A hybrid encoding and decoding method and device based on sub-constellation space

Abstract

The present invention provides a hybrid coding method and device based on a sub-constellation space. The method comprises the following steps: generating bit information and separating it into two signals; performing LDPC and Polar coding on the two signals respectively, and performing polarization multiplexing processing to obtain X polarization and Y polarization; determining whether to perform amplitude shift on the Y polarization based on the constellation segmentation results of the X polarization and the Y polarization, and outputting the signal; loading the signal onto a laser for transmission; during reception, separating the signal into three parts: a 4D signal generated by the X polarization, a 4D signal generated by the Y polarization, and a quaternary phase shift keying (QPSK) signal; calculating the log-likelihood ratio of each bit in each symbol; determining the bit data of each symbol based on the log-likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; restoring the amplitude shift of the Y polarization based on the bit information of the QPSK signal to obtain the Y polarization; further demodulating and decoding the X polarization and the Y polarization to output the bit information.

Description

Technical Field

[0001] The present invention relates to the field of digital communication technology, and in particular to a hybrid encoding and decoding method and device based on a sub-constellation space. Background Art

[0002] With the rapid development of the internet industry, the massive rise of services such as artificial intelligence, big data, and the Internet of Things (IoT), and the rapidly growing demand for information, the requirements for the speed, capacity, and reliability of communication systems are constantly increasing. Meeting the rapidly growing network capacity demand has become a serious challenge for communication technology. As the backbone of communication networks, fiber optic communications plays an important role in researching fiber optic transmission technologies with faster speeds, greater capacity, and better performance to meet the needs of information development.

[0003] In traditional communication systems, the encoder and modulator operate independently, each fulfilling its own function of reducing the bit error rate and increasing the information transmission rate. Trellis Coded Modulation (TCM) breaks this paradigm, achieving excellent decoding performance without increasing system bandwidth or reducing data transmission rate. The basic concept of TCM is to introduce parity bits into the transmitted multi-bit symbols. This introduced coding redundancy allows for a set segmentation (SP) of signal constellation points. This two-dimensional trellis coded modulation scheme improves signal interference immunity without increasing transmission bandwidth or sacrificing information transmission rate, achieving significant coding gain. This increases the minimum Euclidean distance of the constellation while maintaining constant transmit power, thereby reducing the bit error rate.

[0004] Polarization multiplexing (PM) technology is often used in optical fiber communications to expand system capacity by transmitting two signals in the X and Y polarizations, respectively. The two polarization directions, along with their in-phase and quadrature components, together form a four-dimensional signal. When generating four-dimensional probabilistically shaped (TCM) signals, the Y polarization uses an amplitude shifting technique, resulting in slightly worse signal transmission performance in the Y polarization than in the X polarization. Therefore, encoding methods with slightly different performance can be used in the two polarizations to achieve the same performance in the Y polarization as in the X polarization. Because Polar codes significantly outperform LDPC in the case of short and medium code lengths, and their overall complexity is significantly lower, LDPC coding is used in the X polarization and Polar codes in the Y polarization. TCM mapping is then performed on each polarization to form constellation points, forming a 4D symbol for transmission. This hybrid coding followed by TCM is a reliable method for further improving communication system performance. However, when polarization multiplexing (PM) technology and probability shaping technology are used in combination during application, it is necessary to use amplitude shift technology for the Y polarization, but the existing technology lacks a method for effectively decoding the shifted coded signal. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a hybrid coding method based on a sub-constellation space to eliminate or improve one or more defects in the prior art.

[0006] One aspect of the present invention provides a hybrid coding method based on a sub-constellation space, the method comprising signal generation and signal reception and demodulation:

[0007] The step of generating the signal comprises:

[0008] generating bit information and separating the bit information into two signals using a bit distributor; performing advanced FEC coding on the two signals, i.e., using LDPC coding for the bit data that will form the X-polarization signal and using Polar coding for the bit data that will form the Y-polarization signal; and performing polarization multiplexing processing to obtain X-polarization and Y-polarization signals; performing constellation subset segmentation based on the X-polarization and Y-polarization signals; determining whether to perform amplitude shifting on the Y-polarization based on the constellation segmentation results of the X-polarization and Y-polarization signals, generating a quadrature phase shift keying signal based on the amplitude shift result of the Y-polarization, and outputting the signal;

[0009] The steps of receiving and demodulating the signal include:

[0010] The signal is separated into three parts: a 4D signal generated by X polarization, a 4D signal generated by Y polarization, and a quaternary phase shift keying (QPSK) signal; a hard decision is performed on the QPSK signal to obtain bit information of a marker bit carried by the signal; a log likelihood ratio of each bit in the 4D signal generated by X polarization and the 4D signal generated by Y polarization is calculated; data of each bit is determined based on the log likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; and the amplitude shift of the amplitude-shifted Y polarization is restored based on the bit information obtained by demodulating the QPSK signal to obtain the Y polarization. The X polarization and the Y polarization are further demodulated and decoded to output the bit information.

