Multi-element channel coding and decoding system based on iterative feedback

Abstract

The invention discloses a multivariate channel coding and decoding system based on iterative feedback, and belongs to the technical field of communication. Comprising a transmitting end and a receiving end, wherein the transmitting end comprises a multivariate LDPC (Low Density Parity Check) coding module, a serial-to-parallel conversion / mapping module and a modulator module; and the receiving end comprises a soft demodulator module, a serial-to-parallel conversion / demapping module and a multivariate LDPC soft decoding module. According to the invention, on the basis of multivariate LDPC coding, a special modulation constellation diagram is designed, so that each 16 QAM symbol corresponds to two mutually associated GF4 multivariate LDPC coding code elements, the average Euclidean distance of each code element is maximum, and iterative feedback decoding is introduced, so that the anti-noise performance is improved, and the performance of an optical communication system is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and more particularly to a multi-channel coding system based on iterative feedback. Background Art

[0002] With the rapid growth of users' demand for transmission distance and communication capacity, the speed increase rate and capacity expansion of modern optical fiber communication systems are facing severe challenges. Binary coding is a common coding method in existing optical communication systems. In its decoding process, the minimum Hamming distance rule is usually used. This method focuses on finding the minimum distance between the received signal and the transmitted signal.

[0003] Based on the existing spectrum resources, high-order modulation is an inevitable way to improve the spectrum efficiency of optical fiber communication systems. The minimum Euclidean distance rule is usually used for its judgment. This method is mainly based on the Euclidean distance between the received signal sample and the ideal signal sample. However, the final decoding process needs to be based on the minimum Hamming distance rule, which leads to performance loss. Unlike binary decoding, multi-element decoding directly encodes and decodes the received symbols, which can effectively avoid the aforementioned performance differences. This method considers the relationship between multiple symbols, not just binary bits, thereby improving the performance and efficiency of the system. By directly processing symbols, the channel capacity can be better utilized, the bit error rate can be reduced, and the overall communication quality can be improved.

[0004] LDPC code is a high-performance FEC error correction code that can approach the Shannon limit in the case of high code length and has broad prospects. Compared with binary LDPC coding, multi-dimensional LDPC coding is more suitable for high-order modulation and has better noise resistance.

[0005] In a binary channel coding system, an iterative feedback decoding system is introduced to provide feedback to the received signal through each decoding result, which can improve the anti-noise performance. Currently, there is no technical solution to combine it with multi-element coding.

[0006] Therefore, how to provide a multi-channel coding system based on iterative feedback is a problem that those skilled in the art need to solve urgently. Summary of the invention

[0007] In view of this, the present invention provides a multi-channel coding system based on iterative feedback. On the basis of multi-dimensional LDPC coding, a specially designed constellation diagram is used so that each symbol corresponds to two multi-dimensional LDPC coding codewords, and the average Euclidean distance of each codeword is maximized. At the same time, iterative feedback decoding is introduced to improve the anti-noise performance.

[0008] In order to achieve the above object, the present invention provides the following technical solutions:

[0009] A multi-channel coding system based on iterative feedback, comprising: a transmitting end and a receiving end;

[0010] Wherein, the transmitting end includes: a multi-element LDPC encoding module, a serial-to-parallel conversion / mapping module, and a modulator module;

[0011] The receiving end comprises: a soft demodulator module, a serial-to-parallel conversion / demapping module, and a multi-element LDPC soft decoding module.

[0012] The multi-element LDPC coding module adopts multi-element LDPC coding based on the finite field GF4, and the symbol values ​​of input and output are between 0 and 3.

[0013] The serial-to-parallel conversion / mapping module converts the symbols output by two consecutive LDPC encoding modules into a symbol S with a value between 0 and 15. n As the original signal, the calculation formula is:

[0014] S n =s 2n-1 ×4+s 2n ;

[0015] In the formula, s 2n-1 、s 2n are the symbols output by the two LDPC coding modules, S n The original signal.

