A method for adaptively alleviating MPI noise in IMDD system and IMDD system

Abstract

The present invention discloses a method for adaptively mitigating MPI noise in an IMDD system and an IMDD system, belonging to the field of optical communication technology. The method is used at a receiving end and includes: obtaining a received signal and calculating its probability density distribution; obtaining two quantization level values ​​at each trough as a threshold level pair based on the probability density distribution of the received signal; dividing the data points of the received signal into reliable points and stray points based on the threshold level pairs; after performing a decision operation on the reliable points, obtaining a recovered signal corresponding to the reliable points and calculating the decision error; performing MPI noise estimation and MPI noise compensation on the stray points based on the decision error of the reliable points to obtain a recovered signal corresponding to the stray points; and synthesizing the recovered signals corresponding to the reliable points and the recovered signals corresponding to the stray points to obtain a final output signal. Adaptively achieve accurate compensation for MPI noise at stray points in different intervals in the IMDD system link.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical communications, and more particularly, relates to a method for adaptively alleviating MPI noise in an IMDD system and an IMDD system. Background Art

[0002] During transmission, high-order modulation formats (such as PAM4 and PAM8) are more sensitive to damage such as noise and signal jitter due to the reduction in level spacing. In the IMDD (Intensity Modulation and Direct Detection) system, when the interference light caused by the reflection of the signal light between multiple dirty fiber optic connectors reaches the receiver, it will be detected by the PD (Photodetector) together with the signal light, thereby introducing MPI noise. MPI noise can cause jitter in the signal over a large range (thousands of data points) and with large amplitude (a large number of data points approaching or even exceeding the decision level), which is difficult to be compensated by conventional FFE (Feed-Forward Equalizer), resulting in a serious degradation of the BER (Bit Error Ratio). MPI has become one of the main factors currently limiting the deployment of high-order formats in IMDD systems and further increases in speed. However, current construction practices are not standardized, leading to common contamination of fiber optic connector end faces. Problematic connectors still require on-site staff to locate and troubleshoot based on their experience, resulting in a long construction cycle and low efficiency. Therefore, alleviating the signal damage caused by MPI noise to high-order, high-speed IMDD transmission is crucial.

[0003] Numerous studies have been conducted on mitigating the effects of MPI. For example, methods have been proposed to distribute interference noise out-of-band using external phase jitter, disrupt polarization alignment using polarization scramblers, or construct optical branches to simulate interfering light for compensation. However, these solutions require additional optical components and are incompatible with existing networks. Consequently, more economical DSP-based digital domain processing solutions have been proposed in recent years. However, digital domain analog high-pass filter solutions are complex and can degrade transmission performance when MPI noise is low. Hybrid coding and filtering solutions also require high synchronization requirements and additional spread spectrum bands. Consequently, low-complexity error signal reconstruction of MPI noise has emerged as an advantage.

[0004] Current methods for reconstructing MPI noise based on error signals mainly include the ID-MPI-M (intensity-dependent MPI mitigation) scheme, which uses weight factors to compensate for level differences, and the ADT (Adaptive Decision Threshold) scheme, which dynamically adjusts the decision threshold voltage. ID-MPI-M restores the signal by reconstructing and removing MPI noise, while ADT adapts to MPI fluctuations by dynamically adjusting the threshold voltage. However, the ID-MPI-M scheme manually adjusts four weight factors and is not suitable for actual real-time transmission; the ADT scheme requires a training sequence for the initial convergence threshold voltage, which is difficult to achieve in actual short-distance, high-speed transmission services.

[0005] Therefore, the existing methods for reconstructing MPI noise based on error signals in high-speed and high-order IMDD systems have the problems of being difficult to adapt, highly complex, and requiring additional devices. Summary of the Invention

[0006] In view of the shortcomings of the related art, the purpose of the present invention is to provide a method and an IMDD system for adaptively mitigating MPI noise in an IMDD system, aiming to solve the problems of difficulty in adaptability, high complexity and the need for additional devices in the existing high-speed and high-order IMDD system in reconstructing MPI noise based on error signals.

