KR100326307B1 - Apparatus and method for symbol normalizing in cdma system - Google Patents

Abstract

end. FIELD OF THE INVENTION

The present invention relates to a receiver of a mobile communication system using a code division multiple access scheme, and more particularly, to an apparatus and method for normalizing a signal of a channel decoder input stage.

I. The technical problem to be solved by the invention

SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus and method for re-quantizing an input signal of a channel decoder in a code division multiple access mobile communication system and accurately mapping the normalized reference level to the size of a received demodulation symbol in the absence of noise. .

All. Summary of Solution of the Invention

The present invention for achieving the above object is the size of the received demodulation symbol when the normalization device of the channel decoder input stage inputs each symbol value received and demodulated during the corresponding period in the code division multiple access method mobile communication system And a demodulation symbol size calculating unit for calculating a signal, and dividing each symbol value during the corresponding interval by the magnitude of the reception demodulation symbol in the absence of the noise, and multiplying a normalized reference level value, and multiplying the multiplication result by the channel decoder input stage. It is characterized by consisting of a normalizer to map to the quantization level.

la. Important uses of the invention

It allows the channel decoder to have high channel decoding performance.

Description

Device and Method for Normalizing Channel Decoder Input in Code Division Multiple Access Mobile Communication System {APPARATUS AND METHOD FOR SYMBOL NORMALIZING IN CDMA SYSTEM}

The present invention relates to a receiver of a mobile communication system using a code division multiple access scheme, and more particularly, to an apparatus and method for normalizing a signal of a channel decoder input stage.

In a receiver of a mobile communication system of CODE DIVISION MULTIPLE ACCESS (hereinafter referred to as CDMA), a quantization method and a channel decoder for a value of a demodulation symbol output from a symbol demodulator The quantization method for the input signal value of (Channel Decoder) is different. In this case, the quantization method may be a number of bits required for quantization representing a signal size (hereinafter referred to as QB). The problem in that the number of representation bits required for signal processing in the channel decoder is limited is one of the reasons why the demodulation symbol value quantization representation and the quantization representation of the channel decoder input value are different from each other.

On the other hand, the quantized representation of the demodulated symbol values at the input of the channel decoder has a great influence on the channel decoding performance. This is because the decoding performance of the channel decoder varies depending on the quantization representation. In the case where the channel decoder is a turbo decoder, the decoding performance of the channel decoder is largely changed according to the representation method of the demodulation symbol value at the input of the channel decoder. Therefore, when the channel decoder is a turbo decoder, normalization of the turbo decoder input signal may be used to increase decoding performance.

1 illustrates an example of a receiver in a conventional CDMA mobile communication system.

The A / D converter 110 inputs an analog received signal and converts it into a digital signal. The received signal may be a signal converted to baseband by being down-converted through an antenna and an RF unit (not shown). The first symbol demodulator 121 to the Nth symbol demodulator 12N are channels such as PN despreading and Walsh orthogonal demodulation, respectively, for the outputs of the A / D converter 110 of N number of signal paths. The classification code is demodulated. Each symbol demodulator compensates and outputs a symbol accumulated value in a demodulation process of the channel classification code using a channel estimator (not shown). The combiner 130 combines the outputs of the respective symbol demodulators.

However, as shown, it can be seen that the QB of the demodulation symbol value 150 output from the combiner 130 is 10. The channel decoder 140 takes only 6 bits with respect to the demodulation symbol value 150 (QB = 10). That is, the channel decoder 140 inputs only some bits (QB = 6) of the demodulation symbol value 150 (QB = 10) except the upper bit or the lower bit to input a 6-bit signal. The channel decoder 140 performs a decoding operation on the input signal, that is, the channel decoder signal input value 160 (QB = 6), to decode the original data on the transmitter side.

2A to 2C are graphs illustrating an example of a conventional normalization method for an input signal of a channel decoder in a CDMA mobile communication system. Hereinafter, a description will be given with reference to FIG. 1.

