KR20020004479A - Data Rate Matching Method for In 3GPP2 - Google Patents

Abstract

PURPOSE: A data rate matching method is provided to obtain a coding gain as much as a decrease in an available code rate in the case of using a data rate lower than the maximum data rate supported for the operation of a variable data rate in a supplement channel. CONSTITUTION: If 1/4 rate channel coding is carried out for a convolution code, this coded bit stream is outputted with a length of "L"(S30). In case that the length(L) of the coded bit stream is larger than block interleaver size(N), puncturing is carried out. If the length(L) of the coded bit stream is smaller than block interleaver size(N), however, only symbol repetition is carried out(S31). Then the next interleaving step is carried out(S32).

Description

Data Rate Matching Method for In 3GPP2}

The present invention relates to the next generation mobile communication, and more particularly, to a data rate matching method that can give the best performance in a multi-variable data rate or a variable data rate matching method when data is transmitted through an auxiliary channel on a physical layer.

In general, one of the channels for transmitting data of 3GPP2 is called a supplemental channel. This channel supports a relatively high data rate of medium speed or more. Among the operation modes of the supplemental channel, a mode called a variable data rate exists.

That is, in the 3GPP2 system, the base station schedules the transmission of the signal through the forward auxiliary channel among the supplemental channels, and the base station transmits a fixed data rate to the terminal for a predetermined time through the scheduling signal. Will be allocated. However, the channel condition between the base station and the specific terminal may change during that particular time, and the system load of the base station may also change.

In this case, the base station may not have enough transmission power to transmit to the specific terminal at the current data rate. In order to solve this problem, the base station may suspend transmission to the supplemental channel during this time.

However, this solution causes a delay problem in data transmission and also causes inefficient resource management since other terminals cannot use the available transmit power and Walsh code that are already allocated to the terminal.

Another solution in the prior art is to consider how the base station reschedules the transmission data rate after some time. However, this also causes problems such as time delay and waste of Walsh code.

This situation is not unique to the forward link. Similarly, even in the reverse link, the state of the channel between the terminal and the base station may change according to the movement of the terminal, and accordingly, a shortage of transmission power for maintaining proper quality may occur.

Therefore, in order to solve this situation, a mode called a variable data rate is conventionally used.

The multi-variable data rate mode is a method of changing the transmission rate in units of frames according to the channel condition. If the channel condition is determined to be poor, the base station lowers the transmission rate of the supplemental channel. If it is determined that the situation is good again, the transmission is performed at the previous transmission speed.

Specifically, the data rate of the supplemental channel is adjusted between the support rate up to two levels below the maximum rate supported for the current variable channel for the variable data rate.

At this time, if the forward channel is considered, the terminal side should determine the situation where the rate is changed through blind rate detection. Therefore, when the range in which the data rate is variable becomes wide, a problem of increasing the complexity of the terminal occurs.

In addition, the interleaver and Walsh code specified for the maximum transmission rate currently used must be used as it is.

Therefore, when the data rate is lowered to 1/2 of the maximum data rate, twice the symbol repetition is performed to match the length of the interleaver to be used in the channel and the length of the output string of the channel encoder. Likewise, if the data rate is lowered to 1/4 of the maximum data rate, four times symbol repetition is performed to match the length of the interleaver to be used in the channel and the length of the output string of the channel encoder.

If a variable data rate mode for a flexible data rate is supported in the supplemental channel, the channel is the same as the method used in the conventional flexible data rate. Iteratively repeat the output symbols of the encoder to make it equal to or greater than the length of the interleaver used at the maximum data rate, and then use a method of puncturing the bits beyond this length.

1 is a diagram illustrating a rate matching process for supporting a variable data rate and a variable data rate according to the prior art.

Referring to FIG. 1, the variable data rate mode for the supplemental channel may be wasteful in terms of required transmit power.

For example, consider the current forward fourth radio structure (Radio Configuration 4, hereinafter abbreviated as RC4). In this case, a half rate turbo code or a convolution code is used. Assuming that the maximum data rate that can be used in the variable data rate mode is 76.8 kbps, the length of the interleaver used in the forward RC4 is fixed to 3072.

In the case of using the variable data rate method in this mode, in the prior art, the available data rate is used as an appropriate value among {19.2kbps, 38.4kbps, 76.8kbps}, and the length of the interleaver is fixed to 3072. It is.