[0007] To achieve the above objectives, in a first aspect, the present invention provides a method for adaptively mitigating MPI noise in an IMDD system, which is used at a receiving end and includes:

[0008] S1. Obtain the received signal and calculate its probability density distribution;

[0009] S2. According to the probability density distribution of the received signal, two quantization level values ​​are obtained at each trough as a threshold level pair;

[0010] S3, the interval between the threshold level pairs corresponding to each trough is the spurious interval, and the rest are the trustworthy intervals; the data points in the received signal that are within the trustworthy interval are classified as trustworthy points, and the data points that are within the spurious interval are classified as spurious points;

[0011] S4. After performing a judgment operation on the trustworthy point, obtaining a recovery signal corresponding to the trustworthy point, and calculating a judgment error corresponding to the trustworthy point;

[0012] S5. Perform MPI noise estimation and MPI noise compensation on the stray point according to the decision error of the trustworthy point, perform a decision operation on the compensated stray point, and obtain a restored signal corresponding to the stray point;

[0013] S6. The restored signals corresponding to the reliable points and the restored signals corresponding to the stray points are integrated to obtain the final output signal.

[0014] Optionally, S1 specifically includes:

[0015] S11. The receiving end performs data acquisition to obtain received signal data x(t) affected by link MPI noise interference, and records the received signal data as y(t); the transmitted signal data x(t) is the PAM-M transmitted signal data of the transmitting end M-order level in the IMDD optical communication system;

[0016] S12. Uniformly quantize the received signal data y(t) into n groups based on the maximum and minimum values ​​of y(t), orderly divide the data points in y(t) into n equally spaced quantization intervals, and take the midpoint of the quantization interval as the quantization level of the group;

[0017] S13. Obtain a probability density distribution of the level distribution according to the number of data points included in each group, and record the probability density distribution as p(f).

[0018] Optionally, S2 specifically includes:

[0019] S21. The probability density distribution p(f) of the PAM-M signal presents M peaks and (M-1) troughs. On the left side of each trough, three quantization level values ​​are selected from the trough to the peak, and the quantization level whose number of data points in the first group is greater than 125% of the number of data points at the trough is selected as the threshold level, which is recorded as the threshold level th on the m side of the trough level. m,upper ;

[0020] S22, on the right side of each trough, in order from the trough to the peak, select three quantization level values ​​close to the trough, and sequentially select the first quantization level whose number of data points in the group is greater than 125% of the number of data points at the trough, and at the same time, the number of data points contained in this quantization level is greater than the threshold level th on the side of level m. m,upper The number of data points is large, which is recorded as the threshold level th on the m+1 side of the valley level. m+1,lower .

[0021] Optionally, S3 specifically includes:

[0022] S31, the received signal y(t) belonging to the interval (th m,lower ,th m,upper ) is divided into trustworthy points of level m, where the trustworthy point intervals of level “1” and level “M” are (min(y(t)),th 1,upper ) and (th M,lower ,max(y(t)));

[0023] S32, will belong to the interval (th m,upper ,th m+1,lower ) is divided into stray points between level m and level m+1.

[0024] Optionally, the calculating and obtaining the decision error corresponding to the trustworthy point includes:

[0025] Subtract the restored signal of the trustworthy point from the trustworthy point to obtain the error corresponding to the trustworthy point.

[0026] Optionally, S5 specifically includes:

[0027] S51, estimating the MPI noise corresponding to each stray point; the MPI noise estimate of a stray point between level m and level m+1 is calculated by calculating the mean of the errors of the reliable points belonging to level m and the errors of the reliable points belonging to level m+1 among the neighboring data points of the point;

[0028] S52, subtracting the MPI noise corresponding to each stray point one by one to obtain the MPI noise compensation of the stray point;

[0029] S53: Perform PAM-M signal judgment on the compensated stray point to obtain a restored transmission signal corresponding to the stray point.

