WO2022227296A1 - 概率整形pam-4信号传输方法及装置 - Google Patents

概率整形pam-4信号传输方法及装置 Download PDF

Info

Publication number
WO2022227296A1
WO2022227296A1 PCT/CN2021/106996 CN2021106996W WO2022227296A1 WO 2022227296 A1 WO2022227296 A1 WO 2022227296A1 CN 2021106996 W CN2021106996 W CN 2021106996W WO 2022227296 A1 WO2022227296 A1 WO 2022227296A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
pam
bits
probability
processing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/106996
Other languages
English (en)
French (fr)
Inventor
高明义
刘梦丽
沈纲祥
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou University
Original Assignee
Suzhou University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Suzhou University filed Critical Suzhou University
Publication of WO2022227296A1 publication Critical patent/WO2022227296A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/50—Transmitters
    • H04B10/516—Details of coding or modulation
    • H04B10/5161—Combination of different modulation schemes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/50—Transmitters
    • H04B10/516—Details of coding or modulation
    • H04B10/54—Intensity modulation
    • H04B10/541—Digital intensity or amplitude modulation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference

Definitions

  • the invention relates to the technical field of optical communication networks, in particular to a probability shaping PAM-4 signal transmission method and device.
  • PAM Probabilistic Pulse Amplitude Modulation
  • IM/DD intensity-modulation direct-detection
  • PAM-4 four-level pulse amplitude modulation
  • the conventional PS scheme of the coherent optical communication system has high computational complexity, and its complexity is mainly determined by a distributed matcher (DM).
  • DM distributed matcher
  • HiDM Hierarchical Distributed Matcher
  • HiDM has fully parallel input-output interface and pipeline structure, but performs DM/inv-DM at the expense of larger rate loss and higher SNR.
  • the m-out-of-n method utilizes scaling and rounding techniques, so it is difficult to implement in hardware.
  • CCDM constant composition distribution matching
  • Huffman coding has lower implementation complexity, but there are problems of bit rate variation and synchronization. Therefore, PS schemes with low complexity and high performance are more attractive in IM/DD systems.
  • the low complexity condition is met by the cut-and-paste (CAP) method, which divides the transmitted bit sequence into a number of n-symbol groups. For each n-symbol group, after bit-to-symbol mapping, the magnitude bits are extracted and flipped according to a look-up table (LUT), and then the energy of the original magnitude bits and the flipped magnitude bits are calculated and compared to select a larger energy The lower one is transmitted. Finally, label bits are pasted to each n-symbol group to distinguish flip operations [8]. Although different probability distributions can be achieved by changing the symbol length of each group, once the symbol length of each group is chosen, the probability distribution cannot be changed.
  • CAP cut-and-paste
  • the baud rate of a PAM system is always limited by the bandwidth of the available optics and electrical components. Therefore, some improved PAM-4 schemes, such as Faster-than Nyquist (FTN) PAM-4, Partial-response (PR) PAM-4, Constellation switching (CS) )PAM-4, etc. were proposed to adapt to higher baud rate signal transmission in bandwidth-constrained systems [9-12].
  • FTN Faster-than Nyquist
  • PR Partial-response
  • CS Constellation switching
  • FFE feed-forward equalizer
  • DFE decision feedback equalizer
  • THP Tomlinson-Harashima precoding
  • MLSE maximum likelihood bit sequence estimation
  • Volterra equalizer Volterra equalizer
  • the PAM system in the prior art has the problems that the probability distribution cannot be changed after the symbol length is determined, the receiver sensitivity is low, and the communication quality is not ideal.
  • the technical problem to be solved by the present invention is to overcome the problems in the prior art that the probability distribution cannot be changed after the symbol length is determined, the receiver sensitivity is low, and the communication quality is not ideal in the PAM system.
  • the present invention provides a method for performing probability shaping PAM-4 signal transmission by an encoding module, including:
  • the PS-PAM-4 signal is sent to the receiving end, and the receiving end performs channel equalization processing on the PS-PAM-4 signal to obtain a processed output signal, and performs demapping and processing on the output signal. Decoding process to obtain the bit sequence sent by the transmitter.
  • obtaining the amplitude bits with probability distributions of "0" and "1" through energy level assignment includes:
  • bits of high energy level are flipped to the bits of low energy level, and the amplitude bits with the probability distribution of "0" and "1" are obtained.
  • the energy calculation formula of each combination is as follows:
  • x represents PAM-4 symbols and n represents the number of symbols in each combination.
  • the PS-PAM-4 signal is a signal with a variable probability distribution.
  • performing channel equalization processing on the PS-PAM-4 signal by the receiving end includes:
  • the PS-PAM-4 signal is used as the input signal of the trained channel equalization algorithm, and the output signal after channel equalization processing is obtained, wherein the trained channel equalization algorithm is as follows:
  • X(n) is the input signal
  • Y(n) is the output signal
  • M is the number of symbols
  • a k is the tap coefficient
  • the training steps of the channel equalization algorithm are as follows:
  • the present invention also provides a method for performing probability shaping PAM-4 signal transmission by an auxiliary coding module at the receiving end, including:
  • the PS-PAM-4 signal is subjected to channel equalization processing to obtain a processed output signal, wherein the PS-PAM-4 signal is divided by the initial amplitude bit of the bit sequence and the amplitude bit and the symbol bit obtained by the distribution matching process are processed. map get;
  • the present invention also provides a method for performing probability shaping PAM-4 signal transmission by a system, wherein the system includes an encoding module and a receiving end, and the method includes:
  • the distribution matching process includes obtaining amplitude bits with probability distributions of "0" and "1" through energy level assignment;
  • the PS-PAM-4 signal is sent to the receiving end, and the receiving end performs channel equalization processing on the PS-PAM-4 signal to obtain a processed output signal, and performs demapping and processing on the output signal. Decoding processing to obtain the bit sequence sent by the transmitter;
  • the PS-PAM-4 signal is subjected to channel equalization processing to obtain a processed output signal, wherein the PS-PAM-4 signal is divided by the initial amplitude bit of the bit sequence and the amplitude bit and the symbol bit obtained by the distribution matching process are processed. map get;
  • the present invention also provides a coding module, comprising:
  • a sequence dividing unit receiving the bit sequence from the transmitting end, and the sequence dividing unit is used to divide the bit sequence into symbol bits and initial amplitude bits;
  • the probability distribution matching unit is configured to perform distribution matching processing on the initial amplitude bits, wherein the distribution matching processing includes obtaining amplitude bits having probability distributions of "0" and "1" through energy level allocation;
  • mapping unit which is configured to perform mapping processing on the amplitude bits and the sign bits to obtain a PS-PAM-4 signal with probability shaping
