JPH0316488A - Sampling phase detection system for video signal - Google Patents

Abstract

PURPOSE:To attain re-sampling stably even from a video signal including jitter component by superimposing a prescribed periodic signal onto a video signal subjected to sub Nyquist sampling, and detecting and controlling the phase of a sample point of the re-sampling at demodulation by using the periodic signal. CONSTITUTION:A video signal reproduced by a VTR 7 at a receiver side is subjected to elimination of an undesired high frequency component by a low pass filter(LPF) 8 whose cut-off frequency is 2.5MHz and the result is fed to an A/D converter 9. On the other hand, there is a phase difference between a sample point of re-sampling by the A/D converter 9 and a sample point of a sub Nyquist sampling due to jitter of a VTR 7 or the like. An arithmetic processing section 10 calculates a phase difference An between a zero cross point of an extracted periodic signal 4a from an identification signal 4b of an incoming video signal and a sample point of the video signal subjected to Nyquist sampling. Then a clock phase control section 11 controls the phase so that it is advanced or retarded by shift or the like in a way of phase difference An=0, that is, the sample point is converged into a zero cross point of the periodic signal 4a.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a sampling phase detection method for a video signal for performing recombination when demodulating a video signal subjected to subnight sampling.

[Conventional technology]

In conventional systems for transmitting video signals g, sub-Nyquist sampling has been performed to increase resolution. Sub-Nyquist sampling samples the video signal g at a frequency lower than the minimum Nyquist frequency, which is twice the required maximum frequency fn of the video signal, and at the summing frequency selected as the line offset frequency. It is something that

This sub-night strike sampling utilizes frame correlation, a property unique to the video A signal, and can perform band compression to narrow the transmission band. Therefore, broadband video! The signal can be transmitted through a narrowband transmission path, and the wt image quality is improved.

The subnyquist sampled video signal is then demodulated by resampling.

(Problem to be solved by the invention) By the way, 1 m of video sampled by Sapnyquist {
When converting 1 to l[, the sample point for resampling is detected using a color burst signal or the like if there is no jitter. However, in a video signal such as a VTR that includes jitter, it is always difficult to detect a sample point from a video synchronization signal in terms of compatibility with a normal transmission signal.

That is, the phase of the sample point is shifted due to the jitter component, and resampling at this sample point causes deterioration of the image quality, and there is a problem that it is not compatible with the normal transmission signal.

Therefore, the present invention was made in view of the above points, and an object of the present invention is to provide a sampling phase detection method for image signals that can stably re-limb even if the image signal includes a jitter component. do.

[Means to solve the problem]

In order to solve the above-mentioned problem, the present invention superimposes an 18th period signal of an even multiple of the simpling Shumei on the sub-Nyquist sampled video signal, and uses this signal to control the other party at the resampled sample point. It was constructed with a focus on

In other words, the sub-Nyquist sampled video signal 3 is compressed to the sub-Nyquist sampling frequency by sub-Nyquist sampling and transmitted at a predetermined frequency. In this method, a predetermined portion of the sub-Nyquist-sampled video signal is re-sampled with a sampling pulse and demodulated into a reproduced video signal g. Transmit signals by superimposing them.

Then, on the receiving side, the periodic signal 3 whose band is sub-limited to the transmission band of the transmission path is extracted, and the phase difference between the t40 cross point of the periodic signal and the sample point of the resampling is detected. The phase difference of the resampling sample points is 613
00 and demodulates the reproduced video I& signal.

[Effect]

A frequency signal having an even number multiple of the period of the sampling pulse is superimposed on the sub-Nyquist-sampled video m signal on the transmitting side. This signal 3 becomes a signal for detecting a phase difference caused by jitter when being resampled on the receiving side. Therefore, by controlling this phase difference, it is possible to synchronize the sample point of ripple quist sampling on the transmitting side and the ripple point of resampling on the receiving side, allowing stable resampling even for video signals containing jitter components. This makes it possible to ensure the simplicity of . This also makes it possible to automate the tI tuning of the sub-Nyquist sampled video signal. Also,
Subnai. By superimposing a periodic signal on the tossampled video signal, it is possible to distinguish it from the normally transmitted video signal, and compatibility is maintained sufficiently.

