WO2011026330A1 - Method and device for channal decoding - Google Patents

Method and device for channal decoding Download PDF

Info

Publication number
WO2011026330A1
WO2011026330A1 PCT/CN2010/071993 CN2010071993W WO2011026330A1 WO 2011026330 A1 WO2011026330 A1 WO 2011026330A1 CN 2010071993 W CN2010071993 W CN 2010071993W WO 2011026330 A1 WO2011026330 A1 WO 2011026330A1
Authority
WO
WIPO (PCT)
Prior art keywords
state
value
decoded
metric
data
Prior art date
Application number
PCT/CN2010/071993
Other languages
French (fr)
Chinese (zh)
Inventor
杜凡平
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 WO2011026330A1 publication Critical patent/WO2011026330A1/en

Links

Classifications

    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/37—Decoding methods or techniques, not specific to the particular type of coding provided for in groups H03M13/03 - H03M13/35
    • H03M13/39—Sequence estimation, i.e. using statistical methods for the reconstruction of the original codes
    • H03M13/3905—Maximum a posteriori probability [MAP] decoding or approximations thereof based on trellis or lattice decoding, e.g. forward-backward algorithm, log-MAP decoding, max-log-MAP decoding
    • H03M13/3938—Tail-biting

Definitions

  • the present invention relates to the field of communications, and in particular to a channel decoding method and apparatus.
  • BACKGROUND In a wireless communication network, a spatial channel is erratic as a sea surface, and the signal is often distorted, thereby increasing the bit error rate of the transmitted data. In order to reduce the bit error rate, it is necessary to transmit data with certain anti-interference ability.
  • the commonly used method is to encode the transmission data in some way and add redundant protection information, which is channel coding.
  • the channel coding methods used by the 3rd Generation Partnership Project (3GPP) mainly include convolutional codes, turbo codes, and interleaving.
  • the convolutional code is relatively simple compared to the turbo code, and the convolutional coding is generally used in the case of a short code, for example, a broadcast channel in a Long-Term Evolution (LTE).
  • BCH downlink control information
  • DCI Downlink Control Information
  • UCI Uplink Control Information
  • a tail-biting convolutional encoder is used in LTE, and the structure is shown in Fig. 2.
  • the initial value of the encoder's status register is set to the corresponding value of the last 6 information bits of the input data stream, so that the initial and final states of the shift register are the same.
  • the inventors have found that the related art has at least the following problems: When calculating branch metrics, methods such as Hamming metrics and Euclidean metrics are generally used, which are the shortest measures to make the distance, but the metrics and their backtracking The method wastes a lot of resources and the calculation speed is slow. SUMMARY OF THE INVENTION The present invention is directed to a channel decoding method and apparatus to solve the problem of waste of resources and slow calculation speed in the prior art.
  • a channel decoding method including: calculating a bite roll
  • the correlation value of the output value of all the states of the product encoder and the data to be decoded is used as the branch metric value; according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and a larger one is selected.
  • the accumulated value is used as a new metric value of the next state, and the selection result is saved until the data to be decoded ends; according to the selection result, the backtracking is started from the state where the final metric value is the largest, and the decoding is obtained.
  • the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded
  • the data end includes: accumulating the metric values of the corresponding two states and the branch metrics according to the state transition diagram; according to the addition ratio rule, selecting a larger accumulated value as the new metric value of the next state, and The selection result is saved until the data to be decoded ends; the state in which the final metric value is the largest is saved.
  • the backtracking is started from the state in which the final metric value is the largest according to the selection result.
  • the decoding specifically includes: backtracking from the state with the largest metric value, and finding the parent state according to the selection result until all the historical states are found; The lowest bit of the history status output is used as the decoded output bit.
  • the metric values of the corresponding two states are respectively accumulated with the branch metric values, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the translation is to be translated.
  • the method further includes: saving the data to be decoded; accumulating the metric values of the corresponding two states and the branch metric respectively, and selecting a larger accumulated value as the new metric value of the next state until the The decoded data is saved, and the state with the largest metric value is saved as the first termination state.
  • the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded
  • the method further includes: if the state of the final metric value is different from the first termination state, and the maximum number of parameters is not reached, the decoding process is continued.
  • the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers
  • the number of all states is 64
  • the number of different branch metrics is 8.
  • corresponding The metrics of the two states are respectively accumulated with the branch metrics, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the end of the data to be decoded includes:
  • the metric values of the corresponding two states are respectively accumulated with the branch metric values; according to the addition ratio rule, the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends; Among them, if you select the above path, the result will be 0, otherwise it will be 1.
  • three soft bits exist in parallel in one address.
  • a channel decoding apparatus comprising: a branch metric value calculation module, configured to calculate a correlation value of an output value of all states of a tail-biting convolutional encoder and data to be decoded As a branch metric; a comparison selection module, configured to accumulate the metric values of the corresponding two states and the branch metric according to the state transition graph, and select a larger accumulated value as the new metric value of the next state. The selection result is saved until the data to be decoded ends; the backtracking module is configured to perform backtracking according to the selection result from the state with the largest metric value to obtain decoding.
  • the comparison selection module is configured to accumulate the metric value of the corresponding state and the branch metric value according to the state transition diagram; according to the addition ratio selection rule, select a larger accumulated value as the new metric value of the next state, and The selection result is saved until the data to be decoded ends; the state in which the final metric value is the largest is saved.
  • the backtracking module is configured to backtrack from the state with the largest metric value, and find the parent state according to the selection result until all the history states are found; the lowest bit of the historical state output is used as the decoding output bit.
  • the method further includes: a saving module, configured to save data to be decoded; and the adding and selecting module is further configured to accumulate the metric values of the corresponding two states and the branch metric respectively, and select The larger accumulated value is used as the new metric value of the next state until the data to be decoded is saved, and the state with the largest metric value is saved as the first termination state.
  • the method further includes: a determining module, configured to determine that if the state in which the final metric value is the largest is different from the first termination state, and the maximum number of parameters is not reached, the decoding process is continued.
  • the comparison and selection module is used for: According to the state transition diagram, the metric value of the corresponding state is accumulated with the branch metric value; according to the addition ratio rule, the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded End; where, if the above path is selected, the result of the selection is marked as 0, otherwise it is 1.
  • the saved data to be decoded three soft bits exist in parallel in one address.
  • FIG. 1 is a flow chart showing a channel decoding method according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram showing a tail biting convolutional encoder in accordance with a preferred embodiment of the present invention
  • FIG. 4 is a schematic illustration of a full state transition of a tail biting convolution in accordance with a preferred embodiment of the present invention
  • FIG. 5 illustrates an addition in accordance with a preferred embodiment of the present invention.
  • FIG. 6 shows a schematic diagram of a backtracking decoding output in accordance with a preferred embodiment of the present invention
  • Figure 7 shows a flow chart of a Viterbi decoding method in accordance with a preferred embodiment of the present invention
  • FIG. 9 is a diagram showing a channel decoding apparatus according to an embodiment of the present invention.
  • Step S10 calculating an output value of all states of a tail-biting convolutional encoder related to data to be decoded The value is used as the branch metric value
  • Step S20 according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection is saved.
  • step S30 the backtracking is started from the state in which the final metric value is the largest according to the selection result, and the decoding is obtained.
  • the step S20 specifically includes: summarizing the metric values of the corresponding two states and the branch metrics according to the state transition map; according to the adding comparison rule, selecting a larger accumulated value as the new state of the next state The metric, and save the selection until the end of the data to be decoded; save the state with the largest metric.
  • Step S30 specifically includes: backtracking from the state with the largest metric value, finding the parent state according to the selection result, until all the history states are found; and the lowest bit output by the history state is used as the decoding output bit.
  • the preferred embodiment provides an accumulation of state metrics according to a comparison selection rule, and saves the state of the addition comparison result and the final maximum state metric value, and finally backtracks from the state according to the result of the comparison selection. Specific embodiment.
  • the method further includes: saving data to be decoded; accumulating the metric values of the corresponding two states and the branch metric respectively, and selecting a larger accumulated value as the new metric value of the next state Until the data to be decoded is saved, the state with the largest metric value is saved as the first termination state.
  • the method further includes: if the state in which the final metric value is the largest is different from the first termination state, and the maximum number of turns is not reached, the decoding process is continued.