  • a sending unit configured to send the PS-PAM-4 signal to a receiving end, and the receiving end performs channel equalization processing on the PS-PAM-4 signal to obtain a processed output signal, and Demap and decode the output signal to obtain the bit sequence sent by the transmitter.
  • the present invention also provides a receiving end, including:
  • the receiving unit is configured to receive the PS-PAM-4 signal from the encoding module
  • a channel equalization unit configured to perform channel equalization processing on the PS-PAM-4 signal to obtain a processed output signal, wherein the PS-PAM-4 signal is divided by the bit sequence and the initial amplitude bits are passed through The amplitude bits obtained by the distribution matching process are obtained by mapping with the sign bits;
  • a demapping unit configured to perform demapping processing on the output signal to obtain a signal after demapping processing
  • a decoding unit configured to perform decoding processing on the demapped signal to obtain a bit sequence sent by the transmitter.
  • the present invention can obtain PS-PAM-4 signal with variable probability distribution through energy level assignment of various combinations of amplitude bits, so that it can provide various probability distributions for any n symbol types, and has constant PS coding Redundancy greatly enhances the flexibility of the system and reduces the use of additional comparators, and uses channel equalization to mitigate inter-symbol interference caused by limited bandwidth. Compared with the traditional PAM-4 signal transmission method, it can Improves receiver sensitivity by up to 4.2dB for ideal communication quality.
  • FIG. 1 is a schematic diagram of the principle of a probability shaping PAM-4 signal transmission method in Embodiment 1 of the present invention.
  • FIG. 2 is a schematic diagram of a DM implementation process in Embodiment 1 of the present invention.
  • FIG. 3 is a schematic diagram of the energy of each combination in FIG. 2 .
  • FIG. 4 is a PS-PAM-4 signal diagram with four different probability distributions (P1-P4) obtained by using the ELA-CAP method based on 3 symbols in the first embodiment of the present invention.
  • FIG. 5 is a GMI graph of a uniform PAM-4 signal and four PS-PAM-4 signals in Embodiment 1 of the present invention.
  • FIG. 6 is a schematic structural diagram of an FFE in Embodiment 1 of the present invention.
  • FIG. 7 is a graph of a mean square error convergence curve in Embodiment 1 of the present invention.
  • FIG. 8 is a schematic diagram of a point-to-point experiment setup for 25G-Baud IM/DD PAM-4 used in the data center internal network in Embodiment 1 of the present invention.
  • FIG. 9 is a BER curve diagram of a uniform PAM-4 signal and four PS-PAM-4 signals in Embodiment 1 of the present invention.
  • FIG. 10 is an eye diagram of a uniform PAM-4 signal and four PS-PAM-4 signals passing through FFE at -20 dBm in Embodiment 1 of the present invention.
  • ELA-CAP Cut and Paste of Energy Level Assignment
  • DM Distribution Matcher
  • PS Probability Shaping
  • PAM Pulse Amplitude Modulation
  • CCDM Arithmetic Coded Constant Component Distribution Matching
  • FFE Feed Forward Equalizer
  • PRBS Pseudo Random binary sequence
  • WDM wavelength division multiplexing
  • AWG arbitrary waveform generator
  • EDFA fiber amplifier
  • VOA variable optical attenuator
  • SSMF single-mode fiber
  • BTB back-to-back
  • MZM single-drive modulator.
  • a method for transmitting a probability shaping PAM-4 signal performed by an encoding module in this embodiment includes the following steps:
  • S100 Receive the bit sequence from the transmitter, and divide the bit sequence into sign bits and initial amplitude bits.
  • S200 Perform distribution matching processing on the initial amplitude bits, and obtain amplitude bits with probability distributions of "0" and "1" through energy level allocation.
  • obtaining the amplitude bits with the probability distribution of "0" and “1” through energy level allocation includes the following steps: S210: Perform PAM-4 symbol mapping on the initial amplitude bits, arrange and combine them, and calculate the energy of each combination , the energy calculation formula is In the formula, x represents the PAM-4 symbol, and n represents the number of symbols in each combination; S220: Define different energy levels according to the energy of each combination, and calculate the energy level probability; S230: Calculate according to the probability of different energy levels Probability of "0" and "1” in the probability distribution; S240: Flip the bits of high energy level to bits of low energy level to obtain the amplitude bits with probability distributions of "0" and "1".
  • S300 Perform a mapping process on the amplitude bits and the sign bits to obtain a probability-shaped PS-PAM-4 signal, where the PS-PAM-4 signal is a signal with variable probability distribution.
  • S400 Send the PS-PAM-4 signal to the receiving end, and the receiving end performs channel equalization processing on the PS-PAM-4 signal to obtain a processed output signal, and performs demapping and decoding processing on the output signal to obtain the transmitting end The transmitted bit sequence.
  • performing channel equalization processing on the PS-PAM-4 signal by the receiving end includes using the PS-PAM-4 signal as an input signal of the trained channel equalization algorithm, and obtaining an output signal after channel equalization processing, wherein the trained signal is obtained.
  • the channel equalization algorithm is as follows:
  • X(n) is the input signal
  • Y(n) is the output signal
  • M is the number of symbols
  • a k is the tap coefficient
  • the training content of the channel equalization algorithm includes: acquiring training symbols, using the training symbols as the input value of the channel equalization algorithm, and solving the error between the output value and the set reference value; using the normalized least mean square algorithm to calculate the mean square error and adjust the tap coefficient; and then through iterative training, the tap coefficient of the optimal weight is obtained.
  • the uniform bit sequence u is divided into sign bits and amplitude bits.
  • sign bits are not allowed to enter the DM to keep them evenly distributed.
  • the DM inverse operation is implemented using the tag bits generated at the transmitting end. For the tag bits, "0" means invert the operation, and "1" means do nothing. Since 1 bit is added every n symbols, the code rate of DM can be expressed as:
  • m represents the magnitude of bits per PAM symbol.
  • x represents PAM-4 symbols and n represents the number of symbols in each combination.
  • n represents the number of symbols in each combination.
  • Figure 4 shows the energy for each combination.
  • the uniform 3-symbol combination has 4 energy levels, namely E1, E2, E3 and E4, and its probabilities can be calculated as 1/8, 3/8, 3/8 and 1/8 respectively according to Fig. 2, as shown in Fig. 2 shown in the first row, where E1>E2>E3>E4.
  • FIG 4 shows PS-PAM-4 signals with four different probability distributions (P1-P4) obtained by the 3-symbol based ELA-CAP method.
  • P1-P4 probability distributions
  • the P1 and P4 distributions can achieve 1.15-dB and 0.7-dB shaping gains, respectively.
  • the received signal y k can be written as:
  • y k represents the k-th received signal
  • x k represents the k-th transmitted signal
  • a g is the ambiguity function determined by the channel impulse response
  • ⁇ represents the compression factor of FTN transmission
  • T is the sampling time interval
  • p is the generation The number of ISI future and past symbols.
  • pre-cursor ISI and post-cursor ISI should be considered in the design of the channel equalizer.
