Error correction method based on differential modulation polarization demultiplexing system

Abstract

This application relates to an error correction method based on a differential modulation polarization demultiplexing system, belonging to the field of optical communication technology. The method includes: acquiring N sets of in-phase polarization electrical signals E required for electrical domain polarization demultiplexing through differential detection. H E H d * and E V E V d * and cross-polarized electrical signal E H E V d * and E V E H d *; An adaptive algorithm is used to demultiplex the obtained electrical signal to obtain information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d * Corresponding actual relative symbols Δxy and Δyx; the i-th group of actual relative symbols that is estimated incorrectly in each group; obtaining the correct i-th group of absolute symbols and the correct i+1-th group of absolute symbols corresponding to the i-th group of actual relative symbols; using the i-th group of absolute symbols and the i+1-th group of absolute symbols to correct the i-th group of actual relative symbols; can improve the accuracy of polarization demultiplexing systems with added redundant copies; improve the sensitivity of the receiver.

Description

[0001] This application claims priority to Chinese Patent Application No. 202211459295.X, filed on November 17, 2022, entitled "Error Correction Method Based on Differential Modulation Polarization Demultiplexing System", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to an error correction method based on a differential modulation polarization demultiplexing system, belonging to the field of optical communication technology. Background Technology

[0003] Differential modulation is a digital signal modulation method that uses the difference between consecutive symbols to represent information. Without using a local oscillator, differentially modulated signals can be demultiplexed at the receiving end to recover phase difference information. Polarization-multiplexed differential signals undergo polarization rotation during transmission, requiring polarization demultiplexing before phase information recovery.

[0004] Traditional demultiplexing methods for polarization-multiplexed differential signals are implemented in the optical domain, using optical polarization tracking for demultiplexing. During differential detection, the polarization impairment matrix changes, becoming a singular and irreversible matrix, making adaptive demultiplexing impossible in subsequent digital signal processing. Therefore, traditional methods perform polarization tracking before differential detection. Another approach is to increase receiver complexity by implementing polarization demultiplexing in the electrical domain. Specifically, refer to... Figure 1 After the receiver receives the optical signal, it is split into H-polarized and V-polarized signals by a polarization beam splitter (PBS). The H-polarized optical signal passes through a variable optical attenuator (VOA), and one of the signals is then delayed by a timer to obtain the delayed H-polarized signal H. d The other signal is an undelayed H-polarized signal; the V-polarized optical signal is changed to the same polarization state (denoted as H*) by a polarization rotator, and then one signal is delayed to obtain the delayed V-polarized signal H*. d The other signal is an undelayed V-polarized signal H*. Here, H* indicates that the polarization directions are the same. Then, H... d The polarization of H is obtained by passing H through a mixer and two balanced detectors (BD), and then the conjugate product of H polarization delayed by one symbol time is E. H ·E H d *, H*, and H are combined with a Hybrid and two BDs to obtain the conjugate multiplication E of H-polarization and V-polarization delayed by one symbol time. H ·EV d *、H d The conjugate multiplication E of V-polarization and H-polarization delayed by one symbol time is obtained by passing H* through Hybrid and 2 BDs. V ·E H d *, and H* and H* d After hybridization and two BDs, we obtain the conjugate multiplication E of V-polarization and V-polarization delayed by one symbol time. V ·E V d *

[0005] according to Figure 1 As can be seen from the receiver shown, by adding an optical domain redundant copy (i.e., E... H E V d *, coherence between H-polarization and V-polarization delays; E V E H d * (coherence between V-polarization and H-polarization delays), compared to traditional methods, this receiver structure adds two mixers (Hybrid) and four balance detectors (BD). This receiver structure expands the polarization rotation matrix from 2×2 to 4×4, making the rotation matrix a non-singular matrix again. This allows polarization tracking to be achieved in the electrical domain using a traditional 4×4 adaptive algorithm, which has low algorithm complexity and does not require the participation of a local oscillator.

[0006] However, adding redundant copies doubles the complexity of the polarization demultiplexing receiver structure, but does not improve accuracy. Summary of the Invention

[0007] This application provides an error correction method based on a differential modulation polarization demultiplexing system. In systems based on existing receiver architectures, it can achieve a sensitivity improvement of approximately 1 dB while maintaining a bit error rate (BER) of 1E-3. This application provides the following technical solution:

[0008] The N sets of electrical signals required for electric domain polarization demultiplexing are obtained by differential detection. Each set of electrical signals includes E H ·E H d *Electrical signal, E V ·E V d *Electrical signal, E H ·E V d *Electrical signals and E V ·E H d*Electrical signal, where N is an integer greater than 1;

[0009] The received electrical signal is demultiplexed using an adaptive algorithm to obtain the actual relative symbols Δx, Δy, Δxy, and Δyx, wherein the actual relative symbols include the information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d *The corresponding actual relative symbols Δxy and Δyx;

[0010] Among the actual relative symbols in each group, the i-th group of actual relative symbols that was estimated incorrectly, where i is a positive integer less than N;

[0011] Obtain the correct absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols, and the correct absolute symbols X3 and Y3 corresponding to the (i+1)-th group of absolute symbols; each group of absolute symbols includes the absolute symbol corresponding to the X polarization and the absolute symbol corresponding to the Y polarization.

[0012] The i-th group of absolute symbols X2 and Y2, and the (i+1)-th group of absolute symbols X3 and Y3 are used to correct the i-th group of actual relative symbols.

[0013] Optionally, the i-th group of actual relative symbols that was estimated incorrectly in each group of actual relative symbols includes:

[0014] Use the first group of actual relative symbols Δx1, Δy1, Δxy1 and Δyx1 to determine the correct first group of absolute symbols X1 and Y1 with the second group of absolute symbols X2 and Y2;

[0015] Using the first group of absolute symbols, the second group of absolute symbols, and the actual relative symbols corresponding to the cross terms in each group of actual relative symbols, determine the estimated value of each group of absolute symbols;

[0016] Use the absolute signs of each group of estimates to determine the estimated values ​​corresponding to the information items in the relative signs of each group;

[0017] Compare the information items in each group of actual relative symbols with the corresponding estimated values ​​of the information items to determine the estimation error locations corresponding to X polarization and Y polarization.

[0018] When the estimated error location is discontinuous, the estimated error location is determined to be an error in the information item Δx2 or Δy2 in the corresponding i-th group of actual relative symbols;

[0019] When the estimation error locations are consecutive, the first estimation error location is determined to be an error in the cross term Δxy2 or Δyx2 in the corresponding i-th group of actual relative symbols.

[0020] Optionally, obtaining the correct absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols and the correct absolute symbols X3 and Y3 of the (i+1)-th group includes:

[0021] If the information item Δx2 or Δy2 in the i-th group of actual relative symbols is incorrect, the i-th group of absolute symbols X2 and Y2 are obtained based on the previous group of relative symbols, and the correct i-th group of absolute symbols X2 and Y2 are obtained.

[0022] Determine the error location of the information item in the i-th group of actual relative symbols, wherein the error location is X-polarization or Y-polarization;

[0023] Use the absolute symbol X2 or Y2 in the i-th group corresponding to the error-free position and the cross term Δyx2 or Δxy2 of the actual relative symbol in the i-th group to determine the (i+1)-th absolute symbol Y3 or X3 corresponding to the error-free position;

[0024] Use the absolute symbol Y3 or X3 in the (i+1)th group corresponding to the position without error and the actual relative symbol in the (i+1)th group to determine the absolute symbols X4 and Y4 in the (i+2)th group.

