KR0171013B1 - Improved symbol synchronization of multi-level quadrature amplitude modulation - Google Patents

Abstract

This symbol synchronization extraction method is an improved method to extract symbol synchronization more simply by reducing the calculation amount without degrading the synchronous extraction performance in the multi-stage quadrature amplitude modulator. This method takes an absolute value when data is input. step; Determining whether a synchronization value for the input data has been extracted; Outputting a synchronization value if it has been extracted; If the synchronization value has not been extracted, obtaining a maximum value and a minimum value during a symbol period; Determining whether the maximum and minimum values obtained based on the minimum and maximum values obtained during the symbol period are valid regions from which the synchronization value can be extracted; Judging whether the position value of the obtained maximum value is the same as the position value of the maximum value of the previous symbol when the result corresponds to a valid region; If the result of the determination is the same, increasing the count value; And comparing the incremented count value with a predetermined threshold value and extracting a position value of the maximum value obtained as a synchronization value of a symbol if the increased count value is greater than the predetermined threshold value.

Description

Symbol Synchronization Method for Improved Multistage Orthogonal Amplitude Modulation

1 is a waveform diagram of an inphase component of a 16 square amplitude modulated (QAM) signal passing through a nonlinear filter.

2 is a flowchart of an early-late synchronous extraction method using a gradient.

3 is a flowchart of a symbol synchronization extraction method of an improved multi-stage QAM according to the present invention.

4 is a diagram showing an effective sync search area used in the present invention.

5 is a performance result of the present invention.

6 is a performance result of the conventional synchronous extraction method.

The present invention relates to multi-level Quadrature Amplitude Modulation (QAM), and more particularly to a method for extracting symbol synchronization of a multi-stage QAM of 16 or more stages.

Multi-stage QAMs with 16 or more stages are more difficult to extract symbol synchronization than QPSK (Quadrature Phase Shift Keyed, QPSK) because they have high power and spectral efficiency. In particular, if the waveform is generated by a raised-cosine filter, the synchronization problem becomes more difficult. Therefore, in multi-stage QAMs, waveforms are generated as non-linear filters, and early-late synchronous extraction using gradients has been used, which leads to a large amount of computational complexity and computational complexity. As a result, when implementing a high-speed transmission / reception system, a large burden is placed on real-time processing. That is, the conventional symbol synchronization extractor of the multi-stage QAM finds the positive peak and the negative peak of the waveform generated as shown in FIG. 1 through the nonlinear filter using the slope between adjacent samples, The found peak position was used as a symbol synchronous.

2 is a flowchart of an early-late sync extraction method using a gradient. As shown here, after sampling 210, the positive weights are calculated according to the positive gradients (steps 202, 204, 207, and 208). Also, the negative weights are determined according to the negative gradients. The weight is calculated. (Steps 203, 206, 210, and 211) The process of calculating the positive weight includes calculating a positive slope (step 202), +2, +3, -2,-to the slope value G. The slope is calculated by adding 3, +23, and -23, respectively, and comparing whether each value is greater than '0' (step 204). Similarly, the negative slope is +2, +3,-to the slope value (-G). Compute the slope by adding 2, -3, +23, and -23, respectively (step 203), and compare whether each value is greater than '0' (step 206). If the inclination value is smaller than '0', the PLL returns to sampling (step 201) via free oscillation (step 205), and the slope values are in mode '0'. If larger, the positive peak confirmed (step 207) and negative peak confirmed (step 210) are used to calculate the respective weights (step 208) and (step 211). Pscore) is Pscore = G2 + G3 + G-2) + (G-3), and Nscore is Nscore = (-G2) + (-G3) + (-G-2) + (-G-) 3) compare the positive and negative weights (PscoreNscore?) (Step 209), and determine the dominant positive peak if the positive weight is large (step 212), and the dominant negative if the negative weight is large. The peak is determined. (Step 213). The PLL synchronization according to the weight difference is extracted (step 214), and the process is repeated again from the sampling (step 201).

At each sample time, gradient calculations are performed six times in steps 202 and 203, respectively, in order to obtain positive and negative peaks according to the flow chart shown in FIG. 2. The weight and negative peak values are calculated according to the calculated positive peak values. Since the synchronization is extracted by comparing the weights according to the 209th step, when the sampling ratio is 8, there are too many problems because there are 96 calculations and comparisons during the symbol period.

Accordingly, an object of the present invention is to provide a symbol synchronization extraction method of an improved multi-stage quadrature amplitude modulator which can extract symbol synchronization more simply by reducing the calculation amount without degrading the synchronous extraction performance in the multi-stage quadrature amplitude modulator.

