0030] Determine the amplitude-frequency peaking amount of the local maximum point relative to the average amplitude-frequency response in the low-frequency flat region;

[0031] If the amplitude-frequency peaking amount exceeds the preset peaking threshold, then the amplitude-frequency peaking feature is determined to exist.

[0032] Optionally, the process of determining the roll-off characteristics of the fine-scan amplitude-frequency dataset includes:

[0033] Within the high-frequency roll-off region, the high-frequency roll-off slope is obtained by taking logarithmic coordinates of the amplitude-frequency response data and performing linear regression.

[0034] If the difference between the absolute value of the high-frequency roll-off slope and the quotient of 20dB divided by ten octaves is less than a preset difference threshold, it is determined to be a single-pole roll-off characteristic.

[0035] If the absolute value of the high-frequency roll-off slope is greater than or equal to 40dB divided by a decade, it is determined to be a multi-pole roll-off characteristic.

[0036] Optionally, the single-pole low-pass model is expressed as:

[0037] ;

[0038] The multi-pole low-pass model includes at least a second-order two-pole model, expressed as:

[0039] ;

[0040] in, , These are the expressions for the single-pole low-pass model and the second-order double-pole model, respectively. This represents the system's low-frequency and zero-frequency gain. This is the frequency for the sweep test; The -3dB cutoff frequency to be measured; The system resonant frequency is determined by the multipole distribution; The damping coefficient characterizes the degree of peaking in the amplitude-frequency curve. The time curve shows peaking. It is the critical damping.

[0041] Optionally, after obtaining the standard amplitude-frequency response curve, the method further includes:

[0042] Calculate the root mean square error between the standard amplitude-frequency response curve and the fine-scan amplitude-frequency dataset;

[0043] If the root mean square error is greater than the preset error threshold, and the current model is a single-pole low-pass model, then switch to a multi-pole low-pass model for refitting.

[0044] If the root mean square error is greater than the preset error threshold, and the current model is a multi-pole low-pass model, then the model order is increased and the model is refitted.

[0045] If the root mean square error is still greater than the preset error threshold after increasing the model order, an abnormal alarm is triggered, indicating that there is nonlinear distortion or link failure at the optical receiver under test.

[0046] An optical receiver bandwidth testing system, comprising:

[0047] The optical transmission link consists of a tunable laser, an intensity modulator, a microwave signal source, and an optical tunable attenuator connected in sequence. The microwave signal source is used to output a frequency-tunable sinusoidal modulation signal to drive the intensity modulator, and the optical tunable attenuator is used to adjust the output optical power.

[0048] The beam splitter has its input end connected to the output end of the optical adjustable attenuator, its first output end connected to the receiver of the optical module under test, and its second output end connected to the optical power meter.

[0049] An optical power meter is used to measure the amplitude-frequency response of the optical transmission link during no-load calibration, generate calibration coefficients, and continuously monitor the optical power at the second output of the splitter during bandwidth testing.

[0050] The synchronous acquisition unit includes a clock source and an ADC sampling circuit, wherein the clock source provides clock signals to both the microwave signal source and the ADC sampling circuit.

[0051] The host computer processing unit is connected to the microwave signal source, the optical adjustable attenuator, the optical power meter, and the ADC sampling circuit, respectively, and is used to execute each step of the optical receiver bandwidth testing method as described above.

[0052] Optionally, the host computer processing unit includes:

[0053] The first frequency sweep module is used to control the microwave signal source to perform a full-band sparse coarse sweep of the optical receiver under test with the first frequency sweep step, generate an initial amplitude-frequency response curve based on the amplitude-frequency response data corresponding to each coarse sweep frequency point, and divide the initial amplitude-frequency response curve into a low-frequency flat region, a transition core region and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0054] The step adjustment module is used to calculate the local amplitude-frequency attenuation slope in the transition core region and determine the second sweep frequency step based on the local amplitude-frequency attenuation slope, wherein the second sweep frequency step is smaller than the first sweep frequency step and is negatively correlated with the local amplitude-frequency attenuation slope.

[0055] The second frequency sweep module is used to control the microwave signal source to sample using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region;

[0056] The signal-to-noise ratio (SNR) sampling module is used to obtain the real-time SNR of each frequency point and adjust the sampling duration and Fourier transform averaging times of each frequency point according to the real-time SNR to generate a fine-scan amplitude-frequency dataset.

[0057] The model fitting module is used to perform peaking detection and roll-off characteristic determination on the fine scan amplitude-frequency dataset. Based on the detection results, it selects a single-pole low-pass model or a multi-pole low-pass model from the preset model library to perform global curve fitting and obtain the standard amplitude-frequency response curve.

[0058] The bandwidth calculation module is used to calculate the frequency value corresponding to attenuation to -3dB based on the standard amplitude-frequency response curve, and use it as the bandwidth test result of the optical receiver under test.

[0059] Optionally, the host computer processing unit further includes a pre-calibration module for performing no-load link frequency response calibration and optical power linear region locking calibration;

[0060] The process of calibrating the frequency response of the unloaded link includes:

[0061] Disconnect the first output terminal of the beam splitter from the receiver terminal of the optical module under test, and connect only the optical power meter;

[0062] The microwave signal source is controlled to output sinusoidal modulation signals at each frequency point across the entire frequency band in sequence, and the output optical power at each frequency point is recorded by an optical power meter.

[0063] Using the output optical power at the lowest frequency point as a benchmark, the power attenuation at each frequency point is calculated, and a frequency response calibration coefficient table is generated to correct the amplitude-frequency response data at each frequency point.

[0064] The optical power linear region locking calibration process includes:

[0065] Connect the receiver of the optical module under test to the first output of the beam splitter, and adjust the optical adjustable attenuator to gradually increase the optical power input to the receiver of the optical module under test from low to high.

[0066] The output amplitude of the receiver of the optical module under test is acquired synchronously to generate an input optical power-output amplitude curve;

[0067] Identify the linear operating region in the input optical power-output amplitude curve, and adjust the optical adjustable attenuator to the attenuation amount corresponding to the midpoint of the linear operating region, so that the input optical power remains constant within the linear operating region of the photodiode during the full-band frequency sweep.

[0068] An optical receiver bandwidth testing device, comprising a memory and a processor;

[0069] The memory is used to store programs;

[0070] The processor is configured to execute the program to implement the various steps of the optical receiver bandwidth testing method as described in any of the preceding claims.

[0071] A readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the optical receiver bandwidth testing method as described in any of the preceding claims.

[0072] A computer program product includes a computer program that, when executed by a processor, performs the steps of the optical receiver bandwidth testing method as described in any of the preceding claims.

[0073] As can be seen from the above technical solutions, the optical receiver bandwidth testing method, system, and related equipment provided in this application first control a microwave signal source to perform a full-band sparse coarse sweep with a first sweep step, generating an initial amplitude-frequency response curve and dividing it into a low-frequency flat region, a transition core region, and a high-frequency roll-off region; then, based on the local amplitude-frequency attenuation slope of the transition core region, a smaller second sweep step is determined, and different step sampling is used in different regions; subsequently, the sampling duration and the average number of Fourier transforms are adjusted in combination with the real-time signal-to-noise ratio to generate a fine-sweep amplitude-frequency dataset; finally, through peaking detection and roll-off characteristic determination, a suitable low-pass model is matched to perform full-domain curve fitting, and the -3dB frequency is solved as the test result.