〔Example〕

Hereinafter, preferred embodiments of the present invention will be explained with reference to the drawings.

FIG. 1 is a block system diagram showing an embodiment of the system of the present invention. Note that this embodiment is an example in which the present invention is applied to a VTR. In Figure 1, digital analog (D
/A) The sub-night sampled digital video signal that enters the converter 1 is converted into a 5MH frequency signal by a low-pass filter out of the wideband NTSC color video signal.
After removing frequency components higher than z, the app 0g digital (
A/D) converter sampling frequency < fS) 5MH
This is a sampled signal of z.

This sub-Nyquist sampled digital video signal is sent to a clock generator 2 in a D/A converter 1.
Clock frequency 5MH supplied via frequency division N3
It is converted into an analog video m signal based on z. Then, a periodic signal 4a and an identification {Fi number 4b (see FIG. 2}), which will be described later, are added to the analog video signal output from the D/A converter 1 by a superimposing means 4 synchronized with the clock of the clock generator 2. It is superimposed by the adder 5.The analog video signal on which the frequency signal 4a is superimposed has a cutoff frequency of 2.
．． A 5 MHz low-pass filter (LPF) 6 converts the signal into a p-wave analog video signal having a frequency equal to or less than I times the sampling frequency, and supplies the signal to the recording system of the VrR 7.

On the receiving side, the video played on the VTR 7 is filtered through a low-pass filter with a cut-off frequency of 2.5 MHz (
After unnecessary high frequency components are removed by LPF) 8, A/
The signal is supplied to the D converter 9. In the A/D converter 9, the signal is re-linked with a sampling pulse of the clock frequency from the clock phase 11ftil1 section 11, which will be described later, to generate a digital video signal.

Here, the relimbling signal from the A/D converter 9 is
Arithmetic IN! ! ! A predetermined i calculation is performed by the unit 10 and the result is supplied to the clock phase control tm unit 11. On the other hand, the analog video signal that has passed through PF8 is Hli5] Bright signal separation circuit 1
2, and here only the h same signal is supplied to the phase loop circuit 13.

In addition, the PLL circuit 13 changes the sampling frequency fs to zero.
The signal is supplied to the phase control section 11. The clock phase tjJrs section 11 includes an A/D converter 9 and an arithmetic processing section 10.
Then, the phase controlled clock frequency loss is fed back. Thereby, the A/D converter 9 generates a resampled digital video signal synchronized with sub-Nyquist sampling. Note that the above details will be described later.

Next, the periodic signal 4a in FIG.
The case where the signal is superimposed on an analog video signal subjected to sub-Nyquist sampling will be explained with reference to FIGS. 2 and 3. The superposition stage 4 independently generates a periodic signal j34a having the same period as the H domei signal, and this period [signal 4a is subjected to sapnite strike sampling as shown in FIG. 2 (8). Analog fl! l! (2) It is superimposed on a portion other than the Domei signal in the V blanking portion of the image signal. Here, the V-planking portion is a portion that does not appear physically but can be recognized as a video signal. Further, the superimposing means 4, as shown in FIG.
The identification signal 4b (hatched portion) is also superimposed on the previous stage of the hatched portion). Although not shown, this identification signal 4b is generated by a pulse train of a frequency different from that of the frequency signal 4a.
Made, Subnai [Strimbling &! This is a signal g for identifying that it is a recording transmission waveform, and is for distinguishing it from a normal video signal 3.