According to an aspect of the present invention, there is provided a symbol synchronization extraction method comprising: a symbol synchronization extraction method of a multi-stage quadrature amplitude modulator comprising: taking an absolute value when data is input; Determining whether a synchronization value for the input data has been extracted; Outputting a synchronization value if it has been extracted; If the synchronization value has not been extracted, obtaining a maximum value and a minimum value during a symbol period; Determining whether the maximum and minimum values obtained based on the minimum and maximum values obtained during the symbol period are valid regions from which the synchronization value can be extracted; Judging whether the position value of the obtained maximum value is the same as the position value of the maximum value of the previous symbol when the result corresponds to a valid region; If the result of the determination is the same, increasing the count value; And comparing the incremented count value with a predetermined threshold value and extracting a position value of the maximum value obtained as a synchronization value of a symbol if the increased count value is greater than the predetermined threshold value.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The signal of one channel of 16QAM passing through the nonlinear filter shown in FIG. 1 may be displayed at the same frequency as the peak value probabilistic about the magnitude zero. Areas with more than 16 multi-stage QAMs that are more difficult to detect than the QPSK method are those where the size increases sequentially to -3, -1, +1, +3, and vice versa With the portions whose size does not change, the conventional synchronous extraction method obtains the inclination of each sampling point to determine whether it is valid for such portions, but the determination is made in the present invention. Except in the effective judgment area, the valid synchronization extraction area is secured by using the largest and smallest values during the symbol period regardless of the sign, and synchronization is performed using the jumping method rather than the early-rate method. Extract.

3 is a flowchart of a symbol synchronization extraction method according to the present invention.

As shown in the flow chart shown in Fig. 3, before the symbol synchronization extraction, the portion where the peak value is smaller than the surrounding values, such as +3, +1, +3, -3, -1, -3 Exclude.

These valid and invalid areas are as shown in FIG.

To do this, if input data (INPUT) is applied, the absolute value is taken in step 301. In operation 302, it is checked whether the position value i for the input data corresponds to the symbol synchronization value SET_POSITION. As a result of the check, if the symbol synchronization is extracted, the process proceeds to the extracted step 303 and outputs an input signal corresponding to the extracted symbol synchronization SET_POSITION. However, if the check result of step 302 does not correspond to the symbol synchronization, the process proceeds to step 304 and it is checked whether the position value i for the input data for determining the symbol synchronization period is greater than or equal to 8 (8 times oversampled). Example of the case). If i is less than 8, the flow proceeds to step 305. In step 305, the data input in step 301 is compared with a maximum value MAX. As a result of the comparison, if the input data is smaller than the maximum value, the flow proceeds to step 306. In step 306, the data input in step 301 is compared with the minimum value MIN. As a result of the comparison, if the input data is larger than the minimum value, the process returns to step 301. However, if the data input in step 305 is greater than the maximum value, the process proceeds to step 307 and stores the input data as the maximum value, sets the i value for the currently input data as the position value MAX_INDEX of the maximum value, and step 301. Proceeds to step. In addition, if the data input in step 306 is smaller than the minimum value, the process proceeds to step 308 and stores the input data as the minimum value, sets the i value at this time as the position value MIN_INDEX of the minimum value, and then proceeds to step 301. Is returned. As a result of the determination in step 304, if the i value is greater than 8, it is determined that the symbol synchronization period is finished. In step 309, the i value is set to 0, and then in step 310, the minimum value set in step 308 is determined. MIN) to 2.0. As a result of the comparison, if the minimum value is not less than 2.0, it is determined as an invalid area, and the process proceeds to step 311, where the count value is set to 0 (COUNTER = 0), and the process returns to step 301. However, if it is determined in step 310 that the minimum value is less than 2.0, the process proceeds to step 312 and compares whether the value obtained by subtracting the minimum value MIN from the maximum value MAX is greater than 1.0. As a result of the comparison, if it is not large, it is also determined as an invalid area, and the flow proceeds to step 311 where the count value is reset to 0 and the flow returns to step 301. Here, in step 310 and step 312, the threshold value 2.0 for the minimum value MIN and the threshold value 1.0 for the difference between the maximum value and the minimum value are 16 QAM. The voltage values of the waveforms are ± 3 V and ± 1 V. It only has a value. In this case, MAX and MIN are used to select the value of the effective area of the symbol synchronous. The MIN value should not exceed 2 of the square size of the waveform and the difference between the maximum value and the minimum value should be greater than 1. The values that distinguish this valid decision are defined as their respective thresholds. As a result of the determination in step 312, if the difference between the maximum value and the minimum value is greater than 1.0, the process proceeds to step 313 and the value stored as the maximum position value MAX-INDEX and the position of the peak of the previous symbol MAX_b_INDEX in step 307. Compare with That is, it is determined whether the position of the current MAX value is the same as the position of MAX previously determined. Therefore, only the region of the waveform of FIG. 4 whose minimum value is 2 or less and the difference from the maximum value is 1 or more is determined as an effective region. If the comparison result is small, the process proceeds to step 311 and resets to 0 as the count value, and returns to step 301. However, if the determination result of step 313 is the same as the position value of the extracted peak and the peak position value of the previous symbol, the process proceeds to step 314 to increase the count value. In operation 315, it is checked whether the increased count value is greater than a predetermined threshold value. The threshold value THRESHOLD is a section for accumulating the input signal until the symbol synchronization is extracted, and sets a typical 500 symbol section using an oscillator having a precision of 10 ^{-6 in} the system. Here, the optimum value of the predetermined threshold value can be adjusted because there is some difference according to the signal-to-noise ratio, and the convergence and divergence rate of the synchronization extraction can be adjusted by varying the threshold value. As a result of the check in step 315, if the incremented counter value is not greater than a predetermined threshold value, the process returns to step 301.