Suppose that the situation of the channel is not good in the state of transmitting at the maximum transmission rate of 76.8kbps, so that it is impossible to maintain the required quality of service with the current power.

At this time, the data rate is reduced to 1/2 to send data at a rate of 38.4 kbps. Then, 744 bits are dropped for 20 ms at 38.4 kbps, and 16-bit cyclic superposition code (CRC) and 8-bit tail portion are added to input the 768-bit frame to the 1 / 2-rate channel encoder. Then, the output of the channel encoder is 1536 bits, and M = 2 symbol repetition is performed to send it to the 3072 interleaver.

Here, the gain obtained through the symbol repetition is only the repetition gain, and in practice, the repetition may play a role of reducing the effective code rate, but due to this, only a slight increase in time diversity is achieved. You can get it, but it doesn't actually increase the coding gain.

That is, the problem at this time is that the effective code rate is reduced, but the coding gain is not increased accordingly.

In other words, assuming that a 1/2 rate channel code is used and symbol repetition of M = 2 is performed, the effective code rate can be determined to be 1/4. However, it is considered a problem of the prior art that there is almost no increase in coding gain.

Variable Data Rate mode is not a technology used in the Walsh code shortage state. This technique can be thought of as a technique that can efficiently adjust the transmission power to solve the load on the system.

Therefore, if the gain of the transmission power can be obtained by increasing the coding gain when the same Walsh code is used, it can be determined that this will be a more efficient way to adjust the load of the entire system. .

In the same way, even when applied to the supplemental channel of the reverse link, the transmission rate is automatically adjusted according to the channel conditions, so that data can be transmitted without re-scheduling when the transmission power of the terminal is insufficient. It seems to be a more efficient way.

Accordingly, an object of the present invention has been made in view of the above-mentioned problems of the prior art, and the data rate is lower than the maximum data rate supported in the operation of a variable data rate in a supplemental channel. In order to provide a data rate matching method capable of obtaining the coding gain according to the effect of reducing the effective code rate.

According to an aspect of the present invention for achieving the above object, in the process of interleaving the channel-coded data to be mapped to the physical layer, the channel-coded data is transmitted at a data rate other than the maximum data rate in the multivariate data rate mode Puncturing or symbolic repetition of the data of the channel-coded different bit rates for transmission at a channel code of the lowest available data rate, or in the case of a variable data rate mode being transmitted at an arbitrary data rate not defined on each radio structure. Is selectively used to match the interleaving size.

Preferably, a coding rate of 1/4 rate is used when the channel coded data is a convolutional code, and a coding rate of 1/5 rate is used when a turbo code is used.

In addition, when the length of the output string of the channel coded data is larger than the size of the desired interleaver, the puncturing amount obtained by subtracting the interleaver size from the length of the output string of the channel coded data is for each code group unit in the output bit strings of all channel encoders. It is characterized by puncturing.

Preferably, a code group that needs to be punctured by a maximum integer not exceeding a value obtained by dividing the total puncturing amount by the total number of code groups is a first type code group, and the total puncturing amount of the entire code group The code groups to be punctured by a minimum integer greater than or equal to the value divided by the number may be divided into second type code groups so that each is evenly arranged.

Preferably, the amount of puncturing allocated to each code group is included in each code group through a bit inversion operation on the index of the remaining bits except the systematic bits and addition of the total puncturing amount up to the group. It is characterized by determining the position of puncturing in.

On the other hand, regardless of the coding rate of the turbo code, the bit string of the entire channel output sequence is divided into a systematic bit group, a bit group for parity bits of the upper component encoder, and a bit group for parity bits of the lower component encoder. It is characterized by performing puncturing by dividing.

Advantageously, exclude the puncturing in the bit group for the systematic code bits, and overall puncturing for the bit group for the parity bits of the upper component encoder and the bit group for the parity bits of the lower component encoder. Distribute half of the amount.

Further, to determine which bit group each code bit belongs to from the sequential index i of the code bits of the channel coding sequence for the puncturing, the sequential index j in each bit group has a coding rate inverse of n The coding rate inverse of each component encoder is " If "," in _c It is determined by ".

Preferably, when the sequential bit index (i) of the channel coded sequence belongs to an upper component coder bit group, the sequential bit index is mapped to a sequential bit index (j) in the upper component coder bit group. When the sequential bit index i belongs to the lower component encoder bit group, the sequential bit index is mapped to the sequential bit index j in the lower component encoder bit group.