Here, the waveforms of the sub-ninth sampled analog video signal and periodic signal 4a are shown in FIG.
Figure 3 (A) shows a sampling frequency (fs) of 5MHz.
This is the video signal 3 that was sampled at the time of delivery, and the black dots indicate sample points. Figure 3 (B) shows the sub-equipment strimbling frequency < fs) an even multiple of 5 MHz (
Here it is 10MHz periodic bamboo signal @4a,
The black dots indicate reference sample points of the video signal g, and the white dots indicate sample points at which the frequency rs is doubled.

? Here, the frequency of the periodic bamboo signal 4a is 10 MHz, but if the black circle point is taken as the Biroku Ds point, its fundamental component I.
It is i2.5M port 2. Therefore, L. If it passes through P F 6, the frequency will be 2.5M H as shown in Figure 3 (C).
It becomes a sine wave of z. Note that in this embodiment, the periodic signal M4
Although a is set to twice the sub-nine first sampling frequency fS, even if it is set to an even multiple such as 4 times, 6 times, etc., there is no problem as the fundamental component is 2.5 Ml-tz. do not have. Further, although the periodic signal 4a is expressed as a single sine wave, it does not necessarily have to be a pressure sinusoidal wave, but may be a periodic wave.

Next, an arithmetic processing section 10, a clock appearance control section 11, and so on on the receiving side. H synchronization signal separation circuit 12 and PL11 path 13
The sampling phase detection in will be explained. First, as shown in FIG.
The eye same brightness signal is extracted by b, and the phase of this and the eye synchronization signal extracted from the video signal of the LPF 8 by the H synchronization signal separation circuit 12 is compared by the phase comparator 13c. and,
This H synchronization signal is clocked at the clock phase υ1 by the VCO13a.
Part 11 is supplied with the basic sampling frequency t's.

On the other hand, the sample points resampled by the A/D converter 9 have a phase difference from the sample points obtained by sub-Nyquist sampling due to jitter in the VTR 7 or the like. When the sample point of this resampling is made to correspond to the analog periodic signal 4a input to the A/D converter 9, it is expressed as aπ as shown in FIG. This sample point a'n, that is, the phase difference, is calculated by the calculation processing section 10. This is A/
Sample point aπ resampled by D converter 9
This is because the value usually includes a DC component, so it is difficult to detect the phase difference from the Virocross point in Figure @4. Here, in the arithmetic processing unit 10, from the incoming video signal;! The phase difference An between the zero crossing point of the periodic signal @4a extracted by the I-specific signal 4b and the sample point of the resampled video signal is calculated by the following equation.

ATL = <! 4) (an-++ ann)
1aT+ (where n = -1.0.1, 2, ...) In other words, the calculation process in the above equation is performed by subtracting the sample points in between from the average value of the limble points in one period. The phase difference AT+=α or -α (α is a positive number) from the zero crossing point is obtained, and a constant value α or -α is obtained at a period twice the sampling frequency f'S. This is done every cycle at twice the above-mentioned cycle.

Then, the clock phase III 1211 unit 11 calculates the phase difference Ay+=α or -α obtained by the arithmetic processing unit 10 and PL
The sampling frequency fs of the LOG path 13 controls the partner phase of the resampling pulse to approach the sampling frequency fS, and the A/D converter 9
and feeds back to the theater department 10. That is,
As shown in FIG. 6, the clock phase controller 11 determines the direction of phase shift depending on whether the phase difference An is negative (FIG. 6(A)) or positive (FIG. 6(B)). Then, determine the phase shifting tomb based on its absolute value.Then, the phase difference ATL =O, ″
The clock phase tIllt1 section 11 advances or delays the phase by shifting or the like so that the sample point g, that is, the sample point, converges on the pirox point of the periodic signal 4a. This Ill control is repeated every (2) period, thereby supplying the clock phase to the A/D converter 9 and the arithmetic processing section 10 with the clock phase fixed. This means that the phase 1 for 1 clock is
This means that 111 is possible.

In other words, the sample points of the clock phase II control section 11 are:
This is a resampling limbo point that is synchronized with the sub-Nyquist sampling sample point on the transmitting side, stably securing a resampling sample point that is compatible with normal transmission even if the video signal contains some jitter. becomes possible.

