JP2010098395A - Method and apparatus for equalization processing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the accuracy in distortion correction of received teletext signals. <P>SOLUTION: In the equalization processing apparatus 100, a first equalizer 130 equalizes CRI signals and FRC signals included in the teletext signals, and a second equalizer 140 equalizes text data signals included in the teletext signals. The first equalizer 130 operates at a sampling rate higher than a sampling rate corresponding to the symbol rate of the teletext signals, and the second equalizer 140 operates at the sampling rate corresponding to the symbol rate of the teletext signals. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for equalizing a teletext signal.

  Character multiplex broadcasting (teletext) that broadcasts information such as characters and simple figures using a gap between television signals is performed. In teletext, a teletext signal is transmitted during a vertical blanking period (VBI) in a baseband composite signal.

  The upper part of FIG. 6 shows the structure of the teletext signal. The teletext signal includes a CRI (Clock Run In) signal, an FRC (Framing Code) signal, and a text data signal.

  The CRI signal at the top is a reference signal that indicates the cycle of the text data signal, and the FRC signal that follows is a reference signal that indicates the starting point of the text data signal.

  The text data signal following the FRC signal is obtained by NRZ encoding (NRZ: Not Return to Zero) serial digital data (referred to as text data) representing CRI characters or simple figures.

  Of the above three signals of the teletext signal, the CRI signal and the FRC signal have a predetermined signal value arrangement pattern, and the signal value arrangement pattern of the text data signal differs depending on the contents of the text data.

  With such a configuration of the teletext signal, it is possible to detect the CRI signal to know the cycle of the text data signal and to detect the FRC signal to know the starting point of the text data signal.

  The lower part of FIG. 6 shows the detection timing of the teletext signal on the receiving side. First, the CRI signal is detected, and the CRI detection signal becomes active during the CRI signal period. The FRC signal is detected after the CRI signal period, and the FRC detection signal becomes active in the FRC signal period. The text data detection signal becomes active at the end of the FRC signal period, and the text data signal is detected.

  Since the transmission of the teletext signal is baseband transmission, the waveform of the signal is likely to be distorted by a ghost or the like. For this reason, on the receiving side, waveform distortion is corrected using an equalizer.

  The equalizer usually has an equalization filter (transversal filter) and a coefficient correction circuit for correcting the coefficient of the equalization filter. The equalization filter performs waveform distortion correction on the teletext signal, and the coefficient correction circuit compares the processing result of the equalization filter with the expected value, and feeds back the comparison result to the equalization filter. Specifically, the feedback is a process of correcting the coefficient of the equalization filter so that the processing result approaches the expected value based on the error obtained by the comparison.

  Although not limited to the equalization processing of the teletext signal, various methods have been proposed for improving the performance of the equalizer.

  For example, the technique disclosed in Patent Document 1 is different in the training period and the tracking period for a reception burst having a training period in which known training data is inserted and a tracking period in which unknown tracking data is inserted. Is used to update the coefficient of the equalization filter. According to this method, it is said that the convergence to the pull-in state can be increased during the training period, and the amount of calculation can be reduced during the tracking period.

  Non-Patent Document 1 discloses a technique for effectively removing a near ghost and a far ghost from a received signal. This will be described with reference to FIGS.

  FIG. 7 is FIG. 7A of Non-Patent Document 1, and shows an equalizer to which the technique of the document is applied. This equalizer includes two equalization filters (T / 2 Filter and T Filter) having different tap intervals. The T / 2 Filter has a tap interval of T / 2 and is effective for removing near ghosts. The T Filter has a tap interval of T and is effective for removing distant ghosts. The equalizer shown in FIG. 7 removes both the near ghost and the far ghost in the received signal by further processing the signal processed by the T / 2 filter with the T filter. “T” is the reciprocal of the symbol rate.

  FIG. 8 is FIG. 7B of Non-Patent Document 1, and shows another equalizer disclosed in the document. This equalizer is substantially provided with two equalizers, and is an equipping unit configured by a T / 2 filter and an equivalent unit configured by a T filter in cascade connection. When the distance is long, it is more effective than the equalizer shown in FIG.

  Further, Patent Document 2 discloses a method that pays attention to the fact that the frequency spectrum of the received signal varies greatly depending on the data contents.