The method further performs uniform puncturing on the parity bits of each component encoder by checking the sequentially mapped bit indexes in each component encoder after the bit indexes sequentially mapped in each component encoder. It is done.

On the other hand, when the length of the output string of the channel encoder is smaller than the length of the desired interleaver, the amount of the length of the interleaver minus the length of the output string of the channel encoder is repeated at uniform intervals.

Advantageously, the iterative algorithm for the turbo code prioritizes the iterative process in a group for systematic code bits.

In addition, the iterative algorithm for the turbo code extracts the entire code bit stream regardless of the coding rate of the turbo code, for the system bit group, the group of parity bits of the upper component encoder, and the parity bits of the lower component encoder. It is characterized by performing the division into groups.

Further, the repetition algorithm compares the number of bits that should have one larger repetition factor with the number of systematic bits compared to other positions, and all these values until the value is smaller than or equal to the number of systematic bits. To the systematic bit group.

In addition, the repetition algorithm compares the number of bits that should have one larger repetition factor with the number of systematic bits compared to other positions, and if the value is larger than the number of systemic bits, the amount is larger. It is characterized by dividing and assigning half of each of the upper component encoder group and the lower component encoder group.

1 is a diagram illustrating a rate matching process for supporting a variable data rate and a variable data rate according to the prior art.

2 is a view showing the configuration of a general turbo code

3 is a flow chart of a channel of a variable data rate and variable data rate matching method of a turbo code according to the present invention;

4 is a flow chart of a channel of a multivariate data rate and variable data rate matching method of a convolutional code in accordance with the present invention.

Fig. 5 is a block diagram showing the configuration of a decoder of a cyclic scheme for a typical turbo code.

Hereinafter, a configuration and an operation according to an exemplary embodiment of the present invention will be described with reference to the accompanying drawings.

In the present invention, in a variable data rate mode, at a transmission rate other than the maximum transmission rate, when a turbo code is used, a turbo code of 1/5 rate is always promised.

For example, if the maximum data rate is 76.8 kbps in the multivariate data rate mode, if the transmission rate is lowered to 1/2 rate due to the deterioration of the channel environment, 744 bits are dropped for 20 ms, and 16 bits of cyclic superposition code (CRC) and 8 bits are added thereto. A frame of 768 bits is inputted to the channel encoder with a tail of.

In this case, the 1/5 rate turbo encoder receives 768-bit frames as input and outputs a total of 3840 bits as output. The length of this output string is greater than the length 3072 of the interleaver used in the current mode. Therefore, in this case, puncturing of 768 bits (= 3840-3072) should be performed on the output string of the turbo encoder. Therefore, the effective puncturing distance is 5 (= L / P = 3840/768).

As a result, rate matching may be performed by puncturing as much as 1 bit every 5 bits. In this case, the actual coding rate can be thought of as 1/4 rate, which makes it possible to obtain the coding gain itself rather than the time diversity gain.

As another example, when the data rate is lowered to a quarter rate, information from the upper side becomes 360 bits for 20 ms. The 16-bit cyclic superposition code (CRC) and 8-bit tail are added to this, and the length of one frame is 384 bits. In this case, if a turbo code of 1/5 rate is used, 384 bits are inputted from an upper layer, and 1920 (= 5 * 384) code bits are output. To adjust this value to the length of the interleaver of 3072, the 1152 (= 3072-1920) bits may be repeated.

In addition, the method of using the 1/5 rate turbo code may be applied to a variable data rate.

In this case, in the flexible data rate or the variable data rate, only uniform puncturing is used when the length of the output stream of the channel encoder is larger than that of the desired interleaver. If the length of the output string of the encoder is smaller than the length of the interleaver, only a small amount of uniform iterations is used.

It is possible to perform only the uniform puncturing or the uniform repetition, which is basically a code obtained through puncturing from the 1/2 rate, 1/3 rate, and 1/4 rate code as shown in FIG. Therefore, blind rate detection can be performed in the decoding process without any increase in hardware complexity.

In addition, in the present invention, when applied to a convolutional code, in a variable data rate mode, a convolutional code of 1/4 rate is always used regardless of RC at a transmission rate except the maximum transmission rate. Promise to be.

And as with the turbo code, it is promised to always use a convolutional code of 1/4 rate even at a flexible data rate.

In this case, as in the turbo code, a rate matching of one step such as always using puncturing or only symbol repetition for rate matching between the length of the output stream of the channel encoder and the desired length of the interleaver. matching) method.

