Method and apparatus for measuring frequency response characteristics of optoelectronic devices

Abstract

The application relates to the technical field of electric digital data processing and provides a method and device for measuring frequency response characteristics of an optoelectronic device. The method comprises: obtaining an input waveform of an input signal at an input node of the device to be measured by a first collector, and obtaining an output waveform of an output signal at an output node by a second collector; and performing a waveform adjustment operation on the waveform corresponding to the first optical signal by a processor according to the waveform corresponding to the first electric signal, so as to convert the first optical signal to obtain a second electric signal, and performing a differential processing on the first and second electric signals to obtain a first differential electric signal. The waveform adjustment operation is used for reducing the common mode component of the first differential electric signal, and the differential mode component of the first differential electric signal is used for calculating the frequency response characteristics of the device to be measured. Thus, the system frequency response characteristics of the electric-to-optical conversion or the optical-to-electric conversion are accurately measured, and the miniaturization design, high integration and low power consumption of the chip are facilitated.

Description

Technical Field

[0001] This application relates to the field of electronic digital data processing technology, and in particular to a method and apparatus for measuring the frequency response characteristics of optoelectronic devices. Background Technology

[0002] In applications such as high-speed optical communication systems and data center optical interconnects, data transmission and processing between different nodes and machines may require frequent signal amplification or attenuation to match differences between upstream and downstream circuits. It may also necessitate adjusting the frequency components of the electrical signal to counteract inter-symbol interference (ISI) caused by the signal transmission path, such as printed circuit boards (PCBs) and cables, thereby improving the error-free transmission distance. For example, an equalizer can be used to correct the frequency response characteristics of the transmission path, reduce ISI, and provide compensation, helping to restore signal amplitude, rise time, and fall time. Furthermore, when the system includes optoelectronic devices—that is, devices that convert optical signals into electrical signals or vice versa—such as photodiodes (PDs), photodetectors, phototransistors, photoresistors, and photoelectric switches, it also involves the system frequency response characteristics of electro-to-optical converters (such as drivers) and optical-to-electrical converters (such as trans-impedance amplifiers, where an electrical signal is converted into an optical signal after being regulated by an equalizer. During system operation, factors such as device aging and wear can cause the performance of optoelectronic devices in the system to deviate from the design goals, thus affecting the compensation effect and signal transmission performance. Therefore, it is necessary to obtain and evaluate the frequency response characteristics of the optoelectronic devices in the system, which can provide a reference for adjusting the configuration and calibrating the equipment.

[0003] However, existing methods for measuring the frequency response characteristics of optoelectronic devices, such as measuring photoelectric scattering parameters (S-parameters), typically use a vector network analyzer (VNA) to drive a special optical base to generate modulated light for the test, which is then fed back to the electrical input interface of the VNA via the target's output. This method is expensive and requires special calibration to measure the frequency response characteristics of electro-optical or optical-electrical systems. Chinese patent CN119519833A discloses an optoelectronic S-parameter testing device that does not use an optical base. Instead, it uses a Mach-Zehnder modulator (MZM) to modulate the light intensity and amplifies the radio frequency signal generated by the VNA for testing to obtain the modulated optical signal. However, the optoelectronic S-parameter testing device disclosed in Chinese patent CN119519833A requires optical modulation technology to load changes in the electrical signal onto the optical signal, thus necessitating additional instruments and resources. Chinese Patent CN106878207B discloses a method and apparatus for measuring filtering characteristics. This method utilizes the inherent characteristics of the transmitter or transceiver to determine the filtering characteristics of the transmitting end without requiring additional optical modulation and demodulation instruments. It measures the average power of multiple transmitted signals after passing through the transmitting end's filtering module and modulation, and then determines the filtering characteristics corresponding to different frequency points based on these average power values ​​and signal amplitudes, ultimately determining the filtering characteristics of the transmitting end. However, the method and apparatus disclosed in Chinese Patent CN106878207B require transmitting multiple signals at different frequencies or multiple signals at the same frequency but different amplitudes. Therefore, a frequency scanning strategy or amplitude scanning strategy is needed. This means that one signal is transmitted at a time, and the frequency or amplitude of each transmitted signal is changed to determine the filtering characteristics corresponding to different frequency points based on the average power, amplitude, and characteristics of the optical modulator, thus determining the filtering characteristics of the transmitting end. This results in a complex and time-consuming overall measurement scheme. US Patent No. 6718276B2 discloses a method for measuring frequency response. Instead of using a vector network analyzer, it utilizes a Bit Error Rate Tester (BERT) to measure the bit error rate distribution, thereby determining the voltage level transitions. These voltage level transitions are then converted to the frequency domain to obtain the frequency response characteristics. However, US Patent No. 6718276B2 only addresses the case of electrical signal input to electrical signal output and does not cover the frequency response characteristics of systems that convert from electro-optical or optical-to-electrical signals.Chinese patent publication number CN117200916A discloses an optical module that uses frequency response models of the transmitter and receiver to simulate input and output signals, and determines the optimal preset bandwidth using the bit error rate at different preset bandwidths. However, Chinese patent publication number CN117200916A only uses a bit error rate analyzer to compare the input and output signals to obtain the bit error rate corresponding to different optical signal-to-noise ratios at different preset bandwidths, and does not involve the frequency response characteristics of the electro-optical or optical-electrical system.

[0004] To address these challenges, this application proposes a method and apparatus for measuring the frequency response characteristics of optoelectronic devices. This method not only enables accurate measurement of the frequency response characteristics of electro-optical or opto-electrical systems, but also saves on additional instruments and resources by not relying on optical modulation and demodulation technology. The frequency response characteristics of the device under test are calculated using digital processing algorithms, which is beneficial for miniaturized chip design, high integration, and low power consumption. Summary of the Invention

[0005] In a first aspect, this application provides a method for measuring the frequency response characteristics of an optoelectronic device. The method includes: obtaining an input waveform of an input signal at an input node of the device under test (DUT) using a first acquisition device; and obtaining an output waveform of an output signal at an output node of the DUT using a second acquisition device. The input signal includes multiple frequency components, and the DUT has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal, and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal, and the other is the waveform corresponding to the first optical signal. A processor performs a waveform adjustment operation on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal. The first electrical signal and the second electrical signal are then differentially processed to obtain a first differential electrical signal. The waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the DUT.

[0006] The first aspect of this application sets requirements for the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function matching the frequency response characteristics of the device under test using digital processing algorithms. A time-domain waveform acquisition device is used to acquire the time-domain waveforms of the first electrical signal and the first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into the second electrical signal. This provides a basis for waveform analysis and comparison and simplifies the process. The design of the digital processing algorithm improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. Thus, accurate measurement of the frequency response characteristics of electro-optical or optical-electrical systems is achieved without relying on optical modulation and demodulation technology, saving additional instruments and resources. This is beneficial for chip miniaturization, high integration, and low power consumption.

[0007] In one possible implementation of the first aspect of this application, the input signal is a data signal including the plurality of frequency components, and the amplitude distribution of each of the plurality of frequency components conforms to a preset pattern.

[0008] In one possible implementation of the first aspect of this application, the preset mode indicates that the amplitudes of at least two of the plurality of frequency components are inconsistent.

[0009] In one possible implementation of the first aspect of this application, the preset mode indicates that the amplitudes of the plurality of frequency components are the same.

[0010] In one possible implementation of the first aspect of this application, the input signal is a wave packet signal with a wide frequency range.

[0011] In one possible implementation of the first aspect of this application, the data signal includes a test code pattern, which is an identical codeword having a preset segment length.

[0012] In one possible implementation of the first aspect of this application, the test code includes a pseudo-random binary sequence.

[0013] In one possible implementation of the first aspect of this application, the identical codewords having a preset segment length correspond to the low-frequency portion of the plurality of frequency components and are used to provide a reference for the waveform adjustment operation.

[0014] In one possible implementation of the first aspect of this application, the low-frequency portion of the plurality of frequency components corresponding to the same codeword having a preset segment length is used to calculate the gain ratio or compression ratio of the device under test.

[0015] In one possible implementation of the first aspect of this application, the measurement method further includes: inputting the first electrical signal into a plurality of system transfer functions to obtain a plurality of reference output electrical signals corresponding one-to-one with the plurality of system transfer functions; then, performing differential processing on the first electrical signal and the plurality of reference output electrical signals to obtain a plurality of reference differential electrical signals corresponding one-to-one with the plurality of reference output electrical signals; by comparing the differential mode components of the first differential electrical signal and the differential mode components of the plurality of reference differential electrical signals respectively, iterating using an error function, and selecting the system transfer function with the smallest error from the plurality of system transfer functions, wherein the selected system transfer function with the smallest error is used to determine the frequency response characteristics and system bandwidth of the device under test.

[0016] In one possible implementation of the first aspect of this application, the error function is a function of the number of poles and zeros of the system transfer function, and the model characteristics of each of the plurality of system transfer functions include delay alignment time, rise time, fall time, and amplitude variation.

[0017] In one possible implementation of the first aspect of this application, the processor performs a waveform adjustment operation on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into the second electrical signal. This includes: determining the average value of the waveform corresponding to the first optical signal; then subtracting the average value of the waveform corresponding to the first optical signal from the average value of the waveform corresponding to the first optical signal; and then performing a waveform alignment operation in the waveform adjustment operation based on the waveform corresponding to the first electrical signal.

[0018] In one possible implementation of the first aspect of this application, the input signal is a wave packet signal with a wide frequency range including the plurality of frequency components. The wave packet form of the wave packet signal is determined according to the data format of the pseudo-random code generated by the data generator. The consecutive identical digital segments in the data format are the reference for the amplitude adjustment operation in the waveform adjustment operation. The consecutive identical digital segments are constant 0 segments or constant 1 segments.

[0019] In one possible implementation of the first aspect of this application, the differential-mode component of the first differential electrical signal is used to determine the system transfer function with the minimum error, the system transfer function with the minimum error characterizing the overall change in rise time and fall time of the wave packet signal in the frequency domain after passing through the device under test.

[0020] In one possible implementation of the first aspect of this application, the device under test has an electro-optic conversion function, the input signal is the first electrical signal, the output signal is the first optical signal, the input waveform is the waveform corresponding to the first electrical signal, the output waveform is the waveform corresponding to the first optical signal, and the device under test includes a light emitting device.

[0021] In one possible implementation of the first aspect of this application, the device under test has a photoelectric conversion function, the input signal is the first optical signal, the output signal is the first electrical signal, the input waveform is the waveform corresponding to the first optical signal, the output waveform is the waveform corresponding to the first electrical signal, and the device under test includes a photodetector.

[0022] In one possible implementation of the first aspect of this application, the first collector is different from the second collector, or the first collector and the second collector are the same collector with the capability to acquire electrical signal time-domain waveforms and optical signal time-domain waveforms.

[0023] In one possible implementation of the first aspect of this application, the device under test is any one of a plurality of devices to be calibrated, the input signals of the plurality of devices to be calibrated each come from the same signal generator, and the input signals and output signals of the plurality of devices to be calibrated each are used to determine the frequency response characteristics of the plurality of devices to be calibrated in order to achieve parallel calibration of the plurality of devices to be calibrated.

[0024] Secondly, this application provides a measurement device for the frequency response characteristics of an optoelectronic device. The measurement device includes: a first acquisition unit for acquiring an input waveform of an input signal at an input node of the device under test (DUT); a second acquisition unit for acquiring an output waveform of an output signal at an output node of the DUT, wherein the input signal includes multiple frequency components, and the DUT has an electro-optical conversion function or a photoelectric conversion function, one of the input signal and the output signal is a first electrical signal and the other is a first optical signal, and one of the input waveform and the output waveform is a waveform corresponding to the first electrical signal and the other is a waveform corresponding to the first optical signal; and a processor for performing waveform adjustment operations on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal, and performing differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal, wherein the waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the DUT.

[0025] The second aspect of this application sets requirements for the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function matching the frequency response characteristics of the device under test using digital processing algorithms. A time-domain waveform acquisition device is used to acquire the time-domain waveforms of the first electrical signal and the first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into the second electrical signal. This provides a basis for waveform analysis and comparison and simplifies the process. The design of the digital processing algorithm improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. Thus, accurate measurement of the frequency response characteristics of electro-optical or optical-electrical systems is achieved without relying on optical modulation and demodulation technology, saving additional instruments and resources. This is beneficial for chip miniaturization, high integration, and low power consumption.

[0026] Thirdly, this application provides a fault detection method for optoelectronic devices. The fault detection method includes: obtaining an input waveform of an input signal at an input node of the device under test (DUT) using a first acquisition device; and obtaining an output waveform of an output signal at an output node of the DUT using a second acquisition device. The input signal includes multiple frequency components, and the DUT has electro-optical conversion or photoelectric conversion functions. One of the input signal and the output signal is a first electrical signal, and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal, and the other is the waveform corresponding to the first optical signal. A processor performs waveform adjustment on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal to convert the first optical signal into a second electrical signal. The first electrical signal and the second electrical signal are then differentially processed to obtain a first differential electrical signal. The waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to measure and calibrate the frequency response characteristics of the DUT in real time for fault detection.

[0027] The third aspect of this application sets requirements for the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function matching the frequency response characteristics of the device under test using digital processing algorithms. A time-domain waveform acquisition device is used to acquire the time-domain waveforms of the first electrical signal and the first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into the second electrical signal. This provides a basis for waveform analysis and comparison and simplifies the process. The design of the digital processing algorithm improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. Thus, accurate measurement of the frequency response characteristics of electro-optical or optical-electrical systems is achieved without relying on optical modulation and demodulation technology, saving additional instruments and resources. This is beneficial for chip miniaturization, high integration, and low power consumption. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart illustrating a method for measuring the frequency response characteristics of an optoelectronic device according to a first embodiment of this application.

[0030] Figure 2 A schematic diagram of a measuring device for the frequency response characteristics of an optoelectronic device according to a first embodiment of this application;

[0031] Figure 3 A schematic flowchart illustrating a fault detection method for optoelectronic devices provided in an embodiment of this application;

[0032] Figure 4 A schematic flowchart illustrating a waveform adjustment operation provided in an embodiment of this application;

[0033] Figure 5 A schematic diagram illustrating the measurement of the frequency response characteristics of an optoelectronic device in a vehicle-mounted lidar system, provided as an embodiment of this application.

[0034] Figure 6 A schematic diagram illustrating the measurement of frequency response characteristics of an optoelectronic device in an optical interconnect for a data center, provided as an embodiment of this application;

[0035] Figure 7 A flowchart illustrating a method for measuring the frequency response characteristics of an optoelectronic device according to a second embodiment of this application.

[0036] Figure 8 A schematic diagram of a measurement device for the frequency response characteristics of an optoelectronic device according to a second embodiment of this application;

[0037] Figure 9 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0038] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0039] It should be understood that in the description of this application, "at least one" means one or more, and "multiple" means two or more. In addition, the words "first," "second," etc., unless otherwise stated, are used only for the purpose of distinguishing descriptions and should not be construed as indicating or implying relative importance or order.

[0040] Figure 1 This is a schematic flowchart illustrating a method for measuring the frequency response characteristics of an optoelectronic device according to a first embodiment of this application. Figure 1 As shown, the measurement method includes the following steps.

[0041] Step S101: The input waveform of the input signal is obtained at the input node of the device under test by the first acquisition device, and the output waveform of the output signal is obtained at the output node of the device under test by the second acquisition device. The input signal includes multiple frequency components, and the device under test has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal and the other is the waveform corresponding to the first optical signal.

[0042] Step S103: The processor performs waveform adjustment on the waveform corresponding to the first optical signal according to the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal. The processor also performs differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal. The waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the device under test.

[0043] Figure 1The method for measuring the frequency response characteristics of optoelectronic devices according to the first embodiment shown can be used in applications such as high-speed optical communication systems and data center optical interconnects. It measures the frequency response characteristics of electro-optical or optical-electrical systems, and acquires and evaluates the frequency response characteristics of optoelectronic devices in the system, thereby providing a reference for configuration adjustment and equipment calibration. Here, an optoelectronic device refers to a device that converts optical signals into electrical signals or vice versa, including light-emitting devices and light-detecting devices. Exemplary optoelectronic devices include, but are not limited to, photodiodes (PDs), photodetectors, phototransistors, photoresistors, and photoelectric switches. Generally, light-emitting devices use voltage and current to generate electromagnetic radiation (i.e., light), for example, for lighting purposes or as indicator lights. Light-detecting devices, such as phototransistors, are designed to convert received electromagnetic energy (i.e., light) into current or voltage, for example, by using photons to release bound electrons within semiconductor materials. Optical detection devices are generally used in optical sensing and communication. It should be understood that the method and apparatus for measuring the frequency response characteristics of optoelectronic devices provided in the embodiments of this application include... Figure 1The method for measuring the frequency response characteristics of an optoelectronic device according to the first embodiment shown can be used to calculate the frequency response characteristics of a device under test (DUT), wherein the DUT has an electro-optical conversion function or a photoelectric conversion function. Therefore, the DUT internally includes optoelectronic devices, such as light emitting devices or light detecting devices, and there are no restrictions on the number and composition of the optoelectronic devices internally, as long as the DUT ultimately has an electro-optical conversion function or a photoelectric conversion function. In other words, the input and output sides of the DUT must be from an electrical signal input to an optical signal output, or from an optical signal input to an electrical signal output, that is, converting the input electrical signal into an output optical signal or converting the input optical signal into an output electrical signal. Therefore, calculating the frequency response characteristics of the DUT means calculating the frequency response characteristics of an electro-optical or optical-electrical system. In one case, the internal structure of the DUT may correspond to the frequency response characteristics of an electro-optical system, that is, from an electrical signal input to an optical signal output. For example, the DUT internally includes a driver and a laser for converting the input electrical signal into an output optical signal. In another scenario, the internal structure of the device under test (DUT) may correspond to the frequency response characteristics of a photoelectric conversion system, that is, from optical signal input to electrical signal output. For example, the DUT may contain a photodiode (PD) and a transimpedance amplifier (TIA) to convert the input optical signal into an output electrical signal. During system operation, factors such as device aging and wear may cause the performance of the optoelectronic devices in the system to deviate from the design goals, thus affecting the compensation effect and signal transmission performance. Therefore, it is necessary to obtain and evaluate the frequency response characteristics of the optoelectronic devices in the system, which can provide a reference for adjusting the configuration and calibrating the equipment. The measurement scheme for the system frequency response characteristics from electrical signal input to electrical signal output can be achieved by connecting test equipment such as a vector network analyzer (VNA) or a bit error rate tester (BERT) to both the input and output sides. By sending a test signal and then comparing the test signal with the signal after passing through the DUT, the frequency response characteristics of the DUT can be measured. However, for devices under test with electro-optical or photoelectric conversion functions, because the frequency response characteristics of the system involving electro-optical or photoelectric conversion are involved, the test equipment also needs to be combined with an additional optical modulation instrument to load the changes in the electrical signal provided by the test equipment onto the optical signal, or combined with an additional optical demodulation instrument to demodulate the changes in the electrical signal loaded onto the optical signal from the optical signal. This results in the occupation of additional equipment and resources, and also makes the overall measurement scheme complex and time-consuming.While frequency scanning or amplitude scanning strategies can be employed, transmitting one signal at a time, and by changing the frequency or amplitude of each transmitted signal, the filtering characteristics corresponding to different frequency points can be determined based on the average power, amplitude, and characteristics of the optical modulator, thus determining the filtering characteristics of the transmitting end. However, measurement schemes with only a single frequency component as input and those that change that single frequency component for frequency scanning, or similar amplitude scanning-based measurement schemes, require dedicated equipment and control mechanisms for constructing the input signal, and also require dedicated equipment on the output side to determine the average power. This results in a resource-intensive, complex, and time-consuming overall measurement scheme. The following description, in conjunction with specific embodiments of this application, illustrates this point. Figure 1 The method for measuring the frequency response characteristics of optoelectronic devices according to the first embodiment not only achieves accurate measurement of the frequency response characteristics of electro-optical or opto-electric systems, but also saves additional instruments and resources by not relying on optical modulation and demodulation technology. The frequency response characteristics of the device under test are calculated by using digital processing algorithms, which is beneficial for chip miniaturization design, high integration and low power consumption.