[0025] Based on the (i+2)th absolute symbols X4 and Y4 corresponding to the correct position, and the information items Δx3 and Δy3 of the (i+1)th actual relative symbols, the (i+1)th absolute symbol Y3 or X3 corresponding to the incorrect position can be deduced.

[0026] Optionally, the step of correcting the actual relative symbols of the i-th group using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols includes:

[0027] Using the i-th absolute symbols X2 and Y2 corresponding to the error position and the (i+1)-th absolute symbols X3 and Y3 corresponding to the error position, determine the relative symbol Δx2 or Δy2 of the actual information item corresponding to the error position in the i-th group of actual relative symbols.

[0028] Optionally, obtaining the correct absolute symbols X2 and Y2 of the i-th group and the correct absolute symbols X3 and Y3 of the (i+1)-th group corresponding to the actual relative symbols of the i-th group includes:

[0029] If the cross term Δxy2 or Δyx2 in the i-th group of actual relative symbols is incorrect, obtain the i-th group of absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols.

[0030] Using the i-th group of absolute symbols X2 and Y2, and the i-th group of actual relative symbols corresponding to the information items Δx2 and Δy2, determine the (i+1)-th group of absolute symbols X3 and Y3.

[0031] Optionally, the step of correcting the intersection terms Δxy2 and Δyx2 of the actual relative symbols of the i-th group using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 includes:

[0032] Based on the relative symbol relationship between the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3, the relative symbols Δxy2 and Δyx2 corresponding to the cross terms in the i-th group of actual relative symbols are corrected, and after the symbol correction, the step of using the first group of actual relative symbols to determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2 is triggered.

[0033] Optionally, the step of using the first group of actual relative symbols to determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2 includes:

[0034] From each group of candidate absolute symbols, select a group of untraversed absolute symbols as the first group of absolute symbols to be determined;

[0035] Using the information items relative symbols Δx1 and Δy1 in the first group of actual relative symbols, and the first group of absolute symbols to be determined, determine the second group of absolute symbols to be determined corresponding to the information items;

[0036] Using the cross-term relative symbols Δxy1 and Δyx1 in the first group of actual relative symbols, and the first group of absolute symbols to be determined, determine the second group of absolute symbols to be determined corresponding to the cross-term;

[0037] If the second group of absolute symbols to be determined corresponding to the information item is equal to the second group of absolute symbols to be determined corresponding to the cross item, then the first group of absolute symbols to be determined is determined to be the correct first group of absolute symbols, and the second group of absolute symbols to be determined is determined to be the correct second group of absolute symbols.

[0038] If the second group of absolute symbols to be determined corresponding to the information item is not equal to the second group of absolute symbols to be determined corresponding to the cross item, the step of determining a group of untraversed absolute symbols from each group of candidate absolute symbols as the first group of absolute symbols to be determined is executed again to update the assignment of the first group of absolute symbols to be determined, and the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2 are re-determined.

[0039] Optionally, the i-th group of actual relative symbols that was estimated incorrectly in each group of actual relative symbols includes:

[0040] For each set of actual relative symbols, determine the information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d *Do the corresponding actual relative symbols Δxy and Δyx satisfy the constraint Δxy + Δyx = Δx + Δy, where Δxy represents the cross term E? x E y d *Corresponding actual relative sign, Δyx represents the cross term E y E x d *Corresponding actual relative sign, Δx represents E x E x d *Corresponding actual relative symbol, Δy represents E y E y d *Corresponding to the actual relative symbol;

[0041] Determine the index of the actual relative sign error location where the constraint condition is not met;

[0042] Determine whether the actual relative positions of the errors are consecutive;

[0043] The error index of the actual relative symbol that is not consecutively erroneous is determined as the i-th group of actual relative symbols that are erroneous, and error correction is performed by combining the two groups of relative symbols that are not erroneous before and after it.

[0044] Optionally, obtaining the correct absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols and the correct absolute symbols X3 and Y3 of the (i+1)-th group includes:

[0045] Use the (i-1)th group of actual relative symbols to determine the correct (i-1)th group of absolute symbols X1 and Y1 and the i-th group of absolute symbols X2 and Y2;

[0046] Using the i-th group of absolute symbols X2 and Y2 and the actual relative symbols Δx2 and Δy2 corresponding to the information items in the i-th group of actual relative symbols, determine the first estimated symbol corresponding to X polarization and the second estimated symbol corresponding to Y polarization in the (i+1)-th group of absolute symbols;

[0047] Using the i-th group of absolute symbols X2 and Y2 and the actual relative symbols Δxy2 and Δyx2 corresponding to the cross terms in the i-th group of actual relative symbols, determine the third estimated symbol corresponding to X polarization and the fourth estimated symbol corresponding to Y polarization in the (i+1)-th group of absolute symbols;

[0048] If the first estimated symbol is equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, then the estimated symbol corresponding to X polarization, Δx3 and Δyx3 in the (i+1)th group of actual relative symbols are used to determine the (i+2)th group of absolute symbols X4 and Y4; using the (i+2)th group of absolute symbols X4 and Y4 and the (i+1)th group of actual relative symbols, the symbol corresponding to Y polarization in the (i+1)th group of absolute symbols is deduced in reverse; the symbol corresponding to Y polarization and the first estimated symbol constitute the (i+1)th group of absolute symbols Y3 and X3.

[0049] If the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is equal to the fourth estimated symbol, then the estimated symbol corresponding to Y polarization, Δy3 and Δxy3 in the (i+1)th group of actual relative symbols are used to determine the (i+2)th group of absolute symbols X4 and Y4; using the (i+2)th group of absolute symbols X4 and Y4 and the (i+1)th group of actual relative symbols, the symbol corresponding to X polarization in the (i+1)th group of absolute symbols is deduced in reverse; the symbol corresponding to X polarization and the second estimated symbol constitute the (i+1)th group of absolute symbols X3 and Y3.

[0050] If the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, then the (i+1)th group of actual relative symbols Δx3, Δy3, Δxy3 and Δyx3 are used to determine the (i+1)th group of absolute symbols X3 and Y3 and the (i+2)th group of absolute symbols X4 and Y4.

[0051] Optionally, the step of correcting the actual relative symbols of the i-th group using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 includes:

[0052] Using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3, the actual relative symbols Δx2 and Δy2 corresponding to the information items in the i-th group of actual relative symbols are corrected.

[0053] The beneficial effects of this application include at least the following: obtaining N sets of electrical signals required for electric domain polarization demultiplexing through differential detection, each set of electrical signals including E H ·E H d *Electrical signal, E V ·E V d *Electrical signal, E H ·EV d *Electrical signals and E V ·E H d *Electrical signal, where N is an integer greater than 1; the electrical signal is demultiplexed using an adaptive algorithm to obtain the actual relative symbol, which includes the information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d * Corresponding actual relative symbols Δxy and Δyx; among the received actual relative symbols, the i-th group of actual relative symbols that is estimated incorrectly; obtain the correct i-th group of absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols, and the correct i+1-th group of absolute symbols X3 and Y3; each group of absolute symbols includes the absolute symbol corresponding to X polarization and the absolute symbol corresponding to Y polarization; use the i-th group of absolute symbols X2 and Y2 and the i+1-th group of absolute symbols X3 and Y3 to correct the i-th group of actual relative symbols; this can improve the accuracy of polarization demultiplexing systems with added redundant copies and improve the receiver sensitivity; since the cross terms obtained during demultiplexing can be used to verify and correct the information, the accuracy of polarization demultiplexing systems with added redundant copies can be improved.

[0054] In addition, by determining the actual relative symbol of the error through the constraint relationship between the relative symbol of the information item and the relative symbol of the cross item, it is not necessary to redetermine the actual relative symbol of the error when the cross item is wrong, which can improve the efficiency of symbol error correction.