[0016] The modulator module converts S n Modulate the upper channel to obtain 16QAM symbols

[0017] The preset constellation diagram is:

[0018] Each 16QAM symbol is mapped from two multi-dimensional LDPC codewords. The first multi-dimensional LDPC codeword corresponds to the four quadrants of the 16QAM constellation diagram; the second multi-dimensional LDPC codeword corresponds to the center, two sides and four vertices of the 16QAM constellation diagram. The two codewords are modulated into one symbol to obtain a constellation diagram viewed from the perspective of two multi-dimensional LDPC codewords at the same time. This constellation diagram construction method maximizes the average Euclidean distance between the two codewords.

[0019] The soft demodulator module is connected to an optical receiver, which receives an optical signal from an optical fiber and obtains a 16QAM symbol after DSP processing. n , the 16QAM symbol r n Input to the soft demodulator module, output the original signal S n Take the prior probability vector for each value.

[0020] The serial-to-parallel conversion / demapping module converts the prior probability vector into s output by the multivariate LDPC soft decoding module of the previous iteration in each iteration. 2n-1 、s 2n The corresponding corrected probability vector Weighted feedback, converted to s 2n-1 、s 2n The corresponding pre-correction probability vector

[0021] The interior of the serial-to-parallel conversion / demapping module is composed of a serial-to-parallel converter and two weighters;

[0022] The serial-to-parallel converter converts the feedback corrected probability vector sequence Divide and Two-way;

[0023] The first weighting device and the second weighting device are used to Weighted and

[0024] Then the serial-to-parallel converter and Merge into Output.

[0025] The multivariate LDPC soft decoding module is used to input a pre-correction value probability vector sequence Output corrected probability vector sequence and the decoding results c0, c1, ... c 2n-1 ,c 2n …and complete the verification.

[0026] The multivariate LDPC soft decoding module is internally divided into a soft decoding module, a verification module, and a decision module;

[0027] Among them, the soft decoding module input After soft decoding, the updated probability vector sequence is obtained

[0028] The updated probability vector sequence Input the decision module, and decide the code value by taking the maximum value in each vector

[0029] The check module outputs a code element sequence To perform the check, let the check matrix be H and the checksum be when When verification is established.

[0030] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a multi-channel coding system based on iterative feedback decoding, which combines 16QAM modulation and GF4 multi-dimensional LDPC coding (or other multi-dimensional coding), increases the Euclidean distance of code elements through a special 16QAM constellation diagram and introduces outer loop iterative decoding, thereby improving the error correction capability of the multi-dimensional LDPC code and thus improving the anti-noise performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0032] Figure 1 It is a schematic diagram of the overall structure of the system of the present invention;

[0033] Figure 2 A schematic diagram of the structure of the transmitter provided by the present invention;

[0034] Figure 3 A schematic diagram of the receiving end structure provided by the present invention;

[0035] Figure 4 The 16QAM constellation diagram provided by the present invention;

[0036] Figure 5 A constellation diagram corresponding to the first multi-element LDPC symbol corresponding to each 16QAM symbol provided by the present invention;

[0037] Figure 6 A constellation diagram corresponding to the second multi-element LDPC symbol corresponding to each 16QAM symbol provided by the present invention;

[0038] Figure 7 The present invention provides a constellation diagram viewed from the perspective of two multi-element LDPC symbols at the same time;

[0039] Figure 8 A schematic diagram of the structure of a serial-to-parallel conversion / demapping module provided by the present invention;

[0040] Fig. 9 A schematic diagram of the structure of a weighting device 1 provided by the present invention;

[0041] Fig.10 A schematic diagram of the structure of the weighter 2 provided by the present invention;

[0042] Fig.11 This is a schematic diagram of the structure of the multi-dimensional LDPC soft decoding module provided by the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The purpose of the present invention is to provide a multi-channel coding system based on iterative feedback, including: a transmitting end and a receiving end; wherein the transmitting end includes: a multi-LDPC coding module, a serial-to-parallel conversion / mapping module, and a modulator module; the receiving end includes: a soft demodulator module, a serial-to-parallel conversion / demapping module, and a multi-LDPC soft decoding module. 16QAM modulation and GF4 multi-LDPC coding (or other multi-coding) are combined, the Euclidean distance of the code element is increased through a special 16QAM constellation diagram, and an outer loop iterative feedback decoding is introduced to improve the error correction capability of the multi-LDPC code, thereby improving the anti-noise performance, and providing a solution for introducing an outer loop iterative decoding based on a special constellation diagram to improve the error correction capability of the multi-LDPC code soft decision algorithm.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Embodiment 1:

[0047] See also Figure 1-3 Embodiment 1 of the present invention discloses a multi-channel coding system based on iterative feedback, including: a transmitting end and a receiving end; wherein the transmitting end includes: a multi-LDPC encoding module, a serial-to-parallel conversion / mapping module, and a modulator module; the receiving end includes: a soft demodulator module, a serial-to-parallel conversion / demapping module, and a multi-LDPC soft decoding module.