  • the state value of the initial state is initialized in the process of saving the data to be decoded, thereby reducing the influence of the initial state uncertainty of the tail-biting convolutional code.
  • the decoding can be continued until the initial and final states are matched or the maximum number of cycles is reached, which is improved.
  • the performance of the decoding Preferably, when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, all states are 64, and different branch metrics have 8.
  • the corresponding two The metrics of the states are respectively accumulated with the branch metrics, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the end of the data to be decoded includes:
  • the corresponding The metrics of the two states are respectively accumulated with the branch metrics; according to the addition ratio rule, the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends; If you select one of the above paths, the result of the selection is marked as 0, otherwise it is 1.
  • three soft bits exist in parallel in one address.
  • the three soft bits can be simultaneously read and correlated with the state code output bits.
  • the above preferred embodiment provides a specific embodiment of the Viterbi decoding method of the present invention when the tail biting convolutional encoder is as shown in Fig. 2 with a coding rate of 1/3.
  • Viterbi decoding is the maximum likelihood, that is, to find the most similar set of all possible coding combinations with the data to be decoded, because the coding combination and the length of the encoded data are exponential power, so the exact match coding combination is Impossible, the Viterbi algorithm compares the two sets of state metrics arriving at the same state according to the state transition diagram of the encoder, selects the most likely state, discards the other, the so-called plus comparison, thus guaranteeing The amount of state remaining is always equal to the state quantity of the encoder, so that the amount of decoding calculation is acceptable.
  • a schematic diagram of the full state transition is shown in Figure 4.
  • ⁇ First, according to the structure of the tail bit encoder (as shown in Figure 2), the output values of all states are obtained and saved. Because the length of the encoder is 6 registers, there are 64 states. Each state outputs 3 bits, so the size of the state code output table is 64x3. Then, the output value of the state is correlated with the data to be decoded, and 64 correlation values are obtained as branch metric values.
  • the status coded output bit can be expressed as: a(n,m)a, where n represents the state value and m represents the bit number. In practical applications, 0 means positive and 1 means negative. For example: In state 32, the output bit is 011, then the correlation value with the data to be coded is +D0-D 1-D2.
  • the state point with the largest metric value is used as the backtracking point, the previous state is calculated according to the comparison result of the state record, and so on, and all the historical states are obtained.
  • the backtracking method is the inverse of FIG. 5, that is, it is inferred from the comparison result that the previous state is the current state shifted to the right by one bit or 32. Since the change of state is related to the input bit, that is, the lowest bit of the state is the input encoded data, therefore, the last bit of the backtracked history state is directly regarded as the decoded output bit. For example, if the initial ⁇ ⁇ state is 2 and the input bit is 1, then ⁇ !
  • the relationship between the decoded output and the state backtracking is shown in Figure 6. This is a right shifting process. It can be seen that the decoded output can be seen as the inverse of the encoded input.
  • the Viterbi decoding method of the above preferred embodiment can be applied not only to tail-biting convolution but also to zero-revolution, which is characterized by low resource consumption and high speed.
  • the preferred embodiment of the present invention also proposes the following methods: First, when storing data to be decoded, the metric value of the state is simultaneously calculated, but it is not necessary to save the comparison result. Because of the tail biting, when the data to be decoded is saved, the initial state value of starting decoding is obtained. In this way, the data can be decoded while being decoded, and the speed can be made faster. Secondly, the state value of the maximum state metric in the initial state is saved. When the decoding ends, the state value of the maximum state metric of the termination state is compared. If it is the same, the decoding is stopped, otherwise the loop decoding is continued until the above condition is met or arrived. The maximum number of cycles.
  • the parallel computation method can be used, that is, the state metrics of all 64 states are calculated in one step, and the 64 branch metrics need to be calculated simultaneously.
  • the parallel storage mode is used when storing the data to be decoded, that is, three soft bits have one address in parallel, so that it can be read out at the same time and correlated with the state code output bit.
  • Step S102 before decoding, establishing a state code output table, and obtaining output values of all possible states according to the structure of the tail bit encoder
  • Step S104 saving data to be decoded, and 3 data to be decoded
  • the soft bits are stored in parallel in an address, and the state metrics are accumulated according to the addition ratio rule, and the comparison result is not saved until the data is saved, and the state value of the maximum metric value is saved
  • Step S106 read and save The data to be decoded is accumulated according to the addition selection rule, and the result of the comparison is saved, and the result of the comparison is saved for the backtracking at the end of the decoding, until the end of the decoding, the state value of the maximum metric value is saved
  • Step S108 If the termination state is different from the previous termination state and the maximum number of parameters is not reached, then return to step 4 to gather S 106; step S 110, otherwise
  • Step S202 Before decoding, establish a state code output table; Step S204, set K time The data to be decoded is correlated with the coded output values of all states, and all the correlation values are recorded as the branch metric values; Step S206, according to the state transition diagram (as shown in FIG. 4), the metric value and the branch metric of the corresponding state are determined. The values are accumulated, and the accumulated result is taken as the metric of the next state, because the state at the next moment according to the state transition diagram comes from two states, so the result of selecting one of the accumulated values is saved by using "add comparison".
  • Step S208 determining whether the data to be decoded is finished, if not, returning to step S202; Step S210, if yes, searching for the state value of the maximum state metric value; Step S212, The state is backtracked, and according to the record of the saved comparison result, the parent state is searched until all history states are found; step S214, the state calendar is The output of the LSB bit decode output as FIG 9 shows a schematic embodiment UN channel decoding apparatus of the embodiment of the present invention, the apparatus comprising:
  • the branch metric calculation module 10 is configured to calculate a correlation value of the output value of all states of the tail-biting convolutional encoder and the data to be decoded as a branch metric value; and the comparison and selection module 20 is configured to The metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends;
  • the comparison selection module 20 is configured to accumulate the metric values of the corresponding two states and the branch metric values according to the state transition diagram; according to the addition ratio selection rule, select a larger accumulated value as the new state of the next state. The metric, and save the selection until the end of the data to be decoded; save the state with the largest metric.
  • the backtracking module 30 is configured to backtrack from the state with the largest metric value, find the parent state according to the selection result, until all the history states are found; the lowest bit of the historical state output is used as the decoding output bit.
  • the preferred embodiment provides an accumulation of state metrics according to a comparison selection rule, and saves the state of the addition comparison result and the final maximum state metric value, and finally backtracks from the state according to the result of the comparison selection. Specific embodiment.
  • the method further includes: a saving module 40, configured to save data to be decoded; and the comparison and selection module 20 is further configured to accumulate the metric value of the corresponding state and the branch metric value, and select a larger The accumulated value is used as the new metric value of the next state, until the data to be decoded is saved, and the state with the largest metric value is saved as the first termination state; the determining module 50 is configured to determine if the final metric value is the largest. The state is different from the first termination state, and the maximum number of turns is not reached, and the decoding process continues.
  • a saving module 40 configured to save data to be decoded
  • the comparison and selection module 20 is further configured to accumulate the metric value of the corresponding state and the branch metric value, and select a larger The accumulated value is used as the new metric value of the next state, until the data to be decoded is saved, and the state with the largest metric value is saved as the first termination state
  • the determining module 50 is configured to determine if the
  • the state value of the initial state is initialized in the process of saving the data to be decoded, thereby reducing the influence of the initial state uncertainty of the tail-biting convolutional code.
  • the decoding can be continued until the initial and final states are matched or the maximum number of cycles is reached, which is improved.
  • the performance of the decoding Preferably, when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, all states are 64, and different branch metrics are 8.
  • the comparison and selection module is used to transfer the map according to the state.
  • the result of the selection is marked as 0, otherwise it is marked as 1.
  • three soft bits exist in parallel in one address. In this way, by using the parallel storage method, the three soft bits can be simultaneously read and correlated with the state code output bits.
  • the present invention achieves the following technical effects: Since the state correlation value is used as the branch metric, the record comparison and selection result is used as the backtracking roadmap, and the lowest bit of the backtracking state is regarded as the decoding output, and the solution is solved. The problem of wasted resources in the prior art and the calculation speed is slow. The use of this embodiment makes the decoded occupied resources less and the calculation speed is fast. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices.
  • the invention is not limited to any specific combination of hardware and software.
  • the above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Landscapes

