A method for optimizing network parameters and locating faults of an optical module based on PRBS detection

Abstract

The application relates to a PRBS detection-based optical module network parameter optimization and fault positioning method, which comprises the following steps: generating a PRBS sequence in a tester FPGA and sending the PRBS sequence via an optical module interface; meanwhile, the PRBS sequence is looped back to a receiving end through a tester optical module outer ring or a to-be-tested device outer ring; the transmission parameters of the optical module are modified through a management bus of the optical module; the bit error rate of the PRBS sequence returned by the optical module under different transmission parameter combinations is measured; whether there is a parameter combination meeting the Ethernet standard is judged; if there is a parameter combination meeting the Ethernet standard, the optimal configuration is determined based on the bit error rate; the optimal configuration is written into the optical module; if the bit error rates corresponding to all the parameter combinations do not meet the Ethernet standard after all the parameter combinations are traversed, it is determined that the optical module interface has a fault, the optical module interface is identified, and a fault indication is triggered.

Description

Technical Field

[0001] This application relates to the technical field of network equipment, and in particular to a method for optimizing on-network parameters and locating faults in optical modules based on PRBS detection. Background Technology

[0002] In large-scale network equipment testing scenarios, a large number of optical modules and cables are typically used for networking. Due to factors such as differences in production batches, wear and tear during use, and storage environment, these physical materials often have implicit physical reliability issues such as signal attenuation, inter-symbol interference, and eye diagram closure. These physical obstacles can lead to packet loss and packet errors in the underlying link communication, thus seriously affecting the overall testing progress.

[0003] Because optical modules do not support auto-negotiation and link training in their underlying physical design, their underlying physical faults are difficult for the system to directly identify. Existing technologies typically employ methods such as... Figure 1 The dedicated L1 layer tester shown (such as the VIAVIONT-800 series) sends test signals to the optical module and measures the bit error rate in offline mode, thereby performing material screening.

[0004] This technical solution faces significant technical obstacles: dedicated L1 equipment is expensive and cumbersome to operate, and the offline testing process makes it impossible to synchronize material screening with network testing, resulting in an excessively long testing cycle. This technical problem urgently needs to be solved. Summary of the Invention

[0005] To address the aforementioned technical issues, this application provides a method for optimizing on-network parameters and locating faults in optical modules based on PRBS detection.

[0006] This application provides a method for optimizing on-network parameters and locating faults in optical modules based on PRBS detection, which adopts the following technical solution: A PRBS sequence is generated in the FPGA of the test instrument and transmitted via the optical module interface; at the same time, the PRBS sequence is looped back to the receiving end through the outer loop of the optical module of the test instrument or the outer loop of the device under test. Based on the optical module type and CMIS standard, the transmission parameters of the optical module are modified by traversing the management bus of the optical module. Measure the bit error rate of the PRBS sequence transmitted back by the optical module under different combinations of transmission parameters to determine whether there is a parameter combination that meets the Ethernet standard; If a parameter combination that satisfies the Ethernet standard exists, the optimal configuration is determined based on the bit error rate. Write the optimal configuration into the optical module; If the bit error rate corresponding to all the parameter combinations does not meet the Ethernet standard, then the optical module interface is determined to be faulty, the optical module interface is identified, and a fault indication is triggered.

[0007] Preferably, determining the optimal configuration based on the bit error rate specifically includes: Select the parameter combination with the lowest bit error rate as the optimal configuration; If there are multiple parameter combinations that satisfy the Ethernet standard, then the median value is taken as the optimal configuration.

[0008] The PRBS sequence is looped back to the receiving end via the outer loop of the optical module of the tester or the outer loop of the device under test, including any of the following loopback steps: Inside the optical module of the tester, the transmitted signal is physically looped back directly to the receiving port; or, The device under test loops its own transmitted signal back to the receiving port to establish an outer loop that supports the PMA function.

[0009] Preferably, the transmission parameters of the optical module include at least one of pre-emphasis, deemphasis, and output amplitude; The step of modifying the transmission parameters of the optical module through the management bus of the optical module includes modifying the transmission end configuration of the optical module according to the CMIS standard via the I2C bus.

[0010] Preferably, the fault indication is triggered by at least one method, such as a constant indicator light on the tester panel, a pop-up notification in the host computer software, or a fault information output via serial port; the physical number of the optical module interface and the corresponding optical module information are clearly marked in the fault indication.

[0011] Preferably, the PRBS sequence generated in the test instrument FPGA is a PRBS31 sequence or a test sequence specified by other Ethernet standards.

[0012] Preferably, in the step of traversing and modifying the transmission parameters of the optical module, the optimal configuration is quickly approximated by partial sampling of the transmission parameters instead of exhaustive traversal; and if a parameter combination that meets the Ethernet standard cannot be found after sampling traversal, a fault identifier and indication for the optical module interface are triggered.

