JP4560963B2 - QAM decoder - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a QAM decoding apparatus, and more particularly, in decoding of a radio communication signal of orthogonal frequency division multiplexing (OFDM) using a multi-level QAM modulation scheme, a data signal that has fluctuated due to fading or noise or the like. It relates to prevention of judgment errors.
[0002]
[Prior art]
As a method of estimating amplitude fluctuation and phase fluctuation of a data signal from the level of a pilot signal included in a received signal in decoding of an orthogonal frequency division multiplexing (OFDM) wireless communication signal using a multi-level QAM modulation scheme A method using an average of several pilot carriers or a method of linearly interpolating between pilot carriers is generally known. However, although the former average method is effective for amplitude fluctuations due to thermal noise, etc., it cannot cope with phase fluctuations due to fading, etc., and the latter linear interpolation method makes erroneous estimations when pilot carriers deteriorate. there's a possibility that. Further, for the amplitude correction of the data signal based on the above estimation, conventionally, an AGC (automatic gain control) amplifier provided in the intermediate frequency circuit was controlled. Had the disadvantage that processing was delayed.
[0003]
[Problems to be solved by the invention]
In view of the above, the present invention estimates amplitude fluctuation and phase fluctuation using both the average method and the linear interpolation method based on the amplitude of the pilot signal, and for data signal and data determination based on the estimation result. It is an object of the present invention to provide a QAM decoding apparatus that prevents an error in decoded data by correcting the threshold value.
[0004]
[Means for Solving the Problems]
The present invention relates to a baseband I obtained by frequency-converting a received radio frequency signal in decoding of a radio communication signal of an orthogonal frequency division multiplexing (OFDM) using a modulation scheme of multilevel QAM (Quadrature Amplitude Modulation). A time domain signal for each channel and Q channel is converted into a frequency domain signal, and a pilot signal and a data signal are output for each channel, and a pilot signal from the Fourier transform processing unit is used as a basis. Next, the fluctuations of the amplitude and phase of the data signals of the I channel and Q channel forming the OFDM are estimated based on the average level of the pilot signals and the linear interpolation of the pilot signals. and estimation means for, in advance based on the estimation result by the estimating means A data signal from the Fourier transform processing unit based on the threshold value correcting means for correcting the reference threshold value for data determination and the pilot signal from the Fourier transform processing unit and the correction threshold value by the threshold value correcting means. Provided is a QAM decoding device comprising: a data correction unit for correcting data; and a determination unit that performs data determination on a data signal from the data correction unit using a threshold value from the threshold value correction unit and outputs decoded data. is there.
[0005]
Further, the data correction unit for the I channel or the Q channel is a set of two pilot signals, and a first phase error estimation unit that estimates a phase error of the data signal for each set, The first phase error average unit that averages the estimated values estimated by the phase error estimation unit 91, and the phase with respect to all the subcarriers constituting the data signal based on the average value from the first phase error average unit A data correction unit that performs correction to rotate the data, a data temporary determination unit that performs temporary data determination using the correction threshold value from the threshold value correction unit for the data signal from the data correction unit, and the data correction unit A second phase error estimator that estimates the data phase error from the data signal and the data provisionally determined by the data tentative determination unit, and all the estimated values in the second phase error estimator And second phase error averaging unit, composed of a data re-correcting unit that rotates the phase for the entire sub-carriers forming based on the data signals an average value of from the second phase error averaging unit.
[0006]
DETAILED DESCRIPTION OF THE INVENTION
DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below based on examples with reference to the drawings. FIG. 1 is a principal block diagram showing an embodiment of a QAM decoding apparatus according to the present invention. The modulation signal is, for example, an orthogonal frequency division multiplexing (OFDM) using 16QAM. In FIG. 1, reference numeral 1 denotes a synchronization unit, which determines the timing of extracting symbols (described later) when Fourier transforming an input 16QAM I channel (ch) signal and Q channel (ch) signal. The input Ich signal and the Qch signal are baseband signals obtained by frequency-converting the received radio frequency signal, and are signals obtained by digitizing analog signals by an A / D converter (not shown). Each of these signals includes a predetermined number (four in this embodiment) of pilot signals in addition to the data signal. Reference numeral 2 denotes a Fourier transform processing unit, which converts each of the time domain Ich signal and the Qch signal, which are symbols set by the synchronization unit 1, into a frequency domain signal.
