Maximum likelihood sequence estimation method and device for multi-ary shaped signal

Abstract

The application provides a maximum likelihood sequence estimation method and device for a multi-ary shaping signal. The method comprises: obtaining a plurality of discrete level symbols corresponding to the multi-ary shaping signal, and dividing a state space of the multi-ary shaping signal into logical regions corresponding to the respective discrete level symbols; pruning signal transition paths corresponding to each logical region respectively to obtain a reserved path set composed of signal transition paths reserved by each logical region, wherein the pruning comprises: adaptive state pruning based on a preset pruning rule and branch pruning based on path rejection; and taking the reserved path set as effective state transition branches to perform Viterbi decoding to obtain maximum likelihood sequence estimation result data corresponding to the multi-ary shaping signal. The application can effectively reduce the calculation complexity of the maximum likelihood sequence estimation algorithm and improve the applicability, thereby reducing the power consumption of the signal receiving end and improving the spectral efficiency and practicability of the optical communication system.

Description

Technical Field

[0001] This application relates to the field of optical fiber communication technology, and in particular to a method and apparatus for maximum likelihood sequence estimation of multi-level shaped signals. Background Technology

[0002] With the access of new services such as ultra-high-definition video, cloud services, virtual reality, and mobile data transmission, the data transmission traffic of communication systems such as optical backbone networks and data center virtualization solutions (DCs) interconnection has experienced explosive growth. The bandwidth limitations of existing equipment have become a major bottleneck restricting further improvement in information capacity. Therefore, alleviating equipment bandwidth limitations through advanced digital signal processing (DSP) technology has become a key approach to promoting the development of next-generation optical communication networks. Faster-Than-Nyquist signaling (FTN) optical transmission systems significantly improve the system's spectral efficiency (SE) by employing multi-level shaping to compress signal energy within the Nyquist bandwidth, providing a promising solution for reducing DSP power consumption and alleviating device bandwidth limitations. In the receiver, the Maximum Likelihood Sequence Estimation (MLSE) decoder can effectively compensate for inter-symbol interference (ISI) caused by multi-level shaping. The basic idea of ​​the MLSE algorithm is to find the most likely transmission sequence based on the representation of the information sequence in a lattice graph.

[0003] Traditional maximum likelihood sequence estimation algorithms inherently possess extremely high computational complexity, especially for high-order modulation or systems with long delay extensions, where the number of states in the lattice graph becomes very large, significantly increasing the computational burden. This complexity limitation largely restricts the implementation of FTN optical transmission systems to Quadrature Phase Shift Keying (QPSK) using Quadrature Duobinary (QDB) shaping. However, with the continuous increase in symbol rates, device bandwidth constraints become even more severe, necessitating higher-order modulation schemes and more advanced multi-level shaping techniques. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method and apparatus for maximum likelihood sequence estimation of multi-level shaped signals, so as to eliminate or improve one or more defects existing in the prior art.

[0005] One aspect of this application provides a maximum likelihood sequence estimation method for multi-level shaped signals, comprising:

[0006] The process involves acquiring multiple discrete level symbols corresponding to a multi-level shaped signal, and dividing the state space of the multi-level shaped signal into logic regions corresponding to each discrete level symbol; the multi-level shaped signal includes: an orthogonal double binary shaped signal or a quaternary shaped signal;

[0007] The signal transfer paths corresponding to each of the logical regions are pruned to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions. The pruning includes: adaptive state pruning based on preset pruning rules and branch pruning based on path discarding.

[0008] The reserved path set is used as an effective state transition branch for Viterbi decoding to obtain the maximum likelihood sequence estimation result data corresponding to the multi-level shaped signal.

[0009] In some embodiments of this application, obtaining multiple discrete level symbols corresponding to a multi-level shaped signal includes:

[0010] The multi-level shaped signal is subjected to multi-level pulse amplitude modulation decision processing to obtain multiple discrete level symbols corresponding to the multi-level shaped signal.

[0011] In some embodiments of this application, if the multi-level shaped signal is the orthogonal double binary shaped signal, then correspondingly, before obtaining the multiple discrete level symbols corresponding to the multi-level shaped signal, the method further includes:

[0012] The receiver receives an orthogonal double binary shaped signal transmitted via optical fiber, wherein the orthogonal double binary shaped signal is generated in advance by the transmitter after the modulated signal is orthogonally shaped into a double binary shape.

[0013] In some embodiments of this application, the step of pruning the signal transfer paths corresponding to each of the logical regions to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions includes:

[0014] Based on the amplitude of the orthogonal double binary shaped signal and the prior probability of each discrete level symbol of the pre-acquired orthogonal double binary shaped signal, the maximum a posteriori probability decision processing is performed on each of the logic regions to obtain the target decision value corresponding to each of the logic regions.

[0015] Adaptive state pruning based on preset pruning rules is performed on the signal transfer paths corresponding to the logical regions whose target decision value is not 0, and branch pruning based on path discarding is performed on the signal transfer paths corresponding to the logical regions whose target decision value is 0, so as to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions.

[0016] In some embodiments of this application, the discrete level symbols corresponding to the orthogonal double binary shaped signal include: 7 discrete level symbols;

[0017] Correspondingly, the adaptive state pruning based on a preset pruning rule for the signal transfer path corresponding to the logic region where the target decision value is not 0 includes:

[0018] If the target decision value corresponding to the logical region is +6 or -6, then retain a state corresponding to that logical region;

[0019] If the target decision value corresponding to the logical region is +4 or -4, then the two states corresponding to the logical region are retained.

