KR20020000614A - Flexible Rate Puncturing Method - Google Patents

Abstract

PURPOSE: An adaptive rate puncturing method is provided to offer an adaptive rate puncturing technique of a transmitting unit which can execute uniform puncturing to improve the decoding performance of a receiving unit. CONSTITUTION: If a data stream having L bits comes down from an upper layer, a data stream of more than L bits is formed as a CRC(Cyclic Redundancy Check) code and a tail are added to the transmitted data stream(S1). For example, if the length of the data stream from the upper layer is 150, the code bit stream becomes 170 bits after a CRC code of 12 bits and a zero tail of 8 bits are attached. At this moment, the output stream of a 1/4 rate encoder becomes 680 bits for the 170-bit input stream. Because the size of a block interleaver is 768 and the output stream of the 1/4 rate encoder becomes 680 bits, symbol repetition by a symbol repeater becomes 2. Accordingly, the length of the output stream after the symbol repeater becomes 1360(680*2) bits(S2). As the length of this output stream has a value larger than the size of the block interleaver, a puncturing of 592(=1360-768) bits is required(S3).

Description

Flexible Rate Puncturing Method

The present invention relates to a next generation mobile communication, and more particularly, to an adaptive rate puncturing method for a next generation mobile communication for supporting adaptive rate puncturing on a physical layer of cdma2000, which is a synchronous scheme.

In the next generation mobile communication system, which is being standardized for global roaming, cdma2000, which is a synchronous method mainly in the United States, is based on the code division multiple access (CDMA) method.

In cdma2000, symbol repetition and puncturing are applied to different transmission channels to adjust the channel symbol rate at an optimal level.

In the puncturing scheme, there is an adaptive rate puncturing scheme.

The adaptive data rate method is a transmission method that can support arbitrary data rates in addition to the standard data rates supported on each radio structure of cdma2000, and is one of 3GPP speech codecs on the physical layer. -Rate) was introduced to support the codec.

That is, in the case of the adaptive multi-rate (AMR), data bits that do not match the standard data rate supported by each radio structure of the cdma2000 can be descended during the frame period for 20 ms.

In this case, as an example, there is a method of adjusting the data size of a rate supported by the RC through zero padding on the physical layer.

As another example, a symbol repetition for optimizing the data rate between the channel encoder and the block interleaver, and an adaptive data rate method in the process of puncturing are given.

In general, if the length of the output string of the channel encoder is L and the data size of the block interleaver used in the current radio structure is N, the number of symbol repetitions M is set by the following equation.

here Is the smallest integer greater than x.

Then, the length of the output string after passing through the symbol repeater is L * M.

If the length of the output string of the symbol repeater is larger than the size N of the block interleaver to be transmitted to the receiver, puncturing is required to match the length of the output string to the desired length of the interleaver.

Therefore, the puncturing amount P required for one frame is obtained as in Equation 2 below.

P = LM-N

In order to perform such puncturing, the puncturing distance D is first defined as follows.

From here Is the maximum integer not exceeding x.

That is, puncturing of P bits is performed at intervals of D for all (L * M) code bit strings.

In this case, the puncturing occurs at D, 2D, 3D ... when the bit index of the code bit string has a value from 1 to (L * M).

On the other hand, when the original puncturing distance is represented by D _org , D _org is expressed by the following equation.

At this time, the above D _org has a problem that generally do not have an integer value. Therefore, the method used to make it an integer performs a rounding operation on the actual puncturing distance D _org as in Equation 3.

The reason why it is not possible to calculate the puncturing distance through the rounding operation on the D _org is that if the puncturing distance is increased in this way, all the number P of puncturing in the (L * M) code bit strings is not performed. Because you can't.

In the above manner, when the puncturing distance is reduced through the lowering operation on the actual puncturing distance, and the puncturing is performed through this, as shown in FIG. 1, puncturing is intensively performed on only some code bit strings.

This causes the energy distribution to be disproportionately transmitted so that the magnitude of each signal is unbalanced in trellis decoding at the receiving end.

Therefore, a problem occurs when the decoder determines the signal, which ultimately becomes an important cause of deterioration of the performance of the decoder.

Accordingly, an object of the present invention is to provide an adaptive rate puncturing technique of a transmitting end capable of uniform puncturing that improves the decoding performance of a receiving end in view of the above-mentioned problems of the prior art.

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, it is repeatedly output so that the data of the channel-coded other bit rate is more than a certain symbol rate And puncturing by the first puncturing distance that rounds down the actual puncturing distance and the second puncturing distance that rounds up the actual puncturing distance is uniformly arranged for the entire bit string. Puncturing by a puncturing algorithm.

Preferably, the puncturing amount n according to the second puncturing distance has an encoding length of "L", an interleaving size of "N", and the puncturing amount "P" from this L and N is "LM-N." Calculated as "and the symbol repetition number M is" "When calculated as, the puncturing occurrence probability by the second puncturing distance" It is characterized by being P times of ".

If the difference between the puncturing amount and the puncturing amount by the second puncturing distance is positive so that puncturing by each puncturing distance is uniformly performed for the entire bit sequence of the repeated symbols, the first puncturing distance is positive. If negative, puncturing is performed by the second puncturing distance.

1 is a view showing an example of a problem of the puncturing algorithm according to the prior art.

2 illustrates an adaptive puncturing algorithm in accordance with the present invention.

The present invention proposes a uniform puncturing technique of a transmitter to improve decoding performance at a receiver.

In particular, in order to propose this uniform puncturing technique, two puncturing distances are defined, and a method for obtaining uniform puncturing using the two puncturing distances is presented.