[0055] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of an optical receiver structure provided in one embodiment of this application;

[0057] Figure 2 This is a schematic diagram of a noise source provided in one embodiment of this application;

[0058] Figure 3 This is a flowchart of an error correction method based on a differential modulation polarization demultiplexing system provided in one embodiment of this application;

[0059] Figure 4 This is a flowchart of a first error correction method provided in an embodiment of this application;

[0060] Figure 5 This is a schematic diagram illustrating the estimation of absolute signs of each group using the actual relative signs corresponding to the cross terms, provided in one embodiment of this application.

[0061] Figure 6 This is a schematic diagram illustrating the location of the error correction provided in one embodiment of this application;

[0062] Figure 7 This is a flowchart of a second error correction method provided in one embodiment of this application;

[0063] Figure 8 This is a schematic diagram illustrating the location of the error correction provided in another embodiment of this application;

[0064] Figure 9 This is a schematic diagram illustrating the location of the error correction provided in yet another embodiment of this application;

[0065] Figure 10 This is a graph showing the system performance Q factor versus optical signal-to-noise ratio per bit provided in one embodiment of this application;

[0066] Figure 11 This is a graph showing the system performance Q factor versus signal-to-noise ratio per bit provided in one embodiment of this application. Detailed Implementation

[0067] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0068] according to Figure 1 The optical receiver structure shown realizes an electrical polarization demultiplexing receiver for differentially modulated signals by doubling the number of optical domain elements, and then polarization tracking can be achieved in the electrical domain using adaptive digital signal processing.

[0069] Specifically, according to Figure 1 It can be seen that, through optical domain processing and balance detection, the conjugate product E of H-polarization and H-polarization delay by one symbol time can be obtained. H ·E H d *, the conjugate multiplication of V-polarization and V-polarization delayed by one symbol time, E V ·E V d * The conjugate multiplication of H-polarization and V-polarization delayed by one symbol time, E H ·E V d* and the conjugate multiplication of V-polarization and H-polarization delayed by one symbol time, E V ·E H d *

[0070] By obtaining the conjugate multiplication terms between cross-polarizations, the received signal can be reconstructed into a 4-dimensional vector. Therefore, polarization demultiplexing can be achieved by multiplying by a 4×4 inverse matrix.

[0071]

[0072] Among them, E x E represents the X-polarized light signal. x d E represents the delay of the X-polarized light signal by one symbol time. y E represents the Y-polarized optical signal. y d E represents the delay of the Y-polarized optical signal by one symbol time. H This represents the received H-polarization signal, E V This represents the received V-polarized signal, and <·>* represents the conjugate operation.

[0073] Let M denote the above matrix, then we get:

[0074]

[0075] Here, M is a non-singular matrix, meaning it has an inverse matrix M. -1 M -1 =M H .

[0076] Therefore, demultiplexing can be achieved at the receiving end using an adaptive algorithm:

[0077]

[0078] The information item E can be obtained at the demultiplexing output through the above demultiplexing process. x E x d * and E y E y d *and the cross term E x E y d * and E y E x d *

[0079] Generally, when obtaining information item E x E x d * and E y Ey d *and the cross term E x E y d * and E y E x d Subsequently, the cross-items are discarded as useless information. However, in this application, the performance of the obtained cross-items and information items is the same. By making a decision on the information items and cross-items, the relative symbols Δx, Δy representing the information and the relative symbols Δxy, Δyx between the cross-items are obtained. Then, Δx, Δy and Δxy, Δyx can be used to correct erroneous symbols, thereby improving the accuracy of the system.

[0080] There are two possible causes for symbol errors in the system, see reference. Figure 2 :

[0081] The first type: When the main source of noise in the system is electrical noise, the four signals received after photoelectric detection are each affected by one independent noise, and the situation of continuous errors is extremely rare.

[0082] The second type: When the main source of noise in the system is optical noise, the two signals before differential demodulation in the optical domain are each affected by independent noise, and the number of consecutive errors increases, but the single error location is still the main one.

[0083] Based on the above characteristics, this application proposes the following error correction method based on a differential modulation polarization demultiplexing system.

[0084] Figure 3 This is a flowchart of an error correction method based on a differential modulation polarization demultiplexing system provided in one embodiment of this application. The method includes at least the following steps:

[0085] Step 301: Obtain N sets of electrical signals required for electric domain polarization demultiplexing through differential detection. Each set of electrical signals includes E H ·E H d *Electrical signal, E V ·E V d *Electrical signal, E H ·E V d *Electrical signals and E V ·E H d *Electrical signal, where N is an integer greater than 1.

[0086] Step 302: Demultiplex the electrical signal using an adaptive algorithm to obtain the actual relative symbol, which includes the information item E. x E x d* and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d *The corresponding actual relative symbols are Δxy and Δyx.

[0087] Specifically, the inverse of the aforementioned non-singular matrix M is used to demultiplex each group of electrical signals to obtain the information item E corresponding to each group of electrical signals. x E x d * and E y E y d * and cross term E x E y d * and E y E x d * Based on the relationship between the relative phases before and after X-polarization and Y-polarization, the actual relative signs are set for the information term and the cross term, resulting in the cross term E. x E y d *Corresponding actual relative symbols Δxy and cross term E y E x d *Corresponding actual relative symbols Δyx, E x E x d *Corresponding actual relative symbols Δx, E y E y d * Corresponds to the actual relative symbol Δy.

[0088] The relationship between relative phase and actual relative sign is illustrated in Table 1 below:

[0089] Table 1:

[0090] relative phase Actual relative symbols π 00 π / 2 01 3π / 2 10 0 11

[0091] Step 303: Among the received groups of actual relative symbols, estimate the i-th group of actual relative symbols that is incorrect. Here, i is a positive integer less than N.

[0092] Step 304: Obtain the correct absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols, as well as the correct absolute symbols X3 and Y3 corresponding to the (i+1)-th group of symbols; each group of absolute symbols includes the absolute symbol corresponding to the X polarization and the absolute symbol corresponding to the Y polarization.

[0093] Step 305: Correct the actual relative symbols of the i-th group using the absolute symbols X2 and Y2 of the i-th group and the absolute symbols X3 and Y3 of the (i+1)-th group.

[0094] In this embodiment, among the actual relative symbols in each group, the methods for estimating the i-th group of actual relative symbols that are incorrect include at least the following two. The methods for determining the correct i-th group of absolute symbols X2 and Y2 and the correct i+1 group of absolute symbols X3 and Y3, as well as the corresponding correction methods, are different for different estimation methods. The following describes the different estimation methods, the methods for determining the correct i-th group of absolute symbols and the correct i+1 group of absolute symbols under each estimation method, and the correction methods under each estimation method.

[0095] refer to Figure 4 The first error correction process shown includes at least the following steps:

[0096] Step 41: Use the first set of actual relative symbols Δx1, Δy1, Δxy1 and Δyx1 to determine the correct first set of absolute symbols X1 and Y1 and the second set of absolute symbols X2 and Y2.

[0097] Specifically, determining the correct first-group absolute symbols X1 and Y1 and the second-group absolute symbols X2 and Y2 using the first group of actual relative symbols involves the following steps:

[0098] Step 1: Select a group of untraversed absolute symbols from each group of candidate absolute symbols as the first group of absolute symbols to be determined.

[0099] The candidate absolute symbols include all possibilities of absolute symbols. For example, the candidate absolute symbols are shown in Table 2 below. Each group of absolute symbols includes 2 bits corresponding to X polarization and 2 bits corresponding to Y polarization. Suppose that the first group of absolute symbols to be determined selected from Table 2 is the first group of symbols 0000 in Table 2. At this time, the first group of symbols 0000 is an absolute symbol that has already been traversed.