[0074] This application uses a coarse scan to divide the optical receiver's amplitude-frequency response into different regions, and adaptively adjusts the sweep step based on the local amplitude-frequency attenuation slope in the transition core region. A larger first sweep step is used in the low-frequency flat region and the high-frequency roll-off region to reduce frequency redundancy, effectively shorten test time, and improve test efficiency, adapting to the actual needs of batch testing on production lines. A smaller second sweep step is used in the transition core region for encrypted sampling, accurately capturing the rapid amplitude-frequency attenuation characteristics, ensuring test accuracy, and achieving a dual improvement in test efficiency and accuracy. By dividing the region based on the attenuation distribution of the initial amplitude-frequency response curve and using differentiated sampling, test resources can be rationally allocated, avoiding resource waste in the low-frequency flat region while ensuring sampling density in the transition core region, significantly improving sampling rationality and resource utilization. By acquiring the real-time signal-to-noise ratio at each frequency sweep point and dynamically adjusting the sampling duration and the Fourier transform averaging, the influence of interference factors such as circuit noise, optical power fluctuations, and signal jitter can be effectively suppressed. Combined with the characteristics of the optical receiver's low-pass system, by using peaking detection and roll-off characteristic determination to match a suitable single-pole or multi-pole low-pass model for global curve fitting, random errors in the sampling process can be effectively eliminated. This makes the generated standard amplitude-frequency response curve more closely match the actual amplitude-frequency characteristics of the optical receiver, thereby improving the stability and repeatability of the -3dB bandwidth determination and ensuring the reliability of the test results. In summary, this application, through the combination of adaptive non-uniform frequency sweeping and model fitting, achieves accurate, efficient, and stable testing of optical receiver bandwidth, significantly improving test performance. Attached Figure Description

[0075] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0076] Figure 1 This is a schematic diagram of an optical receiver bandwidth testing system disclosed in an embodiment of this application;

[0077] Figure 2 This is a flowchart of an optical receiver bandwidth testing method disclosed in an embodiment of this application;

[0078] Figure 3 This is a hardware structure block diagram of an optical receiver bandwidth testing device disclosed in an embodiment of this application. Detailed Implementation

[0079] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0080] The following section introduces the solution proposed in this application. The technical solution is as follows, and details are provided below.

[0081] This application applies to, for example Figure 1 The optical receiver bandwidth testing system shown includes:

[0082] The optical transmission link consists of a tunable laser, an intensity modulator, a microwave signal source, and an optical tunable attenuator connected in sequence. The microwave signal source is used to output a frequency-tunable sinusoidal modulation signal to drive the intensity modulator, and the optical tunable attenuator is used to adjust the output optical power.

[0083] The beam splitter has its input end connected to the output end of the optical adjustable attenuator, its first output end connected to the receiver of the optical module under test, and its second output end connected to the optical power meter.

[0084] An optical power meter is used to measure the amplitude-frequency response of the optical transmission link during no-load calibration, generate calibration coefficients, and continuously monitor the optical power at the second output of the splitter during bandwidth testing.

[0085] The synchronous acquisition unit includes a clock source and an ADC sampling circuit, wherein the clock source provides clock signals to both the microwave signal source and the ADC sampling circuit.

[0086] The host computer processing unit is connected to the microwave signal source, the optical adjustable attenuator, the optical power meter, and the ADC sampling circuit, respectively, and is used to execute each step of the optical receiver bandwidth testing method.

[0087] Specifically, the optical transmission link, as the core of test signal generation, consists of a tunable laser, an intensity modulator, an RF signal source (i.e., a microwave signal source), and an optically adjustable attenuator connected sequentially. The tunable laser outputs a stable fundamental optical carrier, providing the signal carrier for the modulation process. The RF signal source, synchronized with a common clock source, outputs a continuously adjustable sinusoidal modulated signal. This signal directly drives the intensity modulator, loading the frequency variation of the electrical signal onto the optical carrier to generate an intensity-modulated optical signal that meets the test requirements. The optically adjustable attenuator is located at the output of the intensity modulator, flexibly adjusting the power of the output optical signal to match the operating range of the receiver of the optical module under test, avoiding test distortion due to abnormal optical power and ensuring the stability of the test reference.

[0088] The beam splitter employs a 50:50 splitting ratio. Its input is connected to the output of the adjustable optical attenuator in the optical transmission link, achieving a one-to-two split of the test optical signal. The first output of the beam splitter is connected to the adjustable optical attenuator (secondary adjustment unit), which is then connected to the receiver of the optical module under test (DUT), providing the DUT with a test input signal. The second output is directly connected to an optical power meter, transmitting a portion of the optical signal to the power meter for optical power calibration and real-time monitoring, ensuring the consistency of optical signal power during testing.

[0089] The optical power meter plays a dual core role in the entire testing process: During the no-load calibration phase before the formal start of the test, the optical power meter is used to measure the amplitude-frequency response characteristics of the optical transmission link itself. Based on the measurement data, a calibration coefficient is generated, which is used to compensate for subsequent test data, eliminate the interference of the optical transmission link hardware characteristics on the test results, and ensure the accuracy of the test benchmark; During the bandwidth test, the optical power meter continuously monitors the optical power at the second output end of the splitter and feeds back the optical power changes to the host computer processing unit in real time. The host computer can dynamically adjust the optical adjustable attenuator according to the monitoring data to ensure stable optical power throughout the test and avoid the impact of optical power fluctuations on test accuracy.

[0090] The synchronous acquisition unit includes a clock and an ADC sampling circuit. The clock provides a unified clock signal for the RF signal source and the ADC sampling circuit, ensuring that their working timing is completely synchronized, avoiding sampling deviations caused by clock asynchrony, and ensuring the accuracy of the acquired data. The ADC sampling circuit acquires the electrical signal output by the receiver of the optical module under test, converts the analog signal into a digital signal, and provides a resolvable digital signal for subsequent data processing.

[0091] The host computer processing unit, as the core of the system's control and processing, establishes communication connections with the RF signal source, the optical adjustable attenuator, the optical power meter, and the ADC sampling circuit through circuits. On one hand, the host computer processing unit sends control commands to each hardware module to coordinate and regulate the test process, including controlling the sweep parameters of the microwave signal source, adjusting the attenuation of the optical adjustable attenuator, and triggering ADC sampling, thereby automating the test process. On the other hand, the host computer receives the amplitude-frequency response data collected by the ADC and the monitoring data from the optical power meter, executes the entire process of the optical receiver bandwidth test method, including coarse scan region division, adaptive step adjustment, signal-to-noise ratio optimized sampling, low-pass model fitting, and -3dB bandwidth calculation, and finally outputs the bandwidth test results of the optical receiver under test, completing the entire test process.

[0092] The host computer processing unit includes:

[0093] The first frequency sweep module is used to control the microwave signal source to perform a full-band sparse coarse sweep of the optical receiver under test with the first frequency sweep step, generate an initial amplitude-frequency response curve based on the amplitude-frequency response data corresponding to each coarse sweep frequency point, and divide the initial amplitude-frequency response curve into a low-frequency flat region, a transition core region and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0094] The step adjustment module is used to calculate the local amplitude-frequency attenuation slope in the transition core region and determine the second sweep frequency step based on the local amplitude-frequency attenuation slope, wherein the second sweep frequency step is smaller than the first sweep frequency step and is negatively correlated with the local amplitude-frequency attenuation slope.

[0095] The second frequency sweep module is used to control the microwave signal source to sample using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region;

[0096] The signal-to-noise ratio (SNR) sampling module is used to obtain the real-time SNR of each frequency point and adjust the sampling duration and Fourier transform averaging times of each frequency point according to the real-time SNR to generate a fine-scan amplitude-frequency dataset.

[0097] The model fitting module is used to perform peaking detection and roll-off characteristic determination on the fine scan amplitude-frequency dataset. Based on the detection results, it selects a single-pole low-pass model or a multi-pole low-pass model from the preset model library to perform global curve fitting and obtain the standard amplitude-frequency response curve.

[0098] The bandwidth calculation module is used to calculate the frequency value corresponding to attenuation to -3dB based on the standard amplitude-frequency response curve, and use it as the bandwidth test result of the optical receiver under test.

[0099] Specifically, the first frequency sweep module is mainly responsible for performing coarse frequency sweep operations. It sends control commands to the microwave signal source to control the microwave signal source to perform a sparse coarse sweep of the entire frequency band of the optical receiver under test with the first frequency sweep step. During the coarse sweep process, the amplitude-frequency response data corresponding to each coarse sweep frequency point is collected synchronously. Based on these data, an initial amplitude-frequency response curve is generated. Then, according to the attenuation distribution characteristics of the initial amplitude-frequency response curve, the entire test frequency band is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region, providing a basis for regional division for subsequent differentiated frequency sweeps.

