WO2011095115A1 - 一种解交织方法和装置 - Google Patents

一种解交织方法和装置 Download PDF

Info

Publication number
WO2011095115A1
WO2011095115A1 PCT/CN2011/070764 CN2011070764W WO2011095115A1 WO 2011095115 A1 WO2011095115 A1 WO 2011095115A1 CN 2011070764 W CN2011070764 W CN 2011070764W WO 2011095115 A1 WO2011095115 A1 WO 2011095115A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
code stream
deinterleaved
column
outputted
Prior art date
Application number
PCT/CN2011/070764
Other languages
English (en)
French (fr)
Inventor
矫渊培
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2011095115A1 publication Critical patent/WO2011095115A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/004Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056Systems characterized by the type of code used
    • H04L1/0071Use of interleaving

Definitions

  • the present invention relates to the field of communications, and more particularly to a de-interleaving method and apparatus.
  • the CQI (Channel Quality Information) decoding part of the PUSCH (Physical Uplink Shared Channel) link in the LTE (Long Time Evolution) protocol is currently in use.
  • Decoding rate matching is required before CC (Convolutional Code) decoding. From the UE (User Equipment) side coding to the base station side decoding process, in order to reduce the burst noise interference and achieve the balance between resources and transmission efficiency, rate matching is performed after coding on the UE side and on the base station side. The operation of de-rate matching is performed before decoding.
  • a coded code stream of three equal lengths is formed.
  • the processes corresponding to the rate matching are interleaving, bit convergence, and bit selection.
  • the effect of the interleaving is to scramble the coded code stream with a CRC (Cyclic Redundancy Check) code according to a preset rule to avoid the uplink channel.
  • CRC Cyclic Redundancy Check
  • the process of interleaving is: 1) inputting the encoded code stream into a pre-interleaving matrix having 32 columns by line; for an encoded code stream having a bit number less than 32 integer multiples, A corresponding number of dummy elements are added at the front end thereof so that the number of encoded code stream bits after adding dummy is up to an integral multiple of 32.
  • Table 1 is a column permutation table of the CQI-CC specified in the protocol, and the lower right digit string indicates the column distribution of the interleaved matrix, where the number is the column number in the matrix before interleaving. 2) After the column permutation according to Table 1, the interleaved matrix of the 0th encoded code stream shown in Fig. 2 is obtained, and the interleaving methods of the other two encoded code streams are the same.
  • the bit convergence is performed by sequentially circulating the data in the three interleaved matrices in the order of the 0th, 1st, and 2nd channels to form a converged code stream.
  • the bit selection is to erase the dummy element in the aggregated code stream and output it bit by bit until the predetermined output bit number is completed, for example, 1000, or the coded code stream length is 49, for example, in the lOOOO bit.
  • the number of bits of valid information is 49. Since there are 3 encoded code streams, the 1000 bits are repeated with a period of 49 ⁇ 3.
  • One of the code streams in each cycle is output in the following order: 33, 17, 49, 41, 25, 57, 37...
  • the solution rate matching includes de-duplication, shunting, and de-interleaving.
  • the process of deduplication is to first obtain the length of the valid information of each way, the number of dummy elements to be added, and the position to add dummy, and add dummy. Or use the above example to illustrate: The length of the 0th code stream in the data stream is 49 (the number of bits of valid information obtained before the base station), 15 dummy elements need to be added, calculate the position of adding the dummy element and add dummy.
  • the code stream that is greater than or less than 1/3 code rate the code rate of the code stream is converted to 1/3 by the method of accumulating saturation or zero padding; the solution repetition of the other two code streams The method is the same.
  • the deduplication corresponds to bit selection.
  • the dummy added data is then split into three channels of data, and the offloading process corresponds to the above bit convergence.
  • the function of deinterleaving is to restore the scrambled code stream to the original code stream for later decoding.
  • the process of deinterleaving is: inputting the three-way code with dummy added into the three pre-interleaving matrices by column (the pre-interleaving matrix, the interleaved matrix, the de-interleaving matrix, and the de-interleaved matrix are all indeterminate
  • the matrix with the number of columns is 32.
  • the matrix before de-interleaving is reverse-column-replaced according to Table 1, and the matrix after de-interleaving is obtained.
  • the 15 dummy elements of the first row of the matrix before de-interleaving are destroyed and outputted according to the row, and three are obtained.
  • the road outputs data for later decoding.
  • the inventor has found that at least the following problems exist in the prior art:
  • the base station side de-duplicates the dummy element is added to the calculated dummy position; and the CQI-CC decoding chain in LTE is used.
  • an embodiment of the present invention provides a de-interleaving method, including:
  • the number of the data to be output after the de-interleaved code stream is virtually added with the dummy element is determined by the virtual dummy number added to the de-interleaved code stream;
  • the data is sequentially read from the data address of the data to be deinterleaved corresponding to the data number to be output, to form a deinterleaved code stream.
  • the embodiment of the present invention further provides a de-interleaving device, including:
  • a calculation module configured to calculate, according to a valid information length of the to-be-deinterleaved code stream, a number of dummy elements to be virtually added, a number of rows of the pre-de-interleaved matrix, and a dummy added dummy element in the de-interleaving Column coordinates in the front matrix;
  • a numbering module configured to determine, by using the dummy number of dummy bits, a number of each to-be-outputted data after the de-interleaved code stream is virtually added with a dummy element at a front end thereof;
  • a row coordinate obtaining module configured to obtain, by the number of the data to be outputted, a row coordinate of each to-be-output data number in the deinterleaved matrix in the deinterleaved matrix;
  • a column coordinate obtaining module configured to obtain, according to a mapping relationship between the data number to be output and the column coordinates in the matrix before deinterleaving, the column coordinates of the data to be outputted in the matrix before deinterleaving;
  • the address obtaining module configured to Determining the row coordinates and column coordinates of the output data number in the matrix before deinterleaving, the number of rows of the matrix before deinterleaving, and the number of dummy cells in the column before the column coordinates, and acquiring the number corresponding to the number of the data to be output Deinterleaving the stream data address;
  • the code stream forming module is configured to sequentially read data from the data address of the code to be deinterleaved corresponding to the data number to be output, to form a deinterleaved code stream.
  • the method and apparatus for deinterleaving calculates the number of dummy elements to be virtually added and the position of the virtual added dummy element according to the effective information length of the to-be-deinterleaved code stream, but does not actually add dummy.
  • the operation of the element; the virtual addition of the dummy element is to determine the number of each data to be outputted in the code stream after deinterleaving, and then obtain the data number of each to be outputted by the mapping relationship between the data number to be output and the column coordinates in the matrix before deinterleaving.
  • the column coordinates in the matrix before interleaving according to the row coordinates and column coordinates in the matrix before deinterleaving, the number of rows of the matrix before deinterleaving, and the number of dummy cells in the column before the column coordinates, according to the data number to be output, obtain the data to be outputted
  • the data address of the data stream to be de-interleaved corresponding to the number thereby avoiding the process of adding the dummy element when the repetition is repeated and removing the dummy element before outputting the de-interleaved code stream, shortening the input to which the number of dummy elements needs to be added and the code rate is low.
  • the time of code stream deinterleaving enhances the throughput of CQI decoding.
  • FIG. 1 is a schematic diagram of a prior art matrix before interleaving
  • FIG. 2 is a schematic diagram of a conventional interleaved matrix
  • FIG. 3 is a schematic flowchart of an embodiment of a deinterleaving method according to an embodiment of the present invention
  • FIG. 4 is a schematic flowchart of another embodiment of a deinterleaving method according to an embodiment of the present invention
  • FIG. 6 is a schematic structural diagram of an embodiment of a deinterleaving apparatus according to an embodiment of the present invention. detailed description
  • the deinterleaving method according to an embodiment of the present invention includes:
  • the base station calculates, according to the effective information length of the to-be-deinterleaved code stream, the number of dummy elements to be de-interleaved to be de-interleaved, the number of rows of the matrix before de-interleaving, and the dummy added dummy matrix before the de-interleaving matrix.
  • the effective information length of the to-be-deinterleaved code stream is the number of bits of valid data in the data stream received by the base station, and the base station can obtain the length of the effective information in the received data stream from the UE before deinterleaving, and can adopt the prior art.
  • the implementation of the present invention is not limited thereto.
  • the data stream received by the base station from the UE is divided into three to-be-deinterleaved code streams, and the effective information length of the to-be-deinterleaved code stream is obtained according to the length of the effective information in the received data stream obtained by the base station from the UE.
  • the pre-de-interleaving matrix needs to be used, so it is necessary to calculate the number of dummy elements to be de-interleaved to be de-interleaved, the number of rows of the matrix before de-interleaving, and the dummy dummy added before the de-interleaving.
  • the column coordinates in the matrix ensure that the length of the processed data string is an integer multiple of 32. Since the dummy elements are virtually added in the 0th line of the matrix before deinterleaving, it is sufficient to obtain the column coordinates of the dummy added dummy elements in the matrix before deinterleaving.
  • the number of the data to be output in the deinterleaved code stream is arranged after the virtual dummy is added. If the number of bits of the code stream after deinterleaving is 49, the number of dummy elements to be added to the front end is 15, and the code stream to be outputted after the deinterleaving is numbered 15 to 63.
  • the row to-be-output data has a row coordinate of 0 in the matrix before deinterleaving, and the coordinate of the data row to be output, numbered 48, is 1.