That is, as shown in FIGS. 3 and 4, the rate matching block according to the present invention always performs only one of two processes of puncturing or repetition. The puncturing or repeated amount is "| N-L |" As long as it takes place.

Specifically, when the N-L value is positive, the block interleaver size becomes large. Therefore, the interleaver size is matched by symbol repetition. When the N-L value is negative, the block interleaver size is matched through puncturing.

Referring to FIG. 3, if a 1/4 rate channel encoding is performed on a convolutional code, this coded bit string is output in L length (S30).

If larger than the block interleaver size N, which is the code bit string, puncturing is performed. If smaller than N, only symbol repetition is performed (S31).

Therefore, the puncturing process, which was performed in the two steps, is performed through one step, and the next interleaving process is performed.

This operation is similarly performed in the case of using the turbo code in FIG.

4 differs from FIG. 3 in that the coding rate is different. As described above, data having 1/2 rate, 1/3 rate, and 1/4 rate are generally 1/5. This is because puncturing or symbol repetition is generated from a bit string having a rate.

In the present invention, first, the puncturing patterns of the data portion and the tail portion are defined as shown in Table 1 below to make a turbo code of 1/5 rate.

In general, since decoding of a turbo code is performed cyclically using two constituent decoders 50 and 52 as shown in FIG. 5, in order to show optimal performance of the two constituent decoders, The decoding performance of the two constituent decoders 50 and 52 should be balanced.

To this end, the present invention proposes a puncturing algorithm that can equally divide the total puncturing amount into parity bits of two constituent encoders 50 and 52. (1)

In addition, in the present invention, since the systematic bit plays an important role as compared to the parity bit in the iterative decoding, a puncturing algorithm for removing the puncturing for the systematic bit is proposed.

Then, we propose a puncturing algorithm that performs puncturing at uniform intervals on each parity bit so that each decoder operates at the maximum decoding performance.

In order to present a puncturing algorithm satisfying the above (1) to (3), first, a code bit string having a coding rate of 1/5 rate is divided into a code group consisting of five code bits. The code group is divided into two types of code groups having different bit patterns and used in the next algorithm. The total number of the separated different types of code groups is the length of the input bit string of the channel encoder, which is defined as I here.

In the equations defined below, I is the length of the cyclic superposition code (CRC) and the encoder input string including the tail, and the inverse of the coding rate is five.

In the following equations, L is the length of the output string of the encoder and has a value of "5 * I".

P is the amount of puncturing required for the interleaved bit strings encoded with the turbo code or the convolutional code.

From here Is the maximum integer not exceeding x.

Equation 1 is the number of bits punctured in the code group of the first type.

Where LCEIL {x} RCEIL is the smallest integer value greater than or equal to x.

Equation 2 is the number of bits punctured in the code group of the second type.

K _{1, P2} = P mod I

Equation 3 is the number of the second type code group in the entire code group.

K _{1, GR1} = P ₁

Equation 4 is the number of puncturing amounts P ₁ to be generated in the first type code group.

K _{1, GR2} = K _{1, GR1} + 1

Equation 5 is the number of puncturing amounts P ₂ to be generated in the second type code group.

Equation 6 represents the number of generation of the puncturing amount P ₂ up to the (i-1) th time, and is used as an offset value when the test is to be performed in the i th code group.

In the present invention, the following first algorithm is proposed by using Equation 1 to Equation 6.

As described above, the algorithm describes a puncturing algorithm using turbo codes.

for (i = 0; i <I; i ++) {

if (((i + 1) K _{1, P2} ) mod I <k _{1, P2} ) {

for (j = 1; j <5; j ++) {

if ((BRO ₂ (j-1) + O _i ) mod 4 ≤K _{1, GR1} ) puncture (5i + j) ^th bit

else {

for (j = 1; j <5; j ++)

if ((BRO ₂ (j-1) + O _i ) mod 4 <K _{1, GR1} ) puncture (5i + j) ^th bit

}

}

That is, the algorithm preferentially defines two amounts of puncturing, P ₁ and P ₂ , and assigns these values to code groups, respectively.

In this case, P ₂ is defined as the minimum integer greater than or equal to the value of the total puncturing divided by the length of the code group.

P ₁ is defined as the largest integer less than or equal to the total amount of puncturing divided by the length of the code group.