  When the data content of the received signal includes the bit values “1” and “0” almost evenly, the frequency spectrum of the received signal exhibits a substantially uniform spread, while the data content is “ When “1” or “0” continues, the frequency spectrum of the received signal is biased. When a signal having a biased frequency spectrum continues to some extent, the equalizer performs coefficient correction of the equalizing filter based on the signal having the biased frequency spectrum. In this case, when the data content returns to a random state, a correction result based on a signal having a biased frequency spectrum is dragged, and in some cases, erroneous correction is performed.

  The technique disclosed in Patent Document 2 demodulates a reception signal that has been equalized by an equalizer into bit data, and whether or not “1” or “0” continues for this demodulated data. Confirm. If “1” or “0” continues as a result of the confirmation, the optimization correction operation to the equalizer (which is considered to correspond to the above-described coefficient correction of the equalization filter) is stopped.

Any of the above methods can be applied to the equalization processing of the teletext signal, and it is considered that the effect of the method can be obtained.
Japanese Patent Laid-Open No. 6-188785 JP-A-6-326634 Yu Sakurai and Kazuo Ozeki, “A Study on the Tap Interval in an Automatic Equalizer for Digital Television Receivers”, IEICE Transactions, March 2004, Vol. J87-A No. 3, pp. 424-427

  Regarding the performance of the equalizer, it is also desired to improve the accuracy of distortion correction. In order to improve the correction accuracy, a method of increasing the sampling rate of the equalizer can be considered. Note that increasing the sampling rate of the equalizer means increasing the frequency of coefficient correction of the equalization filter included in the equalizer.

  By the way, since the coefficient correction of the equalization filter is performed based on the comparison result between the signal value processed by the equalization filter and the expected value, in order to increase the frequency of coefficient correction, the number of expected values is set. Need to increase. However, in the case of a teletext signal, since the content of the text data signal included in the teletext signal is unknown, a temporary expected value is used. If the temporary expected value is used, the correction accuracy cannot be expected to increase even if the sampling rate is increased. Further, when the provisional expected value is different from the actual data content, there is a possibility that the correction accuracy is lowered due to the increase.

  One aspect of the present invention is an equalization processing method for a received teletext signal. In this method, equalization processing is performed on the CRI signal and FRC signal included in the teletext signal and the text data signal at different sampling rates.

  It should be noted that the above-described equalization processing method is replaced with an equalization processing apparatus or system, and is effective as an aspect of the present invention.

  According to the technique according to the present invention, it is possible to improve the accuracy of distortion correction of a received teletext signal.

Prior to a specific embodiment of the present invention, first, the principle of the present invention will be described.
The CRI signal and the FRC signal (hereinafter, the CRI signal and the FRC signal are collectively referred to as a CRI & FRC signal) included in the teletext signal and the text data signal have different properties.

  The CRI & FRC signal is a fixed known pattern, and an accurate expected value can be prepared at the time of equalization processing. On the other hand, since the content of the text data signal is unknown, only a provisional expected value can be prepared.

  Further, the CRI & FRC signal and the text data signal have different frequency characteristics. This will be described with reference to FIGS.

  FIG. 1 shows an example of the frequency characteristic of the CRI & FRC signal, and FIG. 2 shows an example of the frequency characteristic of the text data signal. In the figure, the horizontal axis represents the frequency normalized by the symbol rate, and the vertical axis represents the signal amplitude.

  As shown in FIG. 1, in the case of the CRI & FRC signal, the frequency band is limited. This is because the CRI & FRC signal has a fixed pattern with little variation.

  On the other hand, as shown in FIG. 2, in the case of a text data signal, the frequency band is wider than that in the case of a CRI & FRC signal. This is because the contents of the text data signal include “1” and “0” at random. In addition, when the data content is “1” or “0” continuously, the frequency band of the text data signal is biased.

  The present invention has been made paying attention to the above-described different properties of the CRI & FRC signal and the text data signal.

  FIG. 3 is a schematic diagram of an equalization processing apparatus for explaining the technique according to the present invention. The equalization processing apparatus 10 includes an equalization filter 20 and a coefficient correction circuit 30 that corrects a coefficient of the equalization filter 20. The coefficient correction circuit 30 further includes an error detection unit 32 and a coefficient calculation unit 34.

  The error detection unit 32 obtains an error by comparing the teletext signal processed by the equalization filter 20 with the expected value. The coefficient calculation unit 34 calculates a new coefficient of the equalization filter 20 based on the error obtained by the error detection unit 32 and sets the equalization filter 20 so that the processing result by the equalization filter 20 approaches the expected value. To do.