In this case, it is assumed that no noise is added to the -8 and +8 levels, which are 1/4 of each of the 32 levels, among the 64 quantization levels (-32 to +32 levels) for the channel decoder input signal. Receive demodulation symbol size Wow It is assumed that the position of the channel decoder 140 is normalized. Where the quantization level Eight levels can be the desired normalization reference level. The horizontal axis of the graph represents quantization levels of the received signal, and the vertical axis of the graph represents the frequency of demodulation symbol values (PDF (Probability Density Function)).

2A illustrates a conventional normalization method for a received signal when the channel decoder 140 inputs the lower 6 bits of the symbol demodulation value (QB = 10) 150.

The first quantization period 21a (Quantization Range) when the channel decoder 140 inputs the lower 6 bits among the symbol demodulation value (QB = 10) 150 is the first QMIN 22a (Quantization Minimum) as shown. Level) and the first QMAX 23a (Quantization Maximum Level). The first quantization section 21a is divided into 64 levels at regular intervals. The received signal is mapped to a corresponding level between the first quantization sections 21a according to the strength of the signal. At this time, the received signal of the first QMIN 22a or less is mapped to the first QMIN 22a, and the received signal of the first QMAX 23a or more is mapped to the first QMAX 23a.

However, at this time, the reception demodulation symbol at the receiving end in the case where the position of the first -8th level 24a and the first + 8th level 25a of the first quantization section 21a is assumed that the noise is not added, respectively. size And It can be seen that there is a difference between and as much as NA (26a).

2B illustrates a conventional normalization method for a received signal when the channel decoder 140 inputs the first bit to the sixth bit from the symbol demodulation value (QB = 10) 150.

The second quantization period 21b (Quantization Range) when the channel decoder 140 inputs the first to sixth bits from the symbol demodulation value (QB = 10) 150 is represented by the second QMIN ( 22b) the interval between the (Quantization Minimum Level) and the second QMAX (23b) (Quantization Maximum Level). The second quantization section 21b is divided into 64 levels at regular intervals. The received signal is mapped to the corresponding level between the second quantization sections 21b according to the strength of the signal. At this time, the received signals below the second QMIN 22b are mapped to the second QMIN 22b, and the received signals above the second QMAX 23b are mapped to the second QMAX 23b.

As shown, reception demodulation at the receiving end in the case where the positions of the second -8th level 24b and the second + 8th level 25b of the second quantization section 21b are assumed that the noise is not added, respectively. The size of the symbol And It can be seen that there is a difference by as much as the interval NB 26b. However, at this time, it can be seen that the interval difference of the NB 26b is reduced compared to the NA 26a of FIG. 2A.

2C illustrates a conventional normalization method for a received signal when the channel decoder 140 inputs the second to seventh bits from the symbol demodulation value (QB = 10) 150.

The third quantization section 21c (Quantization Range) when the channel decoder 140 inputs the second to seventh bits from the symbol demodulation value (QB = 10) 150 is represented by the third QMIN ( 22c) the interval between the (Quantization Minimum Level) and the third QMAX (23c) (Quantization Maximum Level). The third quantization section 21c is divided into 64 levels at regular intervals. The received signal is mapped to a corresponding level between the third quantization sections 21c according to the strength of the signal. At this time, the received signal of the third QMIN 22c or less is mapped to the third QMIN 22c, and the received signal of the third QMAX 23c or more is mapped to the third QMAX 23c.

As shown, reception demodulation at the receiving end in the case where it is assumed that the positions of the third -8th level 24c and the third + 8th level 25c of the third quantization section 21c are not added to the noise, respectively. The size of the symbol And It can be seen that there is a difference between and as much as the interval NC 26c. At this time, it can be seen that the NC 26c is closer to the reception demodulation symbol size at the receiving end in the case where it is assumed that there is no target noise due to a decrease in the interval difference compared to the NB 26b of FIG. 2B.

2A to 2C, the conventional normalization expresses the received demodulation symbol value in order to represent the received demodulation symbol values (QB = 10) to the limited input bits (QB = 6) of the channel decoder 140. The bit QB = 10 selects the input bit QB = 6 at a fixed position and quantizes it back to 64 levels. In FIG. 2B, the requantization level interval is twice the requantization level interval of FIG. 2A. In FIG. 2C, the requantization level interval is twice the requantization level interval of FIG. 2B. In other words, it can be seen that in the conventional normalization method, the quantization level interval is fixed.

