JPH104507A - Waveform equalization device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent production of discontinuous point at both extreme tap coefficients of a TF 102 by setting a gain with respect to the frequency component of a digital filter constant with respect to a frequency component with a small gain. SOLUTION: A CPU 606 discriminates it that an amplitude of each frequency component of a D(f) resulting from an error signal d(t) with high speed Fourier transform is -20dB or over with respect to a DC component. When discriminated to be over -20dB, the frequency characteristic of a TF 602 is calculated by using the division method being the same as a conventional method. When discriminated not, a frequency characteristic H(f) is set to 0dB. The processing is executed for all frequency components. Then the H(f) is subject to inverse high speed Fourier transform to obtain h(t) so as to calculate a tap coefficient C(i) of the TF602. The coefficients C(i),d(t) are convoluted to calculate an error signal d'(t) and it is used to calculate a tap coefficient of the TF604. Then the C(i) is set to the TF602, and the tap coefficient obtained is set to the TF504.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a waveform equalizer and, more particularly, to a device for improving a non-recursive filter so that discontinuous points do not occur at both ends of the filter.

[0002]

2. Description of the Related Art In recent years, with the deterioration of reception conditions due to ghosts or the like caused by reflections of buildings, etc., a waveform equalizer has been used in a television receiver for the purpose of removing ghosts. In addition, as the size of the screen increases and the resolution of the screen increases due to high-definition broadcasting, waveform equalization is becoming an increasingly important technology. 1
The same applies to the second generation EDTV broadcasting started in 995. The waveform equalizer used for a television receiver or the like is required to have higher performance despite the deterioration of the reception condition.

[0003] A conventional waveform equalizer will be described below. FIG. 10 is a diagram showing a basic configuration of a conventional waveform equalizer. In FIG. 1, reference numeral 101 denotes an input television signal S.
A / that converts 121 from an analog signal to a digital signal
D conversion means. 102 is an input television signal S1
The television signal S123 is obtained by performing the waveform equalization processing of No. 22
Is generated. The transversal filter 102 constitutes a non-recursive filter.

An adder 103 adds the television signal S123 and the television signal S125 to generate an output television signal S124, and 104 performs a waveform equalization process on the output television signal S124. And a transversal filter for generating a television signal S125 to be supplied to the adder 103. The adder 103 and the transversal filter 104 constitute a recursive filter. 105 is an input television signal S1
Reference signal storage means 106 for extracting a reference signal for waveform equalization superimposed in the vertical blanking period of each of the output television signal S124 and the output television signal S124, and storing the extracted reference signal. Control means for performing the control of The above non-recursive filter corrects transmission distortion and removes near ghosts, and the recursive filter removes long-range ghosts.

The operation of the waveform equalizer configured as described above will be described below. FIG. 11 is a flowchart showing an outline of the processing in the control means 106 of the waveform equalizer of FIG. Hereinafter, the transversal filter is abbreviated as TF.

The TF 102 is initialized so that the input television signal S122 and the television signal S123 are equal (step 201). Next, the television signal S
Initialization of TF 104 (TF
104 (setting of tap coefficients) (step 202).

Then, a reference signal is extracted from the input television signal S122 by the reference signal storage means 105 (step 203). This reference signal is read (step 20).
4) Perform synchronous addition (procedure 205). By repeatedly adding the reference signal by the synchronous addition, the influence of noise can be eliminated and the SN ratio can be improved.

The above steps 203 to 205 are repeated until the predetermined number of times of synchronous addition is completed. The synchronously added reference signal is subjected, if necessary, to a field sequence process or a differentiation process to generate an error signal d (t) (procedure 20).
7). An SN ratio is obtained using the error signal d (t) (step 208). Specifically, the pedestal level portion of the error signal d (t) is added, and the SN is obtained by referring to a look-up table created in advance based on the added value.
A ratio is determined.

The error signal d (t) obtained in step 207 is divided on the frequency axis, that is, the error signal d (t)
Is subjected to fast Fourier transform to obtain D (f), and an ideal D (f) having no distortion or ghost is defined as R (f), and R (f) is obtained.
The frequency characteristic H (f) of the digital filter is obtained by dividing (f) by D (f) (step 209).

