[go: up one dir, main page]

CN119921862B - Compensation method and device for multipath crosstalk of optical communication link - Google Patents

Compensation method and device for multipath crosstalk of optical communication link Download PDF

Info

Publication number
CN119921862B
CN119921862B CN202510413452.0A CN202510413452A CN119921862B CN 119921862 B CN119921862 B CN 119921862B CN 202510413452 A CN202510413452 A CN 202510413452A CN 119921862 B CN119921862 B CN 119921862B
Authority
CN
China
Prior art keywords
multipath
feedforward
crosstalk
analog
clock cycle
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202510413452.0A
Other languages
Chinese (zh)
Other versions
CN119921862A (en
Inventor
史学明
张良俊
王卫明
陈新鹏
陈伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Core Tide Zhuhai Technology Co ltd
Original Assignee
Core Tide Zhuhai Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Core Tide Zhuhai Technology Co ltd filed Critical Core Tide Zhuhai Technology Co ltd
Priority to CN202510413452.0A priority Critical patent/CN119921862B/en
Publication of CN119921862A publication Critical patent/CN119921862A/en
Application granted granted Critical
Publication of CN119921862B publication Critical patent/CN119921862B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Optical Communication System (AREA)

Abstract

The invention provides a multipath crosstalk compensation method and device of an optical communication link, wherein the method comprises the steps that an analog-to-digital converter collects signals output by an analog front end and converts the signals into digital signals to be output to a feedforward equalizer, the feedforward equalizer uses feedforward equalized data to subtract target value calculation used in a feedforward equalization process to obtain feedforward equalization error values, whether the feedforward equalizer converges or not is judged, if so, the feedforward equalization error values are output to a filter, the filter filters the feedforward equalization error values to obtain multipath crosstalk noise quantity, the multipath crosstalk noise quantity is used for compensation calculation, and the compensated data are output to a decision feedback equalizer, wherein when the analog-to-digital converter collects the signals output by the analog front end, the signals output by the analog front end are extracted according to preset proportions in each clock period. The device can realize the method. The invention can adapt to the rapid change of multipath crosstalk noise of the optical communication link and perform effective compensation processing.