[0044] See Figure 1In step S101, the first acquisition device obtains the input waveform of the input signal at the input node of the device under test (DUT), and the second acquisition device obtains the output waveform of the output signal at the output node of the DUT. The input signal includes multiple frequency components, and the DUT has electro-optical conversion or photoelectric conversion functions. One of the input signal and the output signal is a first electrical signal, and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal, and the other is the waveform corresponding to the first optical signal. Thus, the first acquisition device with time-domain waveform acquisition function obtains the input waveform of the input signal at the input node of the DUT, i.e., actually acquires the time-domain effect diagram of the input signal. Similarly, the second acquisition device with time-domain waveform acquisition function obtains the output waveform of the output signal at the output node of the DUT, i.e., actually acquires the time-domain effect diagram of the output signal. It should be understood that in some embodiments, the first acquisition device and the second acquisition device are independent of each other, i.e., two different acquisition devices. For example, the first acquisition device is different from the second acquisition device. In some embodiments, the first and second acquisition devices can be the same acquisition device, but they acquire the time-domain waveforms of electrical signals and optical signals through different ports. For example, the first and second acquisition devices can be the same acquisition device, such as an oscilloscope, that has the capability to acquire both electrical and optical signal time-domain waveforms. Thus, the input waveform of the input signal actually acquired by the time-domain waveform acquisition device, and the output waveform of the output signal actually acquired by the time-domain waveform acquisition device, serve as reference benchmarks for subsequent processes. Through waveform comparison and analysis in subsequent processes, the frequency response characteristics of the device under test (DUT) can be calculated. Here, the DUT has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal, and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal, and the other is the waveform corresponding to the first optical signal. Therefore, the frequency response characteristics of the DUT include the system frequency response characteristics of the electro-optical or photoelectric conversion of the optoelectronic device. The input and output sides of the device under test (DUT) must be either from an electrical signal input to an optical signal output, or from an optical signal input to an electrical signal output. In other words, it converts the input electrical signal into an output optical signal or vice versa. When the input signal is a first electrical signal and the output signal is a first optical signal, this corresponds to the frequency response characteristics of an electro-optical system, meaning it converts from an electrical signal input to an optical signal output. For example, the DUT may internally include a driver and a laser to convert the input electrical signal into an output optical signal. This means that the input waveform of the input signal is the waveform corresponding to the first electrical signal, and the output waveform of the output signal is the waveform corresponding to the first optical signal.When the input signal is a first optical signal and the output signal is a first electrical signal, this corresponds to the frequency response characteristics of the optical-to-electrical system, that is, from optical signal input to electrical signal output. For example, the device under test (DUT) internally includes a photodiode and a transimpedance amplifier to convert the input optical signal into the output electrical signal. This means that the input waveform of the input signal is the waveform corresponding to the first optical signal, and the output waveform of the output signal is the waveform corresponding to the first electrical signal. Thus, in step S101, the time-domain waveforms of the electrical signal and the optical signal are acquired using a time-domain waveform acquisition device. This allows for subsequent time-domain waveform analysis and comparison, as well as digital processing algorithms, to measure the frequency response characteristics of the DUT. Furthermore, depending on whether the DUT has an electro-optical conversion function or a photoelectric conversion function, the time-domain waveform of the corresponding signal is acquired in step S101. It should be understood that regardless of whether the device under test (DUT) has electro-optical conversion or photoelectric conversion capabilities, the two time-domain waveforms acquired are the waveform corresponding to the first electrical signal and the waveform corresponding to the first optical signal. The difference lies in the internal structure of the DUT, which may involve either an input of the first electrical signal to an output of the first optical signal or vice versa. Subsequent time-domain waveform analysis and comparison utilize digital processing algorithms, which do not require strict distinction between electrical and optical signal inputs. This simplifies the design of digital processing algorithms and improves computational efficiency.

[0045] Continue reading Figure 1The input signal includes multiple frequency components. As described above, the input waveform of the input signal actually acquired by the time-domain waveform acquisition device, and the output waveform of the output signal actually acquired by the time-domain waveform acquisition device, serve as reference benchmarks for subsequent processes. Through waveform comparison and analysis in subsequent processes, the frequency response characteristics of the device under test can be calculated. Unlike measurement schemes using a single frequency point or a single frequency component input signal, here, an input signal including multiple frequency components is used. This means that the input signal includes at least two different frequency points or at least two different frequency components. Therefore, by using an input signal including multiple frequency components, without significant changes between the frequency range of the input signal and the frequency range of the output signal, the frequency domain transfer function can be analyzed using digital processing algorithms based on the input waveform of the input signal and the output waveform of the output signal. For example, the higher-order system transfer function corresponding to the frequency response characteristics of the device under test can be derived, the equalization value (e.g., gain ratio or compression ratio) can be calculated, and the peak frequency point, equalization frequency point, and Nyquist frequency point can also be calculated. Clock and data recovery (CDR) circuits and waveform shaping circuits can cause significant variations between the frequency range of the input and output signals, affecting the calculation of the frequency response characteristics of the device under test (DUT). Therefore, DUTs generally do not include clock and data recovery or waveform shaping circuits internally, thus minimizing these variations. In this case, without internal components like clock and data recovery or waveform shaping circuits that would cause significant variations between the input and output frequency ranges, using an input signal with multiple frequency components allows for the construction of a high-order system transfer function corresponding to the frequency response characteristics of the DUT using digital processing algorithms, enabling frequency domain transfer function analysis. Furthermore, the presence of multiple frequency components means that, from a time-domain perspective, the input signal's waveform may exhibit irregular amplitude variations, and the distribution of peaks and troughs in both the input and output time-domain waveforms may also be irregular. Therefore, the digital processing algorithm used to derive the higher-order system transfer function that matches the frequency response characteristics of the device under test does not compare the rise and fall times in the time domain, but determines the overall change of rise and fall times in the frequency domain. That is, by taking advantage of the key design requirement that the input signal has multiple frequency components, the algorithm derives the higher-order system transfer function with the smallest error, thereby characterizing the overall change of rise and fall times in the frequency domain.

[0046] Continue reading Figure 1In step S103, the processor adjusts the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal to convert the first optical signal into a second electrical signal. Then, differential processing is performed on the first and second electrical signals to obtain a first differential electrical signal. The waveform adjustment operation reduces the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the device under test (DUT). Thus, using an input signal comprising multiple frequency components, and through a time-domain waveform acquisition device, the time-domain waveforms of the input node and the output node are actually acquired. Here, because the DUT has electro-optical or photoelectric conversion functions, the two acquired time-domain waveforms are the waveform corresponding to the first electrical signal and the waveform corresponding to the first optical signal, depending on the internal structure of the DUT, which may be from the input of the first electrical signal to the output of the first optical signal or vice versa. This is because the time-domain waveform of the first electrical signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal. Therefore, the processor adjusts the waveform of the first optical signal (when the input signal is the first electrical signal, the output signal is the first optical signal; or vice versa) based on the waveform corresponding to the first electrical signal (which may correspond to either the input signal or the output signal). This adjustment is performed to convert the first optical signal into a second electrical signal. Then, the processor performs differential processing on the first and second electrical signals to obtain a first differential electrical signal. Thus, based on digital processing algorithms, the frequency domain transfer function is analyzed, and the higher-order system transfer function corresponding to the frequency response characteristics of the device under test is derived. Furthermore, the waveform adjustment operation reduces the common-mode component of the first differential electrical signal, and the differential-mode component is used to calculate the frequency response characteristics of the device under test. Here, through the waveform adjustment operation, the first optical signal is converted into a second electrical signal based on the time-domain waveform of the acquired first optical signal. The time-domain waveform of the second electrical signal can then be used for waveform analysis and comparison with the time-domain waveform of the first electrical signal. Digital processing algorithms used to derive high-order system transfer functions that match the frequency response characteristics of the device under test can take into account not only the changes in rise time and fall time, but also other factors in the changes between the input and output signals, thereby improving calculation accuracy, such as amplitude changes and delay alignment.Therefore, the first and second electrical signals are differentially processed to obtain a first differential electrical signal. This differential processing allows for better capture of the rise time, fall time, and level changes of the time-domain waveform of the electrical signals. The differential-mode component of the first differential electrical signal obtained through differential processing reflects the changes between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test, enabling frequency domain transfer function analysis based on digital processing algorithms. Furthermore, the waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal. This means that the common-mode component of the first differential electrical signal obtained through differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves calculation accuracy.

[0047] In short, Figure 1 The first embodiment of the method for measuring the frequency response characteristics of optoelectronic devices, as shown, imposes requirements on the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function matching the frequency response characteristics of the device under test using digital processing algorithms. A time-domain waveform acquisition device is used to acquire the time-domain waveforms of a first electrical signal and a first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into a second electrical signal, thus providing waveform analysis capabilities. The comparison simplifies the design of digital processing algorithms and improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. In this way, the frequency response characteristics of electro-optical or optical-electrical systems can be accurately measured. Moreover, it does not rely on optical modulation and demodulation technology, thus saving additional instruments and resources, which is beneficial for chip miniaturization, high integration and low power consumption.

[0048] Figure 2 This is a schematic diagram of a measurement device for the frequency response characteristics of an optoelectronic device, provided as an embodiment of this application. Figure 2As shown, the measuring device A200 includes: a first data acquisition unit A201, a second data acquisition unit A203, and a processor A205. The first data acquisition unit A201 is used to obtain the input waveform A222 of the input signal at the input node A212 of the device under test A210. The second data acquisition unit A203 is used to obtain the output waveform A224 of the output signal at the output node A214 of the device under test A210. The input signal includes multiple frequency components, and the device under test A210 has electro-optical conversion or photoelectric conversion functions. One of the input signal and the output signal is a first electrical signal and the other is a first optical signal. One of the input waveform A222 and the output waveform A224 is the waveform corresponding to the first electrical signal and the other is the waveform corresponding to the first optical signal. Processor A205 is configured to perform waveform adjustment operation on the waveform corresponding to the first optical signal according to the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal, and to perform differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal, wherein the waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the device under test A210.

[0049] Figure 2The first embodiment of the measuring device A200 for the frequency response characteristics of optoelectronic devices shown can be used in applications such as high-speed optical communication systems and data center optical interconnects. It measures the frequency response characteristics of electro-optical or optical-electrical systems, and acquires and evaluates the frequency response characteristics of optoelectronic devices within the system, providing a reference for configuration adjustments and equipment calibration. The measuring device A200 imposes requirements on the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function matching the frequency response characteristics of the device under test (A210) using digital processing algorithms. Through a time-domain waveform acquisition device, the time-domain waveforms of the first electrical signal and the first optical signal are actually acquired. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into a second electrical signal. This provides a basis for waveform analysis and comparison and simplifies the process. The design of the digital processing algorithm improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing through the A205 processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. Thus, accurate measurement of the frequency response characteristics of electro-optical or optical-electrical systems is achieved, and the absence of reliance on optical modulation and demodulation technology saves additional instruments and resources, which is beneficial for chip miniaturization, high integration, and low power consumption.

[0050] Figure 3 This is a schematic flowchart illustrating a fault detection method for optoelectronic devices provided in an embodiment of this application. Figure 3 As shown, the fault detection method includes the following steps.

[0051] Step S301: The first acquisition device obtains the input waveform of the input signal at the input node of the device under test, and the second acquisition device obtains the output waveform of the output signal at the output node of the device under test. The input signal includes multiple frequency components, and the device under test has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal and the other is the waveform corresponding to the first optical signal.

[0052] Step S303: The processor performs waveform adjustment on the waveform corresponding to the first optical signal according to the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal. The processor also performs differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal. The waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to measure and calibrate the frequency response characteristics of the device under test in real time to detect faults in the device under test.

[0053] Figure 3 The fault detection method for optoelectronic devices shown can be used in applications such as high-speed optical communication systems and data center optical interconnects. It is used to measure the frequency response characteristics of electro-optical or optical-electrical systems, and to obtain and evaluate the frequency response characteristics of optoelectronic devices in the system, thereby providing a reference for fault detection. Figure 3 The fault detection method for optoelectronic devices described herein imposes requirements on the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function that matches the frequency response characteristics of the device under test using digital processing algorithms. A time-domain waveform acquisition device is used to acquire the time-domain waveforms of both the first electrical signal and the first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into the second electrical signal. This provides the basis for waveform analysis and comparison. It also simplifies the design of digital processing algorithms and improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. In this way, the frequency response characteristics of electro-optical or optical-electrical systems can be accurately measured. Moreover, it does not rely on optical modulation and demodulation technology, thus saving additional instruments and resources, which is beneficial for chip miniaturization, high integration and low power consumption.

[0054] Figure 3The fault detection method for optoelectronic devices shown can be used not only for calibrating individual devices and measuring frequency response characteristics in real time, but also for large-scale parallel calibration. A single signal generator generates a compliant input signal, and a single time-domain waveform acquisition unit acquires the time-domain waveform of the input signal. Then, the time-domain waveforms of the corresponding output signals of multiple devices requiring calibration are acquired separately at their respective output sides using the time-domain waveform acquisition unit. The waveforms of the multiple devices are then compared and analyzed to obtain the frequency response characteristics of each device. This allows for parallel fault detection for multiple devices requiring calibration. Therefore, Figure 3 The fault detection method for optoelectronic devices shown is suitable for scenarios such as data centers where a large number of optoelectronic devices (including optical emitting devices and optical detection devices) are used, as well as for applications such as lidar that utilize a large number of lasers to construct laser emitting arrays. As mentioned above, during system operation, the performance of optoelectronic devices in the system may deviate from the design target due to factors such as device aging and wear, thus affecting the compensation effect and signal transmission performance. Therefore, it is necessary to obtain and evaluate the frequency response characteristics of the optoelectronic devices in the system. For example, the calculated electro-optical or optical-electrical system frequency response characteristics of the device under test can be compared with the system frequency response characteristics of a reference target. This allows for efficient identification of faulty devices and can be used for device fault detection and replacement. Furthermore, Figure 3 The fault detection method for optoelectronic devices shown can also be used in automated test equipment (ATE) for chips and automated testing before chip manufacturing. With the development of silicon-based optoelectronic integration technology, various optoelectronic devices may be integrated on chips. Therefore, it is necessary to test the performance of these optoelectronic devices. However, it is difficult to add optical substrates and optical modulation devices inside the chip for testing; therefore, testing must rely on modules within the chip itself. Figure 3 The fault detection method for optoelectronic devices shown can be used in various applications requiring the measurement of frequency response characteristics of electro-optical or optical-electrical systems. It eliminates the need for complex optical base hardware and optical modulation devices, and can calculate the frequency response characteristics of the device under test between input and output nodes. This saves the overhead of de-embedding algorithms, simplifying the overall measurement scheme design. Hardware requirements include only a signal generator to generate compliant input signals, a time-domain waveform acquisition unit capable of acquiring time-domain waveforms of both electrical and optical signals, and a processor to execute digital processing algorithms. This facilitates the design and manufacturing of large-scale integrated optoelectronic chips that incorporate optoelectronic devices.

[0055] See Figure 1 , Figure 2 and Figure 3In one possible implementation, the input signal is a data signal comprising the plurality of frequency components, and the amplitude distribution of each of the plurality of frequency components conforms to a preset pattern. Thus, the input signal is optimized, defining it as a data signal comprising the plurality of frequency components, and the amplitude distribution of each of the plurality of frequency components conforms to a preset pattern. The amplitude distribution of each of the plurality of frequency components in the input signal references a certain rule, for example, having consistent or inconsistent amplitudes. The input signal is input to the device under test (DUT) from the input node and then output to the outside of the DUT via the output node. Therefore, during the transmission of the input signal within the DUT, the plurality of frequency components may be adjusted, for example, amplified or attenuated. This means that the plurality of frequency components included in the input signal undergo an overall change in the frequency domain due to the transmission process within the DUT. Therefore, by inputting a data signal containing multiple frequency components into the device under test (DUT) and obtaining a corresponding output signal, this conversion process from input to output signal is influenced by the internal components and circuits of the DUT, which adjust the frequency components of the signal. For example, it is affected by the internal signal compensation functions of the DUT (e.g., pre-emphasis, de-emphasis, equalization, etc.). Thus, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce the information about the frequency response characteristics of the DUT carried by these changes. This allows for the derivation of a higher-order system transfer function (SFC) that matches the frequency response characteristics of the DUT, which can be used for configuration adjustments, equipment calibration, and other applications. Here, the digital processing algorithm used to derive the higher-order SFC that matches the frequency response characteristics of the DUT can employ any suitable algorithm model and principle. For example, an error convergence algorithm can be used to select the system transfer function with the smallest error from multiple SFCs. Alternatively, the error function can be defined as a function of the number of poles and zeros, and then iterative and adaptive algorithms can be used to filter out the system transfer function with the smallest error. Another example is the use of digital fitting algorithms to approximate and calculate a suitable system transfer model. Furthermore, considering that the input signal is limited to a data signal with multiple frequency components, this means that, from a time-domain analysis perspective, the time-domain waveform of the input signal exhibits irregular amplitude variations, and the distribution of peaks and troughs in the time-domain waveform may also be irregular. Therefore, the digital processing algorithm used to derive the higher-order system transfer function that matches the frequency response characteristics of the device under test does not compare the rise and fall times in the time domain. Instead, it determines the overall change in rise and fall times in the frequency domain. That is, by utilizing the key design requirement that the input signal is a data signal including the aforementioned multiple frequency components, it derives the higher-order system transfer function with the minimum error, thereby characterizing the overall change in rise and fall times in the frequency domain.In addition, digital processing algorithms used to derive high-order system transfer functions that match the frequency response characteristics of the device under test can take into account not only the changes in rise time and fall time, but also other factors in the changes between the input and output signals, thereby improving the calculation accuracy, such as amplitude changes and delay alignment.

