KR100451716B1 - Rate matching method of mobile communication system - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a rate matching method of a mobile communication system for optimizing puncturing or repetition patterns in rate matching in mobile communication, particularly based on wideband code division multiple access (W-CDMA). In the algorithm for rate matching, an initial value for determining a bit to be punctured or repeated is determined so that the puncturing or repeating pattern is optimal.

Description

Rate matching method of mobile communication system

The present invention relates to mobile communication, and more particularly, to a rate matching method of a mobile communication system for optimizing puncturing or repetition patterns in rate matching in a mobile communication system based on a wideband code division multiple access (W-CDMA) scheme. will be.

In general, in a mobile communication system using a wideband code division multiple access scheme, a rate matching algorithm punctures an output bit string to adjust a code rate of a bit string output from a channel encoder to an arbitrary value. Puncturing or repetition algorithm.

These rate matching algorithms are currently being jointly researched in the 3rd Generation Common Project (3GPP), which sets the technical standard for next generation mobile communication systems. In the 3rd generation joint project, transport multiplexing of the next generation mobile communication system is applied. To this end, puncturing and iterative algorithms for convolutional codes are specified and proposed in the specification.

The following describes the rate matching algorithm of the convolution code proposed by the third generation joint project in detail.

According to the rate matching algorithm proposed up to now, puncturing or repetition of the bit stream output by channel coding from the channel encoder is performed at uniform equal intervals. This means puncturing or repetition so that uniformity is established for the entire bit string.

Therefore, in the conventional rate matching algorithm, the puncturing or repetition distance n is obtained in units of bits with respect to the entire bit stream channel-coded and output, and the puncturing or repetition is performed for each obtained n-bit distance. For example, if puncturing distance n is '5', puncturing is performed once for every bit distance of '5' for the entire output bit stream. If puncturing distance n is '5.5', '5' is for all bit strings. The puncturing with the bit distance of is performed by half and the puncturing is performed with the other half having a bit distance of '6'.

Therefore, the rate matching algorithm of the conventional convolutional code has the following flow.

1. Specify the initial value (e).

2. Subtract a certain value from the initial value.

3. If the repeatedly subtracted value meets a certain condition (ie, 0 or less), puncture or repeat.

The rate matching algorithm with this flow is expressed as follows.

N _c : Block size of the entire bit stream to be rate matched

N _i : Block size of the entire bit string after rate matching

S: parameter related to the first interleaving of the uplink

S ₀ = (d ₁ , d ₂ , ..., d _Nc ): N _c data block set

y = N _c -N _i

e = (2 * S (k) * y + N _c ) mod 2N _c : difference between the current puncturing code rate and the desired puncturing code rate, S (k) is the shifting parameter

m = 1: index of the current bit

do while m ≤ N _c

e = e-2 * y: update error

if e ≤ 0 then: Check whether the bit at index m is punctured

puncture bit m from set S ₀

e = e + 2 * N _c : update error

end if

m = m + 1: index of the next bit

end do

else

y = N _i -N _c

e = (2 * S (k) * y + N _c ) mod 2N _c : The difference between the current puncturing code rate and the desired puncturing code rate, S (k) is the shifting parameter

m = 1: current bit index

do while m ≤ N _c

e = e-2 * y: update error

do while m ≤ N _c : Check if the bit at index m repeats

repeat bit m from set S ₀

e = e + 2 * N _c: update error

end do

m = m + 1: index of the next bit

end do

end if

However, the rate matching algorithm for the conventional channel code can specify an initial value as an arbitrary value when the puncturing or repetition algorithm is executed, but is currently set to the block size before the rate matching, that is, N _c . .

Considering that the proposed rate matching algorithm is composed of units of bits, only a certain bit position may be punctured or repeated among the entire bit strings, and when only a specific bit position is punctured or repeated, a code There is a problem that performance degradation may occur.

For example, assuming a case of rate matching a channel coded bit string by a convolutional code having a code rate of 1/3, a specific pattern in which only bits of the bit string generated according to the third generation polynomial are punctured or repeated. In this case, code performance is relatively degraded when puncturing or repeating the first bit string and the second bit string.

Accordingly, an object of the present invention has been made in view of the above-mentioned problems of the prior art, and a mobile communication system capable of puncturing or repetition according to a desired pattern by designating an initial value in rate matching with respect to a channel code. To provide a rate matching method.

According to an aspect of the present invention for achieving the above object, the step of inputting a channel coded bit string according to a predetermined coding rate, the value of the size of the input bit string divided by a predetermined puncturing or repeated number of bits The puncturing algorithm within the range according to the bit position in the symbol unit so that any one of periodic puncturing and repetition occurs at a specific bit position in the symbol unit of the input bit stream when the integer multiple of the inverse of the coding rate. And determining an initial value of any one of the iteration algorithm and performing one of the puncturing algorithm and the iteration algorithm using the initial value of the determined variable.

