JP2004295490A - Normalization device and normalization method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the possible overflow of mantissa of valid data from a memory for mantissa memory even in the presence of a number of invalid data inputted. <P>SOLUTION: A data memory 1 receives a data series and stores the mantissa of each of data. A shift number memory 2 receives the data series and stores a data shift number showing the index part of each of data. A reference shift number calculation means receives a control data series transmitted by a control channel, and calculates an average value of a control data shift number showing the index part of each of control data in this control data series to generate a reference shift number. A normalization part 5 subtracts the reference shift number from each data shift number stored by the shift number memory 2, and shifts to the right the mantissa of the data corresponding to the data shift number for the value of this result. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a normalization device and a normalization method, and more particularly to a normalization device and a normalization method for receiving a data sequence and normalizing each data in the data sequence.
[0002]
[Prior art]
In general, in a mobile communication system, in order to reduce the circuit scale of a decoding circuit that receives and decodes data transmitted from a base station, a processing unit that is processed at one time by the decoding circuit at a preceding stage of the decoding circuit ( By matching indices with slots, frames, and the like constituted by a plurality of data, a normalization process for reducing the bit length of the input of the decoding circuit is performed.
[0003]
The conventional normalization device is provided in a stage preceding the decoding circuit, and includes a data memory 10, a shift number memory 11, a reference shift number calculation unit 12, and a normalization unit 13, as shown in FIG. Upon receiving the data transmitted from the base station, the data is normalized.
[0004]
The data memory 10 stores data series ((D [i], S [i]) such as slots and frames, i = 0 to n-1, where D is a mantissa of data and S is an exponent of data. The mantissa part (D [i], i = 0 to n-1) of each data in the received data series is stored as a processing unit (n).
[0005]
The shift number memory 11 stores a shift number (S [i], i = 0 to n−1) indicating an exponent part of each data for the processing unit (n).
[0006]
The reference shift number calculation unit 12 calculates the average value ((S [0] + S [1] +... + S [n−1]) / n) of the shift numbers of the respective data stored in the shift number memory 11. To generate a reference shift number (Sb).
[0007]
The normalizing unit 13 subtracts the reference shift number (Sb) from each of the shift numbers (S [i], i = 0 to n-1) stored in the shift number memory 11, and subtracts these values by the resulting value. Is shifted rightward (N [i] = D [i] >> (S [i] -Sb), i = 0 to n-1).
[0008]
As a result, each data in the data series ((D [i], S [i]), i = 0 to n-1) received by the normalization device is normalized ((N [i], Sb ), I = 0 to n−1).
[0009]
Here, the symbol >> is a symbol indicating that data on the left side of the symbol is shifted rightward by a value on the right side of the symbol.
[0010]
The normalization device has a mantissa memory of a predetermined size (for example, 16 bits), and stores D [i] on this memory.
[0011]
On the other hand, although the base station transmits data on the data channel, there are times when data is actually transmitted and times when data is not transmitted depending on the presence or absence of data to be transmitted. The normalizer receives data on this data channel even when the base station is not actually transmitting data.
[0012]
Compared with valid data received when the base station is actually transmitting data, invalid data (hereinafter referred to as DTX) received when not transmitting data is generally smaller in amplitude. Power is small.
[0013]
Therefore, in the method using the average value of the power of the data (approximate by the exponent part of the data) (for example, the aforementioned reference shift number (Sb)) as the index of the normalization, DTX is included in each of the received data. When a large number of DTXs exist, the average value (Sb) of the power of the data in the processing unit becomes small because the DTX has a small amplitude, and the exponent part (S [i]) of the effective data in the processing unit and the average value The difference from (Sb) increases. If the mantissa part of the valid data is subjected to this difference right-shift normalization, the mantissa memory has only a predetermined size, so that the mantissa of the valid data is more likely to overflow from this memory. .
[0014]
A method using the average value of the amplitude values of the data signals in the processing unit as an index similar to the above is proposed in, for example, Patent Document 1.
[0015]
[Patent Document 1]
JP 2002-232302 A (paragraph “0034”, FIG. 1)
[0016]
[Problems to be solved by the invention]
In the conventional normalization device described above, the reference shift number calculation unit calculates the average value of the shift numbers of the respective data stored in the shift number memory to generate the reference shift number (Sb), and the normalization unit The reference shift number (Sb) is subtracted from each of the shift numbers (S [i], i = 0 to n-1), and the mantissa part of each data corresponding to these shift numbers is right-shifted by a value resulting from the subtraction. (N [i] = D [i] >> (S [i] -Sb), i = 0 to n-1). Therefore, when a large number of invalid data exists in each of the received data, Since the average value (Sb) of the power of the data becomes small and the difference between the exponent part (S [i]) and the average value (Sb) of the effective data in the processing unit increases, the mantissa part of this effective data After normalization, the mantissa of the valid data is There is a problem that the possibility of overflow from Li increases.