  • This embodiment adopts the FFE of the T symbol interval for channel equalization to eliminate pre-cursor ISI and post-cursor ISI. Its structure diagram is shown in Figure 6, and the mathematical expression of the channel equalization algorithm is as follows:
  • X(n) is the input signal
  • Y(n) is the output signal
  • M is the number of symbols that cause pre-cursor ISI and post-cursor ISI respectively
  • a k is the tap coefficient.
  • X(n) is the input signal
  • Y(n) is the output signal
  • M is the number of symbols that cause pre-cursor ISI and post-cursor ISI respectively
  • a k is the tap coefficient.
  • the above-mentioned energy level assignment to various combinations of amplitude bits can obtain PS-PAM-4 signals with variable probability distributions, which can provide various probability distributions for any n symbol types, and have constant PS coding redundancy.
  • the redundancy greatly enhances the flexibility of the system and reduces the use of additional comparators, and uses channel equalization processing to alleviate the inter-symbol interference caused by limited bandwidth.
  • Figure 8 shows the point-to-point experimental setup of 25G-Baud IM/DD PAM-4 for the data center internal network.
  • the PRBS is firstly transmitted into the PS encoder, and a probability-shaped PS-PAM-4 signal is obtained through bit-to-symbol mapping.
  • the symbol sequence is then sampled and pulse-shaped by applying a root raised cosine finite impulse response filter with a roll-off factor of 0.4 to mitigate the performance degradation caused by the limited bandwidth at the signal transmitter.
  • the filtered and S21 compensated data is loaded into the AWG.
  • a PN sequence is added at the transmitting end for signal synchronization.
  • the PS-PAM-4 electrical signal from the AWG was modulated into a 1550.112 nm continuous wave laser by MZM.
  • the output power of the modulated PS-PAM-4 optical signal is about 5.7dBm.
  • VOA and EDFA are used to control the noise level of the BER measurement.
  • Another VOA is used to control the optical signal power entering the 10-GHz photodetector.
  • the received electrical signal is sampled by the RTO and processed by a synchronization algorithm.
  • the discrete digital signal is then resampled to 1 sps after passing through a matched RRC FIR filter.
  • the PS-PAM-4 signal is recovered using the channel equalization algorithm.
  • the demapping and PS decoding operations are implemented, and the BER is calculated.
  • Figure 9 shows the BER curves for a uniform PAM-4 signal and four PS-PAM-4 signals.
  • the 25G-Baud PAM-4 signal was transmitted in the experiment, and the information entropy of the uniform signal and PS signal (P1-P4) were 2, 1.9486, 1.9176, 1.8741 and 1.812bit/symbol, respectively. Therefore, the transmission bit rates are 50, 48.7149, 47.9390, 46.7853 and 45.3132bit/s, respectively. It is evident from Fig. 8 that the loss caused by the 2km SSMF transmission is negligible compared to the BTB case, as shown by the solid and dashed lines.
  • the received optical powers are -20dBm, -21.2dBm, -22dBm, -22.6dBm and -24.2dBm, respectively. Therefore, the PS-PAM-4 signal with the P1-P4 probability distribution can improve the receiver sensitivity by 1.2dB, 2dB, 2.6dB and 4.2dB, respectively, compared with the uniform PAM-4 signal.
  • the eye diagrams of the two through FFE at -20dBm are shown in Figure 10. Without FFE, severe ISI effects would result in eye closure. It can be seen from the figure that by reducing the probability of high-amplitude signals, the degree of eye opening can be enlarged, thereby increasing the margin of error caused by additional noise, for example, PS-PAM-4 signal with probability distribution P4 has a larger margin quantity.
  • a probability shaping PAM-4 signal transmission method of the present invention mainly includes three execution bodies, and the three execution bodies are respectively a transmitter, an encoding module and a receiver. Therefore, the present invention does not describe the execution subject of the transmitting end in more detail here.
  • the present invention describes the specific content of the encoding module, the receiving end, and the system formed by the encoding module and the receiving end as the execution body respectively, wherein the specific content of the encoding module as the execution body has been described in detail above, The present invention will not be described in detail here.
  • This embodiment provides a method for performing probability shaping PAM-4 signal transmission by an auxiliary coding module at the receiving end, including:
  • the PS-PAM-4 signal is subjected to channel equalization processing to obtain a processed output signal, wherein the PS-PAM-4 signal is obtained by mapping the amplitude bits and the symbol bits obtained by the initial amplitude bits divided by the bit sequence through distribution matching processing;
  • This embodiment provides a method for performing probability shaping PAM-4 signal transmission by a system.
  • the system includes an encoding module and a receiving end, and the method includes:
  • the distribution matching processing includes obtaining amplitude bits with probability distributions of "0" and "1" through energy level assignment;
  • the PS-PAM-4 signal is sent to the receiving end, and the receiving end performs channel equalization processing on the PS-PAM-4 signal to obtain the processed output signal, and performs demapping and decoding processing on the output signal to obtain the signal sent by the transmitting end. bit sequence;
  • the PS-PAM-4 signal is subjected to channel equalization processing to obtain a processed output signal, wherein the PS-PAM-4 signal is obtained by mapping the amplitude bits and the symbol bits obtained by the initial amplitude bits divided by the bit sequence through distribution matching processing;
  • This embodiment provides an encoding module, and the encoding module includes:
  • a sequence dividing unit receiving the bit sequence from the transmitting end, and the sequence dividing unit is used to divide the bit sequence into symbol bits and initial amplitude bits;
  • the probability distribution matching unit is configured to perform distribution matching processing on the initial amplitude bits, wherein the distribution matching processing includes obtaining amplitude bits with probability distributions of "0" and "1" through energy level allocation;
  • mapping unit is used to perform mapping processing on the amplitude bits and the sign bits to obtain a PS-PAM-4 signal of probability shaping
  • a sending unit the sending unit is used to send the PS-PAM-4 signal to the receiving end, and the receiving end performs channel equalization processing on the PS-PAM-4 signal, obtains the processed output signal, and performs a channel equalization process on the PS-PAM-4 signal.
  • the output signal is de-mapped and decoded to obtain the bit sequence sent by the transmitter.
  • This embodiment provides a receiving end, and the receiving end includes:
  • the receiving unit is used to receive the PS-PAM-4 signal from the encoding module
  • the channel equalization unit is used to perform channel equalization processing on the PS-PAM-4 signal, and obtain a processed output signal, wherein the PS-PAM-4 signal is divided by the bit sequence of the initial amplitude bit The amplitude obtained by the distribution matching process The bit and the sign bit are mapped to obtain;
  • demapping unit configured to perform demapping processing on the output signal to obtain a signal after demapping processing
  • the decoding unit is configured to perform decoding processing on the demapped signal to obtain the bit sequence sent by the transmitting end.
  • the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions
  • the apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Power Engineering (AREA)
  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
  • Dc Digital Transmission (AREA)