[0013] Preferably, the step of measuring the bit error rate of the PRBS sequence transmitted back by the optical module under different combinations of transmission parameters further includes extracting eye diagram opening and jitter as evaluation indicators of physical dimensions; selecting the transmission parameter combination with the best comprehensive performance index by weighted summation of the bit error rate, eye diagram opening and jitter; and determining the optical module interface fault and giving an indication when the comprehensive index of all parameter combinations does not meet the standard.

[0014] In summary, this application includes at least one of the following beneficial technical effects: 1. It breaks the hardware dependence on expensive dedicated L1 layer offline testers, and completes online packet sending and testing by reusing the original FPGA underlying logic processing capabilities of the system's testers, thereby reducing equipment costs and shortening the test cycle.

[0015] 2. By automatically scanning and dynamically applying optimal pre-emphasis and other waveform parameters, the problem of high-frequency signal attenuation and inter-symbol interference in the physical transmission medium is solved, effectively eliminating hidden faults that occur when multiple materials are combined, and enhancing the physical stability of the underlying data link in complex electromagnetic environments.

[0016] 3. The built-in deterministic fault diagnosis logic can physically locate and explicitly alarm the failed interface when parameter traversal optimization fails, thereby transforming the traditional blind plug-and-play test into rapid physical component replacement under system guidance, which greatly reduces the system's operation and maintenance delay and the cognitive burden on operators. Attached Figure Description Figure 1 This is a block diagram illustrating the offline testing principle of the optical module in a traditional L1 layer tester.

[0017] Figure 2 This is a diagram illustrating a large-scale Ethernet device networking test application scenario.

[0018] Figure 3 This is a schematic diagram of the system architecture and loopback point location in an embodiment of this application.

[0019] Figure 4 This is a flowchart of one specific implementation of this application. Detailed Implementation

[0020] The specific embodiments of the present invention will be described in detail below, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0021] The following is in conjunction with the appendix Figure 1-4 The present application will be further described with reference to specific embodiments.

[0022] The testing system of this invention mainly consists of a device under test (DUT), a tester (with a built-in optical module and FPGA for logic processing), and a transmission medium (such as cable or optical fiber) forming a closed-loop network environment. In large-scale Ethernet network testing scenarios, the underlying optical modules do not inherently support auto-negotiation (AN) and link training (LT) mechanisms, causing upper-layer protocols to be unable to detect and repair physical layer attenuation obstacles. Therefore, this invention achieves adaptive optimization and fault locking of the underlying hardware performance without disrupting or interfering with the existing real network topology by embedding detection and control operators in the FPGA of the tester.

[0023] The following is a detailed description of the core business steps of this invention: Step S1: Generate a PRBS sequence in the FPGA of the test instrument and send the PRBS sequence via the optical module interface; at the same time, loop the PRBS sequence back to the receiving end through the outer loop of the optical module of the test instrument or the outer loop of the device under test; Specifically, since the optical module lacks active detection capability, a standard test sequence, such as PRBS31, is actively generated by the FPGA deployed within the test instrument. The test instrument converts this digital sequence into a high-speed photoelectric signal, which is then transmitted externally via the optical module interface, which serves as the service data channel. (Refer to...) Figure 3 To ensure the transmitted signal can be read back and verified, the system is configured with two highly fault-tolerant physical loopback paths. The first is the optical module outer loop (loopback point 1), where the transmitted signal (Tx) is directly folded back to its receiving port (Rx) within the optical module hardware of the test instrument before transmission. The second is the device under test outer loop (loopback point 2), which utilizes the Physical Media Access Layer (PMA) function of the peer device under test. Following standards such as IEEE 802.3ba, standard Ethernet devices typically support PMA outer loops, allowing the received physical signal to be transmitted back along the original path. By establishing these dual-path loopbacks, the test system can acquire physical signals that have undergone complete media transmission loss in a real network environment.

[0024] Step S2: Based on the optical module type and CMIS standard, modify the transmission parameters of the optical module by traversing the management bus of the optical module.

[0025] Specifically, the management bus used to issue control commands is an I2C interface; the transmission parameters correspond to the key waveform control variables of the optical / electrical signals transmitted by the optical module, including pre-emphasis, deemphasis, and output amplitude; discrete register write commands conforming to the CMIS standard are issued through the I2C bus.