Description

Compensation method and device for multipath crosstalk of optical communication link
Technical Field
The invention relates to the technical field of optical communication, in particular to a method for compensating multipath crosstalk of an optical communication link and a device for realizing the method.
Background
With the development of 5G communication technology, artificial intelligence technology and big data technology, the demands on the computing power of the chip are increased exponentially. Optical interconnection schemes between switches, whether data centers or telecommunications networks, are almost one generation evolving every three years. With the rapid increase and iteration of the capacity of the switching chips of data centers and telecommunications, the high-speed connection between the stages in the network architecture needs to be matched and upgraded. Due to the rapid evolution of data rates, more advanced modulation techniques are needed, i.e. the evolution from NRZ to PAM4 technology has been realized at present. For optical fiber communication of the PAM4 direct alignment detection technology of 100G, the influence of multipath crosstalk (MPI, multi-PATH INTERFERENCE, multipath crosstalk) on signals is more and more emphasized, and optical communication systems often need to compensate for the multipath crosstalk.
Referring to fig. 1, a typical high-speed optical communication system uses a first optical module 11 and a second optical module 21 as carriers for signal transmission and reception, a host 10 communicates with a slave 30 through the first optical module 11, an optical fiber 20 and the second optical module 21, a first DSP chip 12 is disposed in the first optical module 11, a first transmitting DSP13 and a first receiving DSP14 are disposed in the first DSP chip 12, a second DSP chip 22 is disposed in the second optical module 21, and a second transmitting DSP23 and a second receiving DSP24 are disposed in the second DSP chip 22. In the downlink, a signal sent by the host 20 sequentially passes through the first transmitting DSP13 and the first optoelectronic device 15 to reach the optical fiber 20, and then passes through the third optoelectronic device 25 and the second transmitting DSP23 to reach the slave 30. In the uplink, the signal transmitted from the slave 30 passes through the second reception DSP24 and the fourth optoelectronic device 26 in order to reach the optical fiber 20, and passes through the second optoelectronic device 16 and the first reception DSP14 to reach the host 10. Since the opto-electronic channel formed by the individual opto-electronic devices and the optical fiber 20 is not an ideal channel, various impairments and noise may result in signal distortion or degradation of the signal-to-noise ratio.
Referring to fig. 2, in a simplified typical 500m to 40km data center and multi-drop optical fiber transmission channel of a telecommunications network, there are a plurality of optical fiber splices, including for example a first optical fiber splice 32, a second optical fiber splice 33 and a third optical fiber splice 34, between a first optical module end face 31 and a second optical module end face 35, each of which is connected by an optical fiber. Because each optical fiber connector has reflection, and the light reflection is more serious along with the aging of the optical fiber, the pollution of the optical fiber connector and the like, multiple reflections are formed in an optical fiber channel, and finally multipath crosstalk noise is formed, and multipath crosstalk is formed on a received signal. Because the number of optical fiber connectors, the reflectivity of the end face and the length of the optical fiber section are not very strictly regulated and required, various combinations can be adopted to adapt to the actual application scene, the number of connectors can cause uncertainty of noise variable, the increase of the reflectivity can cause increase of noise energy, and the change of the length of the optical fiber can cause change of crosstalk symbol delay. Thus, for a DSP, there is no way for multipath crosstalk noise to be compensated and detected by finding the source location, and there is no way to track compensation with a conventional feed forward equalizer (FFE, feed Forward Equalizer) scheme in engineering applications when there are multiple large delays.
After multipath crosstalk noise is subjected to square detection at a receiving end, high-frequency noise and low-frequency noise coexist, but due to interference of signals and noise, the high-frequency noise is weakened, and the influence of the multipath crosstalk noise is mainly concentrated at low frequency finally. By analysis of experimental data of application scenarios, multipath crosstalk noise frequencies are typically less than 100MHz, and in severe cases the noise to signal ratio may be greater than 5%. The low-frequency noise of the multipath crosstalk noise needs to be eliminated through the DSP algorithm processing.
Referring to fig. 3, in a conventional multipath crosstalk noise compensation method, in a receiving DSP, an analog-to-digital converter 42 is used to collect data of an analog front end 41, then a low-frequency noise compensator 43 averages the collected data segments or performs a sliding window averaging process to approximate a low-pass filtering process, so as to obtain a low-frequency noise, and then the low-frequency noise is subtracted by a signal to realize the multipath crosstalk noise compensation. The compensated data is output to the feedforward equalizer 44 and finally output to the decision feedback equalizer 45 for re-equalization. However, the scheme of data segmentation averaging has the problems of large tracking error and large jitter, while the scheme of sliding window averaging has the problems of high complexity and large implementation difficulty.
On the other hand, the invention patent application publication CN118554892a discloses a noise compensation method of a transimpedance amplifier, which filters a signal using a filter, but is mainly applied to noise compensation of the transimpedance amplifier. In the application scene of the optical communication link, the method cannot meet the requirement of multipath crosstalk noise compensation in the optical communication link because the variation period of multipath crosstalk noise is faster than the noise variation period of the transimpedance amplifier.
Disclosure of Invention
A first object of the present invention is to provide a compensation method of multipath crosstalk of an optical communication link capable of satisfying the multipath crosstalk noise variation period requirement of the optical communication link.
