WO2021238976A1 - 概率整形编码装置、系统及方法 - Google Patents

概率整形编码装置、系统及方法 Download PDF

Info

Publication number
WO2021238976A1
WO2021238976A1 PCT/CN2021/096031 CN2021096031W WO2021238976A1 WO 2021238976 A1 WO2021238976 A1 WO 2021238976A1 CN 2021096031 W CN2021096031 W CN 2021096031W WO 2021238976 A1 WO2021238976 A1 WO 2021238976A1
Authority
WO
WIPO (PCT)
Prior art keywords
probability interval
bitstream
data
unit
symbol
Prior art date
Application number
PCT/CN2021/096031
Other languages
English (en)
French (fr)
Inventor
马博伦
刘衡祁
缪众林
彭俊峰
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2021238976A1 publication Critical patent/WO2021238976A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0002—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0014—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the source coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056—Systems characterized by the type of code used
    • H04L1/0057—Block codes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/32—Carrier systems characterised by combinations of two or more of the types covered by groups H04L27/02, H04L27/10, H04L27/18 or H04L27/26
    • H04L27/34—Amplitude- and phase-modulated carrier systems, e.g. quadrature-amplitude modulated carrier systems

Definitions

  • This application relates to the field of optical communication technology.
  • nonlinear interference mainly depends on the size and symbol distribution of the input signal optical power, which limits the maximum signal-to-noise ratio that can be transmitted.
  • An aspect of the embodiments of the present application provides a probabilistic shaping encoding device, including: a plurality of data buffers, a data scheduling module, and a probabilistic shaping encoding module; any data buffer of the plurality of data buffers is configured to buffer input Data block; the data scheduling module is configured to determine the target buffer according to the set priority rules, and schedule the data in the target buffer; the probability shaping module is configured to obtain the bit stream probability interval and the symbol probability interval according to the target buffer, and According to the bit stream probability interval and the symbol probability interval, the currently scheduled data is probabilistically shaped and coded; wherein the number of multiple data buffers is determined by the pulse period required for coding by the probability-shaped coding module.
  • a probabilistic shaping coding system which includes at least one probabilistic shaping coding device provided in an embodiment of the present application; the at least one probabilistic shaping coding device works in parallel.
  • Another aspect of the embodiments of the present application provides a probabilistic shaping coding method, including: dividing the input data into blocks according to transmission requirements and probability distribution parameters to obtain multiple data blocks; and inputting the multiple data blocks into multiple buffers Perform caching; determine the target cache according to the set priority rules; among them, the target cache is the currently scheduled cache; obtain the bitstream probability interval and the symbol probability interval according to the target cache; and, according to the bitstream probability interval and the symbol probability interval, compare the current The scheduled data is probabilistically shaped and encoded.
  • FIG. 1 is a schematic structural diagram of a probabilistic shaping encoding device provided by an embodiment of the present application.
  • Fig. 2 is a schematic structural diagram of a probabilistic shaping coding device provided by an embodiment of the present application.
  • FIG. 3 is a schematic diagram of an update process of the number of symbol probability intervals provided by an embodiment of the present application.
  • Fig. 4 is a schematic structural diagram of a probabilistic shaping coding device provided by an embodiment of the present application.
  • Fig. 5 is a schematic diagram of probabilistic shaping coding provided by an embodiment of the present application.
  • Fig. 6 is a schematic structural diagram of a probabilistic shaping coding system provided by an embodiment of the present application.
  • Fig. 7 is a flow chart of a probabilistic shaping coding method provided by an embodiment of the present application.
  • the method of related technologies to achieve different equivalent spectral efficiency is to combine two quadrature amplitude modulation (Quadrature Amplitude Modulation, QAM) to obtain an equivalent quadrature amplitude modulation signal.
  • QAM Quadrature Amplitude Modulation
  • This technology is also called time-domain hybrid modulation ( Time domain hybrid modulation, TDHM).
  • TDHM is still based on the combination of ordinary QAM, and the performance is between the two combined QAMs, which is equivalent to exchanging the speed for the transmission distance, and it cannot narrow the gap with the Shannon limit.
  • Probabilistic shaping technology is used to change the probability of symbols appearing in different positions of the constellation, so that the appearance frequency of the outer circle constellation points is reduced, which is conducive to reducing the average power, which is equivalent to increasing the minimum Euclidean distance, and thus has better transmission performance.
  • the probabilistic shaping encoding technology of the related technology uses an inherent rate to encode data, which is difficult to meet the requirements of large throughput data.
  • the embodiments of the present application provide a probabilistic shaping coding device, system, and method, which implement probabilistic shaping coding during data transmission and can increase the coding rate.
  • FIG. 1 is a schematic diagram of a structure of a probabilistic shaping encoding device provided by an embodiment of the application. As shown in FIG. 1, the device may include: a plurality of data buffers 110, a data scheduling module 120, and a probability shaping coding module 130.
  • any one of the multiple data buffers 110 can be configured to buffer the input data block; the data scheduling module 120 can be configured to determine the target buffer according to a set priority rule, and schedule the data in the target buffer Data; the probability shaping module 130 can be configured to obtain the bit stream probability interval and the symbol probability interval according to the target buffer, and perform probability shaping coding on the currently scheduled data according to the bit stream probability interval and the symbol probability interval; wherein, multiple data buffers
  • the number of converters 110 is determined by the pulse period required by the probability shaping encoding module 130 when encoding.
  • the input data is first divided into blocks to obtain multiple data blocks, and then the multiple data blocks are respectively cached in the buffer.
  • the data buffer 110 adopts a buffer to avoid that when the subsequent probability shaping coding module 130 is occupied, the previous-level data can still be input without being blocked.
  • the probability shaping encoding module 130 adopts a divider to perform encoding, which usually requires multiple clock cycles to complete encoding. Assuming that P clock cycles are required, P data buffers 110 are configured.