  The coefficient correction circuit 30 corrects the coefficient of the equalization filter 20 with different correction frequencies for the CRI & FRC signal included in the teletext signal and the text data signal. That is, the coefficient correction circuit 30 operates at different sampling rates for the CRI & FRC signal and the text data signal.

  As described above, the CRI & FRC signal is a fixed known pattern, and the frequency band is limited. Therefore, when the CRI & FRC signal is equalized, the number of expected values can be increased by interpolation calculation of expected values. Therefore, the coefficient correction circuit 30 can correct the coefficient of the equalization filter 20 at a sampling rate higher than, for example, twice the sampling rate corresponding to the symbol rate of the teletext signal. Thereby, the accuracy of distortion correction of the CRI & FRC signal can be improved.

  On the other hand, for text data signals for which only temporary expected values can be obtained, erroneous correction is performed by performing coefficient correction at a sampling rate different from that of the CRI & FRC signal, for example, a sampling rate corresponding to the symbol rate of the teletext signal. Can be avoided.

  As a result, it is possible to improve the distortion correction accuracy of the entire teletext signal.

  FIG. 4 is another schematic diagram of an equalization processing apparatus for explaining the technique according to the present invention. The equalization processing apparatus 50 includes a first equalizer 60 and a second equalizer 70, and performs equalization processing of the CRI & FRC signal and the text data signal by the two equalizers. For easy understanding, in FIG. 4, functional blocks for detecting the CRI & FRC signal and the text data signal from the teletext signal are omitted.

  The first equalizer 60 is responsible for equalization processing of the CRI & FRC signal, and outputs the text data signal to the second equalizer 70 at the subsequent stage without processing. As illustrated, the first equalizer 60 includes an equalization filter 62 and a coefficient correction circuit 64 that corrects the coefficient of the equalization filter 62. The coefficient correction circuit 64 further includes an error detection unit 66 and a coefficient calculation unit 68. The error detection unit 66 obtains an error by comparing the CRI & FRC signal processed by the equalization filter 62 with the expected value (expected value of the CRI & FRC signal). The coefficient calculation unit 68 calculates a new coefficient of the equalization filter 62 based on the error obtained by the error detection unit 66 and sets the equalization filter 62 so that the processing result by the equalization filter 62 approaches the expected value. To do.

  The second equalizer 70 is responsible for equalizing the text data signal and does not process the CRI & FRC signal. As illustrated, the second equalizer 70 includes an equalization filter 72 and a coefficient correction circuit 74 that corrects the coefficient of the equalization filter 72. The coefficient correction circuit 74 further includes an error detection unit 76 and a coefficient calculation unit 78. The error detection unit 76 obtains an error by comparing the text data signal processed by the equalization filter 72 with an expected value (expected value of the text data signal). The coefficient calculation unit 78 calculates a new coefficient of the equalization filter 72 based on the error obtained by the error detection unit 76 and sets the equalization filter 72 so that the processing result by the equalization filter 72 approaches the expected value. To do.

  The coefficient correction circuit 64 in the first equalizer 60 and the coefficient correction circuit 74 in the second equalizer 70 operate at different sampling rates. Specifically, for example, the coefficient correction circuit 64 operates at a sampling rate higher than, for example, twice the sampling rate corresponding to the symbol rate of the teletext signal, and the coefficient correction circuit 74 has a sampling different from that of the coefficient correction circuit 64. It operates at a sampling rate corresponding to the rate, for example the symbol rate of the teletext signal.

  That is, the equalization processing device 50 also corrects the coefficients of the equalization filter with different correction frequencies for the CRI & FRC signal and the text data signal. Therefore, the same effect as that of the equalization processing apparatus 10 shown in the schematic diagram of FIG. 3 can be obtained.

  Further, since the equalization processing device 50 uses different equalizers for the equalization processing of the CRI & FRC signal and the equalization processing of the text data signal, the frequency characteristics of the CRI & FRC signal and the text data signal are mutually equalized. The influence can be avoided, and the accuracy of equalization processing of the entire teletext signal can be further increased.

Based on the above, an equalization processing apparatus that embodies the principles of the present invention will be described.
FIG. 5 shows an equalization processing apparatus 100 according to the embodiment of the present invention. The equalization processing apparatus 100 performs equalization processing on the received teletext signal, and includes a CRI detection unit 112, an FRC detection unit 114, a text data detection unit 116, an OR gate 120, a first gate, Equalizer 130, second equalizer 140, binarization unit 150, and variance calculation unit 160.