[0100] Table 2:

[0101]

[0102]

[0103] Step 2: Using the information items relative symbols Δx1 and Δy1 in the first group of actual relative symbols, and the first group of absolute symbols to be determined, determine the second group of absolute symbols to be determined corresponding to the information items.

[0104] Specifically, the symbol corresponding to the X polarization in the first group of absolute symbols to be determined is calculated with the information item Δx in the first group of actual relative symbols to obtain the symbol corresponding to the X polarization in the second group of absolute symbols to be determined; the symbol corresponding to the Y polarization in the second group of absolute symbols to be determined is calculated with the information item Δy in the first group of actual relative symbols to obtain the symbol corresponding to the Y polarization in the second group of absolute symbols to be determined.

[0105] Schematic representation: The sign relationship between the (k-1)th X polarization and the (k-1)th Δx and the kth X polarization is shown in Table 3 below. In Table 3, k represents the current time, k-1 represents the previous time, and X... I and X Q The two bits representing the absolute phase of X polarization, Δx I and Δx Q This represents the two bits indicating the relative phase between two adjacent symbols before and after X polarization. According to Table 3 (XOR cannot include all correspondences), the phase corresponding to the absolute symbol X(k-1) plus the phase corresponding to the relative symbol Δx equals the phase corresponding to the absolute symbol X(k). From the relationship between phases, the logical relationship between symbol bits can be obtained. Table 3 uses X polarization as an example. In actual implementation, the symbol relationship for Y polarization is the same as in Table 3, i.e., X in Table 3... I and X Q Change to Y I and Y Q 1. Δx in Table 3 I and Δx Q Change to Δy I and Δy Q The specific details of this embodiment will not be listed here.

[0106] For example: if the first group of absolute symbols to be determined is initialized as (X1: 00, Y1: 00), and the first group of actual relative symbols is (Δx1: 11, Δy1: 10), then according to the relationship between the symbols shown in Table 3, the second group of absolute symbols to be determined corresponding to the information item is (X2: 00, Y2: 01).

[0107] Table 3:

[0108]

[0109]

[0110] Step 3: Using the cross-term relative symbols Δxy1 and Δyx1 in the first group of actual relative symbols, and the first group of absolute symbols to be determined, determine the second group of absolute symbols to be determined corresponding to the cross-term.

[0111] Specifically, the symbol corresponding to the X polarization in the first group of absolute symbols to be determined is calculated with the cross term Δxy in the first group of actual relative symbols to obtain the symbol corresponding to the X polarization in the second group of absolute symbols to be determined corresponding to the cross term; the symbol corresponding to the Y polarization in the first group of absolute symbols to be determined is calculated with the cross term Δyx in the first group of actual relative symbols to obtain the symbol corresponding to the Y polarization in the second group of absolute symbols to be determined corresponding to the cross term.

[0112] Schematic representation: The sign relationships between the (k-1)th Y polarization, the (k-1)th Δxy, and the kth X polarization are shown in Table 4 below. In Table 4, k represents the current time, k-1 represents the previous time, and Y... I and Y Q The two bits representing the absolute phase of the Y-polarization, Δxy I and Δxy Q This represents the two bits indicating the relative phase between two adjacent symbols before Y polarization and after X polarization. According to Table 4, the phase corresponding to the absolute symbol Y(k-1) plus the phase corresponding to the cross term relative symbol Δxy equals the phase corresponding to the absolute symbol X(k). Similarly, the logical relationship between symbol bits can be obtained from the phase relationship. Table 4 uses the relationship between the symbols before Y polarization and the current X polarization as an example. In actual implementation, the symbol relationship before X polarization and after Y polarization is the same as in Table 4, i.e., the X in Table 4... I and X Q Change to Y I and Y Q 1. Y in Table 4 I and Y Q The change to X I and X Q 1. Δxy in Table 4 I and Δxy Q Change to Δyx I and Δyx Q The specific details of this embodiment will not be listed here.

[0113] For example: if the first group of absolute symbols to be determined is initialized as (X1: 00, Y1: 00), and the first group of actual relative symbols is (Δxy1: 01, Δyx1: 00), then according to the relationship between the symbols shown in Table 4, the second group of absolute symbols to be determined corresponding to the cross term is (X2: 10, Y2: 11).

[0114] Table 4:

[0115] <![CDATA[Y I (k-1)]]> <![CDATA[Y Q (k-1)]]> <![CDATA[Δxy I (k-1)]]> <![CDATA[Δxy Q (k-1)]]> <![CDATA[X I (k)]]> <![CDATA[X Q (k)]]> 0 0 0 0 1 1 0 0 0 1 1 0 0 0 1 0 0 1 0 0 1 1 0 0 0 1 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 1 1 0 1 1 0 0 0 0 1 1 0 0 1 1 1 1 0 1 0 0 0 1 0 1 1 1 0 1 1 0 0 0 0 1 1 0 1 0 1 1 1 1 0 1 0 1 1 1 1 1 1

[0116] Step 4: If the second group of absolute symbols to be determined for the information item is equal to the second group of absolute symbols to be determined for the cross item, determine that the first group of absolute symbols to be determined is the correct first group of absolute symbols and the second group of absolute symbols to be determined is the correct second group of absolute symbols.

[0117] Step 5: If the second group of absolute symbols to be determined corresponding to the information item is not equal to the second group of absolute symbols to be determined corresponding to the cross item, execute the step of determining a group of untraversed absolute symbols from each group of candidate absolute symbols as the first group of absolute symbols to be determined again, so as to update the assignment of the first group of absolute symbols to be determined and re-determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2.

[0118] For example, for the same first group of absolute symbols to be determined, the second group of absolute symbols to be determined corresponding to the information item determined in step 2 is (X2: 00, Y2: 01), and the second group of absolute symbols to be determined corresponding to the intersection item determined in step 3 is (X2: 10, Y2: 11). The two are not equal. Therefore, the first group of absolute symbols to be determined is incorrect. At this time, a group of untraversed absolute symbols can be re-determined from the candidate absolute symbols of each group as the first group of absolute symbols to be determined. That is, step 1 is executed again to re-determine the correct first group of absolute symbols and the second group of absolute symbols.

[0119] Optionally, if the correct first group of absolute symbols and second group of absolute symbols are not determined using the first group of actual relative symbols, then the first group of actual relative symbols is determined to be incorrect, and the next group of actual relative symbols is used as the first group of actual relative symbols to execute step 41 again.

[0120] Step 42: Using the first group of absolute symbols, the second group of absolute symbols, and the actual relative symbols corresponding to the cross terms in each group of actual relative symbols, determine the estimated value of each group of absolute symbols.

[0121] Specifically, assuming a total of N sets of actual relative symbols are obtained, after obtaining the first and second sets of absolute symbols, the symbol corresponding to the X polarization in the second set of absolute symbols can be logically operated with the intersection term Δyx2 in the second set of actual relative symbols, and the symbol corresponding to the Y polarization in the second set of absolute symbols can be logically operated with the intersection term Δxy2 in the second set of actual relative symbols to obtain the estimated value of the third set of absolute symbols; the symbol corresponding to the X polarization in the third set of estimated absolute symbols can be logically operated with the intersection term Δyx3 in the third set of actual relative symbols, and the symbol corresponding to the Y polarization in the third set of estimated absolute symbols can be logically operated with the intersection term Δxy3 in the third set of actual relative symbols to obtain the estimated value of the fourth set of absolute symbols. This process is repeated until the estimated value of the N+1 set of absolute symbols is estimated using the Nth set of estimated absolute symbols and the Nth set of actual relative symbols, thus obtaining the estimated values ​​of each set of absolute symbols.