[0100] The step adjustment module is used to adaptively adjust the sweep frequency step. It first analyzes the amplitude-frequency response data in the transition core region, calculates the local amplitude-frequency attenuation slope in this region, and then determines the second sweep frequency step based on this local amplitude-frequency attenuation slope. The second sweep frequency step is smaller than the first sweep frequency step, and the two are negatively correlated; that is, the larger the local amplitude-frequency attenuation slope, the smaller the second sweep frequency step. This achieves encrypted sampling in the transition core region, ensuring accurate capture of the amplitude-frequency variation characteristics of this region.

[0101] The second frequency sweep module is responsible for performing fine frequency sweep operations. Based on the three regions divided by the first frequency sweep module, it controls the microwave signal source to use differentiated frequency sweep steps for sampling. The first frequency sweep step is used for sampling in the low-frequency flat region and the high-frequency roll-off region, which can effectively reduce redundant frequency points and shorten the test time. The second frequency sweep step is used for encrypted sampling in the transition core region, which can accurately capture the rapid amplitude and frequency attenuation characteristics, balancing test efficiency and test accuracy.

[0102] The signal-to-noise ratio (SNR) sampling module is used to optimize sampling quality. It acquires the real-time SNR of each swept frequency point and dynamically adjusts the sampling duration and Fourier transform averaging times for each frequency point based on the SNR level. For frequencies with low SNR, the sampling duration is appropriately extended and the Fourier transform averaging times are increased to suppress interference and improve data accuracy. For frequencies with high SNR, the sampling duration is appropriately shortened and the Fourier transform averaging times are reduced to improve testing efficiency. Finally, all optimized sampling data are integrated to generate a fine-scan amplitude-frequency dataset.

[0103] The model fitting module is responsible for processing and fitting the fine-scan amplitude-frequency dataset. It first performs peaking detection and roll-off characteristic determination on the fine-scan amplitude-frequency dataset, analyzes the overall characteristics of the amplitude-frequency response curve, and then selects a suitable low-pass model from the preset model library based on the detection and calculation results. This model library includes single-pole low-pass models and multi-pole low-pass models, which can be flexibly selected according to the actual amplitude-frequency characteristics. The selected model is used to perform full-domain curve fitting on the fine-scan amplitude-frequency dataset to eliminate random errors in the sampling process and obtain a standard amplitude-frequency response curve that fits the actual amplitude-frequency characteristics of the optical receiver under test.

[0104] The bandwidth calculation module is responsible for solving the final bandwidth result. Based on the standard amplitude-frequency response curve generated by the model fitting module, it analyzes the attenuation characteristics of the curve and solves for the frequency value corresponding to the attenuation of the amplitude-frequency response to -3dB. This frequency value is the bandwidth test result of the optical receiver under test, thus completing the entire bandwidth test process.

[0105] This application achieves accurate, efficient, and stable testing of optical receiver bandwidth through the collaborative operation of the aforementioned testing system and its modules. By first coarsely scanning and dividing the amplitude-frequency response region with a first sweep step, and then determining a smaller second sweep step based on the attenuation slope of the transition core region, differentiated sampling is used in different regions. This effectively reduces frequency redundancy, shortens testing time, and accurately captures the rapid attenuation characteristics of the amplitude-frequency response, balancing testing efficiency and accuracy to meet the needs of mass production line testing. By adjusting the sampling duration and the Fourier transform averaging by combining real-time signal-to-noise ratio, the influence of various interference factors can be suppressed. Through peaking detection and roll-off characteristic determination, a suitable low-pass model is matched for full-domain fitting, eliminating random sampling errors and ensuring that the standard amplitude-frequency response curve closely matches actual characteristics, thus improving the stability and repeatability of bandwidth determination. In summary, this method significantly improves testing performance through the organic combination of adaptive non-uniform sweeping and model fitting.

[0106] Furthermore, the host computer processing unit also includes a pre-calibration module for performing no-load link frequency response calibration and optical power linear region locking calibration;

[0107] The process of calibrating the frequency response of the unloaded link includes:

[0108] Disconnect the first output terminal of the beam splitter from the receiver terminal of the optical module under test, and connect only the optical power meter;

[0109] The microwave signal source is controlled to output sinusoidal modulation signals at each frequency point across the entire frequency band in sequence, and the output optical power at each frequency point is recorded by an optical power meter.

[0110] Using the output optical power at the lowest frequency point as a benchmark, the power attenuation at each frequency point is calculated, and a frequency response calibration coefficient table is generated to correct the amplitude-frequency response data at each frequency point.

[0111] The optical power linear region locking calibration process includes:

[0112] Connect the receiver of the optical module under test to the first output of the beam splitter, and adjust the optical adjustable attenuator to gradually increase the optical power input to the receiver of the optical module under test from low to high.

[0113] The output amplitude of the receiver of the optical module under test is acquired synchronously to generate an input optical power-output amplitude curve;

[0114] Identify the linear operating region in the input optical power-output amplitude curve, and adjust the optical adjustable attenuator to the attenuation amount corresponding to the midpoint of the linear operating region, so that the input optical power remains constant within the linear operating region of the photodiode during the full-band frequency sweep.

[0115] Specifically, the optical receiver bandwidth testing system of this application further adds a pre-calibration module to the host computer processing unit on top of the basic architecture. Through the dual calibration process of no-load link frequency response calibration and optical power linear region locking calibration, system errors are eliminated from the source and the optimal test conditions are locked.

[0116] 1. Unloaded link frequency response calibration:

[0117] This calibration process is used to eliminate the interference of the amplitude-frequency characteristics of the optical transmission link itself on the test results. The specific process is as follows: First, disconnect the first output of the beam splitter from the receiver of the optical module under test, leaving only the second output connected to the optical power meter to establish an unloaded test link. Then, the host computer controls a microwave signal source to sequentially output sinusoidal modulation signals at each frequency point across the entire frequency band, driving the intensity modulator to complete optical carrier modulation. The optical power meter simultaneously records the output optical power corresponding to each frequency point. Using the lowest output optical power across the entire frequency band as a benchmark, the power attenuation of each other frequency point relative to the benchmark is calculated, generating a frequency response calibration coefficient table. In subsequent formal testing, this coefficient table is used to correct the collected amplitude-frequency response data at each frequency point in real time, offsetting the frequency response deviation of the optical transmission link itself, and ensuring that the test data only reflects the true amplitude-frequency characteristics of the receiver under test.

[0118] 2. Optical power linear region locking calibration:

[0119] This calibration process is used to lock the optimal input optical power range of the optical receiver under test, avoiding nonlinear distortion from affecting test accuracy. Specifically, the process involves: connecting the receiver of the optical module under test to the first output of the beam splitter to restore the complete test link; the host computer controls the adjustable optical attenuator to gradually adjust the attenuation, increasing the optical power input to the receiver of the optical module under test from low to high; simultaneously, the output amplitude of the receiver of the optical module under test is synchronously acquired through the ADC sampling circuit, generating an input optical power-output amplitude curve; the host computer performs feature analysis on this curve, identifying the linear operating region where the input and output have a linear correspondence, and adjusting the adjustable optical attenuator to the attenuation value corresponding to the midpoint of the linear operating region. This ensures that during subsequent full-band sweep frequency testing, the input optical power remains constant within the linear operating region of the photodiode, fundamentally avoiding nonlinear distortion caused by optical power overload or underload, and guaranteeing the accuracy of the test data.

[0120] The dual calibration of the pre-calibration module eliminates the influence of the system's own frequency response deviation and nonlinear distortion from the source, providing an accurate benchmark and stable operating conditions for testing. By first coarsely scanning and dividing the amplitude-frequency response region with the first frequency sweep step, and then determining a smaller second frequency sweep step based on the attenuation slope of the transition core region, differentiated sampling is used in different regions, which effectively reduces frequency redundancy, shortens test time, and accurately captures the rapid attenuation characteristics of amplitude and frequency, balancing test efficiency and accuracy, and adapting to the batch testing needs of production lines.

[0121] Figure 2 This is a flowchart of an optical receiver bandwidth testing method disclosed in an embodiment of this application.