A second object of the present invention is to provide a multipath crosstalk compensation device for implementing the above-mentioned method for compensating multipath crosstalk of an optical communication link.
The method for compensating multipath crosstalk of the optical communication link comprises the steps of acquiring signals output by an analog-to-digital converter, converting the signals into digital signals and outputting the digital signals to a feedforward equalizer, subtracting target values used in a feedforward equalization process by feedforward equalizer to obtain feedforward equalization error values, judging whether the feedforward equalizer converges or not, outputting the feedforward equalization error values to a filter if the feedforward equalization error values converge, filtering the feedforward equalization error values by the filter to obtain multipath crosstalk noise amounts, performing compensation calculation by using the multipath crosstalk noise amounts, and outputting the compensated data to a decision feedback equalizer, wherein when the analog-to-digital converter acquires the signals output by the analog front end, the signals output by the analog front end are extracted according to preset proportions in each clock period.
According to the scheme, aiming at the characteristic that the change period of multipath crosstalk noise in an optical communication link is relatively quick, the analog-to-digital converter acquires signals in a mode of extracting a certain proportion of signals in each clock period when acquiring signals of an analog front end, so that the data volume required to be processed is greatly reduced. In addition, as each clock period has a certain signal to be extracted, each clock period can be ensured to acquire a certain signal, so that the accuracy of multipath crosstalk noise processing is ensured.
In a preferred embodiment, the extracting the signals output by the analog front end at each clock cycle according to a predetermined ratio includes extracting a predetermined number of signals output by the analog front end at each clock cycle.
Therefore, the number of the signals extracted in each clock cycle is preset, and can be adjusted according to actual conditions, so that the use requirements in different use situations can be met.
Further, when the signal output by the analog front end in the preset number is extracted in each clock cycle, the position of the signal extracted in each clock cycle is determined in a polling mode.
It can be seen that determining the position of the signal extracted for each clock cycle in a polling manner can avoid the problem of inaccurate calculation of multipath crosstalk noise caused by the use of fixed-position extracted signals.
Further, the clock recovery operation is also performed before the feedforward equalization error value is calculated. By performing the clock recovery operation, it can be ensured that the clock signal can remain synchronized during the calculation of the feedforward equalization error value.
The method comprises the steps of obtaining the feedforward equalization data of a feedforward equalizer in a current clock cycle, wherein the feedforward equalization data of the feedforward equalizer in the current clock cycle is obtained by subtracting the multipath crosstalk noise amount from the feedforward equalization data of the feedforward equalizer in the current clock cycle.
Therefore, the backward compensation mode can compensate for the multipath crosstalk noise of the current clock period, so that the multipath crosstalk noise compensation has higher timeliness.
The method comprises the steps of obtaining the compensation data of the next clock period by subtracting the calculated multipath crosstalk noise quantity of the current period from the feedforward balanced data of the feedforward equalizer of the next clock period when the compensation calculation is carried out by using the multipath crosstalk noise quantity.
It follows that the forward compensation mode can apply the multipath crosstalk noise amount of the current clock cycle to compensate the data of the next clock cycle.
Further, after calculating the compensation data of the next clock cycle, the method further comprises the step of calculating the feedforward equalization error value of the clock cycle in the next clock cycle.
It can be seen that the feedforward equalization error value for each clock cycle is calculated and the multipath crosstalk noise for that clock cycle is compensated.
In order to achieve the second objective, the invention provides a multipath crosstalk compensation device of an optical communication link, which comprises an analog-to-digital converter, a feedforward equalizer, a filter, an adder, a decision feedback equalizer, and a decision feedback equalizer, wherein the analog-to-digital converter is used for collecting signals output by an analog front end and converting the signals into digital signals, the feedforward equalizer is used for receiving the signals of the analog front end and performing feedforward equalization processing, target value calculation used in a feedforward equalization process is subtracted from data after feedforward equalization to obtain feedforward equalization error values, the filter is used for receiving feedforward equalization error values and filtering the feedforward equalization error values to obtain multipath crosstalk noise when the feedforward equalizer converges, the adder is used for performing compensation calculation by using the multipath crosstalk noise, the decision feedback equalizer is used for receiving the compensated data and performing equalization processing on the compensated data again, and the analog front end output signals are extracted according to a preset proportion in each clock cycle when the analog front end output signals are collected by the analog front end.
As can be seen from the above scheme, the optical communication link has the characteristic of faster variation period of multipath crosstalk noise, and the analog-to-digital converter acquires signals in a mode of extracting a certain proportion of signals in each clock period when acquiring analog front-end signals, so that the data volume required to be processed is greatly reduced. In addition, as each clock period has a certain signal to be extracted, each clock period can be ensured to acquire a certain signal, so that the accuracy of multipath crosstalk noise processing is ensured.
In a preferred embodiment, the sampling rate of the analog-to-digital converter is greater than or equal to the baud rate of the analog front end output signal.
Therefore, the sampling rate of the analog-to-digital converter is higher than the baud rate of the output signal of the analog front end, so that the analog-to-digital converter can accurately extract the signals of each clock period.