  • mapping relationship between the data number to be output and the column coordinates in the matrix before deinterleaving may be corresponding through a preset reverse column mapping table, as shown in Table 2:
  • Each of the data numbers to be output corresponds to the position in Table 2, and the number at the position is the column coordinate of the number in the matrix before deinterleaving.
  • the position of each data to be outputted in Table 2 can be obtained by calculating the remainder of the data number pair 32 to be outputted, and the remainder is the position of the data to be outputted in Table 2.
  • the data address of the data to be deinterleaved corresponding to the number of the data to be outputted may be calculated according to the following formula:
  • the data address of the data to be decompressed corresponding to the data number to be output in the deinterleaved code stream the column coordinate X
  • the "column before the column coordinates" includes the current column in the embodiment of the present invention.
  • S106 Read data from the data address of the to-be-deinterleaved code stream corresponding to the data number to be output in sequence, to form a deinterleaved code stream.
  • the dummy number to be de-interleaved and the position of the virtual dummy added are calculated, but the operation of adding the dummy is not actually performed.
  • the base station after receiving the data stream sent by the UE, the base station performs a traffic off operation on the data stream, and when performing deinterleaving, the input code stream includes three channels to be deinterleaved after the splitting.
  • the code stream, the deinterleaved output code stream includes three parallel deinterleaved code streams.
  • the code to be deinterleaved in this embodiment is one code stream in the three-way code stream after the received data stream is split.
  • the three-way de-interleaved code stream after the offloading may be deinterleaved in two manners;
  • the first type deinterleaving the three channels to be deinterleaved code streams in the same manner; second: deinterleaving one of the channels to be deinterleaved, and then directly using the solution of the code to be deinterleaved The interleaving results in the other two de-interleaved code streams.
  • the deinterleaving method of the de-interleaved code stream includes:
  • the base station calculates, according to the effective information length of the to-be-deinterleaved code stream, the dummy number to be de-interleaved to be de-interleaved, the number of rows of the pre-de-interleaved matrix, and the dummy added dummy in the matrix before the de-interleaving. Column coordinates in .
  • the pre-de-interleaving matrix needs to be used, so it is necessary to calculate the number of dummy elements to be de-interleaved to be de-interleaved, the number of rows of the matrix before de-interleaving, and the dummy dummy added before the de-interleaving.
  • the column coordinates in the matrix ensure that the length of the processed data string is an integer multiple of 32. Since the dummy elements are added in the 0th line of the matrix before deinterleaving, it is sufficient to obtain the column coordinates of the dummy in the matrix before deinterleaving.
  • the prior art can be used to calculate the process of adding the dummy position in the code stream before the splitting, and then calculating the dummy matrix before the de-interleaving according to the position of the dummy.
  • the virtual add dummy position is 0, 4, 8, 54 and 58 bits, so that each channel to be deinterleaved code
  • the number of the data to be output in the deinterleaved code stream is arranged after the virtual dummy is added. If the number of bits in the code stream after deinterleaving is 49, such as a0, al, a2, a3 a48, the number of dummy bits that need to be virtually added to the front end is
  • the number of streams to be outputted after the deinterleaving is (a0 ⁇ a48) is numbered 15 to 63.
  • Table 2 and Table 1 are mutually reverse column mapping tables.
  • Table 2 needs to be pre-stored for use in de-interleaving.
  • Each data to be output in the code stream after deinterleaving has a column coordinate in the matrix before deinterleaving.
  • Each of the to-be-output data numbers corresponds to the position (0th to 31st bits) in Table 2, and the number at the position is the column coordinate of the number in the matrix before deinterleaving.
  • the position of each data number to be outputted in Table 2 can be obtained by calculating the remainder of the data number pair 32 to be output, and the remainder is The position of the output data number in Table 2 is described.
  • the data to be output is 25, and the remainder of the pair is 25, the position in Table 2 is 25.
  • the number on the 25th bit is 3, and the data to be output is number 25.
  • the column coordinate in the matrix before deinterleaving is 3; if the data to be output is 60, and the remainder of 32 is 28, its position in Table 2 is 28, and the number in the 28th bit is queried by Table 2 as 23, the data of the data to be output, numbered 60, has a column coordinate of 23 in the matrix before deinterleaving.
  • Determining, according to the mapping relationship between the data number to be output and the column coordinates in the matrix before the de-interleaving, obtaining the column coordinates of the data to be outputted in the de-interleaved matrix in the de-interleaved code stream may include: The five-digit binary number corresponding to the number is rearranged in reverse order and the high order is inverted, and the column coordinates of the data number to be outputted in the matrix before deinterleaving are obtained.
  • the five-digit binary number corresponding to the output data number may be specifically calculated as: taking the remainder of the data number pair 32 to be output, and the five-digit binary number of the remainder is reversed in reverse order and the high bit is inverted, that is, the data to be output is Number the column coordinates in the matrix before deinterleaving.
  • the number of data to be output is 25, and the remainder of 32 is 25, the five-digit binary number of the remainder is 11001, and the reverse order is rearranged and the high bit is inverted to become 00011 or 3, and the number is 25 to be output.
  • the column number of the data in the matrix before deinterleaving is 3. If the number is 60, the remainder of the pair 32 is 28, and the five-digit binary number of the remainder is 11100. The reverse order is rearranged and the high order is inverted and becomes 10111 or 23, and the data to be output numbered 60 is before deinterleaving.
  • the column number in the matrix is 23.
  • the inventor made a hard work on the reverse column mapping table in Table 2 above. After a lot of serious comparative research, the mapping relationship between the data number to be output and the column coordinates in the matrix before de-interleaving and its simple FPGA implementation method were found. , greatly improving the efficiency of obtaining the column coordinates in the corresponding pre-interleaved matrix from the data number to be output.
  • the column coordinates in the matrix before deinterleaving can be obtained by step S203, and the row coordinates can be directly calculated when the number is formed, and the corresponding input data is
  • the number of dummy elements before the third column is 2.
  • the data is sequentially read from the data address of the data stream to be de-interleaved corresponding to the data number to be output, to form a de-interleaved code stream, and the code stream after de-interleaving Corresponding to the code stream after the convolutional coding on the UE side.
  • the demultiplexed code streams to be deinterleaved may be deinterleaved by using the operations of S201 to S206 described above while one of the channels to be deinterleaved is deinterleaved.
  • the method may further include the step S207: directly, according to the data address of the data stream to be de-interleaved. Obtaining data addresses of the to-be-deinterleaved code streams corresponding to the numbers of the respective to-be-output data in the other two demultiplexed code streams.
  • the data addresses of the to-be-deinterleaved code streams corresponding to the same data to be outputted in the three-way de-interleaved code stream are also necessarily the same, so One of the de-interleaved code streams is deinterleaved, and then the data addresses of the data to be de-interleaved corresponding to the data numbers to be outputted in the de-interleaved code stream are directly used to form other two-way de-interleaved code streams.
  • a deduplication is further included.
  • the deduplication is to directly convert the code rate of the received data stream into an input code stream of a set code rate.
  • the set code rate is 1/3. Since the dummy element is not inserted, the function of the deduplication part of the embodiment of the present invention needs to be completed, that is, the zero padding or the accumulation saturation is saved, and the addition is saved. The time of the dummy.
  • the number of dummy elements to be de-interleaved to be de-interleaved is dynamically added, and the position of the virtual dummy is added, but it is not true.
  • the number of the data corresponds to the data address of the data stream to be deinterleaved, thereby avoiding the process of adding the dummy element when repeating and removing the dummy element before outputting the deinterleaved code stream, shortening the number of dummy elements to be added and having a low code rate
  • the time to deinterleave the de-interleaved code stream enhances the throughput of the CQI decoding.
  • the base station after receiving the data stream sent by the UE, the base station does not need to perform the offloading operation on the data stream, but includes the three channels to be deinterleaved.
  • a data string is used as a deinterleaved input code stream, and the three channels to be deinterleaved are sequentially connected end to end in the input code stream, and the output code stream is still a three-way parallel deinterleaved code stream.
  • the base station can identify the starting and ending points of each to-be-deinterleaved code stream in the input code stream according to the effective information length of the input code stream.
  • the data to be deinterleaved in this embodiment is a data string corresponding to one code stream before the received data stream is split.
  • the input code stream may be deinterleaved in two manners
  • the first one deinterleaving the three de-interleaved code streams in the input code stream in the same manner;
  • the second method deinterleaving a de-interleaved code stream in the input code stream, and then directly using the de-interleaving result of the to-be-deinterleaved code stream to obtain another two de-interleaved code streams.
  • the de-interleaving method for a de-interleaved code stream in the input code stream includes:
  • the base station calculates, according to the effective information length of the input code stream, the number of dummy elements that need to be virtually added in each of the to-be-deinterleaved code streams in the input code stream, the number of rows of the matrix before de-interleaving, and the dummy added dummy elements before the de-interleaving.
  • the column coordinates in the matrix.
  • the base station side can obtain the effective information length of the input code stream, and the method for obtaining the data can be used in the prior art.
  • the effective information length of the data stream to be de-interleaved in the second embodiment is equal to the effective information length of the input code stream in this embodiment.