In order to satisfy the maximum uniformity, the positions of code groups to which P ₂ puncturing should be allocated are uniformly determined among the positions of all I code groups.

Next, if the test is performed through the above algorithm and it is determined to be a code group to perform P ₂ puncturing, the bit reversal function and the number of puncturings to the bit index within the code group are offset. Select P ₂ indexes to be punctured in the code group.

In addition, the test in each code group is not performed on the systematic bit positions. In addition, when determining the location of puncturing in each code group, it is possible to maximize the location of the puncturing in the code group through bit reversal on the code bit indices in the code group.

Although the first puncturing algorithm performs a puncturing process in units of code groups, it is insufficient to satisfy the conditions of (1) to (3) described above, and therefore, another second puncturing algorithm is proposed. To do this, first, the entire code bit string is defined by grouping the systematic bit string, the parity bit strings of the upper component encoder (RSC), and the bit strings of the lower component encoder (RSC).

For example, the output sequence of turbo code at 1/5 rate is "x ₀ , y ₀ , z ₀ , y ' ₀ , z' ₀ , x ₁ , y ₁ , z ₁ , y ' ₁ , z' ₁ Let's say you can write in "..." bit form.

The output strings can then be thought of as three groups. That is, the systematic bit group {x ₀ , x ₁ , x ₂ , x ₃ , _.... } and the group of output strings of the first encoder {y ₀ , z ₀ , y ₁ , z ₁ , y ₂ , z ₂ , y ₃ , z _{3 ....} } and a group of output strings of the second encoder {y ' ₀ , z' ₀ , y ' ₁ , z' ₁ , y ' ₂ , z' ₂ , y ' ₃ , z' _{3 ....} }

Of the three types of code groups, a puncturing is excluded from the systematic group, and a method of allocating the total amount of puncturing to the remaining two groups by half is used.

Through this method, condition (1) of exclusion of puncturing for cymatic bits is satisfied, and condition (2), which is a balanced puncturing condition for two decoders, is satisfied. The final condition (3) was a uniform puncturing condition for each decoder. Therefore, we apply a uniform puncturing algorithm for each group to satisfy condition (3).

In the following equation, I is the length of the information bit string including the tail and the reserved bits.

P is the required puncturing amount as L-N.

Equation (7) is the amount of puncturing for the component decoder at the top.

Equation 8 is the amount of puncturing for the lower decoder.

The second algorithm using Equations 7 and 8 is as follows.

for (i = 0; i <L; i ++) {

j = i-3

if (0 <(i mod 5) <3) & ((j * P _D1 ) mod 2I <P _D1 ) puncture i ^th bit

else if (2 <(i mod 5) <5) & ((j * P _D2 ) mod 2I <P _D2 ) puncture i ^th bit

)

That is, the second algorithm may determine whether the bit index belongs to the group of the upper component coder (RSC) or the group of the lower component coder (RSC) by examining each bit index i.

The index j indicates a sequential number in the upper component coder group or a sequential number in the lower component coder group.

For example, if I is 4, the output string of the 1/5 rate turbo code is 20 in length.

The indices for the code bits are then 0, 1, 2, 3, ..., 17, 18, 19.

In this case, the relationship between the code index i and the sequential index j in the group for the upper component encoder and the sequential index j in the group for the lower component encoder is defined as follows.

The index mapping for the top constituent encoder is "1,2,6,7,11,12,16,17 (i) -------> 1,2,3,4,5,6,8 (j ), And the index mapping for the lower constituent encoder is "3,4,8,9,13,14,18,19 (i) ------> 3,4,5,6,7," 8,9,10 (j) ".

In this case, the index j is "i-3 Is obtained.

Using the mapping relationship above, performing a group check on the upper component coder (RSC) and a uniform puncturing on the j index results in a puncturing test on the resulting bit index i.

In this case, when performing the uniform puncturing on the index j, it is performed through a modulo operation with a 2I value that is twice the length of all information bits.

The second algorithm may be modified to the following third algorithm, which is also applicable to a general 1 / n turbo code.

In the third algorithm, like the second algorithm, the puncturing equations for the upper and lower component decoders of Equations 7 and 8 are used.

In addition, the inverse n _c of the coding rate of each component encoder is " ", And using the inverse n _c of this coding rate, the index j is defined as in the following equation (9).

Then, two events A and B are defined as follows and puncturing is performed at the corresponding bit positions where these two events occur independently.