Abstract

本发明涉及一种概率整形PAM-4信号传输方法，包括接收比特序列，将比特序列划分为符号比特和初始幅度比特；对初始幅度比特进行分布匹配处理，通过能级分配获得具有"0"和"1"概率分布的幅度比特；对幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；将PS-PAM-4信号发送至接收端，由接收端对PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对输出信号进行解映射和解码处理，获得比特序列。本发明能够为任何n个符号类型提供多种概率分布，并具有恒定的PS编码冗余度，大大增强了系统的灵活性并减少了附加比较器的使用，而且与传统PAM-4信号传输方法相比，其可以将接收器灵敏度提高多达4.2dB，通信质量理想。

Description

概率整形PAM-4信号传输方法及装置 技术领域
本发明涉及光通信网络技术领域，尤其是指一种概率整形PAM-4信号传输方法及装置。
背景技术
概率整形(Probabilistic shaped，PS)脉冲幅度调制(Pulse amplitude modulation，PAM)由于其优越的性能而在数据中心内部网络中有广泛的应用前景。其中强度调制直接检测(Intensity-modulation direct-detection，IM/DD)四级脉冲幅度调制(PAM-4)因具有更简单的结构和更低的成本能耗而优于其他方案。
为了进一步提高PAM系统的性能，概率整形可以通过球形地限制信号空间中的调制级别来放宽信噪比要求，从而降低平均信号能量。但是，相干光通信系统的常规PS方案具有较高的计算复杂度，其复杂度主要由分布匹配器(Distributed matcher，DM)决定。其中分层分布匹配器(Hierarchical distributed matcher，HiDM)具有完全并行的输入-输出接口和流水线结构，却以较大的速率损失和较高的SNR为代价执行DM/inv-DM。m-out-of-n方法利用缩放和舍入技术,因此在硬件实施方面较为困难。另外基于算术编码的恒定成分分布匹配(Constant composition distribution matching，CCDM)在码长无限制的条件下具有零速率损失的特性，但由于串行编码而在解码时会引起一连串的错误。而基于霍夫曼编码的DM具有较低的实现复杂性，却存在比特率变化和同步问题。因此，具有低复杂性和高性能的PS方案在IM/DD系统中更具吸引力。
剪切和粘贴(cut-and-paste，CAP)方法满足了低复杂度这一条件，该方法将发送比特序列切分为许多个n符号组。对于每个n符号组，经过比特到符号映射后，提取幅度位并根据查找表(Look up table，LUT)进行翻转，然后计算并比较原始幅度位和翻转后幅度位的能量，以选择能量较低的一个进行传输。最后，将标签位粘贴到每个n符号组以区分翻转操作[8]。虽然通过改变每个组的符号长度可以实现不同的概率分布，但是一旦选择了每个组的符号长度，则概率分布将不可更改。
此外，PAM系统的波特率始终受可用光学器件和电气组件的带宽限制。因此，一些改进的PAM-4方案，例如更快的奈奎斯特(Faster-than Nyquist，FTN)PAM-4，部分响应(Partial-response，PR)PAM-4，星座切换(Constellation switching，CS)PAM-4等被提出来以适应更高的波特率的信号在带宽受限的系统中传输[9-12]。这时，各种能减轻带宽限制带来的符号间干扰(Inter-symbol interference，ISI)的均衡技术是必不可少的，例如前馈均衡器(Feed-forward equalizer，FFE)、判决反馈均衡器(Decision feedback equalizer，DFE)、Tomlinson-Harashima预编码(Tomlinson-Harashima precoding，THP)、最大似然比特序列估计(Maximum likelihood sequence estimation，MLSE)和沃尔泰拉均衡器(Volterra equalizer，VE)等，但是接收器的灵敏度依然存在较低的问题，导致通信质量不理想。
综上所述，现有技术的PAM系统存在符号长度确定后概率分布不可更改、接收器灵敏度低以及通信质量不理想的问题。
发明内容
为此，本发明所要解决的技术问题在于克服现有技术中PAM系统存在符号长度确定后概率分布不可更改、接收器灵敏度低以及通信质量不理想的问题。
为解决上述技术问题，本发明提供一种由编码模块执行概率整形PAM-4信号传输方法，包括：
接收来自发射端的比特序列，将所述比特序列划分为符号比特和初始幅度比特；
对所述初始幅度比特进行分布匹配处理，通过能级分配获得具有“0”和“1”概率分布的幅度比特；
对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列。
在本发明的一个实施例中，所述通过能级分配获得具有“0”和“1”概率分布的幅度比特包括：
对所述初始幅度比特进行PAM-4符号映射并对其进行排列组合，计算每个组合的能量；
根据每个组合的能量定义不同的能级，并计算其能级概率；
根据不同能级的概率计算概率分布中“0”和“1”的概率；
将高能级的比特翻转为低能级的比特，获得具有“0”和“1”概率分布的幅度比特。
在本发明的一个实施例中，每个组合的能量计算公式如下：
Figure PCTCN2021106996-appb-000001
式中，x表示PAM-4符号，n表示每个组合中的符号数。
在本发明的一个实施例中，所述PS-PAM-4信号为具有可变概率分布的信号。
在本发明的一个实施例中，由所述接收端对所述PS-PAM-4信号进行信道均衡处理包括：
将所述PS-PAM-4信号作为训练好的信道均衡算法的输入信号，得到经过信道均衡处理后的输出信号，其中训练好的信道均衡算法如下：
Figure PCTCN2021106996-appb-000002
式中，X(n)为输入信号，Y(n)为输出信号，M为符号个数，a k为抽头系数。
在本发明的一个实施例中，所述信道均衡算法的训练步骤如下：
获取训练符号，将所述训练符号作为所述信道均衡算法的输入值，求解其输出值与设定的参考值的误差；
使用归一化最小均方算法计算均方误差并调整抽头系数；
再经过迭代训练得到最优权值的抽头系数。
并且，本发明还提供一种由接收端辅助编码模块执行概率整形PAM-4信号传输方法，包括：
接收来自编码模块的PS-PAM-4信号；
将所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中所述PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
对所述输出信号进行解映射处理，得到解映射处理后的信号；
对所述解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
此外，本发明还提供一种由系统执行概率整形PAM-4信号传输方法，所述系统包括编码模块和接收端，方法包括：
由编码模块执行以下步骤：
接收来自发射端的比特序列，将所述比特序列划分为符号比特和初始幅度比特；
对所述初始幅度比特进行分布匹配处理，其中分布匹配处理包括通过能级分配获得具有“0”和“1”概率分布的幅度比特；
对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列；
由接收端执行以下步骤：
接收来自编码模块的PS-PAM-4信号；
将所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中所述PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
对所述输出信号进行解映射处理，得到解映射处理后的信号；
对所述解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
此外，本发明还提供一种编码模块，包括：
序列划分单元，接收来自发射端的比特序列，所述序列划分单元用于将所述比特序列划分为符号比特和初始幅度比特；
