KR20050013322A - STBC/TCM decoding method - Google Patents

Abstract

PURPOSE: A STBC/TCM decoding method is provided to reduce an unnecessary calculation process and the complexity of a TCM decoder by using a bias point after setting the bias point by using signals and channel information of a STBC decoder. CONSTITUTION: A bias point is set by an output signal of a STBC decoder(16) and channel information and an effectual decision symbol is determined thereby. The determined symbol information is transmitted to a TCM decoder in order to calculate a branch metric of a symbol determined by the TCM decoder. The process of setting the bias point includes a process for detecting signs of a real number part and an imaginary number part of an output signal to determine an effectual quadrant and a process for setting bias points of subsets within the determined quadrant.

Description

STBC / TCM decoding method {STBC / TCM decoding method}

The present invention relates to an STBC / TCM decoding method, and more particularly, to Trellis Coded Modulation (TCM) using channel information of Space Time Block Codes (STBC) in an M-ary Quadrature Amplitude Modulation (QAM) STBC / TCM combined system. A decoding method for reducing the amount of calculation of the decoder.

As shown in FIG. 1, the STBC / TCM combined system converts data bits into coded symbols through a TCM encoder 10, and transmits them with a time difference to several antennas through the STBC encoder 12. A system that gains space time code gains.

The TCM encoder 10 generates a modulated symbol (PSK, QAM) while applying a convolution code by inputting a bit.

In this case, the coding bit is used as a bit for creating a subset of the TCM, which increases the Euclidean distance of each symbol to obtain a code gain.

The STBC encoder 12 transmits a signal between two cycles using two or more transmit antennas using mutual orthogonality of a plurality of antennas, and receives and combines (STBC decoders) antenna diversity gains at a receiver. Way.

Reference numeral 14 is a fading AWGN channel.

In the STBC / TCM system, the TCM decoder 18 decodes the complex signal from the STBC decoder 16 because there are two trellis signal paths for each subset according to each symbol. When decoding by the decoding method, two parallel transitions appear when transitioning from state to state, and the received signal and the symbol with the smallest Euclidean distance during the parallel transition for each subset are shown. Will be decided.

In the decoding method of the TCM coding, a branch-metric for each QAM symbol must be calculated for each received symbol.

At this time, in order to calculate one Euclidean distance in TCM, eight multiplications and three addition operations are performed.

Therefore, the complexity increases as the constellation number of the TCM symbol increases.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and the bias point is set using the signal and channel information from the STBC decoder, and the biased signal is calculated using the set bias point to be closest to the discrimination force subset. The purpose of the present invention is to provide a STBC / TCM decoding method that can reduce the complexity of the TCM decoder without degrading performance by finding signal points in advance and reducing unnecessary operations.

In order to achieve the above object, the present invention provides a method for decoding a signal received in an STBC / TCM combined system using M-ary QAM;

A first step of setting a bias point by using the output signal a + jb and the channel information α of the STBC decoder to determine the discriminative force symbol;

It is intended to provide a STBC / TCM decoding method comprising the step of passing the determined symbol information to the TCM decoder and selectively calculating branch values only for the symbols determined by the TCM decoder.

1 is a block diagram of a conventional STBC / TCM coupled system.

2 is a block diagram of an STBC / TCM decoding system according to the present invention.

3 is a detailed block diagram of the bias point generator shown in FIG. 2.

4 is a graph for determining the discriminating potential quadrant using the sign of the received signal.

5 illustrates an example of bias point setting in an STBC / TCM system using an 8-state 16QAM.

6 is a graph showing the positions of symbols in 8 state 16QAM.

FIG. 7 is a detailed functional diagram of the second bias point generator shown in FIG. 3.

<Description of the symbols for the main parts of the drawings>

10: TCM encoder 12: STBC encoder

14 fading AWGN channel 16 STBC decoder

18: TCM decoder 20: bias point generator

22: first bias point generator 24: second bias point generator

26: area discriminator

Hereinafter, the configuration and operation of the present invention will be described with reference to the accompanying drawings.

2 is a STBS / TCM decoding system configured to receive channel information from an STBC decoder in order to reduce the complexity of the TCM decoder in the conventional STBC / TCM system.

The STBC decoder 16 performs a combination process on a signal received for two periods at the receiver and outputs a decoding symbol through a maximum likelihood decoding (MLD) process.

At this time, since the decoded symbol is a complex signal including channel information as shown in Equation 1, the output signal and the channel information of the STBC decoder 16 instead of calculating the branch values for the paths of all symbols during decoding in the existing TCM decoder. Since the TCM decoder 18 selects only the symbols that are prominent in the TCM decoder 18 and performs the branch value operation, unnecessary operations can be reduced without degrading performance.

here , , Is the discrimination symbol, , Fading channel.