  • the storage depth of each data buffer 110 can be set to the size of the data block. Or set to the number of consecutive probability shaping K and the time consumed by the divider to complete one division encoding, that is, P clock cycles, then the storage depth of the data buffer 110 is K*P.
  • the priority setting rule may be that the smaller the amount of scheduled data in the buffer, the higher the priority.
  • the data buffer 110 may be sorted, and the higher the sort, the higher the priority of the buffer.
  • the data scheduling module 120 may include a first scheduling unit and a second scheduling unit.
  • the first scheduling unit may be configured to determine the buffer with the smallest amount of scheduled data in each buffer as the target buffer; the second scheduling unit may be configured to set the target buffer when there are multiple target buffers. The buffer with the lowest ranking among the buffers is determined as the buffer that is finally scheduled.
  • FIG. 2 is a schematic structural diagram of a probabilistic shaping encoding device provided by an embodiment of this application.
  • the probability shaping coding module 130 may include: a first bitstream probability interval generating unit 131, a symbol probability interval generating unit 132, a comparing unit 133, a second bitstream probability interval generating unit 134, and a bitstream probability interval update Unit 135 and symbol probability interval update unit 136.
  • the first bitstream probability interval generating unit 131 is respectively connected to the data scheduling module 120 and the comparing unit 133, and is configured to initialize the bitstream probability interval according to the scheduled data, and send the initialized bitstream probability interval to the comparing unit 133.
  • the symbol probability interval generation unit 132 is connected to the comparison unit 133, and is configured to initialize the symbol probability interval and send the initialized symbol probability interval to the comparison unit 133.
  • the comparing unit 133 is respectively connected to the second bitstream probability interval generating unit 134, the bitstream probability interval updating unit 135, and the symbol probability interval updating unit 136; the comparing unit 133 is configured to compare the bitstream probability interval and the symbol probability interval, if the bitstream If the probability interval is not a subset of the symbol probability interval, the first signal is generated, and the first signal is sent to the second bitstream probability interval generating unit 134; if the bitstream probability interval is a subset of the symbol probability interval, the second signal is generated Signal and output data, and send the second signal to the bit stream probability interval update unit 135 and the symbol probability interval update unit 136 to output the output data.
  • the second bitstream probability interval generating unit 134 is also connected to the data scheduling module 120, and is configured to update the bitstream probability interval according to the scheduled data when the first signal is received, and send the updated bitstream probability interval To the comparison unit 133.
  • the bitstream probability interval updating unit 135 is configured to update the bitstream probability interval updating unit 135 when receiving the second signal.
  • the symbol probability interval update unit 136 is configured to update the symbol probability interval when the second signal is received.
  • the first signal can be understood as a "non-signal”
  • the second signal can be understood as a "signal”.
  • the input bit data is "0" or "1”
  • the first bit stream probability interval generating unit 131 initializes the bit stream probability interval according to the value of the bit data, that is, the probability corresponding to the value of the bit data Select an interval in the interval.
  • the second bit stream probability interval generating unit 134 updates the bit stream probability interval according to the value of the bit data, that is, reselects an interval from the probability interval corresponding to the value of the bit data.
  • the bitstream probability interval update unit 135 may include a first divider and a second divider.
  • the first divider is configured to update the upper limit of the probability interval of the bitstream; the second divider is configured to update the lower limit of the probability interval of the bitstream.
  • the upper and lower limits of the bit stream probability interval are updated according to the upper and lower limits of the bit stream probability interval and the upper and lower limits of the symbol probability interval before the update.
  • the symbol probability interval update unit 136 may use a divider to update the symbol probability interval.
  • the symbol probability interval update unit 136 can be a lookup table update unit; the lookup table update unit is configured to generate a lookup table address according to the symbol probability interval before update, and look up the updated table address according to the lookup table address.
  • Symbol probability interval That is, the symbol probability interval update unit 136 adopts the hardware design of LUT (Look-Up-Table), and for multiple probability shaping modules in parallel, the LUT table can be reused, thereby reducing resource consumption.
  • LUT Look-Up-Table
  • FIG. 4 is a schematic structural diagram of a probabilistic shaping encoding device provided in an embodiment of the present application.
  • the probability shaping coding module 130 may also include a plurality of bitstream probability interval buffers 137, a plurality of symbol probability interval buffers 138, and a probability interval scheduling unit 139; the number and symbols of the bitstream probability interval buffers 137
  • the number of probability interval buffers 138 is the same as the number of data buffers 110, and the bit stream probability interval buffer 137, the symbol probability interval buffer 138, and the data buffer 110 correspond one-to-one.
  • the multiple bitstream probability interval buffers 137 and the multiple symbol probability interval buffers 138 are all connected to the probability interval scheduling unit 139; the bitstream probability interval buffer 137 is configured to buffer the updated bitstream probability interval; the symbol probability interval buffer The processor 138 is configured to buffer the updated symbol probability interval; the probability interval scheduling unit 139 is configured to schedule the bitstream probability interval to the bitstream probability interval according to the set priority rule when the comparing unit 133 generates multiple second signals The update unit 135 updates, and schedules the symbol probability interval to the symbol probability interval update unit 136 for update.
  • the second signal carries information of the data buffer 110, and the priority setting rule may be that the smaller the amount of data that has been scheduled in the buffer, the higher the priority of the buffer.