[0056] In some embodiments, the preset mode indicates that the amplitudes of at least two of the plurality of frequency components are inconsistent. Thus, the input signal is optimized, defining it as a data signal comprising the plurality of frequency components, and the amplitude distribution of each of the plurality of frequency components conforms to the preset mode. The amplitude distribution of the plurality of frequency components in the input signal follows a certain pattern, such as having consistent or inconsistent amplitudes. Here, the preset mode indicates that the amplitudes of at least two of the plurality of frequency components are inconsistent, meaning that different amplitude designs may be used at different frequency points. This implies that the pattern of amplitude distribution of the plurality of frequency components is flexible, and ultimately, the design objective of deriving a higher-order system transfer function that matches the frequency response characteristics of the device under test can still be achieved. Therefore, unlike the frequency scanning method that uses an input signal with a single frequency component and changes that single frequency component, this frequency scanning method generally uses a bit error rate tester (BERT) and a pseudo-random binary sequence (PRBS) to compare data and calculate the bit error rate. Therefore, signals of the same amplitude are generally used at different frequency points. In contrast, the method and apparatus for measuring the frequency response characteristics of optoelectronic devices provided in the specific embodiments and implementation methods of this application can use signals of different amplitudes at different frequency points, which provides better flexibility in the composition of the input signal and is conducive to its widespread application.

[0057] In some embodiments, the preset mode indicates that the amplitudes of the plurality of frequency components are the same. This optimizes the design of the input signal, defining it as a data signal comprising the plurality of frequency components, and ensuring that the amplitude distribution of each frequency component conforms to the preset mode. The amplitude distribution of the plurality of frequency components in the input signal follows a certain pattern, such as having consistent or inconsistent amplitudes. Here, the preset mode indicates that the amplitudes of the plurality of frequency components are the same, meaning that the same amplitude design can be used at different frequency points, and ultimately, the design objective of deriving a high-order system transfer function that matches the frequency response characteristics of the device under test can still be achieved. This provides greater flexibility in the configuration of the input signal, facilitating wider application.

[0058] In some embodiments, the input signal is a wave packet signal with a wide frequency range. As mentioned above, the input signal has been optimized to be a data signal including the multiple frequency components, and the amplitude distribution of each of the multiple frequency components conforms to a preset pattern. Here, the input signal is further defined as a wave packet signal with a wide frequency range. Therefore, requirements are placed on the composition of the input signal, defining it as a wave packet signal with a wide frequency range including multiple frequency components, and the amplitude distribution of each of the multiple frequency components in the input signal refers to a certain rule (preset pattern). The wave packet signal corresponding to the input signal is input to the device under test (DUT) from the input node and then output to the outside of the DUT via the output node. Therefore, during the transmission of the wave packet signal inside the DUT, the frequency components of the wave packet signal may be adjusted, for example, amplified or attenuated. This means that the multiple frequency components included in the wave packet signal undergo an overall change in the frequency domain due to the transmission process inside the DUT. Therefore, by inputting a wide-frequency-range wave packet signal containing multiple frequency components into the device under test (DUT), and obtaining the corresponding output signal, this conversion process from input to output signal is influenced by the internal components and circuits of the DUT, which adjust the frequency components of the signal. For example, it is affected by the internal signal compensation functions of the DUT (such as pre-emphasis, de-emphasis, and equalization). Thus, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce the information about the frequency response characteristics of the DUT carried by these changes. Furthermore, a higher-order system transfer function matching the frequency response characteristics of the DUT can be derived. Specifically, by utilizing the key design requirement that the input signal is a wide-frequency-range wave packet signal with multiple frequency components, instead of comparing rise and fall times in the time domain, the overall changes in rise and fall times are determined in the frequency domain. This allows for the derivation of a higher-order system transfer function with minimal error, thus characterizing the overall changes in rise and fall times in the frequency domain. Therefore, compared to inputting an input signal with only a single frequency component and changing that single frequency component to perform frequency scanning, inputting a wave packet signal with a wide frequency range and multiple frequency components as the input signal to the device under test results means that the amplitude distribution of the multiple frequency components in the wave packet signal follows a certain pattern, and the multiple frequency components are pre-designed and do not change with the scanning mode.Furthermore, the method and apparatus for measuring the frequency response characteristics of optoelectronic devices provided in the specific embodiments and implementation methods of this application can be used in various application scenarios that require measuring the frequency response characteristics of electro-optical or optical-electrical systems. It does not require complex optical base hardware or optical modulation devices, and can calculate the frequency response characteristics of the device under test between the input and output nodes. This saves the overhead of de-embedding algorithms, and the overall measurement scheme design is simplified. Hardware requirements only include a signal generator to generate a compliant input signal, a time-domain waveform acquisition device capable of acquiring the time-domain waveforms of electrical and optical signals, and a processor capable of executing digital processing algorithms. This facilitates the design and manufacturing of large-scale integrated optoelectronic chips that integrate optoelectronic devices.

[0059] In some embodiments, the data signal includes a test code pattern, which is an identical codeword with a preset segment length. As mentioned above, the input signal has been optimized, defining it as a data signal including the multiple frequency components, and the amplitude distribution of each of the multiple frequency components conforms to a preset pattern. Here, requirements are imposed on the code pattern of the input signal, further defining that the data signal includes a test code pattern, and that the test code pattern is an identical codeword with a preset segment length. Utilizing the changes between the input and output signals, information about the frequency response characteristics of the device under test carried by the changes between the input and output signals can be calculated using digital processing algorithms. This allows for the derivation of a higher-order system transfer function matching the frequency response characteristics of the device under test, used for configuration adjustment, device calibration, and other purposes. Here, by utilizing the requirements for the input signal code pattern, a reference for adjusting the amplitude is provided by inserting a test code pattern with an identical codeword with a preset segment length into the data signal, for example, inserting consecutive identical digits (CID) segments with a preset segment length, such as constant 0 segments or constant 1 segments with a preset segment length. The advantage of this design is that, considering the irregular amplitude variation of the input signal's time-domain waveform, and the potentially irregular distribution of peaks and troughs, the design requirement of the input signal's code pattern—that is, a test code pattern with identical codewords of a preset segment length (e.g., consecutive identical digital segments of a preset segment length)—can be used. For example, utilizing the mechanism of consecutive identical digital segments within a pseudo-random binary sequence, multiple consecutive identical codewords may appear, such as nine consecutive segments of constant 0 or 23 consecutive segments of constant 1. This can serve as a benchmark for waveform comparison and analysis. It should be noted that the input signal is a data signal with multiple frequency components. Therefore, the test code pattern with identical codewords of a preset segment length included in the input signal corresponds to the low-frequency portion of the multiple frequency components in the input signal. Therefore, using the test code pattern with identical codewords of a preset segment length as a benchmark is equivalent to using the low-frequency portion of the input signal to derive the equalization value of the device under test (DUT), such as the gain ratio or compression ratio. This allows for amplitude adjustment to improve the accuracy of the final calculation of the DUT's frequency response characteristics.

[0060] In some examples, the test code pattern includes a pseudo-random binary sequence. As mentioned above, requirements are imposed on the composition of the input signal, specifying that the input signal is a data signal including the multiple frequency components, and that the amplitude distribution of each of the multiple frequency components conforms to a preset pattern. Furthermore, requirements are imposed on the code pattern of the input signal, specifying that the data signal includes a test code pattern, and that the test code pattern is a set of identical codewords with a preset segment length. Here, the test code pattern is specified to include a pseudo-random binary sequence (PRBS). Thus, the input signal can be generated using a pseudo-random sequence such as a pseudo-random binary sequence, or other random generation algorithms, to generate a binary code sequence of 0s and 1s with random characteristics. This means that the digital logic of the input signal has certain random characteristics, as long as the requirements for the composition and code pattern of the input signal are met, i.e., a data signal including the multiple frequency components and a test code pattern with identical codewords of a preset segment length. Thus, the design requirements of the input signal's code pattern are essentially test code patterns with identical codewords of a preset segment length (e.g., consecutive identical digital segments of a preset segment length). For example, utilizing the mechanism of consecutive identical digital segments within a pseudo-random binary sequence, multiple consecutive identical codewords may appear, such as nine consecutive segments of constant 0 or 23 consecutive segments of constant 1. This can serve as a benchmark for waveform comparison and analysis. It should be noted that the input signal is a data signal with multiple frequency components. Therefore, the test code pattern with identical codewords of a preset segment length included in the input signal corresponds to the low-frequency portion of the multiple frequency components in the input signal. Therefore, using the test code pattern with identical codewords of a preset segment length as a benchmark is equivalent to using the low-frequency portion of the input signal to derive the equalization value of the device under test (DUT), such as deriving the gain ratio or compression ratio. This allows for amplitude adjustment to improve the accuracy of the final calculation of the frequency response characteristics of the DUT.

[0061] In some examples, the identical codewords with a preset segment length correspond to the low-frequency portion of the multiple frequency components and serve as a reference for the waveform adjustment operation. Thus, utilizing the design requirements of the input signal's code pattern—that is, the test code pattern with identical codewords of a preset segment length (e.g., consecutive identical digital segments of a preset segment length)—for example, leveraging the mechanism of consecutive identical digital segments within a pseudo-random binary sequence, multiple consecutive identical codewords may appear, such as nine consecutive segments of constant 0 or 23 consecutive segments of constant 1. This can serve as a reference for waveform comparison and analysis. It should be noted that the input signal is a data signal with multiple frequency components. Therefore, the test code pattern with identical codewords of a preset segment length included in the input signal corresponds to the low-frequency portion of the multiple frequency components in the input signal. Therefore, using the test code pattern with identical codewords of a preset segment length as a reference is equivalent to using the low-frequency portion of the input signal to derive the equalization value of the device under test (DUT), such as deriving the gain ratio or compression ratio. This allows for amplitude adjustment to improve the accuracy of the final calculation of the frequency response characteristics of the DUT.

[0062] In some examples, the low-frequency components of the multiple frequency components corresponding to the same codeword with a preset segment length are used to calculate the gain ratio or compression ratio of the device under test (DUT). As mentioned above, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce information about the frequency response characteristics of the DUT carried by these changes, and thus, a higher-order system transfer function matching the frequency response characteristics of the DUT can be derived. Specifically, by utilizing the key design requirement that the input signal is a data signal with multiple frequency components, instead of comparing rise and fall times in the time domain, the overall change in rise and fall times is determined in the frequency domain, deriving the higher-order system transfer function with the smallest error, thereby characterizing the overall change in rise and fall times in the frequency domain. Therefore, compared to inputting a signal with only a single frequency component and changing that single frequency component for frequency scanning, inputting a data signal with multiple frequency components as the input signal to the DUT ensures that the amplitude distribution of each frequency component conforms to a preset pattern and therefore follows certain rules. Furthermore, the multiple frequency components are pre-designed and do not change with the scanning mode. Thus, by utilizing the requirements of the input signal's code pattern, a benchmark for adjusting the amplitude is provided by inserting test code patterns with the same codewords of a preset segment length into the data signal. This could be done by inserting consecutive identical digital segments of a preset segment length, such as constant 0 segments or constant 1 segments. The advantage of this design is that, considering the irregular amplitude variations in the input signal's time-domain waveform, and the potentially irregular distribution of peaks and troughs, the design requirements of the input signal's code pattern—namely, test code patterns with the same codewords of a preset segment length (e.g., consecutive identical digital segments of a preset segment length)—can be used. For example, by utilizing the mechanism of consecutive identical digital segments within a pseudo-random binary sequence, multiple consecutive identical codewords may appear, such as nine consecutive constant 0 segments or 23 consecutive constant 1 segments. This can serve as a benchmark for waveform comparison and analysis. It should be noted that the input signal is a data signal with multiple frequency components. Therefore, the test code pattern with the same codeword of a preset segment length included in the input signal corresponds to the low-frequency part of the multiple frequency components in the input signal. That is, the same codeword with the preset segment length corresponds to the low-frequency part of the multiple frequency components. Therefore, using the test code pattern with the same codeword of the preset segment length as a reference is equivalent to using the signal of the low-frequency part of the input signal to derive the equalization value of the device under test, such as deriving the gain ratio or compression ratio, which is used to calculate the gain ratio or compression ratio of the device under test. In this way, the accuracy of the final calculation of the frequency response characteristics of the device under test can be improved by adjusting the amplitude.

[0063] In one possible implementation, the measurement method further includes: inputting the first electrical signal into multiple system transfer functions to obtain multiple reference output electrical signals corresponding one-to-one with the multiple system transfer functions; then, performing differential processing on the first electrical signal and the multiple reference output electrical signals to obtain multiple reference differential electrical signals corresponding one-to-one with the multiple reference output electrical signals; by comparing the differential mode components of the first differential electrical signal and the differential mode components of each of the multiple reference differential electrical signals, iterating using an error function, and selecting the system transfer function with the smallest error from the multiple system transfer functions; the selected system transfer function with the smallest error is used to determine the frequency response characteristics and system bandwidth of the device under test. Thus, the digital processing algorithm used to derive the high-order system transfer function matching the frequency response characteristics of the device under test can employ any suitable algorithm model and principle. For example, an error convergence algorithm can be used to select the system transfer function with the smallest error from multiple system transfer functions; another example is that the error function can be set as a function of the number of poles and zeros, and then iterative and adaptive algorithms can be used to filter out the system transfer function with the smallest error; yet another example is that a digital fitting algorithm can be used to approximate and calculate a suitable system transfer model. Here, multiple reference output electrical signals and corresponding reference differential electrical signals are obtained using multiple system transfer functions. Then, by comparing the differential-mode components of the first differential electrical signal and the differential-mode components of each of the multiple reference differential electrical signals, an error function is used for iteration to select the system transfer function with the smallest error from the multiple system transfer functions. In this way, a high-order system transfer function matching the frequency response characteristics of the device under test is efficiently determined through iterative and adaptive algorithms. The error function can be set as a function of the number of poles and zeros.

[0064] In some embodiments, the error function is a function of the number of poles and zeros of the system transfer function, and the model characteristics of each of the multiple system transfer functions include delay alignment time, rise time, fall time, and amplitude variation. Considering that the input signal includes multiple frequency components, this means that, from a time-domain analysis perspective, the time-domain waveform of the input signal has irregular amplitude variations, and the distribution of peaks and troughs in the time-domain waveform may also be irregular. Therefore, the digital processing algorithm used to derive a higher-order system transfer function that matches the frequency response characteristics of the device under test does not compare rise and fall times in the time domain, but rather determines the overall variation of rise and fall times in the frequency domain. That is, by utilizing the key design requirement that the input signal includes multiple frequency components, the algorithm derives a higher-order system transfer function with the minimum error, thereby characterizing the overall rise and fall time variations in the frequency domain. Furthermore, the digital processing algorithm used to derive a higher-order system transfer function that matches the frequency response characteristics of the device under test can consider other factors in the variation between the input and output signals, in addition to the variations in rise and fall times, thereby improving calculation accuracy, such as amplitude variation and delay alignment.

[0065] In one possible implementation, the processor performs a waveform adjustment operation on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal to convert the first optical signal into the second electrical signal. This includes: determining the average value of the waveform corresponding to the first optical signal; then subtracting the average value of the waveform corresponding to the first optical signal from the average value of the waveform corresponding to the first optical signal; and then performing a waveform alignment operation in the waveform adjustment operation based on the waveform corresponding to the first electrical signal. Thus, using an input signal including multiple frequency components, and through a time-domain waveform acquisition device, the time-domain waveforms of the input node and the output node are actually acquired. Here, because the device under test (DUT) has electro-optical conversion or photoelectric conversion functions, the two acquired time-domain waveforms are the waveform corresponding to the first electrical signal and the waveform corresponding to the first optical signal, depending on the internal structure of the DUT, which may be from the input of the first electrical signal to the output of the first optical signal or from the input of the first optical signal to the output of the first electrical signal. This is because the time-domain waveform of the first electrical signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal. Therefore, the processor adjusts the waveform of the first optical signal (when the input signal is the first electrical signal, the output signal is the first optical signal; or vice versa) based on the waveform corresponding to the first electrical signal (which may correspond to either the input signal or the output signal). This adjustment is performed to convert the first optical signal into a second electrical signal. Then, the processor performs differential processing on the first and second electrical signals to obtain a first differential electrical signal. Thus, based on digital processing algorithms, the frequency domain transfer function is analyzed, and the higher-order system transfer function corresponding to the frequency response characteristics of the device under test is derived. Furthermore, the waveform adjustment operation reduces the common-mode component of the first differential electrical signal, and the differential-mode component is used to calculate the frequency response characteristics of the device under test. Here, through the waveform adjustment operation, the first optical signal is converted into a second electrical signal based on the time-domain waveform of the acquired first optical signal. The time-domain waveform of the second electrical signal can then be used for waveform analysis and comparison with the time-domain waveform of the first electrical signal. By utilizing the processor's internal algorithms, a first optical signal can be converted into a second electrical signal through waveform adjustment operations. For example, waveform alignment can be achieved by removing the DC component. The frequency response characteristics of the device under test can then be calculated by using the changes between the first and second electrical signals. Therefore, it can be applied to scenarios that require a large number of optoelectronic devices, such as optical interconnects in data centers.For example, the amplitude range of the waveform corresponding to the first electrical signal is from -100 millivolts (mV) to 100 mV, and the amplitude range of the waveform corresponding to the first optical signal is from 300 mW to 500 mW. To align the waveforms, the average values ​​of the waveforms can be aligned, i.e., the corresponding average values ​​are subtracted from each of the two original signals. Specifically, the average value of the waveform corresponding to the first optical signal is determined, and then the average value of the waveform corresponding to the first optical signal is subtracted from the average value of the waveform corresponding to the first optical signal. Then, the waveform alignment operation in the waveform adjustment operation is performed according to the waveform corresponding to the first electrical signal. In this way, through waveform adjustment and waveform alignment, and based on the analysis of the frequency domain transfer function using digital processing algorithms, the higher-order system transfer function corresponding to the frequency response characteristics of the device under test is derived. Thus, the frequency response characteristics of electro-optical or optical-electrical systems can be accurately measured, and since it does not rely on optical modulation and demodulation technology, it saves additional instruments and resources, which is beneficial for chip miniaturization, high integration, and low power consumption.