  • the base station calculates, according to the effective information length, the number of dummy elements to be virtually added, the number of rows of the matrix before deinterleaving, and the virtual number to be deinterleaved code streams.
  • the number range of the three-way de-interleaved code stream is the same, that is, the number of the Mth to-be-output data in the code stream after de-interleaving is the same, M For natural numbers.
  • the row coordinates can be obtained while determining the data numbers to be output.
  • the column coordinates in which the data numbers to be output are in the matrix before deinterleaving are further obtained by the mapping relationship.
  • the operation of obtaining the column coordinates of the data to be outputted in the pre-de-interleaving matrix is performed by using the mapping relationship.
  • Step 5305 Obtain, according to the row coordinate and the column coordinate in the matrix before deinterleaving, the number of rows of the matrix before deinterleaving, and the number of dummy cells in the column before the column coordinate, according to the to-be-output data number, acquiring the data to be output The data address of the data stream to be deinterleaved corresponding to the number.
  • the data address of the data to be deinterleaved corresponding to the number of the data to be outputted is calculated according to the following formula:
  • the data address of the data to be decompressed corresponding to the data number to be output in the deinterleaved code stream the coordinate of the column X is deinterleaved The number of rows of the front matrix + the row coordinates - the number of dummy cells in the column before the column coordinates.
  • Read data from the input code stream data address corresponding to the data number to be outputted in sequence to form an output code stream.
  • the de-interleaving code stream to be deinterleaved in the input code stream may be deinterleaved, and the other two channels to be de-interleaved in the input code stream are also used in the above S301.
  • the operation process of ⁇ S306 is deinterleaved.
  • the method may further include the step S307: outputting each to be output according to the de-interleaved code stream.
  • the data to be deinterleaved code stream data address and the effective information length of the code to be deinterleaved Degrees, the data addresses of the to-be-deinterleaved code streams corresponding to the numbers of the data to be outputted in the other two demultiplexed code streams are obtained.
  • the deinterleaved code stream is a 0th channel deinterleaved code stream; and the data address of the to-be-deinterleaved code stream corresponding to the number of each data to be outputted in the code channel after the demultiplexing is obtained as follows: Adding the effective information length of the acquired address by one time, and obtaining a first road to be deinterleaved code stream data address corresponding to the number of each data to be outputted in the first demultiplexed code stream; The obtained address is increased by 2 times of the effective information length, and the second-way de-interleaved code stream data address corresponding to the number of each data to be outputted in the second-channel deinterleaved code stream is obtained.
  • the address is increased by 1 time and 2 times of the effective information length, that is, the first channel and the second channel are respectively obtained.
  • Lbit data after that, the three data can be temporarily stored in three internal registers, and they are serially input into the FIFO (First In First Out), and then the three-way de-interleaved code stream is calculated according to this method.
  • the code stream is deinterleaved, and the three demultiplexed code streams in the FIFO can be synchronously transmitted as an output code stream to the decoding core for subsequent decoding.
  • deduplication Before deinterleaving, it also includes deduplication: directly converts the code rate of the received data stream into an input code stream of a set code rate.
  • the solution repetition of this embodiment does not need to add dummy elements, and directly performs zero padding or accumulated saturation, which saves time for adding dummy elements.
  • no shunting is required, and when the de-interlacing is completed, the three-way output data is started.
  • the method in this embodiment is to perform an operation of inversely calculating a storage address for one of the three channels to be deinterleaved.
  • Another method can be obtained.
  • the storage address of the two-way deinterleaved code stream that is, the address is calculated once, and the address is used three times, further shortening the time required for understanding the interleaving; moreover, the method in this embodiment completes the input code stream while deinterleaving
  • the method of the present embodiment can be used for the de-interleaving operation after the three-way de-interleaved code stream after the splitting is stored in three RAMs (random access memory).
  • the internal buffer is used to divide the three-way de-interleaved code stream, and a three-way de-interleaved code stream is synchronized by using one FIFO, thereby saving three RAM resources.
  • the deinterleaving method of the embodiment according to the effective information length of the to-be-deinterleaved code stream, the dummy number to be de-interleaved and the position of the virtual dummy added are calculated, but the operation of adding the dummy is not actually performed.
  • the embodiment of the present invention provides a de-interleaving device. As shown in FIG. 6, the method includes: a computing module 10, a numbering module 20, a row coordinate acquiring module 30, a column coordinate acquiring module 40, an address obtaining module 50, and a code stream forming module 60. .
  • the calculating module 10 is configured to calculate, according to the effective information length of the to-be-deinterleaved code stream, the number of dummy elements that need to be virtually added, the number of rows of the matrix before de-interleaving, and the dummy added dummy elements in the solution.
  • the column coordinates in the matrix before interleaving.
  • the numbering module 20 is configured to determine, by using the dummy number of dummy bits, a number of each to-be-outputted data after the de-interleaved code stream is virtually added with a dummy element at a front end thereof.
  • the row coordinate obtaining module 30 is configured to obtain row coordinates of the data numbers to be outputted in the deinterleaved matrix in the deinterleaved matrix by the numbers of the data to be outputted.
  • the column coordinate obtaining module 40 is configured to obtain, according to the mapping relationship between the data number to be output and the column coordinates in the matrix before deinterleaving, the column coordinates of the data to be outputted in the matrix before deinterleaving.
  • the column coordinate acquisition module 40 can include:
  • a first column coordinate acquiring unit configured to obtain the deinterlacing by using a preset reverse column mapping table
  • Column coordinates of each data to be output in the code stream before deinterleaving, the reverse column mapping table is shown in Table 2;
  • a second column coordinate obtaining unit configured to reorder the five-digit binary numbers corresponding to the data numbers to be outputted in the deinterleaved code stream in reverse order and invert the high bits, to obtain the data numbers to be outputted in the pre-interleaving matrix Column coordinates.
  • the second column coordinate acquisition unit can adopt register cross-map and connect a non-gate to the high-order output.
  • the 32-branch logic statement implemented by 32 multiplexers through a non-gate saves logic resources and increases the maximum frequency supported here.
  • the address obtaining module 50 is configured to acquire row coordinates and column coordinates in the matrix before deinterleaving, the number of rows of the matrix before deinterleaving, and the number of dummy pixels in the column before the column coordinates according to the data number to be outputted.
  • the address of the data to be deinterleaved corresponding to the number of the output data is mentioned.
  • the address obtaining module 50 obtains the data address of the to-be-deinterleaved code stream corresponding to the number of the to-be-output data, and can be calculated according to the following formula:
  • the data address of the data to be de-interleaved corresponding to the data number to be output in the de-interleaved code stream the number of rows of the matrix before the de-interleaving of the column coordinates X + the row coordinates - the number of dummy cells in the column before the column coordinates.
  • the code stream forming module 60 is configured to sequentially read data from the data address to be deinterleaved corresponding to the data number to be output, to form a deinterleaved code stream.
  • the to-be-deinterleaved code stream may be a code stream in a three-way code stream after the received data stream is split, or a data string corresponding to one code stream before the received data stream is split.
  • the address obtaining module 50 may further include:
  • a first acquiring unit configured to: when the to-be-deinterleaved code stream is the one-way code stream, directly obtain the corresponding information in the other two de-interleaved code streams according to the data address of the to-be-deinterleaved code stream The number of the data to be output corresponds to the address of the data stream to be deinterleaved; or
  • a second acquiring unit configured to: when the to-be-deinterleaved code stream is a data string corresponding to one code stream before the offloading, according to the to-be-deinterleaved code stream data of each data to be outputted in the deinterleaved code stream The address and the effective information length of the to-be-deinterleaved code stream are obtained, and the data addresses of the to-be-deinterleaved code streams corresponding to the numbers of the data to be outputted in the other two demultiplexed code streams are obtained.
  • the deinterleaved code stream is a 0th deinterleaved code stream; and the second obtaining unit includes:
  • a first sub-unit configured to increase the effective information length by one time for the obtained address, and obtain a first de-interleave code corresponding to the number of each to-be-output data in the first de-interleaved code stream.
  • a second sub-unit configured to increase the effective information length by 2 times the obtained address, and obtain a second de-interleaved code stream data corresponding to the number of each to-be-output data in the second de-interleaved code stream. address.
  • the apparatus may further include: a conversion module, configured to directly convert a code rate of the received data stream into an input code stream of a set code rate before deinterleaving.
  • a conversion module configured to directly convert a code rate of the received data stream into an input code stream of a set code rate before deinterleaving.
  • the deinterleaving apparatus of this embodiment calculates the number of dummy elements to be de-interleaved and the position of the virtual added dummy according to the effective information length of the to-be-deinterleaved code stream, but does not actually perform the operation of adding the dummy element.
  • Deinterleaving the code stream data address thereby avoiding the process of adding the dummy element when repeating and outputting the deinterleaved code stream before removing the dummy element, shortening the deinterleaving of the input code stream with a large number of dummy elements and low code rate The time enhances the throughput of CQI decoding.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Error Detection And Correction (AREA)