  • Physics & Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Error Detection And Correction (AREA)

Abstract

A method and device for channel decoding is provided. The method includes: calculating the correlated values of all state output values of a tail biting convolution coder and the data to be decoded as branch metric values; accumulating the metric values of corresponding two states and the branch metric values respectively according to a state transition graph, and selecting the bigger accumulated values as the new metric values of the next states, and storing the selected results until the end of the data to be decoded; backtracking from the state with the final maximum metric value according to the selected results to obtain the decoded values.

Description

一种信道译码方法和装置 技术领域 本发明涉及通信领域， 具体而言， 涉及一种信道译码方法和装置。 背景技术 在无线通信网络中， 空间信道如同海面， 变化无常， 往往使信号发生畸 变， 从而增加了传输数据的误码率。 为了降低误码率， 需要传输数据具有一 定的抗千扰能力， 通常釆用的方法是对传输数据进行某种方式的编码， 增加 冗余的保护信息， 这就是信道编码。 第三代合作伙伴计划 （ 3rd Generation Partnership Project, 简称为 3GPP )釆用的信道编码方法主要有卷积码、 turbo 码以及交织等手段。 其中， 由于卷积码与 turbo码相比编译码相对简单， 在短码情况下一般 釆用卷积编码， 例如， 长期演进（Long-Term Evolution, 简称为 LTE ) 中的 广播信道（ Broadcast channel,简称为 BCH ),下行控制信息（ Downlink Control Information, 简称为 DCI )、 上行控制信息（ Uplink Control Information, 简称 为 UCI ) 等信道中釆用卷积编码进行编码。 为了提高卷积码的码率， 在 LTE中釆用了咬尾卷积编码器， 结构如图 2 所示。 编码器的状态寄存器的初始值设为输入数据流的最后 6个信息比特的 对应值， 使得移位寄存器的初始和最终状态相同。 但是， 因为初始状态与结 束状态是不确定的， 因此釆用传统的 Viterbi译码方法性能要有所下降。 发明人发现相关技术至少存在如下问题： 在计算分支度量时，一般釆用 汉明度量、 欧几里德度量等方法， 这些都是使距离最短小的度量方法， 但是 这种度量方法及其回溯方法会浪费很大的资源， 计算速度较慢。 发明内容 本发明旨在提供一种信道译码方法和装置，以解决现有技术存在的浪费 资源， 计算速度较慢的问题。 根据本发明的一个方面， 提供了一种信道译码方法， 包括： 计算咬尾卷 积编码器的所有状态的输出值与待译码的数据的相关值作为分支度量值； 根 据状态转移图， 将对应的两个状态的度量值分别与分支度量值进行累加， 并 选择较大的累加值作为下一状态的新的度量值， 保存选择结果， 直到待译码 数据结束； 根据选择结果从最终的度量值最大的状态开始回溯， 得到译码。 优选地， 根据状态转移图， 将对应的两个状态的度量值分别与分支度量 值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 保存选择结 果， 直到待译码数据结束具体包括： 根据状态转移图， 将对应的两个状态的 度量值分别与分支度量值进行累加； 按照加比选法则， 选择较大的累加值作 为下一状态的新的度量值， 并保存选择结果， 直到待译码数据结束； 保存最 终的度量值最大的状态。 优选地， 根据选择结果从最终的度量值最大的状态开始回溯， 得到译码 具体包括： 由最终的度量值最大的状态进行回溯，根据选择结果寻找父状态， 直到找出所有的历史状态； 将历史状态输出的最低比特作为译码输出比特。 优选地， 在根据状态转移图， 将对应的两个状态的度量值分别与分支度 量值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 保存选择 结果， 直到待译码数据结束之前还包括： 保存待译码的数据； 将对应的两个 状态的度量值分别与分支度量值进行累加， 并选择较大的累加值作为下一状 态的新的度量值， 直到待译码的数据保存完毕， 保存度量值最大的状态作为 第一次的终止状态。 优选地， 根据状态转移图， 将对应的两个状态的度量值分别与分支度量 值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 保存选择结 果， 直到待译码数据结束之后还包括： 如果最终的度量值最大的状态与第一 次的终止状态不同， 且未达到最大的圏数， 则继续执行译码过程。 优选地， 当咬尾卷积编码器的编码率为 1/3 , 长度为 6个寄存器时， 所 有状态的个数为 64个， 不同的分支度量值有 8个， 根据状态转移图， 将对 应的两个状态的度量值分别与分支度量值进行累加， 并选择较大的累加值作 为下一状态的新的度量值， 保存选择结果， 直到待译码数据结束包括： 根据 状态转移图， 将对应的两个状态的度量值分别与分支度量值进行累加； 按照 加比选法则， 选择较大的累加值作为下一状态的新的度量值， 并保存选择结 果， 直到待译码数据结束； 其中， 如果选择的是上面的一条路径， 则将选择 结果 ΐ己为 0, 否则 ΐ己为 1。 优选地， 保存的待译码的数据中， 3个软比特并行存在一个地址中。 根据本发明的另一个方面， 还提供了一种信道译码装置， 包括： 分支度 量值计算模块， 用于计算咬尾卷积编码器的所有状态的输出值与待译码的数 据的相关值作为分支度量值； 加比选模块， 用于根据状态转移图， 将对应的 两个状态的度量值分别与分支度量值进行累加， 并选择较大的累加值作为下 一状态的新的度量值， 保存选择结果， 直到待译码数据结束； 回溯模块， 用 于根据选择结果从最终的度量值最大的状态开始回溯， 得到译码。 优选地， 加比选模块用于根据状态转移图， 将对应状态的度量值与分支 度量值进行累加； 按照加比选法则， 选择较大的累加值作为下一状态的新的 度量值， 并保存选择结果， 直到待译码数据结束； 保存最终的度量值最大的 状态。 优选地， 回溯模块用于由度量值最大的状态进行回溯， 根据选择结果寻 找父状态， 直到找出所有的历史状态； 将历史状态输出的最低比特作为译码 输出比特。 优选地， 在上述的装置中，还包括： 保存模块， 用于保存待译码的数据； 加比选模块还用于将对应的两个状态的度量值分别与分支度量值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 直到待译码的数据保存完 毕， 保存度量值最大的状态作为第一次的终止状态。 优选地， 在上述的装置中， 还包括： 判断模块， 用于判断如果最终的度 量值最大的状态与第一次的终止状态不同， 且未达到最大的圏数， 则继续执 行译码过程。 优选地， 当咬尾卷积编码器的编码率为 1/3 , 长度为 6个寄存器时， 所 有状态的个数为 64个， 不同的分支度量值有 8个， 加比选模块用于： 根据 状态转移图，将对应状态的度量值与分支度量值进行累加；按照加比选法则， 选择较大的累加值作为下一状态的新的度量值， 并保存选择结果， 直到待译 码数据结束； 其中， 如果选择的是上面的一条路径， 则将选择结果记为 0, 否则 ΐ己为 1。 优选地， 保存的待译码的数据中， 3个软比特并行存在一个地址中。 由于釆用状态相关值作为分支度量， 记录加比选结果作为回溯路标， 将 回溯状态的最低比特当作译码输出， 解决了现有技术存在的浪费资源， 计算 速度较慢的问题， 从而使得译码的占用资源少、 计算速度快。 附图说明 此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部 分， 本发明的示意性实施例及其说明用于解释本发明， 并不构成对本发明的 不当限定。 在附图中： 图 1示出了根据本发明实施例的信道译码方法的流程图； 图 2示出了 居本发明优选实施例的咬尾卷积编码器的示意图； 图 3示出了根据本发明优选实施例的咬尾卷积栅格示意图； 图 4示出了根据本发明优选实施例的咬尾卷积的全状态转移示意图； 图 5示出了根据本发明优选实施例的加比选路径关系图； 图 6示出了根据本发明优选实施例的回溯译码输出示意图； 图 7示出了根据本发明优选实施例一的 Viterbi译码方法的流程图； 图 8示出了根据本发明优选实施例二的单圏的 Viterbi译码方法的流程 图； 图 9示出了根据本发明实施例的信道译码装置的示意图。 具体实施方式 下面将参考附图并结合实施例， 来详细说明本发明。 图 1 示出了才艮据本发明实施例的信道译码方法的流程图， 包括以下步 骤： 步骤 S 10, 计算咬尾卷积编码器的所有状态的输出值与待译码的数据的 相关值作为分支度量值； 步骤 S20 , 根据状态转移图， 将对应的两个状态的度量值分别与分支度 量值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 保存选择 结果， 直到待译码数据结束； 步骤 S30, 根据选择结果从最终的度量值最大的状态开始回溯， 得到译 码。 该实施例由于釆用状态相关值作为分支度量，记录加比选结果作为回溯 路标， 将回溯状态的最低比特当作译码输出， 解决了现有技术存在的浪费资 源， 计算速度较慢的问题。 使用该实施例使得译码的占用资源少、 计算速度 快。 优选地， 步骤 S20具体包括： 才艮据状态转移图， 将对应的两个状态的度 量值分别与分支度量值进行累加； 按照加比选法则， 选择较大的累加值作为 下一状态的新的度量值， 并保存选择结果， 直到待译码数据结束； 保存最终 的度量值最大的状态。 步骤 S30具体包括： 由度量值最大的状态进行回溯，根据选择结果寻找 父状态， 直到找出所有的历史状态； 将历史状态输出的最低比特作为译码输 出比特。 该优选实施例提供了根据加比选法则进行状态度量值的累加，并保存加 比选结果以及最终的最大状态度量值的状态， 最后根据加比选的结果由该状 态进行回溯得到译码的具体实施方案。 优选地， 在步骤 S20之前还包括： 保存待译码的数据； 将对应的两个状 态的度量值分别与分支度量值进行累加， 并选择较大的累加值作为下一状态 的新的度量值， 直到待译码的数据保存完毕， 保存度量值最大的状态作为第 一次的终止状态。 在步骤 S20之后还包括： 如果最终的度量值最大的状态与 第一次的终止状态不同， 且未达到最大的圏数， 则继续执行译码过程。 这样， 为了适应咬尾卷积的特点， 在保存待译码数据的过程中， 初始化 了初始状态的状态值， 因此减少了咬尾卷积码初始状态不确定的影响。 另夕卜， 由于咬尾卷积的栅格图如图 3所示是周期循环的， 因此为了提高性能， 译码 可以一直循环进行下去， 直到初始与终止状态吻合或者达到最大循环次数， 提高了译码的性能。 优选地， 当咬尾卷积编码器的编码率为 1/3 , 长度为 6个寄存器时， 所 有状态为 64个， 不同的分支度量值有 8个， 根据状态转移图， 将对应的两 个状态的度量值分别与分支度量值进行累加， 并选择较大的累加值作为下一 状态的新的度量值， 保存选择结果， 直到待译码数据结束包括： 根据状态转 移图， 将对应的两个状态的度量值分别与分支度量值进行累加； 按照加比选 法则， 选择较大的累加值作为下一状态的新的度量值， 并保存选择结果， 直 到待译码数据结束； 其中， 如果选择的是上面的一条路径， 则将选择结果记 为 0 , 否则 ΐ己为 1。 优选地， 保存的待译码的数据中， 3个软比特并行存在一个地址中。 这 样， 釆用并行存储的方式， 可以将该 3个软比特同时读出， 与状态编码输出 比特进行相关计算。 上述优选实施例提供了当咬尾卷积编码器如图 2所示，编码率为 1/3时， 本发明的 Viterbi译码方法的具体实施方案。 TECHNICAL FIELD The present invention relates to the field of communications, and in particular to a channel decoding method and apparatus. BACKGROUND In a wireless communication network, a spatial channel is erratic as a sea surface, and the signal is often distorted, thereby increasing the bit error rate of the transmitted data. In order to reduce the bit error rate, it is necessary to transmit data with certain anti-interference ability. The commonly used method is to encode the transmission data in some way and add redundant protection information, which is channel coding. The channel coding methods used by the 3rd Generation Partnership Project (3GPP) mainly include convolutional codes, turbo codes, and interleaving. The convolutional code is relatively simple compared to the turbo code, and the convolutional coding is generally used in the case of a short code, for example, a broadcast channel in a Long-Term Evolution (LTE). Abbreviated as BCH), downlink control information (Dlinking Control Information, DCI for short) and Uplink Control Information (UCI) are encoded by convolutional coding. In order to increase the code rate of the convolutional code, a tail-biting convolutional encoder is used in LTE, and the structure is shown in Fig. 2. The initial value of the encoder's status register is set to the corresponding value of the last 6 information bits of the input data stream, so that the initial and final states of the shift register are the same. However, because the initial state and the end state are uncertain, the performance of the conventional Viterbi decoding method is degraded. The inventors have found that the related art has at least the following problems: When calculating branch metrics, methods such as Hamming metrics and Euclidean metrics are generally used, which are the shortest measures to make the distance, but the metrics and their backtracking The method wastes a lot of resources and the calculation speed is slow. SUMMARY OF THE INVENTION The present invention is directed to a channel decoding method and apparatus to solve the problem of waste of resources and slow calculation speed in the prior art. According to an aspect of the present invention, a channel decoding method is provided, including: calculating a bite roll The correlation value of the output value of all the states of the product encoder and the data to be decoded is used as the branch metric value; according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and a larger one is selected. The accumulated value is used as a new metric value of the next state, and the selection result is saved until the data to be decoded ends; according to the selection result, the backtracking is started from the state where the final metric value is the largest, and the decoding is obtained. Preferably, according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded The data end includes: accumulating the metric values of the corresponding two states and the branch metrics according to the state transition diagram; according to the addition ratio rule, selecting a larger accumulated value as the new metric value of the next state, and The selection result is saved until the data to be decoded ends; the state in which the final metric value is the largest is saved. Preferably, the backtracking is started from the state in which the final metric value is the largest according to the selection result. The decoding specifically includes: backtracking from the state with the largest metric value, and finding the parent state according to the selection result until all the historical states are found; The lowest bit of the history status output is used as the decoded output bit. Preferably, according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric values, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the translation is to be translated. Before the end of the code data, the method further includes: saving the data to be decoded; accumulating the metric values of the corresponding two states and the branch metric respectively, and selecting a larger accumulated value as the new metric value of the next state until the The decoded data is saved, and the state with the largest metric value is saved as the first termination state. Preferably, according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded After the end of the data, the method further includes: if the state of the final metric value is different from the first termination state, and the maximum number of parameters is not reached, the decoding process is continued. Preferably, when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, the number of all states is 64, and the number of different branch metrics is 8. According to the state transition diagram, corresponding The metrics of the two states are respectively accumulated with the branch metrics, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the end of the data to be decoded includes: According to the state transition map, The metric values of the corresponding two states are respectively accumulated with the branch metric values; according to the addition ratio rule, the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends; Among them, if you select the above path, the result will be 0, otherwise it will be 1. Preferably, of the saved data to be decoded, three soft bits exist in parallel in one address. According to another aspect of the present invention, there is also provided a channel decoding apparatus, comprising: a branch metric value calculation module, configured to calculate a correlation value of an output value of all states of a tail-biting convolutional encoder and data to be decoded As a branch metric; a comparison selection module, configured to accumulate the metric values of the corresponding two states and the branch metric according to the state transition graph, and select a larger accumulated value as the new metric value of the next state. The selection result is saved until the data to be decoded ends; the backtracking module is configured to perform backtracking according to the selection result from the state with the largest metric value to obtain decoding. Preferably, the comparison selection module is configured to accumulate the metric value of the corresponding state and the branch metric value according to the state transition diagram; according to the addition ratio selection rule, select a larger accumulated value as the new metric value of the next state, and The selection result is saved until the data to be decoded ends; the state in which the final metric value is the largest is saved. Preferably, the backtracking module is configured to backtrack from the state with the largest metric value, and find the parent state according to the selection result until all the history states are found; the lowest bit of the historical state output is used as the decoding output bit. Preferably, in the foregoing apparatus, the method further includes: a saving module, configured to save data to be decoded; and the adding and selecting module is further configured to accumulate the metric values of the corresponding two states and the branch metric respectively, and select The larger accumulated value is used as the new metric value of the next state until the data to be decoded is saved, and the state with the largest metric value is saved as the first termination state. Preferably, in the foregoing apparatus, the method further includes: a determining module, configured to determine that if the state in which the final metric value is the largest is different from the first termination state, and the maximum number of parameters is not reached, the decoding process is continued. Preferably, when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, the number of all states is 64, and the number of different branch metrics is 8. The comparison and selection module is used for: According to the state transition diagram, the metric value of the corresponding state is accumulated with the branch metric value; according to the addition ratio rule, the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded End; where, if the above path is selected, the result of the selection is marked as 0, otherwise it is 1. Preferably, of the saved data to be decoded, three soft bits exist in parallel in one address. Since the state correlation value is used as the branch metric, the record comparison result is used as the backtracking roadmap, The lowest bit of the backtracking state is used as the decoding output, which solves the problem of wasteful resources and slow calculation speed in the prior art, so that the decoding consumes less resources and the calculation speed is fast. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are set to illustrate,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In the drawings: FIG. 1 is a flow chart showing a channel decoding method according to an embodiment of the present invention; FIG. 2 is a schematic diagram showing a tail biting convolutional encoder in accordance with a preferred embodiment of the present invention; A schematic diagram of a tail biting convolution grid in accordance with a preferred embodiment of the present invention; FIG. 4 is a schematic illustration of a full state transition of a tail biting convolution in accordance with a preferred embodiment of the present invention; FIG. 5 illustrates an addition in accordance with a preferred embodiment of the present invention. Figure 6 shows a schematic diagram of a backtracking decoding output in accordance with a preferred embodiment of the present invention; Figure 7 shows a flow chart of a Viterbi decoding method in accordance with a preferred embodiment of the present invention; A flowchart of a single-pass Viterbi decoding method according to a preferred embodiment 2 of the present invention; FIG. 9 is a diagram showing a channel decoding apparatus according to an embodiment of the present invention. BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail with reference to the accompanying drawings in conjunction with the embodiments. 1 is a flow chart showing a channel decoding method according to an embodiment of the present invention, including the following steps: Step S10, calculating an output value of all states of a tail-biting convolutional encoder related to data to be decoded The value is used as the branch metric value; Step S20, according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection is saved. As a result, until the data to be decoded ends; in step S30, the backtracking is started from the state in which the final metric value is the largest according to the selection result, and the decoding is obtained. In this embodiment, since the state correlation value is used as the branch metric, the record comparison and selection result is used as the backtracking roadmap, and the lowest bit of the backtracking state is regarded as the decoding output, which solves the problem that the prior art wastes resources and the calculation speed is slow. . The use of this embodiment makes the decoded occupied resources less and the calculation speed is fast. Preferably, the step S20 specifically includes: summarizing the metric values of the corresponding two states and the branch metrics according to the state transition map; according to the adding comparison rule, selecting a larger accumulated value as the new state of the next state The metric, and save the selection until the end of the data to be decoded; save the state with the largest metric. Step S30 specifically includes: backtracking from the state with the largest metric value, finding the parent state according to the selection result, until all the history states are found; and the lowest bit output by the history state is used as the decoding output bit. The preferred embodiment provides an accumulation of state metrics according to a comparison selection rule, and saves the state of the addition comparison result and the final maximum state metric value, and finally backtracks from the state according to the result of the comparison selection. Specific embodiment. Preferably, before step S20, the method further includes: saving data to be decoded; accumulating the metric values of the corresponding two states and the branch metric respectively, and selecting a larger accumulated value as the new metric value of the next state Until the data to be decoded is saved, the state with the largest metric value is saved as the first termination state. After the step S20, the method further includes: if the state in which the final metric value is the largest is different from the first termination state, and the maximum number of turns is not reached, the decoding process is continued. Thus, in order to adapt to the characteristics of the tail-biting convolution, the state value of the initial state is initialized in the process of saving the data to be decoded, thereby reducing the influence of the initial state uncertainty of the tail-biting convolutional code. In addition, since the raster