[0017]
An object of the present invention is to eliminate such a conventional drawback, so that even if there are a lot of invalid data among the received data, the normalization of the mantissa part of the valid data is unlikely to overflow from the memory for the mantissa. An apparatus and a normalization method are provided.
[0018]
[Means for Solving the Problems]
A first normalization device of the present invention is a normalization device that receives a data sequence, normalizes each data in the data sequence, and outputs each data.
A data memory that receives the data sequence sent on the data channel and stores a mantissa of each of the data in the data sequence;
A shift number memory that receives the data sequence sent on the data channel and stores a data shift number indicating an exponent part of each of the data in the data sequence;
Reference shift number calculating means for receiving a control data sequence sent on the control channel, calculating an average value of control data shift numbers indicating an exponent part of each control data in the control data sequence, and generating a reference shift number,
Each of the reference shift numbers is subtracted from each of the data shift numbers stored in the shift number memory, and the mantissa part of the data stored in the data memory corresponding to the data shift number is shifted to the right by a value resulting from the subtraction. A normalization unit,
It is configured with.
[0019]
The reference shift number calculating means of the first normalization device of the present invention includes:
An information memory that receives the control data sequence sent on the control channel and stores the control data shift number indicating the exponent part of each control data in the control data sequence,
A reference shift number calculation unit that calculates an average value of the control data shift numbers stored in the information memory to generate the reference shift number,
It is configured with.
[0020]
A second normalization device of the present invention is a normalization device that receives a data sequence, normalizes each data in the data sequence, and outputs each data.
A data memory that receives the data sequence sent on the data channel and stores a mantissa of each of the data in the data sequence;
A shift number memory that receives the data sequence sent on the data channel and stores a data shift number indicating an exponent part of each of the data in the data sequence;
In response to the control data sequence sent on the control channel, it is determined whether or not each of the data in the data sequence is invalid data based on the control data shift number indicating the exponent part of each control data in the control data sequence. Calculating a DTX determination threshold value, comparing each of the data shift numbers stored in the shift number memory with the DTX determination threshold value, and determining whether the data corresponding to the data shift number is invalid data according to the comparison result; Reference shift number calculation means for determining whether or not each of the data, the average value of the data shift number of the data other than invalid data to generate a reference shift number,
Each of the reference shift numbers is subtracted from each of the data shift numbers stored in the shift number memory, and the mantissa part of the data stored in the data memory corresponding to the data shift number is shifted to the right by a value resulting from the subtraction. A normalization unit,
It is configured with.
[0021]
The reference shift number calculating means of the second normalization device of the present invention includes:
An information memory for receiving the control data sequence sent on the control channel and storing the control data shift number indicating an exponent part of each control data in the control data sequence;
A DTX determination threshold for determining whether or not each of the data in the data series is invalid data is calculated from each of the control data shift numbers stored in the information memory, and each of the control data stored in the shift number memory is calculated. The data shift number is compared with the DTX determination threshold, and it is determined whether or not the data corresponding to the data shift number is invalid data according to the comparison result, and the data shift of the data other than invalid data is performed. A reference shift number calculation unit that calculates an average value of the numbers to generate a reference shift number,
It is configured with.
[0022]
A first normalization method according to the present invention is a normalization method for receiving a data sequence, normalizing each data in the data sequence, and outputting each data.
A first step of receiving the data sequence sent on the data channel and storing respective mantissas of each of the data in the data sequence;
A second step of receiving the data sequence sent on the data channel and respectively storing a data shift number indicating an exponent of each of the data in the data sequence;
A third step of receiving a control data sequence sent on the control channel, calculating an average value of control data shift numbers indicating an exponent part of each control data in the control data sequence, and generating a reference shift number;
Subtracting the reference shift number generated in the third step from each of the data shift numbers stored in the second step, in the first step corresponding to the data shift number by a value resulting from the subtraction; A fourth step of right shifting each of the mantissa parts of the stored data;
It is comprised including.
[0023]
A second normalization method according to the present invention is a normalization method for receiving a data sequence, normalizing each data in the data sequence, and outputting each data.
