KR100811376B1 - Method for coding and decoding in a communication system - Google Patents

Abstract

The present invention relates to encoding and decoding for correcting errors, etc., generated during transmission and reception of data in a communication system.

In the data encoding method of the present invention, a plurality of initial states and / or a plurality of final states are formed in convolutional code to be included in a portion of the initial state and / or the final state. It is characterized by the transmission including the additional information.

In addition, in the present invention, the decoding method may set the PM value of 0 when the initial state is used as additional information when time t = 0, and / or the PM value of the initial state which is not used when time t = 0 is infinite. An initialization step of setting; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; Selecting the last survival path of the survival path when the time t reaches the end of the decryption window length, selecting a path having the minimum PM value among the final state determined to use as additional information; It is characterized by including.

Description

Method for coding and decoding in a communication system

1 is an encoder of a convolution code including a shift register.

2 is a trellis diagram in a general case where both initial state and final state are 0

3 is a frame structure with tail bits for sending the final state to zero at the end of the frame.

4 is a frame structure of a tail biting convolutional code carrying additional information in tail bits

5 illustrates a conventional case of the present invention, and a trellis diagram of a convolutional code showing one embodiment of various examples when the initial state is 0 is illustrated.

FIG. 6 is a diagram illustrating shift register (SR) values starting at all zeros in order to start when the initial state of FIG. 5 is zero.

7 is a diagram illustrating a trellis diagram when the initial state of the present invention is plural.

FIG. 8 is a diagram illustrating a shift register state for starting in the initial state of FIG. 7, that is, when the initial state is all 0 or all 1. FIG.

9 illustrates a conventional case of the present invention, and a trellis diagram of a convolutional code showing one embodiment of various examples when the final state is 0                 

FIG. 10 is a diagram illustrating that tail bits are all terminated at zero in order to end when the final state of FIG. 9 is zero.

11 is a diagram showing a trellis diagram when the final state of the present invention is plural;

FIG. 12 illustrates a tail bit state for ending in the final state of FIG. 11, that is, when the final state is all zeros or all ones.

FIG. 13 is a table summarizing the state classification of convolutional codes and a method for setting each state. FIG.

14 is a table summarizing the state differences between the conventional method and the present invention.

FIG. 15 is a table showing bits to which additional information shown in FIG. 14 can be added. FIG.

16 is a block diagram of a Viterbi decoder which is a conventional convolutional code decoder applied to the prior art and the present invention.

17 is a flowchart illustrating a decoding process of a conventional Viterbi Decoder repeatedly performed for one frame in consideration of a time relationship.

18 is a convolutional code decoding flowchart using an initial state

19 is a convolutional code decoding flow chart using the final state

20 is a convolutional code decoding flowchart using both an initial state and a final state

Fig. 21 is a flowchart of division decoding in convolutional codes using state pairs of initial and final states.

The present invention relates to encoding and decoding for correcting an error occurring when a data is transmitted or received through a transmission line in a communication system. In particular, a shift register state value in a convolutional coding method is used. When the initial state or the final state is set to a plurality of cases, and the initial state or the final state value is set to be the same, only one piece of information is sent, so that other information is transmitted to the surplus portion to improve the performance of the system. A channel coding and decoding method using a coding method.

First, the coding method will be described.

In general, a convolutional coder, a coding scheme for correcting errors occurring during transmission of information in a communication system, includes a shift register for storing and moving input information. The data is input to the input terminal, and the input data moves the shift registers one by one by a clock, and the output data is made by adding the input data and the respective data using the output of each shift register to produce one data. When input, two bits are output from the two output terminals.

In general, State diagram, Tree Diagram, and Trellis Diagram are used for encoding, and Maximum Likelihood decoding and viterbi algorithms are used for decoding.

Hereinafter, the configuration and operation of the conventional convolutional encoding will be described.

Conventional convolutional codes are encoded and decoded by one frame.

In addition, an encoder is composed of a shift register. At this time, the initial state of the shift register is normally set to all zeros, and specific bit strings are put at the end of the frame to make the shift register end state constant.