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

The present invention changes the initial value of the rate matching algorithm to exclude specific puncturing or repetitive patterns that result in degradation of performance when rate matching the channel code, and to move to generate specific puncturing or repetitive patterns resulting in improved performance. A rate matching method of a communication system is proposed.

To this end, the present invention adjusts an initial value that is not determined in a conventional rate matching algorithm so that a puncturing or repetitive pattern of a specific bit position that causes degradation of code performance when puncturing or repetition for a channel code does not occur. Here, the channel code means a convolutional code.

Before describing the rate matching method according to the present invention, it is assumed that a convolutional code having a code rate of 1 / n is as follows.

N _c : Block size of the entire bit stream to be rate matched

N _i : Block size of the entire bit string after rate matching

P: puncturing rate

N _p : The number of bits punctured or repeated (ie P * N _c )

I _E : Initial value

Then, according to the present invention, the condition of the initial value for the puncturing or repetition according to a specific pattern to be periodically repeated is given by Equation 1 below.

If (l is any integer greater than 0)

Referring to equation 1, Is when the first bit is punctured or repeated, Is when the second bit is punctured or repeated, Is an initial value condition when the nth bit is punctured or repeated.

Here, m means '0' or any integer greater than '0', [a] means 'a-1' if a is an integer, and less than 'a' if a is not an integer and a Means the nearest integer to '. Also 'nl' means Is a multiple of n.

Accordingly, in the present invention, Equation 1 may be generated by changing from a puncturing or repetitive position of a specific pattern that results in degradation of code performance when rate matching the channel code, to a puncturing or repetitive position that results in a relatively improved code performance. Change the initial value I _E.

Therefore, the initial value I _E can be determined as shown in Equation 2 below.

While satisfying the condition that is

_{2nmN p <I E ≤ 2 (} nm + 1) N p

2 (nm + 1) N _p <I _E ≤ 2 (nm + 2) N _p

........

2 (nm + n-1) N _p <I _E ≤ 2 (nm + n) N _p

Referring to Equation 2, 2 nm N _p <I _E ≤ 2 (nm + 1) N _p is the condition of the initial value that the first bit is punctured or repeated, 2 (nm + 1) N _p <I _E ≤ 2 ( nm + 2) N _p is the condition of the initial value where the second bit is punctured or repeated, ....., 2 (nm + n-1) N _p <I _E ≤ 2 (nm + n) N _p Is the condition of the initial value at which the nth bit is punctured or repeated.

Accordingly, when the initial value is arbitrarily adjusted as described in Equation 2 during rate matching of channel codes, puncturing or repetitive patterns having a desired pattern may be generated.

The rate matching method of the channel code according to the present invention will now be described in more detail by applying to a convolutional code having a code rate of 1/3.

In the downlink of a mobile communication system based on a wideband code division multiple access scheme, a convolutional code having a code rate of 1/3 is used as a channel code. At this time, one-third convolution code is generated through the generation polynomial of '557', '663', and '711' as an octal number.

Here, the code string generated through the generation polynomial of '557' is assumed to be 'first code string', and the code string generated through the generation polynomial of '663' is assumed to be 'second code string'. Assuming that a code string generated through the generation polynomial of 'is a third code string, each bit of the code strings generated by each generation polynomial forms one code symbol.

At this time, considering the rate matching algorithm of the conventional convolutional code, a specific pattern of puncturing or repeating only the bits of the 'third code string' configured in one code symbol may occur.

In this case, the initial value is adjusted according to the rate matching method of the convolutional code according to the present invention to exclude the occurrence of a specific pattern in which only the bits of the 'third code string' are punctured or repeated.

In more detail, (L is any integer greater than 0), so n = 3 The condition of the initial value is defined as follows.

here Is the case that the first bit of a code symbol is punctured or repeated, Is the case that the second bit of one code symbol is punctured or repeated, Denotes a case where the third bit of one code symbol is punctured or repeated.

Then, as in the prior art, when the initial value is determined according to the block size N _c of the bit string to be rate matched, the puncturing rate P is determined as in Equation 3 below.

⇒

⇒

here, Considering that In the case of the rate matching of the convolution code, since only the third bits of one code symbol generate a puncturing or repeating pattern, the puncturing or repeating pattern of only the third bit is eliminated and the code performance is improved. In order to generate a puncturing or repetition pattern for relatively good bits, the initial value is adjusted as in Equation 4.

(m is 0 or any integer greater than 0),

Where _{E E} <2 N _p

Referring to Equation 4, when m is '0' for easy implementation, the initial value I _E is not punctured or repeated in the third bits when an arbitrary constant value smaller than 2 N _p is set. The other bits, ie puncturing or repetition of the first bits, are performed.