[0020] If the target decision value corresponding to the logical region is +2 or -2, then the three states corresponding to the logical region are retained;

[0021] Correspondingly, the branch pruning based on path discarding for the signal transfer path corresponding to the logic region with a target decision value of 0 includes:

[0022] If there exists a target decision value of 0 corresponding to the logical region, then the symbol pair specified by the preset pruning rule is discarded in that logical region.

[0023] In some embodiments of this application, if the multi-level shaped signal is the quaternary shaped signal, then correspondingly, before obtaining the multiple discrete level symbols corresponding to the multi-level shaped signal, the method further includes:

[0024] The system receives a quaternary shaped signal transmitted via optical fiber, wherein the quaternary shape is generated in advance by the transmitter after performing quaternary shaping on the modulated signal.

[0025] In some embodiments of this application, the step of pruning the signal transfer paths corresponding to each of the logical regions to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions includes:

[0026] Based on the discrete level symbols corresponding to the quaternary shaped signal, adaptive state trimming is performed on each of the logic regions according to preset trimming rules.

[0027] Using the preset inter-dependency data between each discrete level symbol as a pruning criterion, the signal transfer paths corresponding to each of the logic regions are pruned respectively.

[0028] For the logic region corresponding to the maximum absolute value in each of the discrete level symbols, branch pruning based on path discarding is performed.

[0029] In some embodiments of this application, the discrete level symbols corresponding to the quaternary shaped signal include: 5 discrete level symbols;

[0030] Correspondingly, the adaptive state trimming based on a preset trimming rule for each of the discrete level symbols corresponding to the quaternary shaped signal and each of the logic regions includes:

[0031] For the logic region whose discrete level symbol is +4 or -4, retain the current state corresponding to the logic region and a state adjacent to the current state;

[0032] For the logic region whose discrete level symbol is +2, -2 or 0, retain the current state corresponding to the logic region and the states adjacent to both sides of the current state;

[0033] Correspondingly, the branch pruning based on path discarding for the logic region corresponding to the maximum absolute value in each of the discrete level symbols includes:

[0034] For the logic region where the discrete level symbol is +4 or -4, the symbol pair specified by the preset pruning rule is discarded in the logic region.

[0035] Another aspect of this application provides a maximum likelihood sequence estimation apparatus for multi-level shaped signals, comprising:

[0036] The symbol discretization and region division module is used to acquire multiple discrete level symbols corresponding to the multi-level shaped signal, and to divide the state space of the multi-level shaped signal into logic regions corresponding to each of the discrete level symbols; the multi-level shaped signal includes: orthogonal double binary shaped signal or quaternary shaped signal;

[0037] The pruning module is used to prune the signal transfer paths corresponding to each of the logical regions to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions. The pruning includes: adaptive state pruning based on preset pruning rules and branch pruning based on path discarding.

[0038] The Viterbi decoding module is used to perform Viterbi decoding on the set of retained paths as effective state transition branches to obtain the maximum likelihood sequence estimation result data corresponding to the multi-level shaped signal.

[0039] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the maximum likelihood sequence estimation method for multi-level shaped signals.

[0040] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the described maximum likelihood sequence estimation method for multi-ary shaped signals.

[0041] A fifth aspect of this application provides a computer program product including a computer program that, when executed by a processor, implements the described maximum likelihood sequence estimation method for multi-ary integer signals.

[0042] The maximum likelihood sequence estimation method for multi-level shaped signals provided in this application obtains multiple discrete-level symbols corresponding to the multi-level shaped signal and divides the state space of the multi-level shaped signal into logic regions corresponding to each discrete-level symbol. The multi-level shaped signal includes orthogonal binary shaped signals or quaternary shaped signals. The signal transition paths corresponding to each logic region are pruned to obtain a set of retained paths composed of the retained signal transition paths of each logic region. The pruning includes adaptive state pruning based on preset pruning rules and branch pruning based on path discarding. The set of retained paths is used as effective state transition branches for Viterbi decoding to obtain the maximum likelihood sequence estimation result data corresponding to the multi-level shaped signal. By leveraging the signal symmetry generated by multi-level shaping, combined with adaptive state pruning and branch optimization strategies, the computational complexity of the maximum likelihood sequence estimation algorithm can be significantly reduced while maintaining decoding performance. It can be applied to orthogonal double binary shaped signals and quaternary shaped signals, effectively improving the applicability of the maximum likelihood sequence estimation process. It can effectively solve the problem of high power consumption caused by high complexity in traditional technologies, thereby reducing the power consumption of the signal receiver and improving the spectral efficiency and practicality of the optical communication system.

[0043] Additional advantages, objectives, and features of this application will be set forth in part in the description which follows, and will in part become apparent to those skilled in the art upon review of the following description, or may be learned by practice of the application. The objectives and other advantages of this application can be realized and obtained by means of the structures specifically pointed out in the specification and drawings.

[0044] Those skilled in the art will understand that the purposes and advantages that can be achieved with this application are not limited to those specifically described above, and that the above and other purposes that this application can achieve will be more clearly understood from the following detailed description. Attached Figure Description

[0045] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, do not constitute a limitation thereof. The components in the drawings are not drawn to scale but are merely for illustrating the principles of this application. For ease of illustration and description of certain parts of this application, corresponding portions in the drawings may be enlarged, i.e., may appear larger relative to other components in an exemplary device actually manufactured according to this application. In the drawings:

[0046] Figure 1 This is a schematic diagram of the first step in the maximum likelihood sequence estimation method for multi-base shaped signals according to an embodiment of this application.

[0047] Figure 2 This is a schematic diagram of a second flowchart of a maximum likelihood sequence estimation method for multi-base shaped signals according to an embodiment of this application.