As above, specific bit strings that are put at the end of the frame are called tail bits. In convolutional code, if all zeros are inserted, shift shift is automatically set to zero.

However, if the above method is used, the information corresponding to the tail bits cannot be transmitted, resulting in waste in terms of data rate. Therefore, the proposed method to overcome the above points is tail biting convolutional code.

The tail biting convolutional code has an advantage of increasing data rate by inserting information into tail bits, but additional decoding methods such as an increase in the number of decoding during decoding should be used. The disadvantage is that system performance is poor.

As described above, the encoder of the convolution code includes a shift register (SR) as shown in FIG. 1.

In addition, the state of the codeword may be represented by a trellis diagram. FIG. 2 is a trellis diagram in a general case in which both an initial state and a final state are zero.

As shown in the figure, FIG. 1A shows a case in which the hatching rate is 1/2 and the constraint length K is 3, while FIG. 1B shows a coding rate 1/3 and a constraint length K is 3 persons. It's right.

Here, the coding rate indicates how many bits are coded for one bit of the original bit, and the 1/2 coding rate means that one bit of the original bit is coded into two bits.

In addition, the constraint length K is a parameter representing a coding characteristic and means a register number of an encoder. That is, the number of registers + 1 represents K.

The trellis diagram above is one of several methods for interpreting convolutional codes, and it is an expression to show how the encoding process works.

3 is a frame structure with tail bits for sending a final state to 0 at the end of a frame.

However, in the conventional convolutional code as described above, the initial state of the SR (shift register) is set to 0 to fix the initial start point, and the shift register is set to 0 while the last point is fixed by attaching tail bits to fix the last point. To be

However, the conventional method does not put any information on the transmitted tail bits and is wasted.                         

Therefore, in order to prevent such waste, a tail biting convolutional code that carries additional information on tail bits is proposed in FIG. And the amount of memory and complexity also increases.

Therefore, in order to solve the above-mentioned conventional problem, the present invention configures a plurality of initial states and / or a plurality of final states in a convolutional code, where the state values are the same. The present invention proposes an encoding method capable of transmitting power control information, which is additional information, at least one bit or more, and a decoding method.

In the data encoding method of the present invention, a plurality of initial states and / or a plurality of final states are formed in convolutional code to be included in a portion of the initial state and / or the final state. It is characterized by the transmission including the additional information.

 In addition, when using a plurality of initial state or a plurality of final state in the convolutional code of the present invention, each shift register value corresponding to the state is formed so that additional information can be transmitted.

Also, in the present invention, the decoding method may set the PM value of 0 when the initial state is used as additional information when time t = 0 and / or set the PM value of the initial state which is not used when time t = 0 is infinite. An initialization step of performing;

The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating;

Selecting the last survival path of the survival path when the time t reaches the end of the decryption window length, selecting a path having the minimum PM value among the final state determined to use as additional information; It is characterized by including.

In addition, the present invention provides information that can be added to the shift register value of the initial state and / or the final state, the format indicator for providing information of the frame format (frame format) of the currently transmitted frame or frame structure information of another channel A format indicator or a power control indicator that provides information of power control or an ACK / NACK indicator that provides information or an indicator indicating a signal-to-noise ratio (SIR) or a selectable modulation / coding method for indicating channel status. It characterized in that it comprises at least one.

Other objects and features of the present invention will become apparent from the detailed description of the embodiments with reference to the accompanying drawings.

Hereinafter, an encoding and decoding method according to the present invention will be described with reference to the accompanying drawings.

First, a coding method will be described. In the present invention, a convolutional code using an initial state, a convolutional code using a final state, and a case in which each initial / final state is plural in the case of using both an initial state and a final state Divided into to transmit additional information that is the object of the present invention.

FIG. 5 illustrates a conventional case of the present invention, which is a trellis diagram of a convolutional code showing one embodiment of various examples when the initial state is 0, which is another embodiment of FIG. 2.

FIG. 6 illustrates that the shift registers SR all start at all zeros to start when the initial state of FIG. 5 is zero.