[0011] Using the above scheme, at the receiving end, the present invention proposes a multidimensional soft decision algorithm for amplitude shift set segmentation, which performs bit-by-bit decoding on the data on X polarization and Y polarization respectively. In X polarization, the Euclidean distance can be directly calculated with the subset of the received symbol, that is, the constellation point in the bit-related dimension BRD, to obtain the log-likelihood ratio for decoding; in Y polarization, each bit in the two directions is decoded bit by bit based on the marker bits of the received in-phase component and orthogonal component. If the marker bit is 0, it can be decoded according to the log-likelihood ratio in the bit-related dimension. If the marker bit is 1, it is necessary to determine which bit has changed according to the Gray code mapping rule determined by the transmitting end, and reverse it. After the received signal point is shifted back to its original position, it is decoded in the bit-related dimension. This algorithm has low complexity and, because it can combine probability amplitude shaping (PAS) and set segmentation technology, has relatively superior bit error rate performance and can be applied to effectively decode the shifted coded signal.

[0012] In some embodiments of the present invention, in the step of encoding the two signals separately and performing polarization multiplexing processing to obtain X polarization and Y polarization, LDPC encoding and Polar encoding are respectively used on the two signals to compensate for the fact that the effect of the Y polarization signal in subsequent processing will be slightly worse than that of the X polarization signal.

[0013] In some embodiments of the present invention, the constellation points of the signal in each polarization are divided into multiple subsets. In the step of performing constellation subset division based on the X-polarization and Y-polarization signals, the subsets to which the X-polarization and Y-polarization signals belong are determined. The four subsets for each of the X-polarization and Y-polarization signals are then combined according to the determined subfamily constraints to form a 4D subfamily.

[0014] In some embodiments of the present invention, in the step of determining whether to perform amplitude shifting on the Y polarization based on the constellation segmentation results of the X polarization and the Y polarization, it is determined based on the subset to which the X polarization belongs whether the subfamily consisting of the subsets of the X polarization and the Y polarization belongs to a combination corresponding to a preset subfamily constraint corresponding to the subset of the X polarization. If so, no amplitude shifting is required; if not, the Y polarization is amplitude shifted to the subfamily corresponding to the subset of the X polarization.

[0015] In some embodiments of the present invention, the output signal includes a four-phase phase shift keying signal, and the Y polarization includes an in-phase component and an orthogonal component (for the 64QAM signal adopted, each symbol carries a total of 6 bits of information, of which the first 3 bits are the in-phase component information and the last 3 bits are the orthogonal component information). In the step of outputting the signal, if amplitude shifting is not required, the bit information carried by the four-phase phase shift keying signal is 00; if amplitude shifting is required, and the in-phase component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 10; if amplitude shifting is required, and the orthogonal component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 01; if amplitude shifting is required, and the in-phase component and the orthogonal component of the Y polarization are shifted, the bit information carried by the four-phase phase shift keying signal is 11.

[0016] In some embodiments of the present invention, in the steps of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization; determining the data of each bit based on the log-likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; and restoring the amplitude shift of the Y polarization based on the bit data of the quadrature phase shift keying signal to obtain the Y polarization, and further demodulating and decoding the X polarization and the Y polarization;

[0017] When demodulating the quaternary phase shift keying signal, a hard decision method is used to directly decode the bit information of the mark bit;

[0018] In the 4D signal generated by X polarization, the Euclidean distance is directly calculated with the subset in which it is located, and the log-likelihood ratio is obtained for decoding;

[0019] In the 4D signal generated by Y polarization, each bit in the two directions is decoded bit by bit based on the marker bits of the received in-phase component and orthogonal component, that is, the two marker bits of the quadrature phase shift keying signal. If the marker bit is 0, decoding is performed based on the log-likelihood ratio in the bit-related dimension. If the marker bit is 1, it is necessary to determine the changed bit in the symbol according to the Gray code mapping rules determined by the transmitter, invert it, and then decode it.

[0020] In some embodiments of the present invention, in the step of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization, the log-likelihood ratio of each bit is calculated according to the following formula:

[0021]

[0022] in, represents the log-likelihood ratio of the k-th bit, represents the standard constellation point where the kth bit is 1, represents the standard constellation point where the kth bit is 0, represents the conditional probability.

[0023] In some embodiments of the present invention, in the step of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization, the k-th bit information in the received symbol is respectively compared with the standard constellation point ( ) and take the minimum value; at the same time, it is also compared with the standard constellation point with the kth position being 1 ( ) to compare the Euclidean distance between them, take the minimum value, and then calculate , in Gaussian channel, the formula for calculating log-likelihood ratio is expressed as:

[0024] ;

[0025] Where y represents the received 4D signal, x represents the standard constellation point, represents the log-likelihood ratio of the k-th bit, represents the standard constellation point where the kth bit is 1, represents the standard constellation point where the kth bit is 0, Represents the variance characteristics of the Gaussian channel itself, Indicates the total number of dimensions to be calculated.