[0122] like Figure 2 As shown, the method may include:

[0123] Step S1: Control the microwave signal source to perform a full-band sparse coarse scan of the optical receiver under test with the first sweep frequency step. Generate an initial amplitude-frequency response curve based on the amplitude-frequency response data corresponding to each coarse scan frequency point, and divide the initial amplitude-frequency response curve into a low-frequency flat region, a transition core region, and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0124] Specifically, a sparse coarse scan is used to obtain the preliminary amplitude-frequency response characteristics of the optical receiver under test, thereby dividing the test areas into different amplitude-frequency characteristics. This provides a basis for subsequent differentiated fine scan sampling, ensuring a balance between test efficiency and accuracy. Specifically, the first frequency sweep module of the host computer processing unit sends control commands to the microwave signal source, controlling the microwave signal source to perform a sparse coarse scan of the entire frequency band of the optical receiver under test in the first frequency sweep step. During the frequency sweep, the ADC sampling circuit of the synchronous acquisition unit collects the amplitude-frequency response data corresponding to each coarse scan frequency point in real time and transmits the collected data to the host computer processing unit. The host computer processing unit organizes, filters, and analyzes the received amplitude-frequency response data, removing abnormal interference data, and generates an initial amplitude-frequency response curve that can intuitively reflect the amplitude-frequency attenuation characteristics of the optical receiver under test across the entire frequency band. Subsequently, based on the attenuation distribution of the initial amplitude-frequency response curve, the entire test frequency band is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region. The specific division process includes:

[0125] ① The frequency bands with attenuation less than the first preset threshold are divided into the low-frequency flat region;

[0126] ② The frequency bands with attenuation greater than or equal to the first preset threshold and less than or equal to the second preset threshold are divided into the transition core area;

[0127] ③ The frequency bands with attenuation greater than the second preset threshold are divided into the high-frequency roll-off region;

[0128] Wherein, the first preset threshold corresponds to the starting point when the amplitude-frequency response begins to enter a significant decay, and the second preset threshold is greater than the first preset threshold and corresponds to the critical point when the amplitude-frequency response enters the rapid roll-off stage.

[0129] Frequency bands with attenuation less than a first preset threshold are classified as low-frequency flat regions, where amplitude-frequency response attenuation is gradual and signal characteristics are stable. Frequency bands with attenuation greater than or equal to the first preset threshold and less than or equal to a second preset threshold are classified as transition core regions, which are critical areas where amplitude-frequency response transitions from gradual attenuation to rapid attenuation, directly affecting the accuracy of bandwidth testing. Frequency bands with attenuation greater than the second preset threshold are classified as high-frequency roll-off regions, where amplitude-frequency response attenuates rapidly and signal strength continuously decreases. The first preset threshold corresponds to the starting point where amplitude-frequency response begins to show significant attenuation, and the second preset threshold is greater than the first preset threshold and corresponds to the critical point where amplitude-frequency response enters the rapid roll-off stage. This classification method allows for the precise differentiation of frequency bands with different amplitude-frequency characteristics.

[0130] Step S2: Calculate the local amplitude-frequency attenuation slope in the transition core region, and determine the second sweep step based on the local amplitude-frequency attenuation slope, wherein the second sweep step is smaller than the first sweep step and is negatively correlated with the local amplitude-frequency attenuation slope.

[0131] Specifically, by analyzing the amplitude-frequency attenuation characteristics of the transition core area, a suitable encrypted frequency sweep step is determined to ensure the accuracy of the sampling data in the transition core area, thereby improving the overall test accuracy. The step adjustment module of the host computer processing unit performs in-depth analysis of the amplitude-frequency response data within the divided transition core area, calculates the local amplitude-frequency attenuation slope in this area. The calculation process includes: selecting adjacent coarse sweep frequency points within the transition core area, performing differential calculations on the amplitude-frequency attenuation of adjacent frequency points, obtaining the amplitude-frequency attenuation slope of each sub-interval through the calculation results, and then integrating and smoothing the slopes of each sub-interval to finally obtain the local amplitude-frequency attenuation slope of the entire transition core area. This slope can accurately reflect the attenuation rate of the amplitude-frequency response within the transition core area. Subsequently, the second sweep step is determined based on the calculated local amplitude-frequency attenuation slope. The second sweep step is smaller than the first sweep step, and the two are negatively correlated. That is, the larger the local amplitude-frequency attenuation slope, the more drastic the amplitude-frequency response change in the region, and the smaller the required second sweep step. This achieves encrypted sampling of the transition core region, ensuring that the rapid change characteristics of the amplitude-frequency response in the region can be accurately captured.

[0132] The process of calculating the local amplitude-frequency attenuation slope within the transition core region includes:

[0133] Within the transition core region, differential operations are performed on adjacent coarse sweep frequency points to obtain the amplitude-frequency attenuation slope of each sub-interval;

[0134] The negative correlation between the second sweep step and the local amplitude-frequency attenuation slope is expressed as follows:

[0135] ;

[0136] in, For the second sweep frequency step, The slope of the local amplitude-frequency attenuation. This is a preset proportional coefficient. To prevent division by zero, a regularization constant is used.

[0137] Step S3: Control the microwave signal source to perform sampling using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region.

[0138] Specifically, based on the frequency band division and determined sweep step, differentiated fine-scan sampling is performed to maximize test efficiency and avoid wasting test resources while ensuring test accuracy. The second sweep module of the host computer processing unit integrates the frequency band division information from step S1 and the second sweep step parameters determined in step S2, and sends fine-scan control commands to the microwave signal source to control the microwave signal source to perform differentiated fine-scan operations in different regions. In the low-frequency flat region and the high-frequency roll-off region, since the amplitude-frequency response changes gradually in these two regions, dense sampling is not required. Therefore, the first sweep step is continued for sampling, which ensures that the sampled data can fully reflect the amplitude-frequency characteristics of the region and effectively reduces redundant frequency points, shortening the test time. In the transition core region, since the amplitude-frequency response changes drastically in this region, it is a key area that determines the accuracy of bandwidth testing. Therefore, the second sweep step is used for encrypted sampling. By increasing the sampling density, the rapid attenuation details of the amplitude-frequency response are accurately captured, ensuring the integrity and accuracy of the sampled data, and achieving a balance between test efficiency and test accuracy.

[0139] Step S4: Obtain the real-time signal-to-noise ratio of each frequency point, and adjust the sampling duration and the average number of Fourier transforms of each frequency point according to the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset.

[0140] Specifically, during the fine-scan sampling process, the signal-to-noise ratio (SNR) sampling module of the host computer processing unit collects signal and noise data at each scanned frequency point in real time. Through data processing, it calculates the real-time SNR of each frequency point to determine the sampling quality and identify frequencies with severe interference. Subsequently, based on the real-time SNR of each frequency point, it dynamically adjusts the sampling duration and the number of Fourier transforms for that frequency. For frequencies with low SNR and severe interference, the sampling duration is appropriately extended and the number of Fourier transforms is increased to suppress the influence of circuit noise, signal jitter, and other interference factors, thereby improving the accuracy of the sampled data. For frequencies with high SNR and stable signals, the sampling duration is appropriately shortened and the number of Fourier transforms is reduced to further improve testing efficiency while ensuring sampling quality. After all frequency points have been sampled, the optimized sampling data for each frequency point is organized, filtered, and integrated, eliminating invalid data and retaining valid and accurate sampling data to finally generate a fine-scan amplitude-frequency dataset.

[0141] Step S5: Perform peaking detection and roll-off characteristic determination on the fine scan amplitude-frequency dataset. Based on the detection results, select a single-pole low-pass model or a multi-pole low-pass model from the preset model library to perform global curve fitting and obtain the standard amplitude-frequency response curve.

[0142] Specifically, by analyzing the features of the fine-scanned data, a suitable low-pass model is matched for curve fitting to eliminate random errors in the sampling process, resulting in a standard amplitude-frequency response curve that closely matches the actual amplitude-frequency characteristics of the optical receiver under test. The model fitting module of the host computer processing unit performs in-depth processing on the fine-scanned amplitude-frequency dataset generated in step S4. First, peaking detection and roll-off characteristic determination are performed on the dataset. Peaking detection is used to determine whether the amplitude-frequency response curve has peaking characteristics, clarifying the overall shape of the curve; roll-off characteristic determination is used to analyze the attenuation rate of the amplitude-frequency response in the high-frequency region, determining the overall characteristics of the amplitude-frequency response corresponding to the fine-scanned amplitude-frequency dataset. Subsequently, based on the results of peaking detection and roll-off characteristic determination, a suitable low-pass model is matched and selected from the preset model library for global curve fitting. The preset model library includes single-pole low-pass models and multi-pole low-pass models, among which the multi-pole low-pass models include at least a second-order double-pole model. The appropriate model can be flexibly selected according to the actual characteristics of the amplitude-frequency response to ensure that the model can accurately match the amplitude-frequency response characteristics of the optical receiver under test. By performing global curve fitting on the fine scan amplitude-frequency dataset using the selected low-pass model, random errors generated during the sampling process can be effectively eliminated, deviations in the sampling data can be corrected, and the fitted standard amplitude-frequency response curve can better match the actual amplitude-frequency characteristics of the optical receiver under test.

