JP4591291B2 - Turbo decoding apparatus and method and program thereof - Google Patents

Description

  The present invention relates to a turbo decoding apparatus, a method thereof, and a program, and more particularly to an improvement of a turbo decoding method for decoding a high-performance and high-reliability turbo code used in the field of communication systems and information processing systems.

  As a technique often used in a communication system having a line state that changes with time with low reliability, an automatic repeat request (ARQ) method using packet retransmission control and an error correction decoding (FEC) ) There is a technique called hybrid ARQ (HARQ) in combination with a method (see Patent Document 1).

  A functional block diagram for retransmission in the HARQ technique is shown in FIG. In FIG. 5, when an error is detected in the packet by the CRC check unit 34 using a known cyclic redundancy check (CRC) code, the packet including the error at the receiver of the communication system is correctly decoded. In order to improve the probability, a retransmission request for the packet is made to the transmitter. In response to this retransmission request, the packet is retransmitted from the transmitter.

  In the receiver, the retransmission packet is demodulated by the demodulator 31, and the content of the retransmission packet and the content of the received packet before retransmission are combined by the HARQ combiner 32. Then, an error correction decoding process is performed on the output of the HARQ combiner 32 in the error correction decoder 33. As described above, in the HARQ scheme, the error correction capability of the receiver is improved by reusing it in the HARQ synthesizer 32 without discarding the previous received information.

  As an encoding method used for the error correction decoder 33, there is a turbo code method, which was proposed as an error correction code close to the Shannon limit by C. Berrou et al. This is a high-performance, high-reliability encoding method used in a wide range of communication fields and information processing fields including systems, information recording systems, digital broadcasting systems, and the like.

  First, a general turbo encoder and turbo decoder will be described. 6 and 7 are block diagrams showing the configuration of a known turbo encoder and turbo decoder. FIG. 6 is a schematic block diagram of a turbo encoder with a coding rate = 1/3, and includes element encoders E11 and E22 and an interleaver E21. An encoded bit sequence (Information Bit to be encoded) sequence E01 is branched and transmitted as a systematic bit sequence E02, and is input to an element encoder E11 and an interleaver E21.

  The element encoder E11 encodes the encoded bit sequence E01 using an error correction code, and outputs a redundant bit sequence (Parity Bit) E12. The interleaver E21 generally writes the encoded bit sequence E01 once in the memory and reads it out in a different order, thereby sending the data order to the element encoder E22 in a mixed manner. The element encoder E22 encodes the interleaved encoded bit sequence with an element code and sends a redundant bit sequence E23. As the element encoders E11 and E22, a recursive systematic convolutional encoder (RSC) is usually used.

  The turbo decoder shown in FIG. 7 includes element decoders (Decoders) D04 and D16, an interleaver D12, a deinterleaver D23, and a hard decision output block D25. The element decoder D04 receives a systematic information sequence (Systematic Information) D01 corresponding to the systematic bit sequence E02, a redundant information sequence (Parity Information) D02 corresponding to the redundant bit sequence E12, and external information (Extrinsic Information) D03. Is done. The output external information D11 is used for the next element decoder D16.

  Further, the obtained external information D11 and organization information series D01 are input to the element decoder D16 through the interleaver D12 together with the redundant information series (Parity Information) D14 corresponding to the redundant bit series E23. The soft output information D21 and external information D22 obtained by the element decoder D16 are sent to the deinterleaver D23.

  The deinterleaver D23 outputs the data in the reverse order to the data replacement order by the interleaver D12. That is, the interleaved soft output information D21 and external information D22 are returned to the order before the interleaving and output as soft output information D24 and external information D03. Further, the hard decision & output block D25 makes a hard decision on the soft output information D24 and outputs the final decoding result. The external information D03 is fed back to the element decoder D04 for the next decoding process.

  As described above, in the turbo decoder shown in FIG. 7, the decoding process is repeatedly performed while updating the external information D03 and D15 of the two element decoders. After a plurality of loops, the soft output information D24 is a hard decision. Is done.