Here, if N _p = 1, assuming that the number of punctured or repeated bits is one, since I _E <2 * 1 = 2, an initial value is set as I _E = 1. Here, in the case of repetition, unlike the puncturing, since the initial value is not largely influenced, the initial value may be set based on the puncturing.

As described above, when the initial value is set to '1' in the rate matching of the downlink 1/3 convolution code in the mobile communication system based on the wideband code division multiple access method, a puncturing or repetitive pattern of a bit string having excellent code performance is generated. You can.

Meanwhile, the rate matching method of the channel code according to the present invention can be similarly applied to the uplink, and even when a convolutional code having a 1/3 code rate is applied to the uplink, the initial value is set to '1'. You can improve code performance.

As described above, the rate matching method according to the present invention adjusts an initial value used in a conventional puncturing or repetition algorithm, thereby excluding a puncturing or repetition pattern of a specific code pattern resulting in deterioration of code performance. There is an effect that can result in excellent puncturing or repeating patterns.

In addition, according to the rate matching method according to the present invention, it is possible to easily implement a system and an algorithm by setting an initial value to a constant.

Claims (22)

Inputting a channel coded bit stream according to a predetermined coding rate; Any one of periodic puncturing and repetition at a specific bit position in a symbol unit of the input bit string when the value obtained by dividing the size of the input bit string by a predetermined puncturing or repeating bit number is an integer multiple of the inverse of the coding rate. Determining an initial value of any one of a puncturing algorithm and an iteration algorithm within a range of bit positions in the symbol unit; And performing one of the puncturing algorithm and the repetition algorithm by using the determined initial value. The method of claim 1, wherein the channel coding is convolutional coding. The method of claim 1, wherein the initial value is "1". A channel coder for channel coding the input data; An interleaver for interleaving channel coded symbol unit data; Under a specific condition, an initial value is determined within a range according to the bit position of the symbol unit so that any one of periodic puncturing and repetition occurs at a specific bit position in the symbol unit of the bit string output through the interleaver. And a rate matcher for adjusting a code rate for the output bit stream of the interleaver using the determined initial value. A channel coder for channel coding the input data; For a bit string output through the channel coder, an initial value is determined within a range according to the bit position of the symbol unit so that any one of periodic puncturing and repetition occurs at a specific bit position of a symbol unit under a specific condition. And a rate matcher for adjusting a code rate for the output bitstream of the channel coder using the determined initial value. The rate matching of claim 1, wherein the specific bit position is one of positions except for a position where the impact on the performance of bits in the symbol unit is relatively large in the case of puncturing. Way. The rate of the mobile communication system according to claim 1, wherein the specific bit position is one of positions except for the position of a bit having a relatively small effect on the performance of bits in the symbol unit in the case of repetition. Matching method. 2. The method of claim 1, wherein the generation polynomials of 557, 664, and 711 are used when the predetermined coding rate is 1/3 and the channel coding is convolutional coding. delete Convolutional coding at 1 / n coding rate; K (1 ≦ k ≦ n) in a symbol of the coded output, under the condition that the number of bits N _C of the coded output divided by a predetermined puncturing or repetition bit number N _P is an integer multiple of the n. Setting an initial value for puncturing or repetition of the first bit within a range greater than 2 kmN _{P and} less than or equal to 2 (km + k) N _P when m is an integer greater than or equal to 0; Performing a rate matching algorithm on the coded output using the initial value. delete 11. The method of claim 10, wherein the k-th bit is a bit at a position excluding a bit having a relatively large impact on performance in the case of puncturing, and the impact on the performance in the symbol in the case of repetition is relatively high. Rate matching method of the mobile communication system, characterized in that the bit of the position excluding small bits. 11. The method of claim 10, wherein, for the convolutionally coded output at the 1/3 coding rate, the initial value is a constant within a range greater than 0 and less than 2N _P. The method of claim 13, wherein the constant is one. 14. The rate matching method of claim 13, wherein a generation polynomial of 557, 663, and 711 is used for convolutional coding of the 1/3 coding rate. Inputting a convolutionally coded bit string at a 1/3 coding rate; Setting the initial value to a constant within a range not exceeding twice the predetermined number of puncturing or repeated bits N _{P for the} rate matching algorithm; And performing the rate matching algorithm by using the set initial value. 17. The method of claim 16, wherein the initial value is for determining bits to be punctured or repeated in the rate matching algorithm. 17. The method of claim 16, wherein the constant is one. Inputting a convolutional coded bit string; Performing the repetition or puncturing algorithm by setting an initial value for determining a bit to be punctured or repeated in the puncturing or repetition algorithm for the input bit string to 1; Rate Matching Method. 20. The method of claim 19, wherein a coding rate of the convolutional coding is 1/3. 20. The method of claim 19, wherein a generation polynomial of 557, 663, and 711 is used for the convolutional coding. 17. The method of claim 16, wherein a generation polynomial of 557, 663, and 711 is used for the convolutional coding.