The conventional coding schemes of FIG. 5 and FIG. 6 have the disadvantage that additional information cannot be transmitted separately.

Therefore, as shown in FIG. 7, the trellis diagram in the case where the initial state is plural is proposed.

That is, it is a method of using all the initial states without fixing the initial state to one state. In this case, the number of initial states to be used is determined in consideration of performance and environment of use.

The simplest example of using this initial state is to use two cases where the initial state is all zeros and the initial state is all ones. Examples of trellis diagrams at this time are shown in FIG. 7.                     

FIG. 8 is a diagram illustrating a shift register state for starting in the initial state of FIG. 7, that is, when the initial state is all zeros or all ones.

As shown in the figure, it can be seen that the shift register initial values all start at 0 or 1.

Unlike the case of FIGS. 7 and 8, when three or more initial states are used, the shift register may be initialized to three or more initial states to give each initial state.

The above description is explained further.

In the present invention, a state value eventually corresponds to a value read by one bit of each value contained in the shift register.

That is, if the shift register currently has a value of 010110, the current state is 010110, where the initial state represents the first state of the shift register value, and the final state means the last state of the shift register at the end of one frame. .

Thus, the state value means the values of the shift registers of the encoder.

Therefore, if the initial state is to start at 00000, the shift register should have a value of 00000 when the first shift register is started, and if the initial state is to be set to 010010, the first shift register can be set to 010010.

In general, the simplest way of initial state is to write all zeros according to the conventional method. If you use several initial states, mathematically, you can use as many as 2 ^{shift registers} .

Initial state is possible.

In addition, the final state means the shift register value in the last encoding stage. In order to match the final state, the value cannot be forcibly inserted, so the tail bits are put at the end of the frame so that the final state becomes the value of the tail bits.

Therefore, as in the case of the initial state, if the final state is to be 001100, for example, the tail bits at the end of the frame may be 001100.

In addition, blind rate detection means that the receiver detects the transmission rate by itself, even when there are several types of transmission rates, even though the transmitting side does not inform the receiving side.

In the above, the encoding method using the initial state has been described, and convolutional code using the final state will be described below.

FIG. 9 illustrates a conventional case of the present invention, which is a trellis diagram of a convolutional code showing one embodiment of various examples when the final state is 0, which is another embodiment of FIG.

FIG. 10 illustrates that all tail bit values end at all zeros in order to end when the final state of FIG. 9 is zero.

9 and 10 have a disadvantage in that additional information cannot be transmitted separately.

Therefore, as shown in FIG. 11, the trellis diagram in the case where the final state is plural is proposed.

FIG. 12 is a diagram illustrating a tail bit state for ending in the final state of FIG. 11, that is, when the final state is all zeros or all ones.

As can be seen in the figure, it can be seen that the tail bits state all ends at 0 or 1.

Unlike the case of FIGS. 11 and 12, when three or more final states are used, tail bits may be set to the corresponding three or more final states to give each final state.

Hereinafter, convolutional code using both initial and final states will be described.

That is, at least four cases may occur in the embodiment of the present invention by using both the state of FIGS. 7, 8 and 11, 12.

The present invention is summarized according to each state below.

Fig. 13 summarizes the state classification of convolutional codes and a method for setting each state.

As shown in the figure, the shift register state of the convolutional encoder is divided into an initial state and a final state. The method of setting each state is to set an initial value of the shift register in the initial state. In the final state, the tail bit of the shift register is set.                     

Figure 14 summarizes the state differences between the conventional method and the present invention.

As shown in the figure, in the prior art, there was only one state value in which the initial state and the final state of the shift register were both zero or all ones.

However, in the present invention, at least two cases (all 0's or all 1's) may exist and state values up to 2 ^k-1 cases may exist.

K denotes the constraint length which is the number of registers.

FIG. 15 illustrates bits for adding additional information shown in FIG. 14.

As shown in the figure, if only the initial state and the final state are used independently, at least 1 bit of information can be added, and if all of the final state other than the initial state is used, at least 2 bits of information can be added.