概率分布匹配单元，所述概率分布匹配单元用于对所述初始幅度比特进行分布匹配处理，其中分布匹配处理包括通过能级分配获得具有“0”和“1”概率分布的幅度比特；
映射单元，所述映射单元用于对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
发送单元，所述发送单元用于将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列。
此外，本发明还提供一种接收端，包括：
接收单元，所述接收单元用于接收来自编码模块的PS-PAM-4信号；
信道均衡单元，所述信道均衡单元用于将所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中所述PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
解映射单元，所述解映射单元用于对所述输出信号进行解映射处理，得到解映射处理后的信号；
解码单元，所述解码单元用于对所述解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
本发明的上述技术方案相比现有技术具有以下优点：
本发明通过对幅度比特的各种组合的能级分配，可以获得具有可变概率分布的PS-PAM-4信号，如此能够为任何n个符号类型提供多种概率分布，并具有恒定的PS编码冗余度，大大增强了系统的灵活性并减少了附加比较器的使用，而且采用信道均衡处理来减轻由于带宽有限而引起的符号间干扰，与传统PAM-4信号传输方法相比，其可以将接收器灵敏度提高多达4.2dB，通信质量理想。
附图说明
为了使本发明的内容更容易被清楚的理解，下面根据本发明的具体实施例并结合附图，对本发明作进一步详细的说明，其中
图1是本发明实施例一中概率整形PAM-4信号传输方法的原理示意图。
图2是本发明实施例一中DM实现过程的示意图。
图3是图2中每种组合的能量示意图。
图4是本发明实施例一中基于3符号采用ELA-CAP方法获得的具有四个不同概率分布(P1-P4)的PS-PAM-4信号图。
图5是本发明实施例一中均匀PAM-4信号和四个PS-PAM-4信号的GMI曲线图。
图6是本发明实施例一中FFE的结构示意图。
图7是本发明实施例一中均方误差收敛曲线图。
图8是本发明实施例一中用于数据中心内部网络的25G-Baud IM/DD PAM-4的点对点实验设置示意图。
图9是本发明实施例一中均匀PAM-4信号和四个PS-PAM-4信号的BER曲线图。
图10是本发明实施例一中均匀PAM-4信号和四个PS-PAM-4信号在-20dBm时经过FFE的眼图。
具体实施方式
下面结合附图和具体实施例对本发明作进一步说明，以使本领域的技术人员可以更好地理解本发明并能予以实施，但所举实施例不作为对本发明的限定。
首先本发明对下文中需要出现的英文标记做出如下释义：
ELA-CAP：能级分配的剪切和粘贴；DM：分布匹配器；PS：概率整形；PAM：脉冲幅度调制；CCDM：算术编码的恒定成分分布匹配；FFE：前馈均衡器；PRBS：伪随机二进制序列；WDM：波分复用；AWG：任意波形发生器；EDFA：光纤放大器；VOA：可变光衰减器；SSMF：单模光纤；BTB：背靠背；MZM：单驱动调制器。
实施例一
下面对本发明实施例一提供的一种由编码模块执行概率整形PAM-4信号传输方法进行详细的阐述。
请参阅图1所示，本实施例一种由编码模块执行概率整形PAM-4信号传输方法，包括以下步骤：
S100：接收来自发射端的比特序列，将比特序列划分为符号比特和初始幅度比特。
S200：对初始幅度比特进行分布匹配处理，通过能级分配获得具有“0”和“1”概率分布的幅度比特。
示例地，通过能级分配获得具有“0”和“1”概率分布的幅度比特包括以下步骤：S210：对初始幅度比特进行PAM-4符号映射并对其进行排列组合，计算每个组合的能量，能量计算公式为
Figure PCTCN2021106996-appb-000003
式中，x表示PAM-4符号，n表示每个组合中的符号数；S220：根据每个组合的能量定义不同的能级，并计算其能级概率；S230：根据不同能级的概率计算概率分布中“0”和“1” 的概率；S240：将高能级的比特翻转为低能级的比特，获得具有“0”和“1”概率分布的幅度比特。
S300：对幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号，其中PS-PAM-4信号为具有可变概率分布的信号。
S400：将PS-PAM-4信号发送至接收端，由接收端对PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对输出信号进行解映射和解码处理，获得发射端发送的比特序列。
示例地，由接收端对PS-PAM-4信号进行信道均衡处理包括将PS-PAM-4信号作为训练好的信道均衡算法的输入信号，得到经过信道均衡处理后的输出信号，其中训练好的信道均衡算法如下：
Figure PCTCN2021106996-appb-000004
式中，X(n)为输入信号，Y(n)为输出信号，M为符号个数，a k为抽头系数。
示例地，信道均衡算法的训练内容包括：获取训练符号，将训练符号作为信道均衡算法的输入值，求解其输出值与设定的参考值的误差；使用归一化最小均方算法计算均方误差并调整抽头系数；再经过迭代训练得到最优权值的抽头系数。
为了更加详细的阐述ELA-CAP方法实现DM的内容，请继续参阅图1所示，首先将均匀的比特序列u分为符号比特和幅度比特。为了实现双边MB分配，不允许符号比特进入DM以保持其均匀分布。在进行幅度比特和符号比特多路复用、符号映射、信道传输和符号解映射之后，使用在发射端生成的标签位来实现DM逆运算。对于标签位，“0”表示反转操作，“1”表示不执行任何操作。由于每n个符号增加1个比特，则DM的码率可表示为：
Figure PCTCN2021106996-appb-000005
式中，m表示每个PAM符号的幅度位数。
以3符号编码为例进行举例说明。请参阅图2所示，使用格雷码进行PAM-4符号映射。对于符号比特，“0”和“1”分别表示“-”和“+”。对于初始幅度比特，“0”和“1”分别对应于“3”和“1”。一个PAM-4符号由一个符号比特和一个初始幅度比特组成，即3个PAM-4符号具有3个初始幅度比特。其中3个初始幅度比特具有2 3种不同的排列组合，如图2的第一行所示。每个组合的能量计算方式如下：
Figure PCTCN2021106996-appb-000006
式中，x表示PAM-4符号，n表示每个组合中的符号数。例如，“000”表示幅度为“3”的3符号序列，其能量为3 2+3 2+3 2＝27。图4展示了每种组合的能量。
均匀的3符号组合具有4个能级，即E1、E2、E3和E4，根据图2可以计算其概率分别为1/8、3/8、3/8和1/8，如图2中的第一行所示，其中E1>E2>E3>E4。
由于输入比特是均匀的，所以可以根据不同能级的不同概率计算P1-P4分布中的“0”和“1”的概率，计算公式如下：
Figure PCTCN2021106996-appb-000007
从上述公式中可以看出，降低高能量等级的概率可以降低传输序列的能量，实现方式也极为简单，只需将高能级的比特转换为低能级的比特。如此通过能级分配实现了如图2所示的四种概率分布(P1-P4)。当然还可以适用于n(n>3)个符号，且可以使用更低的冗余度实现更多的概率分布。
图4显示了基于3符号的ELA-CAP方法获得的具有四个不同概率分布 (P1-P4)的PS-PAM-4信号。经过AWGN通道具有上述P1-P4分布，均匀分布的PAM-4信号的GMI曲线和基于CCDM的PS分布的PAM-4信号的GMI曲线如图5所示，其中PS分布具有相同的冗余度(R DM＝3/4)。如图5所示，以均匀分布和基于CCDM的PS分布的交点作为参考(GMI＝1.682bit/symbol)，P1和P4分布分别可以实现1.15-dB和0.7-dB的整形增益。
同时，由于传输装置的带宽限制，接收信号y k可以写为：
Figure PCTCN2021106996-appb-000008