[0007]
An Ich signal estimation unit 3 extracts an Ich side pilot signal from the output signal of the Fourier transform processing unit 2, and estimates the amplitude fluctuation of the Ich data signal from the amplitude of the pilot signal. Reference numeral 4 denotes an estimation unit for the Qch signal, which extracts a Qch side pilot signal from the output signal of the Fourier transform processing unit 2 and estimates the amplitude fluctuation of the Qch data signal from the amplitude of the pilot signal. The estimation unit 3 for the Ich signal and the estimation unit 4 for the Qch signal are used as estimation means. Reference numeral 5 denotes an Ich signal threshold correction unit which corrects a preset threshold on the Ich side used for data determination in QAM based on the estimation by the estimation unit 3 for the Ich signal. A Qch signal threshold correction unit 6 corrects a preset threshold value on the Qch side used for data determination in QAM based on the estimation by the Qch signal estimation unit 4. The threshold value correcting unit 5 for the ch signal and the threshold value correcting unit 6 for the Qch signal are used as the threshold value correcting means.
[0008]
Reference numeral 7 denotes a delay unit for the Ich signal, which delays the Ich-side data signal extracted from the output signal of the Fourier transform processing unit 2 for a predetermined time, and the time between the data and the threshold value at the time of data determination in the determination unit 10 below Consistent timing. Reference numeral 8 denotes a delay unit for the Qch signal, which delays the data signal on the Qch side extracted from the output signal of the Fourier transform processing unit 2 for a predetermined time, and the time between the data and the threshold value at the time of data determination in the determination unit 10 below Consistent timing. Reference numeral 9 denotes a data correction unit. The Ich and Qch data signals from the delay units 7 and 8 are used as the threshold correction data from the threshold correction units 5 and 6 and the pilot signal from the Fourier transform processing unit 2. Data correction is based on the original. Reference numeral 10 denotes a determination unit that determines data for the corrected data signal from the data correction unit 9 based on the correction threshold values from the threshold correction units 5 and 6. The determination output becomes data (decoded data) of each signal point in 16QAM.
[0009]
Next, the present invention will be described in more detail with reference to FIG. In FIG. 2, (A) is an arrangement relation diagram of each signal point (black circle mark) (1 signal point = 4 bits) when 16QAM is set. Reference numerals 21 to 24 in the figure are threshold values used for data determination. In the case of 16QAM, two threshold values of + and − are required for each of the I axis and the Q axis. In FIG. 1, each of the threshold correction units 5 and 6 has + and − represent the polarity of the threshold. Each of these threshold values is set in advance (basic threshold value). FIG. 4B shows the relationship between the amplitude (vertical axis) of the data signal and the pilot signal with respect to the frequency (horizontal axis) drawn on the Ich side. The frequency domain signal is output from the Fourier transform processing unit 2 as described above. In the figure, 25 is a data signal, P1 to P4 are pilot signals, and 26 is a threshold value [corresponding to reference numeral 23 in FIG.
[0010]
Further, FIG. (B) shows a band of ± 20 MHz as an example with respect to the origin (= 0: direct current), and this band is divided into ± 32 (that is, 64 frequency divisions). 1 signal point (4 bits) data for 60 subcarriers of the carrier), and pilot signals P1 to P4 are sequentially assigned to 4 subcarriers. The order of this assignment is from the leftmost end (−20 MHz) to the rightmost end (+20 MHz) in the figure, and after assigning to the rightmost end, it returns to the leftmost end again, and this is repeated thereafter. One group obtained by dividing the 64 frequencies becomes one symbol. If the positions of the pilot signals P1 to P4 are expressed by the relationship of the ± 32 division, P2 and P3 are positions of ± 7, and P1 and P4 are positions of ± 21. Therefore, the number of data subcarriers between pilot signals is 13. “−20 MHz” means a frequency having a complex conjugate relationship with “+20 MHz”.
[0011]
Further, the data signal shown in FIG. (B) shows a case where only the data of the signal point So (eg, 1111 4-bit data) in FIG. . In FIG. (B), four pilot signals P1 to P4 are used, and among these, P4 is negative. However, the number or polarity of the pilot signals is arbitrarily determined depending on the communication method. Here, if reception and data determination can be performed in the state of FIG. However, in practice, the radio wave propagation path is affected by phase fluctuation (phase noise) due to fading and the like, and the receiver side is affected by thermal noise and the like in its processing circuit. In this case, the amplitude and phase of each signal point on the orthogonal axes of I and Q vary, and an error occurs in data determination.