The additional information of FIG. 15 will be described in detail.

For example, in the case of using the initial state, assuming that the initial state can have two cases of all 0's and all 1's, one piece of information if the receiver can figure out one of the two initial states. Is the result of sending.

In the above, one piece of information means that one bit of information is eventually sent, and similarly, when the initial state may have four cases (extension of FIG. 7), two of four initial states on the receiving side as in the above method are used. If you can find a dog, you have two pieces of information.                     

This means that two bits of additional information can be sent. This way, you can think of ways to send multiple bits of information using the initial state.

The same applies to the initial state in the case of the final state.

On the other hand, in the case of using both the initial / final state, it is possible to at least two initial state, two final state is possible. Then, the combination of initial state and final state is 2 * 2 = 4, and there are four possible cases.

Therefore, when both initial and final states are used, at least 2 bits of information can be additionally sent.

Secondly, the decoding method will be described. The present invention relates to a method of using an initial state and a final state that are not currently used and are fixed, and to decoding a convolutional code using state pairs of initial and final states.

The decoding method will be described together with the present invention and the present invention.

16 shows a block diagram of a Viterbi decoder, which is a conventional convolutional code decoder applied to the prior art and the present invention.

The convolutional code decoding method uses Maximum Likelihood Decoding (MLD). The MLD method is an optimal decoding method that finds the most similar format for a given input and decodes it, and the Viterbi algorithm is used as the MLD method.

As shown in the figure, the Viterbi algorithm performs MLD on the received data through variables called BM (Bran Metric Metric) and PM (Path Metric).

The BM is defined as the Hamming distance by obtaining the difference between the reference data generated by the state diagram and the transmitted data.

The PM stores the sum of consecutive BMs delivered through the survival pathway.

In the above description, the survival path is referred to as the survival path by the decoding result decoded using the ML method.

The decoding process adds PM and BM for each state of the state transition diagram, selects a portion having a small value, stores it in the PM, and stores the path. (ACS) (Add-Compare-Select), (Storage of Survival Path), (PM Search).

The route data collected through the above process is traced back after a certain time (TB) (Trace Back) to find the minimum route.

At this time, the selected path is output as a decoding result. (Decoded Data).

The decoding process of the conventional Viterbi Decoder is repeatedly performed for one frame, and the flowchart is shown in FIG. 17 in consideration of the time relationship.

As shown in the figure, the process proceeds in the order of initialization, repetition and final decoding.

However, the conventional convolutional code is a conventional Viterbi decoder of FIG. 16 is an optimal decoder. However, in the case of the convolutional code using the initial and final state, the conventional Viterbi decoder should be changed.

Normally, the start of one frame of convolutional code starts with the state of all shift registers being zero at the time of encoding, so the first state is zero.

Therefore, in the initialization process at the time of decoding, when the initial state is 0, the PM is 0 and other non-zero initial PMs are left at infinity.

Also, convolutional code usually appends tail bits with a value of zero to the end of a frame to bring the final state to zero. In these cases, the survival path is usually chosen as the path whose final state starts at zero.

However, in the case of convolutional codes using initial and final states, it is possible for the initial state to have a different state as well as zero. Therefore, not only PM when the initial state is 0, but also PMs of other states should be initialized to 0.

Also, in the case of convolutional codes using initial and final states, the final state may have not only 0 but also other states. The path having the smallest PM is selected among the paths of.

Hereinafter, the decoding method according to the present invention will be described.

1. Weaving code using the initial state (Fig. 18)

In the case of convolutional codes that use the initial state, the initial state is not only 0, but other states that are additionally determined to be used as the initial state. In this case, the decoding method needs only to change the initialization part in FIG. 17 which shows a conventional general decoding process.                     

That is, an initialization step of setting the PM value in the case of using the initial state as additional information at time t = 0 and / or setting the PM value in the initial state not in use at time t = 0 to infinity; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; Decoding step (TB) looking up the survival path when the time t reaches the end of the decoding window length (TB) to find the minimum path.