The analog-to-digital converter extracts a preset number of signals output by the analog front end in each clock cycle when extracting signals output by the analog front end according to a preset proportion in each clock cycle.
Drawings
Fig. 1 is a block diagram of a conventional high-speed optical communication system.
Fig. 2 is a block diagram of a multi-drop fiber transport channel architecture of a prior art data center and telecommunications network.
Fig. 3 is a schematic diagram of a conventional multipath crosstalk noise compensation method.
Fig. 4 is a block diagram of a first embodiment of a compensation apparatus for multipath crosstalk in an optical communication link according to the present invention.
Fig. 5 is a flowchart of a first embodiment of a method of compensating for multipath crosstalk in an optical communication link according to the present invention.
Fig. 6 is a block diagram of a second embodiment of a compensating device for multipath crosstalk in an optical communication link according to the present invention.
Fig. 7 is a flow chart of a second embodiment of a method of compensating for multipath crosstalk in an optical communication link according to the present invention.
The invention is further described below with reference to the drawings and examples.
Detailed Description
The invention calculates and compensates the multipath crosstalk noise aiming at the condition that the multipath crosstalk noise exists in the optical communication link, thereby reducing the signal distortion condition in the optical communication link and ensuring the signal transmission quality of the optical communication link.
First embodiment:
Referring to fig. 4, the compensation apparatus for multipath crosstalk of the optical communication link of the present embodiment has a first analog-to-digital converter 51, a first feedforward equalizer 52, a first-order digital low-pass filter 53, a first decision feedback equalizer 54, and a first adder 55. The first analog-to-digital converter 51 is configured to collect an analog signal output by the analog front end, and convert the collected analog signal into a digital signal. Since the signal change rate of the analog front end is very fast in the optical communication link, the sampling rate of the first analog-to-digital converter 51 needs to be greater than or equal to the baud rate of the output signal of the analog front end in order to ensure accurate sampling of data. In addition, the first analog-to-digital converter 51 converts the collected analog signal into a digital signal, and also converts the serial signal into parallel data, so that the DSP chip can process the parallel data.
The first feedforward equalizer 52 is configured to receive the digital signal output by the first analog-to-digital converter 51, and is configured to compensate for the impairment of Inter-symbol interference (ISI, inter-Symbol Interference) of the received digital signal, and specifically, the first feedforward equalizer 52 obtains a stable feedforward equalization error, denoted as FFE error, through adaptive convergence calculation, where the feedforward equalization process is performed on the received data, such as performing a convolution calculation on the input data and a feedforward equalization coefficient. The feedforward equalization coefficients used in the feedforward equalization calculation by the first feedforward equalizer 52 are adaptively obtained by using a least mean square algorithm, and after the feedforward equalization coefficients converge and stabilize, the mean square error value of the least mean square algorithm will tend to stabilize.
The first-order digital low-pass filter 53 is a fixed-coefficient first-order filter for smoothing and averaging the input signal and the history signal, thereby realizing the effect of low-pass filtering the input signal. In this embodiment, the filter coefficients of the first-order digital low-pass filter 53 are reasonably configured, so that the bandwidth of the first-order digital low-pass filter 53 can be configured and adjusted to adapt to the use requirements of different application scenarios. Since the present embodiment is applied in an optical communication link and tracking and compensation of multipath crosstalk noise are required, the bandwidth of the first-order digital low-pass filter 53 is required to be greater than 10MHz.
The first adder 55 is configured to receive signals output from the first feedforward equalizer 52 and the first-order digital low-pass filter 53, and to perform addition and subtraction processing on the signals. Specifically, the signal output from the first order digital low-pass filter 53 is subtracted from the signal of the first feedforward equalizer 52, and the calculation result is output to the first decision feedback equalizer 54.
The first decision feedback equalizer 54 is configured to perform equalization on the signal after the feedforward processing, adjust a decision feedback equalization coefficient according to a decision result of the signal output by the first feedforward equalizer 52, and perform decision feedback equalization processing according to the decision feedback equalization coefficient, thereby reducing distortion of the signal. Specifically, after determining the current decision symbol, the first decision feedback equalizer 54 needs to adjust the coefficient of the decision feedback equalization according to the decision result of the previous decision symbol, and apply the coefficient to the current decision symbol, and then send the compensated current decision symbol to the decision device for decision, and output the decision result of the current decision symbol.
The following describes a specific flow of the compensation method for multipath crosstalk of the optical communication link according to the present embodiment with reference to fig. 5. First, step S1 is performed, the first analog-to-digital converter 51 collects an analog signal output from the analog front end, and performs analog-to-digital conversion on the analog signal to obtain a digital signal, and converts a serial signal into a parallel signal. The first analog-to-digital converter 51 samples the signal at a certain ratio in each clock cycle, for example, the analog front end outputs 64 or 32 signals in one clock cycle, and the first analog-to-digital converter 51 samples the signal at a certain ratio or number in the plurality of signals output from the analog front end, for example, samples 1 to 4 signals in each clock cycle. The number or proportion of the signals extracted by the first analog-to-digital converter 51 can be adjusted according to actual needs, the smaller the number of the extracted signals is, the smaller the subsequent calculation amount is, the calculation speed can be improved, the delay of compensation is reduced, but the smaller the extracted signals are, the accuracy of compensation calculation can be affected. Therefore, the speed and accuracy of compensation calculation are considered, and the number of extraction is reasonably configured.