  • the data buffer 110 may be sorted, and the higher the sort, the higher the priority of the buffer.
  • FIG. 5 is a schematic diagram of probabilistic shaping coding provided by an embodiment of the present application.
  • the divider in this embodiment realizes the utilization of the maximum speed limit, and prevents any stage of the divider from being idle.
  • pipeline encoding between data blocks and serial encoding within data blocks are used.
  • the device encodes data blocks in parallel, thereby increasing the encoding rate.
  • the probabilistic shaping encoding device may further include: double buffering.
  • the double buffer is configured to store output data.
  • double buffering can use ping pong buffer. Configure an output ping pong buffer for each channel.
  • the pong buffer can be used for data encoding and buffering. Perform data ping and pong output according to this method.
  • the probabilistic shaping coding device may include multiple data buffers, a data scheduling module, and a probabilistic shaping coding module; any one of the multiple buffers is configured to buffer the input data block; the data scheduling module is It is configured to determine the target buffer according to the set priority rules and schedule the data in the target buffer; the probability shaping module is configured to obtain the bit stream probability interval and the symbol probability interval according to the target buffer, and according to the bit stream probability interval and symbol The probability interval performs probabilistic shaping coding on the currently scheduled data; among them, the number of buffers is determined by the pulse period required for encoding by the probability shaping coding module.
  • Buffer the input data into multiple buffers use the data scheduling module to schedule the data in each buffer according to the set priority, and then the probability shaping encoding module will encode it.
  • the input data can be encoded in parallel at the same time. Probability shaping coding in the process of data transmission, and can increase the coding rate.
  • Fig. 6 is a schematic structural diagram of a probabilistic shaping coding system provided by an embodiment of the present application. As shown in Fig. 6, the system includes at least one probability shaping coding device described above; and at least one probability shaping coding device works in parallel.
  • ping pong buffer output For each probability shaping coding device, two register groups with a bit width of n are configured as ping pong buffer output. Adopting n sets of ping pong buffer output can ensure that the output parallelism of the probabilistic shaping coding device meets the data flow demand (that is, the demand of L). Adopting this structure output realizes the coding requirements of arbitrary parallelism structure under large throughput.
  • Fig. 7 is a flow chart of a probabilistic shaping coding method provided by an embodiment of the present application. Based on the above-mentioned implementable manner, the method may include the following steps S210 to S250.
  • the input data is divided into blocks according to transmission requirements and probability distribution parameters to obtain multiple data blocks.
  • transmission requirements can include bit rate and frequency efficiency.
  • the method of dividing the input data into blocks according to transmission requirements and probability distribution parameters may be: determining the amount of data contained in each data block according to the transmission requirements and probability distribution parameters; and dividing the input data according to the amount of data contained in each data block The data is divided into blocks.
  • the amount of data contained in the data block may be 128 bits.
  • the number of buffers is determined by the pulse period required for probabilistic shaping coding. For example, assuming that P pulse periods are required to complete a probability shaping encoding, the number of buffers is P.
  • the storage depth of the buffer can be set to the size of the data block.
  • the data blocks are respectively input into multiple buffers for buffering.
  • the target cache is determined according to the set priority rule.
  • the target cache is the currently scheduled cache.
  • the priority rule can be that the amount of scheduled data in the buffer is the smallest, and the priority of the buffer is the highest. Or, the buffers are sorted, and the higher the sorting, the higher the priority of the buffers.
  • the method of determining the currently scheduled cache according to the principle of setting priority may be: obtaining the amount of scheduled data of each cache; determining the cache with the smallest amount of scheduled data as the target cache .
  • the cache with the lowest order in the target cache is determined as the cache that is finally scheduled.
  • bit stream probability interval and the symbol probability interval are obtained according to the target buffer.
  • the method of obtaining the bit stream probability interval and the symbol probability interval according to the target buffer may be the bit stream probability interval and the symbol probability interval after the last data in the target buffer is encoded by probability shaping.
  • the updated bitstream probability interval and symbol probability interval are stored in a register, and the bitstream probability interval and symbol probability interval can be correspondingly obtained according to the information in the target buffer.
  • the currently scheduled data is probabilistically shaped and encoded according to the bit stream probability interval and the symbol probability interval.
  • the method of probabilistic shaping and encoding the currently scheduled data can be: judging whether the bitstream probability interval is a subset of the symbol probability interval; if so, according to the bitstream probability interval and the symbol probability interval pair
  • the currently scheduled data is encoded with probability shaping; if not, the bitstream probability interval is updated according to the currently scheduled data, and the operation of judging whether the bitstream probability interval is a subset of the symbol probability interval is performed until the bitstream probability interval Is a subset of the symbol probability interval.
  • the method further includes the following step: recombining the shaped and encoded data according to the buffer channel, And output the recombined data to the lower-level module.
  • the input data can be divided into blocks according to transmission requirements and probability distribution parameters to obtain multiple data blocks; the multiple data blocks are input into multiple buffers for buffering;
  • the priority rule determines the target cache; the target cache is the currently scheduled cache; the bitstream probability interval and symbol probability interval are obtained according to the target cache; the current scheduled data is probabilistically shaped and encoded according to the bitstream probability interval and symbol probability interval ; In this way, the probability shaping coding in the data transmission process can be realized, and the coding rate can be improved.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