In addition, the initial state of the survival path obtained through the decoding process of FIG. 18 is determined as the transmitted initial state.

2. Weaving code using the final state (Fig. 19)

In the case of convolutional codes using the final state, not only the final state is 0 but also other states which are additionally determined to be used as the final state. In this case, the decoding method needs only to change the final decoding method, which is the last step in FIG. 17, which shows a conventional general decoding process.

That is, an initialization step of setting the PM value in the case of using the initial state as additional information at time t = 0 and / or setting the PM value in the initial state not in use at time t = 0 to infinity; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; Selecting the last survival path of the survival path when the time t reaches the end of the decryption window length includes selecting a path having a minimum PM value among the final states determined to be used as additional information; consist of.

Also, the final state of the survival path obtained through the decoding process of FIG. 19 is determined as the transmitted final state.

3. Convolutional code using both initial state and final state (FIG. 20)

20 is a decoding method in which both an initial state and a final state are used. In the conventional general decoding process of FIG. 17, both an initialization process and a final decryption process are changed.

That is, an initialization step of setting the PM value in the case of using the initial state as additional information at time t = 0 and / or setting the PM value in the initial state not in use at time t = 0 to infinity; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; Selecting the last survival path of the survival path when the time t reaches the end of the decryption window length includes selecting a path having a minimum PM value among the final states determined to be used as additional information; consist of.

In addition, the initial state and the final state of the survival path obtained through the decoding process of FIG. 20 are determined as the transmitted initial state and the final state.                     

4. Classification Decoding Method in Convolutional Codes Using State / End State Pairs (FIG. 21)

When the initial and final states are divided into state pairs and used, the PM values for each state pair are obtained by decoding each state pair separately. Among the PM values, the smallest state pair is selected, and information bits decoded based on the selected state pair are taken as the final decoded data.

The meaning of dividing the initial and final states into state pairs is to combine the two states into one pair, such as (initial state and final state), and assign one piece of information to this pair. There is no special standard.

For example, assuming that the initial states are all 0 and all 1, and the final states are all 0 and all 1, (initial state, final state) = (all 0, all 0), (all 1) , All 1), (All 0, All 1), (All 1, All 0) are available, and you can use any one of them.

In other words, if there is one additional bit number, if you select two state pairs (all 0, all 0) and (all 1, all 1), 1-bit information can be sent.

In addition, the method of calculating PMs for each state only needs to be modified at the beginning when setting the initial value of PM and selecting the last survival path.

For example, a conventional convolutional code has an initial state of all zeros and a final state of all zeros. Thus, a conventional Viterbi decoder only sets PMs with zero initial states to zero at the time of PM initialization, PM values are set to infinity.

This excludes infinite cases when choosing the minimum PM later.                     

In addition, when choosing the survival path at the final stage, the final state selects all paths ending in 0.

The operation in the initial state and the final state is fixed because the initial state and the final state are fixed to a specific value, and to obtain a PM of any arbitrary state pair, only the PM value equal to the initial state of the corresponding state pair is zero. Initialize and initialize the PM values of other states to infinity.

In addition, when selecting the survival path at the end, the path ending with the final state of the corresponding state pair is selected as the survival path.

Selecting the minimum PM value is the optimal decoding method. In any case, the decoding method is to obtain the PM value of each state pair, select the minimum PM value, and determine the state pair as the sent state pair.

A general decoding process at this time is shown in FIG. 21.

That is, a first step of setting each state pair by using an initial state and a final state;

Selecting a state pair not yet selected among the initial state and the final state and reading respective initial state and final state;

Set the PM value to 0 when the state is equal to the initial state of the state pair selected in step 2 when time t = 0 and / or the PM value other than the initial state selected in the step 2 when time t = 0 is infinite. Setting a third step;

The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating the fourth step;

The last survival path selection of the survival path when the time t reaches the end of the decryption window length is selected as the survival path by selecting the path having the smallest PM value among the final states in the state pair selected in step 2, and surviving. A fifth step of storing the decoded data and the PM value of the survival path for each of the state pairs selected in step 2 while going back through the path (TB);

And comparing a PM value of each state pair to select a minimum PM value, and determining the stored data for the selected state pair as decoded data.