[0030] Optionally, S6 specifically includes:

[0031] S61, replacing all the trustworthy points in the received data y(t) with the recovery signals corresponding to the trustworthy points;

[0032] S62: Replace all the stray points in the received data y(t) with the restored signals corresponding to the stray points to obtain the restored signal y(t) with the MPI noise eliminated. ′ (t), as the final output signal.

[0033] In a second aspect, the present invention further provides a receiving end of an IMDD system, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes any one of the methods described in the first aspect when executing the computer program.

[0034] In a third aspect, the present invention further provides an IMDD system, comprising:

[0035] A transmitting end, a transmission optical fiber, an optical fiber connector, and a receiving end as described in the second aspect;

[0036] The transmitting end is used to generate and modulate a transmitting signal, and transmit the transmitting signal into a transmission optical fiber;

[0037] The transmission optical fiber is used to transmit the transmission signal to the receiving end for detection and reception;

[0038] The optical fiber connector is used to connect multiple transmission optical fibers;

[0039] The receiving end executes the method for adaptively mitigating MPI noise in an IMDD system as described in any one of the first aspects to obtain a final output signal.

[0040] Compared with the prior art, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0041] 1. The present invention provides a method for adaptively mitigating MPI noise in an IMDD system. The method is based on a DSP digital domain processing solution, does not require additional optical devices, and is compatible with existing networks. Multiple threshold levels are selected based on the density distribution of the received signal to distinguish data points of the received signal to obtain trust points and stray points. The received data is used to guide compensation for MPI noise. After judging the trust points and stray points, the compensation obtains a recovered signal, without the need for additional training sequences. The calculation of the MPI noise estimation and compensation process only involves summation and averaging, requiring only dozens of adders and one divider, and has low computational complexity. The method has the advantages of being applicable to actual transmission services, achieving adaptability, low complexity, and low cost, and provides a new approach for mitigating MPI noise in optical communications.

[0042] 2. This invention provides a method for adaptively mitigating MPI noise in IMDD systems. This method establishes threshold level establishment rules. The probability density distribution of the received signal with respect to quantization levels is extracted in the receiving DSP. Two (M-1) threshold levels are adaptively determined according to the proposed threshold level establishment rules. This method, which uses a signal data point partitioning method based on the probability density distribution, preserves the accuracy of reliable points and accurately compensates for MPI noise at stray points in different intervals, adaptively mitigating MPI noise in IMDD system links.

[0043] 3. The IMDD system provided by the present invention is based on a DSP digital domain processing solution and does not require additional optical devices. The calculation of the MPI noise estimation and compensation process only involves summation and averaging, requiring only dozens of adders and one divider, thus simplifying the complexity of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of a method for adaptively mitigating MPI noise in an IMDD system provided by an embodiment of the present invention;

[0045] Figure 2Schematic diagram of a typical application scenario of an embodiment of the present invention; (a) is a diagram of the actual transmission link structure of IMDD in a centralized access network mobile fronthaul, and (b) is a schematic diagram of optical MPI generation in a typical scenario;

[0046] Figure 3 Schematic diagram of a threshold level establishment rule provided by an embodiment of the present invention;

[0047] Figure 4 Schematic diagram of a data point classification method provided by an embodiment of the present invention;

[0048] Figure 5 This is a principle diagram of a method for adaptively alleviating MPI noise in an IMDD system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0050] The contents involved in the above embodiment are described below in conjunction with a preferred embodiment.