式中，y k表示第k个接收信号，x k表示第k个发送信号，A g是由信道脉冲响应决定的模糊函数，τ代表FTN传输的压缩因子，T为采样时间间隔，p为产生ISI的未来符号与过去符号的个数。
因此，在信道均衡器的设计中应同时考虑pre-cursor ISI和post-cursor ISI。
本实施例采用用于信道均衡的T符号间隔的FFE，用以消除pre-cursor ISI和post-cursor ISI，其结构图如图6所示，信道均衡算法的数学表达式如下：
Figure PCTCN2021106996-appb-000009
其中，X(n)为输入信号，Y(n)为输出信号，M为分别引起pre-cursor ISI和post-cursor ISI的符号个数，a k为抽头系数。在训练实验中有32768个符号，1500个符号用于训练，其余用于测试。首先，训练符号经过FFE之后，其输出值与设定的参考信号作差来获取误差，然后使用归一化最小均方算法计算均方误差并调整抽头系数，再经过迭代训练得到最优权值。最后，测试信号经过具有最优权值的FFE之后得到输出信号。其中均方误差收敛曲线如图7所示，从图中可以看到，仅4次迭代即可达到收敛。
上述通过对幅度比特的各种组合的能级分配，可以获得具有可变概率分布的PS-PAM-4信号，如此能够为任何n个符号类型提供多种概率分布，并具有恒定的PS编码冗余度，大大增强了系统的灵活性并减少了附加比较器的使用，而且采用信道均衡处理来减轻由于带宽有限而引起的符号间干扰，与传统PAM-4信号传输方法相比，其可以将接收器灵敏度提高多达4.2dB，通信质量理想。
为了进一步验证本实施例的有益效果，图8展示了用于数据中心内部网络的25G-Baud IM/DD PAM-4的点对点实验设置。在发射端，首先将PRBS发射到PS编码器中，并通过位到符号映射获得概率整形的PS-PAM-4信号。然后对符号序列进行采样并应用滚降系数为0.4的根升余弦有限脉冲响应滤波器进行脉冲整形以减轻信号发射端带宽受限导致的性能下降。经过滤波和S21补偿的数据被加载到AWG中。同时，在发射端添加一个PN序列以进行信号同步。之后，通过MZM将来自AWG的PS-PAM-4电信号调制为1550.112nm的连续波激光器。调制后的PS-PAM-4光信号的输出功率约为5.7dBm。在经过2km的SSMF传输后，使用VOA和EDFA控制BER测量的噪声水平。另一个VOA用于控制进入10-GHz光电探测器的光信号功率。接收到的电信号由RTO采样后由同步算法处理。然后，离散数字信号通过匹配的RRC FIR滤波器后重新采样为1sps。接着，利用信道均衡算法恢复PS-PAM-4信号。最后，实现解映射和PS解码操作，并计算BER。
图9显示了均匀PAM-4信号和四个PS-PAM-4信号的BER曲线。实验中传输了25G-Baud PAM-4信号，并且均匀信号和PS信号(P1-P4)的信息熵分别为2、1.9486、1.9176、1.8741和1.812bit/symbol。因此，传输比特率分别为50、48.7149、47.9390、46.7853和45.3132bit/s。从图8可以明显看出，与BTB情况相比，2km SSMF传输造成的损失可忽略不计，如实线和虚线所示。在硬判决前向纠错(HD-FEC)阈值3.8×10 -3时，un-PAM-4、P1-PAM-4、 P2-PAM-4、P3-PAM-4和P4-PAM-4信号的接收光功率分别为-20dBm、-21.2dBm、-22dBm-22.6dBm和-24.2dBm。因此，与均匀PAM-4信号相比，具有P1-P4概率分布的PS-PAM-4信号可以分别将接收器的灵敏度提高1.2dB、2dB、2.6dB和4.2dB。两者在-20dBm时经过FFE的眼图如图10所示。假如没有FFE，严重的ISI效应将导致眼睛闭合。从图中可以看出，通过降低高振幅信号的概率，可以扩大眼睛的睁开程度，从而增加附加噪声引起误差的余量，例如概率分布为P4的PS-PAM-4信号具有更大的余量。
本发明一种概率整形PAM-4信号传输方法中主要有三个执行主体，三个执行主体分别是发射端、编码模块和接收端，其中发射端的执行步骤较为简单，其只是作为发射信号的主体，因此本发明在这里不对发射端这一执行主体做较为详细的阐述。
下面本发明对编码模块、接收端以及由编码模块和接收端构成的系统分别作为执行主体的具体内容进行阐述，其中由编码模块作为执行主体的具体内容已经在上文做出了详细的阐述，本发明在这里不做赘述。
实施例二
本实施例提供一种由接收端辅助编码模块执行概率整形PAM-4信号传输方法，包括：
接收来自编码模块的PS-PAM-4信号；
将PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
对输出信号进行解映射处理，得到解映射处理后的信号；
对解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
其具体内容已经在实施例一中做出了详尽的阐述，本发明在这里不做赘述。
实施例三
本实施例提供了一种由系统执行概率整形PAM-4信号传输方法，系统包括编码模块和接收端，方法包括：
由编码模块执行以下步骤：
接收来自发射端的比特序列，将比特序列划分为符号比特和初始幅度比特；
对初始幅度比特进行分布匹配处理，其中分布匹配处理包括通过能级分配获得具有“0”和“1”概率分布的幅度比特；
对幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
将PS-PAM-4信号发送至接收端，由接收端对PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对输出信号进行解映射和解码处理，获得发射端发送的比特序列；
由接收端执行以下步骤：
接收来自编码模块的PS-PAM-4信号；
将PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
对输出信号进行解映射处理，得到解映射处理后的信号；
对解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
其具体内容已经在实施例一中做出了详尽的阐述，本发明在这里不做赘述。
下面对本发明提供的一种概率整形PAM-4信号传输装置进行详细的阐述。
实施例四
本实施例提供了一种编码模块，编码模块包括：
序列划分单元，接收来自发射端的比特序列，序列划分单元用于将比特序列划分为符号比特和初始幅度比特；
概率分布匹配单元，概率分布匹配单元用于对初始幅度比特进行分布匹配处理，其中分布匹配处理包括通过能级分配获得具有“0”和“1”概率分布的幅度比特；
映射单元，映射单元用于对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
发送单元，发送单元用于将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列。
其具体内容已经在实施例一中做出了详尽的阐述，本发明在这里不做赘述。
实施例五
本实施例提供了一种接收端，接收端包括：
接收单元，接收单元用于接收来自编码模块的PS-PAM-4信号；
信道均衡单元，信道均衡单元用于将PS-PAM-4信号进行信道均衡处理， 得到处理后的输出信号，其中PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
解映射单元，解映射单元用于对所述输出信号进行解映射处理，得到解映射处理后的信号；
解码单元，解码单元用于对解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
其具体内容已经在实施例一中做出了详尽的阐述，本发明在这里不做赘述。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程 或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，上述实施例仅仅是为清楚地说明所作的举例，并非对实施方式的限定。对于所属领域的普通技术人员来说，在上述说明的基础上还可以做出其它不同形式变化或变动。这里无需也无法对所有的实施方式予以穷举。而由此所引申出的显而易见的变化或变动仍处于本发明创造的保护范围之中。