  The CRI detection unit 112 detects the CRI signal with respect to the teletext signal, and when detecting the CRI signal, activates the CRI detection signal S1. That is, the CRI detection signal S1 is active during the CRI signal period and inactive during other periods.

  The FRC detection unit 114 detects the FRC signal with respect to the teletext signal, and when detecting the FRC signal, activates the FRC detection signal S2. That is, the FRC detection signal S2 is active during the FRC signal period and inactive during other periods.

  The text data detection unit 116 detects the text data signal with respect to the teletext signal, and activates the text data detection signal S3 when detecting the text data signal. That is, the text data detection signal S3 is active during the text data signal period and inactive during other periods.

  The CRI detection signal S1 from the CRI detection unit 112 and the FRC detection signal S2 from the FRC detection unit 114 are input to the OR gate 120.

  The OR gate 120 outputs a logical sum S4 of the CRI detection signal S1 and the FRC detection signal S2 to the first equalizer 130. This logical sum S4 functions as a signal for controlling whether or not the first equalizer 130 operates, and is hereinafter referred to as a control signal.

  When the control signal S4 is active, the first equalizer 130 performs an equalization process on the teletext signal (in this case, the CRI signal or the FRC signal), and sends it to the second equalizer 140. Output. When the control signal S4 is inactive, the teletext signal (in this case, the text data signal) is output to the second equalizer 140 as it is.

  That is, the first equalizer 130 performs equalization processing only on the CRI & FRC signal of the teletext signal. The detailed configuration of the first equalizer 130 will be described later.

  When the text data detection signal S3 is active, the second equalizer 140 performs an equalization process on the signal from the first equalizer 130 and outputs the signal to the binarization unit 150 for control. When the signal S6 is inactive, the signal from the first equalizer 130 is output to the binarization unit 150 as it is. That is, the second equalizer 140 performs an equalization process only on the text data signal. The detailed configuration of the second equalizer 140 will be described later.

  The binarization unit 150 binarizes the signal from the second equalizer 140 to obtain binary data, and outputs it as a processing result of the equalization processing apparatus 100. This binary data is also output to the variance calculation unit 160.

  The variance calculation unit 160 obtains the variance for the binary data obtained by the binarization unit 150 and activates / deactivates the variance signal S5 according to the obtained variance. Specifically, when the obtained variance is equal to or greater than a predetermined threshold, the variance signal S5 is activated, and when the variance is less than the threshold, the variance signal S5 is deactivated.

The variance signal S5 is input to a coefficient correction circuit 144 described later of the second equalizer 140.
The first equalizer 130 and the second equalizer 140 will be described in detail.

  The first equalizer 130 includes a CRI & FRC equalization filter 132 and a coefficient correction circuit 134. The coefficient correction circuit 134 further includes an error detection unit 136 and a coefficient calculation unit 138. The error detection unit 136 obtains an error by comparing the signal processed by the equalization filter 132 (here, the CRI signal or the FRC signal) with the expected value (the expected value of the CRI signal or the FRC signal). The coefficient calculation unit 138 calculates a new coefficient of the equalization filter 1322 and sets it in the equalization filter 1322 so that the processing result by the equalization filter 132 approaches the expected value based on the error obtained by the error detection unit 136. To do.

  In the first equalizer 130, the coefficient correction circuit 134 operates at a sampling rate that is twice the sampling rate corresponding to the symbol rate of the teletext signal.

  The second equalizer 140 includes an equalization filter 142 and a coefficient correction circuit 144 that corrects the coefficient of the equalization filter 142. The coefficient correction circuit 144 further includes an error detection unit 146 and a coefficient calculation unit 148. The error detection unit 146 obtains an error by comparing the signal processed here by the equalization filter 142 (here, the text data signal) and the expected value (expected value of the text data signal). The coefficient calculation unit 148 calculates a new coefficient of the equalization filter 142 based on the error obtained by the error detection unit 146 so that the processing result by the equalization filter 142 approaches an expected value, and sets the new coefficient in the equalization filter 142. To do. The coefficient correction circuit 144 of the second equalizer 140 is controlled by the dispersion signal S5 from the dispersion calculation unit 160, and operates only when the dispersion signal S5 is active.

  That is, the second equalizer 140 performs an equalization process only on the text data signal. In the equalization process, if the variance of the bit data corresponding to the text data signal is equal to or greater than a predetermined threshold, the coefficient of the text data equalization filter 142 is corrected. If the variance is less than the threshold, Does not correct the coefficient of the text data equalization filter 142.

  The equalization processing apparatus 100 is an embodiment corresponding to the equalization processing apparatus 50 shown in FIG. 2, and performs equalization processing of the CRI & FRC signal and the text data signal by two equalizers having different sampling rates. . According to the equalization processing apparatus 100, the effects described in the description of the equalization processing apparatus 50 can be obtained.

  Further, in the equalization processing apparatus 100, the second equalizer 140 performs the text data equalization filter 142 only when the distribution of the bit data corresponding to the text data signal is large in the equalization processing of the text data signal. Correct the coefficient.

  When the data content of the text data signal includes “1” and “0” at random, the variance increases. Therefore, when the coefficient of the text data equalization filter 142 is corrected only when the variance is large, the frequency characteristic of the text data signal is biased, such as a long period of continuous “1” or “0”. A possible erroneous correction can be avoided, and the accuracy of distortion correction can be further increased.

  The present invention has been described above based on the embodiment. The embodiment is an exemplification, and various modifications, increases / decreases, and combinations may be made to the above-described embodiments without departing from the gist of the present invention. It will be understood by those skilled in the art that modifications in which these changes, increases / decreases, and combinations are also within the scope of the present invention.

It is a figure which shows the frequency characteristic of the CRI signal and FRC signal which are contained in a teletext signal. It is a figure which shows the frequency characteristic of the text data signal contained in a teletext signal. It is a schematic diagram of the equalization processing apparatus for demonstrating the technique concerning this invention. It is another schematic diagram of the equalization processing apparatus for demonstrating the technique concerning this invention. It is a figure which shows the equalization processing apparatus concerning embodiment of this invention. It is a figure which shows the structure of a teletext signal. It is a figure which shows the equalizer disclosed by the nonpatent literature 1. It is a figure which shows the equalizer disclosed by the nonpatent literature 1.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Equalization processing apparatus 20 Equalization filter 30 Coefficient correction circuit 32 Error detection part 34 Coefficient calculation part 50 Equalization processing apparatus 60 1st equalizer 62 Equalization filter 64 Coefficient correction circuit 66 Error detection part 68 Coefficient calculation part 70 Second equalizer 72 Equalization filter 74 Coefficient correction circuit 76 Error detection unit 78 Coefficient calculation unit 100 Equalization processor 112 CRI detection unit 114 FRC detection unit 116 Text data detection unit 120 OR gate 130 First equalizer 132 CRI & FRC equalization filter 134 coefficient correction circuit 136 error detection unit 138 coefficient calculation unit 140 second equalizer 142 text data equalization filter 144 coefficient correction circuit 146 error detection unit 148 coefficient calculation unit 150 binarization unit 160 variance calculation Part

Claims (8)

  Equalization processing is performed on the CRI (Clock Run In) signal and FRC (Framing Code) signal included in the received teletext signal and the text data signal included in the teletext signal at different sampling rates. A characteristic equalization processing method.   The equalization processing method according to claim 1, wherein equalization processing is performed on the CRI signal and the FRC signal at a sampling rate higher than a sampling rate corresponding to a symbol rate of the teletext signal.   The equalization processing method according to claim 1 or 2, wherein equalization processing is performed on the text data signal at a sampling rate corresponding to a symbol rate of the teletext signal.   The equalization process according to any one of claims 1 to 3, wherein equalization processing of the CRI signal and FRC signal and equalization processing of the text data signal are performed by different equalizers. Method.   Equalization processing is performed on the CRI (Clock Run In) signal and FRC (Framing Code) signal included in the received teletext signal and the text data signal included in the teletext signal at different sampling rates. A characteristic equalization processing apparatus.   6. The equalization processing apparatus according to claim 5, wherein equalization processing is performed on the CRI signal and the FRC signal at a sampling rate higher than a sampling rate corresponding to a symbol rate of the teletext signal.   The equalization processing apparatus according to claim 5 or 6, wherein equalization processing is performed on the text data signal at a sampling rate corresponding to a symbol rate of the teletext signal. A first equalizer;
A second equalizer,
The first equalizer performs an equalization process on the CRI signal and the FRC signal,
The equalization processing apparatus according to claim 5, wherein the second equalizer performs an equalization process on the text data signal.

Images