[0051] like Figure 1 As shown, the present invention provides a method for adaptively mitigating MPI noise in an IMDD system, which is used at a receiving end and includes:

[0052] S1. Obtain the received signal and calculate its probability density distribution;

[0053] S2. According to the probability density distribution of the received signal, two quantization level values ​​are obtained at each trough as a threshold level pair;

[0054] S3, the interval between the threshold level pairs corresponding to each trough is the spurious interval, and the rest are the trustworthy intervals; the data points in the received signal that are within the trustworthy interval are classified as trustworthy points, and the data points that are within the spurious interval are classified as spurious points;

[0055] S4. After performing a judgment operation on the trustworthy point, obtaining a recovery signal corresponding to the trustworthy point, and calculating a judgment error corresponding to the trustworthy point;

[0056] S5. Perform MPI noise estimation and MPI noise compensation on the stray point according to the decision error of the trustworthy point, perform a decision operation on the compensated stray point, and obtain a restored signal corresponding to the stray point;

[0057] S6. The restored signals corresponding to the reliable points and the restored signals corresponding to the stray points are integrated to obtain the final output signal.

[0058] An embodiment of the present invention provides a method for adaptively mitigating MPI (Multipath Interference) noise introduced by dirty fiber connectors in a IMDD (Intensity Modulation and Direct Detection) system link in a receiving-end DSP (Digital Signal Processing). Based on an input PAM-M (M-order Pulse Amplitude Modulation) signal impaired by MPI, the receiving-end DSP extracts the probability density distribution of the received signal with respect to the quantization level, and determines 2(M-1) threshold levels according to a proposed threshold level establishment rule. Based on threshold level pairs, the received signal data points are divided into reliable points near the transmission level and spurious points near the decision level that are prone to misjudgment. Specifically, the interval between the threshold level pairs corresponding to each trough is the spurious interval, and the remaining intervals are reliable intervals. Reliable points are directly judged to restore the original signal, and the corresponding error is obtained. Spurious points are first compensated for MPI noise, and then judged to restore the original signal. The MPI noise corresponding to each spurious point is obtained by averaging the errors of the two adjacent reliable points of its transmission level. By combining the processing of reliable points and spurious points, the entire input signal can be restored in the receiving end digital signal processor.

[0059] Optionally, S1 specifically includes:

[0060] S11. The receiving end performs data acquisition to obtain received signal data x(t) affected by link MPI noise interference, and records the received signal data as y(t); the transmitted signal data x(t) is the PAM-M transmitted signal data of the transmitting end M-order level in the IMDD optical communication system;

[0061] S12. Uniformly quantize the received signal data y(t) into n groups based on the maximum and minimum values ​​of y(t), orderly divide the data points in y(t) into n equally spaced quantization intervals, and take the midpoint of the quantization interval as the quantization level of the group;

[0062] S13. Obtain a probability density distribution of the level distribution according to the number of data points included in each group, and record the probability density distribution as p(f).

[0063] Optionally, S2 specifically includes:

[0064] S21. The probability density distribution p(f) of the PAM-M signal presents M peaks and (M-1) troughs. On the left side of each trough, three quantization level values ​​are selected from the trough to the peak, and the quantization level whose number of data points in the first group is greater than 125% of the number of data points at the trough is selected as the threshold level, which is recorded as the threshold level th on the m side of the trough level. m,upper ;

[0065] S22, on the right side of each trough, in order from the trough to the peak, select three quantization level values ​​close to the trough, and sequentially select the first quantization level whose number of data points in the group is greater than 125% of the number of data points at the trough, and at the same time, the number of data points contained in this quantization level is greater than the threshold level th on the side of level m. m,upper The number of data points is large, which is recorded as the threshold level th on the m+1 side of the valley level. m+1,lower .

[0066] Reliable points are data points in the received data that are distributed around M transmission levels and are unlikely to be misjudged and are therefore considered correct. Spurious points are signal points in the received data that are distributed around (M-1) decision levels and are prone to misjudgment due to data fluctuations caused by MPI noise. Reliable points and spurious points together constitute the received signal y(t).