[0048] Figure 3 This is a flowchart illustrating the maximum likelihood sequence estimation method for orthogonal dual binary shaped signals according to an embodiment of this application.

[0049] Figure 4 This is a flowchart illustrating the maximum likelihood sequence estimation method for quaternary shaped signals in one embodiment of this application.

[0050] Figure 5 The following are the verification block diagrams used in application examples 1 and 2 of this application.

[0051] Figure 6 The prior probability distribution diagram of the PAM-7 modulated signal after QDB is provided for application example 1 of this application.

[0052] Figure 7 A branch pruning diagram is provided for application example 1 of this application.

[0053] Figure 8 The graph showing the relationship between bit error rate and optical power for a simplified MLSE algorithm for QDB-shaped signals compared to the original MLSE algorithm is provided for application example 1 of this application.

[0054] Figure 9 The signal prior probability distribution diagram provided for application example 2 of this application.

[0055] Figure 10 A branch pruning diagram is provided for application example 2 of this application.

[0056] Figure 11 The graph showing the relationship between bit error rate and signal-to-noise ratio for a simplified MLSE algorithm for quaternary shaped signals compared to the original MLSE algorithm is provided for application example 2 of this application.

[0057] Figure 12 This is a schematic diagram of the maximum likelihood sequence estimation device for multi-level shaped signals according to an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and their descriptions are used to explain this application, but are not intended to limit it.

[0059] It should also be noted that, in order to avoid obscuring this application with unnecessary details, only the structures and / or processing steps closely related to the solution according to this application are shown in the accompanying drawings, while other details that are not closely related to this application are omitted.

[0060] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0061] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0062] In the following description, embodiments of the present application will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0063] To address the high power consumption issue caused by the high complexity of traditional maximum likelihood sequence estimation algorithms, this application provides a maximum likelihood sequence estimation method for multi-ary shaped signals, a maximum likelihood sequence estimation device for multi-ary shaped signals for executing the method, a physical device, a computer-readable storage medium, and a computer program product. These methods effectively reduce the computational complexity of maximum likelihood sequence estimation algorithms and improve their applicability, thereby reducing power consumption at the signal receiver and improving the spectral efficiency and practicality of optical communication systems.

[0064] The following examples will provide a detailed description.

[0065] Based on this, embodiments of this application provide a maximum likelihood sequence estimation method for multi-level shaped signals, which can be implemented by a maximum likelihood sequence estimation device for multi-level shaped signals. See [link to relevant documentation]. Figure 1 The maximum likelihood sequence estimation method for multi-base integer signals specifically includes the following:

[0066] Step 100: Obtain multiple discrete level symbols corresponding to the multi-level shaped signal, and divide the state space of the multi-level shaped signal into logic regions corresponding to each discrete level symbol; the multi-level shaped signal includes: orthogonal double binary shaped signal or quaternary shaped signal.

[0067] The maximum likelihood sequence estimation device for multi-level shaped signals can be set as a functional module in the maximum likelihood sequence estimation decoder of the receiving device (or simply the receiving end).

[0068] In the maximum likelihood sequence estimation algorithm, discrete level symbols can be identified by integers, such as -6, -4, -2, 0, +2, +4, and +6.

[0069] In one or more embodiments of this application, the multi-base shaped signal may be an orthogonal double binary (QDB) shaped signal or a quaternary shaped signal.

[0070] After determining the multiple discrete level symbols corresponding to the multi-level shaped signal in step 100, the prior probabilities corresponding to each discrete level symbol can be obtained simultaneously. Based on a preset decision threshold, the state space is divided into logic regions corresponding to each discrete level symbol. Here, the state space specifically refers to the continuous range of received signal amplitude. The logic region is a plurality of mutually exclusive intervals divided by the decision threshold into which the amplitude space is divided, with each interval corresponding to a discrete level symbol.

[0071] Step 200: Prune the signal transfer paths corresponding to each of the logical regions to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions. The pruning includes: adaptive state pruning based on preset pruning rules and branch pruning based on path discarding.

[0072] In step 200, the multi-level integer signal corresponds to an original trellis diagram. This trellis diagram is a graphical tool used in communication systems to describe the state transitions of signal sequences. The trellis diagram represents the evolution of the signal sequence through a discrete-time state machine, consisting of nodes and edges. Nodes represent the state of the system at a certain moment (such as the input value or filter state at the previous moment); edges represent nodes connecting adjacent moments, labeled with input symbols and corresponding output signal values. The multi-level integer signal requires a trellis diagram to resolve the sequence dependencies caused by memory. It is understood that each logical region also has its own corresponding partial trellis diagram derived from the original trellis diagram, used to display the signal transition paths formed between the various nodes and edges.

[0073] In one or more embodiments of this application, the rule data corresponding to the preset pruning rule can be pre-defined by a human; the rule data corresponding to the path discard can also be pre-defined by a human.

[0074] In other words, although state reduction and branch pruning techniques have been explored to reduce computational overhead and lower the complexity and power consumption of MLSE, these traditional pruning techniques typically employ fixed-state compression strategies, which cannot achieve high-quality signal reception under complex channel conditions. Furthermore, these pruning rules are not effectively applicable when faced with more advanced multi-level shaping schemes. Therefore, step 200 of this application employs adaptive state pruning and branch pruning (i.e., a branch optimization strategy), which can significantly reduce the computational complexity of the MLSE algorithm while maintaining decoding performance.

[0075] Step 300: Use the reserved path set as an effective state transition branch for Viterbi decoding to obtain the maximum likelihood sequence estimation result data corresponding to the multi-level shaped signal.