[0026] In some embodiments of the present invention, since it is necessary to satisfy the subfamily constraint, that is, ;

[0027] ;

[0028] in, represents the subfamily constraint, represents the dimension j in the subfamily constraint, y represents the received 4D signal, and x represents the standard constellation point. express The kth position in the dimension is the standard constellation point i, express The kth standard constellation point in the dimension that satisfies the subfamily constraint is i, express The kth standard constellation point in the dimension that satisfies the jth subfamily constraint is i, represents a subfamily, the subset to which the kth bit belongs is called the bit-related dimension (BRD), and m represents the dimension of the non-bit-related dimension. Indicates the total number of dimensions calculated, represents a 4D signal of non-bit-related dimensions, represents the standard constellation points of non-bit-related dimensions, A set of dimensions representing non-bit-related dimensions, Represents the minimum Euclidean distance in the bit-related dimension.

[0029] In some embodiments of the present invention, the minimum Euclidean distance in the bit correlation dimension is calculated according to the following formula: :

[0030] When the dimension is not shifted or the marker bit in the shifted dimension is 0, :

[0031]

[0032] When the dimension is a translation dimension and the mark bit = 1, that is, the symbol has an amplitude translation, the Gray coding rule determined by the transmitter is used to determine whether the bit has changed. If it has changed, the data bit of the bit is opposite to the constellation point mapping bit. At this time, it is necessary to consider and , Pick and Minimum value of:

[0033]

[0034] in, yes A subset of excluding

[0035] Conclusion Later, similarly, It can be expressed as:

[0036]

[0037] in The collection is:

[0038]

[0039] in, represents the minimum Euclidean distance between the received signal and the standard constellation point with the kth bit i in the bit-related dimension, represents the minimum Euclidean distance between the received signal and the kth standard constellation point 1-i in the bit-correlation dimension, A set of dimensions representing bit-related dimensions in SFC, represents the 4D signal of the received bit-dependent dimensions, represents a standard constellation point in the bit-dependent dimension.

[0040] The second aspect of the present invention also provides a hybrid encoding and decoding device based on a sub-constellation space, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the device implements the steps implemented by the method described above.

[0041] The third aspect of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps implemented by the aforementioned hybrid coding method based on the sub-constellation space.

[0042] Additional advantages, objects, and features of the present invention will be described in part in the following description and will become apparent to those skilled in the art after studying the following or may be learned by practice of the present invention. The objects and other advantages of the present invention may be particularly pointed out and attained in the description and drawings.

[0043] Those skilled in the art will understand that the purposes and advantages that can be achieved by the present invention are not limited to the above specific descriptions, and the above and other purposes that can be achieved by the present invention will be more clearly understood based on the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention.

[0045] Figure 1 Schematic diagram of an implementation of a hybrid coding method based on a sub-constellation space according to the present invention;

[0046] Figure 2 This is a schematic diagram of the fixed-bit puncturing Polar code structure;

[0047] Figure 3 Schematic diagram of the division of signal constellation subsets in each polarization;

[0048] Figure 4 Schematic diagram of signal generation;

[0049] Figure 5 Schematic diagram of signal reception and demodulation. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0051] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, the accompanying drawings only show structures and / or processing steps closely related to the solutions according to the present invention, while other details that are not closely related to the present invention are omitted.

[0052] To solve the above problems, Figure 1 As shown, the present invention proposes a hybrid coding method based on a sub-constellation space, the method comprising step S100 signal generation and step S200 signal reception and demodulation:

[0053] The step of generating the signal in step S100 includes:

[0054] like Figure 4 As shown, in step S110, bit information is generated, and a bit distributor is used to separate the bit information into two signals;

[0055] The bit distributor uses the odd bits of the bit information as one signal and the even bits as one signal.

[0056] Step S120: Encode the two signals separately and perform polarization multiplexing processing to obtain X polarization and Y polarization. That is, encode the two signals separately, that is, use LDPC coding for the bit data that will form the X polarization signal, and use Polar coding for the bit data that will form the Y polarization signal to compensate for the fact that the effect of the Y polarization signal will be slightly worse than that of the X polarization signal in subsequent processing, and perform polarization multiplexing processing to obtain X polarization and Y polarization.

[0057] Step S130, performing constellation subset division based on X-polarization and Y-polarization signals;

[0058] like Figure 3 As shown, in a specific implementation process, the constellation points of the signal in each polarization are divided into multiple subsets, and the subsets include four subsets A, B, C and D.

[0059] The constellation segmentation is completed by determining the subsets to which the X-polarization and Y-polarization signals belong, and combining the four subsets of the X-polarization and Y-polarization signals according to the determined sub-family constraints to form a 4D sub-family.

[0060] Step S140, determining whether to perform amplitude shift on the Y polarization based on the constellation segmentation results of the X polarization and the Y polarization, generating a quadrature phase shift keying signal based on the amplitude shift result of the Y polarization, and outputting the signal;

[0061] In some embodiments of the present invention, in the step of determining whether to perform amplitude shifting on the Y polarization based on the constellation segmentation results of the X polarization and the Y polarization, it is determined based on the subset to which the X polarization belongs whether the subfamily consisting of the subsets of the X polarization and the Y polarization belongs to a combination corresponding to a preset subfamily constraint corresponding to the subset of the X polarization. If so, no amplitude shifting is required; if not, the Y polarization is amplitude shifted to the subfamily corresponding to the subset of the X polarization.