image of the tail-biting convolution is cyclically cycled as shown in FIG. 3, in order to improve performance, the decoding can be continued until the initial and final states are matched or the maximum number of cycles is reached, which is improved. The performance of the decoding. Preferably, when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, all states are 64, and different branch metrics have 8. According to the state transition diagram, the corresponding two The metrics of the states are respectively accumulated with the branch metrics, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the end of the data to be decoded includes: According to the state transition map, the corresponding The metrics of the two states are respectively accumulated with the branch metrics; according to the addition ratio rule, the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends; If you select one of the above paths, the result of the selection is marked as 0, otherwise it is 1. Preferably, of the saved data to be decoded, three soft bits exist in parallel in one address. In this way, by using the parallel storage method, the three soft bits can be simultaneously read and correlated with the state code output bits. The above preferred embodiment provides a specific embodiment of the Viterbi decoding method of the present invention when the tail biting convolutional encoder is as shown in Fig. 2 with a coding rate of 1/3.
Viterbi译码的主要精髓就是最大似然，即找出所有可能的编码组合中与 待译码数据最相似的一组， 因为编码组合与编码数据长度是指数冪的关系， 因此完全匹配编码组合是不可能的， Viterbi算法才艮据编码器的状态转移图对 到达同一状态的两组状态度量值进行比较，选择最似然的状态，丢弃另一只， 即所谓的加比选， 因此保证了存留的状态量总是等于编码器的状态量， 从而 使译码量计算量可以接受。 全状态转移示意图如图 4所示。 具体方法:^下： 首先， 根据咬尾编码器的结构 （如图 2所示）， 得出所有的状态的输出 值并保存， 因为编码器的长度为 6个寄存器， 因此共有 64个状态， 每个状 态输出 3个比特， 因此状态编码输出表的大小为 64x3。 然后， 将状态的输出值与待译码数据做相关， 求出 64个相关值作为分 支度量值。 状态编码输出比特可以表示为： a(n,m)a, 其中 n表示状态值， m 表示比特序号。 在实际应用中， 0表示正、 1表示负。 例如： 状态 32时， 输 出比特为 011 , 那么与待码数据的相关值就是 +D0-D 1-D2。 这样就可以得出 64个状态的相关值。 实际上因为是 1/3码率， 所以最多只有 8种不同的相关 值， 因此实现时只需要计算 8个即可。 才艮据图 5所示的加比选状态转移图，可以知道下一个状态总是从前两个 状态之一加比选出来的， 即选择前一个状态的度量值与分支度量值的累加值 更大的那个作为该状态的度量值， 并记录下比较结果， 即如果选择的是上面 的路径， 则记为 0 , 否则记为 1 , 保存每一个状态的比较记录。 因此， 最后 保存的是 Κχ64个回溯结果， 其中 K为译码的长度， 总共有 64个状态。 回 溯时， 从度量值最大的状态点作为回溯点， 根据该状态记录的比较结果算出 上一个状态， 以此类推， 得到所有的历史状态。 回溯方法是图 5的反过程， 即根据比较结果推断上一个状态是当前状态右移一位还是需要再加 32。 因为状态的变化与输入比特相关，即状态的最低比特就是输入的编码数 据， 因此， 直接将回溯的历史状态的最氐比特当作译码输出比特即可。 例如 初始^ ί态为 2 , 输入比特为 1 , 则^！ 态变为 5 , 那么回溯时， 如果发现回溯到 状态 5 , 那么就知道这个时候输入的数据是 1 , 因此译码输出就是 1。 译码输 出与状态回溯的关系图 6所示， 这是一个右移过程， 可以看出译码输出可以 看作是编码输入的反过程。 上述优选实施例的 Viterbi译码方法不但可以应用于咬尾卷积， 而且也 可应用于归零卷积， 其特点是占用资源少、 速度快。 此外， 为了应用于咬尾卷积， 本发明的优选实施例还提出了以下几种方 法： 首先， 在存入待译码数据时， 同时计算状态的度量值， 但不必保存加比 选结果。 因为咬尾， 所以当待译码数据保存完毕时， 就得到了开始译码的初 始状态值。 这样， 可以一边存待译码数据、 一边译码， 可以使速度更快。 其次， 保存初始状态中最大状态度量的状态值， 当译码结束时， 与终止 状态的最大状态度量的状态值比较， 如果相同， 停止译码， 否则继续循环译 码， 直到上述条件满足或者到达最大循环次数。 这样利用咬尾卷积的特点， 能够保证译码的准确性， 因为如果不同则译码一定不准确。 最后， 因为咬尾卷积可能要循环多次， 可以釆用并行计算的方法， 即一 步计算所有 64个状态的状态度量， 也需要同时计算 64个分支度量值。 为了 加快计算分支度量值的速度， 在存储待译码数据时釆用并行存储方式， 即 3 个软比特并行存在一个地址， 因此可以同时读出， 与状态编码输出比特进行 相关计算。 图 7示出了根据本发明优选实施例一的 Viterbi译码方法的流程图， 包 括以下步 4聚： 步骤 S 102, 在译码前， 建立状态编码输出表， 可以根据咬尾编码器的 结构， 得到所有可能状态的输出值； 步骤 S 104, 保存待译码数据， 将待译码数据的 3个软比特并行存储在 一个地址中， 同时根据加比选法则， 进行状态度量值的累加， 不保存加比选 结果， 直到数据保存完毕， 保存最大度量值的状态值； 步骤 S 106, 读出保存的待译码数据根据加比选法则， 进行状态度量值 的累加， 同时保存加比选结果， 用于译码结束时的回溯， 直到一圏结束， 保 存最大度量值的状态值； 步骤 S 108, 如果终止状态与上一次的终止状态不同且没达到最大圏数 则返回步 4聚 S 106; 步骤 S 110, 否则， 从度量值最大的状态开始回溯。 图 8示出了根据本发明优选实施例二的单圏的 Viterbi译码方法的流程 图， 包括以下步 4聚： 步骤 S202, 在译码前， 建立状态编码输出表； 步骤 S204，将 K时刻的待译码数据与所有状态的编码输出值进行相关， 记下所有的相关值作为分支度量值； 步骤 S206, 根据状态转移图 （如图 4所示）， 将对应状态的度量值与分 支度量值累加， 并将累加结果作为下一状态的度量值， 因为根据状态转移图 下一时刻的状态来自两个状态， 所以利用 "加比选，，方法选择其中一条累加值 较大的结果保存下来， 并将比较结果记录下来用于历史回溯； 步骤 S208, 判断待译码数据是否结束， 若否， 则返回步骤 S202; 步骤 S210, 若是， 则寻找最大状态度量值的状态值； 步骤 S212, 由该状态进行回溯， 并根据保存的比较结果的记录， 寻找 父状态， 直到找出所有历史状态； 步骤 S214, 将状态历史输出的最低位作为译码输出比特。 图 9示出了 居本发明实施例的信道译码装置的示意图， 该装置包括: 分支度量值计算模块 10 ,用于计算咬尾卷积编码器的所有状态的输出值与待 译码的数据的相关值作为分支度量值；加比选模块 20 ,用于根据状态转移图， 将对应的两个状态的度量值分别与分支度量值进行累加， 并选择较大的累加 值作为下一状态的新的度量值， 保存选择结果， 直到待译码数据结束； 回溯 模块 30 ,用于根据选择结果从最终的度量值最大的状态开始回溯，得到译码。 该实施例由于釆用状态相关值作为分支度量，记录加比选结果作为回溯 路标， 将回溯状态的最低比特当作译码输出， 解决了现有技术存在的浪费资 源， 计算速度较慢的问题。 使用该实施例使得译码的占用资源少、 计算速度 快。 优选地， 加比选模块 20用于根据状态转移图， 将对应的两个状态的度 量值分别与分支度量值进行累加； 按照加比选法则， 选择较大的累加值作为 下一状态的新的度量值， 并保存选择结果， 直到待译码数据结束； 保存最终 的度量值最大的状态。 回溯模块 30用于由度量值最大的状态进行回溯， 根据选择结果寻找父 状态， 直到找出所有的历史状态； 将历史状态输出的最低比特作为译码输出 比特。 该优选实施例提供了根据加比选法则进行状态度量值的累加，并保存加 比选结果以及最终的最大状态度量值的状态， 最后根据加比选的结果由该状 态进行回溯得到译码的具体实施方案。 优选地， 在上述的装置中， 还包括： 保存模块 40 , 用于保存待译码的 数据； 加比选模块 20 还用于将对应状态的度量值与分支度量值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 直到待译码的数据保存完 毕， 保存度量值最大的状态作为第一次的终止状态； 判断模块 50 , 用于判断 如果最终的度量值最大的状态与第一次的终止状态不同， 且未达到最大的圏 数， 则继续执行译码过程。 这样， 为了适应咬尾卷积的特点， 在保存待译码数据的过程中， 初始化 了初始状态的状态值， 因此减少了咬尾卷积码初始状态不确定的影响。 另夕卜， 由于咬尾卷积的栅格图如图 3所示是周期循环的， 因此为了提高性能， 译码 可以一直循环进行下去， 直到初始与终止状态吻合或者达到最大循环次数， 提高了译码的性能。 优选地， 当咬尾卷积编码器的编码率为 1/3 , 长度为 6个寄存器时， 所 有状态为 64个， 不同的分支度量值有 8个， 加比选模块用于根据状态转移 图， 将对应状态的度量值与分支度量值进行累加； 按照加比选法则， 选择较 大的累加值作为下一状态的新的度量值， 并保存选择结果， 直到待译码数据 结束； 其中， 如果选择的是上面的一条路径， 则将选择结果记为 0, 否则记 为 1。 优选地， 保存的待译码的数据中， 3个软比特并行存在一个地址中。 这 样， 釆用并行存储的方式， 可以将该 3个软比特同时读出， 与状态编码输出 比特进行相关计算。 上述优选实施例提供了当咬尾卷积编码器如图 2所示，编码率为 1/3时， 本发明的 Viterbi译码方法的具体实施方案。 从以上的描述中， 可以看出， 本发明实现了如下技术效果： 由于釆用状 态相关值作为分支度量， 记录加比选结果作为回溯路标， 将回溯状态的最低 比特当作译码输出，解决了现有技术存在的浪费资源，计算速度较慢的问题。 使用该实施例使得译码的占用资源少、 计算速度快。 显然， 本领域的技术人员应该明白， 上述的本发明的各模块或各步骤可 以用通用的计算装置来实现， 它们可以集中在单个的计算装置上， 或者分布 在多个计算装置所组成的网络上， 可选地， 它们可以用计算装置可执行的程 序代码来实现， 从而， 可以将它们存储在存储装置中由计算装置来执行， 或 者将它们分别制作成各个集成电路模块， 或者将它们中的多个模块或步骤制 作成单个集成电路模块来实现。 这样， 本发明不限制于任何特定的硬件和软 件结合。 以上所述仅为本发明的优选实施例而已， 并不用于限制本发明， 对于本 领域的技术人员来说， 本发明可以有各种更改和变化。 凡在本发明的 ^"神和 原则之内， 所作的任何修改、 等同替换、 改进等， 均应包含在本发明的保护 范围之内。 The main essence of Viterbi decoding is the maximum likelihood, that is, to find the most similar set of all possible coding combinations with the data to be decoded, because the coding combination and the length of the encoded data are exponential power, so the exact match coding combination is Impossible, the Viterbi algorithm compares the two sets of state metrics arriving at the same state according to the state transition diagram of the encoder, selects the most likely state, discards the other, the so-called plus comparison, thus guaranteeing The amount of state remaining is always equal to the state quantity of the encoder, so that the amount of decoding calculation is acceptable. A schematic diagram of the full state transition is shown in Figure 4. Specific method: ^下: First, according to the structure of the tail bit encoder (as shown in Figure 2), the output values of all states are obtained and saved. Because the length of the encoder is 6 registers, there are 64 states. Each state outputs 3 bits, so the size of the state code output table is 64x3. Then, the output value of the state is correlated with the data to be decoded, and 64 correlation values are obtained as branch metric values. The status coded output bit can be expressed as: a(n,m)a, where n represents the state value and m represents the bit number. In practical applications, 0 means positive and 1 means negative. For example: In state 32, the output bit is 011, then the correlation value with the data to be coded is +D0-D 1-D2. This will give you the relevant values for the 64 states. In fact, because it is 1/3 code rate, there are only 8 different correlation values at most, so only 8 calculations are needed for implementation. According to the addition-selection state transition diagram shown in Fig. 5, it can be known that the next state is always selected from one of the first two states, that is, the metric value of the previous state and the accumulated value of the branch metric value are selected. The larger one is used as a measure of the state, and the result of the comparison is recorded, ie if the selection is above The path is marked as 0, otherwise it is marked as 1, and the comparison record for each state is saved. Therefore, the last saved is 64 backtracking results, where K is the length of the decoding, and there are a total of 64 states. When backtracking, the state point with the largest metric value is used as the backtracking point, the previous state is calculated according to the comparison result of the state record, and so on, and all the historical states are obtained. The backtracking method is the inverse of FIG. 5, that is, it is inferred from the comparison result that the previous state is the current state shifted to the right by one bit or 32. Since the change of state is related to the input bit, that is, the lowest bit of the state is the input encoded data, therefore, the last bit of the backtracked history state is directly regarded as the decoded output bit. For example, if the initial ^ ̄ state is 2 and the input bit is 1, then ^! The state becomes 5, then when backtracking, if it is found back to state 5, then it is known that the data input at this time is 1, so the decoding output is 1. The relationship between the decoded output and the state backtracking is shown in Figure 6. This is a right shifting process. It can be seen that the decoded output can be seen as the inverse of the encoded input. The Viterbi decoding method of the above preferred embodiment can be applied not only to tail-biting convolution but also to zero-revolution, which is characterized by low resource consumption and high speed. In addition, in order to apply to tail biting convolution, the preferred embodiment