In addition, the position of the signal extracted for each clock cycle may be determined in a sample-and-poll manner. For example, the number of extracted signals per clock cycle is 4, and in the first clock cycle, the signals at the 4 th positions of 1, 17, 33 and 49 are extracted, in the second clock cycle, the extracted signals are respectively located at the 2 nd, 18, 34 and 50 th positions, in the third clock cycle, the extracted signals are respectively located at the 3 rd, 19 th, 35 th and 51 th positions, and so on. Therefore, in a plurality of adjacent clock cycles, the positions of the signals extracted in each clock cycle are different, the problem that the calculation of the multipath crosstalk noise is inaccurate due to the fact that the signals are extracted in the same position repeatedly can be avoided, the calculation accuracy of the multipath crosstalk noise can be improved, and the calculated compensation result is more accurate.
Then, step S2 is performed, the first feedforward equalizer 52 is started, and the operation of clock recovery (Clock and Data Recovery) is performed. The first feedforward equalizer 52 will calculate a feedforward equalization error value (FFE error), and specifically, the first feedforward equalizer 52 uses the data obtained after feedforward equalization minus the target value used in the feedforward equalization process to calculate the obtained feedforward equalization error value. The target value used in the feedforward equalization process is a data decision value after feedforward equalization.
Next, step S3 is executed to determine whether the first feedforward equalizer 52 has converged, and if not, it indicates that the data error is large and compensation cannot be performed, step S7 is executed to directly output the data to the first decision feedback equalizer 54. If the first feedforward equalizer 52 has converged, step S4 is performed to turn on the first-order digital low-pass filter 53, and the feedforward equalization error value calculated in step S2 is input to the first-order digital low-pass filter 53. In order to better filter the feedforward equalization error value, reasonable parameters of the first-order digital low-pass filter 53 need to be configured according to an application scene, so that the first-order digital low-pass filter 53 has a reasonable low-pass filtering bandwidth. After receiving the feedforward equalization error value, the first-order digital low-pass filter 53 filters the feedforward equalization error value and obtains a filtering result, and the multipath crosstalk noise amount obtained according to the filtering result is estimated.
Next, step S5 is performed to compensate the feedforward equalized data using the multipath crosstalk noise amount. Specifically, the data subjected to feedforward equalization calculation by the first feedforward equalizer 52 is output to the first adder 55, and the multipath crosstalk noise amount calculated by the first-order digital low-pass filter 53 is output to the first adder 55, and the first adder 55 subtracts the multipath crosstalk noise amount from the feedforward equalized data to obtain compensated data.
Finally, step S6 is performed, and the first adder 55 outputs the compensated data to the first decision feedback equalizer 54, and the data is equalized again by the first decision feedback equalizer 54.
It can be seen that the compensation device for multipath crosstalk of the optical communication link in this embodiment is a scheme of backward compensation, that is, the compensation of multipath crosstalk noise is performed after the first feedforward equalizer 52, and this scheme can supplement multipath crosstalk noise of the current clock cycle in the same clock cycle, so as to ensure timeliness of multipath crosstalk noise compensation.
Second embodiment:
Referring to fig. 6, the compensation apparatus for multipath crosstalk of the optical communication link of the present embodiment has a second analog-to-digital converter 61, a second feedforward equalizer 62, a second-order loop filter 63, a second decision feedback equalizer 64, and a second adder 65. The second analog-to-digital converter 61 is configured to collect an analog signal output from the analog front end, and convert the collected analog signal into a digital signal. Since the rate of signal change of the analog front end is very fast in the optical communication link, the sampling rate of the second analog-to-digital converter 61 needs to be greater than or equal to the baud rate of the output signal of the analog front end in order to ensure accurate sampling of data. In addition, the second analog-to-digital converter 61 converts the collected analog signal into a digital signal and also converts the serial signal into parallel data, so that the DSP chip can process the parallel data.
The second feedforward equalizer 62 is configured to receive the digital signal output by the second analog-to-digital converter 61 and is configured to compensate for the inter-symbol interference impairment of the received digital signal, for example, the second feedforward equalizer 62 obtains a stable feedforward equalization error through adaptive convergence calculation, where the feedforward equalization process is to perform feedforward equalization calculation on the received data, such as performing convolution calculation on the input data and the feedforward equalization coefficient.
The second-order loop filter 63 adopts two coefficients to form a fixed second-order filter, the two coefficients are an integral coefficient ki and a proportional coefficient kp respectively, and the value of the proportional coefficient kp needs to be larger than or equal to the value of the integral coefficient ki so as to ensure the stability of a loop, realize the smooth average processing of an input signal and a history signal, and play a role of low-pass filtering. In this embodiment, the bandwidth adjustment of the second-order loop filter 63 can be realized by configuring different filter coefficients, so as to meet the use requirements in various scenes. To meet the need for tracking multipath crosstalk noise, the bandwidth of the second order loop filter 63 of the present embodiment is greater than 10MHz.