[0062] In specific implementations, the subfamilies corresponding to the preset subfamily constraints include A∪D, B∪C, C∪B, and D∪A. The first one that meets the constraint corresponds to the subset for X polarization, and the second one corresponds to the subset for Y polarization. If the actual X and Y polarizations belong to the subfamily A∪C, the Y polarization needs to be amplitude-shifted to the subfamily A∪D corresponding to the preset subfamily constraint for the X polarization subset.

[0063] The step S200 of signal reception and demodulation includes:

[0064] like Figure 5 As shown, in step S210, the signal is separated into three parts: a 4D signal generated by X polarization, a 4D signal generated by Y polarization, and a quaternary phase shift keying signal;

[0065] In the specific implementation process, the output signal includes a four-phase phase shift keying signal, and the Y polarization includes an in-phase component and an orthogonal component (for the 64QAM signal adopted, each symbol carries a total of 6 bits of information, of which the first 3 bits are the in-phase component information and the last 3 bits are the orthogonal component information). In the step of outputting the signal, if amplitude shifting is not required, the bit information carried by the four-phase phase shift keying signal is 00; if amplitude shifting is required, and the in-phase component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 10; if amplitude shifting is required, and the orthogonal component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 10; if amplitude shifting is required, and the orthogonal component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 10; if amplitude shifting is required, and the in-phase component and orthogonal component of the Y polarization are shifted, the bit information carried by the four-phase phase shift keying signal is 11.

[0066] Step S220, performing hard decision on the quaternary phase shift keying signal to obtain bit information of the mark bit;

[0067] Step S230, calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization;

[0068] Step S240 : Determine the data of each bit based on the log-likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; restore the amplitude shift of the Y polarization based on the bit data of the QPSK signal to obtain the Y polarization; further demodulate and decode the X polarization and the Y polarization to output the bit information.

[0069] Using the above scheme, at the receiving end, the present invention proposes a multidimensional soft decision algorithm for amplitude shift set segmentation, which performs bit-by-bit decoding on the data on X polarization and Y polarization respectively. In X polarization, the Euclidean distance can be directly calculated with the subset of the received symbol, that is, the constellation point in the bit-related dimension BRD, to obtain the log-likelihood ratio for decoding; in Y polarization, each bit in the two directions is decoded bit by bit based on the marker bits of the received in-phase component and orthogonal component. If the marker bit is 0, it can be decoded according to the log-likelihood ratio in the bit-related dimension. If the marker bit is 1, it is necessary to determine which bit has changed according to the Gray code mapping rule determined by the transmitting end, and reverse it. After the received signal point is shifted back to its original position, it is decoded in the bit-related dimension. This algorithm has low complexity and, because it can combine probability amplitude shaping (PAS) and set segmentation technology, has relatively superior bit error rate performance and can be applied to effectively decode the shifted coded signal.

[0070] In some embodiments of the present invention, in the step of respectively encoding the two signals and performing polarization multiplexing processing to obtain X polarization and Y polarization, LDPC coding and Polar coding are respectively used for the two signals.

[0071] In specific applications, TCM technology typically uses convolutional codes for coding. This is relatively simple to implement and the encoding and decoding technology is mature, but the gain it provides is limited. Therefore, cascading more efficient coding methods before the TCM module is an effective way to achieve higher coding gain. Among the currently advanced forward error correction (FEC) codes, Polar codes and LDPC codes offer excellent bit error rate performance in short and long code scenarios, respectively. Both also offer significant advantages in decoding speed and complexity, leading them to be used in 5G scenarios as coding methods for control channels and data channels, respectively. Therefore, combining Polar codes and LDPC codes with TCM can further improve system performance.

[0072] Using this solution, the binary bit stream input from the transmitter is first split into two parts, processed separately, and then transmitted on two polarizations. One data stream is QC-LDPC encoded and subsequently transmitted on the X polarization, while the other data stream is Polar encoded and subsequently transmitted on the Y polarization.

[0073] For the first data channel, LDPC coding is essentially a linear block code. Aside from the sparse check matrix, the code itself is no different from any other block code. Therefore, the construction method of a common block code is directly adopted. First, a suitable check matrix H is designed. Then, based on the duality property, its generator matrix G is obtained. Finally, the source code group is multiplied by the generator matrix to obtain the complete codeword. The check matrix H can be obtained through random construction.

[0074] The Polar code used for the other data channel mainly utilizes the channel polarization phenomenon to rearrange the positions of the new sub-channels generated after the channel is combined and split, placing the information bits in a perfect channel with a capacity close to 1 and the frozen bits in a pure noise channel with a capacity close to 0, thereby approaching the Shannon limit and achieving the highest possible transmission rate while minimizing the bit error rate. Due to the limitations of the encoder structure, the encoded code length is fixed to Code length and code rate have certain limitations, which can be addressed using bit puncturing. For system Polar codes, after encoding, the puncture locations are determined in the frozen bits in descending order of channel capacity. Specifically, locations with higher channel levels are selected as known bits for puncturing. The data to be transmitted is placed between the data bits and the punctured bits, and the remaining frozen bits are set to 0. The known bits are deleted at the transmitter and reinserted at the receiver before decoding. This method not only overcomes code length limitations and improves transmission efficiency, but also allows the inserted known bits to aid decoding at the receiver, ensuring a consistent bit error rate reduction.