The frequency characteristic H (f) is subjected to an inverse Fourier transform to obtain an impulse response h on the time axis.
(T) is obtained (procedure 210). This determined response h
From (t), a portion corresponding to TF102 is used as a tap coefficient of TF102 (procedure 211). The tap coefficient represents the impulse response of the filter on the time axis. By adjusting the tap coefficient, the impulse response is adjusted, that is, the frequency characteristic is adjusted.

Next, a convolution operation of the tap coefficient set in the TF 102 and the error signal d (t) obtained in the step 207 is performed, and when the error signal d (t) is input to the TF 102, it is obtained as an output of the TF 102. Error signal d '
(T) is calculated (step 212). Then, the tap coefficient of the TF 104 is calculated using the error signal d '(t) (procedure 213). The tap coefficient obtained in step 211 is TF
It is set to 102 (procedure 214). The tap coefficient obtained in step 213 is set in the TF 104 (step 215). Steps 203 to 215 are repeated to cope with a change in the state of the input television signal S121. Updating of the tap coefficient as described above is performed at a rate of about once every several seconds to several minutes, and is performed during a vertical blanking period to prevent image disturbance.

FIG. 12 is a diagram showing an amplitude frequency characteristic diagram ((a)) and a phase frequency characteristic diagram ((b)) of the input television signal S122 and the output television signal S124. From this figure, it is understood that the distortion shown in the input television signal S122 can be reduced by executing the processing shown in the flowchart of FIG. 11 and controlling the TF 102 and the TF 104. That is, FIG.
(1) and an input television signal S1 having a distorted frequency characteristic as shown in FIGS.
In FIG. 12, the frequency characteristics are corrected by performing a waveform equalization process using the TFs 102 and TF104.
(A)-(2) and output television signal S124 having frequency characteristics as shown in FIGS. 12 (b)-(2).
Becomes

FIG. 13 is a diagram showing an example of an image of the input television signal S122 and the output television signal S124. As shown in FIG. 13A, the input television signal S122 on which a ghost is superimposed on the screen is a TF
A ghost is removed by performing a waveform equalization process using the TF 104 and the TF 104, and as shown in FIG.
124.

[0014]

The conventional waveform equalizer is configured as described above, and by performing signal processing using the device, a ghost-free image can be obtained. The control method using the equalizer has the following three problems.

The first problem is that the number of taps of the TF 102 constituting the non-recursive filter is finite. Response h obtained by inverse Fourier transform in step 210
In (t), in step 211, the tap coefficient of the TF 102 is controlled using only a portion corresponding to the number of taps of the TF 102. Therefore, as shown in FIG.
The portion of the response h (t) that does not correspond to the tap of the TF 102 (the “unused” portion in FIG. 14) is not used, and as a result, the correction that would have been performed by this portion is not performed. . Therefore, the input television signal S121 cannot be corrected to a desired characteristic, and discontinuous points occur at both ends of the tap of the TF.

The second problem is that if the reference signal extracted in step 203 has an abnormality due to a temporary deterioration of the reception state, etc., if the waveform is equalized in accordance with the reference signal, the TF 102 and The point is that an inappropriate tap coefficient is set in the TF 104.

The third problem is that the operation state of the waveform equalizer and the input / output television signal to / from the waveform equalizer cannot be confirmed from the outside, and the operation of the waveform equalizer depends on the screen. This is a point that can be confirmed only by watching the video actually projected on the above, and it is difficult to obtain information for debugging or the like for device development.

The present invention has been made in order to solve the above-mentioned problems, and it is possible to prevent a discontinuous point from being generated at both ends of a tap of a transversal filter and to change the state of an input signal even when the state of an input signal changes. It is an object of the present invention to provide a waveform equalizer capable of performing a normal waveform equalization operation and confirming the operation inside the waveform equalizer.