That is, as described above, in the CDMA mobile communication system, conventional normalization is performed by quantizing a received demodulation symbol value at a fixed interval to a desired normalization reference level (eg, 8 level), the demodulation symbol size of the receiver when the desired normalized reference level is noisy. There is a problem that cannot be mapped correctly.

If the demodulation symbol size Even if the position of is accurately mapped to one of several fixed normalization reference levels, the size of the demodulation symbol is affected by the mobile channel environment. By changing itself, the conventional fixed normalized reference level and the size of the demodulation symbol There is a problem that does not map correctly.

Accordingly, an object of the present invention is to provide an apparatus and method for re-quantizing an input signal of a channel decoder in a code division multiple access mobile communication system and accurately mapping the normalized reference level to a received demodulation symbol size in the absence of noise. Is in.

Another object of the present invention is to divide each received demodulation symbol value by the size of the demodulation symbol in the absence of noise and multiply it by a normalized reference level value in the code division multiple access method of the mobile communication system, thereby demodulating the demodulation symbol in the absence of the noise. An apparatus and method are provided for normalizing received demodulation symbol values such that magnitudes are mapped correctly to the normalization reference level.

It is still another object of the present invention to obtain a variance value, which is an average value of squares of each received and demodulated symbol value during a corresponding period in a code division multiple access method mobile communication system, to obtain the power of noise added during the corresponding interval, Subtracting the noise power added during the interval from the distribution of the received and demodulated symbol values during the interval, taking the square root of the output of the operator to obtain the magnitude of the received demodulation symbol in the absence of noise, The present invention provides an apparatus and method for dividing each symbol value during a period by the magnitude of a received demodulation symbol in the absence of noise, multiplying a normalized reference level value, and mapping the multiplication result to a quantization level of the channel decoder input terminal. .

In order to achieve the above object, the present invention provides a size of a reception demodulation symbol when a normalization device of a channel decoder input stage inputs each received symbol value and demodulated during a corresponding period in a code division multiple access method mobile communication system. And a demodulation symbol size calculating unit for calculating a symbol, and dividing each symbol value during the corresponding interval by the magnitude of the reception demodulation symbol in the absence of the noise and multiplying it by a normalized reference level value, and multiplying the multiplication result by the channel decoder input stage. It is characterized by consisting of a normalizer to map to the quantization level.

The present invention also provides a method for normalizing a channel decoder input stage in a code division multiple access mobile communication system, comprising: a first step of obtaining a size of a received demodulation symbol when there is no noise by using a received demodulation symbol input during a corresponding section; A second step of dividing each received demodulation symbol during the corresponding period by the magnitude of the received demodulation symbol in the absence of the noise and multiplying it again by a normalized reference level; and a third step of mapping the multiplication result to the quantization level of the channel decoder input stage. It is characterized by consisting of steps.

1 is an example illustrating a part of a receiving end in a conventional code division multiple access mobile communication system.

FIG. 2A is a graph showing a conventional normalization method for a received signal when the channel decoder inputs the lower 6 bits among symbol demodulation values. FIG.

FIG. 2B is a graph showing a conventional normalization method for a received signal when the channel decoder inputs the first to sixth bits among symbol demodulation values. FIG.

FIG. 2C is a graph showing a conventional normalization scheme for a received signal when the channel decoder inputs the second to seventh bits among the symbol demodulation values. FIG.

3 is a block diagram showing a normalization apparatus for a channel decoder input signal according to an embodiment of the present invention.

4 is a flowchart illustrating a normalization method for a channel decoder input signal according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. First, in adding reference numerals to the components of each drawing, it should be noted that the same components have the same reference numerals as much as possible even though they are shown in different drawings. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

The normalization apparatus and method according to an embodiment of the present invention first obtain a variance of demodulation symbol values input to a channel decoder. In addition, the present invention uses the relationship that the variance can be represented by the sum of the power of the demodulation symbol and the added noise power in the absence of noise. Obtain

In addition, the present invention is the size of the demodulation symbol when assuming no noise Can be normalized to the channel decoder input signal by accurately mapping to the desired normalization reference level. In this case, the normalization according to the present invention may be referred to as a process of re-expressing the signal of the channel decoder input terminal in a relative ratio with respect to the demodulation symbol size at the receiver in the case where no noise is added on the wireless communication path. In addition, the normalized reference level according to an embodiment of the present invention is a quantization level (e.g., eighth level) of a specific channel decoder input stage in which each received and demodulated symbol value is mapped to a relative ratio with respect to a received demodulation symbol size in the absence of noise. Is defined.