[0026] Step S3: Measure the bit error rate of the PRBS sequence transmitted back by the optical module under different combinations of transmission parameters, and determine whether there is a parameter combination that meets the Ethernet standard, wherein the Ethernet standard is a bit error rate of less than ; Specifically, the bit error rate (BER) is measured by the check operator inside the FPGA, which performs a bit-by-bit hardware XOR comparison between the locally generated original sequence and the sequence transmitted back via a damaged external link, and then calculates the BER value for each set of values ​​in real time. A comparison is then performed. In some embodiments, the system further includes extracting eye opening and jitter as evaluation metrics, and calculating the overall performance through weighted summation.

[0027] Step S4: If there is a parameter combination that satisfies the Ethernet standard, determine the optimal configuration based on the bit error rate; and write the optimal configuration into the optical module.

[0028] Specifically, the engineering rules for determining the optimal configuration are set as follows: if, after traversal, only one set of transmission parameters makes the bit error rate meet the standard, or minimizes the comprehensive loss function in the aforementioned implementation method, then that set of parameters is directly selected; if there are multiple sets of discrete parameters that meet the standard, then the intermediate value of these parameters is calculated and extracted as the final configuration; the significance of selecting the intermediate value is that it drives the electrical signal waveform of the optical module at the physical center of the eye diagram opening, so that it is as far away as possible from the physical failure critical point caused by temperature drift or voltage fluctuation, thereby reserving the most sufficient signal-to-noise ratio margin for the system; the optimal parameter set is solidified into the optical module register through the I2C management bus, which completes the missing link training (LT) process of the optical module and improves the stability of large-scale network testing.

[0029] Step S5: If the bit error rate corresponding to all the parameter combinations does not meet the Ethernet standard, then the optical module interface is determined to be faulty, the optical module interface is identified, and a fault indication is triggered.

[0030] Specifically: If, after exhausting all parameter traversal or reaching the maximum number of iterations through two-stage optimization sampling, the detected bit error rate still cannot be lower than [a certain value], then [the following applies]. If the threshold or loss function fails to converge to the safe domain, the physical number of the optical module interface that caused the current communication failure and the information of the optical module mounted on that interface will be extracted; a fault indicator will be triggered, the LED indicator of the corresponding channel on the front panel of the tester will be constantly lit in red, and the host computer test software will pop up a conspicuous UI warning, such as a pop-up window clearly stating: Interface XX is faulty, PRBS bit error rate does not meet the requirements, please replace the optical module or transmission medium, and the corresponding fault log will be asynchronously output on the underlying debugging serial port.

[0031] After the operator physically replaces the faulty optical module or transmission cable according to the above hardware physical location replacement guidelines, the system can be retried to execute the automatic detection and parameter optimization process of this invention again. This solves the inefficient mode of blindly replacing cables and trying different solutions in traditional testing, avoids testing delays, and reduces the cognitive burden on operators.

[0032] In some embodiments, local sampling is performed using gradient descent or Bayesian optimization algorithms instead of full exhaustive search to quickly approximate the optimal waveform configuration domain.

[0033] As another implementation method: To address the conflicts between the inability to divide and rounding oscillations caused by conventional continuous algorithms when migrating to the discrete CMIS register space, the tendency of local sampling to get trapped in local optima, and the excessive testing time caused by exhaustive search, this scheme also performs the following steps: A three-dimensional discrete search space for the transmission parameters is established and initialized with a step value greater than the minimum hardware resolution; the three-dimensional discrete search space covers pre-emphasis, deemphasis, and output amplitude; The first-level sparse jump test is performed within the pre-emphasis interval. The first-round bit error rate of the current test point is sampled and obtained by issuing the configuration command of the step value. Based on the test results of adjacent sampling points, the slope gradient of the bit error rate decrease in discrete space is calculated; the register step value of the next round is updated according to the slope gradient: when the slope gradient approaches zero and is in the high bit error rate region, it is determined to be an invalid pre-emphasis interval, and the step value is increased exponentially to quickly skip it; When the slope gradient changes drastically and crosses the preset bit error rate threshold, it is determined that it is approaching the error-free zone, and then the boundary of the safe polygon composed of multiple sets of discrete coordinates is defined. After entering the safe polygon boundary, the step value is converged to the minimum physical resolution of the CMIS register; Within the boundary of the secure polygon, a minimum step size is used to traverse and locate the discrete coordinates that result in the lowest bit error rate, which are then used as the final optimal configuration parameters.