[0075] During implementation, after Polar codes are reorganized and split, the binary memoryless channel becomes polarized. While the channel capacity remains unchanged, two extreme types of channels are obtained: a good channel with a channel capacity approaching 1, and a bad channel with a channel capacity approaching 0. Furthermore, as the code length N increases, the proportion of good channels approaches the capacity of the original channel.

[0076] In the specific implementation process, during the step of Polar encoding, bit puncturing is performed to obtain a codeword length of For example, the process is as follows Figure 2 As shown:

[0077] 1. Select the frozen bit and information bit positions by using ether density evolution or Gaussian approximation, the frozen bit , information bit .

[0078] 2. Use channel polarization to calculate the channel capacity of the frozen bit, and select the frozen bit with the larger channel capacity as the fixed bit. The fixed bit is , and the other two frozen bits are input 0.

[0079] 3. After these channels are uniformly encoded with the system Polar code, the output .

[0080] 4. Breaking down known bits , send the remaining signal .

[0081] 5. At the receiving end, the known bits Insert it into the received data and then decode it.

[0082] In practical applications, considering performance factors such as bit error rate and complexity, a medium-short code with a code length of several hundred bits is selected, which will increase the corresponding known bits. After encoding and puncturing according to the above method, the output data is subjected to probability amplitude shaping and signal set segmentation.

[0083] After the two-way encoding, they enter the TCM module respectively. In this part, convolutional code or XOR is usually used to generate check bits. However, since probability amplitude shaping must be performed first, the check bits are no longer generated by XOR of the information bits, but are divided by the constellation set. The generated non-uniform 1D constellation amplitude set is divided into two 1D subsets in an equally spaced manner. At this time, the minimum Euclidean distance of the constellation points in each subset is twice the original one. Then, the set where each 1-dimensional component (i.e., in-phase and orthogonal component) of the 2D constellation is located is divided into two 1D subsets. According to the 1D subset, four 2D subsets A, B, C, and D are formed to form the "subset constraint (SSC)" of the 2D signals in the two polarization states to ensure that the minimum Euclidean distance of the constellation points in each 2D subset is twice that of the original 2D constellation. Finally, the 1D subset is combined into four 4D sub-families (A∪D, B∪C, C∪B, D∪A), forming the "sub-family constraint (SFC)" of the 4D signal to ensure that the minimum Euclidean distance of the constellation points in each 4D family is twice that of the original 4D constellation. At this time, the division of the 64QAM signal constellation subsets on each polarization is as follows Figure 3 shown.

[0084] In this diversity mode, the 2D signal on the X polarization is kept unchanged, and the 4D sub-family under the combined SFC constraint is selected to determine the subset in which the corresponding Y polarization 2D signal should be located. When determining, the 2D signal is decomposed into 1D signals for analysis. When a 1D signal on the Y polarization does not meet the current SFC sub-family constraint, the 1D signal is shifted to the adjacent signal point of the entire 2D constellation (the direction of the shift must be consistent within the same system), that is, moved to the complementary 1D subset, and a binary flag bit 1 is generated, which is recorded as If the 1D signal on the Y polarization satisfies the SFC condition of the 4D subfamily and generates a binary flag bit 0, then . Each 4D signal generates two marker bits. In this module, the final 4D signal is divided into two 2D signals transmitted on X polarization and Y polarization, and is formed by polarization multiplexing. It can be understood that the in-phase component and the orthogonal component existing in the two polarization directions together constitute the four-dimensional signal. The 4D-PS signal generated according to the above method can be collectively segmented during transmission. During the signal generation process, a 4D symbol will generate two binary marker bits. The bits of this part of the marker bits can be directly mapped to a modulation format that is easier to demodulate, such as quadrature phase shift keying (QPSK), but not limited to QPSK.

[0085] After the above hybrid coding, polarization multiplexing, probability shaping, and constellation set segmentation, the hybrid FEC-4D-PS-64QAM signal is finally obtained. The complete generation process is as follows: Figure 4 It is loaded onto a laser and placed into an optical fiber channel for transmission.

[0086] Therefore, at the receiving end, the received signal is first processed through dispersion compensation and polarization equalization, followed by polarization demultiplexing. A decision is made based on the two polarization states. Whether the flag bit is 1 determines whether the corresponding component on the Y polarization needs to be reverse-shifted. Probabilistic amplitude shaping is then used for demodulation. This paper optimizes the decoding algorithm and proposes a multidimensional soft-decision algorithm for amplitude shift set segmentation.

[0087] After obtaining the bit data, the corresponding LDPC decoding and Polar decoding are performed. The BP algorithm is typically used for LDPC decoding. For Polar codes, the known bits obtained by puncturing during encoding are inserted into the TCM-demodulated data. Polar code decoding is then performed. The Serial Cancellation List (SCL) algorithm is used to iteratively calculate the likelihood ratio (LLR) information of the bit channel. The concept of path metric (PM) is introduced, and iterative decoding is performed along multiple paths. The path with the highest transition probability is ultimately selected as the output, which is the final decoding result. The bit error rate is then compared with the original bit data.