[0066] In some embodiments, the input signal is a wave packet signal with a wide frequency range including the plurality of frequency components. The wave packet form of the wave packet signal is determined according to the data format of the pseudo-random code generated by the data generator. The consecutive identical digital segments in the data format serve as the reference for amplitude adjustment operations in the waveform adjustment operation. These consecutive identical digital segments are either constant 0 segments or constant 1 segments. Optimizing the input signal is crucial for calculating the frequency response characteristics of the device under test. Specifically, requirements are set for the composition of the input signal, limiting it to a wave packet signal with a wide frequency range including multiple frequency components. Furthermore, the wave packet form of the wave packet signal is limited to a data format determined according to the pseudo-random code generated by the data generator. The consecutive identical digital segments in the data format serve as the reference for amplitude adjustment operations in the waveform adjustment operation. Thus, the amplitude distribution of each of the multiple frequency components in the input signal follows a certain pattern, such as having consistent or inconsistent amplitudes, ultimately forming a wave packet signal with a wide frequency range. The wave packet signal corresponding to the input signal is input to the device under test (DUT) from the input node and then output to the outside of the DUT via the output node. Therefore, during the transmission of the wave packet signal inside the DUT, the frequency components of the wave packet signal may be adjusted, such as amplified or attenuated. This means that the multiple frequency components included in the wave packet signal undergo an overall change in the frequency domain due to the transmission process inside the DUT. Therefore, by inputting a wide-frequency-range wave packet signal including multiple frequency components as the input signal to the DUT, and obtaining the corresponding output signal, this conversion process from input signal to output signal is affected by the adjustment of the frequency components in the signal by the internal devices and circuits of the DUT, such as the influence of the internal signal compensation functions of the DUT (e.g., pre-emphasis, de-emphasis, equalization, etc.). Thus, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce the information about the frequency response characteristics of the DUT carried by the changes between the input and output signals. Furthermore, a higher-order system transfer function matching the frequency response characteristics of the DUT can be derived for configuration adjustment, equipment calibration, and other purposes. Here, the digital processing algorithm used to derive the high-order system transfer function that matches the frequency response characteristics of the device under test can employ any suitable algorithm model and principle. For example, an error convergence algorithm can be used to select the system transfer function with the smallest error from a variety of system transfer functions. Alternatively, the error function can be set as a function of the number of poles and zeros, and then iterative and adaptive algorithms can be used to screen out the system transfer function with the smallest error. Another example is that a suitable system transfer model can be approximately calculated using a digital fitting algorithm.Furthermore, considering that the input signal is limited to a wave packet signal with a wide frequency range and multiple frequency components, this means that, from a time-domain analysis perspective, the time-domain waveform of the input signal exhibits irregular amplitude variations, and the distribution of peaks and troughs in the time-domain waveform may also be irregular. Therefore, the digital processing algorithm used to derive the higher-order system transfer function matching the frequency response characteristics of the device under test (DUT) does not compare rise and fall times in the time domain, but rather determines the overall variation of rise and fall times in the frequency domain. That is, leveraging the key design requirement that the input signal is a wave packet signal with a wide frequency range and multiple frequency components, the algorithm derives the higher-order system transfer function with the minimum error, thereby characterizing the overall rise and fall time variations in the frequency domain. In addition, the digital processing algorithm used to derive the higher-order system transfer function matching the frequency response characteristics of the DUT can consider other factors in the variation between the input and output signals besides the changes in rise and fall times, thereby improving calculation accuracy, such as amplitude variations and delay alignment. Furthermore, the consecutive identical digital segments in the data format serve as the benchmark for amplitude adjustment in the waveform adjustment operation. These consecutive identical digital segments are either constant 0 segments or constant 1 segments. The advantage of this design is that, considering the irregular amplitude variations in the time-domain waveform of the input signal, and the potentially irregular distribution of peaks and troughs, the consecutive identical digital segments in the data format, such as the mechanism of consecutive identical digital segments inherent in a pseudo-random binary sequence, can generate multiple consecutive identical codewords, such as nine consecutive constant 0 segments or 23 consecutive constant 1 segments. This can serve as a benchmark for waveform comparison and analysis. It should be noted that the input signal is a wave packet signal with a wide frequency range and multiple frequency components. Therefore, the consecutive identical digital segments in the data format correspond to the low-frequency part of the multiple frequency components in the wave packet signal. Thus, using the consecutive identical digital segments in the data format as the reference for the amplitude adjustment operation in the waveform adjustment operation is equivalent to using the signal of the low-frequency part of the wave packet signal to derive the equalization value of the device under test, such as deriving the gain ratio or compression ratio. In this way, the accuracy of the final calculation of the frequency response characteristics of the device under test can be improved by amplitude adjustment.

[0067] In some examples, the differential-mode component of the first differential electrical signal is used to determine the system transfer function with the minimum error. This minimum error system transfer function characterizes the overall rise and fall time variations in the frequency domain of the wave packet signal after passing through the device under test (DUT). As mentioned above, the input signal is limited to a wave packet signal with a wide frequency range and multiple frequency components. This means that, from a time-domain analysis perspective, the time-domain waveform of the input signal exhibits irregular amplitude variations, and the distribution of peaks and troughs in the time-domain waveform may also be irregular. Therefore, the digital processing algorithm used to derive a higher-order system transfer function that matches the frequency response characteristics of the DUT does not compare rise and fall times in the time domain. Instead, it determines the overall changes in rise and fall times in the frequency domain. That is, by utilizing the key design requirement that the input signal is a wave packet signal with a wide frequency range and multiple frequency components, the algorithm derives the higher-order system transfer function with the minimum error, thereby characterizing the overall rise and fall time variations in the frequency domain. In addition, digital processing algorithms used to derive high-order system transfer functions that match the frequency response characteristics of the device under test can take into account not only the changes in rise time and fall time, but also other factors in the changes between the input and output signals, thereby improving the calculation accuracy, such as amplitude changes and delay alignment.

[0068] In one possible implementation, the device under test (DUT) has an electro-optical conversion function. The input signal is the first electrical signal, the output signal is the first optical signal, the input waveform is the waveform corresponding to the first electrical signal, and the output waveform is the waveform corresponding to the first optical signal. The DUT includes a light emitting device. Thus, when the input signal is the first electrical signal and the output signal is the first optical signal, this corresponds to the frequency response characteristics of the electro-optical system, that is, from electrical signal input to optical signal output. For example, the DUT internally includes a driver and a laser for converting the input electrical signal into the output optical signal. This means that the input waveform of the input signal is the waveform corresponding to the first electrical signal, and the output waveform of the output signal is the waveform corresponding to the first optical signal. It should be understood that regardless of whether the DUT has an electro-optical conversion function or a photoelectric conversion function, the two time-domain waveforms finally acquired are the waveform corresponding to the first electrical signal and the waveform corresponding to the first optical signal, depending only on the internal structure of the DUT, which may be from the first electrical signal input to the first optical signal output or from the first optical signal input to the first electrical signal output. Subsequent time-domain waveform analysis and comparison, as well as digital processing algorithms, do not require strict distinction between electrical and optical signal inputs. This simplifies the design of digital processing algorithms and improves computational efficiency.

[0069] In one possible implementation, the device under test (DUT) has a photoelectric conversion function. The input signal is the first optical signal, the output signal is the first electrical signal, the input waveform is the waveform corresponding to the first optical signal, and the output waveform is the waveform corresponding to the first electrical signal. The DUT includes a photodetector. Thus, when the input signal is the first optical signal and the output signal is the first electrical signal, this corresponds to the frequency response characteristics of the photoelectric conversion system, that is, from optical signal input to electrical signal output. For example, the DUT internally includes a photodiode and a transimpedance amplifier to convert the input optical signal into the output electrical signal. This means that the input waveform of the input signal is the waveform corresponding to the first optical signal, and the output waveform of the output signal is the waveform corresponding to the first electrical signal. It should be understood that regardless of whether the DUT has an electro-optical conversion function or a photoelectric conversion function, the two time-domain waveforms finally acquired are the waveform corresponding to the first electrical signal and the waveform corresponding to the first optical signal, depending only on the internal structure of the DUT, which may be from the first electrical signal input to the first optical signal output or from the first optical signal input to the first electrical signal output. Subsequent time-domain waveform analysis and comparison, as well as digital processing algorithms, do not require strict distinction between electrical and optical signal inputs. This simplifies the design of digital processing algorithms and improves computational efficiency.

[0070] In one possible implementation, the first acquisition device is different from the second acquisition device, or the first and second acquisition devices are the same acquisition device with the capability to acquire both electrical signal time-domain waveforms and optical signal time-domain waveforms. Thus, the first acquisition device with time-domain waveform acquisition capability obtains the input waveform of the input signal at the input node of the device under test, i.e., actually acquires the time-domain effect diagram of the input signal. Similarly, the second acquisition device with time-domain waveform acquisition capability obtains the output waveform of the output signal at the output node of the device under test, i.e., actually acquires the time-domain effect diagram of the output signal. It should be understood that in some embodiments, the first and second acquisition devices are independent of each other, i.e., two different acquisition devices; for example, the first acquisition device is different from the second acquisition device. In some embodiments, the first and second acquisition devices can be the same acquisition device and achieve the acquisition of electrical signal time-domain waveforms and optical signal time-domain waveforms through different ports; for example, the first and second acquisition devices are the same acquisition device with the capability to acquire both electrical signal time-domain waveforms and optical signal time-domain waveforms, such as an oscilloscope. Thus, the input waveform of the input signal actually acquired by the time-domain waveform acquisition device, and the output waveform of the output signal actually acquired by the time-domain waveform acquisition device, serve as reference benchmarks for subsequent processes. Through waveform comparison and analysis in subsequent processes, the frequency response characteristics of the device under test can be calculated.

[0071] In one possible implementation, the device under test (DUT) is any one of a plurality of calibration devices, each with its own input signal from the same signal generator. The input and output signals of each DUT are used to determine its own frequency response characteristics to achieve parallel calibration of the plurality of devices. The method and apparatus for measuring the frequency response characteristics of optoelectronic devices provided in this application impose requirements on the composition of the input signal, limiting it to include multiple frequency components. Optimized design of the input signal allows for the derivation of a high-order system transfer function matching the frequency response characteristics of the DUT using digital processing algorithms. A time-domain waveform acquisition device is used to acquire the time-domain waveforms of a first electrical signal and a first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, waveform adjustment is performed on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, thereby converting the first optical signal into a second electrical signal. This provides waveform analysis. The comparison simplifies the design of digital processing algorithms and improves computational efficiency. The differential-mode component of the first differential electrical signal obtained through differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, waveform adjustment can be used to minimize the common-mode component of the first differential electrical signal obtained through differential processing, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. Thus, accurate measurement of the frequency response characteristics of electro-optical or optical-electrical systems is achieved without relying on optical modulation and demodulation technology, saving additional instruments and resources and facilitating chip miniaturization, high integration, and low power consumption. Furthermore, the measurement method and apparatus for the frequency response characteristics of optoelectronic devices provided in this application can not only be used for calibration of individual devices and real-time measurement of frequency response characteristics, but also for large-scale parallel calibration. A single signal generator generates a compliant input signal, and a single time-domain waveform acquisition device acquires the time-domain waveform of the input signal. Then, the time-domain waveforms of the corresponding output signals are acquired separately at the output sides of multiple devices requiring calibration using the same time-domain waveform acquisition device. The waveforms of the multiple devices requiring calibration are then compared and analyzed to obtain the frequency response characteristics of each device. This allows for parallel fault detection for multiple devices requiring calibration. Therefore, the method and apparatus for measuring the frequency response characteristics of optoelectronic devices provided in this application are suitable for scenarios such as data centers where a large number of optoelectronic devices (including optical emitting devices and optical detection devices) are used, as well as for applications such as lidar that utilize a large number of lasers to construct laser emitting arrays.As mentioned above, during system operation, the performance of optoelectronic devices in the system may deviate from the design target due to factors such as device aging and wear, thereby affecting the compensation effect and signal transmission performance. Therefore, it is necessary to obtain and evaluate the frequency response characteristics of the optoelectronic devices in the system. For example, the calculated electro-optical or optical-electrical system frequency response characteristics of the device under test can be compared with the system frequency response characteristics of a reference target. This can efficiently identify faulty devices and can be used for device fault detection and replacement. Furthermore, the measurement method and apparatus for the frequency response characteristics of optoelectronic devices provided in this application can also be used in automated chip testing equipment and automated chip manufacturing testing. With the development of silicon-based optoelectronic integration technology, various optoelectronic devices may be integrated on chips. Therefore, it is necessary to test the performance of these optoelectronic devices. However, it is difficult to add optical substrates and optical modulation devices inside the chip for testing; therefore, testing must rely on modules within the chip. The method and apparatus for measuring the frequency response characteristics of optoelectronic devices provided in this application can be used in various application scenarios that require measuring the frequency response characteristics of electro-optical or optical-electrical systems. It eliminates the need for complex optical base hardware and optical modulation devices, and can calculate the frequency response characteristics of the device under test between the input and output nodes. This saves the overhead of de-embedding algorithms, resulting in a simplified overall measurement scheme design. Hardware requirements include only a signal generator to generate a compliant input signal, a time-domain waveform acquisition device capable of acquiring time-domain waveforms of electrical and optical signals, and a processor capable of executing digital processing algorithms. This facilitates the design and manufacturing of large-scale integrated optoelectronic chips that integrate optoelectronic devices.

[0072] Figure 4 This is a flowchart illustrating a waveform adjustment operation provided in an embodiment of this application. The processor performs the waveform adjustment operation on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, thereby converting the first optical signal into the second electrical signal. Figure 4 As shown, the waveform adjustment operation includes the following steps:

[0073] Step S401: Determine the average value of the waveform corresponding to the first optical signal.

[0074] Step S403: Subtract the average value of the waveform corresponding to the first optical signal from the average value of the waveform corresponding to the first optical signal.

[0075] Step S405: Perform waveform alignment operation in the waveform adjustment operation according to the waveform corresponding to the first electrical signal.

[0076] See Figure 1 , Figure 2 and Figure 3 Using an input signal comprising multiple frequency components, and through a time-domain waveform acquisition device, the time-domain waveforms of the input node and the output node were actually acquired. Here, because the device under test (DUT) has electro-optical or photoelectric conversion capabilities, the two acquired time-domain waveforms are the waveform corresponding to the first electrical signal and the waveform corresponding to the first optical signal, respectively. The difference depends on the internal structure of the DUT, which may involve either an input of the first electrical signal to an output of the first optical signal or vice versa. Since the time-domain waveform of the first electrical signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal, please refer to [reference needed]. Figure 4The waveform adjustment operation shown involves a processor adjusting the waveform corresponding to the first optical signal (either the input signal or the output signal is the first electrical signal) based on the waveform corresponding to the first electrical signal (which may correspond to either the input signal or the output signal). This adjustment is performed to convert the first optical signal into a second electrical signal. Then, the processor performs differential processing on the first and second electrical signals to obtain a first differential electrical signal. Thus, based on digital processing algorithms, the frequency domain transfer function is analyzed, and the higher-order system transfer function corresponding to the frequency response characteristics of the device under test is derived. Furthermore, the waveform adjustment operation reduces the common-mode component of the first differential electrical signal, and the differential-mode component is used to calculate the frequency response characteristics of the device under test. Here, through the waveform adjustment operation, the first optical signal is converted into a second electrical signal based on the time-domain waveform of the acquired first optical signal. The time-domain waveform of the second electrical signal can then be used for waveform analysis and comparison with the time-domain waveform of the first electrical signal. Using the processor's internal algorithms, a first optical signal can be converted into a second electrical signal through waveform adjustment operations. For example, waveform alignment can be achieved by removing the DC component. The frequency response characteristics of the device under test (DUT) are then calculated using the changes between the first and second electrical signals. Therefore, this approach can be applied to scenarios requiring a large number of optoelectronic devices, such as optical interconnects in data centers. For instance, the amplitude range of the waveform corresponding to the first electrical signal is from -100 millivolts (mV) to 100 mV, while the amplitude range of the waveform corresponding to the first optical signal is from 300 mW to 500 mW. To perform waveform alignment, the average values ​​of the waveforms can be aligned, i.e., the corresponding average values ​​are subtracted from each of the two original signals. Specifically, the average value of the waveform corresponding to the first optical signal is determined, and then the average value of the waveform corresponding to the first optical signal is subtracted from the average value of the waveform corresponding to the first optical signal. Then, the waveform alignment operation in the waveform adjustment operation is performed based on the waveform corresponding to the first electrical signal. Thus, through waveform adjustment and alignment, and by analyzing the frequency domain transfer function based on digital processing algorithms, the higher-order system transfer function corresponding to the frequency response characteristics of the DUT is derived. In this way, the frequency response characteristics of electro-optical or optical-electrical systems can be accurately measured, and the absence of reliance on optical modulation and demodulation technology saves additional instruments and resources, which is beneficial for chip miniaturization, high integration and low power consumption.

[0077] Figure 5 This is a schematic diagram illustrating the measurement of the frequency response characteristics of an optoelectronic device in a vehicle-mounted lidar system, as provided in an embodiment of this application. Figure 5As shown, the vehicle-mounted LiDAR system includes a controller 510, a driver 520, a laser 530, a sensor 550, and a data processing system 560. The controller 510 controls the driver 520 to generate a pulsed current, which then drives the laser 530 to emit pulsed light. The emitted light from the laser 530 is received by the sensor 550 after hitting the target 540, obtaining information such as photon time of flight. This information is then calculated by the data processing system 560 to obtain distance data, thus determining the distance between the vehicle-mounted LiDAR system and the target 540. In autonomous driving applications, the vehicle-mounted LiDAR system is used to detect the distance to various targets within its detection range and can also be used to generate laser point cloud maps. To detect farther distances, the driver 520 needs to generate a larger pulsed current, such as 40 amps. Furthermore, the light emitted by laser 530 undergoes significant attenuation after reflection from target 540. This means the intensity of the reflected light detected by sensor 550 is generally much lower than the intensity of the emitted light. To improve detection accuracy and signal-to-noise ratio, driver 520 needs to generate a larger pulse current to drive laser 530 to emit the required pulsed light. The basic structure of driver 520 includes a pre-drive module and a high-power switching transistor. The high-power switching transistor has a high switching frequency, enabling high-frequency switching operations to provide a high-frequency pulsed current. Laser 530 can be a laser diode, generating high-frequency pulsed light under the action of a high-frequency pulsed current. For example, laser 530 can be a VSEL array composed of multiple vertical-cavity surface-emitting lasers (VCSELs). The pre-drive module, together with the high-power switching transistor, constitutes the driving scheme, providing the control signal for the high-power switching transistor to perform high-frequency switching operations. The basic structure of sensor 550 includes single-photon avalanche diode (SPAD) detectors and time-to-digital converters (TDCs). The laser 530 at the transmitter of the vehicle-mounted LiDAR system generates multiple laser pulses at specified time intervals according to a certain time pattern. The area on target 540 illuminated by these laser pulses, also called emitted pulses, may have a non-uniform depth distribution. That is, the propagation time of some pixels in the area illuminated by the laser pulses emitted by the transmitter of the vehicle-mounted LiDAR system on target 540 may differ from the propagation time of other pixels. Therefore, the SPAD detectors at the receiver of the vehicle-mounted LiDAR system, typically a SPAD array composed of multiple SPAD detectors, can output multiple SPAD signals in response to the received optical radiation, including background noise.The SPAD detector is coupled to a time-to-digital converter (TDD), which generates digitized pulses with timing information based on the SPAD signals. The DTD provides statistical results of the timing and corresponding counts of multiple SPAD signals. For example, the DTD generates multiple histograms based on the counts of photons incident at various times on the SPAD detector. Each histogram corresponds to the re-establishment of a reflected pulse. Each histogram includes multiple timing and corresponding counts, and a peak is selected to represent the histogram and to calculate the final flight time of the corresponding emitted pulse. In other words, the DTD applies histogram-based filtering to the statistical results of the timing and corresponding counts of multiple SPAD signals.