A first step of receiving the data sequence sent on the data channel and storing respective mantissas of each of the data in the data sequence;
A second step of receiving the data sequence sent on the data channel and respectively storing a data shift number indicating an exponent of each of the data in the data sequence;
In response to the control data sequence sent on the control channel, it is determined whether or not each of the data in the data sequence is invalid data based on the control data shift number indicating the exponent part of each control data in the control data sequence. A third step of calculating a DTX determination threshold,
The data shift number stored in the second step is compared with the DTX determination threshold calculated in the third step, and the data corresponding to the data shift number is invalid data according to the comparison result. A fourth step of determining whether or not there is, respectively;
A fifth step of calculating an average value of the data shift numbers of the data not determined as invalid data in the fourth step to generate a reference shift number;
The reference shift number generated in the fifth step is subtracted from each of the data shift numbers stored in the second step, and the first step corresponding to the data shift number corresponds to a value resulting from the subtraction. A sixth step of right shifting each of the mantissa parts of the stored data;
It is comprised including.
[0024]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, embodiments of the present invention will be described with reference to the drawings.
[0025]
FIG. 1 is a block diagram showing a first embodiment of the normalizing device of the present invention.
[0026]
The present embodiment shown in FIG. 1 is provided in a preceding stage of a decoding circuit, and includes a data memory 1, a shift number memory 2, an information memory 3, a reference shift number calculation unit 4, and a normalization unit 5. Receive the data transmitted from the base station and normalize this data.
[0027]
The data memory 1 receives a data sequence such as a slot and a frame transmitted by a data channel and stores a mantissa of each data in the data sequence.
[0028]
The shift number memory 2 receives a data sequence transmitted on the data channel and stores a data shift number indicating an exponent part of each data in the data sequence.
[0029]
The information memory 3 receives the control data sequence sent by the control channel and stores the control data shift number indicating the exponent part of each control data in the control data sequence.
[0030]
The reference shift number calculator 4 calculates the average value of the respective control data shift numbers stored in the information memory 3 to generate the reference shift number.
[0031]
The normalizing unit 5 subtracts the reference shift number from each of the data shift numbers stored in the shift number memory 2, and subtracts each of the data stored in the data memory 1 corresponding to the data shift number by the value resulting from the subtraction. Are respectively shifted to the right.
[0032]
The data channel is a channel for transmitting user data, and the control channel is a Dedicated Physical Control Channel (hereinafter, DPCCH) for transmitting a plurality of control information in a physical layer. The control data in the control data sequence transmitted on the DPCCH has a mantissa part B and an exponent part C, and the exponent part C indicates the number of control data shifts.
[0033]
The plurality of pieces of control information are known pattern pilot information (PILOT), transmission power control command (TPC), feedback information (FBI), and transport combination indicator (TFCI) used for channel estimation in synchronous detection. The control data is data relating to at least one control information of a plurality of pieces of control information.
[0034]
Next, the operation of the normalization device of the present embodiment will be described in detail with reference to FIG.
[0035]
FIG. 2 is a flowchart illustrating an example of the operation according to the embodiment of the present invention.
[0036]
In FIG. 1, in step 21 of FIG. 2, data series ((D [i], S [i]) such as slots and frames transmitted by the data channel by the data memory 1, i = 0 to n−1. , D is the mantissa part of the data, S is the exponent part of the data, and the mantissa part (D [i], i = 0 to n-1) of each data in this data series is set as a processing unit (n). Store.
[0037]
In step 22 of FIG. 2, the shift number memory 2 receives the data sequence transmitted on the data channel and receives the data shift number (S [i], i = 0 to n) indicating the exponent of each data in the data sequence. -1) is stored as a processing unit (n).
[0038]
In step 23 of Fig. 2, a control data sequence ((B [i], C [i]) transmitted on the control channel by the information memory 3, i = 0 to m-1, where B is a mantissa of control data , C receive the exponent part of the control data) and store the control data shift numbers (C [i], i = 0 to m-1) indicating the exponent part of each control data in this control data sequence.
[0039]
In step 24 of FIG. 2, the average value ((C [0] + C [1] +... + C [m−1]) of the respective control data shift numbers stored in the information memory 3 is calculated by the reference shift number calculator 4. ) / M) to generate a reference shift number (Sb).