[0122] Refer to Table 4 above for the symbol relationship between each group of absolute symbols and the actual relative symbols corresponding to the cross terms, and obtain the next group of absolute symbols corresponding to each group of absolute symbols.

[0123] refer to Figure 5 Assuming that the first group of absolute symbols determined using the first group of actual relative symbols (Δx1: 11, Δy1: 10, Δxy1: 01, Δyx1: 00) is (X1: 00, Y1: 11) and the second group of absolute symbols is (X2: 00, Y2: 10), then based on Table 4, when the cross term in the second group of actual relative symbols is (Δxy2: 00, Δyx2: 11), the estimated value of the third group of absolute symbols is (estX3: 10, estY3: 11). When the cross term in the third group of actual relative symbols is (Δxy2: 00, Δyx2: 11), the estimated value of the fourth group of absolute symbols is (estX4: 11, estY4: 01). Figure 5 In the diagram, the symbol on the solid line crossing arrow indicates the symbol corresponding to the crossed item, and the symbol on the dot indicates the absolute symbol.

[0124] Step 43: Use the estimated absolute symbols of each group to determine the estimated values ​​corresponding to the information items in the actual relative symbols of each group.

[0125] Specifically, for two adjacent sets of estimated absolute symbols, the symbols corresponding to the X polarization in the two sets of estimated absolute symbols are compared to obtain the estimated value of Δx in the information item; the symbols corresponding to the Y polarization in the two sets of estimated absolute symbols are compared to obtain the estimated value of Δy in the information item. The relationship between the symbols corresponding to the X polarization in the two sets of estimated absolute symbols and Δx, and the relationship between the symbols corresponding to the Y polarization in the two sets of estimated absolute symbols and Δy, can be determined through Table 3.

[0126] For example: Reference Figure 6 ,according to Figure 5 The estimated absolute symbols of each group can be used to obtain the estimated values ​​of the information items in the second group of actual relative symbols (estΔx2: 01, estΔy2: 10) and the estimated values ​​of the information items in the third group of actual relative symbols (estΔx3: 01, estΔy3: 01).

[0127] Step 44: Compare the information items in each group of actual relative symbols with the corresponding estimated values ​​of the information items to determine the estimation error locations corresponding to X polarization and Y polarization, and then proceed to steps 45 or 48.

[0128] For example: The actual relative symbol corresponding to the information item in the second group of actual relative symbols is (Δx2: 01, Δy2: 11), which is not equal to the estimated relative symbol (estΔx2: 01, estΔy2: 01) corresponding to the information item in the second group. The difference is at the position corresponding to the symbol of Y polarization. Therefore, the estimation error position is the estimation error position corresponding to Y polarization.

[0129] Step 45: If the estimated error location is not continuous, determine that the estimated error location is an error in the information item Δx2 or Δy2 in the corresponding i-th group of actual relative symbols.

[0130] Since the estimated value of the absolute symbol is determined using the actual relative symbol corresponding to the cross term in the actual relative symbol, if the actual relative symbol corresponding to one cross term is incorrect, then the estimated values ​​of all subsequent absolute symbols will be incorrect. In this case, the estimated relative symbol corresponding to the information item estimated using the estimated value of that absolute symbol will also be continuously incorrect. Based on this principle, if the estimation errors are not consecutive, it indicates that the actual relative symbol corresponding to the cross term is not incorrect; if the estimation errors are consecutive, it indicates that the actual relative symbol corresponding to the cross term is incorrect.

[0131] In this step, "continuous error location" means that there are at least two consecutive sets of information items whose actual relative signs are not equal to their corresponding estimated relative signs.

[0132] Step 46: If the information item Δx2 or Δy2 in the i-th group of actual relative symbols is incorrect, obtain the estimated absolute symbols X2 and Y2 of the i-th group from the (i-1)-th group of relative symbols to obtain the correct absolute symbols X2 and Y2 of the i-th group; determine the error position of the information item in the i-th group of actual relative symbols, which is X-polarization or Y-polarization; use the i-th absolute symbol X2 or Y2 corresponding to the correct position and the cross term Δyx2 or Δxy2 of the i-th group of actual relative symbols to determine the corresponding (i+1)-th absolute symbol Y3 or X3; use the absolute symbol Y3 or X3 in the (i+1)-th group corresponding to the correct position and the i+1-th group of actual relative symbols to determine the (i+2)-th absolute symbols X4 and Y4; based on the (i+2)-th absolute symbols X4 and Y4 corresponding to the correct position and the information items Δx3 and Δy3 of the (i+1)-th group of actual relative symbols, deduce the (i+1)-th absolute symbol X3 or Y3 corresponding to the error position.

[0133] This step is an alternative to step 304 in the first method of determining the error location when an information item is incorrect.

[0134] Since the error locations are discontinuous, if the information item in the i-th group of actual relative symbols is incorrect, it means that the (i-1)-th group of actual relative symbols is correct. Therefore, the i-th group of estimated absolute symbols determined based on the (i-1)-th group of actual relative symbols is correct. Thus, the i-th group of estimated absolute symbols can be used as the correct i-th group of absolute symbols.

[0135] Figure 6 In the original text, the error location is in the information item of the second actual relative symbol, specifically the position where the Y polarization differs from the comparison result. That is, the error location is the Y polarization. However, it cannot be determined at this point whether it is the estimated relative symbol pair of the Y polarization in the information item or the actual relative symbol pair of the Y polarization, but it can be determined that the information item of the X polarization is correct. Therefore, the symbol corresponding to the X polarization in the second group of absolute symbols (X2: 00) and the information item (Δx2: 01) in the second group of actual relative symbols can be used to obtain the symbol corresponding to the X polarization in the third group of absolute symbols (X3: 10). Combining X3 with the correct relative symbols in the third group (Δx3: 01, Δy3: 01, Δxy3: 00, Δyx3: 11), according to Tables 3 and 4, the fourth group of absolute symbols (X4: 11, Y4: 01) can be obtained. Then, from Table 3, based on Y4: 01 and Δy3: 01, the Y3 corresponding to the error location is deduced to be 11.

[0136] Step 47: Use the i-th absolute symbols X2 and Y2 corresponding to the error position and the (i+1)-th absolute symbols X3 and Y3 corresponding to the error position to correct the relative symbols Δx2 or Δy2 of the information item corresponding to the error position in the i-th group of actual relative symbols; if the actual relative symbols with errors have not been traversed completely, execute step 44 again; if the actual relative symbols with errors have been traversed completely, the process ends.

[0137] This step is an alternative to step 305 when an information item is incorrect in the first method of determining the error location.

[0138] Given Y2: 10 and Y3: 11, according to the logical relationship in Table 3, the information item Δy2 for Y polarization is 01. Therefore, the information item Δy2 at the error location can be corrected from 11 to 01.

[0139] Step 48: If the estimation error positions are consecutive, determine that the first estimation error position is the error of the cross term Δxy2 or Δyx2 in the corresponding i-th group of actual relative symbols.

[0140] Step 49: If the cross term Δxy2 or Δyx2 in the i-th group of actual relative symbols is incorrect, obtain the i-th group of absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols; use the i-th group of absolute symbols X2 and Y2, and the actual relative symbols Δx2 and Δy2 corresponding to the information terms in the i-th group of actual relative symbols, to determine the (i+1)-th group of absolute symbols X3 and Y3.

[0141] This step is an alternative to step 304 when an error occurs in the first method of determining the error location.