[0143] The single-pole low-pass model is expressed as follows:

[0144] ;

[0145] The multi-pole low-pass model includes at least a second-order two-pole model, expressed as:

[0146] ;

[0147] in, , These are the expressions for the single-pole low-pass model and the second-order double-pole model, respectively. This represents the system's low-frequency and zero-frequency gain. This is the frequency for the sweep test; The -3dB cutoff frequency to be measured; The system resonant frequency is determined by the multipole distribution; The damping coefficient characterizes the degree of peaking in the amplitude-frequency curve. The time curve shows peaking. It is the critical damping.

[0148] Step S6: Based on the standard amplitude-frequency response curve, calculate the frequency value corresponding to the attenuation to -3dB, and use it as the bandwidth test result of the optical receiver under test.

[0149] Specifically, based on the fitted standard amplitude-frequency response curve, the bandwidth test result of the optical receiver under test is calculated, completing the entire test process. The bandwidth calculation module of the host computer processing unit performs a comprehensive analysis of the standard amplitude-frequency response curve obtained in step S5, focusing on the amplitude-frequency attenuation law from low frequency to high frequency, and accurately locating the frequency value corresponding to the amplitude-frequency response attenuation to -3dB. This frequency value is the bandwidth test result of the optical receiver under test.

[0150] The above steps work together to effectively solve problems such as the contradiction between accuracy and efficiency, unreasonable sampling, and unstable test results in traditional testing methods. By first coarsely scanning the partitions, then adaptively adjusting the step size, and then finely scanning and sampling the regions, combined with signal-to-noise ratio optimization and model fitting, the method achieves accurate, efficient, and stable testing of optical receiver bandwidth and is suitable for the testing needs of different types of optical receivers.

[0151] As can be seen from the above technical solutions, the optical receiver bandwidth testing method, system, and related equipment provided in this application first control a microwave signal source to perform a full-band sparse coarse sweep with a first sweep step, generating an initial amplitude-frequency response curve and dividing it into a low-frequency flat region, a transition core region, and a high-frequency roll-off region; then, based on the local amplitude-frequency attenuation slope of the transition core region, a smaller second sweep step is determined, and different step sampling is used in different regions; subsequently, the sampling duration and the average number of Fourier transforms are adjusted in combination with the real-time signal-to-noise ratio to generate a fine-sweep amplitude-frequency dataset; finally, through peaking detection and roll-off characteristic determination, a suitable low-pass model is matched to perform full-domain curve fitting, and the -3dB frequency is solved as the test result.

[0152] This application uses a coarse scan to divide the optical receiver's amplitude-frequency response into different regions, and adaptively adjusts the sweep step based on the local amplitude-frequency attenuation slope in the transition core region. A larger first sweep step is used in the low-frequency flat region and the high-frequency roll-off region to reduce frequency redundancy, effectively shorten test time, and improve test efficiency, adapting to the actual needs of batch testing on production lines. A smaller second sweep step is used in the transition core region for encrypted sampling, accurately capturing the rapid amplitude-frequency attenuation characteristics, ensuring test accuracy, and achieving a dual improvement in test efficiency and accuracy. By dividing the region based on the attenuation distribution of the initial amplitude-frequency response curve and using differentiated sampling, test resources can be rationally allocated, avoiding resource waste in the low-frequency flat region while ensuring sampling density in the transition core region, significantly improving sampling rationality and resource utilization. By acquiring the real-time signal-to-noise ratio at each frequency sweep point and dynamically adjusting the sampling duration and the Fourier transform averaging, the influence of interference factors such as circuit noise, optical power fluctuations, and signal jitter can be effectively suppressed. Combined with the characteristics of the optical receiver's low-pass system, by using peaking detection and roll-off characteristic determination to match a suitable single-pole or multi-pole low-pass model for global curve fitting, random errors in the sampling process can be effectively eliminated. This makes the generated standard amplitude-frequency response curve more closely match the actual amplitude-frequency characteristics of the optical receiver, thereby improving the stability and repeatability of the -3dB bandwidth determination and ensuring the reliability of the test results. In summary, this application, through the combination of adaptive non-uniform frequency sweeping and model fitting, achieves accurate, efficient, and stable testing of optical receiver bandwidth, significantly improving test performance.

[0153] In some embodiments of this application, the process of step S4, obtaining the real-time signal-to-noise ratio of each frequency point and adjusting the sampling duration and Fourier transform averaging of each frequency point based on the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset, is described, and may specifically include:

[0154] ① Obtain a preset first mapping function between signal-to-noise ratio and sampling duration, and a second mapping function between signal-to-noise ratio and the average number of Fourier transforms. Both the first mapping function and the second mapping function are monotonically increasing nonlinear functions.

[0155] ② Obtain the real-time signal-to-noise ratio of the current frequency point, and substitute the real-time signal-to-noise ratio into the first mapping function and the second mapping function respectively to calculate the target sampling duration and target average number of times corresponding to the current frequency point;

[0156] ③ Collect the time domain signal of the current frequency point according to the target sampling duration, and perform Fourier transform averaging on the collected time domain signal according to the target averaging number to obtain the amplitude-frequency response data of the current frequency point. After traversing all fine-scan frequency points, generate the fine-scan amplitude-frequency dataset.

[0157] Specifically, for frequency points where the real-time signal-to-noise ratio is lower than the preset low-noise threshold, the calculated target sampling duration is greater than the preset baseline sampling duration, and the target average number of times is greater than the preset baseline average number of times. For frequency points where the real-time signal-to-noise ratio is higher than the preset high-noise threshold, the calculated target sampling duration is less than the preset baseline sampling duration, and the target average number of times is less than the preset baseline average number of times.

[0158] Specifically, firstly, two sets of preset mapping functions are pre-configured and acquired: a first mapping function corresponding to the signal-to-noise ratio (SNR) and sampling duration, and a second mapping function corresponding to the SNR and the average Fourier transform (AFT) count. Both the first and second mapping functions are monotonically increasing non-linear functions. Through these two sets of mapping functions, a precise correspondence between the SNR and sampling parameters can be established, ensuring that the sampling parameters can adaptively adjust according to changes in the SNR to adapt to the signal quality differences at different sweep frequencies. Subsequently, during the fine-scan sampling process, signal and noise data at each sweep frequency are collected in real time. Through corresponding data processing and analysis, the real-time SNR of the current sweep frequency is accurately obtained. This real-time SNR is then substituted into the preset first and second mapping functions, and the target sampling duration and target AFT count (i.e., the target AFT count) are obtained through function calculations, ensuring that the sampling parameters match the signal quality at the current frequency. Subsequently, according to the calculated target sampling duration, the time-domain signal at the current frequency point is acquired to ensure that the acquired time-domain signal can fully reflect the signal characteristics of that frequency point. Simultaneously, the acquired time-domain signal is subjected to Fourier transform averaging according to the target averaging frequency. Multiple averaging operations effectively suppress the influence of interference factors such as circuit noise and signal jitter, improving the accuracy of the amplitude-frequency response data, thus obtaining the effective amplitude-frequency response data for the current frequency point. Following the above process, all fine-scan frequency points are sequentially traversed. After completing the sampling, signal processing, and amplitude-frequency response data acquisition for each frequency point, the amplitude-frequency response data of all frequency points are organized, filtered, and integrated, eliminating invalid data, ultimately generating a high-quality fine-scan amplitude-frequency dataset. This provides accurate and reliable data support for subsequent peaking detection, model fitting, and bandwidth calculation. To further optimize sampling performance and achieve a balance between test quality and efficiency, preset low-noise and high-noise thresholds are used as benchmarks for signal-to-noise ratio (SNR). For frequency points with a real-time SNR lower than the preset low-noise threshold, it indicates that the signal at that frequency point is severely interfered with and has poor signal quality. The target sampling duration calculated by the mapping function is greater than the preset benchmark sampling duration, and the target Fourier transform average number of times is greater than the preset benchmark average number of times. This suppresses interference and improves the accuracy of the amplitude-frequency response data at that frequency point by extending the sampling time and increasing the average number of times. For frequency points with a real-time SNR higher than the preset high-noise threshold, it indicates that the signal at that frequency point is stable, has less interference, and has good signal quality. The calculated target sampling duration is less than the preset benchmark sampling duration, and the target Fourier transform average number of times is less than the preset benchmark average number of times. This effectively shortens the sampling time and improves the overall test efficiency while ensuring the quality of the sampled data, thus achieving a reasonable allocation of test resources.