[0040]
However, at the time of Multi Code in the case of W-CDMA defined by 3GPP, the average value of the number of control data shifts is ((C [0] + C [1] +... + C [m-1]) / m) × p, to generate a reference shift number (Sb). However, p is a parameter for correcting this difference because the power of the DPCCH is larger than the DPDCH (Dedicated Physical Data Channel) at the time of Multi Code. Assuming that the power is, for example, 1.4 times the power of the DPDCH, p = 1 / 1.4.
[0041]
In step 25 of FIG. 2, the normalizing unit 5 subtracts the reference shift number (Sb) from each data shift number (S [i], i = 0 to n−1) stored in the shift number memory 2.
[0042]
In step 26 of FIG. 2, the normalization unit 5 shifts the mantissa part of each data stored in the data memory 1 corresponding to each data shift number by right shift (N [I] = D [i] >> (S [i] -Sb), i = 0 to n-1).
[0043]
As a result, each data in the data series ((D [i], S [i]), i = 0 to n-1) received by the normalization device is normalized ((N [i], Sb ), I = 0 to n−1).
[0044]
Accordingly, the calculation of the reference shift number (Sb) is performed without using the data shift number (S [i], i = 0 to n-1) indicating the exponent part of each data in the data series. Is performed using the control data shift number (C [i], i = 0 to m-1) indicating the exponent part of the control data in the data series, so that the influence of invalid data (DTX) in the data series can be eliminated, and normalization can be performed. Even if there are a lot of invalid data among the data input by the apparatus, the possibility that the mantissa part of the valid data overflows from the memory for the mantissa is reduced.
[0045]
FIG. 3 is a block diagram showing a second embodiment of the normalizing device according to the present invention.
[0046]
The present embodiment shown in FIG. 3 is provided in a preceding stage of a decoding circuit, and is configured by a data memory 1, a shift number memory 2, an information memory 3, a reference shift number calculation unit 6, and a normalization unit 5. Receive the data transmitted from the base station and normalize this data. The difference from the normalization device of the first embodiment is only the reference shift number calculation unit 6, and the other components are the same.
[0047]
The reference shift number calculation unit 6 calculates a DTX determination threshold for determining whether or not each data of the data series is invalid data from the respective control data shift numbers stored in the information memory 3. The stored data shift numbers are compared with a DTX determination threshold value, and it is determined whether or not the data corresponding to the data shift number is invalid data according to the comparison result, and the data shift number of data other than invalid data is determined. Is calculated to generate a reference shift number.
[0048]
Next, the operation of the normalization device of the present embodiment will be described in detail with reference to FIG.
[0049]
FIG. 4 is a flowchart illustrating an example of the operation of the exemplary embodiment of the present invention.
[0050]
In FIG. 3, the operation from step 41 in FIG. 4 to step 43 in FIG. 4 is the same as the operation from step 21 in FIG. 2 to step 23 in FIG. 2 of the normalization device of the first embodiment.
[0051]
In step 44 of FIG. 4, the reference shift number calculating unit 6 determines whether or not each data of the data series is invalid data based on the respective control data shift numbers stored in the information memory 3 in step 43 of FIG. DTX determination threshold T = ((C [0] +... + C [m-1]) / m) × E). Here, E is a coefficient for setting the DTX determination threshold value to an intermediate value, and this value is determined by performing a test using an actual device or the like.
[0052]
However, at the time of Multi Code in the case of W-CDMA defined by 3GPP, the DTX determination threshold T is calculated as ((C [0] + ... + C [m-1]) / m) × E × p. . This p is the same as p in the first embodiment.
[0053]
In step 45 of FIG. 4, the reference shift number calculation unit 6 calculates the respective data shift numbers (S [i], i = 0 to n-1) stored in the shift number memory 2 in step 42 and in step 44. The data is compared with the DTX determination threshold T, and when the comparison result is S [i]> = T, it is determined that the data corresponding to the data shift number is invalid data.
[0054]
In a fourth step 46, the reference shift number calculating unit 6 calculates an average value of the data shift numbers (S [i] satisfying S [i] <T) of the data not determined to be invalid data in step 45. To generate a reference shift number (Sb).
[0055]
In steps 44 to 46 of FIG. 4, explanation will be added. The DTX determination threshold T is a threshold for separating DTX because there is a possibility that DTX is included in the data, and the DTX is separated using the fact that the power is smaller than that of normal data. Here, since the exponent is used as an approximate value of the power, the exponent is averaged. Basically, it is expected that the power (= exponent part) of effective data will be about the same. Assuming that the exponent part is 10, for example, and the exponent part of DTX is, for example, 20 (here, the larger the exponent part is, the smaller the power), the value above the midpoint 15 is DTX, and the value below 15 is valid data. Is determined. In such a case, threshold value 15 can be obtained by setting E = 1.5. This value is determined by conducting a test using an actual device or the like.