  As a soft output decoding algorithm applied to a turbo decoding element decoder, it is said that a method using a maximum a posteriori probability decoding method (hereinafter referred to as MAP) is currently best. Since the device scale and the processing amount are significantly increased, Max is a simplified processing by determining whether the transmitted data is '1' or '0' based on the maximum likelihood value. A method using -Log-MAP (Max Logarithmic Maximum A Posteriori) is widely used.

  The MAP algorithm is a maximum likelihood decoding algorithm using a trellis diagram. FIG. 8A shows an example of the configuration of a certain element encoder, and shows a case where the number of registers is three. As shown in FIG. 8B, the trellis diagram represents the relationship between the output value when a certain value is input to the element encoder and the register state.

The MAP algorithm is roughly divided into the following three types of processing.
(A) Forward processing: The probability (forward path metric) of reaching each state at each time point from the top of the trellis diagram is calculated.
(B) Backward processing: The probability (backward path metric) of reaching each state at each time point from the end of the trellis diagram is calculated.
(C) Soft output generation processing and external value calculation: The soft output value of the tissue bit at each time point is calculated using the processing results of (a) and (b). Then, using this soft output value, external value calculation is performed.

  In the trellis diagram, the forward path metric and the backward path metric calculated in the forward process and the backward process at the time point t and the state s are expressed as Alpha (t, s) and Beta (t, s), respectively. Further, the probability of transition from the state s to the state s ′ at the time t is expressed as Gamma (t, s, s ′) (here, Gamma is called a branch metric). Gamma (t, s, s') is a probability obtained from a received value (organization information series, redundant information series, external information).

Each of the forward processing, backward processing, soft output generation processing, and external value calculation described above is executed as follows.
(A) Forward processing:
Alpha (t, s) = Max {Alpha (t−1, s ′)
+ Gamma (t, s ′, s)}
Here, the processing Max for obtaining the maximum value means that the processing is performed for all states s ′.

  As shown in FIG. 9A, Alpha (t, S3) for a state State (S3) at time t is calculated as follows. The branch metrics Gamma (t, S1, S3), Gamma (t, S2, S3) are added to the path metrics Alpha (t-1, S1) and Alpha (t-1, S2) of the two previous stages, respectively. The larger value is the Alpha value of the State (referred to as Alpha ACS (Add-Compare-Select) operation). This process is performed on all the states of all time transitions t, and the Alpha values of all the states are held.

  In the first calculation of Alpha, there is no previous Alpha value, so it is necessary to set an initial value. Here, since the transition always starts from State # 0 in the trellis diagram, the initial value of Alpha is defined as Alpha of State # 0 being 0, and other states being -MAX values (minimum value). .

(B) Backward processing:
Beta (t, s) = Max {Beta (t + 1, s ′)
+ Gamma (t + 1, s, s ′)}

  As shown in FIG. 9B, Beta (t, S4) for State (S4) at time t is calculated as follows. The branch metrics Gamma (t + 1, S4, S5) and Gamma (t + 1, S4, S6) are added to the path metrics Beta (t + 1, S5) and Beta (t + 1, S6) of the two subsequent stages. The larger value is the Beta value of the State (referred to as Beta ACS operation).

  In this process, Beta calculation processing is performed for all States of all time transition t from the opposite direction to Alpha (from the trellis final State side). Note that, at the first time of Beta calculation, there is no subsequent Beta value, so an initial value needs to be set. Here, at the final end of the trellis diagram, the initial value of Beta is defined as 0 for State # 0 and -MAX value (minimum value) for other states.

(C) Soft output generation processing and external value calculation:
The value of Alpha (t−1, s ′) that has already been calculated is added to Beta (t, s) and Gamma (t, s ′, s) to obtain the total path metric value at time t. Then, the difference between the “maximum path metric of the path whose decoding result is 0” and the “maximum path metric of the path whose decoding result is 1” is the soft output value at time t.

Taking FIG. 10 as an example, Alpha (4, s ′), Beta (5, s) and Gamma (5, s ′, s) for all combinations of (s, s ′) at time t = 5. ) Is added. Among them, the maximum path metric L0 (t) of the path whose decoding result is 0 is obtained. In this example,
L0 (t = 5) = Alpha (t = 4, state # 7)
+ Beta (t = 5, state # 6)
+ Gamma (t = 5, state # 7, state # 6)
It has become.