[0159] In some embodiments of this application, the process of peaking detection and roll-off characteristic determination in step S5 is described, which may specifically include:

[0160] The process of peaking detection for the fine-scan amplitude-frequency dataset includes:

[0161] ① Perform second-order difference operations on the fine-scan amplitude-frequency dataset to identify local maxima;

[0162] ② Determine the amplitude-frequency peaking amount of the local maximum point relative to the average amplitude-frequency response of the low-frequency flat region;

[0163] ③ If the amplitude-frequency peaking amount exceeds the preset peaking threshold, then it is determined that amplitude-frequency peaking characteristics exist.

[0164] Specifically, firstly, a second-order difference operation is performed on the fine-scan amplitude-frequency dataset. This operation effectively amplifies local variation features in the data and filters out redundant information with gradual changes, thereby accurately identifying local maxima in the dataset. These local maxima are the core characteristics of potential peaking in the amplitude-frequency response, and their location and amplitude directly reflect the specific peaking characteristics. Subsequently, the amplitude-frequency peaking amount of the identified local maxima relative to the average amplitude-frequency response in the low-frequency flat region is determined. The low-frequency flat region, as a region with stable signal and no significant attenuation in the amplitude-frequency response, has its average amplitude-frequency response used as a benchmark. The amplitude-frequency peaking amount is then calculated by comparing the local maxima with this benchmark. The difference between the reference values ​​yields the amplitude-frequency peaking amount, which quantifies the degree of peaking at local maxima relative to stable regions, avoiding peaking judgment bias caused by inconsistent benchmarks. Finally, the calculated amplitude-frequency peaking amount is compared with a preset peaking threshold, which is a critical value for determining whether effective peaking exists based on extensive testing experience. If the amplitude-frequency peaking amount exceeds the preset peaking threshold, the amplitude-frequency response of the optical receiver under test is determined to have amplitude-frequency peaking characteristics; otherwise, it is determined not to have amplitude-frequency peaking characteristics. This determination result will directly determine the selection direction of the subsequent low-pass model, ensuring that the model is compatible with the actual characteristics of the amplitude-frequency response.

[0165] The process of determining the roll-off characteristics of the fine-scan amplitude-frequency dataset includes:

[0166] ① Within the high-frequency roll-off region, the high-frequency roll-off slope is obtained by taking the logarithmic coordinates of the amplitude-frequency response data and performing linear regression.

[0167] ②If the difference between the absolute value of the high-frequency roll-off slope and the quotient of 20dB divided by ten octaves is less than a preset difference threshold, it is determined to be a single-pole roll-off characteristic.

[0168] ③ If the absolute value of the high-frequency roll-off slope is greater than or equal to 40dB divided by ten octaves, it is determined to be a multi-pole roll-off characteristic.

[0169] Specifically, the roll-off slope is calculated first by selecting valid amplitude-frequency response data within the high-frequency roll-off region. During selection, outlier and noise interference data at the edges of the high-frequency roll-off region must be removed to ensure that the selected amplitude-frequency response data accurately reflects the amplitude-frequency attenuation pattern in the high-frequency region, avoiding deviations caused by outlier data in subsequent calculations. Next, the selected valid amplitude-frequency response data undergoes a logarithmic coordinate transformation. This transformation converts the nonlinear amplitude-frequency attenuation relationship in the high-frequency region into a linear one, facilitating accurate calculation of the attenuation slope using linear regression. After the logarithmic coordinate transformation, linear regression is performed on the transformed logarithmic coordinate data. A straight line characterizing the amplitude-frequency attenuation trend in the high-frequency roll-off region is fitted using linear regression. The slope of this line is the high-frequency roll-off slope. The absolute value of this slope directly reflects the attenuation rate of the amplitude-frequency response in the high-frequency roll-off region; a larger absolute value indicates a faster attenuation rate in the high-frequency region.

[0170] Secondly, based on the calculated high-frequency roll-off slope, the single-pole roll-off characteristic is determined. A preset difference threshold, based on extensive testing experience, is a critical value used to distinguish single-pole characteristics from other roll-off characteristics. Its purpose is to avoid judgment errors caused by small slope deviations. The specific judgment logic is as follows: calculate the difference between the absolute value of the high-frequency roll-off slope and the quotient of 20dB divided by a decimal octave. Compare this difference with the preset difference threshold. If the difference is less than the preset difference threshold, it indicates that the amplitude-frequency attenuation law in the high-frequency region highly matches the attenuation characteristics of the single-pole system. In this case, the amplitude-frequency response of the optical receiver under test is determined to have single-pole roll-off characteristics, and subsequent curve fitting using a single-pole low-pass model can be prioritized.

[0171] Finally, the multi-pole roll-off characteristic is determined, and its determination logic complements and does not conflict with that of the single-pole roll-off characteristic determination. Specifically, the absolute value of the calculated high-frequency roll-off slope is compared with the quotient of 40dB divided by ten octaves. If the absolute value of the high-frequency roll-off slope is greater than or equal to the quotient of 40dB divided by ten octaves, it indicates that the amplitude-frequency attenuation rate in the high-frequency region is significantly faster than the attenuation rate of the single-pole system, which is consistent with the attenuation characteristics of a multi-pole system. At this time, it is determined that the amplitude-frequency response of the optical receiver under test has multi-pole roll-off characteristics. Subsequently, a multi-pole low-pass model (at least including a second-order bipole model) needs to be matched for curve fitting to ensure that the fitted curve can accurately fit the rapid attenuation law in the high-frequency region.

[0172] The entire roll-off characteristic determination process employs data processing methods such as logarithmic transformation and linear regression to accurately obtain the high-frequency roll-off slope. Then, by applying preset thresholds for stratified determination, it can clearly distinguish between single-pole and multi-pole roll-off characteristics. Roll-off characteristic determination further clarifies the overall characteristics of the amplitude-frequency response, complementing peaking detection results and providing dual criteria for accurate matching of subsequent low-pass models.

[0173] Furthermore, after obtaining the standard amplitude-frequency response curve, to ensure the effectiveness of the curve fitting and the accuracy of the test results, and to avoid bandwidth calculation errors due to fitting deviations, this application also includes a curve fitting verification and anomaly handling process. This process can promptly identify fitting problems and make targeted adjustments to ensure the reliability of the test, as detailed below:

[0174] ① Calculate the root mean square error between the standard amplitude-frequency response curve and the fine-scan amplitude-frequency dataset;

[0175] ② If the root mean square error is greater than the preset error threshold, and the current model is a single-pole low-pass model, then switch to a multi-pole low-pass model for refitting.

[0176] ③ If the root mean square error is greater than the preset error threshold, and the current model is a multi-pole low-pass model, then increase the model order and refit.

[0177] ④ If the root mean square error is still greater than the preset error threshold after increasing the model order, an abnormal alarm is triggered, indicating that there is nonlinear distortion or link failure at the optical receiver under test.