[0056]
The operations from step 47 to step 48 in FIG. 4 are the same as the operations from step 25 to step 26 in FIG. 2 of the normalizing device according to the first embodiment.
[0057]
As a result, each data in the data series ((D [i], S [i]), i = 0 to n-1) received by the normalization device is normalized ((N [i], Sb ), I = 0 to n−1).
[0058]
Thus, since the calculation of the reference shift number (Sb) is performed using only the data shift number indicating the exponent part of the valid data in the data series, the influence of invalid data in the data series can be eliminated, and the normalization device Even if a large number of invalid data exist in the input data, the possibility that the mantissa of valid data overflows from the memory for the mantissa is reduced.
[0059]
In the normalization device according to the first embodiment and the normalization device according to the second embodiment, when the control data sequence sent on the control channel is normalized as the control data sequence of only PILOT in the processing unit slot, Since the PILOT (pilot information) comes before the data (D, S), Sb can be calculated in advance. Therefore, the control data shift number (C [i]) indicating the exponent part of the control data in the information memory 3 , I = 0 to m-1), and the circuit scale can be reduced.
[0060]
【The invention's effect】
As described above, the normalization apparatus according to the present invention is configured such that the reference shift number calculating means receives the control data sequence sent on the control channel and receives the control data shift indicating the exponent part of each control data in the control data sequence. Calculates the average of the numbers to generate the reference shift number, and the normalization unit subtracts the reference shift number from each data shift number stored in the shift number memory, and corresponds to the data shift number corresponding to the value resulting from this subtraction. In order to shift the mantissa part of the data stored in the data memory 1 to the right, the calculation of the reference shift number is performed by using the control data in the control data series without using the data shift number indicating the exponent part of the data in the data series. Control data shift number indicating the exponent part of Can eliminate the influence of invalid data of the inner, even if invalid data out of the data normalization devices are entered there are many, potentially mantissa of valid data overflows from the memory for the mantissa is reduced.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a first embodiment of a normalization device according to the present invention.
FIG. 2 is a flowchart illustrating an example of an operation according to the first exemplary embodiment of the present invention.
FIG. 3 is a block diagram showing a second embodiment of the normalization device of the present invention.
FIG. 4 is a flowchart illustrating an example of an operation according to the second exemplary embodiment of the present invention.
FIG. 5 is a block diagram showing a conventional normalization device.
[Explanation of symbols]
Reference Signs List 1 data memory 2 shift number memory 3 information memory 4 reference shift number calculation section 5 normalization section 6 reference shift number calculation section 10 data memory 11 shift number memory 12 reference shift number calculation section 13 normalization section

Claims (10)

In a normalization device that receives a data series, normalizes each data in the data series and outputs each data,
A data memory that receives the data sequence sent on the data channel and stores a mantissa of each of the data in the data sequence;
A shift number memory that receives the data sequence sent on the data channel and stores a data shift number indicating an exponent part of each of the data in the data sequence;
Reference shift number calculating means for receiving a control data sequence sent on the control channel, calculating an average value of control data shift numbers indicating an exponent part of each control data in the control data sequence, and generating a reference shift number,
Each of the reference shift numbers is subtracted from each of the data shift numbers stored in the shift number memory, and the mantissa part of the data stored in the data memory corresponding to the data shift number is shifted to the right by a value resulting from the subtraction. A normalization unit,
A normalization device comprising: The reference shift number calculating means,
An information memory that receives the control data sequence sent on the control channel and stores the control data shift number indicating the exponent part of each control data in the control data sequence,
A reference shift number calculation unit that calculates an average value of the control data shift numbers stored in the information memory to generate the reference shift number,
The normalization device according to claim 1, further comprising: In a normalization device that receives a data series, normalizes each data in the data series and outputs each data,
A data memory that receives the data sequence sent on the data channel and stores a mantissa of each of the data in the data sequence;
A shift number memory that receives the data sequence sent on the data channel and stores a data shift number indicating an exponent part of each of the data in the data sequence;
In response to the control data sequence sent on the control channel, it is determined whether or not each of the data in the data sequence is invalid data based on the control data shift number indicating the exponent part of each control data in the control data sequence. Calculating a DTX determination threshold value, comparing each of the data shift numbers stored in the shift number memory with the DTX determination threshold value, and determining whether the data corresponding to the data shift number is invalid data according to the comparison result; Reference shift number calculation means for determining whether or not each of the data, the average value of the data shift number of the data other than invalid data to generate a reference shift number,