[0048] Specifically, the multi-element LDPC coding module adopts multi-element LDPC coding based on the finite field GF4, and the code element values ​​of the input and output are 0-3.

[0049] Specifically, the serial-to-parallel conversion / mapping module converts the code elements s output by two consecutive LDPC encoding modules into 2n-1 s 2n Convert to a symbol S with a value between 0 and 15 n The calculation formula is:

[0050] S n =s 2n-1 ×4+s 2n ;

[0051] Specifically, the modulator module converts S n Modulate the upper channel to obtain 16QAM symbols It is a 16QAM signal and enters the channel for transmission.

[0052] Specifically, the preset constellation diagram is: according to the constellation diagram of the first multi-dimensional LDPC codeword corresponding to each 16QAM symbol and the constellation diagram of the second multi-dimensional LDPC codeword corresponding to each 16QAM symbol, two codewords are modulated into one symbol.

[0053] For details, see the constellation diagram Figure 4-Figure 7 As shown, Figure 4 is the 16QAM constellation diagram; from the first code element of each 16QAM symbol, we get Figure 5 The constellation diagram corresponding to the first code element shown in the figure; from the second code element of each 16QAM symbol, we get Figure 6 The constellation diagram corresponding to the second code element shown in the figure; a certain value of each code element contains all the values ​​of another code element. If we look at them together, Figure 7 The constellation diagram shown is viewed from the perspective of two code elements at the same time. The two code elements are modulated into a 16QAM symbol. The red color is the value of the first code element and the blue color is the value of the second code element.

[0054] Specifically, each 16QAM symbol is mapped from two multi-dimensional LDPC codewords. The first multi-dimensional LDPC codeword corresponds to the four quadrants of the 16QAM constellation diagram; the second multi-dimensional LDPC codeword corresponds to the center, two sides and four vertices of the 16QAM constellation diagram. The two codewords are modulated into one symbol to obtain a constellation diagram viewed from the perspective of two multi-dimensional LDPC codewords at the same time. This constellation diagram construction method maximizes the Euclidean distance between the two codewords.

[0055] Specifically, the optical receiver of the soft demodulator module receives the optical signal from the optical fiber and obtains the 16QAM symbol r after DSP processing. n (At the receiving end, the optical signal is received by the receiver and processed by DSP to obtain the 16QAM symbol), which is input into the soft demodulator module and outputs the original signal S n Take the prior probability vector for each value:

[0056]

[0057] Where i is S n The possible values ​​of .

[0058] Specifically, assuming that the channel noise conforms to a two-dimensional normal distribution, that is, N(0,0,σ,σ,0), the calculation method is:

[0059]

[0060] In the formula, θ is the normalization parameter, re means taking the real part, and im means taking the imaginary part;

[0061] σ is the standard deviation of the noise, It represents the probability that the nth code element takes the value i.

[0062] Specifically, the serial-to-parallel conversion / demapping module converts S n The corresponding probability vector After the last iteration, the LDPC decoder outputs s 2n-1 s 2n The corresponding corrected probability vector Weighted feedback, converted to s 2n-1 s 2n The corresponding pre-correction probability vector

[0063] Specifically, before correction:

[0064]

[0065] Code element s 2n-1 The probability of taking value i

[0066]

[0067] Code element s 2n The probability of taking value i

[0068] Correction (Feedback)

[0069]

[0070] Specifically, the calculation method is to normalize the probability of the modulation symbol corresponding to each coded symbol by weighting it with the feedback probability of another symbol after summing it up:

[0071]

[0072] α is a normalization parameter that satisfies

[0073] In the first calculation, there is no feedback value, the default So there is

[0074]

[0075] α is a normalization parameter that satisfies

[0076] For details, see Figure 8The serial-to-parallel conversion / demapping module consists of a serial-to-parallel converter and two weighters. The serial-to-parallel converter feeds back the (corrected) probability vector sequence Divide into two and The first weighting device 1 and the second weighting device 2 are used to respectively Weighted and Then through the parallel-to-serial converter and Merge into Output.