Description

一种解交织方法和装置 本申请要求了 2010年 02月 02日提交的、 申请号为 201010111391.6、发明 名称为 "一种解交织方法和装置"的中国申请的优先权,其全部内容通过引用 结合在本申请中。 技术领域
本发明涉及通信领域, 尤其是一种解交织方法和装置。 背景技术 目前, LTE( Long Time Evolution,长期演进)协议中的 PUSCH( Physical Uplink Shared Channel, 上行物理共享信道 )链路中的 CQI ( Channel Quality Information,信道质量信息)译码部分,在对其中的 CC( Convolutional Code, 卷积码)译码前需要进行解速率匹配。 从 UE ( User Equipment, 用户设备) 侧编码到基站侧译码过程中, 为了降低突发性噪声干扰, 以及达到资源与 传输效率的平衡,通常采用在 UE侧编码后进行速率匹配和在基站侧译码前 进行解速率匹配的操作。
在 UE侧, 原数据信息经过卷积编码(在 LTE中采用 (3, 1, 7 )编码 ) 后, 会形成 3路等长的编码后码流。 所述速率匹配对应的过程为交织、 比 特汇聚和比特选择。其中,对于各路编码后码流,交织的作用是对带有 CRC ( Cyclic Redundancy Check, 循环冗余校验 )码的所述编码后码流按照预置 的规则进行置乱, 避免在上行信道中传输时受到突发性噪声干扰而造成的 一连串数据错误。
以其中一路编码后码流为例, 交织的过程是: 1 )将所述编码后码流按 行输入到具有 32列的交织前矩阵中; 对于比特数不足 32整数倍的编码后 码流, 在其前端加入相应数目的哑元, 以使添加哑元后的编码后码流比特 数达到 32的整数倍。 下面以 3路编码后码流中的第 0路编码后码流为例进 行说明。若第 0路编码后码流长度为 49,则需添加的哑元数为 64-49=15个, 如图 1所示, 其中带阴影的表示哑元, 编码后码流前端的第 0〜14位均为哑 元(图 1中未全部示出) 。 表 1为协议中规定的 CQI-CC的列置换表,其右下方的数字串表示交织 后矩阵的列分布, 其中的数字为交织前矩阵中的列号。 2 )按照表 1进行列 置换后, 得到图 2所示的第 0路编码后码流的交织后矩阵, 其他两路编码 后码流的交织方法相同。
表 1
Figure imgf000004_0001
比特汇聚是按照第 0路、 第 1路、 第 2路的顺序, 将三个交织后矩阵 中的数据循环的、 按列顺次输出, 形成汇聚码流。 比特选择是打掉汇聚码 流中的哑元并逐个比特(bit )进行输出,直到完成预定的输出 bit数,如 1000 个, 还是以编码后码流长度为 49为例, 则所述 lOOObit中有效信息的位数 为 49, 由于有 3路编码后码流, 因此所述 lOOObit是以 49 <3作为周期进行 重复。 每一周期中的其中一路码流按后续顺序输出: 33、 17、 49、 41、 25、 57、 37... ...。
在基站侧,接收到 UE发送的数据流后,在译码前要对接收的数据流进 行解速率匹配。 解速率匹配包括解重复、 分流和解交织。 解重复的过程是 先获得各路所述有效信息的长度、 需添加的哑元数和添加哑元的位置并添 加哑元。 还是使用上面的例子进行说明: 所述数据流中第 0路码流长度为 49 (基站之前获得了有效信息的位数) , 需添加 15个哑元, 计算出添加哑 元的位置并添加哑元; 之后完成码率的转换, 即将大于或小于 1/3码率的码 流, 通过累加饱和或者补零的方法将码流的码率转换为 1/3; 其他两路码流 的解重复方法相同。 所述解重复与比特选择相对应。 然后将添加哑元后的 数据分流成三路数据, 所述分流过程与上述比特汇聚相对应。
解交织的作用是将置乱后的码流还原成原始码流的顺序, 以备后续译 码使用。 解交织的过程是: 将添加有哑元的三路码流行按列输入到三个解 交织前矩阵(所述交织前矩阵、 交织后矩阵、 解交织前矩阵和解交织后矩 阵都是行数不定、 列数为 32的矩阵) , 将解交织前矩阵按照表 1进行逆向 列置换, 得到解交织后矩阵, 打掉各解交织前矩阵第一行的 15个哑元并按 行输出, 得到三路输出数据, 以备后续译码使用。 发明人在实现本发明的过程中, 发现现有技术至少存在如下问题: 基站侧在解重复时, 会先在计算出的哑元位置上添加哑元; 而在 LTE 的 CQI-CC译码链路中, 存在需要添加的哑元数目较多且码率较低的情况, 例如, 输入的 CC数据长度为 33, 则需要加入的哑元数目为 64-33=31, 代 表着 48.4%的时间是在做哑元的添加和累加操作, 若按照单软比特解重复, 在码率为 1/12时会造成 31 x3 (三路数据) x4 (重复 4次) =372个时钟的 浪费。 这样, 大量的时间用于完成哑元的添加和哑元的累加, 造成解速率 匹配耗时较长, 进而译码延迟增大和吞吐量下降。 发明内容
本发明实施例的目的是提供一种解交织方法和装置。
一方面, 本发明实施例提供了一种解交织方法, 包括:
根据待解交织码流的有效信息长度, 计算所述待解交织码流需要虚拟 添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在所述解交织前矩 阵中的列坐标;
通过所述需要虚拟添加哑元数, 确定解交织后码流在其前端虚拟添加 哑元后各待输出数据的编号;
由所述各待输出数据的编号得到解交织后码流中各待输出数据编号在 解交织前矩阵中的行坐标;
根据待输出数据编号与解交织前矩阵中列坐标的映射关系, 获得所述 各待输出数据编号在解交织前矩阵中的列坐标;
根据所述待输出数据编号在解交织前矩阵中的行坐标及列坐标、 解交 织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出数据 的编号对应的待解交织码流数据地址;
依次从与待输出数据编号对应的待解交织码流数据地址读取数据, 形 成解交织后码流。
另一方面, 本发明实施例还提供了一种解交织装置, 包括:
计算模块, 用于根据待解交织码流的有效信息长度, 计算所述待解交 织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在 所述解交织前矩阵中的列坐标; 编号模块, 用于通过所述需要虚拟添加哑元数, 确定解交织后码流在 其前端虚拟添加哑元后的各待输出数据的编号;
行坐标获取模块, 用于由所述各待输出数据的编号得到解交织后码流 中各待输出数据编号在解交织前矩阵中的行坐标;
列坐标获取模块, 用于根据待输出数据编号与解交织前矩阵中列坐标 的映射关系, 获得所述各待输出数据编号在解交织前矩阵中的列坐标; 地址获取模块, 用于根据所述待输出数据编号在解交织前矩阵中的行 坐标及列坐标、 解交织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出数据的编号对应的待解交织码流数据地址;
码流形成模块, 用于依次从与待输出数据编号对应的待解交织码流数 据地址读取数据, 形成解交织后码流。
本发明实施例的解交织方法和装置, 根据待解交织码流的有效信息长 度, 计算待解交织码流需要虚拟添加的哑元数和虚拟添加哑元的位置, 但 是并不真正进行添加哑元的操作; 虚拟添加哑元, 是为了确定解交织后码 流中各待输出数据的编号, 然后通过待输出数据编号与解交织前矩阵中列 坐标的映射关系得到各待输出数据编号在解交织前矩阵中的列坐标, 根据 待输出数据编号在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行 数、 列坐标之前的列中的哑元数, 获取待输出数据的编号对应的待解交织 码流数据地址, 由此避免了解重复时添加哑元和输出解交织后码流前去掉 哑元的过程, 缩短了对需添加哑元数目较多且码率低的输入码流解交织的 时间, 增强了 CQI译码的吞吐能力。
附图说明 为了更清楚地说明本发明实施例或现有技术中的技术方案, 下面将对 实施例或现有技术描述中所需要使用的附图作简单地介绍。
图 1是现有的交织前矩阵的示意图;
图 2是现有的交织后矩阵的示意图;
图 3是本发明实施例的解交织方法的一实施例的流程示意图; 图 4是本发明实施例的解交织方法的另一实施例的流程示意图 图 5是本发明实施例的解交织方法的第再一实施例的流程示意图; 图 6是本发明实施例的解交织装置实施例的结构示意图。 具体实施方式
为使本发明实施例的上述目的、 特征和优点能够更加明显易懂, 下面 结合附图对本发明实施例作进一步详细的说明。