Each of the reference shift numbers is subtracted from each of the data shift numbers stored in the shift number memory, and the mantissa part of the data stored in the data memory corresponding to the data shift number is shifted to the right by a value resulting from the subtraction. A normalization unit,
A normalization device comprising: The reference shift number calculating means,
An information memory for receiving the control data sequence sent on the control channel and storing the control data shift number indicating an exponent part of each control data in the control data sequence;
A DTX determination threshold for determining whether or not each of the data in the data series is invalid data is calculated from each of the control data shift numbers stored in the information memory, and each of the control data stored in the shift number memory is calculated. The data shift number is compared with the DTX determination threshold, and it is determined whether or not the data corresponding to the data shift number is invalid data according to the comparison result, and the data shift of the data other than invalid data is performed. A reference shift number calculation unit that calculates an average value of the numbers to generate a reference shift number,
The normalization device according to claim 3, further comprising: The data channel is a channel for transmitting user data, and the control channel is a Dedicated Physical Control Channel (hereinafter, referred to as Dedicated Physical Control Channel) for transmitting a plurality of control information in a physical layer. , DPCCH.), And each of the control data in the control data sequence transmitted on the DPCCH has a mantissa part B and an exponent part C, and the exponent part C indicates the control data shift number. The normalization device according to claim 1, 2, 3, or 4, wherein The plurality of control information includes pilot information (PILOT) of a known pattern used for channel estimation in synchronous detection, a transmission power control command (TPC), feedback information (FBI), and a transport combination indicator (TFCI). The normalization device according to claim 5, wherein the control data is data relating to at least one control information of a plurality of the control information. In a normalization method of receiving a data series, normalizing each data in the data series and outputting each data,
A first step of receiving the data sequence sent on the data channel and storing respective mantissas of each of the data in the data sequence;
A second step of receiving the data sequence sent on the data channel and respectively storing a data shift number indicating an exponent of each of the data in the data sequence;
A third step of receiving a control data sequence sent on the control channel, calculating an average value of control data shift numbers indicating an exponent part of each control data in the control data sequence, and generating a reference shift number;
Subtracting the reference shift number generated in the third step from each of the data shift numbers stored in the second step, in the first step corresponding to the data shift number by a value resulting from the subtraction; A fourth step of right shifting each of the mantissa parts of the stored data;
A normalization method comprising: In a normalization method of receiving a data series, normalizing each data in the data series and outputting each data,
A first step of receiving the data sequence sent on the data channel and storing respective mantissas of each of the data in the data sequence;
A second step of receiving the data sequence sent on the data channel and respectively storing a data shift number indicating an exponent of each of the data in the data sequence;
In response to the control data sequence sent on the control channel, it is determined whether or not each of the data in the data sequence is invalid data based on the control data shift number indicating the exponent part of each control data in the control data sequence. A third step of calculating a DTX determination threshold,
The data shift number stored in the second step is compared with the DTX determination threshold calculated in the third step, and the data corresponding to the data shift number is invalid data according to the comparison result. A fourth step of determining whether or not there is, respectively;
A fifth step of calculating an average value of the data shift numbers of the data not determined as invalid data in the fourth step to generate a reference shift number;
The reference shift number generated in the fifth step is subtracted from each of the data shift numbers stored in the second step, and the first step corresponding to the data shift number corresponds to a value resulting from the subtraction. A sixth step of right shifting each of the mantissa parts of the stored data;
A normalization method comprising: The data channel is a channel for transmitting user data, and the control channel is a Dedicated Physical Control Channel (hereinafter, referred to as Dedicated Physical Control Channel) for transmitting a plurality of control information in a physical layer. , DPCCH.), And each of the control data in the control data sequence transmitted on the DPCCH has a mantissa part B and an exponent part C, and the exponent part C indicates the control data shift number. 9. The normalization method according to claim 7, wherein: The plurality of control information includes pilot information (PILOT) of a known pattern used for channel estimation in synchronous detection, a transmission power control command (TPC), feedback information (FBI), and a transport combination indicator (TFCI). 10. The normalization method according to claim 9, wherein the control data is data relating to at least one control information of a plurality of the control information.

Images