[0067] The probability density distribution p(f) of the PAM-M signal presents M peaks and (M-1) troughs. For each trough, two threshold levels can be determined on its left and right sides, which serve as the upper boundary th of the trustworthy point of level m. m,upper and the trustworthy point th of level m+1 m+1,lower (Ignore the lower boundary of the lowest level "1" and the upper boundary of the highest level "M"). m,lower ,th m,upper ) is the trustworthy point of level m (wherein the trustworthy point interval between level “1” and level “M” is (min(y(t)),th 1,upper ) and (th M,lower ,max(y(t)))), and in the interval (th m,upper ,th m+1,lower ) is a stray point between level m and level m+1. m,upper , consider selecting three nearby quantization level values ​​from the left side of the trough from the mth trough to the mth peak in the order from trough to peak; and select the quantization level whose number of data points in the first group is greater than 125% of the number of data points at the trough in order as the threshold level th m,upper For the threshold level th m,lower, consider selecting three nearby quantization level values ​​from the right side of the trough from the m-1th trough to the mth peak side in the order from trough to peak; select the quantization level that meets the following two conditions as the threshold level th m,lower , (1) sequentially select the quantization level whose number of data points in the first group is greater than 125% of the number of data points at the trough, (2) the number of data points in the selected quantization level is greater than th m-1,upper Contains a lot.

[0068] Optionally, S3 specifically includes:

[0069] S31, the received signal y(t) belonging to the interval (th m,lower ,th m,upper ) is divided into trustworthy points of level m, where the trustworthy point intervals of level “1” and level “M” are (min(y(t)),th 1,upper ) and (th M,lower ,max(y(t)));

[0070] S32, will belong to the interval (th m,upper ,th m+1,lower ) is divided into stray points between level m and level m+1.

[0071] Based on the probability density distribution p(f), according to the threshold level establishment rule in the specific step of S2 above, 2(M-1) threshold levels are determined, and the combination of 2(M-1) threshold levels is expressed as {th 1,upper ,th 2,lower ,th 2,upper ,…,th m,upper ,th m+1,lower ,…,th M,lower}, which is used to divide the reliable points and stray points in the received data y(t), where m = 1, 2, ..., M, ignoring th 1,lower and th M,upper .

[0072] Optionally, the calculating and obtaining the decision error corresponding to the trustworthy point includes:

[0073] Subtract the restored signal of the trustworthy point from the trustworthy point to obtain the error corresponding to the trustworthy point.

[0074] For the trustworthy point, PAM-M signal judgment is performed to obtain the restored original signal and the corresponding error corresponding to the trustworthy point. The error of the trustworthy point is obtained by subtracting the original signal restored from the trustworthy point from the trustworthy point.

[0075] Optionally, S5 specifically includes:

[0076] S51, estimating the MPI noise corresponding to each stray point; the MPI noise estimate of a stray point between level m and level m+1 is calculated by calculating the mean of the errors of the reliable points belonging to level m and the errors of the reliable points belonging to level m+1 among the neighboring data points of the point;

[0077] S52, subtracting the MPI noise corresponding to each stray point one by one to obtain the MPI noise compensation of the stray point;

[0078] S53: Perform PAM-M signal judgment on the compensated stray point to obtain a restored transmission signal corresponding to the stray point.

[0079] First, the MPI noise corresponding to each stray point is estimated. The MPI noise of a stray point between levels m and m+1 is calculated by taking the mean of the errors of the reliable points at level m and the reliable points at level m+1 in its neighboring data points. The calculated mean error is then subtracted from each stray point to compensate for the MPI noise at that point. Finally, the compensated stray point is judged to obtain the restored original signal corresponding to the stray point.

[0080] Optionally, S6 specifically includes:

[0081] S61, replacing all the trustworthy points in the received data y(t) with the recovery signals corresponding to the trustworthy points;

[0082] S62: Replace all the stray points in the received data y(t) with the restored signals corresponding to the stray points to obtain the restored signal y(t) with the MPI noise eliminated. ′ (t), as the final output signal.