of the present invention also proposes the following methods: First, when storing data to be decoded, the metric value of the state is simultaneously calculated, but it is not necessary to save the comparison result. Because of the tail biting, when the data to be decoded is saved, the initial state value of starting decoding is obtained. In this way, the data can be decoded while being decoded, and the speed can be made faster. Secondly, the state value of the maximum state metric in the initial state is saved. When the decoding ends, the state value of the maximum state metric of the termination state is compared. If it is the same, the decoding is stopped, otherwise the loop decoding is continued until the above condition is met or arrived. The maximum number of cycles. This makes use of the characteristics of the tail-biting convolution to ensure the accuracy of the decoding, because if it is different, the decoding must be inaccurate. Finally, because the tail-biting convolution may have to be repeated multiple times, the parallel computation method can be used, that is, the state metrics of all 64 states are calculated in one step, and the 64 branch metrics need to be calculated simultaneously. In order to speed up the calculation of the branch metric value, the parallel storage mode is used when storing the data to be decoded, that is, three soft bits have one address in parallel, so that it can be read out at the same time and correlated with the state code output bit. FIG. 7 is a flowchart of a Viterbi decoding method according to a preferred embodiment of the present invention, including the following steps: Step S102, before decoding, establishing a state code output table, and obtaining output values of all possible states according to the structure of the tail bit encoder; Step S104, saving data to be decoded, and 3 data to be decoded The soft bits are stored in parallel in an address, and the state metrics are accumulated according to the addition ratio rule, and the comparison result is not saved until the data is saved, and the state value of the maximum metric value is saved; Step S106, read and save The data to be decoded is accumulated according to the addition selection rule, and the result of the comparison is saved, and the result of the comparison is saved for the backtracking at the end of the decoding, until the end of the decoding, the state value of the maximum metric value is saved; Step S108 If the termination state is different from the previous termination state and the maximum number of parameters is not reached, then return to step 4 to gather S 106; step S 110, otherwise, trace back from the state with the largest metric value. 8 is a flowchart of a Viterbi decoding method according to a preferred embodiment of the present invention, including the following steps: Step S202: Before decoding, establish a state code output table; Step S204, set K time The data to be decoded is correlated with the coded output values of all states, and all the correlation values are recorded as the branch metric values; Step S206, according to the state transition diagram (as shown in FIG. 4), the metric value and the branch metric of the corresponding state are determined. The values are accumulated, and the accumulated result is taken as the metric of the next state, because the state at the next moment according to the state transition diagram comes from two states, so the result of selecting one of the accumulated values is saved by using "add comparison". And recording the comparison result for historical backtracking; Step S208, determining whether the data to be decoded is finished, if not, returning to step S202; Step S210, if yes, searching for the state value of the maximum state metric value; Step S212, The state is backtracked, and according to the record of the saved comparison result, the parent state is searched until all history states are found; step S214, the state calendar is The output of the LSB bit decode output as FIG 9 shows a schematic embodiment UN channel decoding apparatus of the embodiment of the present invention, the apparatus comprising: The branch metric calculation module 10 is configured to calculate a correlation value of the output value of all states of the tail-biting convolutional encoder and the data to be decoded as a branch metric value; and the comparison and selection module 20 is configured to The metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends; the traceback module 30 is used for According to the selection result, the backtracking is started from the state where the final metric value is the largest, and the decoding is obtained. In this embodiment, since the state correlation value is used as the branch metric, the record comparison and selection result is used as the backtracking roadmap, and the lowest bit of the backtracking state is regarded as the decoding output, which solves the problem that the prior art wastes resources and the calculation speed is slow. . The use of this embodiment makes the decoded occupied resources less and the calculation speed is fast. Preferably, the comparison selection module 20 is configured to accumulate the metric values of the corresponding two states and the branch metric values according to the state transition diagram; according to the addition ratio selection rule, select a larger accumulated value as the new state of the next state. The metric, and save the selection until the end of the data to be decoded; save the state with the largest metric. The backtracking module 30 is configured to backtrack from the state with the largest metric value, find the parent state according to the selection result, until all the history states are found; the lowest bit of the historical state output is used as the decoding output bit. The preferred embodiment provides an accumulation of state metrics according to a comparison selection rule, and saves the state of the addition comparison result and the final maximum state metric value, and finally backtracks from the state according to the result of the comparison selection. Specific embodiment. Preferably, in the foregoing apparatus, the method further includes: a saving module 40, configured to save data to be decoded; and the comparison and selection module 20 is further configured to accumulate the metric value of the corresponding state and the branch metric value, and select a larger The accumulated value is used as the new metric value of the next state, until the data to be decoded is saved, and the state with the largest metric value is saved as the first termination state; the determining module 50 is configured to determine if the final metric value is the largest. The state is different from the first termination state, and the maximum number of turns is not reached, and the decoding process continues. Thus, in order to adapt to the characteristics of the tail-biting convolution, the state value of the initial state is initialized in the process of saving the data to be decoded, thereby reducing the influence of the initial state uncertainty of the tail-biting convolutional code. In addition, since the raster image of the tail-biting convolution is cyclically cycled as shown in FIG. 3, in order to improve performance, the decoding can be continued until the initial and final states are matched or the maximum number of cycles is reached, which is improved. The performance of the decoding. Preferably, when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, all states are 64, and different branch metrics are 8. The comparison and selection module is used to transfer the map according to the state. And accumulating the metric value of the corresponding state and the branch metric value; according to the addition ratio rule, selecting a larger accumulated value as the new metric value of the next state, and saving the selection result until the data to be decoded ends; If the above path is selected, the result of the selection is marked as 0, otherwise it is marked as 1. Preferably, of the saved data to be decoded, three soft bits exist in parallel in one address. In this way, by using the parallel storage method, the three soft bits can be simultaneously read and correlated with the state code output bits. The above preferred embodiment provides a specific embodiment of the Viterbi decoding method of the present invention when the tail biting convolutional encoder is as shown in Fig. 2 with a coding rate of 1/3. From the above description, it can be seen that the present invention achieves the following technical effects: Since the state correlation value is used as the branch metric, the record comparison and selection result is used as the backtracking roadmap, and the lowest bit of the backtracking state is regarded as the decoding output, and the solution is solved. The problem of wasted resources in the prior art and the calculation speed is slow. The use of this embodiment makes the decoded occupied resources less and the calculation speed is fast. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device, or they may be separately fabricated into individual integrated circuit modules, or they may be Multiple modules or steps are made into a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Claims