[0078] See Figure 5 As can be seen, the control signal provided by the controller 510 from the driver 520 to the laser 530 is converted into an optical signal output. Therefore, the driver 520 and the laser 530 can be regarded as a device under test (DUT) with electro-optical conversion function. Therefore, in order to ensure the performance of the transmitter of the vehicle-mounted lidar system, it is necessary to evaluate the electro-optical system frequency response characteristics of the system composed of the driver 520 and the laser 530. For this purpose, the input terminal of the driver 520 can be used as the input node B562 of the DUT, and the output terminal of the laser 530 can be used as the output node B564 of the DUT. Thus, using the measurement method and apparatus for the frequency response characteristics of optoelectronic devices provided in the embodiments of this application, for example... Figure 1The method for measuring the frequency response characteristics of optoelectronic devices according to the first embodiment shown generates a compliant input signal using a signal generator, for example, by using a controller 510. A time-domain waveform acquisition device actually acquires the time-domain waveforms of a first electrical signal (corresponding to input node B562) and a first optical signal (corresponding to output node B564). Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, a waveform adjustment operation is performed on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, in order to convert the first optical signal into a second electrical signal. This provides a basis for waveform analysis and comparison and simplifies the design of the digital processing algorithm. This improves computational efficiency. The differential-mode component of the first differential electrical signal obtained through differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained through differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. In this way, the frequency response characteristics of the electro-optical system (the joint frequency response characteristics of the driver 520 and the laser 530) can be accurately measured. Moreover, it does not rely on optical modulation and demodulation technology, thus saving additional instruments and resources, which is beneficial for chip miniaturization, high integration and low power consumption.

[0079] Continue reading Figure 5 As can be seen, sensor 550 can be considered as a device under test (DUT) with photoelectric conversion functionality. Therefore, to ensure the performance of the receiver in the vehicle-mounted LiDAR system, it is necessary to evaluate the system frequency response characteristics of the photoelectric conversion of sensor 550. For this purpose, the input terminal of sensor 550 can be used as the input node C572 of the DUT, and the output terminal of sensor 550 can be used as the output node C574 of the DUT. Thus, the measurement method and apparatus for the frequency response characteristics of optoelectronic devices provided in the embodiments of this application can be used, for example... Figure 1The method for measuring the frequency response characteristics of optoelectronic devices according to the first embodiment shown generates a compliant input signal using a signal generator, for example, by providing a compliant input signal using a controller 510. A time-domain waveform acquisition device actually acquires the time-domain waveforms of a first electrical signal (corresponding to output node C574) and a first optical signal (corresponding to input node C572). Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, a waveform adjustment operation is performed on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, in order to convert the first optical signal into a second electrical signal. This provides a basis for waveform analysis and comparison and simplifies digital processing. The algorithm design improves computational efficiency. The differential-mode component of the first differential electrical signal obtained through differential processing by the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained through differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves computational accuracy. Thus, accurate measurement of the frequency response characteristics of the photoelectric conversion system (frequency response characteristics of sensor 550) is achieved. Moreover, it does not rely on optical modulation and demodulation technology, thus saving additional instruments and resources, which is beneficial for chip miniaturization, high integration, and low power consumption.

[0080] Figure 6 This is a schematic diagram illustrating the measurement of the frequency response characteristics of an optoelectronic device used in an optical interconnect for a data center, as provided in an embodiment of this application. Figure 6 The diagram illustrates that within a data center employing optical interconnect technology, different terminal nodes, such as computing nodes, servers, and storage devices, are connected to the network through their respective network interfaces, i.e., adapters, and perform functions such as electrical domain caching, photoelectric conversion, and electro-optical conversion through their respective adapters. Figure 6 Both electrical adapters A610 and B612 are connected to the optical interconnect network. The output channel of electrical adapter A610 exits the terminal node where it resides via an electro-optical converter 630 to enter the optical interconnect network 620, while the input channel of electrical adapter B612 exits the optical interconnect network 620 via a photoelectric converter 632 to enter the terminal node where it resides. The optical interconnect network 620 may internally include a multi-level structure for optical packetization, sequencing, and assembly.

[0081] See Figure 6It can be seen that the performance of optical interconnects in data centers depends on the performance of the electro-optical converter 630 and the optoelectronic converter 632. Although optical interconnect networks can achieve all-optical data transmission channels and may not require optoelectronic or electro-optical conversion during data transmission, the construction of complete input and output channels, as well as the complete data transmission from one end node to another, still involves the conversion between electrical and optical data. To ensure the correct operation of optical interconnects in data centers, it is necessary to evaluate the frequency response characteristics of optoelectronic devices in the optical interconnects, including evaluating... Figure 6 The electro-optic converter 630 shown exhibits electro-optical system frequency response characteristics, and evaluation Figure 6 The photoelectric converter 632 shown exhibits the system frequency response characteristics of its photoelectric conversion. Thus, the measurement method and apparatus for the frequency response characteristics of optoelectronic devices provided in this application embodiment, for example... Figure 1 The method for measuring the frequency response characteristics of optoelectronic devices according to the first embodiment shown generates a required input signal using a signal generator. A time-domain waveform acquisition device actually acquires the time-domain waveforms of a first electrical signal and a first optical signal (the acquisition of the corresponding signal's time-domain waveform depends on the measurement needs of the electro-optic converter 630 and the optoelectronic converter 632). Addressing the issue that the time-domain waveform of the first electrical signal cannot be directly compared with the time-domain waveform of the first optical signal, a waveform adjustment operation is performed on the waveform corresponding to the first electrical signal to convert the first optical signal into a second electrical signal. This provides a basis for waveform analysis and comparison, simplifies the design of digital processing algorithms, and improves computational efficiency. The differential-mode component of the first differential electrical signal obtained by differential processing through the processor reflects the change between the first and second electrical signals. This change information can be used to calculate the frequency response characteristics of the device under test. Furthermore, the common-mode component of the first differential electrical signal obtained by differential processing can be minimized through waveform adjustment, which helps to further highlight the differential-mode component of the first differential electrical signal and improves the calculation accuracy. In this way, the frequency response characteristics of the electro-optical system (electro-optical converter 630) and the optical-electrical system (photoelectric converter 632) can be accurately measured. Moreover, it does not rely on optical modulation and demodulation technology, which saves additional instruments and resources and is conducive to the miniaturization, high integration and low power consumption of the chip.

[0082] See Figure 1 , Figure 2 and Figure 3The first acquisition device obtains the input waveform of the input signal at the input node of the device under test (DUT), and the second acquisition device obtains the output waveform of the output signal at the output node of the DUT. The input signal includes multiple frequency components, and the DUT has electro-optical or photoelectric conversion capabilities. One of the input signal and the output signal is a first electrical signal, and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal, and the other is the waveform corresponding to the first optical signal. Thus, the first acquisition device with time-domain waveform acquisition capability obtains the input waveform of the input signal at the input node of the DUT, i.e., actually acquires the time-domain effect diagram of the input signal. Similarly, the second acquisition device with time-domain waveform acquisition capability obtains the output waveform of the output signal at the output node of the DUT, i.e., actually acquires the time-domain effect diagram of the output signal. Figure 1Taking the first embodiment of the method for measuring the frequency response characteristics of optoelectronic devices as an example, a time-domain waveform acquisition device is used to actually acquire the time-domain waveforms of a first electrical signal and a first optical signal. Addressing the issue that the time-domain waveforms of the first electrical signal and the first optical signal cannot be directly analyzed and compared, a waveform adjustment operation is performed on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal. This provides a basis for waveform analysis and comparison, simplifies the design of digital processing algorithms, and improves computational efficiency. In applications such as high-speed optical communication systems and data center optical interconnects, the frequency response characteristics of systems involving electro-optical conversion are crucial. For example, an electrical signal adjusted by an equalizer is converted into an optical signal and output. An equalizer can be used to correct the frequency response characteristics of the transmission channel, reduce inter-symbol interference, and provide compensation, helping to restore signal amplitude, rise time, and fall time. As chip size shrinks and integration increases, modules such as equalizers are typically integrated inside the chip, for example, deployed within the die as part of a functional chip. Therefore, integrated circuits providing specific functions may include equalizers or similar modules for adjusting the frequency components of electrical signals. Because equalizers or similar modules are integrated inside the chip or as part of an integrated circuit, the input side of the device under test (DUT), i.e., the input node, may be located inside the chip or difficult to locate and detect. This makes it difficult to directly acquire the input waveform using a time-domain waveform acquisition device. Therefore, some DUTs with electro-optical conversion capabilities may exist where the electrical signal input occurs at a location difficult to directly acquire in the time domain. Thus, they can only directly acquire the optical signal output using a time-domain waveform acquisition device. This makes it impossible to perform waveform adjustment and subsequent waveform analysis and comparison based on the directly acquired time-domain waveforms of the electrical and optical signals. The following section will combine... Figure 7 and Figure 8 This section details how to address the issue of the input side of the device under test being located inside the chip or being difficult to locate and detect.

[0083] Figure 7 This is a schematic flowchart illustrating a method for measuring the frequency response characteristics of an optoelectronic device according to a second embodiment of this application. Figure 7 As shown, the method includes the following steps.

[0084] Step S701: Using a simulation waveform generator, a simulated time-domain waveform corresponding to the digital logic of the input signal received by the input node of the device under test is generated as the simulated input waveform of the input signal, based on the digital logic of the input signal received by the input node of the device under test. Additionally, using a time-domain waveform acquisition device, the output waveform of the output signal is obtained at the output node of the device under test. The input signal is a wide-frequency-range wave packet signal including multiple frequency components. The wave packet signal includes a test code pattern, which is a series of identical digital segments with a preset segment length. The input signal is a first electrical signal, and the output signal is a second electrical signal or a first optical signal.

[0085] Step S703: When the output signal is the second electrical signal, the processor calculates the frequency response characteristics of the device under test based on the simulated input waveform and the waveform of the second electrical signal.

[0086] Step S705: When the output signal is the first optical signal, the processor adjusts the output waveform of the output signal according to the simulated input waveform to convert the first optical signal into a third electrical signal. Then, the processor calculates the frequency response characteristics of the device under test based on the simulated input waveform and the waveform of the third electrical signal.

[0087] Figure 7The second embodiment of the method for measuring the frequency response characteristics of optoelectronic devices can be applied in technical fields such as high-speed digital communication, high-speed optical communication, high-speed transmission interfaces, and optical interconnects, for example, in applications such as data centers, high-performance servers, cloud computing, and artificial intelligence. It utilizes a simulation waveform generator, a time-domain waveform acquisition unit, and a processor to calculate the frequency response characteristics of the device under test (DUT). The DUT can be an equalizer or a similar functional module for adjusting the frequency components of an electrical signal. In some embodiments, the DUT can be used in applications such as high-speed optical communication systems and data center optical interconnects, and involves the frequency response characteristics of electro-optical systems, for example, where an electrical signal adjusted by an equalizer is converted into an optical signal and output. In some embodiments, the DUT can be integrated inside a chip or as part of an integrated circuit, which means that it is difficult to obtain the frequency response characteristics of the DUT by connecting test equipment such as a vector network analyzer (VNA) to both the input and output ends. The device under test (DUT) is deployed between its upstream and downstream circuits. This may involve data transmission and processing between different nodes and machines. Frequent signal amplification or attenuation may be necessary to match the differences between upstream and downstream circuits. Adjustments to the frequency components of the electrical signal may also be needed to counteract inter-symbol interference (ISI) caused by the signal transmission path, such as printed circuit boards (PCBs) and cables, thereby improving the error-free transmission distance. During use, the equalizer or similar functional modules within the DUT may deviate from their design performance due to factors such as component aging and wear, affecting compensation effectiveness and signal transmission performance. Therefore, it is necessary to acquire and evaluate the frequency domain characteristics and frequency response of the equalizer or similar functional modules in real time and repeatedly to provide a reference for configuration adjustments and equipment calibration. However, the input terminal or input node of the device under test may be located inside the chip, or it may be difficult to accurately locate by analyzing the circuit structure, such as an on-chip equalizer. This means that it is difficult to obtain the input electrical signal on the input node through the corresponding pin or probe. In contrast, the output terminal or output node of the device under test can generally obtain the output electrical signal through the pin or probe, or the output optical signal can be measured by connecting a VNA to the output terminal.While it's possible to measure the time-domain performance of the entire chip or the entire integrated circuit including the device under test (DUT) and therefore its internal equalizer, this measurement yields a joint frequency response characteristic for the entire chip or integrated circuit. It's difficult to distinguish the DUT's frequency response characteristics from other modules; for example, it might be difficult to differentiate the DUT's frequency response characteristics from those of the preceding or following circuits. This hinders the accurate acquisition of the equalizer's frequency response characteristics. Furthermore, while deploying additional feedback loops and detector circuits may enable real-time detection of high-frequency phase errors in the DUT, this necessitates additional circuit area and increases hardware costs. It also doesn't reduce overall power consumption, thus failing to meet the requirements for chip miniaturization, high integration, and low power consumption.

[0088] See Figure 7In step S701, a simulated time-domain waveform corresponding to the digital logic of the input signal received by the input node of the device under test (DUT) is generated by a simulated waveform generator as the simulated input waveform of the input signal, based on the digital logic of the input signal received by the input node of the DUT. Additionally, an output waveform of the output signal is obtained at the output node of the DUT by a time-domain waveform acquisition device. In some embodiments, the input node of the DUT may be located inside a chip or is difficult to accurately locate by analyzing the circuit structure. Therefore, it is difficult to directly acquire the time-domain waveform of the input signal using a acquisition device. For example, the DUT may be an on-chip equalizer, or it may be integrated inside a die as part of the physical layer of the die. Although it is difficult to directly acquire the time-domain waveform or time-domain effect diagram of the input signal received by the input node, the digital logic of the input signal received by the input node can be determined, for example, the timing diagram of the high level (generally corresponding to digital logic 1 in a digital circuit) and low level (generally corresponding to digital logic 0 in a digital circuit) on the input node can be determined. It should be understood that, depending on the specific circuit design requirements, voltage levels can be set to represent binary "1" and "0". Generally, a high level corresponds to logic 1, and a low level corresponds to logic 0; conversely, in some cases, a high level can correspond to logic 0, and a low level to logic 1. Furthermore, different definitions and standards for voltage levels can be set, such as different logic level standards. By determining the digital logic of the input signal received by the input node of the device under test (DUT), and then combining the corresponding logic level standards and the correspondence between high and low levels and digital logic, a timing diagram reflecting the high and low level changes can be determined based on the digital logic of the input signal. Thus, a simulated time-domain waveform corresponding to the digital logic of the input signal is generated using a simulated waveform generator as the simulated input waveform of the input signal. Without needing to actually acquire the time-domain waveform of the input node, an approximate time-domain waveform, i.e., a simulated time-domain waveform corresponding to the digital logic of the input signal, can be generated based on the digital logic of the input signal received by the input node. Additionally, the output waveform of the output signal is obtained at the output node of the DUT using a time-domain waveform acquisition device, thus actually acquiring the time-domain waveform of the output node. As can be seen, the input node is equivalent to a virtual node because it is not necessary to actually acquire the virtual waveform of the input node. The input node is used to better indicate the digital logic of the input signal. In other words, the reference for subsequent processes is not the time-domain waveform of the actual acquired input node, but the simulated time-domain waveform generated by the simulated waveform generator that corresponds to the digital logic of the input signal, and thus serves as the simulated input waveform of the input signal for subsequent processes.Output nodes are actual physical nodes. For example, the output waveform can be obtained through pins, or it can be accurately located by analyzing the circuit structure and then obtained through probes. Alternatively, it can be measured by connecting a vector network analyzer to measure the output optical signal. The time-domain waveform of the actual acquired output node may be either an electrical signal or an optical signal, serving as a reference for subsequent processes. Similarly, the simulated input waveform of the input signal (i.e., the simulated time-domain waveform generated by the simulated waveform generator corresponding to the digital logic of the input signal), and the output waveform of the output signal actually acquired by the time-domain waveform acquisition device, serve as references for subsequent processes. Through waveform comparison and analysis in subsequent processes, the frequency response characteristics of the device under test (DUT) can be calculated. When the DUT includes an equalizer and a driver, the calculated frequency response characteristics are the combined frequency response characteristics of the equalizer and driver. When the DUT includes an equalizer, a driver, and optoelectronic devices, the calculated frequency response characteristics are the combined frequency response characteristics of the equalizer, driver, and optoelectronic devices. In some embodiments, the internal signal compensation method of the device under test can adopt a variety of technical means, and can also be combined with whether the device under test is deployed at the signal transmitting end or the signal receiving end to adopt the corresponding signal compensation method, such as continuous-time linear equalizer (CTLE), feed-forward equalizer (FEE), decision feedback equalizer (DFE), and pre-emphasis and de-emphasis.