[0088] In some embodiments of the present invention, in the steps of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization; determining the data of each bit based on the log-likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; and restoring the amplitude shift of the Y polarization based on the bit data of the quadrature phase shift keying signal to obtain the Y polarization, and further demodulating and decoding the X polarization and the Y polarization:

[0089] When demodulating the quaternary phase shift keying signal, a hard decision method is used to directly decode the bit information of the mark bit;

[0090] In the 4D signal generated by X polarization, the Euclidean distance is directly calculated with the subset in which it is located, and the log-likelihood ratio is obtained for decoding;

[0091] In the 4D signal generated by Y polarization, each bit in both directions is decoded bit by bit based on the received marker bits in the in-phase component and orthogonal component directions, that is, the two marker bits carried by the quaternary phase-shift keying signal. If the marker bit is 0, decoding is performed based on the log-likelihood ratio in the bit-related dimension. If the marker bit is 1, it is necessary to determine the changed bit in the symbol according to the Gray code mapping rules determined by the transmitter, invert it, and reverse the signal back to its original position before decoding.

[0092] In the specific implementation process, the original calculation formula of the log-likelihood ratio is:

[0093]

[0094] It can be approximated as:

[0095]

[0096] in, represents the log-likelihood ratio of the k-th signal bit, represents the standard constellation point where the kth bit is 1, represents the standard constellation point where the kth bit is 0, represents the conditional probability.

[0097] In a Gaussian channel, the log-likelihood ratio can be further transformed into the following formula: the k-th bit information in the received symbol is compared with the standard constellation point with the k-th bit being 0 ( ) and take the minimum value; at the same time, it is also compared with the standard constellation point with the kth position being 1 ( ) to compare the Euclidean distance between them, take the minimum value, and then calculate .

[0098]

[0099] Where y represents the received 4D signal, x represents the standard constellation point, represents the log-likelihood ratio of the k-th bit, represents the standard constellation point where the kth bit is 1, represents the standard constellation point where the kth bit is 0, Represents the variance characteristics of the Gaussian channel itself, Indicates the total number of dimensions to be calculated.

[0100] In the case of N dimensions, the k-th bit information in the received symbol needs to be compared with the standard constellation point with the k-th bit being 0 in N dimensions ( ) and take the minimum value; also compare it with the standard constellation point with the kth bit being 1 ( ) to compare the Euclidean distance between them, take the minimum value, and then calculate , which can be expressed as:

[0101]

[0102] At the transmitter, the 4D subset to which each bit belongs in the constellation has been determined. The 4D subset to which the kth bit belongs is called the bit-related dimension (BRD). Therefore, the calculation can be divided into two parts: non-BRD and BRD:

[0103]

[0104] After simplification, it can be seen that the log-likelihood ratio of the k-th bit is only related to the BRD. Therefore, in the algorithm, only the Euclidean distance to each constellation point in the BRD needs to be calculated.

[0105] To restore the amplitude-shifted signal, a multi-dimensional soft decision algorithm for amplitude-shifted set segmentation is proposed here:

[0106] In some embodiments of the present invention, since it is necessary to satisfy the subfamily constraint, that is, ;

[0107] ;

[0108] in, represents the subfamily constraint, represents the dimension j in the subfamily constraint, y represents the received 4D signal, and x represents the standard constellation point. express The kth position in the dimension is the standard constellation point i, express The kth standard constellation point in the dimension that satisfies the subfamily constraint is i, express The kth standard constellation point in the dimension that satisfies the jth subfamily constraint is i, represents a subfamily, the subset to which the kth bit belongs is called the bit-related dimension, and m represents the dimension of the non-bit-related dimension. Indicates the total number of dimensions calculated, A 4D signal representing the dimensions of the non-bit-related dimensions, represents the standard constellation points of non-bit-related dimensions, A set of dimensions representing non-bit-related dimensions, Represents the minimum Euclidean distance in the bit-related dimension.

[0109] In some embodiments of the present invention, when the dimension is an unshifted dimension or the dimension is a shifted dimension but the flag bit is 0, ,at this time:

[0110]

[0111] When the dimension is a translation dimension and the mark bit = 1, that is, the symbol has undergone amplitude translation, the Gray coding rule determined by the transmitter is used to determine whether the bit has changed. If it has changed, the data bit of the bit is opposite to the constellation point mapping bit, then it is necessary to consider and Two situations, Pick and Minimum value of:

[0112]

[0113] in, yes A subset of excluding

[0114] Conclusion Later, similarly, It can be expressed as:

[0115]

[0116] in The collection is:

[0117]

[0118] in, represents the minimum Euclidean distance between the received signal and the k-th standard constellation point i in the bit-related dimension, represents the minimum Euclidean distance between the received signal and the k-th standard constellation point 1-i in the bit-related dimension, represents the bit-related dimension in SFC, represents the 4D signal of the received bit-dependent dimensions, represents a standard constellation point in the bit-dependent dimension.