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

FIG. 2 is a diagram illustrating an adaptive data rate encoder and an adaptive data rate algorithm in the process of puncturing symbol repetition for optimizing a data rate between a block interleaver.

Referring to FIG. 2, when a data string having L bits is transmitted from the transport channel, the data string has a cyclic redundancy code (CRC) and a tail added thereto to form a bit string of L bits or more (S1).

For example, in the third radio structure of the current forward link, when the maximum data rate is 9.6 kbps, data is only 172 bits for 20 ms. In addition, when a 12-bit cyclic redundancy code (CRC) and an 8-bit tail bit are added and a convolutional code of 9 restraints is used, a data string having a data size of 170 is 192 bits by channel coding. Is formed.

Since coding at 1 / n rate is used, the length of the code bit string after the coding becomes 768 (when n is 4) bits.

That is, in the above example, the bit string having 170 bits ultimately outputs 768 bits by channel coding. If the 768-bit data string passes block interleaving with a block size N of 768, no puncturing is required.

However, if the length of the data stream descending from the upper is 150, the code bit string after adding 12-bit cyclic redundancy code (CRC) and 8-bit zero-tail becomes 170 bits.

At this time, the output string of the 1 / 4-rate encoder is 680 bits with respect to the 170-bit input string.

Since the size of the block interleaver N is 768 and the output string of the 1/4 rate encoder is 680 bits, the symbol repetition by the symbol repeater is 2 (= ), The length of the output string after passing through the symbol repeater is 1360 (680 * 2) bits. (S2)

Since the length of the output string has a larger value than that of the block interleaver, puncturing of 592 (= 1360-768) bits is necessary. (S3)

In the present invention, this puncturing process is described in terms of uniform puncturing.

First, consider the concept of uniform puncturing from the actual puncturing distance.

If D _org is calculated to be 2.333 ... Then, uniform puncturing means that two-thirds of the total puncturing amount P should be performed with the puncturing distance 2, and the other 1/3 puncturing should be performed with the distance 3.

In other words, 0.3333 ..., which is the fractional part of D _org , can be thought of as a probability value that the puncturing distance will be three.

Therefore, in order to introduce the above concept, two puncturing distances D ₁ and D ₂ may be defined as Equations 5 and 6 below.

At this time, the probability A to have the puncturing distance D ₂ is calculated by the following equation.

That is, if D ₁ is subtracted from the actual puncturing distance D _org , only a portion below the decimal point is obtained, and thus a probability value having a distance D ₂ can be calculated through this value.

Next, the number of times the puncturing distance D ₂ occurs from the probability value can be obtained. This number can be obtained by multiplying the amount P of the overall puncturing by the probability value calculated in Eq.

After obtaining the number n of puncturings having a distance of D ₂ as in Equation 8, the next problem to be solved is how to determine the position of such puncturing.

In fact, to get the best decoding performance, the position of the total P punctures is to place those with a puncturing distance of D ₂ as evenly as possible on the bit.

The following algorithm can be constructed for the purpose of uniform puncturing.

e _ini = P; P is the puncturing amount

e _minus = LM-P ; The number of punctures caused by the distance D ₂ .

e _plus = P; P is the puncturing amount

i = 1; Use this algorithm if more than one puncture occurs.

e = e _ini ; e has an initialization value P.

While (i≤P) do {

e = e-e _minus

if (e> 0)

puncture bit with distance D ₁ ; Puncturing by distance D ₁

else {

puncture bit with distance D ₂ ; Puncturing by distance D ₂

e = e + e _plus

}

i ++

}

In the above algorithm, if the punctured by the respective puncture distances positive difference between the puncturing amount of the whole puncturing amount of the puncture distance D ₂ such that uniformly performed for the entire bit stream of the symbol repetition, the puncture distance D ₁ In the case of a negative number, puncturing is performed by the puncturing distance D ₂ .

As can be seen from the above algorithm, LM-P having a puncturing distance D _{2 out} of the total P punctures It can be arrange | positioned uniformly.

The puncturing algorithm has a shape similar to the data matching algorithm defined on 3GPP.

However, if the above algorithm is used, it is not necessary to repeat the loop for the entire (L * M) code bit strings, but it can repeat the loop as much as P, which is the amount that puncturing actually occurs. It becomes possible.

As described above, the present invention solves the problem of the puncturing method for supporting the current adaptive data rate of 3GPP2. That is, there is an effect of improving the decoding function of the receiver by solving the imbalance of energy distribution and obtaining a uniform puncturing pattern as much as possible.

Compared to the existing 3GPP rate matching pattern determination algorithm, a uniform puncturing pattern is obtained through fewer loop iterations, thereby reducing the delay 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 (3)

In the process of interleaving the channel coded data to be mapped to the physical layer, Iteratively outputting the data of the channel coded different bit rates to have the same symbol rate; The puncturing algorithm applies the repeated symbols to the puncturing algorithm such that puncturing by the first puncturing distance that rounds down the actual puncturing distance and the second puncturing distance that rounds up the actual puncturing distance is uniformly arranged for the entire bit string. Adaptive puncturing method comprising the step of puncturing by. 2. The puncturing amount n according to the second puncturing distance has a coding length of "L", an interleaving size of "N", and the puncturing amount "P" from "L" and "N". LM-N "and the symbol repetition number M is" "When calculated as, the puncturing occurrence probability by the second puncturing distance" Adaptive rate puncturing method, characterized in that P times. The first and second puncturing amounts of the first and second puncturing distances are equal to each other so that the puncturing by the puncturing distances is uniform with respect to the entire symbol sequence. The puncturing method is characterized in that puncturing is performed by the second puncturing distance in the case of a negative number by the puncturing distance.