Claims (10)

  1. 一种由编码模块执行概率整形PAM-4信号传输方法，其特征在于，包括：
    接收来自发射端的比特序列，将所述比特序列划分为符号比特和初始幅度比特；
    对所述初始幅度比特进行分布匹配处理，通过能级分配获得具有“0”和“1”概率分布的幅度比特；
    对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
    将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列。
  2. 根据权利要求1所述的由编码模块执行概率整形PAM-4信号传输方法，其特征在于：所述通过能级分配获得具有“0”和“1”概率分布的幅度比特包括：
    对所述初始幅度比特进行PAM-4符号映射并对其进行排列组合，计算每个组合的能量；
    根据每个组合的能量定义不同的能级，并计算其能级概率；
    根据不同能级的概率计算概率分布中“0”和“1”的概率；
    将高能级的比特翻转为低能级的比特，获得具有“0”和“1”概率分布的幅度比特。
  3. 根据权利要求2所述的由编码模块执行概率整形PAM-4信号传输方法，其特征在于：每个组合的能量计算公式如下：
    Figure PCTCN2021106996-appb-100001
    式中，x表示PAM-4符号，n表示每个组合中的符号数。
  4. 根据权利要求1所述的由编码模块执行概率整形PAM-4信号传输方法，其特征在于：所述PS-PAM-4信号为具有可变概率分布的信号。
  5. 根据权利要求1所述的由编码模块执行概率整形PAM-4信号传输方法，其特征在于：由所述接收端对所述PS-PAM-4信号进行信道均衡处理包括：
    将所述PS-PAM-4信号作为训练好的信道均衡算法的输入信号，得到经过信道均衡处理后的输出信号，其中训练好的信道均衡算法如下：
    Figure PCTCN2021106996-appb-100002
    式中，X(n)为输入信号，Y(n)为输出信号，M为符号个数，a k为抽头系数。
  6. 根据权利要求5所述的由编码模块执行概率整形PAM-4信号传输方法，其特征在于：所述信道均衡算法的训练步骤如下：
    获取训练符号，将所述训练符号作为所述信道均衡算法的输入值，求解其输出值与设定的参考值的误差；
    使用归一化最小均方算法计算均方误差并调整抽头系数；
    再经过迭代训练得到最优权值的抽头系数。
  7. 一种由接收端辅助编码模块执行概率整形PAM-4信号传输方法，其特征在于，包括：
    接收来自编码模块的PS-PAM-4信号；
    将所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中所述PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
    对所述输出信号进行解映射处理，得到解映射处理后的信号；
    对所述解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
  8. 一种由系统执行概率整形PAM-4信号传输方法，其特征在于，所述系统包括编码模块和接收端，方法包括：
    由编码模块执行以下步骤：
    接收来自发射端的比特序列，将所述比特序列划分为符号比特和初始幅度比特；
    对所述初始幅度比特进行分布匹配处理，其中分布匹配处理包括通过能级分配获得具有“0”和“1”概率分布的幅度比特；
    对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
    将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列；
    由接收端执行以下步骤：
    接收来自编码模块的PS-PAM-4信号；
    将所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中 所述PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
    对所述输出信号进行解映射处理，得到解映射处理后的信号；
    对所述解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
  9. 一种编码模块，其特征在于，包括：
    序列划分单元，接收来自发射端的比特序列，所述序列划分单元用于将所述比特序列划分为符号比特和初始幅度比特；
    概率分布匹配单元，所述概率分布匹配单元用于对所述初始幅度比特进行分布匹配处理，其中分布匹配处理包括通过能级分配获得具有“0”和“1”概率分布的幅度比特；
    映射单元，所述映射单元用于对所述幅度比特和符号比特进行映射处理，获得概率整形的PS-PAM-4信号；
    发送单元，所述发送单元用于将所述PS-PAM-4信号发送至接收端，由所述接收端对所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，并对所述输出信号进行解映射和解码处理，获得发射端发送的比特序列。
  10. 一种接收端，其特征在于，包括：
    接收单元，所述接收单元用于接收来自编码模块的PS-PAM-4信号；
    信道均衡单元，所述信道均衡单元用于将所述PS-PAM-4信号进行信道均衡处理，得到处理后的输出信号，其中所述PS-PAM-4信号由比特序列划分的初始幅度比特经过分布匹配处理得到的幅度比特与符号比特进行映射获得；
    解映射单元，所述解映射单元用于对所述输出信号进行解映射处理，得到解映射处理后的信号；
    解码单元，所述解码单元用于对所述解映射处理后的信号进行解码处理，获得发射端发送的比特序列。
PCT/CN2021/106996 2021-04-27 2021-07-19 概率整形pam-4信号传输方法及装置 Ceased WO2022227296A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110461754.7 2021-04-27
CN202110461754.7A CN113193920B (zh) 2021-04-27 2021-04-27 概率整形pam-4信号传输方法及装置