本申请涉及光通信技术领域，并提供一种概率整形编码装置、系统及方法。概率整形编码装置包括：多个数据缓存器、数据调度模块及概率整形编码模块。多个数据缓存器中的任一数据缓存器被配置为缓存输入的数据块；数据调度模块被配置为按照设定优先级规则确定目标缓存器，并调度目标缓存器中的数据；概率整形模块被配置为根据目标缓存器获取比特流概率区间和符号概率区间，并根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码。

Description

概率整形编码装置、系统及方法 技术领域
本申请涉及光通信技术领域。
背景技术
在光纤线路中，影响传输系统的最主要因素是非线性干扰，而非线性干扰主要取决于输入信号光功率的大小和符号分布，从而也就限定了可传输的最大信噪比。
发明内容
本申请实施例的一个方面提供一种概率整形编码装置，包括：多个数据缓存器、数据调度模块及概率整形编码模块；多个数据缓存器中的任一数据缓存器被配置为缓存输入的数据块；数据调度模块被配置为按照设定优先级规则确定目标缓存器，并调度目标缓存器中的数据；概率整形模块被配置为根据目标缓存器获取比特流概率区间和符号概率区间，并根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码；其中，多个数据缓存器的数量由概率整形编码模块编码时所需的脉冲周期确定。
本申请实施例的另一方面提供一种概率整形编码系统，包括至少一个本申请实施例提供的概率整形编码装置；该至少一个概率整形编码装置并行工作。
本申请实施例的再一方面提供一种概率整形编码方法，包括：根据传输需求及概率分布参数对输入的数据进行分块，获得多个数据块；将多个数据块输入多个缓存器中进行缓存；按照设定优先级规则确定目标缓存；其中，目标缓存为当前被调度的缓存；根据目标缓存获取比特流概率区间和符号概率区间；以及，根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码。
附图说明
图1是本申请实施例提供的概率整形编码装置的一种结构示意图。
图2是本申请实施例提供的概率整形编码装置的一种结构示意图。
图3是本申请实施例提供的符号概率区间的数量的更新过程的一种示意图。
图4是本申请实施例提供的概率整形编码装置的一种结构示意图。
图5是本申请实施例提供的概率整形编码的一种示意图。
图6是本申请实施例提供的概率整形编码系统的一种结构示意图。
图7是本申请实施例提供的概率整形编码方法的一种流程图。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
在后续的描述中，使用用于表示元件的诸如“模块”、“部件”或“单元”的后缀仅为了有利于本发明的说明，其本身没有特有的意义。因此，“模块”、“部件”或“单元”可以混合地使用。
为使本申请的目的、技术方案和优点更加清楚明白，下文中将结合附图对本申请的实施例进行详细说明。需要说明的是，在不冲突的情况下，本申请实施例中的可实施方式中的特征可以相互任意组合。
相关技术实现不同等效频谱效率的方法是将两种正交振幅调制(Quadrature Amplitude Modulation，QAM)进行组合，得到一个等效正交振幅调制信号，这种技术也称之为时域混合调制(Time domain hybrid modulation，TDHM)。但是TDHM仍然基于普通QAM的组合，性能介于组合的两种QAM之间，相当于用速率换取了传输距离，并不能缩小与香农极限的差距。
采用概率整形技术改变星座图不同位置符号出现的概率，让外圈星座点出现频率降低，有利于减小平均功率，相当于增加了最小欧 氏距离，从而有更好的传输性能。相关技术的概率整形编码技术采用固有的速率对数据进行编码，难以满足大吞吐量数据的要求。
本申请实施例提供一种概率整形编码装置、系统及方法，实现对数据传输过程中的概率整形编码，可以提高编码的速率。
图1为本申请实施例提供的概率整形编码装置的一种结构示意图。如图1所示，该装置可包括：多个数据缓存器110、数据调度模块120及概率整形编码模块130。
多个数据缓存器110中的任一个数据缓存器110可被配置为缓存输入的数据块；数据调度模块120可被配置为按照设定优先级规则确定目标缓存器，并调度目标缓存器中的数据；概率整形模块130可被配置为根据目标缓存器获取比特流概率区间和符号概率区间，并根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码；其中，多个数据缓存器110的数量由概率整形编码模块130编码时所需的脉冲周期确定。
在一种可实施方式中，首先将输入数据分块，获得多个数据块，然后将多个数据块分别缓存至缓存器中。数据缓存器110采用buffer，避免后续的概率整形编码模块130被占用时，保持前级数据仍然能够输入，而不被阻塞。概率整形编码模块130采用除法器进行编码通常需要消耗多个时钟周期才能完成编码，假设需要P个时钟周期，则配置P个数据缓存器110。每一个数据缓存器110的存储深度可以设置为数据块的大小。或者设置为连续概率整形次数K与除法器完成一次除法编码消耗的时长，即P个时钟周期，则数据缓存器110的存储深度为K*P。