[0077] For details, see Fig. 9 , based on the multivariate coding weighting principle of iterative feedback, in the first weighter 1, the feedback sequence right Weighted The calculation formula is:

[0078]

[0079] For details, see Fig.10 , based on the multi-dimensional decoding weighting principle of iterative feedback, in the second weighter 2, the feedback sequence right Weighted The calculation formula is:

[0080]

[0081] Specifically, in iterative decoding, the demodulation in each iteration needs to use the decoding result of the previous time, but the demodulation in the first iteration does not need to use it. The input here is the output of the previous decoding.

[0082] Specifically, the multivariate LDPC soft decoding module inputs the pre-correction value probability vector sequence Output corrected probability vector sequence and the decoding results c0, c1, ... c 2n-1 ,c 2n …while completing the verification.

[0083] For details, see Fig.11 The internal structure of the multi-element LDPC soft decoding module is divided into a soft decoding module, a check module, and a decision module, among which:

[0084] Soft decoding module: input After soft decoding (such as BP decoding), the updated probability vector sequence is obtained

[0085] Decision module: Input The codeword value is determined by taking the maximum value in each vector

[0086] Verification module: output code element sequence To check, let the check matrix be H and the checksum be when When verification is established.

[0087] Embodiment 2:

[0088] Embodiment 2 of the present invention discloses a communication method using a multi-channel coding system based on iterative feedback provided by the present invention. The communication method is a communication method that combines multi-LDPC channel coding with a special constellation diagram and outer loop information transmission, and adds outer loop iterative decoding to the traditional decoding process, thereby improving its error correction capability and anti-noise capability.

[0089] Specifically, the FEC (forward error correction coding) of the present invention adopts a multivariate LDPC code based on the QC-LDPC code expansion of the 5G standard, and replaces the coefficients of the base matrix of the QC-LDPC code in the 5G standard with random GF4 coefficients to obtain the required base matrix of the multivariate QC LDPC code. Take the base matrix 1 in the 5GQC-LDPC standard for expansion. When the code rate is 0.5, set the expansion coefficient Z = 32, then there are 1408 GF4 code elements in code length and 704 GF4 code elements in information bits, which is converted to a code length of 2816 bits and 1408 bits of information bits. Use 16QAM transmission. Set the upper limit of the number of inner loop iterations to 50 and the upper limit of the number of outer loop iterations to 30 during the decoding process.

[0090] The specific implementation includes the following steps:

[0091] A1. Generate a check matrix H of random coefficients according to the base graph and the code length and code rate.

[0092] A2. The transmitter randomly generates 704 GF4 information sequences [g0, g1, … g 703 ];

[0093] A3. Replace [g0,g1,…g 703 ] Input the multivariate LDPC coding module, encode it according to the H matrix according to the characteristics of QC-LDPC coding, and get the coding sequence [s0, s1, …s 1407 ], the specific process refers to the 5G LDPC (Low-Density Parity-Check) standard.

[0094] A4. Enter the serial-to-parallel conversion / mapping module and obtain the GF16 symbol sequence [S0, S1, ... S 703 ].

[0095] A5. Enter the modulator module and convert it into a QAM16 symbol sequence [v0,v1,…v 703 ], modulated on a fiber channel.

[0096] A6. Receive the signal from the optical fiber and restore the symbol information sequence [r0, r1, ... r 703 ] Specifically, after DSP processing, the specific algorithm depends on the system configuration, and the present invention does not limit it.

[0097] A7. After the soft demodulator module, the probability vector sequence is obtained

[0098] A8. and the feedback (corrected) probability vector sequence of A9 output of the previous iteration (In the first cycle, the output of A9 is replaced by a probability vector of all 1s) Input the serial-to-parallel conversion / demapping module and output the pre-correction probability vector

[0099] A9. Input soft decoding module, after 50 rounds of QSPA decoding, verification is performed. If the verification is successful, the result sequence [c0, c1, ... c 1407 ], if the verification fails, the corrected probability vector sequence is output And jump to A8.