[0019]

According to a first aspect of the present invention, there is provided a digital filter for inputting an input television signal, and a digital filter for waveform equalization superimposed on a vertical blanking period of the input television signal. Signal storage means for storing a reference signal for extracting the reference signal from the input television signal, writing the reference signal in the reference signal storage means, and controlling the digital filter based on the stored reference signal. And a waveform equalizer having
The control means obtains the amplitude frequency characteristic of the reference signal, compares the gain of each frequency component of the amplitude frequency characteristic with a certain level, and determines the frequency component of the digital filter with respect to the frequency component having a smaller gain than this level. The gain relating to the frequency component is set to a constant value.

According to a second aspect of the present invention, a digital filter for inputting an input television signal is superimposed on a vertical blanking period of the input television signal.
Reference signal storage means for storing a reference signal for waveform equalization; extracting the reference signal from the input television signal, writing the reference signal in the reference signal storage means, and storing the reference signal based on the stored reference signal. And a control unit for obtaining the SN ratio of the television signal and controlling the digital filter sequentially. In the waveform equalizer, the second and subsequent sequential control by the control unit are used in the previous sequential control. An amount of change between the first reference signal and the second reference signal acquired by the current sequential control is obtained, and when the amount of change exceeds a threshold value corresponding to the SN ratio, the second reference signal is obtained. Is not used for controlling the digital filter.

According to a third aspect of the present invention, a digital filter to which an input television signal is input and a reference signal for waveform equalization superimposed on a vertical blanking period of the input television signal are stored. Reference signal storage means, and control means for extracting the reference signal from the input television signal or the output television signal output from the digital filter, writing the reference signal in the reference signal storage means, and controlling the digital filter. A waveform equalizer having a switch for changing an operation state of the control means, and a transmission means for transmitting data by the control means, wherein, when the switch is on, the transmission means Is used to transmit the reference signal and the internal information of the control means.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a waveform equalizer according to an embodiment of the present invention will be described. In each of the embodiments described below, a GC which is a reference signal for waveform equalization is used.
An example will be described in which a television signal (hereinafter, referred to as an EDTV signal) in which an R signal is superimposed in the vertical blanking period is an input television signal.

Embodiment 1 FIG. 1 is a configuration diagram of a waveform equalizer according to Embodiment 1 of the present invention. In FIG. 1, reference numeral 601 denotes an input EDTV signal S62.
1 is an A / D converter for converting an analog signal 1 into a digital signal S622. 602 is an input EDTV signal S62
2 is a 52-tap transversal filter (the first tap is the zeroth tap and the 32nd tap is the main tap) that performs the waveform equalization process of No. 2 and generates the EDTV signal S623. It has become something. 603 is an EDTV signal S62
3 and an EDTV signal S625 to generate an output EDTV signal S624. An adder 604 performs a waveform equalization process on the output EDTV signal S624 and outputs an EDTV signal S624.
This is a 150-tap transversal filter that generates 25, and the adder 603 and the transversal filter 604 form a recursive filter.

Reference numeral 605 denotes the input EDTV signal S622,
A buffer memory 606 for extracting a GCR signal which is a reference signal for waveform equalization superimposed on each vertical blanking period of the output EDTV signal S624 and storing the extracted GCR signal. CPU that performs
It is.

FIG. 2 shows the CPU 606 of the first embodiment.
FIG. 6 is a diagram showing a flowchart of a waveform equalization process in FIG. Input EDTV signal S622 and EDTV signal S623
Are set to make the transversal filter 602 equal (step 701). Next, initialization of the transversal filter 604 is performed so that the EDTV signal S625 is not output (step 702). The GCR signal is extracted from the EDTV signal S622 in the buffer memory 605 (procedure 703). This GCR signal (FIG. 3 (a),
(B)) is read (procedure 704), and synchronous addition is performed (procedure 705). The above steps 703 to 705 are repeated until the predetermined number of synchronous additions ends. An error signal d (t) is obtained from the synchronously added GCR signal (FIG. 3C).
Is generated (procedure 707). An SN ratio is obtained from the error signal d (t) (procedure 708). The error signal d (t) obtained in step 707 is subjected to fast Fourier transform to obtain D (f) (step 709).