Next, the case where a reproduced video signal is demodulated using the resampled digital video signal will be explained with reference to the block diagram in FIG. 7 and the schematic diagram in FIG. 8.

In FIG. 7, the resampled video signal synchronized with the sub-Nyquist sampling frequency on the transmitting side by the A/D converter 9 is a digital signal consisting of pixel data.
The data is sequentially input to field memories 15, 16, 17, and 18 through a switch circuit 14 that is switched field by field. The resampled video signal with a 4-field cycle is sequentially input to the field memories 15 to 18.
After pixel data for each field is accumulated in the field memories 14 to 18, it is read out. A composite signal consisting of the resampled video signals of the first and third fields (see FIG. 8) read out from field memories 15 and 17, respectively, and a second field read out from field memories 16 and 18, respectively. and the fourth field (see Figure 8)
The composite signal consisting of the recombined video signals is alternately D/D for each field by the switch circuit 19.
It is selectively output to the A converter 20.

Here, FIG. 8 shows a schematic diagram of the relimbed video signal for each field. As is clear from Fig. 8, the sample points are at the positions indicated by black circles in the first field, black squares in the second field, white circles in the 37th field, white squares in the fourth field, and the limble point is Each field is different, 4 fields (2 frames)
It goes around.

Returning to FIG. 7 again, the signal Ji! taken out by the switch circuit 19 will be explained. 2 signal has clock frequency 10M
Analog video &amp;
converted into a signal. After that, the signal with a frequency of 5 MHz or less, that is, the maximum frequency of 4.2
The p-wave is added to the MHZ signal and the reproduced image signal is demodulated.

As is clear from this embodiment, by being able to synchronize the resampling sample point with the sub-Nyquist sampling sample point, it is possible to automate the demodulation of the video signal transmitted by the sub-Nyquist sampling. Even video signals containing jitter components, such as those from VTRs, can be stably reassembled.

〔Effect of the invention〕

As described above, according to the present invention, a predetermined periodic signal is folded into the subnight sampled video signal g, and the phase of the resampling sample point during demodulation is detected using this periodic hull signal. ! 11, it is possible to stably secure sample points for resampling even for video signals containing jitter components, and to maintain sufficient compatibility with normal transmission signals.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a waveform diagram showing a case where a perceptual signal is superimposed on a video signal subjected to sub-Nyquist summing, and FIG. Figure 4 is a block system diagram showing the PLL circuit, and Figures 5 and 6 are waveform diagrams showing the Zabnyquist sampled video signal and periodic signal. A waveform diagram showing points, FIG. 7 is a block system diagram when demodulating a reproduced video signal, and FIG. 8 is a schematic diagram of sample points of each field in FIG. 1...D/A converter, 4...1,000 stages of superposition, 4a...
Periodic signal, 6.8...Low pass filter (LPF), 7
...VTR, 9...A/D converter, 1o...1 arithmetic processing section, 11... clock phase I1 control section, 12...
H synchronous signal separation circuit, 13... PLL circuit.

Claims (1)

[Claims] A sub-Nyquist sampled video signal is sampled at a predetermined period in order to transmit a video signal with a wider band than the transmission band of an existing transmission path by compressing the band to a sub-Nyquist sampling frequency by sub-Nyquist sampling. In a video signal sampling phase detection method that detects the phase of the sampling pulse when resampling the pulse and demodulating the reproduced video signal, on the transmitting side, the sub-Nyquist sampled video signal is A periodic signal of an even multiple of the sub-Nyquist sampling frequency is superimposed and transmitted, and on the receiving side, the periodic signal band-limited to the transmission band of the transmission path is extracted, and the zero crossing point of the periodic signal is A sampling phase detection method for a video signal, characterized in that a phase difference with a sample point of sampling is detected, and the phase difference of the sample point of resampling is controlled to demodulate a reproduced video signal.