[0083] The final output signal is obtained by combining the reliable and spurious points to recover the signal. The entire input signal can be recovered in the receiving digital signal processor. This output signal is the restored transmitted signal after removing the MPI noise. This method mitigates MPI noise.

[0084] In a specific embodiment, for ease of understanding, as Figure 2 As shown in the figure, a typical application scenario of the present invention is given. The actual transmission link structure of IMDD in the mobile fronthaul of the centralized access network is as follows: Figure 2 As shown in (a), the signal light starts from the transmitting end laser, passes through a series of link nodes connected by optical fiber jumpers or fused fiber connectors of different lengths, and finally reaches the receiver and is received by the internal detector. Figure 2As shown in (b), when the signal light passes through a contaminated optical fiber connector, some of its energy is reflected by the connector and becomes interference light. After an even number of reflections, the interference light is transmitted in the same direction as the signal light. The multiple interference lights formed by reflections between different connectors in the link overlap with the signal light and are detected by the receiver together.

[0085] In this example, a 56 Gbps PAM4 signal is first generated. After upsampling by a factor of 2, the signal is pulse-shaped using a root-raised cosine filter with a roll-off factor of 0.1 to generate the transmit signal. The transmit signal is then loaded into an arbitrary waveform generator with a sampling rate of 64 GSa / s. The output of the arbitrary waveform generator is amplified by an electronic amplifier and then drives a Mach-Zehnder modulator. A continuous-wave laser serves as the laser source, injecting the modulated signal light into a 15.5-kilometer optical fiber link connected by three contaminated connectors. The optical signal, a superposition of the signal and interference light, is transmitted via the optical fiber to a photodetector at the receiving end. Finally, a digital sampling oscilloscope samples the signal at 100 GSa / s for offline signal processing. Subsequent adaptive MPI noise mitigation of the receive signal is performed in the receiver's DSP.

[0086] The received PAM4 signal y(t) affected by MPI interference is sent to the DSP. The signal data points are evenly divided into 120 quantization levels based on their maximum and minimum values. The data points are then grouped into 120 groups based on the quantization level. The probability density distribution of y(t) with respect to quantization level, denoted as p(f), is obtained based on the number of data points in each quantization level group.

[0087] The embodiment of the present invention provides a preferred threshold level establishment rule, specifically as follows: Figure 3 According to the above threshold level establishment rules, six threshold levels {th 1,upper ,th 2,lower ,th 2,upper ,th 3,lower ,th 3,upper ,th 4,lower}. The three quantization levels on the left and right sides of each of the three troughs of p(f) are used as potential values ​​of the threshold level; secondly, for the threshold level th m,upper , in the order from trough to peak, select the potential value where the number of data points in the first group is greater than 125% of the number of data points at the trough as the threshold level th m,upper , in order to determine {th 1,upper ,th 2,upper ,th 3,upper}; Finally, for the threshold level th m,lower, in the order from trough to peak, select the first group whose number of data points is greater than 125% of the number of data points at the trough and whose number of data points is greater than that for th m-1,upper The potential value of the number of data points contained in it is used as the threshold level th m,lower , in order to determine {th 2,lower ,th 3,lower ,th 4,lower}, resulting in six threshold levels. PAM4 has four transmission levels: -3, -1, 1, and 3, so M = 4. In this case, the mth level described corresponds to transmission levels -3, -1, 1, and 3 when m = 1, 2, 3, and 4, respectively.

[0088] According to Figure 4 The input signal point classification method shown in the figure classifies the received signal y(t) belonging to the interval (min(y(t)),th 1,upper )、(th 2,lower ,th 2,upper )、(th 3,lower ,th 3,upper )、(th 4,lower , max(y(t))) are divided into trustworthy points of levels -3, -1, 1 and 3; and those belonging to the interval (th 1,upper ,th 2,lower )、(th 2,upper ,th 3,lower )、(th 3,upper ,th 4,lower ) are divided into stray points between level-3 and level-1, between level-1 and level 1, and between level 1 and level 3.