Accordingly, the biased STBC / TCM decoder according to the present invention uses the channel information α and the output signal a + jb characteristic of the STBC decoder 16 to determine the symbol region of the discriminating potential subchannel to obtain a bias point. And a block for transmitting the selected subchannel symbol information to the TCM decoder 18.

3 is a block diagram in which a bias point generator is applied to a TCM decoder in an STBC / TCM system.

The bias point generator 20 sets the center of the estimated area as a bias so as to discriminate in the subset of the symbols actually transmitted by using the channel information of the received symbol and sets the center of the estimation region as a bias so that the TCM decoder 18 can identify the most likely symbols. It is a device that can selectively calculate the branch value only.

Each bias point generator 22 or 24 receives a signal from a bias shifted received signal from the STBC decoder 16 or the preceding bias point generator, sets a new bias point to calculate branch values, and then calculates the branch value. Passed to the bias point generator to operate in multiple stages to estimate the path of the signal.

Accordingly, the bias point generator is composed of several bias point generators 22 and 24 according to the number of subsets of the TCM decoder 18.

If there are multiple signal points in the subset from the bias set on the basis of the bias at each stage, the branch value calculated by the actual TCM decoder 18 is found only the signal point closest to the subset, and the branch value is determined only for the discriminating force symbol. By performing the calculation, the calculation amount of the TCM decoder can be reduced.

FIG. 4 is a function for discriminating the discriminating potential quadrant using the sign of the received signal. The real part and the imaginary part (a, b) of the complex output symbol of the STBC decoder 16 of the area discriminator 26 of FIG. Detecting the sign shows determining the potent quadrant of the received symbol.

Through this process, the computation of the comparison part is limited from four quadrants to one quadrant, thus reducing the amount of computation.

In this process, the area setting of the quadrant according to the output symbol of the STBC decoder 16 is shown in Table 1.

Once the quadrant region is determined, the bias point is set within the determined quadrant.

At this time, the center point of each subset becomes a bias point.

For example, in the case of 16QAM TCM, the bias point setting for each quadrant is shown in Table 2.

The first bias point generator 22 applies a bias point set according to the quadrant region from Equation 2 from Equation 2 to generate the biased symbols a 'and b' from the output symbols a and b of the STBC decoding. The sign of the biased symbol is determined to set the nearest symbol of the reference QAM constellation of the TCM decoder 18, and the branch value with the symbol is calculated.

In this way, the branch value calculation of the symbol with respect to the discrimination force subset is performed.

5 is an exemplary diagram of bias point configuration in an STBC / TCM system using 8-state, 16QAM.

Assuming that the output signal of the STBC decoder 16 is identified as a quadrant symbol, set the branch value of the subset to (2,2) bias point, which is the center of the first quadrant, and apply a and b to equation (2). Move around to create the coordinates a ', b' and determine the sign again.

At this time, if the coordinate of X is the same as in FIG. 5, a '<0, b' <0, and the nearest symbol can be estimated as symbol 3.

In this way, the symbols of the first discriminating influence subset are discriminated.

Next, the determination on the symbols of the second discrimination influence subset will be described with reference to FIG. 6.

To this end, the up, down, left, and right symbols are selected based on the first discriminating force subset symbols determined above, and the branch value of the set subset symbols is calculated by the TCM decoder.

FIG. 6 is a graph showing the positions of symbols centering on a set symbol after setting a symbol closest to a received symbol in an 8-state 16QAM TCM.

If symbol 3 of the first quadrant is a symbol of the first discrimination potential subset, the symbols 4, 2, 0, and 6 on the top, bottom, left, and right sides of the symbol represent the subset 4, 2, 0, and 6 symbols. It is the closest signal point to the received signal at symbols and subsets 4, 2, 0 and 6.

Here, the Euclidean distance between the determined signal point and the received signal is obtained and used as the branch value in the TCM grid.

As a result, the branch values of the symbols for the subsets 3 and 5 subsets such as 4, 2, 0, and 6 may be calculated.

Next, the determination on the symbols of the third discrimination influence subset will be described with reference to FIG. 7.

Fig. 7 is a detailed functional diagram of the second bias point generator that functions to determine the symbols within the remaining subsets 1, 5, and 7.

In this process, since a symbol of the first discrimination force subset is already determined, the bias is set based on the symbol.

That is, as in the above-described example, if symbol 3 is the first discriminating potential symbol, symbol 3 is a bias point.

When the bias point is determined, the biased symbols (a "and b") are calculated by applying the bias to Equation 3 to determine the symbol region.