在一种可实施方式中，设定优先级规则可以是缓存器中已被调度的数据量越小，优先级越高。在一种可实施方式中，可对数据缓存器110进行排序，排序越靠前的缓存器的优先级越高。
在一种可实施方式中，数据调度模块120可包括第一调度单元和第二调度单元。其中，第一调度单元可被配置为将各个缓存器中已被调度的数据量最小的缓存器确定为目标缓存器；第二调度单元可被配置为当目标缓存器为多个时，将目标缓存器中排序最低的缓存器确 定为最终被调度的缓存。
在一种可实施方式中，图2为本申请实施例提供的概率整形编码装置的一种结构示意图。如图2所示，概率整形编码模块130可包括：第一比特流概率区间生成单元131、符号概率区间生成单元132、比较单元133、第二比特流概率区间生成单元134、比特流概率区间更新单元135和符号概率区间更新单元136。
第一比特流概率区间生成单元131分别与数据调度模块120及比较单元133相连，被配置为根据被调度的数据初始化比特流概率区间，并将初始化的比特流概率区间发送至比较单元133。
符号概率区间生成单元132与比较单元133相连，被配置为初始化符号概率区间，并将初始化的符号概率区间发送至比较单元133。
比较单元133分别与第二比特流概率区间生成单元134、比特流概率区间更新单元135及符号概率区间更新单元136相连；比较单元133被配置为比较比特流概率区间和符号概率区间，若比特流概率区间不是符号概率区间的子集，则生成第一信号，并将第一信号发送至第二比特流概率区间生成单元134；若比特流概率区间是符号概率区间的子集，则生成第二信号及输出数据，并将第二信号发送至比特流概率区间更新单元135和符号概率区间更新单元136，将输出数据输出。
第二比特流概率区间生成单元134还与数据调度模块120相连，被配置为当接收到第一信号时，根据调度的数据对比特流概率区间进行更新，并将更新后的比特流概率区间发送至比较单元133。比特流概率区间更新单元135被配置为接收到第二信号时，对比特流概率区间更新单元135进行更新。符号概率区间更新单元136被配置为接收到第二信号时，对符号概率区间进行更新。
其中，第一信号可以理解为“非信号”，第二信号可以理解为“是信号”。
在一种可实施方式中，输入的比特数据为“0”或“1”，第一比特流概率区间生成单元131根据比特数据的值初始化比特流概率区间，即从比特数据的值对应的概率区间中选择一个区间。第二比特 流概率区间生成单元134根据比特数据的值更新比特流概率区间，即从比特数据的值对应的概率区间中重新选择一个区间。
在一种可实施方式中，比特流概率区间更新单元135可包括第一除法器和第二除法器。第一除法器被配置为更新比特流概率区间的上限；第二除法器被配置为更新比特流概率区间的下限。具体地，根据更新前的比特流概率区间的上下限和符号概率区间的上下限更新比特流概率区间的上下限。
在一种可实施方式中，符号概率区间更新单元136可以采用除法器对符号概率区间进行更新。为了减少系统中使用除法器的数量，符号概率区间更新单元136可为查表更新单元；查表更新单元被配置为根据更新前的符号概率区间生成查表地址，根据查表地址查找更新后的符号概率区间。即符号概率区间更新单元136采用LUT(Look-Up-Table)的硬件设计，对于并行的多个概率整形模块，可以复用该LUT表，从而可以降低资源消耗。在光通信系统中，概率整形模块可输出固定数量的符号，假设总数为n，符号1的个数为n1，符号3的个数为n3，且n1+n3＝n。那么符号概率区间包含的概率区间的数量为
Figure PCTCN2021096031-appb-000001
示例性地，图3为符号1对应的符号概率区间的数量的更新过程的一种示意图。
在一种可实施方式中，图4是本申请实施例提供的概率整形编码装置的一种结构示意图。如图4所示，概率整形编码模块130还可包括多个比特流概率区间缓存器137、多个符号概率区间缓存器138及概率区间调度单元139；比特流概率区间缓存器137的数量和符号概率区间缓存器138的数量均与数据缓存器110的数量相同，且比特流概率区间缓存器137、符号概率区间缓存器138以及数据缓存器110一一对应。
多个比特流概率区间缓存器137、多个符号概率区间缓存器138均与概率区间调度单元139相连；比特流概率区间缓存器137被配置为缓存更新后的比特流概率区间；符号概率区间缓存器138被配置为 缓存更新后的符号概率区间；概率区间调度单元139被配置为当比较单元133生成多个第二信号时，根据设定优先级规则将比特流概率区间调度至比特流概率区间更新单元135进行更新，将符号概率区间调度至符号概率区间更新单元136进行更新。
在一种可实施方式中，第二信号中携带有数据缓存器110的信息，设定优先级规则可以是缓存器中已被调度的数据量越小，缓存器的优先级越高。在一种可实施方式中，可对数据缓存器110进行排序，排序越靠前的缓存器的优先级越高。