Next, it is determined whether or not the amplitude of each frequency component of D (f) is -20 dB or more with respect to the DC component (step 710). In this step 710, -20 dB
If it is determined that the frequency component is higher than the frequency component H (f) of the TF 602 using a division method as in the related art.
Is calculated (procedure 711). On the other hand, in step 710 above,
If the frequency component is not determined to be 20 dB or more, the frequency characteristic of the frequency component is set to H (f) = 0 dB (step 7).
12). Then, steps 710 to 712 are executed for all frequency components.

Next, the frequency characteristic H (f) obtained up to step 713 is subjected to inverse fast Fourier transform to obtain h (t) (step 714), the gain is adjusted, and the tap coefficient c (i) of the TF602 is obtained. (I = 0 to 51) is calculated (procedure 71).
5). The convolution operation of the tap coefficient c (i) and the error signal d (t) is performed, and when the error signal d (t) is input to the TF 602, an error signal d ′ (t) obtained as an output of the TF 602 is calculated. (Step 716). The tap coefficient of the TF 604 is calculated using the error signal d ′ (t) generated in step 716 (step 717). The tap coefficient c (i) obtained in step 715 is set in the TF 602 (step 71).
8). Further, the tap coefficient obtained in step 717 is set to TF6
04 (step 719).

Steps 703 to 719 are repeated to respond to the change in the state of the input EDTV signal S621. By performing the signal processing as described above, of the frequency components of the frequency characteristic D (f) of the error signal, the frequency component whose amplitude is less than −20 dB with respect to the DC component is not divided, and the gain is replaced by the gain. Is set in the TF 602 such that is equal to 1 (FIG. 4A). Generally, when H (f) having an impulse-like frequency characteristic is subjected to inverse fast Fourier transform, a response h having a spread as shown in FIG.
(T). Therefore, division is not performed on the frequency component whose amplitude is less than −20 dB, and the gain of H (f) is set to 1 instead.
To prevent H (f) from having an impulse-like frequency characteristic. As a result, FIG.
As shown in FIG. 5, the frequency response D (f) having the frequency characteristic D (f) as shown in FIG. The spread is suppressed in the present invention, and the tap of the TF 602 generated from the response h (t) does not have a discontinuous point at both ends thereof. As shown in FIG. However, as shown in FIG. 6B, the tap coefficient has a reduced spread. Thus, it is possible to prevent a discontinuous point from being generated due to the finite tap length.

In the first embodiment, the number of taps of the acyclic filter is set to 52, and the main tap position is set to 32.
The number of taps of the recursive filter is set to 150, and the threshold value of the amplitude used in step 710 is set to a constant such as -20 dB. However, these values are not limited to those described above.

Embodiment 2 Hereinafter, a waveform equalizer according to Embodiment 2 of the present invention will be described with reference to the drawings. The configuration of the waveform equalizer is the same as that of the first embodiment, and the description thereof is omitted here.

FIG. 7 is a flowchart showing a waveform equalization process in the CPU 606 according to the second embodiment. In FIG. 7, procedure procedures 701 to 708, procedure 70
9, step 711, and steps 714 to 719 are the same as the processing described in FIG.

The first error signal d obtained in step 707
(T) and the second error signal d obtained in the previous step 707
(T) (this second d (t) is hereinafter referred to as d1 (t)), and the absolute value x (x = | d
(T) -d1 (t) |) is obtained (procedure 1101). It is determined whether or not the absolute value x of the difference is equal to or greater than a threshold value y corresponding to the SN ratio obtained in step 708 (step 1102). If it is determined in step 1102 that the absolute value x of the difference is equal to or larger than the threshold value y, (xy) 2 is added to the accumulated error e. If it is not determined in step 1102 that the absolute value x of the difference is equal to or larger than the threshold value y, the procedure 1104 is executed, and the above procedures 1101 to 1104 are executed for all t.

Next, the accumulated error e becomes the threshold value 50,000,
It is determined whether it is 0 or more (procedure 1105). If it is determined that the accumulated error e is equal to or larger than the threshold value 50,000, the subsequent processing is not performed, and the process returns to the GCR extraction processing (procedure 703). On the other hand, the accumulated error e is equal to the threshold value 5
If it is determined that the value is less than 0000, the process is continued, and the process proceeds to the next step 709. Thereafter, the same processing as in the first embodiment is performed.

As described above, according to the second embodiment,
If it is determined in step 1105 that the accumulated error e is 50,000 or more, it is determined that the change in the error signal d (t) is large. The reason why the change of the error signal d (t) is large is that the reception state is temporarily deteriorated. If waveform equalization is performed using the error signal d (t) obtained when the reception state temporarily deteriorates, inappropriate tap coefficients will be set for the TF 602 and the TF 604. Therefore, when the value of the accumulated error e is large, it is considered that the extracted reference signal has an error, and the GC
By not using the R signal for waveform equalization, TF602,
In addition, it is possible to prevent setting an irregular tap coefficient in the TF 604. In addition, Embodiment 2
50,0 as the threshold used in procedure 1105 in
Although 00 was used, the constants used are not limited to these values.

Embodiment 3 Hereinafter, a waveform equalizer according to Embodiment 3 of the present invention will be described with reference to the drawings. FIG. 8 is a configuration diagram of a waveform equalizer according to the third embodiment. 8, 601 to 606 and S621 to S625 are the same as those of the configuration of the waveform equalizer shown in FIG. In FIG. 8, reference numeral 1201 denotes a control unit (C
A switch 1202 for notifying the PU 606) of a change in the operation state is a transmission circuit for transmitting data from the control unit to the outside.

FIG. 9 shows the control means C according to the third embodiment.
FIG. 14 is a diagram illustrating a flowchart of a waveform equalization process in a PU 606. Note that, in FIG. 9, procedure procedures 701 to 708, procedure 709, procedure 711, and procedure 714 are performed.
Step 719 is the same as the processing shown in the flowchart of FIG. 2, and the description thereof is omitted here.

It is determined whether the switch 1201 is turned on (step 1301). Switch 12 in procedure 1301
When it is determined that 01 is ON, the error signal d (t) generated in step 707 is transmitted to the outside using the transmission circuit 1202 (step 1302). Switch 1 in procedure 1301
If it is not determined that 201 is ON, the procedure proceeds to step 709.
To perform fast Fourier transform.

Next, after performing the division processing in step 711, if it is determined in step 1303 that the switch 1201 is on, the frequency characteristic D of the error signal obtained in step 709 is obtained.
(F) And the frequency characteristic H (f) of the TF 602 obtained in the step 711 is transmitted to the outside using the transmission circuit 1202 (step 1304). On the other hand, if it is not determined in step 1303 that the switch 1201 is on, the procedure 71
4 to perform inverse fast Fourier transform.

If it is determined in step 1305 that the switch 1201 is on, the TF 60 obtained in step 715
The transmission coefficient 1202, the tap coefficient of the TF 604 obtained in step 717, the reference signal of the input / output EDTV signal, and information indicating various operation states are transmitted to the outside using the transmission circuit 1202 (step 1306). On the other hand, if it is not determined in step 1305 that the switch 1201 is on,
Execute 3.

As described above, according to the third embodiment,
By turning on the switch 1201, the CPU 60
6 is changed, and the procedure 1302, procedure 1304, and procedure 1304 are changed.
In step 1306, various information is transmitted to the outside. As a result, the operation state of the waveform equalizer and the state of the input / output EDTV signal can be confirmed as data in addition to the fact that the state is confirmed by observing an image actually projected on the screen. Become. This makes it possible to easily perform debug processing during device development and maintenance after marketing.

The procedure 130 in the third embodiment
2. Needless to say, the contents to be transmitted in step 1304 and step 1306 are not limited to those described above.

[0042]

As described above, according to the first aspect of the invention, the gain of each frequency component of the amplitude frequency characteristic of the reference signal is monitored, and the gain of the digital filter for the frequency component having a small gain is monitored. By making the value constant, the spread of h (t) can be suppressed, and the effect of preventing discontinuous points from being generated at both ends of the tap coefficient of TF 102 generated from h (t) can be prevented. There is.

According to the second aspect of the present invention, the amount of change in the reference signal is monitored successively, and when the amount of change is large, the reference signal is not used for controlling the digital filter.
There is an effect that it is possible to prevent an irregular tap coefficient from being set in a digital filter by using a reference signal that cannot be normally extracted due to a temporary deterioration of a reception state or the like.

According to the third aspect of the present invention, when the switch is on, the control means transmits the reference signal and the information inside the control means using the transmitting means, so that the operation of the waveform equalizer can be performed. The state and the state of the input / output television signal of the waveform equalizer can be externally confirmed, and there is an effect that the debugging process at the time of device development and the maintenance after marketing can be easily performed.

[Brief description of the drawings]

FIG. 1 is a diagram showing a basic configuration of a waveform equalizer in Embodiments 1 and 2.

FIG. 2 is a diagram showing a flowchart of a waveform equalization process of the waveform equalizer in the first embodiment.

FIG. 3 is a diagram for explaining a state of a GCR signal in an EDTV signal.

FIG. 4 is a diagram for explaining setting of frequency characteristics of TF 602 in the first embodiment.

FIG. 5 is a diagram for explaining a state of an inverse Fourier transform by the waveform equalizer according to the first embodiment.

FIG. 6 is a diagram for explaining setting of tap coefficients of TF 602 according to the first embodiment.

FIG. 7 is a diagram showing a flowchart of a waveform equalization process of the waveform equalization device according to the second embodiment.

FIG. 8 is a diagram illustrating a basic configuration of a waveform equalizer according to a third embodiment.

FIG. 9 is a diagram showing a flowchart of a waveform equalization process of the waveform equalizer in the third embodiment.

FIG. 10 is a diagram showing a basic configuration of a conventional waveform equalizer.

FIG. 11 is a diagram showing a flowchart of a waveform equalization process in a conventional waveform equalization device.

FIG. 12 shows amplitude frequency characteristics of a conventional waveform equalizer,
FIG. 7 is a diagram showing a state of correction of phase frequency characteristics.

FIG. 13 is a diagram for explaining a state of ghost removal using a conventional waveform equalizer.

FIG. 14 is a diagram for explaining a problem of the conventional waveform equalizer.

[Explanation of symbols]

101 A / D conversion unit 102, 104, 602, 604 Transversal filter 103, 603 Adder 105 Reference signal storage unit 106 Control unit 601 A / D converter 605 Buffer memory 606 CPU 1201 Switch 1202 Transmission circuit S121 Input television signal S122 Digitized input television signal S123, S125 television signal S124 Output television signal S621 Input EDTV signal S622 Digitized input EDTV signal S623, S625 EDTV signal S624 Output EDTV signal

Claims (3)

[Claims] 1. A digital filter to which an input television signal is input, reference signal storage means for storing a reference signal for waveform equalization superimposed on a vertical blanking period of the input television signal, Control means for extracting the reference signal from the input television signal, writing the reference signal in the reference signal storage means, and controlling the digital filter based on the stored reference signal, wherein the control means Obtains the amplitude frequency characteristic of the reference signal, compares the gain of each frequency component of the amplitude frequency characteristic with a certain level, and determines the frequency component of the digital filter for the frequency component having a smaller gain than the level. A waveform equalizer, wherein a gain of the waveform equalizer is set to a constant value. 2. A digital filter to which an input television signal is inputted, and reference signal storage means for storing a reference signal for waveform equalization superimposed on a vertical blanking period of the input television signal; The reference signal is extracted from the input television signal and written into the reference signal storage means. Based on the stored reference signal, an SN ratio of the television signal is obtained from the reference signal, and the digital filter is sequentially controlled. The waveform equalizer having control means, wherein the control means, after the second successive control, the first reference signal used in the preceding sequential control and the first reference signal captured in the current sequential control. The second reference signal is not used for controlling the digital filter when the amount of change from the second reference signal is obtained and the amount of change exceeds a threshold value corresponding to the SN ratio. A waveform equalizing device. 3. A digital filter to which an input television signal is input, reference signal storage means for storing a reference signal for waveform equalization superimposed on a vertical blanking period of the input television signal, Control means for extracting the reference signal from the input television signal or the output television signal output from the digital filter, writing the reference signal in the reference signal storage means, and controlling the digital filter, and an operation state of the control means A switch for changing data, and a transmitting means for transmitting data by the control means. When the switch is on, the control means uses the transmitting means to transmit the reference signal. And a device for transmitting internal information of the control means.