Event "A": (0 <(i mod n) <n _c ) & ((j * P _D1 ) mod ((n _c -1) * I) <P _D1 )

Event "B": ((n _c -1) <(i mod n) <n) & ((j * P _D2 ) mod ((n _c -1) * I) <P _D2 )

That is, when increasing from 0 to L, as an algorithm puncturing the i-th bit when the event "A" or "B" occurs, it is available for turbo codes of all coding rates.

The third algorithm is also applicable to the design of a repetition algorithm for a 1/5 rate code.

At this time, giving a certain priority to the iteration for the systematic bit component plays an important role in improving the overall decoding performance. Therefore, we first define two repetition factors M ₁ and M ₂ . Further, we define the number of which have a repetition factor of M ₂ from below, and that the K _1.

In the following equation, I is the length of the information bit string including the tail and the reserved bits.

K ₁ = N mod L

In Equation 12, K ₁ is allocated to the systematic bit group, the upper component encoder bit group, and the bit groups of the lower component encoder in the following equation (13).

K _{1, x} = K ₁ K ₁ ≤I

I otherwise

Equation 13 is the number of bits having the repetition factor M ₂ for the systematic bits.

K _{1, D1} = 0 K ₁ ≤I

otherwise

Equation 14 is the number of bits having the repetition factor M ₂ for the upper component encoder bits.

K _{1, D2} = 0 K ₁ ≤I

otherwise

Equation 15 is the number of bits having the repetition factor M ₂ for the lower component encoder bits.

Through Equation 13, Equation 14, and Equation 15, it can be confirmed that an additional amount of up to 100 / n is allocated to the systematic bits with respect to the entire code bit string.

By using Equation 13 to Equation 15, each system code group, a code group of the upper component coder (RSC), and a code group of the lower component coder (RSC) in a manner similar to the following puncturing algorithm We propose the following fourth algorithm for matching uniformity.

for (i = 0; i <L; i ++) {

j = i-3

if ((i mod 5) = 0) & ((i * K _{1, x} / 5) mod I <K _{1, x} )) ∥

((0 <(i mod 5) <3) & ((j * K _{1, D1} ) mod 2I <K _{1, D1} )) ∥

((2 <(i mod 5) <5) & ((j * K _{1, D2} ) mod 2I <K _{1, D2} ))

repeat i ^th bit with repetition factor M ₂

otherwise repeat i ^th bit with repetition factor M ₁

}

The above algorithm constitutes the following fourth algorithm in a general form for a general 1 / n rate turbo code.

In other words, the inverse n _c of the coding rate of the component encoder is " And using different repetition factors M ₁ and M ₂ of Equations 10 and 11, and the number of bits having the repetition factor M ₂ of Equation 12 is used in the following algorithm.

In addition, the number of bits having the iteration factor M ₂ in the systemic bit group of Equation 13, the upper component encoder bit group of Equation 14, and the lower component encoder bit group of Equation 15 is used in the following algorithm. .

In Equation 16, index j indicating a sequential number in the upper component encoder group or the lower component encoder group is defined as follows.

Event "A": ((i mod n = 0) & ((i * K _{1, x} / n) mod I <K _{1, x} ))

Event "B": ((0 <(i mod n) <n _c ) & ((j * K _{1, D1} ) mod (n _c -1) * I <K _{1, D1} ))

Event "C": ((n _c -1 <(i mod n) <n) & ((j * K _{1, D2} ) mod (n _c -1) * I <K _{1, D2} ))

In the algorithm, when an event A, B, or C occurs for an index i that increases from 0 to the length L of the coding coded sequence, the i th bit having the repetition factor M ₂ is repeated.

Otherwise, the i th bit is repeated by the repetition factor M ₁ .

If a 1/4 rate convolution code is used, the puncturing algorithm is performed in the following simple process.

for (i = 0; i <L; i ++) {

if (((i + 1) * P) mod L <P) puncture i ^th bit

}

L represents the length of the output string of the channel encoder, P (= L-N) represents the amount to be punctured in the encoder output string. I is a value representing an index of a bit and has a value between 0 and L-1.

As described above, the present invention provides a practical code in addition to the time diversity gain due to symbol repetition when a variable data rate or a multivariate data rate mode for an auxiliary channel or both are supported in the current next generation communication system (3GPP2). Coding gain can be obtained due to the reduction of the rate, thereby lowering the required transmission power.