On the other hand, the maximum path metric L1 (t) of the path whose decoding result is 1 is also obtained. In this example,
L1 (t = 5) = Alpha (t = 4, state # 4)
+ Beta (t = 5, state # 0)
+ Gamma (t = 5, state # 4, state # 0)
It has become. And the soft output value of t = 5 is
L (t = 5) = L0 (t = 5) -L1 (t = 5)
Is calculated.

  In the Max-Log-MAP algorithm, the value obtained by subtracting the channel value (value obtained from the received value) and the prior value (external information given from the preceding decoder) from the soft output value (post value) obtained above. Becomes external information.

  As described above, in the ideal Max-Log-MAP algorithm, Alpha and Beta operations are performed on all decoding target data at once. However, in data transmission, as the data length increases, Max-Log-MAP decoding requires a huge memory area. In particular, a memory area for storing the path metric information of the entire trellis is required. Moreover, since the decoding process delay increases with the increase in decoding length, it is difficult to apply to an actual real-time system.

  Therefore, a method called a sliding window is widely used (see Non-Patent Document 1). In this method, only the likelihood information in the trellis for the window size is stored, and the window position is shifted until the decoding length is reached, so that a significant memory saving can be achieved.

JP 2004-253354 A A.J.Viterbi, “An Intuitive Justification and a Simplified Implementation of the MAP Decoder for Convolutional Codes,” IEEE J. Select. Areas Commun., Vol.16, pp.260-264, Feb. 1998.

In a general turbo decoding method that is not a sliding window method, it is necessary to set initial values of external values in order to calculate Alpha and Beta in the first decoding process (1st iteration, Decoder # 1). These values are all 0. However, in the case of HARQ retransmission control, the turbo external information before HARQ retransmission has a certain degree of reliability, so if it can be reused, better decoding performance can be obtained. When the decoding result satisfies a predetermined condition, if there is a decoding number control process in which iterative decoding is stopped at that time and a hard decision is made and the decoding result is output, the convergence of the turbo decoding process is It is thought that it is early.

  On the other hand, since the sliding window turbo decoding method processes in the order of windows, in the initial decoding process, the initial value of each of the sub-windows divided for each window is unstable, and thus is forcibly set. Decoding converges slowly, and there is a problem that decoding errors are concentrated at the divided node points (dividing points for each subwindow) due to window division. Therefore, in order to realize effective sliding window decoding, a method for improving the initialization of the decoding process has been an issue, and the same applies to the turbo decoding in the HARQ scheme described above.

  Accordingly, an object of the present invention is to provide a turbo decoding apparatus and method for improving turbo decoding performance during HARQ retransmission and reducing the number of times of turbo decoding and reducing power consumption. .

A turbo decoding device according to the present invention comprises:
A turbo decoding device in a receiving device configured to make a request for automatic retransmission of turbo-coded received data to a transmitting side,
When the received data retransmission, the external information obtained by the retransmission before the turbo decoding, saw including a decoding means for using as an initial value for the path metric calculation,
The decoding means includes
The turbo coding code block which is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding means for performing decoding processing in order from the first subcode block to the last subcode block,
The path metric in the backward direction of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating the path metric in the backward direction .
Another turbo decoding device according to the present invention is:
A turbo decoding device in a receiving device configured to make a request for automatic retransmission of turbo-coded received data to a transmitting side,
Decoding means for using external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when the received data is retransmitted;
The decoding means includes
The turbo encoded code block that is the received data is divided into a plurality of subcode blocks, and includes a sliding turbo decoding unit that performs decoding processing in reverse order from the last subcode block to the first subcode block,
A path metric in the forward direction of each node of the subcode block in the turbo decoding before the retransmission is used as an initial value for calculating the path metric in the forward direction.