权 利 要 求 书  Claims
1. 一种信道译码方法， 其特征在于， 包括： A channel decoding method, comprising:
计算咬尾卷积编码器的所有状态的输出值与待译码的数据的相关 值作为分支度量值；  Calculating a correlation value of an output value of all states of the tail-biting convolutional encoder and data to be decoded as a branch metric value;
根据状态转移图，将对应的两个状态的度量值分别与所述分支度量 值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 保存选 择结果， 直到所述待译码的数据结束；  According to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the to-be-translated The end of the code data;
根据所述选择结果从最终的度量值最大的状态开始回溯， 得到译 码。  According to the selection result, the backtracking is started from the state in which the final metric value is the largest, and the decoding is obtained.
2. 根据权利要求 1所述的方法， 其特征在于， 根据状态转移图， 将对应的 两个状态的度量值分别与所述分支度量值进行累加， 并选择较大的累加 值作为下一状态的新的度量值， 保存选择结果， 直到所述待译码的数据 结束具体包括： 2. The method according to claim 1, wherein, according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the next state. The new metric value, saving the selection result until the end of the data to be decoded specifically includes:
根据状态转移图，将所述对应的两个状态的度量值分别与所述分支 度量值进行累加；  And adding, according to the state transition graph, the metric values of the corresponding two states to the branch metric value respectively;
按照加比选法则，选择较大的累加值作为所述下一状态的新的度量 值， 并保存选择结果， 直到所述待译码的数据结束；  According to the additive selection rule, a larger accumulated value is selected as the new metric value of the next state, and the selection result is saved until the data to be decoded ends;
保存所述最终的度量值最大的状态。  Save the state in which the final metric is the largest.
3. 根据权利要求 1所述的方法， 其特征在于， 根据所述选择结果从最终的 度量值最大的状态开始回溯， 得到译码具体包括： The method according to claim 1, wherein the backtracking is started from the state in which the final metric value is the largest according to the selection result, and the decoding specifically includes:
由所述最终的度量值最大的状态进行回溯 ,根据所述选择结果寻找 父状态， 直到找出所有的历史状态；  Backtracking is performed by the state in which the final metric value is the largest, and the parent state is searched according to the selection result until all the history states are found;
将所述历史状态输出的最低比特作为译码输出比特。  The lowest bit of the history state output is used as a decoded output bit.
4. 根据权利要求 1所述的方法， 其特征在于， 在根据状态转移图， 将对应 的两个状态的度量值分别与所述分支度量值进行累加， 并选择较大的累 加值作为下一状态的新的度量值， 保存选择结果， 直到所述待译码的数 据结束之前还包括： The method according to claim 1, wherein, according to the state transition map, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and a larger accumulated value is selected as the next A new metric of the state, saving the selection result until the end of the data to be decoded includes:
保存所述待译码的数据； 将所述对应的两个状态的度量值分别与所述分支度量值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 直到所述待译码的数 据保存完毕， 保存度量值最大的状态作为第一次的终止状态。 根据权利要求 4所述的方法， 其特征在于， 根据状态转移图， 将对应的 两个状态的度量值分别与所述分支度量值进行累加， 并选择较大的累加 值作为下一状态的新的度量值， 保存选择结果， 直到所述待译码的数据 结束之后还包括： Saving the data to be decoded; And respectively adding the metric values of the corresponding two states to the branch metric value, and selecting a larger accumulated value as a new metric value of the next state, until the data to be decoded is saved, and saved. The state with the largest metric is the first termination state. The method according to claim 4, wherein, according to the state transition map, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and the larger accumulated value is selected as the new state of the next state. The metric value, the selection result is saved, and after the end of the data to be decoded, the method further includes:
如果所述最终的度量值最大的状态与所述第一次的终止状态不同， 且未达到最大的圏数， 则继续执行译码过程。 根据权利要求 5所述的方法， 其特征在于， 当所述咬尾卷积编码器的编 码率为 1/3 , 长度为 6个寄存器时， 所述所有状态的个数为 64个， 不同 的所述分支度量值有 8个，  If the state in which the final metric value is the largest is different from the first termination state, and the maximum number of turns is not reached, the decoding process continues. The method according to claim 5, wherein when the coding rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, the number of all the states is 64, different There are 8 branch metrics.
根据状态转移图，将所述对应的两个状态的度量值分别与所述分支 度量值进行累加， 并选择较大的累加值作为下一状态的新的度量值， 保 存选择结果， 直到所述待译码的数据结束包括：  And according to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value, and a larger accumulated value is selected as a new metric value of the next state, and the selection result is saved until the The end of the data to be decoded includes:
根据状态转移图，将对应的两个状态的度量值分别与所述分支度量 值进行累加；  According to the state transition diagram, the metric values of the corresponding two states are respectively accumulated with the branch metric value;
按照加比选法则， 选择较大的累加值作为下一状态的新的度量值， 并保存选择结果， 直到所述待译码的数据结束；  According to the additive selection rule, a larger accumulated value is selected as a new metric value of the next state, and the selection result is saved until the data to be decoded ends;
其中， 如果选择的是上面的一条路径， 则将所述选择结果记为 0 , 否则 ΐ己为 1。 根据权利要求 6所述的方法， 其特征在于， 保存的所述待译码的数据中， 3个软比特并行存在一个地址中。 一种信道译码装置， 其特征在于， 包括：  Wherein, if the above path is selected, the selection result is recorded as 0, otherwise ΐ is 1. The method according to claim 6, wherein among the data to be decoded stored, three soft bits exist in parallel in one address. A channel decoding device, comprising:
分支度量值计算模块，用于计算咬尾卷积编码器的所有状态的输出 值与待译码的数据的相关值作为分支度量值；  a branch metric calculation module, configured to calculate a correlation value of an output value of all states of the tail-biting convolutional encoder and data to be decoded as a branch metric value;
加比选模块， 用于根据状态转移图， 将对应的两个状态的度量值分 别与所述分支度量值进行累加， 并选择较大的累加值作为下一状态的新 的度量值， 保存选择结果， 直到所述待译码的数据结束； 回溯模块，用于根据所述选择结果从最终的度量值最大的状态开始 回溯， 得到译码。 The ratio selection module is configured to accumulate the metric values of the corresponding two states and the branch metrics according to the state transition graph, and select a larger accumulated value as a new metric value of the next state, and save the selection. As a result, until the data to be decoded ends; The backtracking module is configured to perform backtracking from the state in which the final metric value is the largest according to the selection result, to obtain decoding.
9. 根据权利要求 8所述的装置， 其特征在于， 所述加比选模块用于： 9. The apparatus according to claim 8, wherein the ratio selection module is configured to:
根据状态转移图，将所述对应的两个状态的度量值分别与所述分支 度量值进行累加；  And adding, according to the state transition graph, the metric values of the corresponding two states to the branch metric value respectively;
按照加比选法则， 选择较大的累加值作为下一状态的新的度量值， 并保存选择结果， 直到所述待译码的数据结束；  According to the additive selection rule, a larger accumulated value is selected as a new metric value of the next state, and the selection result is saved until the data to be decoded ends;
保存所述最终的度量值最大的状态。  Save the state in which the final metric is the largest.
10. 根据权利要求 8所述的装置， 其特征在于， 所述回溯模块用于： 由所述度量值最大的状态进行回溯， 根据所述选择结果寻找父状 态， 直到找出所有的历史状态； The apparatus according to claim 8, wherein the backtracking module is configured to: perform backtracking from a state in which the metric value is the largest, and search for a parent state according to the selection result until all history states are found;
将所述历史状态输出的最低比特作为译码输出比特。  The lowest bit of the history state output is used as a decoded output bit.
11. 根据权利要求 8所述的装置， 其特征在于， 还包括： 11. The device according to claim 8, further comprising:
保存模块， 用于保存所述待译码的数据；  a saving module, configured to save the data to be decoded;
所述加比选模块还用于将所述对应的两个状态的度量值分别与所 述分支度量值进行累加， 并选择较大的累加值作为下一状态的新的度量 值， 直到所述待译码的数据保存完毕， 保存度量值最大的状态作为第一 次的终止^! 态。  The ratio selection module is further configured to accumulate the metric values of the corresponding two states and the branch metric value respectively, and select a larger accumulated value as a new metric value of the next state, until the After the data to be decoded is saved, the state with the largest metric value is saved as the first termination state.
12. 根据权利要求 11所述的装置， 其特征在于， 还包括： 12. The device according to claim 11, further comprising:
判断模块，用于判断如果所述最终的度量值最大的状态与所述第一 次的终止状态不同， 且未达到最大的圏数， 则继续执行译码过程。  And a judging module, configured to determine that if the state in which the final metric value is the largest is different from the first termination state, and the maximum number of parameters is not reached, the decoding process is continued.
13. 根据权利要求 12所述的装置， 其特征在于， 当所述咬尾卷积编码器的编 码率为 1/3 , 长度为 6个寄存器时， 所述所有状态为 64个， 不同的所述 分支度量值有 8个， 13. The apparatus according to claim 12, wherein when the code rate of the tail-biting convolutional encoder is 1/3 and the length is 6 registers, all of the states are 64, different places. There are 8 branch metrics.
所述加比选模块用于：  The addition ratio module is used to:
根据状态转移图，将所述对应的两个状态的度量值分别与所述分支 度量值进行累加； 按照加比选法则， 选择较大的累加值作为下一状态的新的度量值， 并保存选择结果， 直到所述待译码的数据结束； And accumulating the metric values of the corresponding two states and the branch metric values according to the state transition diagram; According to the additive selection rule, a larger accumulated value is selected as a new metric value of the next state, and the selection result is saved until the data to be decoded ends;
其中， 如果选择的是上面的一条路径， 则将所述选择结果记为 0 , 否则 ΐ己为 1。 根据权利要求 13所述的装置，其特征在于，保存的所述待译码的数据中 , 3个软比特并行存在一个地址中。  Wherein, if the above path is selected, the selection result is recorded as 0, otherwise ΐ is 1. The apparatus according to claim 13, wherein among the data to be decoded stored, three soft bits exist in parallel in one address.
PCT/CN2010/071993 2009-09-02 2010-04-21 Method and device for channal decoding WO2011026330A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200910168950.4 2009-09-02
CN 200910168950 CN101635611B (en) 2009-09-02 2009-09-02 Channel decoding method and channel decoding device