[0089] Continue reading Figure 7The input signal is a wave packet signal with a wide frequency range including multiple frequency components. The wave packet signal includes a test code pattern, which is a continuous identical digit (CID) segment with a preset segment length. The input signal is a first electrical signal, and the output signal is a second electrical signal or a first optical signal. As described above, the simulated input waveform of the input signal, i.e., the simulated time-domain waveform corresponding to the digital logic of the input signal generated by the simulated waveform generator, and the output waveform of the output signal actually acquired by the time-domain waveform acquisition device, serve as reference benchmarks for subsequent processes. Through waveform comparison and analysis in subsequent processes, the frequency response characteristics of the device under test can be calculated. Therefore, by using a wave packet signal with a wide frequency range including multiple frequency components as the input signal, and by specifically designing the code pattern of the wave packet signal, i.e., requiring the wave packet signal to include a test code pattern and the test code pattern to be a continuous identical digit segment with a preset segment length, the frequency response characteristics of the device under test can be calculated. Thus, by inputting the optimized input signal to the device under test (DUT), and assuming no significant change between the frequency ranges of the input and output signals, the frequency domain transfer function can be analyzed using digital processing algorithms based on the simulated input waveform and the output waveform. This includes deriving the higher-order system transfer function corresponding to the frequency response characteristics of the DUT, calculating equalization values ​​(e.g., gain ratio or compression ratio), and determining peak, equalization, and Nyquist frequencies. Clock and data recovery (CDR) circuits and waveform shaping circuits can cause significant changes between the frequency ranges of the input and output signals, affecting the calculation of the DUT's frequency response characteristics. Therefore, DUTs generally do not include clock and data recovery or waveform shaping circuits internally to ensure minimal changes between the input and output frequency ranges. Thus, in the absence of internal devices or circuits such as clock data recovery circuits or waveform shaping circuits in the device under test (DUT) that would cause significant variations between the frequency range of the input signal and the frequency range of the output signal, a wave packet signal with multiple frequency components over a wide frequency range can be used as the input signal. Furthermore, a test code pattern consisting of consecutive identical digital segments of a preset length can be inserted into the input signal. This allows for the use of digital processing algorithms to construct a high-order system transfer function corresponding to the frequency response characteristics of the DUT, thereby enabling frequency domain transfer function analysis.In some embodiments, the input signal can be generated using a pseudo-random sequence, such as a pseudo-random binary sequence (PRBS), or other random generation algorithms, to produce a binary code sequence of 0s and 1s with random characteristics. This means that the digital logic of the input signal has certain random characteristics, as long as it meets the requirements of a wide-frequency-range wave packet signal with multiple frequency components and the test code pattern requirements of consecutive identical digital segments with a preset segment length. The wave packet signal includes multiple frequency components, that is, at least two different frequency components. The amplitudes of these multiple frequency components may be consistent or inconsistent. Therefore, influenced by multiple frequency components, this means that the time-domain waveform of the input signal (if it can be actually acquired) has the characteristic of irregular amplitude changes. As mentioned above, the input node is equivalent to a virtual node, which is actually impossible to measure or too difficult to measure. Therefore, a simulation waveform generator is used to generate a simulated time-domain waveform corresponding to the digital logic of the input signal, and the code pattern requirements of the input signal are optimized by using a wide-frequency-range wave packet signal with multiple frequency components. The time-domain waveform is actually acquired at the output node. When the device under test (DUT) exhibits electrical signal-to-electrical signal transmission characteristics, the time-domain waveform acquired at the output node is the time-domain waveform of the electrical signal. Conversely, when the DUT exhibits electrical signal-to-optical signal transmission characteristics, the time-domain waveform acquired at the output node is the time-domain waveform of the optical signal. The specific configurations of the DUT's front-end and back-end circuits can be flexibly adjusted according to actual needs. When the DUT is deployed at the receiving end, the front-end circuit can be the clock data recovery circuit within the chip, and the input node of the DUT can be the node where the clock data recovery circuit outputs the recovered signal. When the DUT is deployed at the transmitting end, the front-end circuit can be a signal sequence generator, and the input node of the DUT can be the output node of the transmitted signal sequence generated by the signal sequence generator.

[0090] Continue reading Figure 7The input signal is a first electrical signal, and the output signal is a second electrical signal or a first optical signal. In some embodiments, the device under test (DUT) may have electrical signal-to-electrical signal transmission characteristics. For example, the DUT may internally include an equalizer and a driver, meaning that the output waveform of the output signal obtained at the output node of the DUT by the time-domain waveform acquisition device is the time-domain waveform of the electrical signal. In step S703, when the output signal is the second electrical signal, the processor calculates the frequency response characteristics of the DUT based on the simulated input waveform and the waveform of the second electrical signal. Thus, using a wide-frequency-range wave packet signal including multiple frequency components as the input signal, and utilizing the test code pattern included in the wave packet signal as continuous identical digital segments with a preset segment length, and through a simulated waveform generator, generating a simulated time-domain waveform corresponding to the digital logic of the input signal as the simulated input waveform of the input signal without actually acquiring the time-domain waveform of the input node, and through a time-domain waveform acquisition device, actually acquiring the time-domain waveform of the output node, obtaining the output waveform of the output signal at the output node of the device under test, and finally, using the simulated input waveform of the input signal and the output waveform of the output signal, performing frequency domain transfer function analysis based on digital processing algorithms, and deriving the higher-order system transfer function corresponding to the frequency response characteristics of the device under test. In this way, the frequency response characteristics of the equalizer are accurately obtained, effectively handling situations where the equalizer input is located inside the chip, such as obtaining the frequency response characteristics of an on-chip equalizer, without needing to deploy additional feedback loops and detection circuits into the chip, which is beneficial for chip miniaturization, high integration, and low power consumption.

[0091] Continue reading Figure 7The input signal is a first electrical signal, and the output signal is a second electrical signal or a first optical signal. In some embodiments, the device under test (DUT) may have electrical signal to optical signal transmission characteristics, for example, the DUT may include an equalizer, a driver, and optoelectronic devices. This allows the electrical signal, after being adjusted by the equalizer, to be finally converted into an optical signal and output. This means that the output waveform of the output signal obtained by the time-domain waveform acquisition device at the output node of the DUT is the time-domain waveform of the optical signal. In step S705, when the output signal is the first optical signal, the processor adjusts the output waveform of the output signal according to the simulated input waveform to convert the first optical signal into a third electrical signal. Then, the processor calculates the frequency response characteristics of the DUT based on the simulated input waveform and the waveform of the third electrical signal. Thus, a wide-frequency-range wave packet signal including multiple frequency components is used as the input signal, and the test code pattern included in the wave packet signal is a series of identical digital segments with a preset segment length. Furthermore, a simulated time-domain waveform corresponding to the digital logic of the input signal is generated as the simulated input waveform of the input signal using a simulated waveform generator, without the need to actually acquire the time-domain waveform of the input node. Additionally, the time-domain waveform of the output node is actually acquired using a time-domain waveform acquisition device, obtaining the output waveform of the output signal at the output node of the device under test. Here, the output signal is the first optical signal, therefore the output waveform of the output signal is the time-domain waveform of the first optical signal. The simulated input waveform of the input signal is a simulated digital waveform generated based on the digital logic of the input signal; therefore, the simulated input waveform of the input signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal. Therefore, when the output signal is the first optical signal, the processor performs a waveform adjustment operation on the output waveform of the output signal according to the simulated input waveform, thus converting the first optical signal into a third electrical signal. Then, based on the simulated input waveform and the waveform of the third electrical signal, the frequency response characteristics of the device under test are calculated. Thus, based on digital processing algorithms, the frequency domain transfer function is analyzed, and the higher-order system transfer function corresponding to the frequency response characteristics of the device under test is derived. This achieves accurate acquisition of the equalizer's frequency response characteristics, effectively addressing situations where the equalizer's input is located inside the chip, such as acquiring the frequency response characteristics of an on-chip equalizer. Furthermore, it eliminates the need to deploy additional feedback loops and detection circuits within the chip, which is beneficial for chip miniaturization, high integration, and low power consumption.

[0092] Continue reading Figure 7Optimizing the input signal is crucial for calculating the frequency response characteristics of the device under test (DUT). Specifically, requirements are set for the input signal's composition, limiting it to a wide-frequency-range wave packet signal comprising multiple frequency components. Furthermore, requirements are set for the input signal's code pattern, specifying that the wave packet signal includes a test code pattern, and that the test code pattern consists of consecutive identical digital segments of a predetermined segment length. Thus, the amplitude distribution of the multiple frequency components in the input signal follows a certain pattern, such as having consistent or inconsistent amplitudes, ultimately forming a wide-frequency-range wave packet signal. The wave packet signal corresponding to the input signal is input to the DUT from the input node and then output to the outside of the DUT via the output node. Therefore, during its transmission within the DUT, the frequency components of the wave packet signal may be adjusted, for example, amplified or attenuated. This means that the multiple frequency components included in the wave packet signal undergo an overall change in the frequency domain due to their transmission within the DUT. Therefore, by inputting a wide-frequency-range wave packet signal containing multiple frequency components into the device under test (DUT), and obtaining the corresponding output signal, this conversion process from input to output signal is influenced by the internal components and circuits of the DUT, which adjust the frequency components of the signal. For example, it is affected by the internal signal compensation functions of the DUT (e.g., pre-emphasis, de-emphasis, equalization, etc.). Thus, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce the information about the frequency response characteristics of the DUT carried by these changes. This allows for the derivation of a higher-order system transfer function (SFC) that matches the frequency response characteristics of the DUT, which can be used for configuration adjustments, equipment calibration, and other applications. Here, the digital processing algorithm used to derive the higher-order SFC that matches the frequency response characteristics of the DUT can employ any suitable algorithm model and principle. For example, an error convergence algorithm can be used to select the SFC with the smallest error from multiple SFCs. Alternatively, the error function can be defined as a function of the number of poles and zeros, and then iterative and adaptive algorithms can be used to screen for the SFC with the smallest error. Another example is the use of digital fitting algorithms to approximate a suitable SFC model. Furthermore, considering that the input signal is limited to a wave packet signal with a wide frequency range and multiple frequency components, this means that, from the perspective of time domain analysis, the time-domain waveform of the input signal (if it is assumed that it can be actually acquired at the input node) has irregular amplitude variations, and the distribution of the peaks and troughs of the time-domain waveform of the input signal may also be irregular.Therefore, the digital processing algorithm used to derive the higher-order system transfer function (STF) that matches the frequency response characteristics of the device under test (DUT) does not compare rise and fall times in the time domain. Instead, it determines the overall changes in rise and fall times in the frequency domain. This leverages the key design requirement that the input signal is a wave packet signal with multiple frequency components over a wide frequency range. By deriving the higher-order STF with the minimum error, it characterizes the overall changes in rise and fall times in the frequency domain. Furthermore, the digital processing algorithm for deriving the higher-order STF that matches the frequency response characteristics of the DUT can consider other factors in the changes between the input and output signals, besides the changes in rise and fall times, thereby improving calculation accuracy. These factors include amplitude variations and delay alignment.

[0093] Continue reading Figure 7As mentioned above, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce the information about the frequency response characteristics of the device under test (DUT) carried by these changes. This allows for the derivation of a higher-order system transfer function (STU) that matches the DUT's frequency response characteristics. Specifically, by leveraging the key design requirement that the input signal is a wide-frequency-range wave packet signal with multiple frequency components, instead of comparing rise and fall times in the time domain, the overall changes in rise and fall times are determined in the frequency domain. This allows for the derivation of a higher-order STU with minimal error, thus characterizing the overall rise and fall time changes in the frequency domain. Therefore, compared to inputting a signal with only a single frequency component and then changing that single frequency component for frequency scanning, using a wide-frequency-range wave packet signal with multiple frequency components as the input signal to the DUT allows for a more precise control over the amplitude distribution of the multiple frequency components. Furthermore, these multiple frequency components are pre-designed and do not change with the scanning mode. Here, the wave packet form of the wide-frequency-range wave packet signal with multiple frequency components is determined based on the data format of the input signal's code. As mentioned above, requirements are set for the code pattern of the input signal, specifying that the wave packet signal includes a test code pattern, and that the test code pattern is a series of identical digital segments with a preset segment length. For example, a constant 0 segment or a constant 1 segment is a series of identical digital segments with a preset segment length. The input signal can be generated using a pseudo-random sequence, such as a pseudo-random binary sequence, or other random generation algorithms, to produce a binary code sequence of 0s and 1s with random characteristics. The wave packet form is determined according to the specific pseudo-random code data format used by the algorithm that generates the input signal; for example, a constant 0 or constant 1 segment in that data format is used as a reference to adjust the amplitude. The advantage of this design is that, considering the irregular amplitude variation of the time-domain waveform of the input signal (assuming it can be actually acquired at the input node), and the irregular distribution of the peaks and troughs of the time-domain waveform of the input signal, the design requirements of the input signal code pattern, namely, the test code pattern with consecutive identical digital segments of a preset segment length, are utilized. For example, by utilizing the mechanism of consecutive identical digital segments of the pseudo-random binary sequence itself, multiple consecutive identical codewords may appear, such as nine consecutive constant 0 segments or 23 consecutive constant 1 segments. This can serve as a benchmark for waveform comparison and analysis.It should be noted that the input signal is a wave packet signal with a wide frequency range and multiple frequency components. Therefore, the test code pattern of consecutive identical digital segments with a preset segment length included in the wave packet signal corresponds to the low-frequency part of the multiple frequency components in the wave packet signal. Therefore, using the test code pattern of consecutive identical digital segments with a preset segment length as a reference is equivalent to using the low-frequency part of the wave packet signal to derive the equalization value of the device under test (DUT), such as the gain ratio or compression ratio. This allows for amplitude adjustment to improve the accuracy of the final calculation of the frequency response characteristics of the DUT. Furthermore, the DUT may include optoelectronic devices, such as devices that convert electrical signals into optical signals, like optical emitting devices. When the DUT includes optoelectronic devices, it means that the output signal is the first optical signal. Therefore, the output waveform of the output signal is the time-domain waveform of the first optical signal. The simulated input waveform of the input signal is a simulated digital waveform generated based on the digital logic of the input signal. Therefore, the simulated input waveform of the input signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal. By utilizing the processor's internal algorithm, the first optical signal can be converted into a third electrical signal through waveform adjustment operations. For example, waveform alignment can be achieved by removing the DC component. The frequency response characteristics of the device under test can then be calculated by using the changes between the first and third electrical signals. Therefore, it can be applied to scenarios that require a large number of optoelectronic devices, such as optical interconnects in data centers.

[0094] In short, Figure 7The second embodiment of the method for measuring the frequency response characteristics of optoelectronic devices imposes requirements on the composition of the input signal, limiting it to a wave packet signal with a wide frequency range including multiple frequency components. It also imposes requirements on the code pattern of the input signal, limiting the wave packet signal to include a test code pattern, where the test code pattern consists of consecutive identical digital segments with a preset segment length. Through optimized design of the input signal, a high-order system transfer function matching the frequency response characteristics of the device under test can be derived using digital processing algorithms, thus improving calculation accuracy. Furthermore, by using a simulation waveform generator, a simulated time-domain waveform corresponding to the digital logic of the input signal is generated without the need for actual acquisition of the time-domain waveform of the input node. The simulated input waveform of the input signal and the actual time-domain waveform of the output node were acquired through a time-domain waveform acquisition device. The output waveform of the output signal was obtained at the output node of the device under test. Finally, the frequency domain transfer function was analyzed using the simulated input waveform of the input signal and the output waveform of the output signal. The first optical signal was converted into a third electrical signal by using waveform adjustment operation, and the frequency response characteristics of the electro-optical system were calculated. The frequency response characteristics of the equalizer were accurately obtained, which can effectively deal with the situation where the input end of the equalizer is located inside the chip, such as obtaining the frequency response characteristics of the on-chip equalizer. Moreover, it does not require the deployment of additional feedback loops and detection circuits into the chip, which is beneficial to the miniaturization design, high integration and low power consumption of the chip.

[0095] See Figure 7 , Figure 7 The second embodiment of the method for measuring the frequency response characteristics of optoelectronic devices, as shown, uses a first electrical signal as the input signal and a first optical signal as the output signal. This corresponds to the case of electrical signal input and optical signal output, thus realizing the measurement of the frequency response characteristics of the device under test (DUT) with electro-optical conversion function, i.e., the system frequency response characteristics of electro-optical conversion. As mentioned above, some DUTs with electro-optical conversion function may exist where the electrical signal input occurs at a location where direct time-domain waveform acquisition is difficult. Therefore, the output optical signal can only be directly acquired using a time-domain waveform acquisition device. This makes it impossible to perform waveform adjustment and subsequent waveform analysis and comparison based on the directly acquired time-domain waveforms of the electrical and optical signals. Figure 7 The second embodiment of the method for measuring the frequency response characteristics of optoelectronic devices can effectively address the problem that the input side of the device under test is located inside the chip or is difficult to locate and detect.

[0096] Figure 8 This is a schematic diagram of a measurement device for the frequency response characteristics of an optoelectronic device, provided as an embodiment of the second version of this application. Figure 8As shown, the measuring device includes a simulated waveform generator D801, a time-domain waveform acquisition unit D803, and a processor D805. The simulated waveform generator D801 generates a simulated time-domain waveform corresponding to the digital logic D820 of the input signal received at the input node D812 of the device under test (D810), serving as the simulated input waveform D822 of the input signal. The time-domain waveform acquisition unit D803 obtains the output waveform D824 of the output signal at the output node D814 of the D810. The input signal is a wide-frequency-range wave packet signal including multiple frequency components, and the wave packet signal includes a test code pattern, which is a series of identical digital segments with a preset segment length. The input signal is a first electrical signal, and the output signal is a second electrical signal or a first optical signal. The processor D805 calculates the frequency response characteristics of the D810 based on the simulated input waveform D822 and the waveform of the second electrical signal when the output signal is the second electrical signal. The processor D805 is further configured to, when the output signal is the first optical signal, perform waveform adjustment on the output waveform D824 of the output signal according to the simulated input waveform D822, so as to convert the first optical signal into a third electrical signal. The processor D805 is also configured to calculate the frequency response characteristics of the device under test D810 based on the simulated input waveform D822 and the waveform of the third electrical signal. Figure 8 The diagram also schematically illustrates the pre-amplifier circuit D880 and the post-amplifier circuit D882 relative to the device under test (D810). It should be understood that the pre-amplifier circuit D880 and the post-amplifier circuit D882 are only used to illustrate possible pre-amplifier circuit relationships for the D810. Depending on the specific deployment of the D810 and the chip design purpose, the pre-amplifier circuit D880 and the post-amplifier circuit D882 can each correspond to different functional modules, sub-circuits, and devices. The specific configuration of the pre-amplifier circuit D880 and the post-amplifier circuit D882 of the D810 can be flexibly adjusted according to actual needs. For example, when the D810 is deployed at the receiving end, the pre-amplifier circuit D880 can be the clock data recovery circuit inside the chip, and the input node D812 of the D810 can be the node where the clock data recovery circuit outputs the recovered signal. For example, when the device under test (DUT) D810 is deployed at the transmitting end, the front-end circuit D880 of DUT D810 can be a signal sequence generator, and the input node D812 of DUT D810 can be the output node of the transmitted signal sequence generated by the signal sequence generator. The back-end circuit D882 of DUT D810 can correspond to chip pins, output modules, etc., depending on the specific application requirements.