The turbo decoding method according to the present invention comprises:
A turbo decoding method in a receiving apparatus configured to make a request for automatic retransmission of turbo-encoded received data to a transmitting side,
When the received data retransmission, the external information obtained by the retransmission before the turbo decoding, saw including a decoding step of using as an initial value for the path metric calculation,
The decoding step includes
The turbo encoded code block which is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding step for performing decoding processing in order from the first subcode block to the last subcode block,
The path metric in the backward direction of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating the path metric in the backward direction .
Another turbo decoding method according to the present invention is:
A turbo decoding method in a receiving apparatus configured to make a request for automatic retransmission of turbo-encoded received data to a transmitting side,
A decoding step of using external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when retransmitting the received data;
The decoding step includes
The turbo coding code block which is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding step for performing decoding processing in reverse order from the last subcode block to the first subcode block,
A path metric in the forward direction of each node of the subcode block in the turbo decoding before the retransmission is used as an initial value for calculating the path metric in the forward direction.

The program according to the present invention is:
A program for causing a computer to execute a turbo decoding method in a receiving apparatus configured to make an automatic retransmission request for turbo-encoded received data to a transmitting side,
When the received data retransmission, the external information obtained by the retransmission before the turbo decoding, saw including a decoding process to be used as an initial value for the path metric calculation,
The decryption process
The turbo encoded code block that is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding process that sequentially performs a decoding process from the first subcode block to the last subcode block,
The path metric in the backward direction of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating the path metric in the backward direction .
Other programs according to the present invention are:
A program for causing a computer to execute a turbo decoding method in a receiving apparatus configured to make an automatic retransmission request for turbo-encoded received data to a transmitting side,
A decoding process that uses external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when the received data is retransmitted;
The decryption process
The turbo encoded code block that is the received data is divided into a plurality of subcode blocks, and a sliding turbo decoding process that performs decoding processing in reverse order from the last subcode block to the first subcode block,
A path metric in the forward direction of each node of the subcode block in the turbo decoding before the retransmission is used as an initial value for calculating the path metric in the forward direction.

  According to the present invention, at the time of HARQ retransmission, the turbo external information (external value) before retransmission is not discarded, but is reused as initial information (external value initial value), thereby improving the decoding performance. Thus, it is possible to reduce the number of iterative decoding operations, thereby reducing the power consumption.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an embodiment of the present invention, in which the present invention is applied to a general turbo decoding apparatus that is not a sliding window turbo system.

  Here, turbo decoding means iterative decoding, and in one decoding process, as shown in FIG. 7, an element decoding process (Decoder # 1) using two element decoders D04 and D16 is used. , # 2), and extrinsic information obtained by the element decoder is passed to the next element decoder via the interleaver D12 or deinterleaver D23. The circuit is realized by a processing circuit. FIG. 1 shows the circuit block.

  Referring to FIG. 1, data after HARQ processing, which is an output of the HARQ processing unit 11 for HARQ processing described in FIG. 5, is stored in the turbo input RAM 12. In the turbo decoding process, using the data from the RAM 12 and the external information obtained by the previous element decoder, the Alpha calculation unit 13 and Beta calculation unit 14 perform Alpha calculation (forward processing). , Beta calculation (backward processing) is performed. However, in the case of HARQ retransmission, when the first turbo decoding (Decoder # 1) is started, the external information of the last round of turbo decoding stored in the memory 19 before the HARQ retransmission is set as an external value initial value. It is used in the Alpha calculator 13 and Beta calculator 14. Therefore, switching between the external value calculation unit 17 and the memory 19 is performed by the switch 20.

  Then, using the obtained Alpha value and Beta value, the LLR calculation unit 15 performs LLR (maximum log likelihood ratio, ie, soft output value) calculation. The LLR calculation result is input to the external value calculation unit 17 and external information is calculated. The external information, which is the calculation result, is stored in the memory 19 for use as an external value initial value at the time of retransmission by HARQ processing, and the external information of the next element decoder via the switch 20 and the interleaver / deinterleaver 21. Used as

  Further, in order to further improve the decoding efficiency, the decoding number control unit 18 controls the number of decodings. That is, when the soft output value satisfies a certain predetermined condition, iterative decoding is stopped and the hard decision unit 16 performs control so that a hard decision is made. This hard decision result is output as a decoding result.