示例性地，图5是本申请实施例提供的概率整形编码的一种示意图。如图5所示，本实施例中的除法器实现了最大限速的利用，避免除法器任意一级流水出现空闲状态。本申请实施例中数据块间流水编码，数据块内串行编码。该装置对数据块进行并行编码，从而提高编码速率。
在一种可实施方式中，概率整形编码装置还可包括：双缓存。双缓存被配置为存储输出数据。其中，双缓存可以采用ping pong buffer。对于每个通道配置一个输出ping pong buffer，当ping buffer进行编码数据输出时，pong buffer可进行数据编码缓存。按照这种方法进行数据ping pong输出。
本申请实施例提供的概率整形编码装置可包括多个数据缓存器、数据调度模块及概率整形编码模块；多个缓存器中的任一缓存器被配置为缓存输入的数据块；数据调度模块被配置为按照设定优先级规则确定目标缓存器，并调度目标缓存器中的数据；概率整形模块被配置为根据目标缓存器获取比特流概率区间和符号概率区间，并根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码；其中，缓存器的数量由概率整形编码模块编码时所需的脉冲周期确定。将输入数据缓存至多个缓存器中，利用数据调度模块对各缓存器中的数据按照设定优先级调度后，由概率整形编码模块进行编码，可以同时对输入的数据并行进行编码，不仅实现对数据传输过程中的概率整形编码，而且可以提高编码的速率。
图6是本申请实施例提供的概率整形编码系统的一种结构示意 图。如图6所示，该系统包括至少一个上述概率整形编码装置；且至少一个概率整形编码装置并行工作。
对于每一个概率整形编码装置均配置一个位宽为n的两个寄存器组作为ping pong buffer输出。采用n组ping pong buffer输出可以保证概率整形编码装置输出并行度满足数据流量需求(即，L的需求)。采用这种结构输出实现了大吞吐量下的任意并行度结构编码要求。
图7是本申请实施例提供的概率整形编码方法的一种流程图。以上述可实施方式为基础，该方法可包括如下步骤S210～S250。
在S210中，根据传输需求及概率分布参数对输入的数据进行分块，获得多个数据块。
其中，传输需求可以包括比特率及频率效率。具体地，根据传输需求及概率分布参数对输入的数据进行分块的方式可以是：根据传输需求及概率分布参数确定每个数据块包含的数据量；根据每个数据块包含的数据量对输入的数据进行分块。示例性地，数据块包含的数据量可为128比特。
在S220中，将多个数据块输入多个缓存器中进行缓存。
其中，缓存器的数量由概率整形编码所需的脉冲周期来确定。例如，假设需要P个脉冲周期完成一次概率整形编码，则缓存器的数量为P。缓存器的存储深度可以设置为数据块的大小。
具体地，对输入数量分块之后，将数据块分别输入多个缓存器进行缓存。
在S230中，按照设定优先级规则确定目标缓存。
其中，目标缓存为当前被调度的缓存。设定优先级规则可以是缓存器中已被调度的数据量最小，缓存器的优先级最高。或者，对缓存器进行排序，排序越靠前的缓存器的优先级越高。
在一种可实施方式中，按照设定优先级原则确定当前被调度的缓存的方式可以是：获取各路缓存已被调度的数据量；将已被调度的数据量最小的缓存确定为目标缓存。
在一种可实施方式中，若目标缓存有多个，则将目标缓存中排 序最低的缓存确定为最终被调度的缓存。
在S240中,根据目标缓存获取比特流概率区间和符号概率区间。
根据目标缓存获取比特流概率区间和符号概率区间的方式可以是目标缓存中的上一个数据被概率整形编码后更新后的比特流概率区间和符号概率区间。在本申请实施例中，更新的比特流概率区间和符号概率区间存储于寄存器中，可以根据目标缓存的信息对应的获取到比特流概率区间和符号概率区间。
在S250中,根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码。
根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码的方式可以是：判断比特流概率区间是否为符号概率区间的子集；若是，则根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码；若否，则根据当前调度的数据对比特流概率区间进行更新，并返回执行判断比特流概率区间是否为符号概率区间的子集的操作，直到比特流概率区间为符号概率区间的子集。
在一种可实施方式中，在根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码之后，该方法还包括如下步骤：对整形编码后的数据按照缓存器通道进行重新组合，并将重新组合后的数据输出至下级模块。
根据本申请实施例提供的概率整形编码方法，可根据传输需求及概率分布参数对输入的数据进行分块，获得多个数据块；将多个数据块输入多个缓存器中进行缓存；按照设定优先级规则确定目标缓存；其中，目标缓存为当前被调度的缓存；根据目标缓存获取比特流概率区间和符号概率区间；根据比特流概率区间和符号概率区间对当前调度的数据进行概率整形编码；从而可实现对数据传输过程中的概率整形编码，可以提高编码的速率。