[0178] Specifically, firstly, the root mean square error (RMSE) between the standard amplitude-frequency response curve and the fine-scan amplitude-frequency dataset is calculated. The RMSE quantifies the degree of fit between the fitted curve and the measured fine-scan data. A smaller error indicates a closer fit between the fitted curve and the actual amplitude-frequency characteristics of the receiver under test, resulting in a better fit. Conversely, a larger error indicates a deviation in the fit, requiring adjustment. If the calculated RMSE is greater than the preset error threshold, and a single-pole low-pass model is currently used, it means the single-pole model cannot accurately adapt to the current amplitude-frequency response characteristics. In this case, it is necessary to switch to a multi-pole low-pass model and re-fit the global curve. The complex characteristics of the multi-pole model improve the fitting accuracy and reduce the fitting deviation. If the RMSE is still greater than the preset error threshold after switching to the multi-pole low-pass model... If the error exceeds the preset threshold, it indicates that the current multi-pole model order is insufficient to match the amplitude-frequency response characteristics. In this case, the model order is increased to refit the data, further optimizing the model's fitting ability and making the fitted curve more closely match the measured data. If the root mean square error still exceeds the preset error threshold after increasing the model order, it means that the fitting deviation is not caused by a model adaptation problem, but by an anomaly in the optical receiver under test or a fault in the test link. In this case, an anomaly alarm is triggered, which will be promptly fed back to the staff, clearly indicating that the optical receiver under test may have nonlinear distortion or a fault in the test link. This alerts the staff to promptly check related issues such as test link connection, optical signal power, and equipment operating status to ensure the normal conduct of the test process and the reliability of the test results.

[0179] The optical receiver bandwidth testing device provided in this application embodiment can be applied to optical receiver bandwidth testing equipment. Figure 3 The hardware structure block diagram of the optical receiver bandwidth testing equipment is shown below. Figure 3 The hardware structure of the optical receiver bandwidth test equipment may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4.

[0180] In this embodiment of the application, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4;

[0181] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0182] Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device;

[0183] The memory stores a program, which the processor can call. The program is used for:

[0184] The microwave signal source is controlled to perform a full-band sparse coarse scan of the optical receiver under test with the first sweep frequency step. An initial amplitude-frequency response curve is generated based on the amplitude-frequency response data corresponding to each coarse scan frequency point. The initial amplitude-frequency response curve is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0185] Calculate the local amplitude-frequency attenuation slope within the transition core region, and determine the second sweep step based on the local amplitude-frequency attenuation slope, wherein the second sweep step is smaller than the first sweep step and is negatively correlated with the local amplitude-frequency attenuation slope;

[0186] The microwave signal source is controlled to perform sampling using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region;

[0187] The real-time signal-to-noise ratio of each frequency point is obtained, and the sampling duration and the number of Fourier transforms are adjusted according to the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset.

[0188] Peaking detection and roll-off characteristic determination are performed on the fine scan amplitude-frequency dataset. Based on the detection results, a single-pole low-pass model or a multi-pole low-pass model is matched and selected from the preset model library for global curve fitting to obtain the standard amplitude-frequency response curve.

[0189] Based on the standard amplitude-frequency response curve, the frequency value corresponding to the attenuation to -3dB is calculated and used as the bandwidth test result of the optical receiver under test.

[0190] Optionally, the refined and extended functions of the program can be referred to the above description.

[0191] This application embodiment also provides a readable storage medium that can store a program suitable for execution by a processor, the program being used for:

[0192] The microwave signal source is controlled to perform a full-band sparse coarse scan of the optical receiver under test with the first sweep frequency step. An initial amplitude-frequency response curve is generated based on the amplitude-frequency response data corresponding to each coarse scan frequency point. The initial amplitude-frequency response curve is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0193] Calculate the local amplitude-frequency attenuation slope within the transition core region, and determine the second sweep step based on the local amplitude-frequency attenuation slope, wherein the second sweep step is smaller than the first sweep step and is negatively correlated with the local amplitude-frequency attenuation slope;

[0194] The microwave signal source is controlled to perform sampling using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region;

[0195] The real-time signal-to-noise ratio of each frequency point is obtained, and the sampling duration and the number of Fourier transforms are adjusted according to the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset.

[0196] Peaking detection and roll-off characteristic determination are performed on the finely scanned amplitude-frequency dataset. Based on the detection results, a single-pole low-pass model or a multi-pole low-pass model is matched and selected from the preset model library for global curve fitting to obtain the standard amplitude-frequency response curve.

[0197] Based on the standard amplitude-frequency response curve, the frequency value corresponding to the attenuation to -3dB is calculated and used as the bandwidth test result of the optical receiver under test.

[0198] Optionally, the refined and extended functions of the program can be referred to the above description.

[0199] This application also provides a computer program product, including a computer program, wherein the computer program is executed by a processor using the following method:

[0200] The microwave signal source is controlled to perform a full-band sparse coarse scan of the optical receiver under test with the first sweep frequency step. An initial amplitude-frequency response curve is generated based on the amplitude-frequency response data corresponding to each coarse scan frequency point. The initial amplitude-frequency response curve is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve.

[0201] Calculate the local amplitude-frequency attenuation slope within the transition core region, and determine the second sweep step based on the local amplitude-frequency attenuation slope, wherein the second sweep step is smaller than the first sweep step and is negatively correlated with the local amplitude-frequency attenuation slope;

[0202] The microwave signal source is controlled to perform sampling using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region;

[0203] The real-time signal-to-noise ratio of each frequency point is obtained, and the sampling duration and the number of Fourier transforms are adjusted according to the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset.

[0204] Peaking detection and roll-off characteristic determination are performed on the finely scanned amplitude-frequency dataset. Based on the detection results, a single-pole low-pass model or a multi-pole low-pass model is matched and selected from the preset model library for global curve fitting to obtain the standard amplitude-frequency response curve.

[0205] Based on the standard amplitude-frequency response curve, the frequency value corresponding to the attenuation to -3dB is calculated and used as the bandwidth test result of the optical receiver under test.

[0206] Optionally, the refined and extended functions of the program can be referred to the above description.

[0207] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0208] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0209] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for testing the bandwidth of an optical receiver, characterized in that, The methods include: The microwave signal source is controlled to perform a full-band sparse coarse scan of the optical receiver under test with the first sweep frequency step. An initial amplitude-frequency response curve is generated based on the amplitude-frequency response data corresponding to each coarse scan frequency point. The initial amplitude-frequency response curve is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve. Calculate the local amplitude-frequency attenuation slope within the transition core region, and determine the second sweep step based on the local amplitude-frequency attenuation slope, wherein the second sweep step is smaller than the first sweep step and is negatively correlated with the local amplitude-frequency attenuation slope; The microwave signal source is controlled to perform sampling using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region; The real-time signal-to-noise ratio of each frequency point is obtained, and the sampling duration and the number of Fourier transforms are adjusted according to the real-time signal-to-noise ratio to generate a fine-scan amplitude-frequency dataset. Peaking detection and roll-off characteristic determination are performed on the fine scan amplitude-frequency dataset. Based on the detection results, a single-pole low-pass model or a multi-pole low-pass model is matched and selected from the preset model library for global curve fitting to obtain the standard amplitude-frequency response curve. Based on the standard amplitude-frequency response curve, the frequency value corresponding to the attenuation to -3dB is calculated and used as the bandwidth test result of the optical receiver under test.

2. The method according to claim 1, characterized in that, Based on the attenuation distribution of the initial amplitude-frequency response curve, the region is divided into a low-frequency flat region, a transition core region, and a high-frequency roll-off region, including: The frequency band with attenuation less than the first preset threshold is divided into the low-frequency flat region; The frequency band with attenuation greater than or equal to the first preset threshold and less than or equal to the second preset threshold is divided into the transition core region; The frequency band with attenuation greater than the second preset threshold is divided into the high-frequency roll-off region; Wherein, the first preset threshold corresponds to the starting point when the amplitude-frequency response begins to enter a significant decay, and the second preset threshold is greater than the first preset threshold and corresponds to the critical point when the amplitude-frequency response enters the rapid roll-off stage.

3. The method according to claim 1, characterized in that, The process of calculating the local amplitude-frequency attenuation slope within the transition core region includes: Within the transition core region, differential operations are performed on adjacent coarse sweep frequency points to obtain the amplitude-frequency attenuation slope of each sub-interval; The negative correlation between the second sweep step and the local amplitude-frequency attenuation slope is expressed as follows: ； in, For the second sweep frequency step, The slope of the local amplitude-frequency attenuation. This is a preset proportional coefficient. To prevent division by zero, a regularization constant is used.