In step 302, the position value i is increased for each input data (i ++) to compare whether the symbol synchronization value (SET_POSITION) is set as a symbol representative value. In step 316, the number of samples of which a sample is a suitable symbol representative value is stored in a symbol synchronization value SET_POSITION variable. That is, the SET_POSITION value has a value from '0' to '7'. In step 302, an input signal is output when i is equal to the SET_POSITION value, and one sample value is output from the eight sample values that are input. It's possible.

However, if the increased count value is larger than the predetermined threshold value, that is, whether the average of 500 symbols is averaged is determined, the process proceeds to step 316 and sets the maximum position value MAX_INDEX to the extracted synchronization value and resets the count. After setting, the process returns to step 301.

5 and 6 artificially synchronize the 16QAM signal having an Eb / No = 15dB and a sampling rate of 8 with a transmission rate of 20Ksample / sec and a reception rate of 20.004sample / sec using the method according to the present invention and the conventional gradient method. This is the constellation of the results of experiments with 20000 symbols. As shown in FIG. 5 and FIG. 6, the signal points extracted by the symbol synchronization extraction method according to the present invention of FIG. 5 have improved performance as indicated by the small points of the signal points compared to the conventional method of FIG. Can be.

As described above, the symbol synchronization extraction method of the multi-stage QAM according to the present invention divides the synchronization extraction area for the symbol into a valid area and an invalid area, and obtains and uses only the maximum and minimum values for each sample during the symbol period. As a result, the amount of computation is reduced compared to the past, and thus the symbol synchronization of the multi-stage QAM can be extracted more simply.

Claims (3)

CLAIMS 1. A symbol synchronization extraction method of a multi-stage quadrature amplitude modulator, comprising: taking an absolute value when data is input; Determining whether a synchronization value for the input data has been extracted; Outputting a synchronization value if it has been extracted; If the synchronization value has not been extracted, comparing the input data with the current maximum value and the minimum value and updating the maximum value and the minimum value during the corresponding symbol period by the input data when the value is larger than the maximum value or smaller than the minimum value; Comparing whether the minimum value obtained during the symbol period is less than 2 and the difference between the maximum value and the minimum value is greater than 1 to determine whether the maximum value and the minimum value are valid regions from which a synchronization value can be extracted; If the minimum value is less than 2 and the difference between the maximum value and the minimum value is greater than 1, determining whether the obtained maximum value is the same as the maximum value of the previous symbol; Increasing a count value if the position value of the maximum value is the same; When the incremented count value is greater than a predetermined threshold value, a predetermined threshold value is compared with respect to a section in which an increased count value and an input signal are accumulated until extracting a symbol synchronous symbol. The symbol synchronization extraction method of the improved multi-stage quadrature amplitude modulation, characterized in that it comprises the step of extracting the synchronization value of. The method of claim 1, wherein the updating of the maximum value and the minimum value during the corresponding symbol period by the input data comprises: comparing the input data value with the maximum value and the minimum value and storing the maximum value if the input data value is larger than the maximum value. The symbol synchronization extraction method of the improved multi-stage quadrature amplitude modulation, characterized in that if the value is smaller than the minimum value, the minimum value is stored and the position value of the stored maximum value and the minimum value is stored as the current position value (i) of the input data. 2. The method of claim 1, wherein a predetermined threshold value compared with the count value is varied according to a signal-to-noise ratio to adjust the convergence and divergence rate of the synchronization extraction.