The second-order loop filter 63 is capable of receiving the data output from the second feedforward equalizer 62, that is, the data obtained after feedforward equalization, and performing filtering processing on the feedforward equalized data, and the result of the filtering processing is output to the second adder 65. The second adder 65 is configured to receive signals output from the second analog-to-digital converter 61 and the second-order loop filter 63, and perform addition and subtraction processing on the signals. Specifically, the signal output from the second analog-to-digital converter 61 is subtracted by the signal output from the second-order loop filter 63, and the calculation result is fed back to the second feedforward equalizer 62.
The second decision feedback equalizer 64 is configured to perform equalization on the signal after the feedforward processing, adjust a decision feedback equalization coefficient according to a decision result of the signal output by the second feedforward equalizer 62, and perform decision feedback equalization processing according to the decision feedback equalization coefficient, thereby reducing distortion of the signal. Specifically, after determining the current decision symbol, the second decision feedback equalizer 64 needs to adjust the coefficient of the decision feedback equalization according to the decision result of the previous decision symbol, and apply the coefficient to the current decision symbol, and then send the compensated current decision symbol to the decision device for decision, and output the decision result of the current decision symbol.
The following describes a specific flow of the compensation method for multipath crosstalk of the optical communication link according to the present embodiment with reference to fig. 7. First, step S11 is performed, and the second analog-to-digital converter 61 collects an analog signal output from the analog front end, and performs analog-to-digital conversion on the analog signal to obtain a digital signal, and converts a serial signal into a parallel signal. The second analog-to-digital converter 61 samples the signal, and extracts the signal at a certain ratio every clock cycle, for example, extracts 1 to 4 signals in one clock cycle. And, the number or proportion of the signals extracted by the second analog-to-digital converter 61 can be adjusted according to actual needs. In addition, the position of the signal extracted in each clock cycle can be determined in a sampling and polling mode, namely, in a plurality of adjacent clock cycles, the position of the signal extracted in each clock cycle is changed in sequence, for example, one bit is moved backwards, and the problem that multipath crosstalk noise is calculated inaccurately due to the fact that the signal extracted in the same position is repeated can be avoided.
Then, step S12 is performed, the second feedforward equalizer 62 is started, and the operation of clock recovery (Clock and Data Recovery) is performed, and the second feedforward equalizer 62 calculates a feedforward equalization error value (FFE error), specifically, the second feedforward equalizer 62 calculates a feedforward equalization error value by subtracting a target value used in the feedforward equalization process at the current clock cycle from data obtained after the feedforward equalization at the current clock cycle.
Next, step S13 is executed to determine whether the second feedforward equalizer 62 has converged, and if not, it indicates that the data error is large and compensation cannot be performed, step S19 is executed to directly output the data to the second decision feedback equalizer 64. If the second feedforward equalizer 62 has converged, step S14 is performed to turn on the second-order loop filter 63, and the feedforward equalization error value calculated in step S12 is input to the second-order loop filter 63. In order to better filter the feedforward equalization error value, reasonable parameters of the second-order loop filter 63 need to be configured according to an application scene, so that the second-order loop filter 63 has a reasonable low-pass filtering bandwidth. After receiving the feedforward equalization error value, the second-order loop filter 63 filters the feedforward equalization error value and obtains a filtering result, and the multipath crosstalk noise amount of the current clock cycle is estimated according to the filtering result.
On the other hand, in the next clock cycle, the second analog-to-digital converter 61 will collect the analog signal output by the analog front end in the next clock cycle, that is, step S15 is performed, and then step S16 is performed, and the data in the next clock cycle is compensated by using the multipath crosstalk noise amount obtained by calculation in the current clock cycle. Specifically, the second adder 65 subtracts the multipath crosstalk noise amount calculated in the current clock cycle from the data obtained in the second analog-to-digital converter 61 in the next clock cycle to calculate the compensation data in the next clock cycle, that is, subtracts the data output from the second-order loop filter 63 in step S14 from the data obtained in step S15 to obtain the compensation data in the next clock cycle.
Then, step S17 is performed, and the second feedforward equalizer 62 calculates a new feedforward equalization error value for the next clock cycle while performing a clock recovery operation. In the next clock cycle, the data received by the second feedforward equalizer 62 is the data output from the second adder 65, and therefore, the data is the calculation result of the multipath crosstalk noise amount calculated by subtracting the current clock cycle from the data acquired by the second analog-to-digital converter 61 in the next clock cycle. Finally, step S18 is performed to output the compensated data to the second decision feedback equalizer 64, and the data is equalized again by the second decision feedback equalizer 64.
It can be seen that the compensation device for multipath crosstalk of the optical communication link of this embodiment is a forward compensation scheme, that is, the multipath crosstalk noise obtained by calculation of the current clock cycle is compensated for the next clock cycle before the second feedforward equalizer 62 performs feedforward equalization.
The analog-to-digital converter samples the data output by the analog front end in a sampling mode, so that the data volume of subsequent processing can be reduced, and the position of the data extracted in each clock period is determined in a polling mode, thereby being beneficial to improving the accuracy of multipath crosstalk noise compensation.
Finally, it should be emphasized that the foregoing is merely a preferred embodiment of the present invention, and is not intended to limit the invention, but rather that various changes and modifications can be made by those skilled in the art without departing from the spirit and principles of the invention, and any modifications, equivalent substitutions, improvements, etc. are intended to be included within the scope of the present invention.