[0089] The classified data points are as follows Figure 5 The MPI noise mitigation process shown in FIG is used for processing. In which, the trustworthy point is directly judged to obtain the restored original signal corresponding to the trustworthy point, and the restored original signal of the trustworthy point is subtracted from the trustworthy point to obtain the error corresponding to the trustworthy point.

[0090] Specifically, for a stray point between levels -3 and -1, the estimated MPI noise is obtained by taking the mean of the errors of the nearby reliable points belonging to levels -3 and -1. For a stray point between levels -1 and 1, the estimated MPI noise is obtained by taking the mean of the errors of the nearby reliable points belonging to levels -1 and 1. For a stray point between levels 3 and 1, the estimated MPI noise is obtained by taking the mean of the errors of the nearby reliable points belonging to levels 3 and 1. Each stray point is first subtracted from the MPI noise corresponding to that point, and then a decision is made to obtain the restored original signal corresponding to that stray point.

[0091] The embodiment of the present invention processes the received signal through the DSP at the receiving end, selects multiple threshold levels according to the density distribution of the received signal, distinguishes the data points of the received signal, and obtains trust points and stray points; uses the guidance in the received data to compensate for the MPI noise, and after judging the trust points and stray points, compensates to obtain the recovered signal, without the need to add an additional training sequence in this process; wherein, the calculation of the MPI noise estimation and compensation process only involves summation and averaging, requiring only dozens of adders and a divider, and the computational complexity is low. This solves the problems of the method of reconstructing the MPI noise based on the error signal in the existing high-speed and high-order IMDD system, which is difficult to adapt, has high complexity, and requires additional devices. The method adaptively realizes the precise compensation of the MPI noise of stray points in different intervals in the IMDD system link, achieving the beneficial effects of wide application range, adaptability, low complexity, and low cost.

[0092] Based on the above embodiments, the present invention further provides a receiving end of an IMDD system, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes any one of the methods in the above embodiments when executing the computer program.

[0093] Based on the above embodiments, Figure 2 As shown, the present invention also provides an IMDD system, comprising:

[0094] The transmitting end, the transmission optical fiber, the optical fiber connector and the receiving end described in the above embodiments;

[0095] The transmitting end is used to generate and modulate a transmitting signal, and transmit the transmitting signal into a transmission optical fiber;

[0096] The transmission optical fiber is used to transmit the transmission signal to the receiving end for detection and reception;

[0097] The optical fiber connector is used to connect multiple transmission optical fibers;

[0098] The receiving end executes the method for adaptively mitigating MPI noise in the IMDD system as described in the above embodiment to obtain a final output signal.

[0099] A system for adaptively mitigating MPI noise in an IMDD system provided by an embodiment of the present invention is used to execute a method for adaptively mitigating MPI noise in an IMDD system provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects.

[0100] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for adaptively mitigating MPI noise in an IMDD system, characterized in that: For the receiving end, including: S1. Obtain the received signal and calculate its probability density distribution; S2. According to the probability density distribution of the received signal, two quantization level values ​​are obtained at each trough as a threshold level pair; S3, the interval between the threshold level pairs corresponding to each trough is the spurious interval, and the rest are the trustworthy intervals; the data points in the received signal that are within the trustworthy interval are classified as trustworthy points, and the data points that are within the spurious interval are classified as spurious points; S4. After performing a judgment operation on the trustworthy point, obtaining a recovery signal corresponding to the trustworthy point, and calculating a judgment error corresponding to the trustworthy point; S5. Perform MPI noise estimation and MPI noise compensation on the stray point according to the decision error of the trustworthy point, perform a decision operation on the compensated stray point, and obtain a restored signal corresponding to the stray point; S6. The restored signals corresponding to the reliable points and the restored signals corresponding to the stray points are integrated to obtain the final output signal.