In addition, the reduction in the required transmission power due to the increase in the coding gain makes it possible to more efficiently manage the loading of the system.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit of the present invention.

Therefore, the technical scope of the present invention should not be limited to the contents described in the examples, but should be defined by the claims.

Claims (17)

In the process of interleaving the channel coded data to be mapped to the physical layer, When the channel coded data is transmitted at a data rate other than the maximum data rate in the multivariate data rate mode or at a random data rate not defined on each radio structure in the variable data rate mode, the data is transmitted at the lowest available data rate channel code. To match the channel coded data of different bit rates to the interleaving size by selectively using puncturing or symbol repetition. 2. The method of claim 1, wherein a coding rate of 1/4 rate is used when the channel coded data is a convolutional code. The data rate matching method of claim 1, wherein a coding rate of one fifth is used when the channel coded data is a turbo code. The output bit stream of the entire channel encoder according to claim 1, wherein when the length of the output string of the channel-coded data is larger than the size of the desired interleaver, the puncturing amount obtained by subtracting the interleaver size from the length of the output string of the channel-coded data is included in the output bit string of the entire channel encoder. Puncturing, characterized in that the data rate matching method. The data rate matching method of claim 4, wherein the puncturing is performed by dividing the channel-coded output string into units of code groups for puncturing the channel-coded output string. 6. The method of claim 5, wherein a code group that must be punctured by a maximum integer not exceeding a value obtained by dividing the total puncturing amount by the total number of code groups is a first type code group, and the total puncturing amount is the total number. And dividing the code groups to be punctured by a minimum integer greater than or equal to a value divided by the number of code groups into second type code groups so that each is evenly arranged. 6. The method of claim 5, wherein the amount of puncturing allocated to each code group is determined by a bit inversion operation on the index of the remaining bits except the systematic bits and adding the total puncturing amount up to the group. And determining the location of puncturing within the code group. 5. The method of claim 4, wherein the bit stream of the entire channel output sequence is divided into a systematic bit group, a bit group for parity bits of an upper component encoder, and parity bits of a lower component encoder, regardless of a coding rate of the turbo code. And performing puncturing by dividing into bit groups. 9. The method of claim 8, excluding puncturing in a bit group for the systematic code bits, and for a bit group for parity bits of an upper component encoder and a bit group for parity bits of a lower component encoder. A data rate matching method comprising dividing half of the total puncturing amount. The method according to claim 8, wherein in order to determine which bit group each code bit belongs to from a sequential index i of code bits of a channel coding sequence for puncturing, the sequential index j in each bit group is a coding rate. Inverse is n and inverse n _c of the coding rate of each component encoder is " If "," in _c Data rate matching method. 11. The method of claim 10, wherein when the sequential bit index (i) of the channel coding sequence belongs to an upper component encoder bit group, the sequential bit index is mapped to a sequential bit index (j) in the upper component encoder bit group. And if the sequential bit index (i) of the channel coded sequence belongs to a lower component encoder bit group, the sequential bit index is mapped to a sequential bit index (j) in the lower component encoder bit group. 11. The method of claim 10, wherein uniform puncturing is performed on parity bits of each component encoder by checking the sequentially mapped bit indexes in each component encoder after bit indexes sequentially mapped in each component encoder. Performing a data rate matching method. 2. The data rate of claim 1, wherein when the length of the output string of the channel encoder is smaller than the length of the desired interleaver, the data rate is repeated at equal intervals by subtracting the length of the output string of the channel encoder from the length of the interleaver. Matching method. 14. The method of claim 13, wherein the iterative algorithm for the turbo code prioritizes the iterative process in a group for systematic code bits. The method of claim 13, wherein the iterative algorithm for the turbo code is configured to determine the entire code bit stream regardless of the coding rate of the turbo code, the group of the systematic bit group and the parity bits of the upper component encoder, and the parity of the lower component encoder. And performing data by dividing the data into groups for bits. 14. The method of claim 13, wherein the iterative algorithm compares the number of bits that should have one larger repetition factor with respect to the other positions and the number of systematic bits, so that the value is smaller than or equal to the number of systematic bits. Until all these values are assigned to the systematic bit group. The method according to claim 13, wherein the iterative algorithm compares the number of bits that should have one larger repetition factor with respect to the other positions and the number of systematic bits, and if the value is larger than the number of systematic bits, And allocating half of the larger amount for the upper component encoder group and the lower component encoder group.