In addition, the state pair of the initial state and the final state of the survival path obtained through the decoding process of FIG. 21 is determined as the transmitted state pair.

In the case where the information added in the present invention is provided as code words of an error detection code (CRC: Cyclic Redundancy Check) for detecting an error of information bits to be decoded, an error of information bits transmitted through the added error detection code Use this information to determine the presence of

In the present invention, when selecting a survival path, a path having a minimum PM value is selected as a survival path among the paths that are found to be error free by performing an error detection code (CRC) test on each path.

When selecting the survival path, if the error detection code (CRC) test is found to be error-free for the path having the minimum PM value, the survival path is selected. If the error detection code is found to be error, Secondly, the error detection code is checked for the path with the smallest PM value. At this time, if it is found that there is no error for the path with the smallest PM value, the path is selected as the survival path. If the error detection code check result shows that there is an error for the path with the minimum PM value, the error detection code check is performed for the third path with the lowest PM value. Repeat the above process until

As described above, in the present invention, when the initial state or the final state value is set to the same state by setting the initial state or the final state, which are shift register state values, in a convolution coding method, only one piece of information is provided. In the transmission method, the encoding method in the communication system for improving the performance of the system by transmitting other information to the surplus portion, and also in the case of using the initial state / final state or the initial / final state, respectively, in the decoding method. Convolutional code decoding using state pairs.

Although the preferred embodiment of the present invention has been described above, the present invention may use various changes, modifications, and equivalents. It is clear that the present invention can be applied in the same manner by appropriately modifying the above embodiments.

 Accordingly, the above description does not limit the scope of the invention as defined by the limitations of the following claims.

Therefore, according to the present invention, when the initial state or the final state, which is the shift register state value, is set to a plurality of cases, respectively, and the initial state or the final state value is the same, only one piece of information is sent. It can additionally transmit two or more pieces of information, and can effectively decode convolutional codes.

Claims (22)