Publications (1)

Publication Number Publication Date
WO2011026330A1 true WO2011026330A1 (en) 2011-03-10

Family

ID=41594682

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/071993 WO2011026330A1 (en) 2009-09-02 2010-04-21 Method and device for channal decoding

Country Status (2)

Country Link
CN (1) CN101635611B (en)
WO (1) WO2011026330A1 (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101635611B (en) * 2009-09-02 2013-06-05 中兴通讯股份有限公司 Channel decoding method and channel decoding device
CN102291198B (en) * 2010-06-18 2014-08-20 中兴通讯股份有限公司 Channel decoding method and device
CN102377438B (en) * 2010-08-11 2013-12-04 中兴通讯股份有限公司 Channel decoding method and tail biting convolutional decoder
CN102404011B (en) * 2010-09-15 2015-05-20 中兴通讯股份有限公司 Method and device for achieving Viterbi decoding
CN102891690B (en) * 2011-07-19 2017-02-22 上海无线通信研究中心 Tail-biting convolution code decoding method
WO2012163135A1 (en) * 2011-05-27 2012-12-06 上海无线通信研究中心 Channel decoding method and decoder
CN102857233B (en) * 2011-06-28 2017-03-15 中兴通讯股份有限公司 Interpretation method and device
CN102857242B (en) * 2011-06-28 2016-03-02 联芯科技有限公司 Tail-biting convolutional code interpretation method and device
CN103248465B (en) * 2012-02-01 2016-06-08 联芯科技有限公司 A kind of terminal processing device and terminal processing method
CN105763286A (en) * 2014-12-15 2016-07-13 深圳市中兴微电子技术有限公司 Method and device for identifying wrong detection control information
CN106301391B (en) * 2016-08-08 2019-07-16 西安电子科技大学 A kind of improved soft output tail-biting convolutional code interpretation method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6256764B1 (en) * 1997-11-26 2001-07-03 Nortel Networks Limited Method and system for decoding tailbiting convolution codes
US20060242531A1 (en) * 2005-04-26 2006-10-26 Navini Networks, Inc. Method for decoding tail-biting convolutional codes
CN101369817A (en) * 2008-09-27 2009-02-18 华为技术有限公司 Interpretation method and apparatus for tail-biting convolutional code
CN101635611A (en) * 2009-09-02 2010-01-27 中兴通讯股份有限公司 Channel decoding method and channel decoding device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6256764B1 (en) * 1997-11-26 2001-07-03 Nortel Networks Limited Method and system for decoding tailbiting convolution codes
US20060242531A1 (en) * 2005-04-26 2006-10-26 Navini Networks, Inc. Method for decoding tail-biting convolutional codes
CN101369817A (en) * 2008-09-27 2009-02-18 华为技术有限公司 Interpretation method and apparatus for tail-biting convolutional code
CN101635611A (en) * 2009-09-02 2010-01-27 中兴通讯股份有限公司 Channel decoding method and channel decoding device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JIANG, HUI: "Design and Implementation for Symbol Timing and Viterbi Decoding in Digital Communication Systems", NANJING UNIVERSITY OF SCIENCE AND TECHNOLOGY, 31 January 2007 (2007-01-31), pages 50 - 54 *

Also Published As

Publication number Publication date
CN101635611A (en) 2010-01-27
CN101635611B (en) 2013-06-05

Similar Documents

Publication Publication Date Title
WO2011026330A1 (en) Method and device for channal decoding
US6934321B2 (en) W-CDMA transmission rate estimation method and device
JPH07273813A (en) Method and apparatus for generating soft symbol
CN107911195B (en) CVA-based tail-biting convolutional code channel decoding method
JP4806673B2 (en) Decoding device and decoding method
KR102547476B1 (en) Method for controlling decoding process based on path metric value and computing apparatus and mobile device for controlling the same
US8904266B2 (en) Multi-standard viterbi processor
CN114553372B (en) Decoding method and device for digital signals and electronic equipment
US8843811B2 (en) Apparatus and method for decoding in communication system
WO2012034398A1 (en) Method and device for implementing viterbi decoding
KR100853139B1 (en) Transport format detecting apparatus and method
JP4922242B2 (en) Encoding device, encoding method, and Viterbi decoding device
CN114430279B (en) List Viterbi decoding method, device, decoder and storage medium
CN102291198B (en) Channel decoding method and device
JP4286274B2 (en) Error correction device
EP2605410B1 (en) Channel decoding method and tail biting convolutional decoder
CN114499548B (en) Decoding method, device and storage medium
US7797618B2 (en) Parallel decoder for ultrawide bandwidth receiver
JP4228795B2 (en) TF determination method, TF determination method, and program
CN117353755A (en) Decoding method, decoding device, decoding equipment and computer readable storage medium
Hulyalkar et al. Hard and Soft Decision Decoding Using Viterbi Algorithm
US8503585B2 (en) Decoding method and associated apparatus
CN116192155A (en) Viterbi decoding circuit with low time delay and automatic length adjustment of surviving path

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: 10813252

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: 10813252

Country of ref document: EP

Kind code of ref document: A1