[0142] In this embodiment, if the cross-item is determined to be incorrect, the information item is assumed to be correct. Therefore, according to Table 3, the i-th group of absolute symbols can be obtained from the actual relative symbols corresponding to the information items in the i-th group of actual relative symbols and the i-th group of absolute symbols.

[0143] Step 491: Based on the relative symbol relationship between the i-th group of absolute symbols and the (i+1)-th group of absolute symbols, correct the relative symbols Δxy2 and Δyx2 corresponding to the cross terms in the i-th group of actual relative symbols, and after the symbol correction, trigger the execution of the step of using the first group of actual relative symbols to determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2, that is, re-execute step 41.

[0144] This step is an alternative to step 305 when an error occurs in the first method of determining the error location.

[0145] Specifically, according to Table 4, the i-th group of relative symbols can be determined from the i-th group of absolute symbols and the (i+1)-th group of absolute symbols, and the i-th group of relative symbols is determined as the corrected relative symbols.

[0146] refer to Figure 7 The second error correction process shown includes at least the following steps:

[0147] Step 71: For each group of actual relative symbols, determine the information item E. x E x d * and E y E y d *Corresponding actual relative sign, and cross term E x E y d * and E y E x d* Determine whether the corresponding actual relative symbol satisfies the constraint Δxy+Δyx=Δx+Δy; determine the index of the error position of the actual relative symbol that does not satisfy the constraint; determine whether the position of the erroneous actual relative symbol is continuous; determine the error index of the non-continuous erroneous actual relative symbol as the i-th group of erroneous actual relative symbols, and perform error correction in combination with the two groups of non-erroneous relative symbols before and after.

[0148] Where Δxy represents the cross term E x E y d *Corresponding actual relative sign, Δyx represents the cross term E y E x d *Corresponding actual relative sign, Δx represents E x E x d *Corresponding actual relative symbol, Δy represents E y E y d * Corresponds to the actual relative symbol.

[0149] Information item Δx i =x i+1 -x i ;Δy i =y i+1 -y i ; Intersection term Δxy i =x i+1 -y i ;Δyx i =y i+1 -x i The constraint is the relationship between relative phases (relative signs). According to the phase relationship shown in Table 1, assuming the relative sign corresponding to the information item is 0101, the sum of the phases is pi. The actual relative sign corresponding to the cross item is 0011, and the sum of the phases is also pi. If the two are equal, it is not an error location; if the sum of the phases is not equal, it is marked as a potentially error location.

[0150] Step 72: Use the (i-1)th group of actual relative symbols to determine the correct (i-1)th group of absolute symbols X1 and Y1 and the i-th group of absolute symbols X2 and Y2.

[0151] The implementation method of this step is the same as that of step 41, except that the first group is changed to the (i-1)th group and the second group is changed to the ith group.

[0152] Steps 72-74 are alternative steps to step 304 in the second method of determining the error location.

[0153] Step 73: Using the absolute symbols X2 and Y2 of the i-th group and the actual relative symbols Δx2 and Δy2 corresponding to the information items in the i-th group of actual relative symbols, determine the first estimated symbol corresponding to X polarization and the second estimated symbol corresponding to Y polarization in the (i+1)-th group of absolute symbols; using the absolute symbols X2 and Y2 of the i-th group and the actual relative symbols corresponding to the cross terms Δxy2 and Δyx2 in the i-th group of actual relative symbols, determine the third estimated symbol corresponding to X polarization and the fourth estimated symbol corresponding to Y polarization in the (i+1)-th group of absolute symbols.

[0154] refer to Figure 5 Assume the actual relative sign of the i-th group is the second group, and the actual relative sign of the second group is (Δx2: 01, Δy2: 11, Δxy2: 00, Δyx2: 11). Then, the absolute sign of the second group, determined using the actual relative sign of the first group, is (X2: 00, Y2: 10). According to Table 3, from the absolute sign of the second group and (Δx2: 01, Δy2: 11), we can obtain the first estimated sign 10 and the second estimated sign 10. According to Table 4, from the absolute sign of the second group and (Δxy2: 00, Δyx2: 11), we can obtain the third estimated sign 10 and the fourth estimated sign 11.

[0155] Step 74: If the first estimated symbol is equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, then use the estimated symbol corresponding to X polarization, Δx, Δx3, and Δyx3 in the (i+1)th group of actual relative symbols to determine the (i+2)th group of absolute symbols X4 and Y4; use the (i+2)th group of absolute symbols X4 and Y4 and the (i+1)th group of actual relative symbols to reverse-engineer the symbol corresponding to Y polarization in the (i+1)th group of absolute symbols. The symbol corresponding to Y polarization and the first estimated symbol constitute the (i+1)th group of absolute symbols Y3 and X3; if the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is equal to the fourth estimated symbol, then use the estimated symbol corresponding to Y polarization... The estimated symbols, Δy3 and Δxy3 in the (i+1)th group of actual relative symbols are used to determine the (i+2)th group of absolute symbols X4 and Y4. Using the (i+2)th group of absolute symbols X4 and Y4 and the (i+1)th group of actual relative symbols, the symbols corresponding to X polarization in the (i+1)th group of absolute symbols are deduced. The symbols corresponding to X polarization and the second estimated symbols constitute the (i+1)th group of absolute symbols X3 and Y3. If the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, then the (i+1)th group of absolute symbols Δx3, Δy3, Δxy3 and Δyx3 are used to determine the (i+1)th group of absolute symbols and the (i+2)th group of absolute symbols X4 and Y4.

[0156] As shown in the example in step 73, the first estimated symbol X3:10 and the third estimated symbol estX3:10 are the same, but the third estimated symbol Y3:10 and the fourth estimated symbol estY3:11 are different. At this point, the absolutely correct symbol for X polarization is considered to be the estimated symbol 10. Then, according to Table 3, the absolutely correct symbol for X polarization in the fourth group is obtained from the estimated symbol 10 and Δx3 in the actual relative symbols of the third group, as X4:11. According to Table 4, the absolutely correct symbol for Y polarization in the fourth group is derived from the estimated symbol 10 and Δx3 in the actual relative symbols of the third group, as Y4:01. According to Table 3, the symbol corresponding to Y polarization in the third group is deduced from Y4:01 in the fourth group and Δy3 in the actual relative symbols of the third group, as Y3:11. The symbol corresponding to Y polarization, Y3:11, and the first estimated symbol X3:10 constitute the third group of absolute symbols.

[0157] Similarly, when the first and third estimation symbols are different, but the third and fourth estimation symbols are the same, the principle is the same as described above, and this embodiment will not be illustrated further.

[0158] When the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, the (i+1)th group of actual relative symbols is used to determine the (i+1)th group of absolute symbols and the (i+2)th group of absolute symbols. The specific determination process is the same as the implementation of step 41, except that the first group is replaced by the (i+1)th group and the second group is replaced by the (i+2)th group. This embodiment will not be described in detail here.

[0159] Step 75: Use the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 to correct the actual relative symbols Δx2 and Δy2 corresponding to the information items in the i-th group of actual relative symbols.

[0160] This step is an alternative to step 305 in the second method of determining the error location.

[0161] According to Table 3, X in the i-th group of absolute symbols i And the X in the (i+1)th group of absolute symbols i+1 We can obtain Δx i ;Y from the i-th group of absolute symbols i And Y in the (i+1)th group of absolute symbols i+1 We can obtain Δy i .

[0162] Step 76: Determine whether the current i-th group of actual relative symbols is the last group of erroneous actual relative symbols. If yes, the process ends; otherwise, take the next group of erroneous actual relative symbols as the i-th group of actual relative symbols and execute step 72 again.