Publications (1)

Publication Number Publication Date
WO2022227296A1 true WO2022227296A1 (zh) 2022-11-03

Family

ID=76979624

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/106996 Ceased WO2022227296A1 (zh) 2021-04-27 2021-07-19 概率整形pam-4信号传输方法及装置

Country Status (2)

Country Link
CN (1) CN113193920B (zh)
WO (1) WO2022227296A1 (zh)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113938209B (zh) * 2021-09-03 2022-10-28 华中科技大学 一种针对概率整形的脉冲幅度调制信号的整形方法及系统
CN114024818A (zh) * 2021-11-04 2022-02-08 北京邮电大学 基于幅度平移的概率整形四维qam调制方法及系统
CN114938259B (zh) * 2022-04-13 2023-04-25 苏州大学 概率整形pam-8信号短距离传输方法及系统
CN114980143B (zh) * 2022-05-05 2023-06-16 华南理工大学 一种基于概率成形与ftn传输技术的概率因子寻优方法
EP4559101A4 (en) * 2022-07-20 2026-04-01 Qualcomm Inc GREY MATCHING FOR VARIABLE TO FIXED DISTRIBUTION ADAPTATION
CN115441954B (zh) * 2022-08-04 2024-06-04 武汉邮电科学研究院有限公司 一种概率整形的脉冲幅度调制方法、装置及系统
US20250286757A1 (en) * 2022-09-01 2025-09-11 Qualcomm Incorporated Probabilistic amplitude shaping including energy selection, composition determination, and amplitude symbol sequence determination
EP4566247A4 (en) * 2022-10-10 2025-11-12 Huawei Tech Co Ltd RECEIVER DEVICE FOR PULSE AMPLITUDE MODULATION SIGNALS
CN116346239B (zh) * 2022-12-06 2024-09-06 苏州大学 基于概率整形高阶qam相干光通信系统的矩形星座编码方法
CN115964906B (zh) * 2023-03-17 2023-06-02 巨霖科技(上海)有限公司 基于多边沿响应的pam3差分端口统计眼图仿真方法和装置
CN115987404B (zh) * 2023-03-20 2023-06-02 北京理工大学 基于预编码和概率整形联合优化的光子射频信号生成方法
CN116683995A (zh) * 2023-07-05 2023-09-01 南京新基讯通信技术有限公司 基于霍夫曼编码的概率整形八级脉冲幅度调制方法及系统
CN118764138B (zh) * 2024-09-06 2025-01-21 鹏城实验室 传输信号的概率整形方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10523480B1 (en) * 2018-11-08 2019-12-31 Nxp B.V. K-bit enumerative sphere shaping of multidimensional constellations
CN111049586A (zh) * 2019-12-25 2020-04-21 南京信息工程大学 一种基于降幅式概率成形的脉冲幅度位置调制系统
CN111181652A (zh) * 2019-12-26 2020-05-19 华侨大学 一种基于比特加权分布匹配的ps-pam4系统
US10742472B1 (en) * 2019-09-06 2020-08-11 Qualcomm Incorporated Probabilistic amplitude shaping
US20200280468A1 (en) * 2019-02-28 2020-09-03 Nokia Solutions And Networks Oy Probabilistic signal shaping using multiple codebooks

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106160873B (zh) * 2016-06-16 2018-12-11 复旦大学 一种基于脉冲幅度调制系统的相移编码方法和实现装置
CN110278031A (zh) * 2019-06-10 2019-09-24 华侨大学 基于多对一映射ps-pam4的直接探测光纤通信系统及通信方法
CN111628789B (zh) * 2020-05-26 2023-03-28 复旦大学 一种基于混合概率整形和几何整形的信号生成方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10523480B1 (en) * 2018-11-08 2019-12-31 Nxp B.V. K-bit enumerative sphere shaping of multidimensional constellations
US20200280468A1 (en) * 2019-02-28 2020-09-03 Nokia Solutions And Networks Oy Probabilistic signal shaping using multiple codebooks
US10742472B1 (en) * 2019-09-06 2020-08-11 Qualcomm Incorporated Probabilistic amplitude shaping
CN111049586A (zh) * 2019-12-25 2020-04-21 南京信息工程大学 一种基于降幅式概率成形的脉冲幅度位置调制系统
CN111181652A (zh) * 2019-12-26 2020-05-19 华侨大学 一种基于比特加权分布匹配的ps-pam4系统

Also Published As

Publication number Publication date
CN113193920A (zh) 2021-07-30
CN113193920B (zh) 2022-12-30

Similar Documents

Publication Publication Date Title
CN113193920B (zh) 概率整形pam-4信号传输方法及装置
Jana et al. Pre-equalized faster-than-Nyquist transmission
CN109328451B (zh) 用于预编码超奈奎斯特信令的系统和方法
US11128500B1 (en) Method and apparatus for a lookup table-based coding mechanism for communication systems
US11223504B2 (en) System and methods for multi-level signal transmission
CN114938259B (zh) 概率整形pam-8信号短距离传输方法及系统
CN110061761B (zh) 信号均衡方法及装置、光接收机
CN113452638B (zh) 一种正交啁啾复用光传输方法和系统
CN115173950B (zh) 数据中心光互连系统及方法
Du et al. 90-m/660-Mbps underwater wireless optical communication enabled by interleaved single-carrier FDM scheme combined with sparse weight-initiated DNN equalizer
Zou et al. Colored noise suppressed FTN vs PS in peak power constrained IM-DD PAM systems
Qin et al. Recurrent neural network based joint equalization and decoding method for trellis coded modulated optical communication system
Huo et al. Modified DDFTN algorithm for band-limited short-reach optical interconnects
US12143250B2 (en) Optical interconnect system and method for data center
Schmalen et al. Recent advances on machine learning-aided dsp for short-reach and long-haul optical communications
CN107105354B (zh) 一种无源光网络设备以及用于无源光网络设备的方法
WO2020095916A1 (ja) 光伝送システム
CN106453194A (zh) 一种非正交多维无载波幅度相位调制技术传输信号的方法
Liu et al. Multi-distributed probabilistically shaped PAM-4 system for intra-data-center networks
Kuschnerov et al. Advances in deep learning for digital signal processing in coherent optical modems
Mustafa et al. Four single-sideband M-QAM modulation using soft input soft output equalizer over OFDM
WO2026011810A1 (zh) 一种用于光传输系统的非线性均衡方法及装置
Chen et al. Computationally efficient pre-distortion based on adaptive partitioning neural network in underwater visible light communication
Jana et al. Multi-dimensional probabilistic shaping for optical superchannels
US11581944B2 (en) Optical transmission system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21938763

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21938763

Country of ref document: EP

Kind code of ref document: A1