[0100] A10. When the number of loops between A8 and A9 reaches the upper limit of the number of iterations of the outer loop, a decoding failure message is output.

[0101] Based on the multi-element LDPC coding, the present invention designs a special modulation constellation diagram framework, so that each 16QAM symbol corresponds to two multi-element LDPC coding code elements, and the average Euclidean distance of each code element is maximized. At the same time, feedback decoding is introduced to improve the anti-noise performance and enhance the performance of the optical communication system.

[0102] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0103] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-channel coding system based on iterative feedback, characterized in that: include: Transmitter and receiver; Wherein, the transmitting end includes: a multi-element LDPC encoding module, a serial-to-parallel conversion / mapping module, and a modulator module; The receiving end comprises: a soft demodulator module, a serial-to-parallel conversion / demapping module, and a multi-element LDPC soft decoding module.

2. The multi-channel coding system based on iterative feedback according to claim 1, characterized in that: The multi-element LDPC coding module adopts multi-element LDPC coding based on the finite field GF4, and the symbol values ​​of input and output are between 0 and 3.

3. The multi-channel coding system based on iterative feedback according to claim 1, characterized in that: The serial-to-parallel conversion / mapping module converts the symbols output by two consecutive LDPC encoding modules into a symbol S with a value between 0 and 15. n As the original signal, the calculation formula is: S n =s 2n-1 ×4+s 2n ; In the formula, s 2n-1 、s 2n are the symbols output by the two LDPC coding modules, S n is the original signal.

4. The multi-channel coding system based on iterative feedback according to claim 3, characterized in that: The modulator module converts S n Modulate the upper channel to obtain 16QAM symbols 5. The multi-channel coding system based on iterative feedback according to claim 4, characterized in that: The preset constellation diagram is: Each 16QAM symbol is mapped from two multi-dimensional LDPC codewords. The first multi-dimensional LDPC codeword corresponds to the four quadrants of the 16QAM constellation diagram; the second multi-dimensional LDPC codeword corresponds to the center, two sides and four vertices of the 16QAM constellation diagram. The two codewords are modulated into one symbol to obtain a constellation diagram viewed from the perspective of two multi-dimensional LDPC codewords at the same time.

6. The multi-channel coding system based on iterative feedback according to claim 1, characterized in that: The soft demodulator module is connected to an optical receiver, which receives an optical signal from an optical fiber and obtains a 16QAM symbol after DSP processing. n , the 16QAM symbol r n Input to the soft demodulator module, output the original signal S n Take the prior probability vector for each value.

7. The multi-channel coding system based on iterative feedback according to claim 1, characterized in that: The serial-to-parallel conversion / demapping module converts the prior probability vector into s output by the multivariate LDPC soft decoding module of the previous iteration in each iteration. 2n-1 、s 2n The corresponding corrected probability vector Weighted feedback, converted to s 2n-1 、s 2n The corresponding pre-correction probability vector 8. The multi-channel coding system based on iterative feedback according to claim 7, characterized in that: The interior of the serial-to-parallel conversion / demapping module is composed of a serial-to-parallel converter and two weighters; The serial-to-parallel converter converts the feedback corrected probability vector sequence Divide and Two-way; The first weighting device and the second weighting device are used to Weighted and Then the serial-to-parallel converter and Merge into Output.

9. The multi-channel coding system based on iterative feedback according to claim 1, characterized in that: The multivariate LDPC soft decoding module is used to input a pre-correction value probability vector sequence Output corrected probability vector sequence and the decoding results c0, c1, ... c 2n-1 ,c 2n …and complete the verification.

10. The multi-channel coding and decoding system based on iterative feedback according to claim 9, characterized in that: The multivariate LDPC soft decoding module is internally divided into a soft decoding module, a verification module, and a decision module; Among them, the soft decoding module input After soft decoding, the updated probability vector sequence is obtained The updated probability vector sequence Input the decision module, and decide the code value by taking the maximum value in each vector The check module outputs a code element sequence To perform the check, let the check matrix be H and the checksum be when When verification is established.

Images