Claims (9)

1.光通信链路的多径串扰的补偿方法,包括:1. A method for compensating multipath crosstalk of an optical communication link, comprising: 模数转换器采集模拟前端输出的信号并转换成数字信号后输出至前馈均衡器;The analog-to-digital converter collects the signal output by the analog front end and converts it into a digital signal before outputting it to the feedforward equalizer; 其特征在于:Features: 所述前馈均衡器使用前馈均衡后的数据减去前馈均衡过程中使用的目标值计算获得前馈均衡误差值;The feedforward equalizer calculates a feedforward equalization error value by subtracting a target value used in the feedforward equalization process from the data after the feedforward equalization; 判断所述前馈均衡器是否收敛,如果收敛,则将所述前馈均衡误差值输出至滤波器,所述滤波器对所述前馈均衡误差值进行滤波,获得多径串扰噪声量;Determine whether the feedforward equalizer converges, and if converged, output the feedforward equalization error value to a filter, and the filter filters the feedforward equalization error value to obtain a multipath crosstalk noise amount; 使用所述多径串扰噪声量进行补偿计算,并将补偿后的数据输出至判决反馈均衡器;Perform compensation calculation using the multipath crosstalk noise amount, and output the compensated data to a decision feedback equalizer; 其中,所述模数转换器采集所述模拟前端输出的信号时,在各个时钟周期按照预设的比例抽取所述模拟前端输出的信号,且以轮询的方式确定各个所述时钟周期所抽取的信号的位置。When the analog-to-digital converter collects the signal output by the analog front end, it extracts the signal output by the analog front end in each clock cycle according to a preset ratio, and determines the position of the signal extracted in each clock cycle in a polling manner. 2.根据权利要求1所述的光通信链路的多径串扰的补偿方法,其特征在于:2. The method for compensating multipath crosstalk of an optical communication link according to claim 1, characterized in that: 在各个时钟周期按照预设的比例抽取所述模拟前端输出的信号包括:Extracting the signal output by the analog front end according to a preset ratio in each clock cycle includes: 在每一个所述时钟周期抽取预设数量的所述模拟前端输出的信号。A preset number of signals output by the analog front end are extracted in each clock cycle. 3.根据权利要求1或2所述的光通信链路的多径串扰的补偿方法,其特征在于:3. The method for compensating multipath crosstalk of an optical communication link according to claim 1 or 2, characterized in that: 计算所述前馈均衡误差值前,还执行时钟恢复操作。Before calculating the feedforward equalization error value, a clock recovery operation is also performed. 4.根据权利要求1或2所述的光通信链路的多径串扰的补偿方法,其特征在于:4. The method for compensating multipath crosstalk of an optical communication link according to claim 1 or 2, characterized in that: 所述滤波器为一阶数字低通滤波器;The filter is a first-order digital low-pass filter; 使用所述多径串扰噪声量进行补偿计算时,使用所述多径串扰噪声量对所述前馈均衡器前馈均衡后的数据进行补偿:使用当前时钟周期所述前馈均衡器前馈均衡后的数据减去所述多径串扰噪声量获得当前时钟周期的补偿数据。When using the multipath crosstalk noise amount for compensation calculation, the multipath crosstalk noise amount is used to compensate the data after feed-forward equalization of the feed-forward equalizer: the data after feed-forward equalization of the feed-forward equalizer in the current clock cycle is subtracted from the multipath crosstalk noise amount to obtain the compensation data of the current clock cycle. 5.根据权利要求1或2所述的光通信链路的多径串扰的补偿方法,其特征在于:5. The method for compensating multipath crosstalk of an optical communication link according to claim 1 or 2, characterized in that: 所述滤波器为二阶环滤波器;The filter is a second-order ring filter; 使用所述多径串扰噪声量进行补偿计算时,使用下一个时钟周期所述模数转换器采集的数据减去当前周期计算的所述多径串扰噪声量获得下一时钟周期的补偿数据。When the multipath crosstalk noise amount is used for compensation calculation, the compensation data for the next clock cycle is obtained by subtracting the multipath crosstalk noise amount calculated in the current cycle from the data collected by the analog-to-digital converter in the next clock cycle. 6.根据权利要求5所述的光通信链路的多径串扰的补偿方法,其特征在于:6. The method for compensating multipath crosstalk of an optical communication link according to claim 5, characterized in that: 计算下一时钟周期的补偿数据后,还执行:After calculating the compensation data for the next clock cycle, it also executes: 在下一个时钟周期,计算该时钟周期的前馈均衡误差值。In the next clock cycle, the feedforward equalization error value of the clock cycle is calculated. 7.光通信链路的多径串扰的补偿装置,包括:7. A device for compensating multipath crosstalk of an optical communication link, comprising: 模数转换器,用于采集模拟前端输出的信号并转换成数字信号;Analog-to-digital converter, used to collect the signal output by the analog front end and convert it into a digital signal; 其特征在于:Features: 前馈均衡器,用于接收所述模数转换器的信号并进行前馈均衡处理,且前馈均衡后的数据减去前馈均衡过程中使用的目标值计算获得前馈均衡误差值;A feedforward equalizer, used to receive the signal of the analog-to-digital converter and perform feedforward equalization processing, and calculate the feedforward equalization error value by subtracting the target value used in the feedforward equalization process from the data after the feedforward equalization; 滤波器,在所述前馈均衡器收敛时,接收所述前馈均衡误差值输,并对所述前馈均衡误差值进行滤波,获得多径串扰噪声量;a filter, which receives the feedforward equalization error value input when the feedforward equalizer converges, and filters the feedforward equalization error value to obtain a multipath crosstalk noise amount; 加法器,使用所述多径串扰噪声量进行补偿计算;an adder, using the multipath crosstalk noise amount to perform compensation calculation; 判决反馈均衡器,接收补偿后的数据并对补偿后的数据再次进行均衡处理;A decision feedback equalizer receives the compensated data and performs equalization again on the compensated data; 其中,所述模数转换器采集所述模拟前端输出的信号时,在各个时钟周期按照预设的比例抽取所述模拟前端输出的信号,且以轮询的方式确定各个所述时钟周期所抽取的信号的位置。When the analog-to-digital converter collects the signal output by the analog front end, it extracts the signal output by the analog front end in each clock cycle according to a preset ratio, and determines the position of the signal extracted in each clock cycle in a polling manner. 8.根据权利要求7所述的光通信链路的多径串扰的补偿装置,其特征在于:8. The multipath crosstalk compensation device for an optical communication link according to claim 7, characterized in that: 所述模数转换器的采样率大于或者等于所述模拟前端输出信号的波特率。The sampling rate of the analog-to-digital converter is greater than or equal to the baud rate of the analog front-end output signal. 9.根据权利要求7或8所述的光通信链路的多径串扰的补偿装置,其特征在于:9. The multipath crosstalk compensation device for an optical communication link according to claim 7 or 8, characterized in that: 所述模数转换器在各个时钟周期按照预设的比例抽取所述模拟前端输出的信号时,在每一个所述时钟周期抽取预设数量的所述模拟前端输出的信号。When the analog-to-digital converter extracts the signal output by the analog front end according to a preset ratio in each clock cycle, a preset number of signals output by the analog front end are extracted in each clock cycle.
CN202510413452.0A 2025-04-03 2025-04-03 Compensation method and device for multipath crosstalk of optical communication link Active CN119921862B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202510413452.0A CN119921862B (en) 2025-04-03 2025-04-03 Compensation method and device for multipath crosstalk of optical communication link