[0119] To summarize the above derivation, for any case , under the S-dimensional SFC constraint, it can be expressed as:

[0120]

[0121] In the specific implementation process, a decision is made based on the obtained log-likelihood ratio, and demodulation of PAS probability amplitude shaping is performed;

[0122] The demodulated bit data in the X polarization direction is decoded using the BP algorithm, and the bit data in the Y polarization direction is decoded using the SCL algorithm. Note that before performing Polar decoding, the digital known bits extracted by the transmitter need to be inserted into the multi-dimensional soft decision output bits in the Y polarization before decoding.

[0123] After decoding is completed, the bit error rate is calculated.

[0124] The beneficial effects of the present invention include:

[0125] 1. This invention uses parallel coding of Polar and LDPC codes, combined with TCM technology, and incorporates polarization multiplexing (PM) technology, set partitioning (SP) theory, and probability amplitude shaping (PAS) technology to achieve hybrid coded modulation of four-dimensional high-order signals in sub-constellation space. The decoding algorithm is also optimized.

[0126] 2. When PAS shaping and the set segmentation technique used in TCM are applied simultaneously in optical fiber communications, incompatibility arises: if PAS shaping is performed first, followed by set segmentation, the distribution of the shaped symbols will be chaotic; if set segmentation is performed first, followed by PAS shaping, decoding at the receiver will not be improved. Therefore, this invention employs an amplitude shifting method. Simultaneously with set segmentation, the amplitude of the PAS-posted signal on the Y polarization is shifted according to constraints. This generates a marker bit (0 or 1) to aid decoding, overcoming this incompatibility.

[0127] 3. At the receiving end, the present invention proposes a multidimensional soft-decision algorithm for amplitude-shifted set partitioning, which performs bit-by-bit decoding on the X- and Y-polarization data. In the X-polarization, the Euclidean distance is directly calculated with the constellation points in the subset containing the received symbol (i.e., the bit-related dimension (BRD)), and the log-likelihood ratio is obtained for decoding. In the Y-polarization, each bit in the two directions is decoded bit-by-bit based on the marker bits of the received in-phase and orthogonal components. If the marker bit is 0, decoding is performed in the BRD based on the log-likelihood ratio. If the marker bit is 1, it is necessary to determine which bit has changed based on the Gray code mapping rules determined by the transmitter, and then invert it. After the received signal point is shifted back to its original position, decoding is performed in the BRD. This algorithm has low complexity and, due to its ability to combine PAS and set partitioning techniques, exhibits superior bit error rate performance.

[0128] An embodiment of the present invention also provides a hybrid encoding and decoding device based on a sub-constellation space, which includes a computer device, the computer device including a processor and a memory, the memory storing computer instructions, and the processor being used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the device implements the steps implemented by the method described above.

[0129] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the steps implemented by the aforementioned hybrid coding method based on the sub-constellation space. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, floppy disk, hard disk, removable storage disk, CD-ROM, or any other form of storage medium known in the art.

[0130] It should be understood by those skilled in the art that the various exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether to implement the system in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention. When implemented in hardware, it may be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via a data signal carried in a carrier wave.

[0131] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0132] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or replace features of other embodiments.

[0133] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations to the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A hybrid coding method based on a sub-constellation space, characterized in that: The method comprises the steps of generating a signal and demodulating a signal received: The signal generation step comprises: generating bit information, separating the bit information into two signals using a bit distributor; encoding the two signals respectively and performing polarization multiplexing processing to obtain X polarization and Y polarization; performing constellation subset segmentation based on the X polarization and Y polarization signals; determining whether to perform amplitude shift on the Y polarization based on the constellation segmentation results of the X polarization and the Y polarization, generating a four-phase phase shift keying signal based on the amplitude shift result of the Y polarization, and outputting the signal; The steps of receiving and demodulating the signal include: The signal is separated into three parts: a 4D signal generated by X polarization, a 4D signal generated by Y polarization, and a quaternary phase shift keying (QPSK) signal; a hard decision is performed on the QPSK signal to obtain bit information of a marker bit; a log likelihood ratio of each bit in the 4D signal generated by X polarization and the 4D signal generated by Y polarization is calculated; data of each bit is determined based on the log likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; the amplitude shift of the Y polarization is restored based on the bit data of the QPSK signal to obtain the Y polarization; the X polarization and the Y polarization are further demodulated and decoded to output the bit information.

2. The hybrid coding method based on sub-constellation space according to claim 1, characterized in that In the step of respectively encoding the two signals and performing polarization multiplexing processing to obtain X polarization and Y polarization, LDPC coding and Polar coding are respectively used for the two signals.

3. The hybrid coding method based on sub-constellation space according to claim 1, characterized in that: In the step of determining whether to perform amplitude shifting on the Y polarization based on the constellation segmentation results of the X polarization and the Y polarization, it is determined based on the subset to which the X polarization belongs whether the subfamily consisting of the subsets of the X polarization and the Y polarization belongs to a combination corresponding to a preset subfamily constraint corresponding to the subset of the X polarization. If so, no amplitude shifting is required; if not, the Y polarization is amplitude shifted to the subfamily corresponding to the subset of the X polarization.