4. The method according to claim 1, characterized in that, The step of obtaining the real-time signal-to-noise ratio (SNR) of each frequency point and adjusting the sampling duration and Fourier transform averaging times of each frequency point based on the real-time SNR to generate a fine-scan amplitude-frequency dataset includes: Obtain a preset first mapping function between signal-to-noise ratio and sampling duration, and a second mapping function between signal-to-noise ratio and the average number of Fourier transforms, wherein both the first mapping function and the second mapping function are monotonically increasing nonlinear functions; Obtain the real-time signal-to-noise ratio of the current frequency point, and substitute the real-time signal-to-noise ratio into the first mapping function and the second mapping function respectively to calculate the target sampling duration and target average number of times corresponding to the current frequency point; The time-domain signal of the current frequency point is collected according to the target sampling duration, and the collected time-domain signal is subjected to Fourier transform averaging processing according to the target averaging number to obtain the amplitude-frequency response data of the current frequency point. After traversing all fine-scan frequency points, the fine-scan amplitude-frequency dataset is generated. Specifically, for frequency points where the real-time signal-to-noise ratio is lower than the preset low-noise threshold, the calculated target sampling duration is greater than the preset baseline sampling duration, and the target average number of times is greater than the preset baseline average number of times. For frequency points where the real-time signal-to-noise ratio is higher than the preset high-noise threshold, the calculated target sampling duration is less than the preset baseline sampling duration, and the target average number of times is less than the preset baseline average number of times.

5. The method according to claim 1, characterized in that, The process of peaking detection for the fine-scan amplitude-frequency dataset includes: Perform second-order difference operations on the fine-scan amplitude-frequency dataset to identify local maxima. Determine the amplitude-frequency peaking amount of the local maximum point relative to the average amplitude-frequency response in the low-frequency flat region; If the amplitude-frequency peaking amount exceeds the preset peaking threshold, then the amplitude-frequency peaking feature is determined to exist.

6. The method according to claim 1, characterized in that, The process of determining the roll-off characteristics of the fine-scan amplitude-frequency dataset includes: Within the high-frequency roll-off region, the high-frequency roll-off slope is obtained by taking logarithmic coordinates of the amplitude-frequency response data and performing linear regression. If the difference between the absolute value of the high-frequency roll-off slope and the quotient of 20dB divided by ten octaves is less than a preset difference threshold, it is determined to be a single-pole roll-off characteristic. If the absolute value of the high-frequency roll-off slope is greater than or equal to 40dB divided by ten octaves, it is determined to be a multi-pole roll-off characteristic.

7. The method according to claim 1, characterized in that, The single-pole low-pass model is expressed as follows: ； The multi-pole low-pass model includes at least a second-order two-pole model, expressed as: ； in, , These are the expressions for the single-pole low-pass model and the second-order double-pole model, respectively. This represents the system's low-frequency and zero-frequency gain. This is the frequency for the sweep test; The -3dB cutoff frequency to be measured; The system resonant frequency is determined by the multipole distribution; The damping coefficient characterizes the degree of peaking in the amplitude-frequency curve. The time curve shows peaking. It is the critical damping.

8. The method according to claim 1, characterized in that, After obtaining the standard amplitude-frequency response curve, the following steps are also included: Calculate the root mean square error between the standard amplitude-frequency response curve and the fine-scan amplitude-frequency dataset; If the root mean square error is greater than the preset error threshold, and the current model is a single-pole low-pass model, then switch to a multi-pole low-pass model for refitting. If the root mean square error is greater than the preset error threshold, and the current model is a multi-pole low-pass model, then the model order is increased and the model is refitted. If the root mean square error is still greater than the preset error threshold after increasing the model order, an abnormal alarm is triggered, indicating that there is nonlinear distortion or link failure at the optical receiver under test.

9. A system for testing the bandwidth of an optical receiver, characterized in that, include: The optical transmission link consists of a tunable laser, an intensity modulator, a microwave signal source, and an optical tunable attenuator connected in sequence. The microwave signal source is used to output a frequency-tunable sinusoidal modulation signal to drive the intensity modulator, and the optical tunable attenuator is used to adjust the output optical power. The beam splitter has its input end connected to the output end of the optical adjustable attenuator, its first output end connected to the receiver of the optical module under test, and its second output end connected to the optical power meter. An optical power meter is used to measure the amplitude-frequency response of the optical transmission link during no-load calibration, generate calibration coefficients, and continuously monitor the optical power at the second output of the splitter during bandwidth testing. The synchronous acquisition unit includes a clock source and an ADC sampling circuit, wherein the clock source provides clock signals to both the microwave signal source and the ADC sampling circuit. The host computer processing unit is connected to the microwave signal source, the optical adjustable attenuator, the optical power meter, and the ADC sampling circuit, respectively, and is used to execute each step of the optical receiver bandwidth testing method as described in any one of claims 1-8.

10. The system according to claim 9, characterized in that, The host computer processing unit includes: The first frequency sweep module is used to control the microwave signal source to perform a full-band sparse coarse sweep of the optical receiver under test with the first frequency sweep step, generate an initial amplitude-frequency response curve based on the amplitude-frequency response data corresponding to each coarse sweep frequency point, and divide the initial amplitude-frequency response curve into a low-frequency flat region, a transition core region and a high-frequency roll-off region according to the attenuation distribution of the initial amplitude-frequency response curve. The step adjustment module is used to calculate the local amplitude-frequency attenuation slope in the transition core region and determine the second sweep frequency step based on the local amplitude-frequency attenuation slope, wherein the second sweep frequency step is smaller than the first sweep frequency step and is negatively correlated with the local amplitude-frequency attenuation slope. The second frequency sweep module is used to control the microwave signal source to sample using the first frequency sweep step in the low-frequency flat region and the high-frequency roll-off region, and to perform encrypted sampling using the second frequency sweep step in the transition core region; The signal-to-noise ratio (SNR) sampling module is used to obtain the real-time SNR of each frequency point and adjust the sampling duration and Fourier transform averaging times of each frequency point according to the real-time SNR to generate a fine-scan amplitude-frequency dataset. The model fitting module is used to perform peaking detection and roll-off characteristic determination on the fine scan amplitude-frequency dataset. Based on the detection results, it selects a single-pole low-pass model or a multi-pole low-pass model from the preset model library to perform global curve fitting and obtain the standard amplitude-frequency response curve. The bandwidth calculation module is used to calculate the frequency value corresponding to attenuation to -3dB based on the standard amplitude-frequency response curve, and use it as the bandwidth test result of the optical receiver under test.

11. The system according to claim 10, characterized in that, The host computer processing unit also includes a pre-calibration module for performing no-load link frequency response calibration and optical power linear region locking calibration; The process of calibrating the frequency response of the unloaded link includes: Disconnect the first output terminal of the beam splitter from the receiver terminal of the optical module under test, and connect only the optical power meter; The microwave signal source is controlled to output sinusoidal modulation signals at each frequency point across the entire frequency band in sequence, and the output optical power at each frequency point is recorded by an optical power meter. Using the output optical power at the lowest frequency point as a benchmark, the power attenuation at each frequency point is calculated, and a frequency response calibration coefficient table is generated to correct the amplitude-frequency response data at each frequency point. The optical power linear region locking calibration process includes: Connect the receiver of the optical module under test to the first output of the beam splitter, and adjust the optical adjustable attenuator to gradually increase the optical power input to the receiver of the optical module under test from low to high. The output amplitude of the receiver of the optical module under test is acquired synchronously to generate an input optical power-output amplitude curve; Identify the linear operating region in the input optical power-output amplitude curve, and adjust the optical adjustable attenuator to the attenuation amount corresponding to the midpoint of the linear operating region, so that the input optical power remains constant within the linear operating region of the photodiode during the full-band frequency sweep.

12. An optical receiver bandwidth testing device, characterized in that, Including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the optical receiver bandwidth testing method as described in any one of claims 1-8.

13. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements each step of the optical receiver bandwidth testing method as described in any one of claims 1-8.

14. A computer program product, comprising a computer program, characterized in that, The computer program is executed by the processor to perform each step of the optical receiver bandwidth testing method as described in any one of claims 1-8.

Images