如图 3所示, 本发明一实施例的解交织方法包括:
5101 , 基站根据待解交织码流的有效信息长度, 计算所述待解交织码 流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在所述 解交织前矩阵中的列坐标;
所述待解交织码流的有效信息长度, 为基站接收到的数据流中有效数 据的位数, 基站在解交织前可以从 UE 获得接收的数据流中有效信息的长 度, 可以采用现有技术实现, 本发明实施例对此不做限定。 本实施例中, 基站从 UE接收的数据流分为三路待解交织码流,根据基站从 UE获得的接 收的数据流中有效信息的长度即可得到待解交织码流的有效信息长度。
本发明实施例中, 需要使用到解交织前矩阵, 因此需要计算出待解交 织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在 所述解交织前矩阵中的列坐标, 保证被处理的数据串长度为 32的整数倍。 由于哑元都是在解交织前矩阵的第 0行中虚拟添加, 因此获得虚拟添加的 哑元在解交织前矩阵中的列坐标就足够了。
计算虚拟添加的哑元在解交织前矩阵中的列坐标, 可以利用现有技术 在解重复时计算添加哑元位置的方式, 然后根据哑元的位置算出哑元在解 交织前矩阵中的列坐标, 例如, 具体可以使用公式: 虚拟添加的哑元在解 交织前矩阵中的列坐标=虚拟添加哑元的位置 /解交织前矩阵行数;由于虚拟 添加的哑元的位置均为 n*k, 其中 n为解交织前矩阵的行数, k为虚拟添加 的哑元所在列的列坐标, 因此通过上述公式计算的列坐标结果可以保证是 整数。
5102 , 通过所述需要虚拟添加哑元数, 确定解交织后码流在其前端虚 拟添加哑元后的各待输出数据的编号;
由于在解交织后码流的前端虚拟添加哑元, 因此, 解交织后码流中的 待输出数据的编号要在虚拟添加哑元后进行排列。 若解交织后码流的位数 为 49, 则其前端需添入的哑元数为 15个, 那么该路解交织后码流待输出数 据的编号为 15到 63。
S 103, 由所述各待输出数据的编号得到解交织后码流中各待输出数据 编号在解交织前矩阵中的行坐标;
由各待输出数据的编号可以得到各待输出数据编号在解交织前矩阵中 的行坐标,例如,使用公式:待输出数据行坐标 = [待输出数据编号 /32]取整, 如编号为 25的待输出数据在解交织前矩阵中的行坐标为 0,编号为 48的待 输出数据行坐标为 1。
S104, 根据待输出数据编号与解交织前矩阵中列坐标的映射关系, 获 得所述各待输出数据编号在解交织前矩阵中的列坐标;
所述待输出数据编号与解交织前矩阵中列坐标的映射关系可以通过预 置的逆向列映射表进行对应, 所述逆向列映射表如表 2所示:
表 2
Figure imgf000008_0001
各待输出数据编号与表 2 中的位置对应, 所述位置上的数字即为该编 号在解交织前矩阵中的列坐标。 其中, 每个待输出数据编号在表 2 中的位 置可以通过将待输出数据编号对 32求余数, 该余数即为所述待输出数据编 号在表 2中的位置。
5105 , 根据所述待输出数据编号在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出 数据的编号对应的待解交织码流数据地址;
所述获取待输出数据的编号对应的待解交织码流数据地址具体可以按 照如下公式计算: 解交织后码流中的待输出数据编号对应的待解交织码流 数据地址=所述列坐标 X解交织前矩阵的行数 +所述行坐标-所述列坐标之 前的列中的哑元数。
其中, 所述 "列坐标之前的列" 在本发明实施例中包含当前列。
5106, 依次从与待输出数据编号对应的待解交织码流数据地址读取数 据, 形成解交织后码流。
本实施例的解交织方法, 根据待解交织码流的有效信息长度, 计算待 解交织码流需要虚拟添加的哑元数和虚拟添加哑元的位置, 但是并不真正 进行添加哑元的操作; 虚拟添加哑元, 是为了确定解交织后码流中各待输 出数据的编号, 然后通过待输出数据编号与解交织前矩阵中列坐标的映射 关系得到各待输出数据编号在解交织前矩阵中的列坐标, 根据待输出数据 编号在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 列坐标 之前的列中的哑元数, 获取待输出数据的编号对应的待解交织码流数据地 址, 由此避免了解重复时添加哑元和输出解交织后码流前去掉哑元的过程, 缩短了对需添加哑元数目较多且码率低的输入码流解交织的时间, 增强了
CQI译码的吞吐能力。 在本发明另一本实施例中,基站在接收到 UE发送的数据流后,会对所 述数据流进行分流操作, 则在进行解交织时, 输入码流包括分流后的三路 待解交织码流, 解交织后的输出码流包括三路并行解交织后码流。 本实施 例中的待解交织码流为接收到的数据流分流后的三路码流中的一路码流。
本实施例可以采用两种方式对所述分流后的三路待解交织码流进行解 交织;
第一种: 对所述三路待解交织码流采用相同的方式进行解交织; 第二种: 对其中一路待解交织码流进行解交织, 然后直接利用所述待 解交织码流的解交织结果得到另两路的解交织后码流。
无论采用上述哪种方式, 其中一路待解交织码流的解交织方法, 如图 4 所示, 包括:
S201 , 基站根据待解交织码流的有效信息长度, 计算所述待解交织码 流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在所述 解交织前矩阵中的列坐标。
本发明实施例中, 需要使用到解交织前矩阵, 因此需要计算出待解交 织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在 所述解交织前矩阵中的列坐标, 保证被处理的数据串长度为 32的整数倍。 由于哑元都是在解交织前矩阵的第 0行中添加, 因此获得哑元在解交织前 矩阵中的列坐标就足够了。
计算虚拟添加的哑元在解交织前矩阵中的列坐标, 可以利用现有技术 计算在分流前的码流中添加哑元位置的过程, 然后根据哑元的位置算出哑 元在解交织前矩阵中的列坐标, 例如, 具体可以使用公式: 虚拟添加的哑 元在解交织前矩阵中的列坐标 =虚拟添加哑元的位置 /解交织前矩阵行数。对 于有效信息长度为 49的待解交织码流, 参见图 2, 按照以列读取的方式, 虚拟添加哑元位置为第 0、 4、 8 54、 58位, 以使每路待解交织码 流虚拟添加哑元后长度为 32的整数倍,便于利用解交织前矩阵(列数为 32 ) 进行计算。 若虚拟添加的哑元在某路待解交织码流中的位置为 54, 则通过 54/2=27可知,所述虚拟添加的哑元在解交织前矩阵中的列坐标为 27,与图 2 ^艮好的吻合。
S202 , 通过所述需要虚拟添加哑元数, 确定解交织后码流在其前端虚 拟添加哑元后各待输出数据的编号。
由于在解交织后码流的前端虚拟添加哑元, 因此, 解交织后码流中的 待输出数据的编号要在虚拟添加哑元后进行排列。 若解交织后码流的位数 为 49, 如 a0、 al、 a2、 a3 a48 , 则其前端需要虚拟添入的哑元数为
15个, 那么该路解交织后码流待输出数据(a0〜a48 ) 的编号为 15到 63。
S203, 由所述各待输出数据的编号得到解交织后码流中各待输出数据 编号在解交织前矩阵中的行坐标。
由各待输出数据的编号可以得到各待输出数据编号在解交织前矩阵中 的行坐标,例如,使用公式:待输出数据行坐标 = [待输出数据编号 /32]取整, 以步骤 S202中的例子进一步进行说明, 如 alO编号为 25, [25/32]取整 =0, 则 alO在解交织前矩阵中的行坐标为 0, 同理, a33编号为 48, 其待输出数 据行坐标为 1。
S204, 根据待输出数据编号与解交织前矩阵中列坐标的映射关系, 获 得所述各待输出数据编号在解交织前矩阵中的列坐标。
本发明实施例提供了两种获得待输出数据编号在解交织前矩阵中的列 坐标的方式:
A )通过预置的逆向列映射表获得各待输出数据编号在解交织前矩阵中 的列坐标。
发明人通过对表 1进行转换得到如表 2所示的逆向列映射表, 表 2与 表 1互为逆向列映射表, 本实现方式中表 2需要预存起来以备解交织时使 用。 解交织后码流中各待输出数据编号都拥有一个在解交织前矩阵中的列 坐标。 各待输出数据编号与表 2中的位置(第 0〜31位)对应, 所述位置上 的数字即为该编号在解交织前矩阵中的列坐标。 其中, 每个待输出数据编 号在表 2中的位置可以通过将待输出数据编号对 32求余数, 该余数即为所 述待输出数据编号在表 2中的位置。如待输出数据编号为 25, 其对 32的余 数为 25, 则其在表 2中的位置为 25, 通过表 2可以查询到第 25位上的数 字为 3, 则编号为 25的待输出数据在解交织前矩阵中的列坐标为 3; 若待 输出数据编号为 60, 其对 32的余数为 28, 则其在表 2中的位置为 28, 通 过表 2查询第 28位上的数字为 23, 则编号为 60的待输出数据在解交织前 矩阵中的列坐标为 23。