4. The hybrid coding method based on sub-constellation space according to claim 3, characterized in that: The output signal includes a four-phase phase shift keying signal, and the Y polarization includes an in-phase component and a quadrature component. In the step of outputting the signal, if amplitude shifting is not required, the bit information carried by the four-phase phase shift keying signal is 00; if amplitude shifting is required, and the in-phase component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 10; if amplitude shifting is required, and the quadrature component of the Y polarization is shifted, the bit information carried by the four-phase phase shift keying signal is 01; if amplitude shifting is required, and the in-phase component and quadrature component of the Y polarization are shifted, the bit information carried by the four-phase phase shift keying signal is 11.

5. The hybrid coding method based on sub-constellation space according to claim 1, characterized in that: In the steps of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization; determining the data of each bit based on the log-likelihood ratio to obtain the X polarization and the amplitude-shifted Y polarization; and restoring the amplitude shift of the Y polarization based on the bit data of the quaternary phase shift keying signal to obtain the Y polarization, and further demodulating and decoding the X polarization and the Y polarization; When demodulating the quaternary phase shift keying signal, a hard decision method is used to directly decode the bit information of the mark bit; In the 4D signal generated by X polarization, the Euclidean distance is directly calculated with the subset in which it is located, and the log-likelihood ratio is obtained for decoding; In the 4D signal generated by Y polarization, each bit in the two directions is decoded bit by bit based on the marker bits of the received in-phase component and orthogonal component, that is, the two marker bits of the quadrature phase shift keying signal. If the marker bit is 0, decoding is performed based on the log-likelihood ratio in the bit-related dimension. If the marker bit is 1, it is necessary to determine the changed bit in the symbol according to the Gray code mapping rules determined by the transmitter, invert it, and then decode it.

6. The hybrid coding method based on sub-constellation space according to claim 1, characterized in that: In the step of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization, the log-likelihood ratio of each bit is calculated according to the following formula: in, represents the log-likelihood ratio of the k-th bit, represents the standard constellation point where the kth bit is 1, represents the standard constellation point where the kth bit is 0, represents the conditional probability.

7. The hybrid coding method based on sub-constellation space according to claim 6, characterized in that: In the step of calculating the log-likelihood ratio of each bit in the 4D signal generated by the X polarization and the 4D signal generated by the Y polarization, the k-th bit information in the received symbol is respectively compared with the standard constellation point ( ) and take the minimum value; at the same time, it is also compared with the standard constellation point with the kth bit being 1 ( ) to compare the Euclidean distance between them, take the minimum value, and then calculate , in Gaussian channel, the formula for calculating log-likelihood ratio is expressed as: ； Where y represents the received 4D signal, x represents the standard constellation point, represents the log-likelihood ratio of the k-th bit, represents the standard constellation point where the kth bit is 1, represents the standard constellation point where the kth bit is 0, Represents the variance characteristics of the Gaussian channel itself, Indicates the total number of dimensions to be calculated.

8. The hybrid coding method based on sub-constellation space according to claim 7, characterized in that: Since the subfamily constraints need to be satisfied, that is, ; ； in, represents the subfamily constraint, represents the dimension j in the subfamily constraint, y represents the received 4D signal, and x represents the standard constellation point. express The kth position in the dimension is the standard constellation point i, express The kth standard constellation point in the dimension that satisfies the subfamily constraint is i, express The kth standard constellation point in the dimension that satisfies the jth subfamily constraint is i, represents a subfamily, the subset to which the kth bit belongs is called the bit-related dimension, and m represents the dimension of the non-bit-related dimension. Indicates the total number of dimensions calculated, A 4D signal representing the dimensions of the non-bit-related dimensions, represents the standard constellation points of non-bit-related dimensions, A set of dimensions representing non-bit-related dimensions, represents the minimum Euclidean distance in the bit-related dimension, Indicates bit-related dimensions.

9. The hybrid coding method based on sub-constellation space according to claim 8, characterized in that: The minimum Euclidean distance in the bit-related dimension is calculated according to the following formula : When the dimension is not shifted or the dimension is shifted but the flag bit is 0, ,at this time: When the dimension is a translation dimension and the mark bit = 1, that is, the symbol has undergone amplitude translation, the Gray coding rule determined by the transmitter is used to determine whether the bit has changed. If it has changed, the data bit of the bit is opposite to the constellation point mapping bit, then it is necessary to consider and Two situations, Pick and Minimum value of: in, yes A subset of excluding Conclusion Later, similarly, It can be expressed as: in The collection is: in, represents the minimum Euclidean distance between the received signal and the k-th standard constellation point i in the bit-related dimension, represents the minimum Euclidean distance between the received signal and the k-th standard constellation point 1-i in the bit-related dimension, represents the bit-related dimension in SFC, represents the 4D signal of the received bit-dependent dimensions, represents a standard constellation point in the bit-dependent dimension.

10. A hybrid coding and decoding device based on a sub-constellation space, characterized in that: The apparatus includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the apparatus implements the steps implemented by the method according to any one of claims 1 to 9.

Images