[0097] Figure 8The second embodiment of the measurement device for the frequency response characteristics of optoelectronic devices, as shown, imposes requirements on the composition of the input signal. It specifies that the input signal is a wave packet signal with a wide frequency range including multiple frequency components. Furthermore, it imposes requirements on the code pattern of the input signal, specifying that the wave packet signal includes a test code pattern, and that the test code pattern is a series of identical digital segments with a preset segment length. Through optimized design of the input signal, a high-order system transfer function matching the frequency response characteristics of the device under test (D810) can be derived using digital processing algorithms, thus improving calculation accuracy. Using the simulation waveform generator D801, a simulated time-domain waveform corresponding to the digital logic D820 of the input signal is generated as a simulation of the input signal without the need to actually acquire the time-domain waveform of the input node D812. The true input waveform D822, and the actual time-domain waveform of the output node D814 are acquired through the time-domain waveform acquisition unit D803. The output waveform D824 of the output signal is obtained at the output node D814 of the device under test D810. Finally, the frequency domain transfer function is analyzed using the simulated input waveform D822 and the output waveform D824 of the output signal. The first optical signal is converted into a third electrical signal by waveform adjustment operation, realizing the calculation of the frequency response characteristics of the electro-optical system. The frequency response characteristics of the equalizer are accurately obtained, which can effectively deal with the situation where the input end of the equalizer is located inside the chip, such as obtaining the frequency response characteristics of the on-chip equalizer. Moreover, it does not require the deployment of additional feedback loops and detection circuits into the chip, which is beneficial to the miniaturization design, high integration and low power consumption of the chip.

[0098] See Figure 8 , Figure 8 The second embodiment of the measuring device for the frequency response characteristics of optoelectronic devices, as shown, uses a first electrical signal as the input signal and a first optical signal as the output signal. This corresponds to the case of electrical signal input and optical signal output, thus realizing the measurement of the frequency response characteristics of the device under test D810 with electro-optical conversion function, i.e., the system frequency response characteristics of electro-optical conversion. As mentioned above, there may be some devices under test with electro-optical conversion function (e.g., Figure 8 The device under test (D810) is located at a point where direct time-domain waveform acquisition is difficult, therefore it can only be achieved through a time-domain waveform acquisition device (e.g., D810). Figure 8 The time-domain waveform acquisition device (D803) shown directly acquires the output of the optical signal. Therefore, it is impossible to perform waveform adjustment and subsequent waveform analysis and comparison based on the time-domain waveforms of the directly acquired electrical and optical signals. Figure 8 The second embodiment of the measuring device for the frequency response characteristics of optoelectronic devices shown can effectively address the problem that the input side of the device under test D810 is located inside the chip or is difficult to locate and detect.

[0099] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of an optoelectronic device, as shown, in one possible implementation, has at least two frequency components whose amplitudes are inconsistent, or the amplitudes of the multiple frequency components are identical. Requirements are placed on the composition of the input signal, limiting it to a wave packet signal with a wide frequency range including multiple frequency components, and requirements are placed on the code pattern of the input signal, limiting the wave packet signal to include a test code pattern, and the test code pattern to be a series of identical digital segments with a preset segment length. Thus, the amplitude distribution of the multiple frequency components in the input signal refers to a certain pattern, for example, having consistent or inconsistent amplitudes, ultimately forming a wave packet signal with a wide frequency range. The wave packet signal corresponding to the input signal is input to the device under test (DUT) from the input node and then output to the outside of the DUT via the output node. Therefore, during the transmission of the wave packet signal inside the DUT, the frequency components of the wave packet signal may be adjusted, for example, amplified or attenuated. This means that the multiple frequency components included in the wave packet signal undergo an overall change in the frequency domain due to the transmission process inside the DUT. Therefore, by inputting a wide-frequency-range wave packet signal containing multiple frequency components into the device under test (DUT), a corresponding output signal is obtained. This conversion process from input signal to output signal is influenced by the internal devices and circuits of the DUT, which adjust the frequency components in the signal. For example, it is affected by the internal signal compensation functions of the DUT (such as pre-emphasis, de-emphasis, equalization, etc.). Thus, by utilizing the changes between the input and output signals, digital processing algorithms can be used to deduce the information about the frequency response characteristics of the DUT carried by the changes between the input and output signals. Furthermore, a higher-order system transfer function matching the frequency response characteristics of the DUT can be derived for configuration adjustment, equipment calibration, and other purposes. Therefore, the wave packet signal corresponding to the input signal may adopt the same amplitude design at different frequency points, that is, the amplitudes of the multiple frequency components are the same, or it may adopt the different amplitude design at different frequency points, that is, the amplitudes of at least two of the multiple frequency components are inconsistent. This means that the amplitude distribution of the multiple frequency components in the wave packet signal is flexible. It is possible to adopt a design where the amplitudes of the frequency components are consistent or inconsistent, and ultimately it is still possible to achieve the design goal of deriving a high-order system transfer function that matches the frequency response characteristics of the device under test.Therefore, unlike the frequency scanning method that uses an input signal with a single frequency component and changes that single frequency component, this frequency scanning method generally uses a bit error rate tester (BERT) and a pseudo-random binary sequence (PRBS) to compare data and calculate the bit error rate. Therefore, signals of the same amplitude are generally used at different frequency points. In contrast, the measurement method for the frequency response characteristics of optoelectronic devices provided in the specific embodiments and implementation methods of this application uses a wave packet signal with a wide frequency range and multiple frequency components as the input signal. Signals of the same or different amplitudes can be used at different frequency points, which provides better flexibility in the composition of the input signal and is conducive to its widespread application.

[0100] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of optoelectronic devices, in one possible implementation, includes a test code pattern comprising a pseudo-random binary sequence. The input signal can be generated using a pseudo-random sequence such as a pseudo-random binary sequence (PRBS), or other random generation algorithms, to produce a binary code sequence of 0s and 1s with random characteristics. This means that the digital logic of the input signal has certain random characteristics, as long as it meets the requirements of a wide frequency range wave packet signal with multiple frequency components and the requirements of a test code pattern with consecutive identical digital segments of a preset segment length. The wave packet form is determined according to the specific pseudo-random code data format used by the algorithm that generates the input signal; for example, a constant 0 or constant 1 segment in that data format is used as a reference to adjust the amplitude. The advantage of this design is that, considering the irregular amplitude variation of the time-domain waveform of the input signal (assuming it can be actually acquired at the input node), and the irregular distribution of the peaks and troughs of the time-domain waveform of the input signal, the design requirements of the input signal code pattern, namely, the test code pattern with consecutive identical digital segments of a preset segment length, are utilized. For example, by utilizing the mechanism of consecutive identical digital segments of the pseudo-random binary sequence itself, multiple consecutive identical codewords may appear, such as nine consecutive constant 0 segments or 23 consecutive constant 1 segments. This can serve as a benchmark for waveform comparison and analysis. It should be noted that the input signal is a wave packet signal with a wide frequency range and multiple frequency components. Therefore, the test code pattern of consecutive identical digital segments with a preset segment length included in the wave packet signal corresponds to the low-frequency part of the multiple frequency components in the wave packet signal. Therefore, using the test code pattern of consecutive identical digital segments with a preset segment length as a reference is equivalent to using the signal of the low-frequency part of the wave packet signal to derive the equalization value of the device under test, such as deriving the gain ratio or compression ratio. In this way, the accuracy of the final calculation of the frequency response characteristics of the device under test can be improved by adjusting the amplitude.

[0101] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of optoelectronic devices, in one possible implementation, involves consecutive identical digital segments of a preset segment length corresponding to the low-frequency portion of the plurality of frequency components. The low-frequency portion of the plurality of frequency components corresponding to the consecutive identical digital segments of the preset segment length is used to calculate the gain ratio or compression ratio of the device under test (DUT). As mentioned above, by utilizing the change between the input and output signals, information about the frequency response characteristics of the DUT carried by the change between the input and output signals can be deduced through digital processing algorithms. This allows for the derivation of a higher-order system transfer function that matches the frequency response characteristics of the DUT. Specifically, by utilizing the key design requirement that the input signal is a wave packet signal with a wide frequency range and multiple frequency components, instead of comparing rise and fall times in the time domain, the overall change in rise and fall times is determined in the frequency domain. This allows the derivation of a higher-order system transfer function with minimal error, thereby characterizing the overall change in rise and fall times in the frequency domain. Therefore, compared to inputting a signal with only a single frequency component and then changing that single frequency component for frequency scanning, inputting a wave packet signal with a wide frequency range and multiple frequency components as the input signal to the device under test (DUT) allows for amplitude distribution of the multiple frequency components within the wave packet signal to follow a certain pattern. Furthermore, these multiple frequency components are pre-designed and do not change with the scanning mode. Here, the wave packet form of the wide frequency range wave packet signal with multiple frequency components is determined based on the data format of the input signal's code pattern. As mentioned above, requirements are imposed on the input signal's code pattern, limiting the wave packet signal to include a test code pattern, and stating that the test code pattern is a series of identical digital segments with a preset segment length. For example, a constant 0 segment or a constant 1 segment is a series of identical digital segments with a preset segment length. The input signal can be generated using a pseudo-random sequence, such as a pseudo-random binary sequence, or other random generation algorithms to produce a binary code sequence of 0s and 1s with random characteristics. The wave packet form is determined based on the specific pseudo-random code data format used by the algorithm that generates the input signal; for example, a constant 0 or constant 1 segment from that data format is used as a reference to adjust the amplitude. The advantage of this design is that, considering the irregular amplitude variation of the time-domain waveform of the input signal (assuming it can be actually acquired at the input node), and the irregular distribution of the peaks and troughs of the time-domain waveform of the input signal, the design requirements of the input signal code pattern, namely, the test code pattern with consecutive identical digital segments of a preset segment length, are utilized. For example, by utilizing the mechanism of consecutive identical digital segments of the pseudo-random binary sequence itself, multiple consecutive identical codewords may appear, such as nine consecutive constant 0 segments or 23 consecutive constant 1 segments. This can serve as a benchmark for waveform comparison and analysis.It should be noted that the input signal is a wave packet signal with a wide frequency range and multiple frequency components. Therefore, the test code pattern of consecutive identical digital segments with a preset segment length included in the wave packet signal corresponds to the low-frequency part of the multiple frequency components in the wave packet signal. Therefore, using the test code pattern of consecutive identical digital segments with a preset segment length as a reference is equivalent to using the signal of the low-frequency part of the wave packet signal to derive the equalization value of the device under test, such as deriving the gain ratio or compression ratio. In this way, the accuracy of the final calculation of the frequency response characteristics of the device under test can be improved by adjusting the amplitude.

[0102] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8 The second embodiment of the measurement device for the frequency response characteristics of optoelectronic devices, as shown, includes, in one possible implementation, an equalizer and a driver for the device under test (DUT). The equalizer can be a continuous-time linear equalizer (CTLE), a pre-emphasis equalizer, or a de-emphasis equalizer. The internal signal compensation method of the DUT can employ various techniques and can be tailored to whether the DUT is deployed at a signal transmitter or receiver. Examples include continuous-time linear equalizers (CTLE), feed-forward equalizers (FEE), decision-feedback equalizers (DFE), and, for instance, pre-emphasis and de-emphasis. In some embodiments, the driver is based on current-mode logic or low-voltage differential signals. Thus, the driver can employ various driving methods and differential signaling techniques, such as Current Mode Logic (CML) and Low Voltage Differential Signaling (LVDS), which helps to provide beneficial technical effects such as low noise and low power consumption.

[0103] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of an optoelectronic device, in one possible implementation, indicates that the frequency response characteristics of the device under test (DUT) indicate the equalization value and Nyquist frequency of the equalizer. Thus, by inputting an optimized input signal to the DUT, and assuming no significant change between the frequency range of the input signal and the frequency range of the output signal, the frequency domain transfer function can be analyzed using digital processing algorithms based on the simulated input waveform and the output waveform of the output signal. For example, the higher-order system transfer function corresponding to the frequency response characteristics of the DUT can be derived, the equalization value (e.g., gain ratio or compression ratio) can be calculated, and the peak frequency, equalization frequency, and Nyquist frequency can also be calculated.

[0104] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of an optoelectronic device, in one possible implementation, when the output signal is the second electrical signal, calculates the frequency response characteristics of the device under test based on the simulated input waveform and the waveform of the second electrical signal by a processor. This includes: performing differential processing on the first electrical signal and the second electrical signal using the simulated input waveform and the waveform of the second electrical signal to obtain a first differential electrical signal, wherein the differential mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the device under test. Thus, using a wide-frequency-range wave packet signal including multiple frequency components as the input signal, and utilizing the test code pattern included in the wave packet signal as continuous identical digital segments with a preset segment length, and through a simulated waveform generator, generating a simulated time-domain waveform corresponding to the digital logic of the input signal as the simulated input waveform of the input signal without actually acquiring the time-domain waveform of the input node, and through a time-domain waveform acquisition device, actually acquiring the time-domain waveform of the output node, obtaining the output waveform of the output signal at the output node of the device under test, and finally, using the simulated input waveform of the input signal and the output waveform of the output signal, performing frequency domain transfer function analysis based on digital processing algorithms, and deriving the higher-order system transfer function corresponding to the frequency response characteristics of the device under test. In this way, the frequency response characteristics of the equalizer are accurately obtained, effectively handling situations where the equalizer input is located inside the chip, such as obtaining the frequency response characteristics of an on-chip equalizer, without needing to deploy additional feedback loops and detection circuits into the chip, which is beneficial for chip miniaturization, high integration, and low power consumption. Furthermore, differential processing can better capture the rise time, fall time, and level changes of the time-domain waveform of the electrical signal. The differential mode component of the first differential electrical signal obtained by differential processing reflects the change of the second electrical signal relative to the first electrical signal. In this way, the frequency response characteristics of the device under test can be calculated using the information of the change, and the frequency domain transfer function can be analyzed based on digital processing algorithms.

[0105] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The measurement device for the frequency response characteristics of optoelectronic devices according to the second embodiment shown includes, in some embodiments, the method further comprising: inputting the first electrical signal into a plurality of system transfer functions to obtain a plurality of reference output electrical signals corresponding one-to-one with the plurality of system transfer functions; then, performing differential processing on the first electrical signal and the plurality of reference output electrical signals to obtain a plurality of reference differential electrical signals corresponding one-to-one with the plurality of reference output electrical signals; by comparing the differential mode components of the first differential electrical signal and the differential mode components of the plurality of reference differential electrical signals respectively, iterating using an error function, and selecting the system transfer function with the smallest error from the plurality of system transfer functions; the selected system transfer function with the smallest error is used to calculate the frequency response characteristics and system bandwidth of the device under test. Thus, the digital processing algorithm for deriving a high-order system transfer function matching the frequency response characteristics of the device under test can employ any suitable algorithm model and algorithm principle. For example, an error convergence algorithm can be used to select the system transfer function with the smallest error from multiple system transfer functions. Alternatively, the error function can be set as a function of the number of poles and zeros, and then iterative and adaptive algorithms can be used to filter out the system transfer function with the smallest error. Another example is the use of a digital fitting algorithm to approximate and calculate a suitable system transfer model. Here, multiple reference output electrical signals and corresponding reference differential electrical signals are obtained using multiple system transfer functions. Then, by comparing the differential mode components of the first differential electrical signal and the differential mode components of each of the multiple reference differential electrical signals, the error function is used for iteration to select the system transfer function with the smallest error from the multiple system transfer functions. In this way, a high-order system transfer function matching the frequency response characteristics of the device under test is efficiently determined through iterative and adaptive algorithms. The error function can be set as a function of the number of poles and zeros.

[0106] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of an optoelectronic device, in one possible implementation, involves the processor adjusting the output waveform of the output signal based on the simulated input waveform, when the output signal is the first optical signal, to convert the first optical signal into the third electrical signal. Then, the processor calculates the frequency response characteristics of the device under test (DUT) based on the simulated input waveform and the waveform of the third electrical signal. This includes: performing differential processing on the first electrical signal and the third electrical signal using the simulated input waveform and the waveform of the third electrical signal to obtain a second differential electrical signal. The differential-mode component of the second differential electrical signal is used to calculate the frequency response characteristics of the DUT. Since the output signal is the first optical signal, its output waveform is the time-domain waveform of the first optical signal. The simulated input waveform of the input signal is a simulated digital waveform generated based on the digital logic of the input signal; therefore, the simulated input waveform of the input signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal. Therefore, when the output signal is the first optical signal, the processor adjusts the output waveform of the output signal according to the simulated input waveform, thus converting the first optical signal into a third electrical signal. Then, based on the simulated input waveform and the waveform of the third electrical signal, the frequency response characteristics of the device under test (DUT) are calculated. Thus, based on digital processing algorithms, the frequency domain transfer function is analyzed, and the higher-order system transfer function corresponding to the frequency response characteristics of the DUT is derived. Furthermore, differential processing can better capture the rise time, fall time, and level changes of the electrical signal's time-domain waveform. The differential-mode component of the second differential electrical signal obtained through differential processing reflects the change of the third electrical signal relative to the first electrical signal. This change information can be used to calculate the frequency response characteristics of the DUT, and the frequency domain transfer function analysis is achieved based on digital processing algorithms.

[0107] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The measurement device for the frequency response characteristics of optoelectronic devices according to the second embodiment shown includes, in some embodiments, the method further comprising: inputting the first electrical signal into a plurality of system transfer functions to obtain a plurality of reference output electrical signals corresponding one-to-one with the plurality of system transfer functions; then, performing differential processing on the first electrical signal and the plurality of reference output electrical signals to obtain a plurality of reference differential electrical signals corresponding one-to-one with the plurality of reference output electrical signals; by comparing the differential mode components of the second differential electrical signal and the differential mode components of the plurality of reference differential electrical signals respectively, iterating using an error function, and selecting the system transfer function with the smallest error from the plurality of system transfer functions; the selected system transfer function with the smallest error is used to calculate the frequency response characteristics and system bandwidth of the device under test. Thus, the digital processing algorithm for deriving a high-order system transfer function matching the frequency response characteristics of the device under test can employ any suitable algorithm model and algorithm principle. For example, an error convergence algorithm can be used to select the system transfer function with the smallest error from multiple system transfer functions. Alternatively, the error function can be set as a function of the number of poles and zeros, and then iterative and adaptive algorithms can be used to filter out the system transfer function with the smallest error. Another example is the use of a digital fitting algorithm to approximate and calculate a suitable system transfer model. Here, multiple reference output electrical signals and corresponding reference differential electrical signals are obtained using multiple system transfer functions. Then, by comparing the differential mode components of the first differential electrical signal and the differential mode components of each of the multiple reference differential electrical signals, the error function is used for iteration to select the system transfer function with the smallest error from the multiple system transfer functions. In this way, a high-order system transfer function matching the frequency response characteristics of the device under test is efficiently determined through iterative and adaptive algorithms. The error function can be set as a function of the number of poles and zeros.