On the other hand, if its size Is the size of the demodulated symbol in the absence of the noise Dividing by gives the value of '1'. If you want to map this to a normalization reference level (eg, +8 level), multiply it by the normalization reference level value. That is, the size of the demodulation symbol Is mapped to '8', which is a value corresponding to eight levels among the quantization levels of the channel decoder input terminal. The size of the demodulation symbol for all received demodulation symbol values Divide by and multiply the normalization reference level again. Then, it is mapped to the quantization level of the channel decoder input terminal, that is, from the -32 level to the +31 level and input to the channel decoder. Of course, values less than -32 levels are mapped to -32 levels, while values greater than +31 levels are mapped to +31 levels.

On the other hand, the size of the demodulation symbol in the absence of noise In order to find the method, simply use the average of demodulation symbols in a certain period. The above method is accurate for channel environments with very high signal to noise ratios. Can be obtained. However, in the environment in which the channel decoder operates, the signal-to-noise ratio is very poor. Does not match the average of the demodulation symbol over a period of time.

Therefore, in the present invention, the signal size of the received demodulation symbol at the receiver when the noise is not added. In order to obtain, use the relationship represented by the sum of the power of the received demodulation symbol at the receiver and the noise power added to the actually received demodulation symbol when it is assumed that the distribution of received and demodulated symbol values is noisy.

The above relationship can be represented by Equation 1 below.

At this time, the <Equation 1> Is the size of demodulation symbol in the absence of noise Means, said Denotes the k-th demodulation symbol value in the N-th predetermined section. In addition, the left side of Equation 1 indicates the variance of demodulation symbol values in the N-th section. And the right side of Equation 1 Denotes the power of the received demodulation symbol at the receiver in the absence of noise. And the right side of Equation 1 Denotes a variance of noise added to the demodulation symbol actually received in the N-th period.

On the other hand, the CDMA receiver can obtain the variance of the noise using a data unmodulated pilot channel or a sync channel continuously received. That is, the received signal input to the receiver includes a signal of a pilot channel and a synchronization channel. In addition, a plurality of symbol demodulators, which are divided into the receivers, despread the received signals input in several paths, respectively, and multiply the code channel classification codes such as corresponding Walsh codes to obtain symbols of each channel. By considering that each symbol includes not only a signal component but also a noise component, it is possible to obtain a noise variance value. The noise variance is obtained from a noise meter. The noise variance value used here is obtained by using the symbols input for a certain period, which is input to the noise meter. Symbols The noise variance of this section can be obtained as follows. here, The number of squared values over one interval of symbols It can be seen that / 2. At this time, the noise measuring instrument is used for each symbol demodulator ( Above) / 2 squared values Divide by the number of symbols and take the average to find the noise variance of each signal path. The noise measurer adds the noise dispersion values of the S number to obtain the noise dispersion value.

On the other hand, the signal size of the received demodulation symbol at the receiver when it is assumed that noise in Equation 1 is not added ( ) Can be obtained as Equation 2 below.

The left side of Equation 2 Is the signal size of the received demodulation symbol at the receiver when no noise is added. And the right side of Equation 2 is Denotes the variance of received and demodulated symbol values during N intervals. K in Equation 2 denotes an order of each symbol value. And the right side of Equation 2 Denotes a variance of noise added to the demodulation symbol actually received in the N-th period.

3 is a block diagram illustrating a normalization apparatus for a channel decoder input signal according to an embodiment of the present invention. Hereinafter, a description will be given with reference to FIGS. 1 to 2C.

In an embodiment of the present invention, in order to input the received demodulation symbol values 150 to the channel decoder 140 through the requantization process, first, the size of the demodulation symbol in the absence of noise It is necessary to obtain the process. For this purpose, Equation 2 is used. That is, the configuration of the demodulated symbol squared averager 310, the noise detector 320, and the squared square root 330 of FIG. 3 is the magnitude of the demodulated symbol in the absence of noise. Can be obtained.