[0012]
FIG. 5C shows an example of the state of FIG. 5B in which the level has fluctuated due to phase noise such as fading among the above-mentioned influences. The data signal 27 and the four pilot signals (P1 -P4) shows that the level varies in comparison with those in FIG. Actually, the amplitude fluctuation due to the thermal noise is further added to the fluctuation due to the phase noise such as fading. FIG. (D) shows the relationship between the two, and 28 is the phase due to the fading or the like. A data signal whose level has been fluctuated due to noise is shown, and 29 is a data signal whose level has fluctuated further due to thermal noise. Along with this data variation, the levels of the four pilot signals also vary. From the above, it is possible to estimate the fluctuation of the data signal by grasping the level fluctuation of the four pilot signals. It is the estimation unit 3 for the Ich signal and the estimation unit 4 for the Qch signal that perform this estimation, and the summation of both estimations is the amplitude variation and phase variation of each signal point on the I and Q orthogonal axes. It becomes.
[0013]
In the above estimation, each of the estimation unit 3 and the estimation unit 4 calculates the following formulas (1) and (2). In addition, a formula is described as a general formula.
g1 = (P1 + P2 + ... + (Pi-1) + Pi + ... + Pn) / n (1)
g2 (k) = [{Pi- (Pi-1)} k / .N] + (Pi-1) (2)
In the above equation, g1 represents an average level of n pilot signals (P1 to Pn), and g2 (k) is a linear equation using k as a function, and represents linear interpolation between pilot signals. Further, k is a numerical value indicating the position of the subcarrier, and N is a number obtained by adding the number of pilot signals “1” to the number of subcarriers between pilot signals. For these N and k, for example, in the case of FIG. (B), the number of pilot signals n = 4 (lines), and there are 13 subcarriers between the pilot signals as described above. N = 13 + 1 = 14. In addition, in {} in the expression (2), it is a level difference between adjacent pilot signals (except between P1 and P4), and k is for each of 13 subcarriers between these pilot signals. It is a numerical value of 1, 2, 3,. In this case, the first subcarrier on the (Pi-1) side is set to “1”.
[0014]
In this way, g2 (k) means that when a level change occurs between adjacent pilot signals, it is estimated that the level of each subcarrier in the meantime changes linearly. . Based on the results of the above formulas (1) and (2), a correction threshold g (k) is obtained by the following formula (3), and a preset threshold is corrected to this threshold.
g (k) = norm * {(N / 2) * g1 + N * g2 (k)} / {N + (N / 2)} (3)
In the above equation, norm is a preset threshold value (basic threshold value), and N and k are the same as in the above equations (1) and (2). As described above, the correction threshold value g (k) is obtained by weighting and adding elements of both the level average and the linear interpolation of the conventional pilot signal.
[0015]
The correction is performed by the threshold correction unit 5 for Ich signal and the threshold correction unit 6 for Qch signal, and the correction value is set for the determination unit 10. The state of this threshold correction is indicated by reference numeral 30 in FIG. That is, the original threshold value 26 is corrected to the new threshold value 30. As a result, the portion (31) that causes an error in data determination at the original threshold 26 does not cause an error due to the setting of the correction threshold 30. In FIG. 2C, the influence of thermal noise (FIG. D) is ignored. However, due to this thermal noise, a correction threshold value 30 may be exceeded in some data signals, resulting in an error in data determination. For this reason, it is necessary to correct the data signal in the same manner as the threshold value correction. The data correction unit 9 corrects the data signal.
[0016]
Hereinafter, the data correction unit 9 will be described with reference to FIG. In the figure, the block diagram is omitted for the Qch side having the same configuration as the Ich side. In the figure, 9a is an Ich-side data correction unit, 9b is a Qch-side data correction unit, and the Ich-side data correction unit 9a has a data signal from the delay unit 7, a pilot signal from the Fourier transform processing unit 2, and an Ich The correction threshold value data is input from the signal threshold correction unit 5, the data correction unit 9 b on the Qch side receives the data signal from the delay unit 8, the pilot signal from the Fourier transform processing unit 2, and the threshold correction unit 6 for the Qch signal. Further, correction threshold data is input. In the data correction unit 9a on the Ich side, the first phase error estimation unit 91 uses two sets of pilot signals (P1 and P2, P2 and P3, P3 and P4, P1 and P4, P1 and P3, P2 and P4). The phase error of the data signal due to the sampling timing error and the initial phase error is estimated from the level difference in each set. This estimation is performed for all pairs in the parentheses. Here, the sampling timing is an error of the sampling timing in an A / D conversion unit (not shown) when obtaining the digital Ich signal input to the synchronization unit 1, and the initial phase error is a frequency conversion of the radio frequency signal. Is a phase error.