[0108] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The measurement device for the frequency response characteristics of optoelectronic devices in the second embodiment shown, in some examples, uses an error function that is a function of the number of poles and zeros of the system transfer function. The model characteristics of each of the multiple system transfer functions include delay alignment time, rise time, fall time, and amplitude variation. Considering that the input signal is limited to a wave packet signal with a wide frequency range and multiple frequency components, this means that, from a time-domain analysis perspective, the time-domain waveform of the input signal (assuming it can be actually acquired at the input node) has irregular amplitude variations, and the distribution of peaks and troughs in the time-domain waveform may also be irregular. Therefore, the digital processing algorithm used to derive a higher-order system transfer function that matches the frequency response characteristics of the device under test does not compare rise and fall times in the time domain, but rather determines the overall variation of rise and fall times in the frequency domain. That is, by utilizing the key design requirement that the input signal is a wave packet signal with a wide frequency range and multiple frequency components, the algorithm derives a higher-order system transfer function with the minimum error, thereby characterizing the overall variation of rise and fall times in the frequency domain. In addition, digital processing algorithms used to derive high-order system transfer functions that match the frequency response characteristics of the device under test can take into account not only the changes in rise time and fall time, but also other factors in the changes between the input and output signals, thereby improving the calculation accuracy, such as amplitude changes and delay alignment.

[0109] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of an optoelectronic device, in some embodiments, when the output signal is the first optical signal, performs a waveform adjustment operation on the output waveform of the output signal according to the simulated input waveform by the processor to convert the first optical signal into the third electrical signal. This includes: determining the average value of the waveform corresponding to the first optical signal; then subtracting the average value of the waveform corresponding to the first optical signal from the average value of the waveform corresponding to the first optical signal; and then performing a waveform alignment operation in the waveform adjustment operation according to the waveform corresponding to the first electrical signal. The device under test may include an optoelectronic device, such as a device that converts an electrical signal into an optical signal, like a light emitting device. When the device under test includes an optoelectronic device, it means that the output signal is the first optical signal. Therefore, the output waveform of the output signal is the time-domain waveform of the first optical signal, while the simulated input waveform of the input signal is a simulated digital waveform generated based on the digital logic of the input signal. Therefore, the simulated input waveform of the input signal cannot be directly analyzed and compared with the time-domain waveform of the first optical signal. Using the processor's internal algorithms, a first optical signal can be converted into a third electrical signal through waveform adjustment operations. For example, waveform alignment can be achieved by removing the DC component. The frequency response characteristics of the device under test (DUT) are then calculated using the changes between the first and third electrical signals. Therefore, this approach can be applied to scenarios requiring a large number of optoelectronic devices, such as optical interconnects in data centers. For instance, the amplitude range of the waveform corresponding to the first electrical signal is from -100 millivolts (mV) to 100 mV, while the amplitude range of the waveform corresponding to the first optical signal is from 300 mW to 500 mW. To perform waveform alignment, the average values ​​of the waveforms can be aligned, i.e., the corresponding average values ​​are subtracted from each of the two original signals. Specifically, the average value of the waveform corresponding to the first optical signal is determined, and then the average value of the waveform corresponding to the first optical signal is subtracted from the average value of the waveform corresponding to the first optical signal. Then, the waveform alignment operation in the waveform adjustment operation is performed based on the waveform corresponding to the first electrical signal. Thus, through waveform adjustment and alignment, and by analyzing the frequency domain transfer function based on digital processing algorithms, the higher-order system transfer function corresponding to the frequency response characteristics of the DUT is derived. In this way, the frequency response characteristics of the equalizer can be accurately obtained, which can effectively deal with the situation where the input of the equalizer is located inside the chip, such as obtaining the frequency response characteristics of the on-chip equalizer. Moreover, there is no need to deploy additional feedback loops and detection circuits into the chip, which is beneficial to the miniaturization design, high integration and low power consumption of the chip.

[0110] See Figure 7 and Figure 8 ,based on Figure 7The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8 The second embodiment of the measurement device for the frequency response characteristics of an optoelectronic device, as shown, in some examples, has the wave packet form of the wave packet signal determined according to the data format of the test code pattern. The consecutive identical digital segments serve as the reference for amplitude adjustment in the waveform adjustment operation, and the consecutive identical digital segments are either constant 0 segments or constant 1 segments. The wave packet form of a wide frequency range wave packet signal with multiple frequency components is determined according to the data format of the input signal's code pattern. As mentioned above, requirements are imposed on the code pattern of the input signal, limiting the wave packet signal to include a test code pattern, and the test code pattern is a series of consecutive identical digital segments with a preset segment length. For example, constant 0 segments or constant 1 segments are consecutive identical digital segments with a preset segment length. The input signal can be generated using a pseudo-random sequence, such as a pseudo-random binary sequence, or other random generation algorithms to generate a binary code sequence of 0s and 1s with random characteristics. The wave packet form is determined according to the specific pseudo-random code data format used by the algorithm that generates the input signal, for example, using a constant 0 or constant 1 segment in that data format as a reference for amplitude adjustment. The advantage of this design is that, considering the irregular amplitude variation of the time-domain waveform of the input signal (assuming it can be actually acquired at the input node), and the irregular distribution of the peaks and troughs of the time-domain waveform of the input signal, the design requirements of the input signal code pattern, namely, the test code pattern with consecutive identical digital segments of a preset segment length, are utilized. For example, by utilizing the mechanism of consecutive identical digital segments of the pseudo-random binary sequence itself, multiple consecutive identical codewords may appear, such as nine consecutive constant 0 segments or 23 consecutive constant 1 segments. This can serve as a benchmark for waveform comparison and analysis. It should be noted that the input signal is a wave packet signal with a wide frequency range and multiple frequency components. Therefore, the test code pattern of consecutive identical digital segments with a preset segment length included in the wave packet signal corresponds to the low-frequency part of the multiple frequency components in the wave packet signal. Therefore, using the test code pattern of consecutive identical digital segments with a preset segment length as a reference is equivalent to using the signal of the low-frequency part of the wave packet signal to derive the equalization value of the device under test, such as deriving the gain ratio or compression ratio. In this way, the accuracy of the final calculation of the frequency response characteristics of the device under test can be improved by adjusting the amplitude.

[0111] See Figure 7 and Figure 8 ,based on Figure 7 The second embodiment shown illustrates a method for measuring the frequency response characteristics of optoelectronic devices, and based on... Figure 8The second embodiment of the measurement device for the frequency response characteristics of optoelectronic devices, as shown, in one possible implementation, involves the device under test (DUT) being deployed at the receiving end, with the input node corresponding to the output node of the recovered signal; or, the DUT being deployed at the transmitting end, with the input node corresponding to the output node of the transmitted signal sequence. The specific details of the front-end and back-end circuits of the DUT can be flexibly adjusted according to actual needs. When the DUT is deployed at the receiving end, the front-end circuit can be a clock data recovery circuit within the chip, and the input node of the DUT can be the node where the clock data recovery circuit outputs the recovered signal. When the DUT is deployed at the transmitting end, the front-end circuit can be a signal sequence generator, and the input node of the DUT can be the output node of the transmitted signal sequence generated by the signal sequence generator.

[0112] Figure 9 This is a schematic diagram of a computing device 900 provided in an embodiment of this application. The computing device 900 includes one or more processors E910, a communication interface 920, and a memory 930. The processors E910, the communication interface 920, and the memory 930 are interconnected via a bus 940. Optionally, the computing device 900 may further include an input / output interface 950, which is connected to input / output devices for receiving user-set parameters, etc. The computing device 900 can be used to implement some or all of the functions of the device embodiment or system embodiment in the above-described embodiments of this application; the processor E910 can also be used to implement some or all of the operation steps of the method embodiment in the above-described embodiments of this application. For example, the specific implementation of various operations performed by the computing device 900 can be referred to the specific details in the above embodiments, such as the processor E910 being used to execute some or all of the steps or operations in the above-described method embodiments. For example, in the embodiments of this application, the computing device 900 can be used to implement some or all of the functions of one or more components in the above-described device embodiments. In addition, the communication interface 920 can be used for communication functions necessary to implement the functions of these devices and components, and the processor E910 can be used for processing functions necessary to implement the functions of these devices and components.

[0113] It should be understood that, Figure 9 The computing device 900 may include one or more processors E910, and the multiple processors E910 may collaboratively provide processing power in a parallel connection, a serial connection, a serial-parallel connection, or an arbitrary connection manner; or the multiple processors E910 may form a processor sequence or a processor array; or the multiple processors E910 may be divided into a main processor and an auxiliary processor; or the multiple processors E910 may have different architectures, such as adopting a heterogeneous computing architecture. Furthermore, Figure 9 The structural and functional descriptions of the computing device 900 shown are exemplary and non-limiting. In some exemplary embodiments, the computing device 900 may include... Figure 9 The diagram shows more or fewer components, or combinations of some components, or splitting of some components, or different arrangements of components.

[0114] The processor E910 can have various specific implementations. For example, it can include one or more combinations of a central processing unit (CPU), a graphics processing unit (GPU), a neural network processing unit (NPU), a tensor processing unit (TPU), or a data processing unit (DPU). This application does not impose specific limitations on these embodiments. The processor E910 can also be a single-core or multi-core processor. The processor E910 can be a combination of a CPU and hardware chips. These hardware chips can be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. The PLDs can be complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), generic array logic (GALs), or any combination thereof. The processor E910 can also be implemented using logic devices with built-in processing logic, such as FPGAs or digital signal processors (DSPs). The communication interface 920 can be a wired interface or a wireless interface, used to communicate with other modules or devices. The wired interface can be an Ethernet interface, a local interconnect network (LIN), etc., while the wireless interface can be a cellular network interface or a wireless LAN interface, etc.

[0115] Memory 930 may be non-volatile memory, such as read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Memory 930 may also be volatile memory, which may be random access memory (RAM) used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM). The memory 930 can also be used to store program code and data, so that the processor E910 can call the program code stored in the memory 930 to execute some or all of the operation steps in the above method embodiments, or to execute the corresponding functions in the above device embodiments. Furthermore, the computing device 900 may include, compared to... Figure 9 The number of components displayed may be more or less, or there may be different component configurations.

[0116] Bus 940 can be a Peripheral Component Interconnect Express (PCIe) bus, or an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (Ubus or UB), a Compute Express Link (CXL) bus, a Cache Coherent Interconnect for Accelerators (CCIX) bus, etc. Bus 940 can be divided into address bus, data bus, control bus, etc. In addition to the data bus, bus 940 can also include a power bus, control bus, and status signal bus. However, for clarity, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0117] The methods and devices provided in this application are based on the same inventive concept. Since the principles by which the methods and devices solve problems are similar, the embodiments, implementation methods, examples, or methods of implementation of the methods and devices can be referred to each other, and repeated details will not be repeated. This application also provides a system comprising multiple computing devices, the structure of each computing device of which can refer to the structure of the computing devices described above. The functions or operations achievable by this system can refer to the specific implementation steps in the above method embodiments and / or the specific functions described in the above device embodiments, and will not be repeated here.

[0118] This application also provides a computer-readable storage medium storing computer instructions. When these computer instructions are executed on a computer device (such as one or more processors), they can implement the method steps described in the above method embodiments. The specific implementation of the above method steps by the processor of the computer-readable storage medium can refer to the specific operations described in the above method embodiments and / or the specific functions described in the above device embodiments, and will not be repeated here.

[0119] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. This application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Embodiments of this application can be implemented wholly or partially by software, hardware, firmware, or any other combination. When implemented in software, the above embodiments can be implemented wholly or partially as a computer program product. This application can take the form of a computer program product embodied on one or more computer-usable storage media containing computer-usable program code. The computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer-readable storage media can be any available medium that a computer can access, or a data storage device such as a server or data center that contains one or more sets of available media. Available media can be magnetic media (such as floppy disks, hard disks, and magnetic tapes), optical media, or semiconductor media. Semiconductor media can be solid-state drives, random access memory, flash memory, read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, or any other suitable form of storage medium.

[0120] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. Each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0121] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not described in detail in a certain embodiment can be referred to in the relevant descriptions of other embodiments. Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of the embodiments of this application. The steps in the methods of the embodiments of this application can be adjusted in order, combined, or deleted according to actual needs; the modules in the systems of the embodiments of this application can be divided, combined, or deleted according to actual needs. If these modifications and variations of the embodiments of this application fall within the scope of the claims of this application and their equivalents, then this application also intends to include these modifications and variations.

Claims

1. A method for measuring the frequency response characteristics of optoelectronic devices, characterized in that, The measurement method includes: The first acquisition device obtains the input waveform of the input signal at the input node of the device under test, and the second acquisition device obtains the output waveform of the output signal at the output node of the device under test. The input signal includes multiple frequency components, and the device under test has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal and the other is the waveform corresponding to the first optical signal. The processor performs waveform adjustment on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal. It also performs differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal. The waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the device under test.

2. The measurement method according to claim 1, characterized in that, The input signal is a data signal that includes the multiple frequency components, and the amplitude distribution of each of the multiple frequency components conforms to a preset pattern.

3. The measurement method according to claim 2, characterized in that, The preset mode indicates that the amplitudes of at least two of the plurality of frequency components are inconsistent.

4. The measurement method according to claim 2, characterized in that, The preset mode indicates that the amplitudes of the plurality of frequency components are the same.

5. The measurement method according to claim 2, characterized in that, The input signal is a wave packet signal with a wide frequency range.

6. The measurement method according to claim 2, characterized in that, The data signal includes a test code pattern, which is a set of identical codewords with a preset segment length.

7. The measurement method according to claim 6, characterized in that, The test code pattern includes a pseudo-random binary sequence.

8. The measurement method according to claim 6, characterized in that, The identical codewords with a preset segment length correspond to the low-frequency portion of the plurality of frequency components and are used to provide a reference for the waveform adjustment operation.

9. The measurement method according to claim 8, characterized in that, The low-frequency components of the plurality of frequency components corresponding to the same codeword with a preset segment length are used to calculate the gain ratio or compression ratio of the device under test.

10. The measurement method according to claim 1, characterized in that, The measurement method further includes: The first electrical signal is input into multiple system transfer functions to obtain multiple reference output electrical signals that correspond one-to-one with the multiple system transfer functions. Then, the first electrical signal and the multiple reference output electrical signals are differentially processed to obtain multiple reference differential electrical signals that correspond one-to-one with the multiple reference output electrical signals. By comparing the differential-mode components of the first differential electrical signal and the differential-mode components of the plurality of reference differential electrical signals respectively, and using an error function for iteration, the system transfer function with the smallest error is selected from the plurality of system transfer functions. The selected system transfer function with the smallest error is used to determine the frequency response characteristics and system bandwidth of the device under test.

11. The measurement method according to claim 10, characterized in that, The error function is a function of the number of poles and zeros of the system transfer function, and the model characteristics of each of the multiple system transfer functions include the delay alignment time, rise time, fall time, and amplitude variation.

12. The measurement method according to claim 1, characterized in that, The processor performs waveform adjustment on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into the second electrical signal, including: The average value of the waveform corresponding to the first optical signal is determined. Then, the average value of the waveform corresponding to the first optical signal is subtracted from the average value of the waveform corresponding to the first optical signal. Then, the waveform alignment operation in the waveform adjustment operation is performed according to the waveform corresponding to the first electrical signal.

13. The measurement method according to claim 12, characterized in that, The input signal is a wave packet signal with a wide frequency range that includes the multiple frequency components. The wave packet form of the wave packet signal is determined according to the data format of the pseudo-random code generated by the data generator. The consecutive identical digital segments in the data format are the reference for the amplitude adjustment operation in the waveform adjustment operation. The consecutive identical digital segments are either constant 0 segments or constant 1 segments.

14. The measurement method according to claim 13, characterized in that, The differential-mode component of the first differential electrical signal is used to determine the system transfer function with the minimum error. The system transfer function with the minimum error characterizes the overall change in rise time and fall time of the wave packet signal in the frequency domain after passing through the device under test.

15. The measurement method according to claim 1, characterized in that, The device under test has an electro-optic conversion function. The input signal is the first electrical signal, the output signal is the first optical signal, the input waveform is the waveform corresponding to the first electrical signal, and the output waveform is the waveform corresponding to the first optical signal. The device under test includes a light emitting device.

16. The measurement method according to claim 1, characterized in that, The device under test has photoelectric conversion function, the input signal is the first optical signal, the output signal is the first electrical signal, the input waveform is the waveform corresponding to the first optical signal, the output waveform is the waveform corresponding to the first electrical signal, and the device under test includes a photodetector.

17. The measurement method according to claim 1, characterized in that, The first collector is different from the second collector, or the first collector and the second collector are the same collector with the ability to acquire electrical signal time-domain waveforms and optical signal time-domain waveforms.

18. The measurement method according to claim 1, characterized in that, The device under test is any one of a plurality of devices to be calibrated. The input signals of the plurality of devices to be calibrated each come from the same signal generator. The input signals and output signals of the plurality of devices to be calibrated are used to determine the frequency response characteristics of the plurality of devices to be calibrated in order to achieve parallel calibration of the plurality of devices to be calibrated.

19. A measuring device for the frequency response characteristics of optoelectronic devices, characterized in that, The measuring device includes: The first data acquisition unit is used to obtain the input waveform of the input signal at the input node of the device under test; The second data acquisition unit is used to obtain the output waveform of the output signal at the output node of the device under test. The input signal includes multiple frequency components, and the device under test has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal and the other is the waveform corresponding to the first optical signal. The processor is configured to perform waveform adjustment operations on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal, and to perform differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal, wherein the waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to calculate the frequency response characteristics of the device under test.

20. A fault detection method for optoelectronic devices, characterized in that, The fault detection method includes: The first acquisition device obtains the input waveform of the input signal at the input node of the device under test, and the second acquisition device obtains the output waveform of the output signal at the output node of the device under test. The input signal includes multiple frequency components, and the device under test has an electro-optical conversion function or a photoelectric conversion function. One of the input signal and the output signal is a first electrical signal and the other is a first optical signal. One of the input waveform and the output waveform is the waveform corresponding to the first electrical signal and the other is the waveform corresponding to the first optical signal. The processor performs waveform adjustment on the waveform corresponding to the first optical signal based on the waveform corresponding to the first electrical signal, so as to convert the first optical signal into a second electrical signal. It also performs differential processing on the first electrical signal and the second electrical signal to obtain a first differential electrical signal. The waveform adjustment operation is used to reduce the common-mode component of the first differential electrical signal, and the differential-mode component of the first differential electrical signal is used to measure and calibrate the frequency response characteristics of the device under test in real time for fault detection of the device under test.

Images