  FIG. 2 is a flowchart showing the operation of the embodiment of FIG. Referring to FIG. 2, first, the data after HARQ processing in HARQ processing unit 11 is input to the turbo decoder (step S1), but when this data is not retransmission data by HARQ (No in step S2). The initial external values used in the first turbo decoding element decoder (corresponding to D04 in FIG. 7: Decoder # 1) are all 0 (step S3).

  When the data is HARQ retransmission (Yes in step S2), the external value initial value used in the first turbo decoding element decoder (Decoder # 1) is the turbo before HARQ retransmission stored in the memory 19 This is the final external information for decoding (step S4). Then, Alpha and Beta calculations are performed (step S5), and LLR (softening value) is calculated (step S6).

  If the current element decoding process is Decoder # 1 (No in step S7), external value calculation is performed (step S8), and the external value (external information) as the calculation result is the decoder # 2 in the next stage. It will be used as the initial value of the external value. If the current element decoding process is Decoder # 2 (Yes in step S7), it is determined whether or not the decoding count control process, that is, the soft output satisfies a predetermined condition (step S9), and if not satisfied (No in step S10) ), And the external information at that time is delivered to the element decoder 1 next time (next to the number of times of turbo iterative decoding) (step S8).

  If the soft output satisfies the predetermined condition (Yes in step S10), the decoding is stopped and the obtained external information is stored in the memory 19 (step S11). The stored external information is used as an external value initial value when HARQ retransmission is further required. Finally, a hard decision is made based on the obtained soft output, and the final result of turbo decoding is output (step S12).

  FIG. 3 is a block diagram of another embodiment of the present invention, in which the present invention is applied to a sliding window turbo decoding device. Parts equivalent to those in FIG. 1 are denoted by the same reference numerals. In sliding window turbo decoding, as described above, one turbo code block is divided into a plurality of equal-length (equal size) sub-windows (sub-code blocks), and one code block is sequentially assigned to each sub-code block ( Element decryption processing is performed on a subwindow basis.

  Therefore, the difference from FIG. 1 is that the Alpha calculation unit 13 in FIG. 1 becomes the Alpha calculation unit 13a of the subwindow in FIG. 3, and the Beta calculation unit 14 of FIG. 1 changes to the Beta calculation unit 14a of the subwindow in FIG. It has become. Further, a memory 22 for storing the Beta calculation result of the sub window calculation unit 14a and a switch 23 for supplying the output of the memory 22 to the Beta calculation unit 14a of the sub window as an initial value are added. The other configuration is equivalent to that of FIG.

  In FIG. 3, all data after HARQ processing is stored in the turbo input RAM 12, and Alpha and Beta are calculated for each subwindow. However, since the code block is divided into sub-windows that are sub-code blocks, the Beta initial value (in each sub-window) of each intermediate sub-window during the first turbo decoding (Decoder # 1, # 2). Since the last Beta value) is indefinite, the Beta value at the equivalent position of the element decoding retrieval is read from the memory 22 and used as the Beta initial value of each intermediate subwindow before HARQ retransmission.

  Note that the processes such as the external value initial value, LLR calculation, decoding count control, and hard decision are the same as in the example of FIG. When the decoding count control unit 18 determines that the decoding is completed, all code block external information and the Beta value of each subcode block node by code block division are stored in the memories 19 and 22, respectively, and further retransmission is necessary. In this case, it is reused as the initial value of turbo decoding.

  FIG. 4 is a diagram for explaining the operation of another embodiment of the present invention shown in FIG. 3, and is an operation of sliding window turbo decoding processing including the case of HARQ retransmission. At the time of HARQ retransmission of each sub-window, in the first turbo decoding element decoding 1 (Decoder # 1), external information and HA-target Beta information before HARQ retransmission for the sub-window are read from the memories 19 and 22, respectively. Used as an initial value for element decoding.