In the system for transmitting / receiving data, In the convolutional code, at least one of a plurality of initial states or a plurality of final states is configured to transmit additional information including at least one portion of the initial state or the final state. Encoding method in a communication system, characterized in that. 2. The encoding method according to claim 1, wherein when a plurality of initial states or a plurality of final states are used in convolutional codes, respective shift register values corresponding to the states are formed to transmit additional information. Way. The method of claim 1, wherein the additional information of the plurality of initial states or the plurality of final states forms an initial state or tail bit with shift register values of the plurality of initial states or the final states as All 0 or All 1. Encoding method in communication system. The method of claim 3, wherein when one of the plurality of initial / final states All 0 or All 1 uses the same initial (0 or 1) initial / final state and the state value is known, at least one bit or more of information is added. Encoding method in a communication system, characterized in that the transmission. The method according to claim 3, wherein at least two bits of information are added when all the state values of 0 and 1 are used for all 0 or all 1 of the plurality of initial / final states. Encoding method in a communication system, characterized in that the transmission. 3. The communication system according to claim 2, wherein each shift register value represents an initial state or a final state, and each initial state or final state exists by 2 ⁿ (n = number of shift registers). Coding method. The format indicator according to any one of claims 1 to 5, wherein the information which can be added to the shift register value of the initial state and / or the final state provides information of a frame format of a currently transmitted frame. Or a format indicator for providing frame structure information of another channel, a power control indicator for providing power control information, or an ACK / NACK indicator for providing information, or a signal-to-noise ratio (SIR) for indicating a channel state, or And at least one indicator indicating a selectable modulation / coding method. [8] The communication system of claim 7, wherein when the added information is a frame rate, the communication system provides information on a transmission rate of the frame and uses the information as a rate indicator or as information for performing blind rate detection. Encoding method. 8. The method of claim 7, wherein when the added information is a power control indicator providing information of power control, the reverse link is the power control information of the forward link, the reverse link is A coding method in a communication system, characterized by indicating power control information of a reverse link. 8. The method of claim 7, wherein when the added information provides ACK or NACK, ACK or NACK information of the forward link in the reverse link, ACK or NACK information of the reverse link in the reverse link. Encoding method in a communication system, characterized in that. 8. The method of claim 7, wherein when the added information provides a signal-to-noise ratio (SIR) or a selectable modulation / coding method for indicating a channel state, the reverse link indicates the channel state of the forward link. And information for informing the state of the channel of the reverse link at the time of the reverse link. The information transmitted through the added error detection code according to claim 7, wherein the additional information is provided as code words of an error detection code (CRC: Cyclic Redundancy Check) for detecting an error of information bits to be decoded. Encoding method in a communication system, characterized in that used as information for determining the presence or absence of errors of bits In decoding the convolutional code using a Viterbi decoder, Initializing the PM value when the initial state is used as additional information at time t = 0 or at least one or more PM values of the initial state not used when time t = 0; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; And And a deciphering step (TB) looking up the survival path when the time t reaches the end of the decoding window length (TB). The decoding method of claim 13, wherein the initial state of the survival path obtained through the decoding process is determined as the transmitted initial state. In decoding the convolutional code using a Viterbi decoder, Initializing the PM value when the initial state is used as additional information at time t = 0 or at least one or more PM values of the initial state not used when time t = 0; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; and Selecting the last survival path of the survival path when the time t reaches the end of the decryption window length, selecting a path having the minimum PM value among the final state determined to use as additional information; Decryption method in a communication system comprising a. 16. The decoding method of claim 15, wherein the final state of the survival path obtained through the decoding process is determined as the transmitted final state. In decoding the convolutional code using a Viterbi decoder, Initializing the PM value when the initial state is used as additional information at time t = 0 or at least one or more PM values of the initial state not used when time t = 0; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating; And Selecting the last survival path of the survival path when the time t reaches the end of the decryption window length, selecting a path having the minimum PM value among the final state determined to use as additional information; Decryption method in a communication system comprising a. 18. The decoding method of claim 17, wherein the initial state and the final state of the survival path obtained through the decoding process are determined as the transmitted initial state and the final state. In decoding the convolutional code using a Viterbi decoder, A first step of setting each state pair by using an initial state and a final state; Selecting a state pair not yet selected among the initial state and the final state and reading respective initial state and final state; When time t = 0, the PM value is set to 0 when the state is equal to the initial state of the state pair selected in step 2, or when the time t = 0, the PM value which is not the initial state selected in step 2 is set to infinity A third step set to at least one case of; The time t is increased to calculate the BM value for the branch and the PM value for the path, and the PM value is compared and the survival path is found and stored with the PM value until the time t reaches the end of the decoding window length. Repeating the fourth step; The last survival path selection of the survival path when the time t reaches the end of the decryption window length is selected as the survival path by selecting the path having the smallest PM value among the final states in the state pair selected in step 2, and surviving. A fifth step of storing the decoded data and the PM value of the survival path for each of the state pairs selected in step 2 while going back through the path (TB); and And comparing a PM value of each state pair to select a minimum PM value, and determining the stored data for the selected state pair as the decoded data. 20. When selecting a survival path according to claim 15 or 17 or 19, selecting a path having a minimum PM value among paths that are found to be error-free by performing an error detection code (CRC) check on each path. A decoding method in a communication system characterized by 20. When selecting a survival path according to claim 15 or 17 or 19, if an error detection code (CRC) test is found to be error-free on a path having a minimum PM value, the survival path is selected. If an error is found in the path, the error detection code is checked for the path with the second minimum PM value. If the path with the second minimum PM value is found to be error-free, the path is saved. If it is found that there is an error on the path with the second minimum PM value, the error detection code is checked and the error detection code is checked for the path with the third minimum PM value. And repeating the above process until it is determined that there is no error. 20. The method of claim 19, wherein the state pairs of the initial state and the final state of the survival path obtained through the decoding process are determined as the transmitted state pairs.

Images