For symbol discrimination located at subsets 1, 5, and 7, the symbols of the second discrimination potential subset are determined using the equations of two straight lines (b "= a", b "=-a") with a slope of 45 degrees. .

If a "> b", -a "> b", symbols 1 and 1 in quadrant 4 and 5 and 7 in quadrant 4 are the closest symbols to the signals received in subsets 1, 5 and 7, respectively. Pass to TCM decoder to calculate branch value.

Through this process, it is possible to perform selective branch value calculation of symbols in all subsets of STBC / TCM of 8-state 16QAM such as subset 3, 4, 2, 0, 6, 1, 5, 7.

In addition, the repetitive operation of the first and second bias point generators 22 and 24 may be applied to an M-ary QAM TCM or STBC / TCM decoder having more constellations such as 64QAM and 256QAM.

According to the TCM decoding method according to the present invention, the TCM decoder 18 by selecting the area of the received symbol and applying the bias point, and selects and calculates the symbols of the influential subset by only calculating the branch value of the eight symbols when there are eight subsets Can be decoded with reduced complexity without compromising performance.

Specifically, compared with the conventional TCM decoding method, the multiplication, 37%, 41%, and 25% of the computation amount can be reduced.

If the 8-state, 16QAM TCM decoding process combined with STBC is performed by the conventional method, after calculating 16 Euclidean distances for each of 16 symbols, only the eight nearest signal points in the subset are branch metric. To be used.

However, in the TCM decoding method according to the present invention, since the signal point closest to the received signal and the subset is determined in advance by the bias point setting, only the eight Euclidean distances for the eight symbols except the insignificant symbols are calculated.

In addition, even if the number of QAM constellations, such as 32QAM and 64QAM, increases, the TCM decoder using the same subset number decodes only with operations of almost the same complexity without performance degradation.

If the constellation is divided into more levels and the subset is increased, the amount of computation increases because there are more kinds of branch values to calculate.

Therefore, in the case of TCM decoding having the same eight subsets, according to the present invention, six times a, b, a ', b', a ", b" code comparison, four times multiplication and four times necessary for setting a bias point are required. Instead of adding a subtraction operation, the Euclidean distance operation, which requires eight multiplications and three additions, is not performed by eight times the number of subsets, which reduces complexity.

Claims (9)

CLAIMS What is claimed is: 1. A method for decoding a received signal in an STBC / TCM combined system using M-ary QAM; A first step of setting a bias point by using the output signal a + jb and the channel information α of the STBC decoder to determine the discriminative force symbol; And transmitting the determined symbol information to a TCM decoder to selectively calculate branch values only for the symbols determined by the TCM decoder. The method of claim 1, wherein the bias point setting, First-a step of determining the discriminating potential quadrant by detecting the codes of the real part (a) and the imaginary part (b) of the output signal; STBC / TCM decoding method, characterized in that performed in step 1-b to set a bias point for each subset in the determined quadrant. The STBC / TCM decoding method according to claim 2, wherein the center point of each subset is a bias point in the first-b step. The method according to claim 2 or 3, in the case of the eight-state (16-state) 16QAM in step 1-b, If the determined quadrant is one quadrant, the bias point (bias_x, bias-y) is determined as (2,2), If it's two quadrants, we determine the bias point as (-2,2), If it is a quadrant, we determine the bias point as (-2, -2), If the quadrant is four, the bias point is determined as (2, -2). The method according to claim 1 or 2, wherein the determination of the discriminating power symbol, Steps 1-c of making a biased signal a 'and b' from the output signals a and b using the set bias points, and determining the sign of the discriminating symbol by re-determining the sign of the biased signal; And a first-d step of determining a symbol of a discriminating influential subset consisting of first, second, and third discriminating insignificant subset symbols centering on the determined symbols. 6. The STBC / TCM decoding method according to claim 5, wherein the biased symbols (a ', b') are made by the calculation of equation (4). The symbol of claim 5, wherein the symbol of the first discriminating influence subset is: STB / TCM decoding method comprising moving a and b to make a ', b' coordinate of biased signal, and determine the sign of a ', b' to determine nearest symbol. The symbol of claim 7, wherein the symbol of the second discrimination force subset is: STBC / TCM decoding method, characterized in that determined by the symbols in the top, bottom, left and right centered on the symbols of the first discriminating power subset. The symbol of claim 7 or 8, wherein the symbol of the third discriminating influence subset is: Based on the symbols of the first discrimination force subset, a biased signal (a ", b") is generated from the output signals (a, b) using bias points, and the diagonal equation (b "= a ", b" =-a ") to determine the closest symbol to the received signal.