B )通过 FPGA ( Field Programmable Gate Array, 现场可编程门阵列) 进行具体实现。
所述根据待输出数据编号与解交织前矩阵中列坐标的映射关系, 获得 所述解交织后码流中各待输出数据编号在解交织前矩阵中的列坐标具体可 以包括: 将待输出数据编号对应的五位二进制数倒序重排并高位取反, 得 到待输出数据编号在解交织前矩阵中的列坐标。
其中, 待输出数据编号对应的五位二进制数具体可以计算为: 将待输 出数据编号对 32取余数, 所述余数的五位二进制数倒序重排并高位取反, 即为所述待输出数据编号在解交织前矩阵中的列坐标。
例如, 待输出数据的编号为 25, 其对 32取余数后为 25, 该余数的五 位二进制数为 11001, 倒序重排并高位取反后变成 00011即 3, 则编号为 25 的待输出数据在解交织前矩阵中的列号为 3。 若编号为 60, 其对 32取余数 后为 28,该余数的五位二进制数为 11100,倒序重排并高位取反后变成 10111 即 23, 则编号为 60的待输出数据在解交织前矩阵中的列号为 23。
在通过 FPGA具体实现时, 可以通过寄存器交叉映射, 并在高位输出 端后接一个非门。通过一个非门实现了原有的由 32个多路选择器实现的 32 分支的逻辑语句, 节省了逻辑资源, 提高了此处支持的最大频率。
发明人对上述表 2 中的逆向列映射表进行了辛勤的付出, 在大量认真 的比较研究后, 发现了待输出数据编号与解交织前矩阵中列坐标的映射关 系以及其简便的 FPGA 实现方式, 大幅提高了从待输出数据编号获得其对 应的解交织前矩阵中列坐标的效率。
S205 , 根据所述待输出数据编号在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出 数据的编号对应的待解交织码流数据地址。
所述获取待输出数据的编号对应的待解交织码流数据地址按照如下公 式计算: 解交织后码流中的待输出数据编号对应的待解交织码流数据地址= 所述列坐标 X解交织前矩阵的行数 +所述行坐标-所述列坐标之前的列中的 哑元数。
例如, 对于编号为 25的待输出数据, 通过步骤 S203可以获得其在解 交织前矩阵中的列坐标为 3,所述行坐标在形成编号的时候就可以直接计算 出来, 则其对应的输入数据地址为: 3x2+0-2=4 (可以参见图 2 ), 其中, 编 号为 25 的待输出数据在解交织前矩阵中的位置为 (0, 3 ), 解交织前矩阵 的行数为 2, 解交织前矩阵中, 第 3列之前的哑元数为 2。
S206, 依次从与待输出数据编号对应的待解交织码流数据地址读取数 据, 形成解交织后码流。
在获得各待输出数据编号对应的待解交织码流数据地址后, 依次从与 待输出数据编号对应的待解交织码流数据地址读取数据, 形成解交织后码 流, 解交织后码流与 UE侧卷积编码后的码流对应。
至此, 已完成对一路待解交织码流的解交织操作。
对于在本实施例开始部分提及的第一种方式, 可以在其中一路待解交 织码流解交织的同时, 其余两路待解交织码流也采用上述 S201〜S206的操 作过程进行解交织。
对于第二种方式, 在获取其中一路解交织后码流中待输出数据的编号 对应的待解交织码流数据地址后, 还可以包括步骤 S207: 根据所述待解交 织码流数据地址, 直接获取另两路解交织后码流中各相应待输出数据的编 号分别对应的待解交织码流数据地址。 由于三路待解交织码流的有效信息 长度相同, 交织方式相同, 因此三路解交织后码流中相同待输出数据编号 分别对应的待解交织码流数据地址也必然相同, 所以可以先对其中一路待 解交织码流解交织, 然后直接使用其解交织后码流中各待输出数据编号对 应的待解交织码流数据地址, 形成其他两路解交织后码流。
在解交织前, 还包括解重复, 对于本发明实施例, 解重复是将接收到 的数据流的码率直接转换为设定码率的输入码流。 目前在 LTE系统中, 所 述设定码率为 1/3, 由于不用插入哑元, 所以本发明实施例的解重复部分需 要完成的功能更简单, 即实现补零或累加饱和, 节省了添加哑元的时间。
本实施例的解交织方法, 根据待解交织码流的有效信息长度, 计算待 解交织码流需要虚拟添加的哑元数和虚拟添加哑元的位置, 但是并不真正 进行添加哑元的操作; 虚拟添加哑元, 并确定解交织后码流中各待输出数 据的编号, 然后通过待输出数据编号与解交织前矩阵中列坐标的映射关系 得到各待输出数据编号在解交织前矩阵中的列坐标, 根据待输出数据编号 在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 列坐标之前 的列中的哑元数, 获取待输出数据的编号对应的待解交织码流数据地址, 由此避免了解重复时添加哑元和输出解交织后码流前去掉哑元的过程, 缩 短了对需添加哑元数目较多且码率低的待解交织码流解交织的时间, 增强 了 CQI译码的吞吐能力。 与上一实施例不同,在本发明再一实施例中,基站在接收到 UE发送的 数据流后, 不需对所述数据流进行分流操作, 而是将包含三路待解交织码 流的一个数据串作为解交织的输入码流, 所述三路待解交织码流在输入码 流中依次首尾相接, 而输出码流仍然为三路并行解交织后码流。 基站能够 根据输入码流的有效信息长度识别出每路待解交织码流在输入码流中的起 止点。 本实施例中的待解交织码流为接收到的数据流分流前与一路码流对 应的数据串。
本实施例可以采用两种方式对所述输入码流进行解交织;
第一种: 对所述输入码流中的三路待解交织码流采用相同的方式进行 解交织;
第二种: 对所述输入码流中的一路待解交织码流进行解交织, 然后直 接利用所述待解交织码流的解交织结果得到另两路解交织后码流。
无论采用上述哪种方式, 对输入码流中一路待解交织码流的解交织方 法, 如图 5所示, 包括:
S301 , 基站根据输入码流的有效信息长度, 计算输入码流中各待解交 织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在 所述解交织前矩阵中的列坐标。
基站侧能够获得输入码流的有效信息长度, 获取的方式可使用现有技 术, 本发明实施例对此不做限定。对于从 UE接收到的数据流, 无论是否分 流, 其有效信息长度都是不变的, 即实施例二中待解交织码流的有效信息 长度等于本实施例中输入码流的有效信息长度。 基站根据该有效信息长度 计算各路待解交织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚 拟添加的哑元在所述解交织前矩阵中的列坐标。
S302 , 通过所述需要虚拟添加哑元数, 确定解交织后码流在其前端虚 拟添加哑元后的各待输出数据的编号。
由于三路解交织后码流是各自独立进行编号的, 因此所述三路解交织 后码流的编号范围相同, 即各路解交织后码流中第 M个待输出数据的编号 相同, M为自然数。
S303, 由所述各待输出数据的编号得到解交织后码流中各待输出数据 编号在解交织前矩阵中的行坐标。
所述行坐标在确定各待输出数据编号的同时就可以得出。
5304 , 根据待输出数据编号与解交织前矩阵中列坐标的映射关系, 获 得所述各待输出数据编号在解交织前矩阵中的列坐标。
对步骤 S303中计算过行坐标的解交织后码流, 进一步通过所述映射关 系获得其中各待输出数据编号在解交织前矩阵中的列坐标。 通过所述映射 关系获得各待输出数据编号在解交织前矩阵中的列坐标的操作, 可以参见 实施例二中的描述, 此处不再赘述。
5305 , 根据所述待输出数据编号在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出 数据的编号对应的待解交织码流数据地址。
所述获取待输出数据的编号对应的待解交织码流数据地址按照如下公 式计算: 解交织后码流中的待输出数据编号对应的待解交织码流数据地址= 所述列坐标 X解交织前矩阵的行数 +所述行坐标-所述列坐标之前的列中的 哑元数。
5306 , 依次从与待输出数据编号对应的输入码流数据地址读取数据, 形成输出码流。