Claims (14)

  1. 一种概率整形编码装置，包括：多个数据缓存器、数据调度模块及概率整形编码模块；其中：
    所述多个数据缓存器中的任一数据缓存器被配置为缓存输入的数据块；
    所述数据调度模块被配置为按照设定优先级规则确定目标缓存器，并调度所述目标缓存器中的数据；
    所述概率整形模块被配置为根据所述目标缓存器获取比特流概率区间和符号概率区间，并根据所述比特流概率区间和所述符号概率区间对当前调度的数据进行概率整形编码；以及
    所述多个数据缓存器的数量由所述概率整形编码模块编码时所需的脉冲周期确定。
  2. 根据权利要求1所述的装置，其中，所述数据调度模块包括第一调度单元和第二调度单元；
    所述第一调度单元被配置为将各个数据缓存器中已被调度的数据量最小的数据缓存器确定为所述目标缓存器；以及
    所述第二调度单元被配置为当所述目标缓存器为多个时，将所述目标缓存器中排序最低的数据缓存器确定为最终被调度的缓存。
  3. 根据权利要求1所述的装置，其中，所述概率整形编码模块包括：第一比特流概率区间生成单元、符号概率区间生成单元、比较单元、第二比特流概率区间生成单元、比特流概率区间更新单元和符号概率区间更新单元；
    所述第一比特流概率区间生成单元分别与所述数据调度模块及比较单元相连，被配置为根据被调度的数据初始化比特流概率区间，并将初始化的比特流概率区间发送至所述比较单元；
    所述符号概率区间生成单元与所述比较单元相连，被配置为初始化符号概率区间，并将初始化的符号概率区间发送至所述比较单元；
    所述比较单元分别与所述第二比特流概率区间生成单元、所述 比特流概率区间更新单元及符号概率区间更新单元相连；所述比较单元被配置为比较所述比特流概率区间和所述符号概率区间，响应于确定所述比特流概率区间不是所述符号概率区间的子集，生成第一信号，并将所述第一信号发送至所述第二比特流概率区间生成单元；响应于确定所述比特流概率区间是所述符号概率区间的子集，生成第二信号及输出数据，并将所述第二信号发送至所述比特流概率区间更新单元和所述符号概率区间更新单元，将所述输出数据输出；
    所述第二比特流概率区间生成单元与所述数据调度模块相连，被配置为响应于接收到第一信号，根据调度的数据对所述比特流概率区间进行更新，并将更新后的比特流概率区间发送至所述比较单元；
    所述比特流概率区间更新单元被配置为响应于接收到第二信号，对所述比特流概率区间更新单元进行更新；以及
    所述符号概率区间更新单元被配置为响应于接收到第二信号，对所述符号概率区间进行更新。
  4. 根据权利要求3所述的装置，其中，所述比特流概率区间更新单元包括第一除法器和第二除法器；
    所述第一除法器被配置为更新所述比特流概率区间的上限；以及
    所述第二除法器被配置为更新比特流概率区间的下限。
  5. 根据权利要求3所述的装置，其中，所述符号概率区间更新单元为查表更新单元；以及
    所述查表更新单元被配置为根据更新前的符号概率区间生成查表地址，并根据所述查表地址查找更新后的符号概率区间。
  6. 根据权利要求3所述的装置，还包括多个比特流概率区间缓存器、多个符号概率区间缓存器，以及概率区间调度单元；其中：
    所述多个比特流概率区间缓存器的数量以及所述多个符号概率区间缓存器的数量均与所述多个数据缓存器的数量相同，且所述多个 比特流概率区间缓存器、所述多个符号概率区间缓存器以及所述多个数据缓存器一一对应；
    所述多个比特流概率区间缓存器、所述多个符号概率区间缓存器均与所述概率区间调度单元相连；
    所述多个比特流概率区间缓存器中的任一比特流概率区间缓存器被配置为缓存更新后的比特流概率区间；
    所述多个符号概率区间缓存器中的任一符号概率区间缓存器被配置为缓存更新后的符号概率区间；以及
    所述概率区间调度单元被配置为响应于确定所述比较单元生成多个第二信号，根据所述设定优先级规则将比特流概率区间调度至所述比特流概率区间更新单元进行更新，并将符号概率区间调度至所述符号概率区间更新单元进行更新。
  7. 根据权利要求3所述的装置，还包括：双缓存，被配置为存储输出数据。
  8. 一种概率整形编码系统，包括至少一个根据权利要求1-7任一项所述的概率整形编码装置；其中，所述至少一个概率整形编码装置并行工作。
  9. 一种概率整形编码方法，包括：
    根据传输需求及概率分布参数对输入的数据进行分块，获得多个数据块；
    将所述多个数据块输入多个缓存器中进行缓存；
    按照设定优先级规则确定目标缓存；其中，目标缓存为当前被调度的缓存；
    根据所述目标缓存获取比特流概率区间和符号概率区间；以及
    根据所述比特流概率区间和所述符号概率区间对当前调度的数据进行概率整形编码。
  10. 根据权利要求9所述的方法，其中，根据所述传输需求及所述概率分布参数对所述输入的数据进行分块，包括：
    根据所述传输需求及所述概率分布参数确定每个数据块包含的数据量；其中，所述传输需求包括比特率及频率效率；以及
    根据所述每个数据块包含的数据量对所述输入的数据进行分块。
  11. 根据权利要求9所述的方法，其中，按照所述设定优先级原则确定所述目标缓存，包括：
    获取各路缓存已被调度的数据量；以及
    将已被调度的数据量最小的缓存确定为所述目标缓存。
  12. 根据权利要求11所述的方法，其中，将所述已被调度的数据量最小的缓存确定为所述目标缓存，包括：
    若所述目标缓存有多个，则将所述目标缓存中排序最低的缓存确定为所述当前被调度的缓存。
  13. 根据权利要求9所述的方法，其中，根据所述比特流概率区间和所述符号概率区间对所述当前调度的数据进行概率整形编码，包括：
    循环执行以下步骤，直至确定所述比特流概率区间为所述符号概率区间的子集，根据所述比特流概率区间和所述符号概率区间对所述当前调度的数据进行概率整形编码：
    判断所述比特流概率区间是否为所述符号概率区间的子集；以及
    响应于确定所述比特流概率区间不是所述符号概率区间的子集，根据所述当前调度的数据更新所述比特流概率区间。
  14. 根据权利要求9所述的方法，在根据所述比特流概率区间和所述符号概率区间对所述当前调度的数据进行概率整形编码之后，还包括：
    对所述整形编码后的数据按照缓存器通道进行重新组合。
PCT/CN2021/096031 2020-05-29 2021-05-26 概率整形编码装置、系统及方法 WO2021238976A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010479404.9A CN113746594A (zh) 2020-05-29 2020-05-29 概率整形编码装置、系统及方法
CN202010479404.9 2020-05-29