2. The method according to claim 1, wherein S1 specifically includes: S11. The receiving end performs data acquisition to obtain received signal data x(t) affected by link MPI noise interference, and records the received signal data as y(t); the transmitted signal data x(t) is the PAM-M transmitted signal data of the transmitting end M-order level in the IMDD optical communication system; S12. Uniformly quantize the received signal data y(t) into n groups based on the maximum and minimum values ​​of y(t), orderly divide the data points in y(t) into n equally spaced quantization intervals, and take the midpoint of the quantization interval as the quantization level of the group; S13. Obtain a probability density distribution of the level distribution according to the number of data points included in each group, and record the probability density distribution as p(f).

3. The method according to claim 1, wherein S2 specifically includes: S21. The probability density distribution p(f) of the PAM-M signal presents M peaks and (M-1) troughs. On the left side of each trough, three quantization level values ​​are selected from the trough to the peak, and the quantization level whose number of data points in the first group is greater than 125% of the number of data points at the trough is selected as the threshold level, which is recorded as the threshold level th on the m side of the trough level. m,upper ; S22, on the right side of each trough, in order from the trough to the peak, select three quantization level values ​​close to the trough, and sequentially select the first quantization level whose number of data points in the group is greater than 125% of the number of data points at the trough, and at the same time, the number of data points contained in this quantization level is greater than the threshold level th on the side of level m. m,upper The number of data points is large, which is recorded as the threshold level th on the m+1 side of the valley level. m+1,lower .

4. The method according to claim 3, wherein S3 specifically includes: S31, the received signal y(t) belonging to the interval (th m,lower ,th m,upper ) is divided into trustworthy points of level m, where the trustworthy point intervals of level "1" and level "M" are (min(y(t)),th 1,upper ) and (th M,lower ,max(y(t))); S32, will belong to the interval (th m,upper ,th m+1,lower ) is divided into stray points between level m and level m+1.

5. The method according to claim 1, wherein The calculating and obtaining the decision error corresponding to the trustworthy point includes: Subtract the restored signal of the trustworthy point from the trustworthy point to obtain the error corresponding to the trustworthy point.

6. The method according to claim 5, wherein S5 specifically includes: S51, estimating the MPI noise corresponding to each stray point; the MPI noise estimate of a stray point between level m and level m+1 is calculated by calculating the mean of the errors of the reliable points belonging to level m and the errors of the reliable points belonging to level m+1 among the neighboring data points of the point; S52, subtracting the MPI noise corresponding to each stray point one by one to obtain the MPI noise compensation of the stray point; S53: Perform PAM-M signal judgment on the compensated stray point to obtain a restored transmission signal corresponding to the stray point.

7. The method according to claim 6, wherein S6 specifically includes: S61, replacing all the trustworthy points in the received data y(t) with the recovery signals corresponding to the trustworthy points; S62: Replace all the stray points in the received data y(t) with the restored signals corresponding to the stray points to obtain the restored signal y(t) with the MPI noise eliminated. ′ (t), as the final output signal.

8. A receiving end of an IMDD system, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor executes the method according to any one of claims 1 to 7 when executing the computer program.

9. An IMDD system, characterized in that: include: A transmitting end, a transmission optical fiber, an optical fiber connector, and a receiving end according to claim 8; The transmitting end is used to generate and modulate a transmitting signal, and transmit the transmitting signal into a transmission optical fiber; The transmission optical fiber is used to transmit the transmission signal to the receiving end for detection and reception; The optical fiber connector is used to connect multiple transmission optical fibers; The receiving end executes the method for adaptively mitigating MPI noise in an IMDD system according to any one of claims 1 to 7 to obtain a final output signal.

Images