至此, 完成了对输入码流中的其中一路待解交织码流的解交织。
对于本实施例开始部分提及的第一种方式, 可以在输入码流中的一路 待解交织码流解交织的同时, 对输入码流中的其他两路待解交织码流也采 用上述 S301〜S306的操作过程进行解交织。
对于第二种方式, 在获取所述解交织后码流中待输出数据的编号对应 的待解交织码流数据地址后, 还可以包括步骤 S307: 根据所述解交织后码 流中各待输出数据的待解交织码流数据地址和待解交织码流的有效信息长 度, 获取另两路解交织后码流中各待输出数据的编号对应的待解交织码流 数据地址。
优选的, 所述解交织后码流为第 0路解交织后码流; 则所述获取另两 路解交织后码流中各待输出数据的编号对应的待解交织码流数据地址为: 对所述获取的地址增加 1倍的所述有效信息长度, 获得第 1路解交织 后码流中各待输出数据的编号对应的第 1路待解交织码流数据地址; 以及, 对所述获取的地址增加 2倍的所述有效信息长度, 获得第 2路解交织后码 流中各待输出数据的编号对应的第 2路待解交织码流数据地址。
第 0路解交织后码流的第 lbit在从相应的输入码流数据地址读取出数 据后, 对该地址分别增加 1倍和 2倍的有效信息长度, 即分别得到第 1路 和第 2路解交织后码流的第 lbit的输入码流数据地址,由此可以在得到第 0 路解交织后码流的第 lbit数据的同时获得第 1路和第 2路解交织后码流的 第 lbit数据; 之后, 可以通过三个内部寄存器暂存这三个数据, 并将它们 串行输入到 FIFO ( First In First Out, 先进先出) 中, 然后按照此方法计算 三路解交织后码流的其他比特的输入码流数据地址。 当三路解交织后码流 都解交织完毕后,就可以将 FIFO中的三路解交织后码流作为输出码流同步 传输给译码核, 以进行后续译码。
在解交织前, 还包括解重复: 将接收到的数据流的码率直接转换为设 定码率的输入码流。 本实施例的解重复不需添加哑元, 直接进行补零或累 加饱和, 节省了添加哑元的时间。 而且, 解重复后不需分流, 在完成解交 织的同时就开始分流成三路输出数据。
本实施例的方法是对三路待解交织码流中的其中一路进行逆推计算存 储地址的操作, 利用一路待解交织码流的有效数据串长度和所述得到的地 址, 就可以得到另两路解交织后码流的存储地址, 即计算一次地址, 对该 地址使用三次, 进一步缩短了解交织所需的时间; 而且, 本实施例的方法 是在解交织的同时完成对输入码流的分流操作, 相比于现有技术将分流后 的三路待解交织码流存储到三个 RAM ( Random Access Memory, 随机存取 存储器)后再进行解交织操作而言, 本实施例的方法可以使用内部寄存器 对三路解交织后码流进行分流, 使用一个 FIFO对三路解交织后码流进行同 步, 节约了 3个 RAM资源。 本实施例的解交织方法, 根据待解交织码流的有效信息长度, 计算待 解交织码流需要虚拟添加的哑元数和虚拟添加哑元的位置, 但是并不真正 进行添加哑元的操作; 虚拟添加哑元, 并确定解交织后码流中各待输出数 据的编号, 然后通过待输出数据编号与解交织前矩阵中列坐标的映射关系 得到各待输出数据编号在解交织前矩阵中的列坐标, 根据待输出数据编号 在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 列坐标之前 的列中的哑元数, 获取待输出数据的编号对应的待解交织码流数据地址, 由此避免了解重复时添加哑元和输出解交织后码流前去掉哑元的过程, 缩 短了对需添加哑元数目较多且码率低的输入码流解交织的时间; 而且对一 个地址使用三次, 进一步缩短了解交织所需的时间; 增强了 CQI译码的吞吐 能力; 本实施例的方法是在解交织的同时使用内部寄存器对输入码流进行 分流、 使用 FIFO进行三路解交织后码流的同步操作, 节约了 3个 RAM资源。
当然, 本领域技术人员可以基于本发明实施例公开的思想, 对上述实 施例进行组合, 进而发展出更多的实施例, 这些实施例均应包含在本发明 实施例的保护范围内。 第四实施例
本发明实施例提供了一种解交织装置, 如图 6所示, 包括: 计算模块 10, 编号模块 20, 行坐标获取模块 30, 列坐标获取模块 40, 地址获取模块 50, 码流形成模块 60。
计算模块 10, 用于根据待解交织码流的有效信息长度, 计算所述待解 交织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元 在所述解交织前矩阵中的列坐标。
编号模块 20, 用于通过所述需要虚拟添加哑元数, 确定解交织后码流 在其前端虚拟添加哑元后的各待输出数据的编号。
行坐标获取模块 30, 用于由所述各待输出数据的编号得到解交织后码 流中各待输出数据编号在解交织前矩阵中的行坐标。
列坐标获取模块 40, 用于根据待输出数据编号与解交织前矩阵中列坐 标的映射关系, 获得所述各待输出数据编号在解交织前矩阵中的列坐标。
所述列坐标获取模块 40可以包括:
第一列坐标获取单元, 用于通过预置的逆向列映射表获得所述解交织 后码流中的各待输出数据编号在解交织前矩阵中的列坐标, 所述逆向列映 射表如表 2所示; 和 /或
第二列坐标获取单元, 用于将所述解交织后码流中各待输出数据编号 对应的五位二进制数倒序重排并高位取反, 得到各待输出数据编号在解交 织前矩阵中的列坐标。 在通过 FPGA具体实现时, 第二列坐标获取单元可 以采用寄存器交叉映射, 并在高位输出端后接一个非门。 通过一个非门实 现原有的由 32个多路选择器实现的 32分支的逻辑语句, 节省了逻辑资源, 提高了此处支持的最大频率。
地址获取模块 50, 用于根据所述待输出数据编号在解交织前矩阵中的 行坐标及列坐标、 解交织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出数据的编号对应的待解交织码流数据地址。
所述地址获取模块 50获取所述待输出数据的编号对应的待解交织码流 数据地址具体可以按照下列公式计算: 解交织后码流中的待输出数据编号 对应的待解交织码流数据地址=所述列坐标 X解交织前矩阵的行数 +所述行 坐标-所述列坐标之前的列中的哑元数。
码流形成模块 60, 用于依次从与待输出数据编号对应的待解交织码流 数据地址读取数据, 形成解交织后码流。
所述待解交织码流可以为接收到的数据流分流后的三路码流中的一路 码流、 或接收到的数据流分流前与一路码流对应的数据串。 则地址获取模 块 50还可以包括:
第一获取单元, 用于当所述待解交织码流为所述分流后的一路码流时, 根据所述待解交织码流数据地址, 直接获取另两路解交织后码流中各相应 待输出数据的编号分别对应的待解交织码流数据地址; 或
第二获取单元, 用于当所述待解交织码流为所述分流前与一路码流对 应的数据串时, 根据所述解交织后码流中各待输出数据的待解交织码流数 据地址和所述待解交织码流的有效信息长度, 获取另两路解交织后码流中 各待输出数据的编号对应的待解交织码流数据地址。
优选的, 所述解交织后码流为第 0路解交织后码流; 则所述第二获取 单元包括:
第一子单元, 用于对所述获取的地址增加 1 倍的所述有效信息长度, 获得第 1路解交织后码流中各待输出数据的编号对应的第 1路待解交织码 流数据地址;
第二子单元, 用于对所述获取的地址增加 2倍的所述有效信息长度, 获得第 2路解交织后码流中各待输出数据的编号对应的第 2路待解交织码 流数据地址。
所述装置还可以包括: 转换模块, 用于在解交织前, 将接收到的数据 流的码率直接转换为设定码率的输入码流。
本实施例的解交织装置, 根据待解交织码流的有效信息长度, 计算待 解交织码流需要虚拟添加的哑元数和虚拟添加哑元的位置, 但是并不真正 进行添加哑元的操作; 虚拟添加哑元, 并确定解交织后码流中各待输出数 据的编号, 然后通过待输出数据编号与解交织前矩阵中列坐标的映射关系 得到各待输出数据编号在解交织前矩阵中的列坐标, 根据待输出数据编号 在解交织前矩阵中的行坐标及列坐标、 解交织前矩阵的行数、 列坐标之前 的列中的哑元数, 获取待输出数据的编号对应的待解交织码流数据地址, 由此避免了解重复时添加哑元和输出解交织后码流前去掉哑元的过程, 缩 短了对需添加哑元数目较多且码率低的输入码流解交织的时间,增强了 CQI 译码的吞吐能力。
以上所述仅为本发明的较佳实施例而已, 并非用于限定本发明的保护 范围。 凡在本发明的精神和原则之内所作的任何修改、 等同替换、 改进等, 均包含在本发明的保护范围内。