Publications (1)

Publication Number Publication Date
WO2021238976A1 true WO2021238976A1 (zh) 2021-12-02

Family

ID=78725108

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/096031 WO2021238976A1 (zh) 2020-05-29 2021-05-26 概率整形编码装置、系统及方法

Country Status (2)

Country Link
CN (1) CN113746594A (zh)
WO (1) WO2021238976A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024130578A1 (en) * 2022-12-21 2024-06-27 Qualcomm Incorporated Probabilistic shaping based on block codes

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108092712A (zh) * 2017-12-08 2018-05-29 武汉邮电科学研究院 基于概率整形编码的可见光通信系统、方法及应用设备
CN110140330A (zh) * 2017-01-02 2019-08-16 杜塞尔多夫华为技术有限公司 用于整形数据序列概率分布的装置和方法
US10516503B1 (en) * 2019-02-20 2019-12-24 Mitsubishi Electric Research Laboratories, Inc. Distribution matcher
US10608749B1 (en) * 2018-01-24 2020-03-31 Inphi Corporation Probabilistic shaping techniques for high performance coherent optical transceivers

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110140330A (zh) * 2017-01-02 2019-08-16 杜塞尔多夫华为技术有限公司 用于整形数据序列概率分布的装置和方法
CN108092712A (zh) * 2017-12-08 2018-05-29 武汉邮电科学研究院 基于概率整形编码的可见光通信系统、方法及应用设备
US10608749B1 (en) * 2018-01-24 2020-03-31 Inphi Corporation Probabilistic shaping techniques for high performance coherent optical transceivers
US10516503B1 (en) * 2019-02-20 2019-12-24 Mitsubishi Electric Research Laboratories, Inc. Distribution matcher

Also Published As

Publication number Publication date
CN113746594A (zh) 2021-12-03

Similar Documents

Publication Publication Date Title
EP3718228B1 (en) Communication system and method using a set of distribution matchers
US20190052509A1 (en) Virtual lookup table for probabilistic constellation shaping
US10749554B1 (en) System and method for short block length distribution matching
US8619762B2 (en) Low power deserializer and demultiplexing method
US10516503B1 (en) Distribution matcher
WO2019134681A1 (en) Probabilistically shaped multi-level pulse modulation with gray code mapping
Fehenberger et al. Parallel-amplitude architecture and subset ranking for fast distribution matching
JP2022529393A (ja) 短ブロック長分布マッチングアルゴリズム
WO2021238976A1 (zh) 概率整形编码装置、系统及方法
CN107018113A (zh) 发射机、接收机和信号处理的方法
CN110401489A (zh) 一种光纤信号产生方法、装置及系统
JPWO2016121341A1 (ja) 光送信器、光通信システム、および光通信方法
WO2024103712A1 (zh) 一种信号扫描方法、装置、电子设备及存储介质
CN112888019A (zh) 数据发送方法、数据接收方法、芯片、调制解调器及终端
WO2024016649A1 (zh) 总线传输结构及方法、芯片
JP2022007503A (ja) 受信装置及び復号化方法
CN103488611B (zh) 基于IEEE802.11.ad协议的FFT处理器
Hou et al. Cognitive radio spectrum allocation strategy based on improved genetic algorithm
Li et al. Cognitively reconfigurable mimic-based heterogeneous password recovery system
CN1437316A (zh) 一种数控振荡器及其产生正余弦信号的方法
Cai et al. Research of federated learning communication data transmission algorithm based on user grouping policy
Li et al. Color-Shift Keying Code Design in Visible Light Communications Using Finite-State Machines
Li et al. Intelligent equalization based on RBF LSSVM and adaptive channel decoding in faster-than-nyquist receiver
WO2024159980A1 (zh) 信号传输方法、装置及设备
WO2022042698A1 (zh) 数据发送方法及装置、存储介质、电子装置

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: 21812711

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11/05/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21812711

Country of ref document: EP

Kind code of ref document: A1