[0163] Compared to the first error correction method, the second error correction method is more efficient. The first error correction method is less efficient because it requires correcting the cross term, recalculating the absolute sign, obtaining the relative sign, and comparing them again when an error occurs.

[0164] The second error correction method utilizes the constraint relationship between the relative symbols of information items and the relative symbols of cross items. Each index is independent, and an error in a relative symbol will not cause subsequent index positions to appear consecutively. Therefore, error correction can be performed on each individual error position simultaneously.

[0165] In summary, the error correction method based on a differential modulation polarization demultiplexing system provided in this embodiment obtains N groups of electrical signals from an electrical domain polarization demultiplexing receiver through differential detection. Each group of electrical signals includes E H ·E H d *Electrical signal, E V ·E V d *Electrical signal, E H ·E V d *Electrical signals and E V ·E H d *Electrical signals, where N is an integer greater than 1; each group of electrical signals is demultiplexed using an adaptive algorithm to obtain the actual relative symbol, which includes the information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d * Corresponding actual relative symbols Δxy and Δyx; among the received actual relative symbols, the i-th group of actual relative symbols that is estimated incorrectly; obtain the correct i-th group of absolute symbols X2 and Y2 and the correct i+1-th group of absolute symbols X3 and Y3 corresponding to the i-th group of actual relative symbols; each group of absolute symbols includes the absolute symbol corresponding to X polarization and the absolute symbol corresponding to Y polarization; use the i-th group of absolute symbols X2 and Y2 and the i+1-th group of absolute symbols X3 and Y3 to correct the i-th group of actual relative symbols; can improve the accuracy of the polarization demultiplexing receiver structure with redundant replicas and improve the receiver sensitivity; since the cross terms obtained during demultiplexing can be used to verify and correct the information, the accuracy of the polarization demultiplexing receiver structure with added redundant replicas can be improved.

[0166] In addition, by determining the actual relative symbol of the error through the constraint relationship between the relative symbol of the information item and the relative symbol of the cross item, it is not necessary to redetermine the actual relative symbol of the error when the cross item is wrong, which can improve the efficiency of symbol error correction.

[0167] To facilitate understanding of the above error correction method, several examples are provided below to illustrate it.

[0168] Example 1: The received relative symbols are not consecutively positioned. For example, the second error correction method determines that only Δy2 in the second group of actual relative symbols is incorrect. (Refer to...) Figure 6 Alternatively, by using the second error correction method, it can be determined that only Δy2 and Δyx2 in the second group of actual relative symbols are incorrect, as shown in the reference. Figure 8 The specific error correction process includes:

[0169] Step 1: Obtain the first set of correct absolute symbols X1, Y1 and the second set of correct absolute symbols X2, Y2 from the first set of correct relative symbols.

[0170] Step 2: Obtain the third set of uncertain absolute symbols X3, Y3, estX3, estY3 from the second set of correct absolute symbols X2, Y2 and the second set of uncertain relative symbols.

[0171] Step 3: Determine if X3 and estX3 are equal, and if Y3 and estY3 are equal. If they are equal, the current absolute symbols in the third group are considered correct. Combine the correct relative symbols in the third group to determine the fourth group of absolute symbols. Use the fourth group of absolute symbols and the third group of relative symbols to determine the remaining absolute symbols in the third group. Then, use the already determined absolute symbols in the third and second groups to perform error correction.

[0172] Example 2: If three or more of the actual relative symbols in the same group are incorrect, for example, if Δy2, Δxy2, and Δyx2 in the second group are incorrect, refer to... Figure 9 In Example 1, step 3, the two absolute symbols of the third group X and Y are not equal. The third group of absolute symbols and the fourth group of absolute symbols X3, X4, Y3, Y4 can be determined through the initialization procedure (i.e., the process in step 41) based on the correct actual relative symbols of the third group. Then, the relative symbols are obtained from the determined second and third group of absolute symbols according to the logical relationship, and the information is corrected.

[0173] refer to Figure 10 and Figure 11 The curve marked with hollow symbols is Figure 1The performance curves of the Polarization Multiplexing (PM) Differential Phase Shift Keying (DQPSK) (PM-DQPSK) system without error correction are shown. The solid marks represent the performance curves achieved after using the second error correction method. The slope of the performance curve after error correction is greater than that of the original curve, meaning that the better the system performance, the greater the improvement in system performance.

[0174] Specifically, Figure 10 The curve represents the system performance Q-factor versus the optical signal-to-noise ratio (OSNR) per bit. In an OSNR-constrained system with only optical noise, the system performance can be improved by 0.7 dB when the system Q-factor reaches the forward error correction (FEC) decision threshold (indicated by the dashed line); and the system performance can be improved by more than 0.7 dB when the system Q-factor reaches 10 dB or more.

[0175] Figure 11 The graph shows the system performance (Q factor) versus signal-to-noise ratio per bit (EbN0). In a power-constrained system with only electrical noise, the system is unaffected by noise during transmission. The noise in the system originates from thermal and scattering noise during receiver photoelectric detection, as well as electrical noise introduced by the transimpedance amplifier. When the system Q factor reaches the FEC decision threshold (indicated by the dashed line), the system performance can be improved by 1.9 dB; when the system Q factor reaches 10 dB or more, the system performance can be improved by more than 2 dB after using error correction algorithm 2.

[0176] Optionally, this application also provides a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the error correction method based on a differential modulation polarization demultiplexing system described in the above method embodiments.

[0177] Optionally, this application also provides a computer product including a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the error correction method based on a differential modulation polarization demultiplexing system described in the above method embodiments.

[0178] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0179] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. An error correction method based on a differential modulation polarization demultiplexing system, characterized in that, The method includes: The N sets of electrical signals required for electric domain polarization demultiplexing are obtained by differential detection. Each set of electrical signals includes E H ·E H d *Electrical signal, E V ·E V d *Electrical signal, E H ·E V d *Electrical signals and E V ·E H d *Electrical signal, where N is an integer greater than 1; An adaptive algorithm is used to demultiplex the electrical signal to obtain the actual relative symbol, which includes the information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d *The corresponding actual relative symbols Δxy and Δyx; Among the received actual relative symbols, the i-th actual relative symbol that was estimated incorrectly, where i is a positive integer less than N; Obtain the correct absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols, and the correct absolute symbols X3 and Y3 corresponding to the (i+1)-th group of absolute symbols; each group of absolute symbols includes the absolute symbol corresponding to the X polarization and the absolute symbol corresponding to the Y polarization. The i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 are used to correct the i-th group of actual relative symbols; Among the groups of actual relative symbols, the i-th group of actual relative symbols that was estimated incorrectly includes: Use the first group of actual relative symbols Δx1, Δy1, Δxy1 and Δyx1 to determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2; Using the first group of absolute symbols, the second group of absolute symbols, and the actual relative symbols corresponding to the cross terms in each group of actual relative symbols, determine the estimated value of each group of absolute symbols; Use the estimated absolute sign of each group to determine the estimated value corresponding to the information item in the actual relative sign of each group; Compare the information items in each group of actual relative symbols with the corresponding estimated information items to determine the estimation error locations corresponding to X polarization and Y polarization. When the estimated error location is discontinuous, the estimated error location is determined to be an error in the information item Δx2 or Δy2 in the corresponding i-th group of actual relative symbols; When the estimation error locations are consecutive, the first estimation error location is determined to be an error in the cross term Δxy2 or Δyx2 in the corresponding i-th group of actual relative symbols. The step of obtaining the correct absolute symbols X2 and Y2 corresponding to the i-th group of relative symbols and the correct absolute symbols X3 and Y3 of the (i+1)-th group includes: If the information item Δx2 or Δy2 in the i-th group of actual relative symbols is incorrect, the i-th group of absolute symbols X2 and Y2 are obtained from the (i-1)-th group of relative symbols to obtain the correct i-th group of absolute symbols X2 and Y2. Determine the error location of the information item in the i-th group of actual relative symbols, wherein the error location is in X polarization or Y polarization; Use the absolute symbol X2 or Y2 of the i-th group corresponding to the position without error, and the intersection term Δyx2 or Δxy2 of the actual relative symbol of the i-th group to determine the (i+1)-th absolute symbol Y3 or X3. Use the absolute symbol Y3 or X3 in the (i+1)th group corresponding to the position without error and the actual relative symbol in the (i+1)th group to determine the absolute symbols X4 and Y4 in the (i+2)th group. Based on the (i+2)th absolute symbols X4 and Y4 corresponding to the correct position, and the information items Δx3 and Δy3 of the (i+1)th actual relative symbols, the (i+1)th absolute symbol X3 or Y3 corresponding to the incorrect position can be deduced. The step of correcting the actual relative symbols of the i-th group using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 includes: Using the i-th absolute symbols X2 and Y2 corresponding to the error position, and the (i+1)-th absolute symbols X3 and Y3 corresponding to the error position, correct the relative symbols Δx2 or Δy2 of the information item corresponding to the error position in the i-th group of actual relative symbols.