[0076] In step 300, the set of retained paths can be used as effective state transition branches and input into the Viterbi decoder for sequence metric calculation. Finally, the minimum metric path is output as the decoding result, which is the maximum likelihood sequence estimation result data corresponding to the multi-level shaped signal.

[0077] As can be seen from the above description, the maximum likelihood sequence estimation method for multi-level shaped signals provided in this application, by utilizing the signal symmetry generated by multi-level shaped signals and combining adaptive state pruning and branch optimization strategies, can significantly reduce the computational complexity of the maximum likelihood sequence estimation algorithm while maintaining decoding performance. It can be applied to orthogonal double binary shaped signals and quaternary shaped signals, effectively improving the applicability of the maximum likelihood sequence estimation process. It can effectively solve the problem of high power consumption of equipment caused by high complexity in traditional technologies, thereby reducing the power consumption of the signal receiver and improving the spectral efficiency and practicality of the optical communication system.

[0078] To further improve the effectiveness and reliability of maximum likelihood sequence estimation for multi-ary shaped signals, this application provides a method for maximum likelihood sequence estimation of multi-ary shaped signals, see [link to relevant documentation]. Figure 2 Step 100 in the maximum likelihood sequence estimation method for multi-base shaped signals specifically includes the following:

[0079] Step 110: Perform multi-level pulse amplitude modulation decision processing on the multi-level shaped signal to obtain multiple discrete level symbols corresponding to the multi-level shaped signal.

[0080] The process of Pulse Amplitude Modulation (PAM-M) decision can be as follows: based on the preset M standard level values, calculate M-1 decision thresholds; and then classify the received multi-level shaped signal into M discrete level symbols based on the decision thresholds.

[0081] Step 120: Divide the state space of the multi-level shaped signal into logic regions corresponding to each of the discrete level symbols.

[0082] In step 120, the state space of the multi-level shaped signal is divided into M logic regions that are only associated with a specific amplitude level of the received signal (i.e., the multi-level shaped signal) according to M-1 decision thresholds.

[0083] To further improve the effectiveness and reliability of maximum likelihood sequence estimation for QDB-shaped signals, this application provides a maximum likelihood sequence estimation method for multi-level shaped signals, see [link to relevant documentation]. Figure 3 The method for maximum likelihood sequence estimation of multi-base shaped signals includes the following steps prior to step 100:

[0084] Step 010: Receive the orthogonal double binary shaped signal transmitted via optical fiber, wherein the orthogonal double binary shaped signal is generated in advance by the transmitter after orthogonally shaped into a double binary shape from the modulated signal.

[0085] Correspondingly, to further reduce the computational complexity of maximum likelihood sequence estimation for QDB-shaped signals, in a maximum likelihood sequence estimation method for multi-level shaped signals provided in this application embodiment, see [link to relevant documentation]. Figure 3 Step 200 in the maximum likelihood sequence estimation method for multi-base shaped signals specifically includes the following:

[0086] Step 210: Based on the amplitude of the orthogonal double binary shaped signal and the prior probability of each discrete level symbol of the pre-acquired orthogonal double binary shaped signal, perform maximum a posteriori probability decision processing on each logic region to obtain the target decision value corresponding to each logic region.

[0087] Among them, the maximum a posteriori probability decision (MAP) calculates the posterior probability of the symbol based on the received signal amplitude and the prior probability distribution, and then selects the discrete level symbol with the highest posterior probability as the decision output. The target decision value can also be written as MAP decision or MAP decision result.

[0088] In other words, in this embodiment of the application, compared with the prior art, step 210 makes full use of the prior probability and error distribution of the received signal, effectively reducing computational complexity while maintaining robust decoding performance under strict bandwidth constraints. Here, the error distribution is the amplitude probability distribution of the multi-level shaped signal after being affected by noise.

[0089] Step 220: Perform adaptive state pruning based on preset pruning rules on the signal transfer paths corresponding to the logic regions where the target decision value is not 0, and perform branch pruning based on path discarding on the signal transfer paths corresponding to the logic regions where the target decision value is 0, so as to obtain a set of retained paths composed of the signal transfer paths retained by each logic region.

[0090] To further reduce the computational complexity of maximum likelihood sequence estimation for QDB-shaped signals, in the maximum likelihood sequence estimation method for multi-level shaped signals provided in this application embodiment, the discrete level symbols corresponding to the orthogonal dual binary shaped signals include: 7 discrete level symbols; correspondingly, step 220 in the maximum likelihood sequence estimation method for multi-level shaped signals specifically includes the following:

[0091] Step 221: If there exists a target decision value of +6 or -6 corresponding to the logical region, then retain a state corresponding to that logical region.

[0092] Step 222: If there exists a target decision value of +4 or -4 corresponding to the logical region, then retain the two states corresponding to the logical region.

[0093] Step 223: If there exists a target decision value of +2 or -2 corresponding to the logical region, then retain the three states corresponding to that logical region.

[0094] Step 224: If there exists a target decision value of 0 corresponding to the logical region, then discard the symbol pair specified by the preset pruning rule in that logical region.

[0095] To further improve the effectiveness and reliability of maximum likelihood sequence estimation for quaternary shaped signals, this application provides a method for maximum likelihood sequence estimation for multi-base shaped signals, see [link to relevant documentation]. Figure 4 The method for maximum likelihood sequence estimation of multi-base shaped signals includes the following steps prior to step 100:

[0096] Step 020: Receive the quaternary shaped signal transmitted via optical fiber, wherein the quaternary shaped signal is generated in advance by the transmitter after performing quaternary shaped signal on the modulated signal.

[0097] Correspondingly, to further reduce the computational complexity of maximum likelihood sequence estimation for quaternary integer signals, in a maximum likelihood sequence estimation method for multi-base integer signals provided in this application embodiment, see [link to relevant documentation]. Figure 4 Step 200 in the maximum likelihood sequence estimation method for multi-base shaped signals specifically includes the following:

[0098] Step 230: Based on the discrete level symbols corresponding to the quaternary shaped signal, perform adaptive state trimming based on preset trimming rules for each of the logic regions.

[0099] Step 240: Using the preset inter-dependency data between each of the discrete level symbols as the pruning criterion, prune the signal transfer paths corresponding to each of the logic regions respectively.

[0100] The inter-dependency data between the discrete level symbols can be pre-acquired manually and sent to a maximum likelihood sequence estimation device for multi-level shaped signals for storage.

[0101] Step 250: Perform branch pruning based on path discarding for the logic region corresponding to the maximum absolute value in each of the discrete level symbols.

[0102] To further reduce the computational complexity of maximum likelihood sequence estimation for quaternary integer signals, in a maximum likelihood sequence estimation method for multi-base integer signals provided in this application embodiment, the discrete level symbols corresponding to the quaternary integer signal include 5 discrete level symbols; correspondingly, step 230 in the maximum likelihood sequence estimation method for multi-base integer signals specifically includes the following:

[0103] Step 231: For the logic region whose discrete level symbol is +4 or -4, retain the current state corresponding to the logic region and a state adjacent to the current state;

[0104] Step 232: For the logic region whose discrete level symbol is +2, -2 or 0, retain the current state corresponding to the logic region and the states adjacent to the current state on both sides.

[0105] Correspondingly, step 250 in the maximum likelihood sequence estimation method for multi-base shaped signals specifically includes the following:

[0106] Step 251: For the logic region where the discrete level symbol is +4 or -4, discard the symbol pair specified by the preset pruning rule in the logic region.

[0107] To further illustrate the above embodiments, this application also provides a specific application example of the maximum likelihood sequence estimation method for multi-ary integer signals, namely, a low-complexity maximum likelihood sequence estimation (MLSE) method for multi-ary integer signals. Figure 5 A corresponding verification block diagram is presented, belonging to the field of optical fiber communication technology. This method, by utilizing the signal symmetry generated by multi-level shaping and combining adaptive state pruning and branch optimization strategies, can significantly reduce the computational complexity of the MLSE algorithm while maintaining decoding performance. Specifically, it includes two implementation methods: orthogonal dual binary (QDB) shaping and quaternary shaping, effectively solving the high power consumption problem caused by high complexity in traditional technologies, and improving the spectral efficiency and practicality of optical communication systems.

[0108] The specific explanation is as follows:

[0109] (I) Application Example 1: A low-complexity MLSE method for QDB integer signals, including the following steps:

[0110] Step 1: At the transmitting end, the modulated signal is shaped using a QDB.

[0111] The transfer function for QDB integers is:

[0112] H(z) = 1 + z -1 (1)

[0113] Where z represents a complex variable.

[0114] In the DSP process at the transmitting end, the pseudo-random binary sequence is first mapped into 16QAM symbols, and then QDB shaping is performed. The resulting signal is processed and transmitted through a standard single-mode fiber (SSMF) over a certain length, and then received at the receiving end for offline DSP.

[0115] Step 2: The receiver uses a simplified MLSE algorithm to probe the signal after QDB.

[0116] The simplified MLSE algorithm for QDB-shaped signals is as follows:

[0117] Step 2.1: Perform PAM-7 decision on the PAM-7 signal obtained after QDB transformation to obtain the decision result, which includes 7 discrete level symbols, i.e., the PAM-7 modulated signal after QDB.

[0118] Figure 6 The prior probability plot of the PAM-7 modulated signal after QDB is shown. The prior probabilities of the PAM-7 modulated signal are [1 / 16, 1 / 8, 3 / 16, 1 / 4, 3 / 16, 1 / 8, 1 / 16].

[0119] Step 2.2: Based on the decision result, divide the state space into logical regions associated only with specific amplitude levels of the received signal, including R1 to R7; in the PAM-7 decision scenario, the state space specifically refers to the continuous range of values ​​for the received signal amplitude. The logical regions R1 to R7, divided according to the decision result, are seven mutually exclusive intervals into which the amplitude space is divided by the decision threshold, each interval corresponding to a discrete level symbol.

[0120] Step 2.3: Implement adaptive state pruning for different logical regions;

[0121] When the MAP decision is ±6, only one state is retained; when the decision is ±4, two states are retained; when the decision is ±2, three states are retained; and when the MAP decision is 0, no direct state pruning is performed.

[0122] Step 2.4: For some states that cannot be directly pruned, further optimization is performed on the branch metric calculation (BM) by discarding some symbol pairs that have a very low probability of being close to zero after QDB transformation.

[0123] That is, when the MAP decision is 0, the BM calculation is further optimized by discarding some symbol pairs that have a very low probability of being close to zero after the QDB transformation;

[0124] For example, the symbol pairs {1,3}, {1,1}, {3,1}, {3,3}, {-1,-3}, {-1,-1}, {-3,-1}, {-3,-3}, {-1,3}, {3,-1}, {-3,1}, and {1,-3} in the logical region corresponding to a MAP decision of 0 are discarded.

[0125] Among them, the symbol pairs with a very low probability of approaching zero are: symbol pairs ({1, 3}, {1, 1}, {3, 1}, {3, 3}, {-1, -3}, {-1, -1}, {-3, -1}, {-3, -3}, {-1, 3}, {3, -1}, {-3, 1}, {1, -3}). QDB is the sum of the previous symbol and the next symbol. The probability of these symbols approaching zero after QDB transformation is very low because these symbol pairs will not produce a 0 level after QDB transformation. Only symbol pairs with opposite signs (such as {-1, 1}) will become a 0 level after QDB transformation.

[0126] Among them, MLSE branch pruning for QDB-shaped signals is as follows: Figure 7 As shown, S0 to Sk represent different states.

[0127] Step 2.5: Finally, output the MLSE decoding result via Viterbi.

[0128] in, Figure 8 This is a comparison of the bit error rate (BER) of the simplified MLSE and the original MLSE under different ROPs. Compared with the traditional MLSE, the performance degradation of the proposed simplified MLSE under all ROPs is negligible. Here, ROP represents optical power; BER represents bit error rate.

[0129] (II) Application Example 2: A low-complexity MLSE method for quaternary integer signals, including the following steps:

[0130] Step 1: The transmitting end performs quaternary shaping on the modulated signal;

[0131] The transfer function for quaternion integers is:

[0132] H(z) = 1 + 2z -1 +z -2 (2)

[0133] That is, at the transmitting end, the original pseudo-random binary sequence is QPSK modulated, and the modulated signal is then quaternized and processed to obtain the digital signal to be transmitted. After the transmitted signal is transmitted through an optical fiber over a certain distance, it is coherently received, and the received signal is processed by an offline DSP.

[0134] Step 2: The receiver uses a simplified MLSE algorithm to probe the shaped signal.

[0135] The simplified MLSE algorithm for quaternion integers is as follows:

[0136] Step 2.1: Perform PAM-M decision on the signal obtained after quaternion shaping;

[0137] That is, the PAM-5 signal obtained after quaternary shaping of QPSK is pre-determined to obtain a rough decision result.

[0138] Figure 9 The prior probability plot of the quaternion-shaped PAM-5 signal is shown, as follows: Figure 9 As shown, the prior probabilities of the PAM-5 modulated signal are [1 / 8, 1 / 4, 1 / 4, 1 / 4, 1 / 8].

[0139] Step 2.2: Based on the PAM-5 decision result, divide the state space into five logic regions that are only associated with a specific amplitude level of the received signal, defined as R1 to R5;

[0140] Step 2.3: Implement adaptive state pruning for different logical regions;

[0141] When the PAM-5 pre-decision result is ±4, the current state and one adjacent state are retained; when the decision result is ±2 and 0, the current state and the two states on both sides are retained.

[0142] Step 2.4: By introducing inter-symbol dependencies as a pruning criterion, the branches in the original mesh graph are restricted to the effective paths in the signal space, which greatly reduces the number of search paths;

[0143] Step 2.5: For some states, discard some symbol pairs with lower probabilities after the transformation, and then perform further branch pruning.

[0144] Taking state -4 as an example, although its neighboring state -2 can also reach the next states -4 and -2, the probability is much smaller compared to the paths {-4, -4} and {-4, -2}, so it can be discarded. The situation for state 4 is similar.

[0145] In other words, for states ±4, additional branch pruning can be performed by discarding some symbol pairs with a low probability of reaching the current state after integer transformation. Here, probability refers to the fact that the likelihood of some adjacent symbol pairs reaching state 4 or -4 after multi-base integer transformation is very small. Therefore, this method directly discards some paths without setting a threshold.

[0146] Figure 10 This diagram illustrates branch pruning in the MLSE algorithm for quaternary integer signals. For states ±4, only 4 paths are retained; for states ±2, 9 paths are retained; and for state 0, 11 paths are retained. n and S n+1 This indicates adjacent states.

[0147] Step 2.6: Finally, output the MLSE decoding result via Viterbi.

[0148] Figure 11 This paper shows the bit error rate (BER) and signal-to-noise ratio (OSNR) relationship between the simplified MLSE algorithm for quaternary shaped signals provided in this application and the traditional original MLSE algorithm. Under different SNR conditions, the proposed simplified MLSE scheme maintains comparable BER performance to the traditional MLSE. At 3.8e -3 When the hard-decision forward error correction threshold is reached, only a 0.7dB OSNR loss is generated, while the complexity is reduced to 33% of the original. HD-FEC (Hard-Decision Forward Error Correction) refers to a coding technique that achieves error correction through discrete-level decisions.

[0149] In summary, the adaptive pruning MLSE algorithms for QDB-shaped and quaternary-shaped signals provided in Examples 1 and 2 of this application, compared with existing technologies, fully utilize the prior and error distribution of the signal, effectively reducing computational complexity while maintaining robust decoding performance under strict bandwidth constraints. Further research was conducted on pruning algorithms under existing QDB schemes. Through quaternary shaping, the system's SE was significantly improved. Simultaneously, the adaptive MLSE pruning algorithm effectively utilizes the correlation between symbols, achieving a balance between computational complexity and decoding performance under strict bandwidth constraints.

[0150] From a software perspective, this application also provides an apparatus for performing all or part of the maximum likelihood sequence estimation method for multi-ary shaped signals, see [link to relevant documentation]. Figure 12 The maximum likelihood sequence estimation device for multi-level shaped signals specifically includes the following components:

[0151] The symbol discretization and region division module 10 is used to acquire multiple discrete level symbols corresponding to the multi-level shaped signal, and to divide the state space of the multi-level shaped signal into logic regions corresponding to each of the discrete level symbols; the multi-level shaped signal includes: orthogonal double binary shaped signal or quaternary shaped signal.

[0152] The pruning module 20 is used to prune the signal transfer paths corresponding to each of the logical regions to obtain a set of retained paths composed of the signal transfer paths retained by each of the logical regions. The pruning includes: adaptive state pruning based on preset pruning rules and branch pruning based on path discarding.

[0153] The Viterbi decoding module 30 is used to perform Viterbi decoding on the set of retained paths as effective state transition branches to obtain the maximum likelihood sequence estimation result data corresponding to the multi-level shaped signal.

[0154] The embodiments of the maximum likelihood sequence estimation apparatus for multi-ary integer signals provided in this application can be used to execute the processing flow of the embodiments of the maximum likelihood sequence estimation method for multi-ary integer signals in the above embodiments. Its functions will not be repeated here, but can be referred to the detailed description of the embodiments of the maximum likelihood sequence estimation method for multi-ary integer signals described above.

[0155] The maximum likelihood sequence estimation part of the multi-base integer signal maximum likelihood sequence estimation device can be completed in a client device or a server. The specific choice depends on the processing capability of the client device and the limitations of the user's usage scenario. This application does not impose any limitations in this regard. If all operations are completed in the client device, the client device may further include a processor for the specific processing of the maximum likelihood sequence estimation for the multi-base integer signal.

[0156] As can be seen from the above description, the maximum likelihood sequence estimation device for multi-level shaped signals provided in this application, by utilizing the signal symmetry generated by multi-level shaped signals and combining adaptive state pruning and branch optimization strategies, can significantly reduce the computational complexity of the maximum likelihood sequence estimation algorithm while maintaining decoding performance. It can be applied to orthogonal double binary shaped signals and quaternary shaped signals, effectively improving the applicability of the maximum likelihood sequence estimation process. It can effectively solve the problem of high power consumption of equipment caused by high complexity in traditional technologies, thereby reducing the power consumption of the signal receiver and improving the spectral efficiency and practicality of the optical communication system.

[0157] This application also provides an electronic device, which may include a processor, a memory, a receiver, and a transmitter. The processor is used to execute the maximum likelihood sequence estimation method for multi-level shaped signals mentioned in the above embodiments. The processor and the memory can be connected via a bus or other means, taking a bus connection as an example. The receiver can be connected to the processor and the memory via wired or wireless means.

[0158] The processor can be a central processing unit (CPU). The processor can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0159] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the maximum likelihood sequence estimation method for multi-ary integer signals in the embodiments of this application. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory, thereby implementing the maximum likelihood sequence estimation method for multi-ary integer signals in the above method embodiments.

[0160] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0161] The one or more modules are stored in the memory, and when executed by the processor, they perform the maximum likelihood sequence estimation method for multi-base integer signals in the implementation embodiment.

[0162] In some embodiments of this application, the user equipment may include a processor, a memory, and a transceiver unit. The transceiver unit may include a receiver and a transmitter. The processor, memory, receiver, and transmitter may be connected via a bus system. The memory is used to store computer instructions, and the processor is used to execute the computer instructions stored in the memory to control the transceiver unit to send and receive signals.

[0163] As one implementation method, the functions of the receiver and transmitter in this application can be implemented by transceiver circuits or dedicated transceiver chips, and the processor can be implemented by dedicated processing chips, processing circuits or general-purpose chips.

[0164] As another implementation approach, the server provided in this application embodiment can be implemented using a general-purpose computer. That is, the program code implementing the processor, receiver, and transmitter functions is stored in memory, and the general-purpose processor implements the processor, receiver, and transmitter functions by executing the code in memory.

[0165] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned maximum likelihood sequence estimation method for multi-level shaped signals. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0166] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the aforementioned maximum likelihood sequence estimation method for multi-ary shaped signals.

[0167] Those skilled in the art will understand that the 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application 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 data signals carried on a carrier wave.

[0168] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0169] In this application, 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 in place of features of other embodiments.

[0170] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to the embodiments of this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A maximum likelihood sequence estimation method for a multi-ary shaped signal, characterized by, The method comprises the steps of: obtaining a plurality of discrete level symbols corresponding to a multi-level shaping signal, and dividing a state space of the multi-level shaping signal into a plurality of logical regions corresponding to the plurality of discrete level symbols respectively; the multi-level shaping signal comprises a quadrature dual binary shaping signal or a quaternary shaping signal; pruning a signal transition path corresponding to each of the logical regions respectively to obtain a reserved path set composed of signal transition paths reserved by each of the logical regions respectively, wherein the pruning comprises adaptive state pruning based on a preset pruning rule and branch pruning based on path abandonment; using the reserved path set as valid state transition branches for Viterbi decoding to obtain maximum likelihood sequence estimation result data corresponding to the multi-level shaping signal; the step of obtaining a plurality of discrete level symbols corresponding to a multi-level shaping signal comprises: performing multi-level pulse amplitude modulation decision processing on the multi-level shaping signal to obtain a plurality of discrete level symbols corresponding to the multi-level shaping signal; if the multi-level shaping signal is the quadrature dual binary shaping signal, then before the step of obtaining a plurality of discrete level symbols corresponding to a multi-level shaping signal, the method further comprises the steps of: receiving a quadrature dual binary shaping signal transmitted through an optical fiber, wherein the quadrature dual binary shaping signal is generated by performing quadrature dual binary shaping on a modulated signal at a transmitting end; the step of pruning a signal transition path corresponding to each of the logical regions respectively to obtain a reserved path set composed of signal transition paths reserved by each of the logical regions respectively comprises: performing maximum a posteriori probability decision processing on each of the logical regions according to an amplitude corresponding to the quadrature dual binary shaping signal and a prior probability corresponding to each of the discrete level symbols corresponding to the quadrature dual binary shaping signal respectively to obtain a target decision value corresponding to each of the logical regions respectively; performing adaptive state pruning based on a preset pruning rule on a signal transition path corresponding to a logical region with a target decision value of 0, and performing branch pruning based on path abandonment on a signal transition path corresponding to a logical region with a target decision value of 0 to obtain a reserved path set composed of signal transition paths reserved by each of the logical regions respectively.

2. The method of maximum likelihood sequence estimation for a multi-ary shaped signal according to claim 1, wherein, each of the discrete level symbols corresponding to the quadrature dual binary shaping signal comprises seven discrete level symbols; correspondingly, the step of performing adaptive state pruning based on a preset pruning rule on a signal transition path corresponding to a logical region with a target decision value of 0 comprises: if the target decision value corresponding to the logical region is +6 or -6, one state corresponding to the logical region is reserved; if the target decision value corresponding to the logical region is +4 or -4, two states corresponding to the logical region are reserved; if the target decision value corresponding to the logical region is +2 or -2, three states corresponding to the logical region are reserved; correspondingly, the step of performing branch pruning based on path abandonment on a signal transition path corresponding to a logical region with a target decision value of 0 comprises: If the target decision value corresponding to the logical region is 0, discarding the symbol pair specified by the preset pruning rule in the logical region.

3. The method for maximum likelihood sequence estimation of a multi-ary shaped signal according to claim 1, wherein, If the multi-ary shaping signal is the quaternary shaping signal, before the obtaining of the plurality of discrete level symbols corresponding to the multi-ary shaping signal, the method further includes: Receiving a quaternary shaping signal transmitted through an optical fiber, wherein the quaternary shaping signal is generated by performing quaternary shaping on a modulated signal at a transmitting end.

4. The method of maximum likelihood sequence estimation for a multi-ary shaped signal according to claim 3, wherein, The pruning of the signal transition paths corresponding to each of the logical regions respectively includes: According to each discrete level symbol corresponding to the quaternary shaping signal, performing adaptive state pruning based on a preset pruning rule on each of the logical regions respectively; Taking the preset interdependence relationship data between each of the discrete level symbols as a pruning criterion, pruning the signal transition paths corresponding to each of the logical regions respectively; For the logical region corresponding to the maximum absolute value in each of the discrete level symbols, performing branch pruning based on path discarding.

5. The method for maximum likelihood sequence estimation of a multi-ary shaped signal according to claim 4, wherein, Each discrete level symbol corresponding to the quaternary shaping signal includes five discrete level symbols. Accordingly, the adaptive state pruning based on the preset pruning rule on each of the logical regions respectively according to each discrete level symbol corresponding to the quaternary shaping signal includes: For the logical region in which the discrete level symbol is +4 or -4, retaining the current state corresponding to the logical region and one state adjacent to the current state; For the logical region in which the discrete level symbol is +2, -2 or 0, retaining the current state corresponding to the logical region and the states adjacent to the current state on both sides respectively; Accordingly, the branch pruning based on path discarding on the logical region corresponding to the maximum absolute value in each of the discrete level symbols includes: For the logical region in which the discrete level symbol is +4 or -4, discarding the symbol pair specified by the preset pruning rule in the logical region.

6. A maximum likelihood sequence estimation device for a multi-ary shaped signal, characterized by, The method includes: A symbol discretization and region division module is configured to obtain a plurality of discrete level symbols corresponding to a multi-ary shaping signal, and divide a state space of the multi-ary shaping signal into logical regions corresponding to each of the discrete level symbols respectively; The multi-ary shaping signal includes a quadrature dual binary shaping signal or a quaternary shaping signal; A pruning module is configured to prune signal transition paths corresponding to each of the logical regions respectively to obtain a reserved path set composed of signal transition paths reserved by each of the logical regions respectively, wherein the pruning includes adaptive state pruning based on a preset pruning rule and branch pruning based on path discarding; A Viterbi decoding module is configured to take the reserved path set as valid state transition branches to perform Viterbi decoding to obtain maximum likelihood sequence estimation result data corresponding to the multi-ary shaping signal; The obtaining of the plurality of discrete level symbols corresponding to the multi-ary shaping signal includes: performing a multi-amplitude modulation decision processing on the multi-ary shaping signal to obtain a plurality of discrete level symbols corresponding to the multi-ary shaping signal; If the multi-ary shaping signal is the quadrature dual binary shaping signal, the device is further configured to perform, before the obtaining of the plurality of discrete level symbols corresponding to the multi-ary shaping signal: receiving a quadrature dual binary shaping signal transmitted through an optical fiber, wherein the quadrature dual binary shaping signal is generated by performing quadrature dual binary shaping on a modulated signal at a transmitting end; wherein the pruning, respectively, of the signal transition paths corresponding to each of the logical regions to obtain a reserved path set composed of the signal transition paths reserved by each of the logical regions comprises: performing a maximum a posteriori probability decision processing on each of the logical regions according to the amplitude of the quadrature dual binary shaping signal and the prior probability of each of the discrete level symbols corresponding to the quadrature dual binary shaping signal to obtain a target decision value corresponding to each of the logical regions; performing an adaptive state pruning on the signal transition paths corresponding to the logical region with the target decision value of 0 based on a preset pruning rule, and performing a branch pruning on the signal transition paths corresponding to the logical region with the target decision value of 0 based on path abandonment, to obtain a reserved path set composed of the signal transition paths reserved by each of the logical regions.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the maximum likelihood sequence estimation method for a multi-ary shaping signal according to any one of claims 1 to 5 when executing the computer program.

Images