[0017]
The first phase error averaging unit 92 averages all estimated values estimated by the phase error estimating unit 91. This average value is taken as a first correction value. The data correction unit 93 performs correction for rotating the phase with respect to the entire subcarrier (data signal) based on the first correction value from the first phase error averaging unit 92. The temporary data determination unit 94 temporarily performs data determination in the same manner as the determination unit 10, and uses the correction threshold value from the threshold value correction unit 5 to temporarily determine data for the data signal from the data correction unit 93. The second phase error estimation unit 95 estimates the data phase error from the data signal corrected by the data correction unit 93 and the data provisionally determined by the data provisional determination unit 94. In this case, the data of the data correcting unit 93 through a delay unit 96 for timings for alignment. In addition, the above estimation is performed for all subcarriers with two subcarriers (−1 and +1, −2 and +2, −3 and +3...) Having a target relationship with respect to the center frequency.
[0018]
The second phase error averaging unit 97 averages all the estimated values in the second phase error estimating unit 95. This average value is taken as a second correction value. However, in this average, the phase error estimated by the second phase error estimation unit 95 is much larger than a predetermined value compared to the phase difference between the correction threshold from the threshold correction unit 5 and the reference threshold. Exclude and average. The data re-correction unit 98 uses the second correction value in the second phase error averaging unit 97 to phase the entire subcarrier (data signal) via the delay unit 96 and the delay unit 97 for timing adjustment. The data is corrected by rotating. The above is the data correction on the Ich side, but the same thing is also corrected for the data on the Qch side (data correction unit 9b).
In the above description, 16QAM is used. However, the present invention can be applied to other multilevel QAM.
[0019]
【The invention's effect】
As described above, according to the present invention, a pilot signal is used as a threshold for data determination and a data signal in decoding of an orthogonal frequency division multiplexing (OFDM) radio communication signal using a multi-level QAM modulation scheme. Since both the averaging method and the linear interpolation method are corrected in combination, the influence of phase noise such as fading or thermal noise such as circuit elements can be avoided. As a result, occurrence of errors in data determination is prevented, and the performance of the QAM demodulator can be improved.
[Brief description of the drawings]
FIG. 1 is a main part block diagram showing an embodiment of a QAM demodulator according to the present invention.
FIG. 2 is an explanatory diagram related to FIG. 1;
3 is a principal block diagram showing an example of a data correction unit 9 in FIG. 1; FIG.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Synchronization part 2 Fourier-transform processing part 3, 4 Estimation part 5, 6 Threshold correction part 7, 8 Delay part 9 Data correction part
9a Ich side data correction part
9b Qch side data correction part
10 Judgment part
91, 95 Phase error estimator
92, 97 Phase error averaging section
93 Data correction section
94 Data provisional judgment section
96, 99 Delay part
98 Data re-correction part

Claims (2)

In decoding of a radio communication signal of an orthogonal frequency division multiplexing (OFDM) using a modulation scheme of multilevel QAM (Quadrature Amplitude Modulation), a baseband I channel obtained by frequency conversion of the received radio frequency signal and the Q A Fourier transform processing unit that converts a time domain signal of each channel into a frequency domain signal and outputs a pilot signal and a data signal for each channel; and the OFDM based on the pilot signal from the Fourier transform processing unit Estimating means for estimating fluctuations in the amplitude and phase of each of the data signals of the I channel and Q channel to be formed based on an average level for the plurality of pilot signals and linear interpolation for the plurality of pilot signals; , preset based on the estimation result by the estimating means Data for correcting the data for the data signal from the Fourier transform processing unit based on the threshold value correcting unit for correcting the reference threshold for data determination, the pilot signal from the Fourier transform processing unit, and the correction threshold value by the threshold correcting unit A QAM decoding apparatus comprising: a correction unit; and a determination unit that performs data determination on a data signal from the data correction unit using a threshold value from the threshold value correction unit and outputs decoded data. The data correction unit for I channel or Q channel has a plurality of pilot signals as two sets, a first phase error estimation unit that estimates a phase error of the data signal for each set, and the phase A phase is rotated with respect to all subcarriers constituting the data signal based on a first phase error average unit that averages the estimated values estimated by the error estimation unit and an average value from the first phase error average unit. A data correction unit that performs correction, a data temporary determination unit that performs temporary data determination using a correction threshold value from the threshold value correction unit for the data signal from the data correction unit, and a data signal that has been corrected by the data correction unit A second phase error estimator that estimates a data phase error from the data provisionally determined by the data tentative determination unit, and a second level that averages all the estimated values in the second phase error estimator. An error averaging unit and a data re-correction unit that rotates a phase with respect to all subcarriers constituting a data signal based on an average value from the second phase error averaging unit. Item 4. The QAM decoding device according to Item 1.

Images