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202510413452.0A CN119921862B (en) 2025-04-03 2025-04-03 Compensation method and device for multipath crosstalk of optical communication link

Publications (2)

Publication Number Publication Date
CN119921862A CN119921862A (en) 2025-05-02
CN119921862B true CN119921862B (en) 2025-07-11

Family

ID=95501063

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202510413452.0A Active CN119921862B (en) 2025-04-03 2025-04-03 Compensation method and device for multipath crosstalk of optical communication link

Country Status (1)

Country Link
CN (1) CN119921862B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119602871A (en) * 2024-11-15 2025-03-11 清华大学深圳国际研究生院 Multipath interference noise suppression method, device and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9876581B2 (en) * 2016-02-10 2018-01-23 Inphi Corporation Circuit for multi-path interference mitigation in an optical communication system
US11438197B2 (en) * 2020-11-12 2022-09-06 Texas Instruments Incorporated Adaptive equalizer with a real feedforward filter and a single complex feedforward tap
CN117560090A (en) * 2023-11-13 2024-02-13 无锡飞龙九霄微电子有限公司 Multipath interference elimination method and circuit thereof
CN118554892B (en) * 2024-07-27 2024-12-03 芯潮流(珠海)科技有限公司 Noise compensation method and device for transimpedance amplifier

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119602871A (en) * 2024-11-15 2025-03-11 清华大学深圳国际研究生院 Multipath interference noise suppression method, device and storage medium

Also Published As

Publication number Publication date
CN119921862A (en) 2025-05-02

Similar Documents

Publication Publication Date Title
US7158567B2 (en) Method and apparatus for improved high-speed FEC adaptive equalization
US10498562B2 (en) Electric signal transmission device
JP4968415B2 (en) DIGITAL FILTER DEVICE, DIGITAL FILTERING METHOD, AND DIGITAL FILTER DEVICE CONTROL PROGRAM
CN101283560B (en) Data communication circuit with equalization control
US7003228B2 (en) Method and apparatus for improved high-speed adaptive equalization
US8750442B2 (en) Digital receiver and waveform compensation method
WO2018108008A1 (en) Higher-level clock and data recovery (cdr) in passive optical networks (pons)
US11637684B2 (en) Clock extraction in systems affected by strong intersymbol interference
CN113055321A (en) Optical communication data receiving method and system capable of suppressing burst error propagation
CN110958500B (en) Method and system for receiving a signal
KR101363122B1 (en) Apparatus and Method of equalizing applied adaptation algorithm for high speed transmission
US20070286315A1 (en) Digital signal processor, receiver, corrector and methods for the same
CN115643135B (en) Power efficient nonlinear equalizer and method
CN119921862B (en) Compensation method and device for multipath crosstalk of optical communication link
WO2022002045A1 (en) Signal processing method and device, storage medium, and electronic device
CN118554892A (en) Noise compensation method and device for transimpedance amplifier
JP2009094777A (en) Signal waveform deterioration compensation circuit
WO2015165042A1 (en) Signal receiving method and receiver
CN119011008B (en) Error Propagation Suppression Method for Decision Feedback Equalizer Based on Error Correlation
KR101359691B1 (en) Apparatus and Method of equalizing for high speed transmission
EP4436067A1 (en) Method and apparatus for equalization in pon
CN120378261B (en) Signal processing optimization method and system for training BCJR by using least square method
WO2018086117A1 (en) Error processing method and error processing apparatus
CN118449613A (en) A multipath interference compensation algorithm for incoherent PAM-modulated optical fiber communication systems
CN114697168A (en) Hundred mega Ethernet digital baseband signal processing method and signal processing module

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant