A method and system for digital signal optical communication based on single pilot symbol

Abstract

The application provides a kind of digital signal optical communication method and system based on single pilot symbol, and the sending end periodically inserts X / Y polarized pilot symbol pair, is shaped by orthogonal double binary coding super-Nyquist signal, so that each pair of pilot symbols is merged into equivalent single pilot symbol, phase and amplitude information are retained and mapped to the outermost circle of QAM constellation diagram.Receiving end finds initial pilot symbol sequence based on the characteristics of the highest amplitude of the outermost circle of QAM constellation diagram, constructs sliding window according to the length of the original pilot symbol sequence in the initial pilot symbol sequence to extract sequence, and compares the original pilot symbol sequence to calculate the variance of phase increment, determines the frame starting position based on the minimum variance, and uses pilot symbol to provide accurate initial information and periodic guide decoding for MLSE algorithm.The application can efficiently utilize pilot symbol resources and significantly reduce the complexity of MLSE decoding.

Description

Technical Field

[0001] This invention relates to the field of optical communication technology, and in particular to a digital signal optical communication method and system based on a single pilot symbol. Background Technology

[0002] With the rapid development of cloud computing, the Internet of Things (IoT), and data center (DC) applications, high-bandwidth, low-power optical interconnect technology has become a core requirement for modern communication systems. Traditional Nyquist-rate transmission is limited by spectral efficiency and cannot meet the ever-increasing bandwidth demands. Super Nyquist (FTN) transmission technology, by compressing symbol intervals in the time or frequency domain and introducing controllable inter-symbol interference (ISI), overcomes the limitations of the Nyquist criterion and significantly improves spectral efficiency. However, FTN signal shaping and decoding face many challenges, especially in high-order modulation formats (such as 16QAM) and symbol rate sampling scenarios, where high signal processing complexity and high power consumption limit its application in power-sensitive environments such as data centers. In FTN systems, pilot symbols are typically used for frame synchronization, carrier recovery (CR), and frequency offset estimation (FOE). In existing technologies, the utilization of pilot symbols is relatively limited, mainly focusing on phase recovery or synchronization functions, failing to fully leverage their potential in the decoding stage. Furthermore, after FTN signal shaping, the strong correlation between symbols leads to an exponential increase in the computational complexity of traditional Maximum Likelihood Sequence Estimation (MLSE) decoding algorithms, especially when processing QDB-16QAM (49QAM) signals, where the required state space and branch metric calculations increase significantly. Existing research shows that traditional MLSE decoding schemes, under the near-binary channel response of QDB signals, require handling complex state transitions, significantly increasing memory requirements and processing latency. Some hybrid decoding schemes attempt to combine single-symbol detection and simplified MLSE, but due to the inability to accurately locate error-causing nodes, frequent MLSE activations are easily triggered in bandwidth-constrained scenarios, leading to decreased decoding efficiency and increased latency. These technical limitations indicate an urgent need for a scheme that efficiently utilizes Pilot symbols, optimizes FTN signal shaping, and performs MLSE decoding to achieve a balance between high performance and low power consumption, meeting the needs of next-generation data center optical interconnects. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a digital signal optical communication method and system based on a single pilot symbol, in order to eliminate or improve one or more defects existing in the prior art, and solve the problems of low pilot symbol utilization, high maximum likelihood sequence estimation decoding complexity and high power consumption in existing FTN transmission systems.

[0004] One aspect of the present invention provides a method for transmitting digital signals in optical communication based on a single pilot symbol, the method comprising the following steps:

[0005] In the data frame to be transmitted, a pair of X / Y polarized pilot symbols are inserted periodically at a first set number of data symbols based on a preset pilot symbol sequence.

[0006] The data frame is shaped and output using orthogonal double binary encoding. Based on the symmetry of the transformation, each pair of X / Y polarized pilot symbols is merged into an equivalent single pilot symbol. The single pilot symbol retains the phase and amplitude information of each pair of pilot symbols for subsequent decoding guidance. All single pilot symbols are mapped to the outermost constellation points in the QAM constellation diagram.

[0007] In some embodiments, before inserting a pair of X / Y polarized pilot symbols at a first predetermined number of data symbols periodic intervals in the data frame to be transmitted, the method further includes:

[0008] Generate a pseudo-random binary sequence and map it to an orthogonal amplitude modulation symbol with a set number of bits;

[0009] Pilot symbol sequences are generated for the orthogonal amplitude modulation symbols based on different X / Y polarization seeds and shared with the receiver.

[0010] In some embodiments, the pseudo-random binary sequence is generated using a linear feedback shift register, or by generating random numbers using a pseudo-random number generator and then converting them into a binary sequence.

[0011] In some embodiments, super Nyquist signal shaping of the data frame by orthogonal double binary encoding further includes: introducing controllable inter-symbol interference using a raised cosine filter or a root raised cosine filter, and adjusting the roll-off factor to control the amount of ISI.

[0012] On the other hand, the present invention also provides a method for receiving digital signals in optical communication based on a single pilot symbol, the method comprising the following steps:

[0013] The signal to be processed obtained by the above-mentioned digital signal optical communication transmission method based on single pilot symbols is received and shaped, and IQ orthogonalization, time-domain equalization, carrier recovery and DDLSM equalization are performed.

[0014] Traverse all possible frame start positions, extract one symbol at a time with a first set number of symbols to obtain candidate sequences, calculate the average amplitude of each candidate sequence, and determine the candidate sequence with the highest average amplitude as the initial pilot symbol sequence;

[0015] A sliding window is constructed according to the first length of the original pilot symbol sequence inserted at the transmitting end. Multiple window symbols of the first length are extracted by sliding symbol by symbol from the starting position of the initial pilot symbol sequence. Each window symbol is compared with the original pilot symbol sequence symbol by symbol to obtain a phase increment sequence. The variance of the phase increment sequence corresponding to each window symbol is calculated. The position of the window symbol with the smallest variance is determined as the frame start position, and frame synchronization is completed.

[0016] The known pilot symbol values ​​at the pilot positions provide accurate initial information for the MLSE algorithm, and the pilot symbols constrain the survival path to provide periodic guidance for MLSE to decode and obtain the target data frame.

[0017] In some embodiments, the known pilot symbol values ​​at the pilot locations provide precise initial information for the MLSE algorithm, including:

[0018] During the initialization phase of MLSE decoding, the initial state of the trellis diagram is determined based on the value of the first known pilot symbol at the start position of the received sequence. The path metric corresponding to this initial state is initialized to 0, while the path metrics of other possible states are initialized to infinity.

[0019] In some embodiments, using pilot symbol constraints to constrain the survival path provides periodic guided decoding for MLSE to obtain the target data frame, including:

[0020] When processing reaches the receiving time corresponding to the pilot symbol, only the surviving path whose grid state is consistent with the known pilot symbol value at that time is retained, while other mismatched paths are discarded.

[0021] On the other hand, the present invention also provides a digital signal optical communication system based on a single pilot symbol, comprising:

[0022] The transmitting end digital signal processing module is used to execute the above-described digital signal optical communication transmission method based on a single pilot symbol to obtain the signal to be transmitted;

[0023] Optical transmitter, including:

[0024] An arbitrary waveform generator is used to sample and output the signal to be transmitted according to a first set sampling rate.

[0025] An external high-power laser is used to generate a continuous wave optical signal of a set wavelength;

[0026] A coherent optical modulator is used to perform IQ modulation on the sampled output signal to be transmitted and the optical signal;

[0027] Single-mode optical fiber is used to transmit the optical signal output by the optical transmitter.

[0028] A coherent optical receiver is used to mix the optical signal received from a single-mode optical fiber with a free-running ECL local oscillator to generate an electrical signal.

[0029] An oscilloscope is used to capture the electrical signal at a second set sampling rate;

[0030] The receiving end digital signal processing module is used to execute the above-described digital signal optical communication receiving method based on a single pilot symbol.

[0031] In some embodiments, the receiver digital signal processing module is further configured to perform bit error rate analysis and evaluate system performance.

[0032] On the other hand, the present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0033] The digital signal optical communication method and system based on single pilot symbols described in this invention involves periodically inserting X / Y polarized pilot symbol pairs at the transmitting end. These pairs are then shaped using orthogonal dual binary encoding super Nyquist signal shaping, merging each pair of pilot symbols into an equivalent single pilot symbol. Phase and amplitude information is preserved and mapped to the outermost ring of the QAM constellation diagram. At the receiving end, the initial pilot symbol sequence is located based on the characteristic of the highest amplitude at the outermost ring of the QAM constellation diagram. A sliding window is constructed within the initial pilot symbol sequence according to the length of the original pilot symbol sequence to extract the sequence. The variance of the phase increment is calculated by comparing this sequence with the original pilot symbol sequence. The frame start position is determined based on the minimum variance. The pilot symbols provide accurate initial information and periodically guide decoding for the MLSE algorithm. This invention optimizes two pilot symbols into a single pilot symbol after FTN shaping at the transmitting end, preserving crucial phase and amplitude information, thus improving spectral utilization while ensuring the realization of frame synchronization, carrier recovery, and decoding guidance functions. The single pilot symbol provides periodic auxiliary information and guides the decision for the back-end maximum likelihood sequence estimation (MLSE) decoding. By periodically calibrating the state transition path and optimizing error propagation correction, the decoding efficiency and accuracy are improved.

[0034] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the specification and drawings.

[0035] Those skilled in the art will understand that the objectives and advantages achievable with this invention are not limited to those specifically described above, and that the above and other objectives achievable with this invention will become clearer from the following detailed description. Attached Figure Description

[0036] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings:

[0037] Figure 1 This is a flowchart illustrating a digital signal optical communication transmission method based on a single pilot symbol according to an embodiment of the present invention.

[0038] Figure 2 This is a flowchart illustrating a digital signal optical communication receiving method based on a single pilot symbol according to another embodiment of the present invention.

[0039] Figure 3 This is a frame data structure and its corresponding constellation diagram in a digital signal optical communication transmission method based on a single pilot symbol according to an embodiment of the present invention, wherein (a) is the frame data structure and (b) is the corresponding constellation diagram.

[0040] Figure 4 A comparison of the bit error rate curves of the proposed MLSE scheme and the traditional MLSE scheme provided in an embodiment of the invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0042] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0043] 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.

[0044] This application aims to address the problems of high complexity, high power consumption, and insufficient spectral efficiency faced by Super Nyquist (FTN) transmission. In FTN transmission systems, traditional Nyquist rate transmission is limited by spectral efficiency, while FTN can improve its utilization. However, in scenarios with high-order modulation formats and symbol rate sampling, signal processing complexity is high and power consumption is large. Existing pilot symbols have low utilization, mainly focusing on phase recovery or synchronization functions, failing to fully utilize their potential in the decoding stage. Furthermore, the computational complexity of traditional MLSE decoding algorithms increases exponentially when processing QDB-16QAM signals. To address this, this application optimizes the paired pilot symbols at the transmitting end into a single pilot symbol, retaining key information and enabling it to provide periodic auxiliary information and guiding decisions for MLSE decoding at the receiving end. This reduces decoding complexity, improves decoding reliability and accuracy, and meets the high bandwidth and low power consumption requirements of power-sensitive environments such as data centers.

[0045] Specifically, this invention provides a digital signal optical communication transmission method based on a single pilot symbol, such as... Figure 1 As shown, the method includes the following steps S101~S102:

[0046] Step S101: In the data frame to be transmitted, a pair of X / Y polarized pilot symbols are inserted periodically at a first set number of data symbols based on a preset pilot symbol sequence.

[0047] Step S102: The data frame is shaped and output using orthogonal double binary encoding. Based on the symmetry of the transformation, each pair of X / Y polarized pilot symbols is merged into an equivalent single pilot symbol. The single pilot symbol retains the phase and amplitude information of each pair of pilot symbols for subsequent decoding guidance. All single pilot symbols are mapped to the outermost constellation points in the QAM constellation diagram.

[0048] Before step S101, that is, before inserting a pair of X / Y polarized pilot symbols at a first predetermined number of data symbols periodic intervals in the data frame to be transmitted based on a preset pilot symbol sequence, the method further includes steps S1011 and S1012:

[0049] Step S1011: Generate a pseudo-random binary sequence and map it to an orthogonal amplitude modulation symbol with a set number of bits.

[0050] Step S1012: Generate pilot symbol sequences for the orthogonal amplitude modulation symbols based on different X / Y polarization seeds and share them with the receiver.

[0051] Pseudo-random binary sequences are used to simulate the random data characteristics in real communication, avoid test deviations caused by periodic signals, and provide a basic data stream for subsequent pilot insertion and FTN coding.

[0052] In some embodiments, the pseudo-random binary sequence is generated using a linear feedback shift register (LFSR), or by generating random numbers using a pseudo-random number generator and then converting them into a binary sequence. The sequence must satisfy the uniform distribution characteristic, ensuring that the probability of 0 / 1 appearing is equal. The PRBS is grouped according to a set number of bits m, such as m=4 in 16QAM, with each 4 bits mapped to one complex symbol, and the mapping rule follows the standard QAM constellation diagram.

[0053] In optical communication, X and Y polarizations are two orthogonal polarization states that can be used to transmit independent data streams, increasing system capacity. Pilot symbols are known reference signals used to assist the receiver in channel estimation, synchronization, and other tasks. Generating pilot symbol sequences based on different polarization seeds means generating specific pilot symbol sequences for X and Y polarizations respectively, enabling the receiver to distinguish and process signals from different polarizations, achieving accurate channel estimation and synchronization, and ensuring reliable data transmission.

[0054] In step S101, in the data frame, a pair of orthogonal polarization pilot symbols are inserted after every first set number (denoted as N) of data symbols. The X-polarization pilot is modulated on the X polarization state of the optical signal, and the Y-polarization pilot is modulated on the Y polarization state.

[0055] In step S102, orthogonal double binary coding (QDB coding) performs super Nyquist signal shaping on the data symbols, introducing controlled inter-symbol interference (ISI). Specifically, a raised cosine filter or a root raised cosine filter is used to introduce controllable ISI, and the roll-off factor is adjusted to control the amount of ISI.

[0056] Orthogonal double binary (QDB) shaping is equivalent to passing through a transfer function. The filter, that is, the current symbol, is obtained by adding the two transmitted symbols. So in the original X or Y polarization, there are two pilot symbols. After QDB, the two pilot symbols are added together to obtain a single pilot symbol. This single pilot symbol carries the information of the original two independent pilot symbols. The receiver regards this as an "equivalent single pilot symbol" for subsequent phase estimation, channel estimation and other tasks.

[0057] On the other hand, the present invention also provides a digital signal optical communication receiving method based on a single pilot symbol, such as... Figure 2 As shown, the method includes the following steps S201~S204:

[0058] Step S201: Receive the signal to be processed obtained by the above-mentioned digital signal optical communication transmission method based on single pilot symbols, and perform IQ orthogonalization, time-domain equalization, carrier recovery and DDLSM equalization.

[0059] Step S202: Traverse all possible frame start positions, extract one symbol at a time with a first set number of symbols as the interval to obtain candidate sequences, calculate the average amplitude of each candidate sequence, and determine the candidate sequence with the highest average amplitude as the initial pilot symbol sequence.

[0060] Step S203: Construct a sliding window according to the first length of the original pilot symbol sequence inserted by the transmitter, slide from the starting position of the initial pilot symbol sequence to extract multiple window symbols of the first length, compare each window symbol with the original pilot symbol sequence symbol by symbol to obtain the phase increment sequence, calculate the variance of the phase increment sequence corresponding to each window symbol, and determine the position of the window symbol with the smallest variance as the frame start position to complete frame synchronization.

[0061] Step S204: Use the known pilot symbol values ​​at the pilot positions to provide accurate initial information for the MLSE algorithm, and use the pilot symbols to constrain the survival path to provide periodic guidance for MLSE decoding to obtain the target data frame.

[0062] In step S201, the signal to be processed can be received using polarization diversity coherent reception. The receiver first receives the FTN (Super Nyquist) signal optimized by a single pilot symbol. Then, IQ orthogonalization is performed to decompose the received signal into in-phase (I) and quadrature (Q) components for easier subsequent processing. Next, time-domain equalization is performed to preliminarily estimate the signal from the transmitter and reduce inter-symbol interference (ISI). After that, carrier recovery is performed to eliminate the effects of frequency offset and phase noise in the received signal. Finally, DDLSM (Decision-Driven Least Mean Square) equalization is performed to further reduce signal distortion.

[0063] In step S202, to extract the pilot sequence, the receiver traverses all possible frame start positions (position 1 to position N). For each candidate start position, two symbols are extracted as candidate sequences according to the same data symbol interval (first predetermined number) as the transmitter. The average amplitude of each candidate sequence is calculated. During the encoding process at the transmitter, the pilot symbols are mapped to the outermost constellation points in the QAM constellation diagram. Therefore, the amplitude of the pilot symbols is usually large and fixed. Thus, the candidate sequence with the highest average amplitude is likely to be the sequence containing pilot symbols, and it is determined as the initial pilot symbol sequence.

[0064] In step S203, a sliding window is constructed according to the first length of the original pilot symbol sequence inserted by the transmitter. Multiple window symbols of the first length are extracted symbol by symbol, starting from the beginning position of the initial pilot symbol sequence. The starting positions of the extracted window symbols differ, and the arrangement order of the pilot symbols may overlap with or not overlap with the original pilot symbol sequence. When they overlap, the difference in phase increments at corresponding positions between the window symbol and the original pilot symbol sequence is minimized. Therefore, finding the position with the smallest phase increment difference determines the start position of the data frame. Thus, each window symbol is compared symbol by symbol with the original pilot symbol sequence to obtain a phase increment sequence, and the variance of the phase increment sequence corresponding to each window symbol is calculated. The position of the window symbol with the smallest variance is most likely the frame start position, thereby achieving precise frame synchronization.

[0065] In step S204, during the initialization phase of MLSE decoding, the initial state of the trellis graph is determined based on the value of the first known pilot symbol at the start position of the received sequence. The path metric corresponding to this initial state is initialized to 0, while the path metrics of other possible states are initialized to infinity. When processing reaches the reception time corresponding to the pilot symbol, only the surviving paths whose trellis state matches the known pilot symbol value at that time are retained, while other mismatched paths are discarded.

[0066] In the receiver-side digital signal optical communication method of the present invention, MLSE decoding is the core processing step, which uses a single pilot symbol to provide accurate initial information and periodically guide the decision, specifically including the following sub-steps:

[0067] MLSE decoding initialization:

[0068] During the initialization phase of MLSE decoding, the initial state of the Viterbi algorithm's trellis graph is determined based on the value of the first known single pilot symbol at the start position of the received sequence. Assuming the first symbol of the received sequence is a known single pilot symbol (e.g., an initial pilot value of -3), the initial state of the trellis graph is set to the state corresponding to that pilot symbol, with its path metric initialized to 0. The path metrics for other non-matching states are set to infinity. This pilot-based initialization ensures that the MLSE algorithm starts from a reliable starting point, reducing the risk of error propagation in the initial phase.

[0069] Periodic guided decoding:

[0070] Single pilot symbols are periodically inserted at intervals of a predetermined number of data symbols (e.g., N=64), providing periodic constraints for MLSE decoding. When processing the reception time corresponding to the pilot symbol, the MLSE algorithm only retains surviving paths whose grid state matches the known single pilot symbol value, discarding other mismatched paths. For example, if the pilot symbol value is -6, only the path corresponding to that state is retained at that time, significantly reducing invalid branches in the grid diagram. This periodic guidance is equivalent to providing "checkpoints" for MLSE decoding, periodically calibrating state transition paths, optimizing error propagation correction, and improving decoding reliability.

[0071] Branch metric calculation and path selection:

[0072] MLSE decoding employs the Viterbi algorithm, which evaluates the likelihood of state transitions by calculating the branch metric (BM). The formula for calculating BM is:

[0073] ;

[0074] in, This represents the transition metric from state i at time k to state j at time k+1. It is the received symbol at time k. and These are the symbols transmitted at time k-1 and time k, respectively. At the pilot symbol positions, BM calculations are based on known pilot values, further simplifying the computational complexity.

[0075] The cumulative metric (AM) determines the survival path by accumulating the BM:

[0076] ;

[0077] Within the tracing period (tracing length D=20), the path with the minimum AM is selected as the decoded output. The periodic constraint of the single pilot symbol reduces the accumulation of invalid paths and optimizes path selection efficiency.

[0078] On the other hand, the present invention also provides a digital signal optical communication system based on a single pilot symbol, such as... Figure 3 As shown, it includes: a transmitting digital signal processing module, an optical transmitter, a single-mode optical fiber, a coherent optical receiver, an oscilloscope, and a receiving digital signal processing module.

[0079] The transmitting end digital signal processing module is used to execute the digital signal optical communication transmission method based on a single pilot symbol described in steps S101 to S102 above to obtain the signal to be transmitted.

[0080] An optical transmitter includes: an arbitrary waveform generator for sampling and outputting the signal to be transmitted according to a first set sampling rate; an external high-power laser for generating a continuous wave optical signal of a set wavelength; and a coherent optical modulator for IQ modulation of the sampled output signal to be transmitted and the optical signal.

[0081] Single-mode optical fiber is used to transmit the optical signal output by the optical transmitter.

[0082] A coherent optical receiver is used to mix the optical signal received from a single-mode optical fiber with a free-running ECL local oscillator to generate an electrical signal.

[0083] An oscilloscope is used to capture the electrical signal at a second set sampling rate.

[0084] The receiving end digital signal processing module is used to execute the digital signal optical communication receiving method based on a single pilot symbol described in steps S201 to S104 above.

[0085] In some embodiments, the receiver digital signal processing module is also used to perform bit error rate analysis and evaluate system performance.

[0086] On the other hand, the present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0087] The present invention will now be described with reference to a specific embodiment:

[0088] The purpose of this embodiment is to overcome the problems of low pilot symbol utilization, high complexity of maximum likelihood sequence estimation (MLSE) decoding, and high power consumption in existing FTN transmission systems, and to provide a communication method and system for FTN signal shaping and MLSE decoding based on single-pilot optimization. This embodiment optimizes two pilot symbols into a single pilot symbol after FTN shaping at the transmitting end, and uses this single pilot symbol to provide periodic auxiliary information and guiding decisions for MLSE decoding at the receiving end, thereby achieving efficient and low-power signal processing to meet the high bandwidth and low power consumption requirements of data centers.

[0089] This embodiment proposes a frame structure based on paired Pilot symbols and a Pilot-assisted MLSE decoding method, which specifically includes the following steps:

[0090] 1. Transmitter Digital Signal Processing

[0091] Digital signal processing at the transmitting end mainly includes signal frame structure design. The specific steps are as follows:

[0092] 1.1 Generate pseudo-random binary sequences (PRBS) and map them to QAM symbols.

[0093] 1.2 Generate pilot symbol sequences based on different X / Y polarization seeds.

[0094] 1.3 Insert paired Pilot symbols into the frame structure. Initially, insert two Pilot symbols every N data symbols.

[0095] 1.4 FTN shaping is achieved through orthogonal double binary (QDB) encoding. After FTN shaping, each pair of pilot symbols is merged into an equivalent single pilot symbol. This single pilot symbol retains the phase and amplitude information of the original pilot pair and is used for subsequent decoding guidance. All pilot symbols are mapped to the outermost constellation points of the QAM constellation diagram.

[0096] 2. Digital signal processing at the receiving end

[0097] The receiver's digital signal processing unit is mainly used for signal reception, recovery, and demodulation. The proposed pilot-assisted scheme includes the following core steps. Specifically, the processing steps for each aspect are as follows:

[0098] 2.1 The receiver performs front-end DSP processing on the M-QAM signal after QDB-FTN shaping, including IQ orthogonalization, time-domain equalization, carrier recovery, and DDLSM equalization.

[0099] 2.2 Extraction of pilot sequences: On the received signal, traverse all possible frame start positions, i.e., positions 1 to N; for each candidate start position, select a symbol sequence of length L as a candidate sequence with a symbol interval of N; calculate the average amplitude of each candidate sequence; identify the candidate sequence with the highest average amplitude as the pilot sequence.

[0100] 2.3 Confirming the frame start position: In the extracted pilot sequence, a sliding window of length L is used, sliding symbol by symbol from the start position; for each position of the sliding window, L symbols within the window are extracted, and these L symbols are compared symbol by symbol with the known transmitted pilot sequence to calculate the phase increment (PIs) sequence between corresponding symbols, and the variance of the PIs sequence is calculated; after the sliding window reaches the end of the received pilot sequence, the variances obtained from all window positions are compared; the start position of the sliding window corresponding to the minimum variance value is determined as the final frame start position, and frame synchronization is completed.

[0101] 2.4 The known symbol value of the pilot position provides accurate initial information for the MLSE algorithm. In the initialization stage of MLSE decoding, the initial state of the trellis is determined based on the first known pilot symbol value at the start position of the received sequence. The path metric corresponding to this initial state is initialized to 0, and the path metrics of other possible states are initialized to infinity.

[0102] 2.5 Utilizing pilot symbols to constrain surviving paths provides periodic guidance for MLSE. During the lattice evolution process of MLSE decoding, when processing reaches the reception time corresponding to the pilot symbol, only surviving paths whose lattice state at that time matches the known pilot symbol value are retained, while other mismatched paths are discarded.

[0103] More specifically, the following embodiments use a 100 Gbaud FTN-16QAM optical communication system as an example to illustrate the implementation process of the communication method based on single-pilot optimization of FTN signal shaping and MLSE decoding, but the application scenarios of the present invention are not limited to this.

[0104] Reference Figure 1 As shown in the figure, this embodiment discloses a digital signal processing architecture for a pilot-assisted coherent optical communication system. The specific implementation steps are as follows:

[0105] S1: At the sending end, a pseudo-random binary sequence (PRBS) is generated and mapped to 16QAM symbols with a symbol rate of 100 Gbaud.

[0106] S2: Pilot Symbol Insertion: According to the frame structure design, a pair of Pilot symbols is inserted every N data symbols for subsequent frame synchronization, carrier recovery, and decoding guidance. The insertion ratio of Pilot symbols is 1 / N to balance spectral efficiency and functional requirements.

[0107] S3: The signal is shaped using FTN through QDB encoding to generate a spectrum-compressed FTN-16QAM signal (also known as 49QAM). QDB encoding introduces controllable inter-symbol interference (ISI), improving spectrum utilization.

[0108] S4: After QDB encoding, the symmetry of the QDB transform is used to merge each pair of Pilot symbols into an equivalent single Pilot symbol. Orthogonal binary (QDB) shaping is equivalent to passing through a transfer function... The filter, that is, the current symbol, is obtained by adding the two transmitted symbols. So in the original X or Y polarization, there are two pilot symbols. After QDB, the two pilot symbols are added together to obtain a single pilot symbol. This single pilot symbol carries the information of the original two independent pilot symbols. The receiver regards this as an "equivalent single pilot symbol" for subsequent phase estimation, channel estimation and other tasks.

[0109] S5: The generated FTN-16QAM signal is output via an arbitrary waveform generator (AWG) at a sampling rate of 100 GSa / s, with a 3dB bandwidth of 45 GHz. The signal is then input to a coherent drive modulator (CDM, 3 dB bandwidth 40 GHz) for electro-optic conversion. An external cavity laser (ECL) provides a continuous-wave optical signal with a center wavelength of 1550 nm and an output power of 15 dBm. The modulated optical signal is amplified by an erbium-doped fiber amplifier (EDFA) and transmitted to the receiver via a 10 km standard single-mode fiber (SSMF).

[0110] S6: The receiver uses a polarization diversity coherent receiver (ICR, 3 dB bandwidth 40 GHz) to mix the optical signal with a free-running ECL local oscillator (LO) to generate an electrical signal. The signal is captured by a real-time oscilloscope (RTO) at a sampling rate of 256 GSa / s and processed by an offline DSP.

[0111] S7: The receiver performs front-end DSP processing on the QDB-FTN shaped M-QAM signal, including IQ orthogonalization, time-domain equalization, carrier recovery, and DDLSM equalization.

[0112] S8: Extract pilot sequences and accurately locate pilot symbols.

[0113] S9: The known sign value of the pilot position provides accurate initial information for the MLSE algorithm;

[0114] In the current embodiment, the branch metric calculation formula for the MLSE algorithm is as follows:

[0115] ;

[0116] in, This represents the transition metric from state i at time k to state j at time k+1. It is the received symbol at time k. and These are the symbols sent at time k-1 and time k, respectively.

[0117] Initialization process: After the receiver completes frame synchronization, it determines that the starting position of the received sequence contains the first single pilot symbol (e.g., corresponding to an initial pilot value of -3). In the Viterbi algorithm for MLSE decoding, the initial state of the mesh graph is set to the state corresponding to -3, and its path metric is initialized to 0. The path metric for other states is set to infinity. The branch metric calculation formula is:

[0118] ;

[0119] Where, when time k is 1, This represents the first received symbol, and −3 represents the known pilot value. This is the assumed value for the next symbol. This initialization method ensures that MLSE starts from the correct state, reducing the risk of initial error propagation.

[0120] Initialization advantages: Compared to traditional MLSE which requires traversing all possible initial states, the known value of a single pilot symbol compresses the initial state space from M^L (L is the memory length; M=4 in 16QAM and 16 when L=2) to 1.

[0121] S10: Utilize pilot symbols to constrain surviving paths, providing periodic guidance for MLSE. During the lattice evolution process of MLSE decoding, when processing reaches the reception time corresponding to the pilot symbol, only the surviving path whose lattice state at that time matches the known pilot symbol value is retained, while other mismatched paths are discarded.

[0122] Periodic guidance mechanism: Single pilot symbols are inserted at intervals of N=64 symbols, with a known reference point provided every 64 symbols. During MLSE mesh evolution, when processing to a pilot symbol position (e.g., symbol 64, 128), only surviving paths matching known pilot values ​​(e.g., +6) are retained. For example, if the pilot value is +6, the mesh state must satisfy:

[0123] ;

[0124] Non-matching paths (such as) The error path is discarded directly. This constraint is equivalent to inserting a "forced node" in the mesh graph, which periodically resets the error path and prevents the accumulation of errors.

[0125] Path selection and output: MLSE uses a tracing length D=20, updates AM at each time step, and selects the path with the smallest AM as the surviving path. The mandatory constraint on pilot symbol positions reduces the accumulation of invalid paths and optimizes tracing efficiency. The final output is the symbol sequence corresponding to the path with the smallest AM.

[0126] Figure 4 A comparison of the bit error rate curves of the proposed MLSE scheme in this embodiment with those of the traditional MLSE scheme shows that, compared with the traditional MLSE, the scheme proposed in this invention improves the sensitivity gain of the receiver by 0.15dB.

[0127] Corresponding to the above method, the present invention also provides an apparatus / system including a computer device, the computer device including a processor and a memory, the memory storing computer instructions, the processor executing the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the apparatus / system performs the steps of the method as described above.

[0128] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. 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.

[0129] In summary, the digital signal optical communication method and system based on single pilot symbols described in this invention periodically inserts X / Y polarized pilot symbol pairs at the transmitting end. These pairs are then shaped using orthogonal dual binary encoding super Nyquist signal shaping, merging each pair of pilot symbols into an equivalent single pilot symbol. This preserves phase and amplitude information and maps the symbol to the outermost ring of the QAM constellation diagram. At the receiving end, the initial pilot symbol sequence is located based on the characteristic of the highest amplitude at the outermost ring of the QAM constellation diagram. A sliding window is constructed within this initial sequence, extracting a sequence according to the length of the original sequence. This sequence is then compared with the original sequence to calculate the variance of the phase increment. The frame start position is determined based on the minimum variance. The pilot symbols provide accurate initial information and periodically guide decoding for the MLSE algorithm. This invention optimizes two pilot symbols into a single pilot symbol after FTN shaping at the transmitting end, preserving crucial phase and amplitude information, thus improving spectral utilization while ensuring the implementation of frame synchronization, carrier recovery, and decoding guidance functions. The single pilot symbol provides periodic auxiliary information and guides the decision for the back-end maximum likelihood sequence estimation (MLSE) decoding. By periodically calibrating the state transition path and optimizing error propagation correction, the decoding efficiency and accuracy are improved.

[0130] 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 invention. 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 invention are programs or code segments used to perform the desired 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 in a carrier wave.

[0131] It should be clarified that the present invention 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 the present invention 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 the present invention.

[0132] In this 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 in place of features of other embodiments.

[0133] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for transmitting digital signals in optical communication based on a single pilot symbol, characterized in that, The method includes the following steps: A pseudo-random binary sequence is generated and mapped to an orthogonal amplitude modulation symbol with a set number of bits; a pilot symbol sequence is generated for the orthogonal amplitude modulation symbol based on different X / Y polarization seeds and shared to the receiver; In the data frame to be transmitted, a pair of X / Y polarized pilot symbols are inserted periodically at a first set number of data symbols based on a preset pilot symbol sequence. The data frame is shaped and output using orthogonal double binary encoding. Based on the symmetry of the transformation, each pair of X / Y polarized pilot symbols is merged into an equivalent single pilot symbol. The single pilot symbol retains the phase and amplitude information of each pair of pilot symbols for subsequent decoding guidance. All single pilot symbols are mapped to the outermost constellation points in the QAM constellation diagram. The known pilot symbol values ​​at the pilot positions provide accurate initial information for the MLSE algorithm. The pilot symbols constrain the survival path to provide periodic guidance for MLSE decoding to obtain the target data frame. The super Nyquist signal shaping of the data frame through orthogonal double binary encoding also includes: introducing controllable inter-symbol interference using a raised cosine filter or a root raised cosine filter, and adjusting the roll-off factor to control the amount of ISI.

2. The digital signal optical communication transmission method based on a single pilot symbol according to claim 1, characterized in that, The pseudo-random binary sequence is generated using a linear feedback shift register, or by generating random numbers using a pseudo-random number generator and then converting them into a binary sequence.

3. A method for receiving digital signals in optical communication based on a single pilot symbol, characterized in that, The method includes the following steps: The system receives the signal to be processed obtained by the digital signal optical communication transmission method based on a single pilot symbol as described in any one of claims 1 to 2, and performs IQ orthogonalization, time-domain equalization, carrier recovery, and DDLSM equalization. Traverse all possible frame start positions, extract one symbol at a time with a first set number of symbols to obtain candidate sequences, calculate the average amplitude of each candidate sequence, and determine the candidate sequence with the highest average amplitude as the initial pilot symbol sequence; A sliding window is constructed according to the first length of the original pilot symbol sequence inserted at the transmitting end. Multiple window symbols of the first length are extracted by sliding symbol by symbol from the starting position of the initial pilot symbol sequence. Each window symbol is compared with the original pilot symbol sequence symbol by symbol to obtain a phase increment sequence. The variance of the phase increment sequence corresponding to each window symbol is calculated. The position of the window symbol with the smallest variance is determined as the frame start position, and frame synchronization is completed. The known pilot symbol values ​​at the pilot positions provide accurate initial information for the MLSE algorithm, and the pilot symbols constrain the survival path to provide periodic guidance for MLSE to decode and obtain the target data frame.

4. The digital signal optical communication receiving method based on a single pilot symbol according to claim 3, characterized in that, The known pilot symbol values ​​at the pilot locations provide precise initial information for the MLSE algorithm, including: During the initialization phase of MLSE decoding, the initial state of the trellis diagram is determined based on the value of the first known pilot symbol at the start position of the received sequence. The path metric corresponding to this initial state is initialized to 0, while the path metrics of other possible states are initialized to infinity.

5. The digital signal optical communication receiving method based on a single pilot symbol according to claim 4, characterized in that, Using pilot symbols to constrain the survival path provides periodic guided decoding for MLSE to obtain the target data frame, including: When processing reaches the receiving time corresponding to the pilot symbol, only the surviving path whose grid state is consistent with the known pilot symbol value at that time is retained, while other mismatched paths are discarded.

6. A digital signal optical communication system based on a single pilot symbol, characterized in that, include: The transmitting end digital signal processing module is used to execute the digital signal optical communication transmission method based on a single pilot symbol as described in any one of claims 1 to 2 to obtain the signal to be transmitted; Optical transmitter, including: An arbitrary waveform generator is used to sample and output the signal to be transmitted according to a first set sampling rate. An external high-power laser is used to generate a continuous wave optical signal of a set wavelength; A coherent optical modulator is used to perform IQ modulation on the sampled output signal to be transmitted and the optical signal; Single-mode optical fiber is used to transmit the optical signal output by the optical transmitter. A coherent optical receiver is used to mix the optical signal received from a single-mode optical fiber with a free-running ECL local oscillator to generate an electrical signal. An oscilloscope is used to capture the electrical signal at a second set sampling rate; The receiving end digital signal processing module is used to execute the digital signal optical communication receiving method based on a single pilot symbol as described in any one of claims 3 to 5.

7. The digital signal optical communication system based on a single pilot symbol according to claim 6, characterized in that, The receiver's digital signal processing module is also used to perform bit error rate analysis and evaluate system performance.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method as described in any one of claims 1 to 5.

Images