  Also, in element decoding 2 (Decoder # 2) of the first turbo decoding, Beta information of the turbo node before HARQ retransmission for the subwindow is read from the memory 22 and used as the Beta initial value of the subwindow. When the number of iterations of turbo decoding is 2 or more (after 2 decoding) (assumed to be N), the initial values of Alpha and Beta in the subwindow are as follows.

(1) For the first subwindow # 1 (1 to W in FIG. 4):
Alpha is the same as the setting of the ideal Max-Log-MAP algorithm, and Beta is the top Beta value of the next subwindow (W + 1 to 2W in FIG. 4) in the previous (N-1) iterative turbo decoding.

(2) For intermediate sub-window #i (1 <i <X) excluding the first # 1 and last #X sub-windows:
Alpha is the final Alpha value of the # (i-1) subwindow in the previous (N-1) iterative turbo decoding, and Beta is the top of the # (i + 1) subwindow in the previous (N-1) iterative turbo decoding. Beta value.

(3) For the last subwindow #X:
Alpha is the final Alpha value of the subwindow of # (X-1) in the previous (N-1) iterative turbo decoding, and Beta is the same as the setting of the ideal Max-Log-MAP algorithm.

  Also, external value initial values other than the first turbo decoding element decoding 1 (Decoder # 1) are set by passing information between the decoders # 1 and # 2 via the interleaver / deinterleaver.

  In the example shown in FIGS. 3 and 4, the subwindows are processed in this order from # 1 to #X in the sliding window turbo decoding. However, the subwindows are processed in this order from #X to # 1 ( In that case, the Alpha initial value of the node of each subwindow (the first of each subwindow) needs to read the Alpha value of the node before retransmission from the memory 22, as shown in FIG. In 3 and 4, Alpha and Beta are all reversed.

  It is obvious that the operation of each of the above embodiments can be configured such that the operation procedure is recorded in advance on a recording medium such as a ROM as a program and is read and executed by a computer.

  As described above, according to the present invention, in the case of HARQ retransmission, turbo external information before retransmission can be reused as initial information without being discarded, and decoding performance can be improved. In addition, when the turbo decoding frequency control process is used, the number of iterative decoding times can be reduced, which also saves power. Furthermore, since the initial value of Beta is undecided in the first sliding window turbo decoding by code block division, decoding performance can be improved using Beta Metric information before retransmission.

  Briefly, since the Beta value is calculated in the direction from the end to the beginning of the code block (backward direction), when decoding in code block order (from the first subwindow to the last subwindow), Beta values cannot be passed from the next subwindow (one behind). When the pre-initialization process is not performed, since the initial value is assumed to be a fixed value (for example, 0) in the first decoding process, the likelihood information (LLR) calculated based on this initial value setting is used. It differs from the result of the ideal Max-Log-MAP algorithm.

  That is, since the likelihood information becomes inaccurate, the performance of the iterative decoder also deteriorates. In the present invention, the Beta window information of each sub-window before retransmission is reused as an initial value, thereby improving the sliding window turbo decoding performance.

It is a block diagram which shows one embodiment of this invention. It is a flowchart which shows the operation | movement of embodiment of FIG. It is a block diagram which shows other embodiment of this invention. 4 is a timing chart showing the operation of the embodiment of FIG. It is a functional block diagram of retransmission control by HARQ. It is a functional block diagram of a turbo encoder. It is a functional block diagram of a turbo decoder. (A) is an example of the configuration of an element encoder, and is a diagram showing a case where the number of registers is 3, and (b) is a relationship between an output value and a register state when a certain value is input to this element encoder. FIG. (A) is a figure which shows the ACS calculation example of Alpha, (b) is a figure which shows the ACS calculation example of Beta. It is a figure which shows the example of calculation of a soft output value.