[0034] As another implementation method, in order to solve the problems caused by heat accumulation in optical modules and fiber aging in complex networks, this solution also performs the following logical steps: During the same frequency cycle of PRBS bit error detection, DDMI polling commands are triggered in parallel to collect physical environment reference data of the optical module in real time; the physical environment reference data includes the internal real-time temperature and real-time received optical power; The eye diagram opening and jitter of the current link are extracted using FPGA; Calculate the degradation gradient of underlying environmental physical quantities. Based on the DDMI sampling sequence within a continuous-time sliding window, calculate the temperature drift gradient and optical attenuation gradient to characterize the degradation trend of the physical channel due to thermodynamic and optical interference; Using the logarithmic function Log10(BER) The significant difference in bit error rate at each level converges to a linear evaluation space; the discrete jitter is smoothly mapped to a penalty coefficient of 0 to 1 using a nonlinear Sigmoid function; A dynamic loss function integrating environmental gradients is constructed. The temperature drift gradient and optical attenuation gradient are weighted and summed with the normalized bit error rate, jitter penalty coefficient, and eye diagram opening. When the junction temperature of the optical module continues to rise or the received optical power continues to decrease, but the bit error has not yet occurred, the loss function can detect and amplify the penalty value in advance. During the traversal and optimization process, the combination of transmission parameters that minimizes the dynamic synthesis loss function is selected, and a compensation mapping table is established and updated in system memory. The above-disclosed embodiments are merely a few specific examples of the present invention. However, the embodiments of the present invention are not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for optimizing on-network parameters and locating faults in optical modules based on PRBS detection, characterized in that, include: A PRBS sequence is generated in the FPGA of the test instrument and transmitted via the optical module interface; at the same time, the PRBS sequence is looped back to the receiving end through the outer loop of the optical module of the test instrument or the outer loop of the device under test. Based on the optical module type and CMIS standard, the transmission parameters of the optical module are modified by traversing the management bus of the optical module. Measure the bit error rate of the PRBS sequence transmitted back by the optical module under different combinations of transmission parameters to determine whether there is a parameter combination that meets the Ethernet standard; If a parameter combination that satisfies the Ethernet standard exists, the optimal configuration is determined based on the bit error rate; and the optimal configuration is written into the optical module. If the bit error rate corresponding to all the parameter combinations does not meet the Ethernet standard, then the optical module interface is determined to be faulty, the optical module interface is identified, and a fault indication is triggered.

2. The method for optimizing on-network parameters and locating faults of optical modules based on PRBS detection as described in claim 1, characterized in that, Determining the optimal configuration based on the bit error rate specifically includes: Select the parameter combination with the lowest bit error rate as the optimal configuration; If there are multiple parameter combinations that satisfy the Ethernet standard, then the median value is taken as the optimal configuration.

3. The method for optimizing on-network parameters and locating faults of optical modules based on PRBS detection as described in claim 2, characterized in that, The PRBS sequence is looped back to the receiving end via the outer loop of the optical module of the tester or the outer loop of the device under test, including any of the following loopback steps: Inside the optical module of the tester, the transmitted signal is physically looped back directly to the receiving port; or, The device under test loops its own transmitted signal back to the receiving port to establish an outer loop that supports the PMA function.

4. The method for optimizing on-network parameters and locating faults of optical modules based on PRBS detection as described in claim 3, characterized in that, The transmission parameters of the optical module include at least one of pre-emphasis, deemphasis, and output amplitude; The step of modifying the transmission parameters of the optical module through the management bus of the optical module includes modifying the transmission end configuration of the optical module according to the CMIS standard via the I2C bus.

5. The method for optimizing on-network parameters and locating faults of optical modules based on PRBS detection as described in claim 4, characterized in that, The fault indication is triggered by at least one of the following methods: the indicator light on the tester panel is constantly on, the upper computer software pop-up prompts, or the serial port outputs fault information; the physical number of the optical module interface and the corresponding optical module information are clearly marked in the fault indication.

6. The method for optimizing on-network parameters and locating faults of optical modules based on PRBS detection as described in claim 5, characterized in that, The PRBS sequence generated in the FPGA of the test instrument is a PRBS31 sequence or a test sequence specified by other Ethernet standards.

7. The method for optimizing on-network parameters and locating faults of optical modules based on PRBS detection as described in claim 6, characterized in that, In the step of traversing and modifying the transmission parameters of the optical module, the optimal configuration is quickly approximated by partial sampling of the transmission parameters instead of exhaustive traversal; if a parameter combination that meets the Ethernet standard cannot be found after sampling traversal, a fault identifier and indication are triggered for the optical module interface.

8. The method for optimizing on-network parameters and locating faults of an optical module based on PRBS detection as described in claim 2, further comprising, in the step of measuring the bit error rate of the PRBS sequence transmitted back by the optical module under different combinations of transmission parameters, extracting eye diagram opening and jitter as evaluation indicators of physical dimensions; selecting the transmission parameter combination with the optimal comprehensive performance index by weighted summation of the bit error rate, eye diagram opening, and jitter; and determining the optical module interface fault and providing an indication when the comprehensive index of all parameter combinations does not meet the standard.

Images