The normalizer 340 calculates each received demodulation symbol value 150. Divide by and multiply by the normalized reference level. The normalizer 340 maps the multiplication result to a quantization level (eg, 64 levels) of the input of the channel decoder 140. In an embodiment of the present invention, the normalizer 340 takes the integer part of the result of the multiplication. The normalizer 340 maps the integer value from the -32 level to the +31 level, which is the quantization level of the input of the channel decoder 140. In this case, each of the integer values corresponds to each level among all quantization levels (64 levels) of the input terminal of the channel decoder 140.

That is, the normalizer 340 according to the embodiment of the present invention has the magnitude of the demodulation symbol when there is no noise. Normalize the received symbol values such that is correctly mapped to the normalization reference level. This allows the channel decoder 140 to have a high channel decoding performance.

Hereinafter, the operation of the normalization device configuration according to the embodiment of the present invention will be described with reference to FIG. 3.

The demodulation symbol squared averager 310 squares the value of each received and demodulated symbol during the corresponding interval and obtains an average value of variance. In this case, the corresponding section may be a power control group unit or a transmission frame unit.

The noise measurer 320 accumulates symbol values obtained by accumulating orthogonal spreading code lengths on the demodulated symbol values of the pilot or sync channel output from the first symbol demodulator 121 to the Nth symbol demodulator 12N, respectively, and the next. Calculate and square the difference from the symbol value obtained by accumulating the length of the orthogonal spreading code times. The noise measurer 320 averages the square values corresponding to the N number of signal paths during the corresponding interval, and adds the average value of each signal path to obtain a noise dispersion value.

The operator 350 subtracts the noise variance value for the corresponding interval from the variance of the received and demodulated symbol values during the corresponding interval.

The square root root 330 is a square root squared with respect to the output of the operator 350. ), The size of the received demodulation symbol at the receiver in the case where no noise is added.

The normalizer 340 calculates the size of the received demodulation symbol in the case where there is no noise. Divide by and multiply by the normalized reference level. The normalizer 340 outputs a normalized demodulation symbol in which the multiplication result is mapped to a quantization level (eg, 64 levels) of the input terminal of the channel decoder 140.

4 is a flowchart illustrating a normalization method for a channel decoder input signal according to an embodiment of the present invention. Hereinafter, a description will be given with reference to FIGS. 1 to 3.

When the received demodulated symbol is input in step 410, the size of the received demodulation symbol when there is no noise during the corresponding period (step 420). Calculate At this time, the calculation of step 420 may use the equation (2).

In step 430, the respective received demodulation symbol values are divided by the magnitude of the received demodulation symbol in the absence of the noise, and then multiplied by the normalized reference level. In step 440, the result of the calculation in step 430 is mapped to the quantization level of the channel decoder input terminal.

3 to 4, the normalization apparatus and method for the channel decoder input signal according to an embodiment of the present invention is first the size of the demodulation symbol in the absence of noise Obtain In this case, Equation 2 may be used. In addition, the configuration according to an embodiment of the present invention for implementing Equation 2 may be a demodulated symbol squared average, noise measuring device, square root and arithmetic operator. The present invention calculates each received demodulation symbol value. Divide by and multiply by the normalized reference level. The present invention maps the multiplication result to the quantization level of the channel decoder input terminal. As a result, the present invention provides the size of the demodulation symbol in the absence of noise. Since the received demodulation symbol values can be normalized so that A is correctly mapped to the normalization reference level, the channel decoder can have high channel decoding performance.

Therefore, in the normalization apparatus and method according to an embodiment of the present invention, first, the size of a demodulation symbol in the absence of noise Obtain And wherein each received demodulation symbol value is obtained Dividing by and multiplying by the normalized reference level again, the size of the demodulation symbol in the absence of the noise Since the received demodulation symbol values can be normalized so that H is correctly mapped to the normalization reference level, the channel decoder can have high channel decoding performance.