Explanation of symbols

11 HARQ processor 12 Turbo input RAM
DESCRIPTION OF SYMBOLS 13 Alpha calculation part 14 Beta calculation part 15 LLR calculation part 16 Hard decision part 17 External value calculation part 18 Decoding frequency control part 19,22 Memory 20,23 Switch 21 Interleaver / deinterleaver

Claims (10)

A turbo decoding device in a receiving device configured to make a request for automatic retransmission of turbo-coded received data to a transmitting side,
When the received data retransmission, the external information obtained by the retransmission before the turbo decoding, saw including a decoding means for using as an initial value for the path metric calculation,
The decoding means includes
The turbo coding code block which is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding means for performing decoding processing in order from the first subcode block to the last subcode block,
A turbo decoding device characterized in that a backward direction path metric of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating the backward direction path metric. . A turbo decoding device in a receiving device configured to make a request for automatic retransmission of turbo-coded received data to a transmitting side,
Decoding means for using external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when the received data is retransmitted;
The decoding means includes
The turbo encoded code block that is the received data is divided into a plurality of subcode blocks, and includes a sliding turbo decoding unit that performs decoding processing in reverse order from the last subcode block to the first subcode block,
A turbo decoding device characterized in that a forward path metric of each node of the subcode block in turbo decoding before retransmission is used as an initial value for calculating a forward path metric . The decoding means includes
Means for iterative decoding of the received data, and storage means for storing the external information at the time of the last decoding of the iterative decoding ,
The turbo decoding device according to claim 1 or 2 , wherein the storage information of the storage means is used as an initial value for calculating a path metric at the time of retransmission . The decoding means includes
Means for performing iterative decoding of each of the subcode blocks, and storage means for storing the path metric at the time of the last decoding of the iterative decoding,
The turbo decoding device according to any one of claims 1 to 3, wherein the storage information of the storage means is used as an initial value for calculating a path metric at the time of retransmission . A turbo decoding method in a receiving apparatus configured to make a request for automatic retransmission of turbo-encoded received data to a transmitting side,
A decoding step of using external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when retransmitting the received data;
The decoding step includes
The turbo encoded code block which is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding step for performing decoding processing in order from the first subcode block to the last subcode block,
A turbo decoding method characterized in that a backward direction path metric of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating a backward direction path metric. . A turbo decoding method in a receiving apparatus configured to make a request for automatic retransmission of turbo-encoded received data to a transmitting side,
When the received data retransmission, the external information obtained by the retransmission before the turbo decoding, saw including a decoding step of using as an initial value for the path metric calculation,
The decoding step includes
The turbo coding code block which is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding step for performing decoding processing in reverse order from the last subcode block to the first subcode block,
A turbo decoding method, wherein a forward path metric of each node of the subcode block in turbo decoding before retransmission is used as an initial value for calculating a forward path metric . The decoding step includes
Including iterative decoding of the received data, and storing the external information in the storage means at the time of the last decoding of the iterative decoding,
The turbo decoding method according to claim 5 or 6, wherein the storage information of the storage means is used as an initial value for calculating a path metric at the time of retransmission. The decoding step includes
Performing iterative decoding of each of the subcode blocks; and storing the path metric at the time of the last decoding of the iterative decoding in a storage means;
The turbo decoding method according to claim 5, wherein the storage information of the storage unit is used as an initial value for calculating a path metric at the time of retransmission . A program for causing a computer to execute a turbo decoding method in a receiving apparatus configured to make an automatic retransmission request for turbo-encoded received data to a transmitting side,
A decoding process that uses external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when the received data is retransmitted;
The decryption process
The turbo encoded code block that is the received data is divided into a plurality of subcode blocks, and has a sliding turbo decoding process that sequentially performs a decoding process from the first subcode block to the last subcode block,
A program characterized in that a backward direction path metric of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating the backward direction path metric. A program for causing a computer to execute a turbo decoding method in a receiving apparatus configured to make an automatic retransmission request for turbo-encoded received data to a transmitting side,
A decoding process that uses external information obtained by turbo decoding before retransmission as an initial value for path metric calculation when the received data is retransmitted;
The decryption process
The turbo encoded code block that is the received data is divided into a plurality of subcode blocks, and a sliding turbo decoding process that performs decoding processing in reverse order from the last subcode block to the first subcode block,
A program characterized in that a forward path metric of each node of the subcode block in the turbo decoding before retransmission is used as an initial value for calculating the forward path metric.

Images