Claims

权利要求
1、 一种解交织方法, 其特征在于, 包括:
根据待解交织码流的有效信息长度, 计算所述待解交织码流需要虚拟 添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在所述解交织前矩 阵中的列坐标;
通过所述需要虚拟添加哑元数, 确定解交织后码流在其前端虚拟添加 哑元后各待输出数据的编号;
由所述各待输出数据的编号得到解交织后码流中各待输出数据编号在 解交织前矩阵中的行坐标;
根据待输出数据编号与解交织前矩阵中列坐标的映射关系, 获得所述 各待输出数据编号在解交织前矩阵中的列坐标;
根据所述待输出数据编号在解交织前矩阵中的行坐标及列坐标、 解交 织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出数据 的编号对应的待解交织码流数据地址;
依次从与待输出数据编号对应的待解交织码流数据地址读取数据, 形 成解交织后码流。
2、 如权利要求 1所述的方法, 其特征在于, 所述待解交织码流为接收 到的数据流分流后的三路码流中的一路码流、 或接收到的数据流分流前与 一路码流对应的数据串。
3、 如权利要求 2所述的方法, 其特征在于, 在获取所述解交织后码流 中待输出数据的编号对应的待解交织码流数据地址后, 还包括:
若所述待解交织码流为所述分流后的一路码流, 则根据所述待解交织 码流数据地址, 直接获取另两路解交织后码流中各相应待输出数据的编号 分别对应的待解交织码流数据地址;
若所述待解交织码流为所述分流前与一路码流对应的数据串, 则根据 所述解交织后码流中各待输出数据的待解交织码流数据地址和所述待解交 织码流的有效信息长度, 获取另两路解交织后码流中各待输出数据的编号 对应的待解交织码流数据地址。
4、 如权利要求 3所述的方法, 其特征在于, 所述解交织后码流为第 0 路解交织后码流; 则所述获取另两路解交织后码流中各待输出数据的编号对应的待解交 织码流数据地址具体包括:
对所述获取的地址增加 1倍的所述有效信息长度, 获得第 1路解交织 后码流中各待输出数据的编号对应的第 1路待解交织码流数据地址;
以及, 对所述获取的地址增加 2倍的所述有效信息长度, 获得第 2路 解交织后码流中各待输出数据的编号对应的第 2路待解交织码流数据地址。
5、 如权利要求 1-4任一项所述的方法, 其特征在于, 所述获取待输出 数据的编号对应的待解交织码流数据地址按照如下公式计算:
解交织后码流中的待输出数据编号对应的待解交织码流数据地址 =所 述列坐标 X解交织前矩阵的行数 +所述行坐标-所述列坐标之前的列中的哑 元数。
6、 如权利要求 1-4任一项所述的方法, 其特征在于, 所述根据待输出 数据编号与解交织前矩阵中列坐标的映射关系, 获得解交织后码流中各待 输出数据编号在解交织前矩阵中的列坐标具体包括:
通过预置的逆向列映射表获得各待输出数据编号在解交织前矩阵中的 列坐标。
7、 如权利要求 1-4任一项所述的方法, 其特征在于, 所述根据待输出 数据编号与解交织前矩阵中列坐标的映射关系, 获得解交织后码流中各待 输出数据编号在解交织前矩阵中的列坐标具体包括:
将所述解交织后码流中各待输出数据编号对应的五位二进制数倒序重 排并高位取反, 得到各待输出数据编号在解交织前矩阵中的列坐标。
8、 一种解交织装置, 其特征在于, 包括:
计算模块, 用于根据待解交织码流的有效信息长度, 计算所述待解交 织码流需要虚拟添加的哑元数、 解交织前矩阵的行数和虚拟添加的哑元在 所述解交织前矩阵中的列坐标;
编号模块, 用于通过所述需要虚拟添加哑元数, 确定解交织后码流在 其前端虚拟添加哑元后的各待输出数据的编号;
行坐标获取模块, 用于由所述各待输出数据的编号得到解交织后码流 中各待输出数据编号在解交织前矩阵中的行坐标;
列坐标获取模块, 用于根据待输出数据编号与解交织前矩阵中列坐标 的映射关系, 获得所述各待输出数据编号在解交织前矩阵中的列坐标; 地址获取模块, 用于根据所述待输出数据编号在解交织前矩阵中的行 坐标及列坐标、 解交织前矩阵的行数、 所述列坐标之前的列中的哑元数, 获取所述待输出数据的编号对应的待解交织码流数据地址;
码流形成模块, 用于依次从与待输出数据编号对应的待解交织码流数 据地址读取数据, 形成解交织后码流。
9、 如权利要求 8所述的装置, 其特征在于, 所述待解交织码流为接收 到的数据流分流后的三路码流中的一路码流、 或接收到的数据流分流前与 一路码流对应的数据串。
10、 如权利要求 9所述的装置, 其特征在于, 所述地址获取模块还包 括:
第一获取单元, 用于当所述待解交织码流为所述分流后的一路码流时, 根据所述待解交织码流数据地址, 直接获取另两路解交织后码流中各相应 待输出数据的编号分别对应的待解交织码流数据地址; 或
第二获取单元, 用于当所述待解交织码流为所述分流前与一路码流对 应的数据串时, 根据所述解交织后码流中各待输出数据的待解交织码流数 据地址和所述待解交织码流的有效信息长度, 获取另两路解交织后码流中 各待输出数据的编号对应的待解交织码流数据地址。
11、 如权利要求 10所述的装置, 其特征在于, 所述解交织后码流为第 0路解交织后码流;
则所述第二获取单元包括:
第一子单元, 用于对所述获取的地址增加 1 倍的所述有效信息长度, 获得第 1路解交织后码流中各待输出数据的编号对应的第 1路待解交织码 流数据地址;
第二子单元, 用于对所述获取的地址增加 2倍的所述有效信息长度, 获得第 2路解交织后码流中各待输出数据的编号对应的第 2路待解交织码 流数据地址。
12、 如权利要求 8-11任一项所述的装置, 其特征在于, 所述列坐标获 取模块包括:
第二列坐标获取单元, 用于将所述解交织后码流中各待输出数据编号 对应的五位二进制数倒序重排并高位取反, 得到各待输出数据编号在解交 织前矩阵中的列坐标。
PCT/CN2011/070764 2010-02-02 2011-01-28 一种解交织方法和装置 WO2011095115A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010111391.6 2010-02-02
CN 201010111391 CN101800625B (zh) 2010-02-02 2010-02-02 一种解交织方法和装置

Publications (1)

Publication Number Publication Date
WO2011095115A1 true WO2011095115A1 (zh) 2011-08-11

Family

ID=42596135

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/070764 WO2011095115A1 (zh) 2010-02-02 2011-01-28 一种解交织方法和装置

Country Status (2)

Country Link
CN (1) CN101800625B (zh)
WO (1) WO2011095115A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101800625B (zh) * 2010-02-02 2013-03-20 上海华为技术有限公司 一种解交织方法和装置
WO2012155458A1 (zh) * 2011-05-13 2012-11-22 中兴通讯股份有限公司 一种交织或解交织的实现方法和装置
CN102624484B (zh) * 2012-03-02 2014-12-10 华为技术有限公司 交织实现方法和设备
CN109510688B (zh) * 2018-12-18 2021-06-22 京信通信系统(中国)有限公司 速率匹配方法、解速率匹配方法、装置和基站
CN109768843B (zh) * 2018-12-18 2021-09-03 京信网络系统股份有限公司 速率匹配方法、解速率匹配方法、装置和基站
CN112731282B (zh) * 2020-12-09 2023-12-26 浙江工业大学 一种基于虚拟子阵交织的声阵列空间滤波方法及系统
CN115225203B (zh) * 2022-06-08 2024-04-12 芯翼信息科技(上海)有限公司 一种数据的解交织方法、装置、电子设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006063210A2 (en) * 2004-12-09 2006-06-15 Conexant Systems, Inc. Method and system for dynamic interleaver adaptation scheme
JP2006295861A (ja) * 2005-04-05 2006-10-26 Ist Kk 移動体通信プロトコル解析装置のためのデインターリーブ方法
CN101388751A (zh) * 2008-10-28 2009-03-18 重庆重邮信科通信技术有限公司 解速率匹配方法
CN101510781A (zh) * 2009-03-24 2009-08-19 华为技术有限公司 交织和解交织处理中哑元的填充方法和装置及处理系统
CN101800625A (zh) * 2010-02-02 2010-08-11 上海华为技术有限公司 一种解交织方法和装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006063210A2 (en) * 2004-12-09 2006-06-15 Conexant Systems, Inc. Method and system for dynamic interleaver adaptation scheme
JP2006295861A (ja) * 2005-04-05 2006-10-26 Ist Kk 移動体通信プロトコル解析装置のためのデインターリーブ方法
CN101388751A (zh) * 2008-10-28 2009-03-18 重庆重邮信科通信技术有限公司 解速率匹配方法
CN101510781A (zh) * 2009-03-24 2009-08-19 华为技术有限公司 交织和解交织处理中哑元的填充方法和装置及处理系统
CN101800625A (zh) * 2010-02-02 2010-08-11 上海华为技术有限公司 一种解交织方法和装置

Also Published As

Publication number Publication date
CN101800625A (zh) 2010-08-11
CN101800625B (zh) 2013-03-20

Similar Documents

Publication Publication Date Title
WO2011095115A1 (zh) 一种解交织方法和装置
US8433987B2 (en) Method for high-efficient implementation of de-rate matching including HARQ combining for LTE
US7954016B2 (en) Efficient multi-symbol deinterleaver
US8843799B2 (en) Serial processing method, parallel processing method of bit rate matching and device thereof
WO2011026375A1 (zh) 编码方法及装置、解码方法及装置
WO2014090045A1 (zh) Td-scdma上行传输信道处理方法
WO2011137749A1 (zh) 一种lte下行业务信道解速率匹配的方法及装置
WO2009152770A1 (zh) 一种第二次交织及解交织的方法和装置
WO2018090629A1 (zh) 一种解交织解速率匹配的方法、装置及计算机存储介质
US10110349B2 (en) Rate dematching method, apparatus and receiving-side device
WO2012034097A1 (en) Accessing memory during parallel turbo decoding
WO2017011946A1 (zh) 基于不等差错保护的数据传输方法、装置和设备
JP2023508449A (ja) 復号化方法、装置、ネットワークデバイス及び記録媒体
WO2012034398A1 (zh) 维特比解码实现方法及装置
WO2014048351A1 (zh) 时分复用的前向纠错编码方法及装置
TW201141080A (en) Error correction code encoder, error correction code decoder, interleaving/de-interleaving method and soft-in/soft-out decoding method
WO2002099994A1 (en) System and method for interleaving data in a wireless transmitter
WO2012149741A1 (zh) 一种解速率匹配的解交织方法和装置
CN103544113A (zh) 基于数据块的卷积交织在fpga中ram的实现方法
WO2008028419A1 (fr) Procédé et système d&#39;entrelacement/désentrelacement dans un système de communication
EP2323302A1 (en) Improved HARQ
CN112929127B (zh) 一种用于5g nr并行解交织和解速率匹配的方法和装置
Huo et al. Rate Matching and Interleaved Hardware Sharing Design
CN116015546B (zh) 一种基于FPGA的任意长度turbo码速率匹配方法
WO2011091672A1 (zh) 一种速率匹配实现方法和系统

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: 11739394

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: 11739394

Country of ref document: EP

Kind code of ref document: A1