Claims (13)

In the normalization apparatus of the channel decoder input stage in a code division multiple access mobile communication system, A received demodulation symbol size calculating unit for inputting received and demodulated symbol values during a corresponding interval to obtain a size of a received demodulation symbol in the absence of noise; Each symbol value during the corresponding period is divided by the size of the received demodulation symbol in the absence of the noise, multiplied by a normalized reference level value, and a normalizer for mapping the multiplication result to the quantization level of the channel decoder input terminal. Normalization device. The reception demodulation symbol size calculation unit of claim 1, A normalization device characterized by using Equation 3 below. At this time, the <Equation 3> Is the size of demodulation symbol in the absence of noise Means, said Denotes the k-th demodulation symbol value in the N-th predetermined corresponding section. In addition, the left side of Equation 3 indicates the variance of demodulation symbol values in the N-th section. And the right side of Equation 3 Denotes the power of the received demodulation symbol at the receiver in the absence of noise. And the right side of Equation 3 Denotes a variance of noise added to the demodulation symbol actually received in the N-th period. The method of claim 2, wherein the reception demodulation symbol size calculation unit, A demodulation symbol squared averager that squares each received and demodulated symbol value during the corresponding interval and obtains a variance value that is an average thereof; A noise measuring device for obtaining power of noise added during the corresponding section; An operator for subtracting the noise power added during the interval from the distribution of the received and demodulated symbol values during the interval; And a square root square that obtains the magnitude of the received demodulation symbol in the absence of the noise by taking the square root of the output of the calculator. The method of claim 3, wherein the corresponding section, Normalization apparatus, characterized in that the power control group unit and the transmission frame unit. The method of claim 4, wherein the normalizer, And an integer part of the multiplication result and mapping to each quantization level of the channel decoder input terminal. In the normalization apparatus of the channel decoder input stage in a code division multiple access mobile communication system, A demodulation symbol squared averager that squares each received and demodulated symbol value and obtains an average variance value; A noise measuring device for obtaining power of noise added during the corresponding section; An operator for subtracting the noise power added during the interval from the distribution of the received and demodulated symbol values during the interval; A square root root that takes the square root of the output of the operator to obtain the magnitude of the received demodulation symbol in the absence of noise; Each symbol value during the corresponding period is divided by the size of the received demodulation symbol in the absence of the noise, multiplied by a normalized reference level value, and a normalizer for mapping the multiplication result to the quantization level of the channel decoder input terminal. Normalization device. The method of claim 6, wherein the corresponding section, Normalization apparatus, characterized in that the power control group unit and the transmission frame unit. The method of claim 7, wherein the normalizer, And an integer part of the multiplication result and mapping to each quantization level of the channel decoder input terminal. In the normalization method of the channel decoder input terminal in a code division multiple access mobile communication system, A first step of obtaining a size of a received demodulation symbol in the absence of noise using a received demodulation symbol input during a corresponding section; Dividing each received demodulation symbol by the magnitude of the received demodulation symbol in the absence of the noise and multiplying it by a normalized reference level again; And a third step of mapping the multiplication result to the quantization level of the channel decoder input terminal. The method of claim 9, wherein the first step, A normalization method characterized by using Equation 4 below. At this time, the <Equation 4> Is the size of demodulation symbol in the absence of noise Means, said Denotes the k-th demodulation symbol value in the N-th predetermined corresponding section. In addition, the left side of Equation 4 indicates the variance of demodulation symbol values in the N-th section. And the right side of Equation 4 Denotes the power of the received demodulation symbol at the receiver in the absence of noise. And the right side of Equation 4 Denotes a variance of noise added to the demodulation symbol actually received in the N-th period. The method of claim 10, wherein the first step, A first step of obtaining a variance value which is the square value of each received and demodulated symbol value during the corresponding interval; Obtaining a power of the noise added during the corresponding section; A third step of subtracting the noise power added during the interval from the distribution of the received and demodulated symbol values during the interval; And a fourth step of obtaining a magnitude of a received demodulation symbol in the absence of the noise by taking a square root of the output of the calculator. The method of claim 11, wherein the corresponding section, Normalization apparatus, characterized in that the power control group unit and the transmission frame unit. The method of claim 12, wherein the fourth step, And taking an integer part of the multiplication result and mapping it to each quantization level of the channel decoder input terminal.