2. The method according to claim 1, characterized in that, The step of obtaining the correct absolute symbols X2 and Y2 corresponding to the actual relative symbols of the i-th group and the correct absolute symbols X3 and Y3 of the (i+1)-th group includes: If the cross term Δxy2 or Δyx2 in the i-th group of actual relative symbols is incorrect, obtain the i-th group of absolute symbols X2 and Y2 corresponding to the i-th group of actual relative symbols. Using the i-th group of absolute symbols X2 and Y2, and the i-th group of actual relative symbols corresponding to the information items Δx2 and Δy2, determine the (i+1)-th group of absolute symbols X3 and Y3.

3. The method according to claim 2, characterized in that, The step of correcting the intersection terms Δxy2 and Δyx2 of the actual relative symbols of the i-th group using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 includes: Based on the relative symbol relationship between the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3, the relative symbols Δxy2 and Δyx2 corresponding to the cross terms in the i-th group of actual relative symbols are corrected, and after the symbol correction, the step of using the first group of actual relative symbols to determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2 is triggered.

4. The method according to claim 1, characterized in that, The method of using the first group of actual relative symbols to determine the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2 includes: From each group of candidate absolute symbols, select a group of untraversed absolute symbols as the first group of absolute symbols to be determined; Using the information items relative symbols Δx1 and Δy1 in the first group of actual relative symbols, and the first group of absolute symbols to be determined, determine the second group of absolute symbols to be determined corresponding to the information items; Using the cross-term relative symbols Δxy1 and Δyx1 in the first group of actual relative symbols, and the first group of absolute symbols to be determined, determine the second group of absolute symbols to be determined corresponding to the cross-term; If the second group of absolute symbols to be determined corresponding to the information item is equal to the second group of absolute symbols to be determined corresponding to the cross item, then the first group of absolute symbols to be determined is determined to be the correct first group of absolute symbols, and the second group of absolute symbols to be determined is determined to be the correct second group of absolute symbols. If the second group of absolute symbols to be determined corresponding to the information item is not equal to the second group of absolute symbols to be determined corresponding to the cross item, the step of determining a group of untraversed absolute symbols from each group of candidate absolute symbols as the first group of absolute symbols to be determined is executed again to update the assignment of the first group of absolute symbols to be determined, and the correct first group of absolute symbols X1 and Y1 and the second group of absolute symbols X2 and Y2 are re-determined.

5. The method according to claim 1, characterized in that, Among the various groups of actual relative symbols, the i-th group of actual relative symbols that was estimated incorrectly includes: For each set of actual relative symbols, determine the information item E. x E x d * and E y E y d *The corresponding actual relative symbols Δx and Δy, and the cross term E x E y d * and E y E x d *Do the corresponding actual relative symbols Δxy and Δyx satisfy the constraint Δxy + Δyx = Δx + Δy, where Δxy represents the cross term E? x E y d *Corresponding actual relative sign, Δyx represents the cross term E y E x d *Corresponding actual relative sign, Δx represents E x E x d *Corresponding actual relative symbol, Δy represents E y E y d *Corresponding to the actual relative symbol; Determine the index of the actual relative sign error location where the constraint condition is not met; Determine whether the actual relative positions of the errors are consecutive; The error index of the actual relative symbol that is not consecutively erroneous is determined as the i-th group of actual relative symbols that are erroneous, and error correction is performed in combination with the two groups of relative symbols that are not erroneous before and after it.

6. The method according to claim 5, characterized in that, The step of obtaining the correct absolute symbols X2 and Y2 corresponding to the actual relative symbols of the i-th group and the correct absolute symbols X3 and Y3 of the (i+1)-th group includes: Use the (i-1)th group of actual relative symbols to determine the correct (i-1)th group of absolute symbols X1 and Y1 and the i-th group of absolute symbols X2 and Y2; Using the i-th group of absolute symbols X2 and Y2 and the actual relative symbols Δx2 and Δy2 corresponding to the information items in the i-th group of actual relative symbols, determine the first estimated symbol corresponding to X polarization and the second estimated symbol corresponding to Y polarization in the (i+1)-th group of absolute symbols; Using the i-th group of absolute symbols X2 and Y2 and the actual relative symbols Δxy2 and Δyx2 corresponding to the cross terms in the i-th group of actual relative symbols, determine the third estimated symbol corresponding to X polarization and the fourth estimated symbol corresponding to Y polarization in the (i+1)-th group of absolute symbols; If the first estimated symbol is equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, then the estimated symbol corresponding to X polarization, Δx3 and Δyx3 in the (i+1)th group of actual relative symbols are used to determine the (i+2)th group of absolute symbols X4 and Y4; using the (i+2)th group of absolute symbols X4 and Y4 and the (i+1)th group of actual relative symbols, the symbol corresponding to Y polarization in the (i+1)th group of absolute symbols is deduced in reverse; the symbol corresponding to Y polarization and the first estimated symbol constitute the (i+1)th group of absolute symbols Y3 and X3. If the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is equal to the fourth estimated symbol, then the estimated symbol corresponding to Y polarization, Δy3 and Δxy3 in the (i+1)th group of actual relative symbols are used to determine the (i+2)th group of absolute symbols X4 and Y4; using the (i+2)th group of absolute symbols X4 and Y4 and the (i+1)th group of actual relative symbols, the symbol corresponding to X polarization in the (i+1)th group of absolute symbols is deduced in reverse; the symbol corresponding to X polarization and the second estimated symbol constitute the (i+1)th group of absolute symbols X3 and Y3. If the first estimated symbol is not equal to the third estimated symbol and the second estimated symbol is not equal to the fourth estimated symbol, then the (i+1)th group of actual relative symbols Δx3, Δy3, Δxy3 and Δyx3 are used to determine the (i+1)th group of absolute symbols X3 and Y3 and the (i+2)th group of absolute symbols X4 and Y4.

7. The method according to claim 6, characterized in that, The step of correcting the actual relative symbols of the i-th group using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3 includes: Using the i-th group of absolute symbols X2 and Y2 and the (i+1)-th group of absolute symbols X3 and Y3, the actual relative symbols Δx2 and Δy2 corresponding to the information items in the i-th group of actual relative symbols are corrected.

Images