JPS5935240B2 - Horizontal reference pulse extraction method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for forming a horizontal reference pulse from a horizontal synchronizing signal and a burst signal of an NTSC television signal.

Generally, a horizontal synchronization signal (hereinafter referred to as HD) is used as the horizontal reference pulse of an NTSC television signal, but the reference synchronization signal is obtained by correcting the time axis of a television signal containing jitter, such as the playback signal of a video tape recorder. In some cases, such as a time base corrector that is synchronized to a horizontal reference pulse, which is accurate and also includes reference phase information of the color subcarrier, is required.

In this case, it is a well-known fact that in the NTSq color television signal, a horizontal synchronizing signal and a burst signal are used to form a pulse corresponding to one burst pulse, and the pulse is preferably used as a horizontal reference pulse. However, the phases of the horizontal synchronizing signal and the noise signal do not have a fixed relationship, and there is often jitter between the two signals. Furthermore, the burst signal is transmitted for each scanning line (hereinafter referred to as H). The phase changes by 1800. Therefore, some effort is required to form the horizontal reference pulse with a relationship between the two signals. Note that the horizontal reference pulse does not always have to have the phase of the falling (or rising) edge of a specific burst pulse;
Once it is operated, it is sufficient that it can be stably formed as a pulse having a different 1/2 cycle phase for each horizontal period without fluctuation only during the operation.

This is because, for example, in a time base corrector, the write side and the readout side to the memory are interlocked and use the same reference pulse. Here, a block diagram of an example of a conventional forming method is shown in FIG. 1, and a waveform diagram of its input signal is shown in FIG.

Further, parts a and b of Fig. 2 are enlarged to show HD on the same time axis, and a single waveform diagram is shown in Fig. 3. In the figure, 1 is HI), 2 is burst pulse, and 3 is monostable multivibrator (MV
), 4 is the MV output signal, and 5 is the flip-flop (
(hereinafter referred to as F-F), 6 is F-F output pulse, VR is M
The output pulse width regulator of V is shown. To explain the operation, M
VI triggers MV3 and MV's output pulse width regulator VR
The pulse width T is set by VL. The phase of the M output signal 4, that is, 11D, having the same value is delayed and a signal of 1 is generated. The trailing edge of this MV output signal 4 is F. F5f (set. - side burst pulse 2 is the M output signal 4 is the F.F5
At the next falling edge (B2 or B2 in FIG. 3 in this example) after setting the F. Reset F5. Therefore F.
The F output pulse 6 is Twa or T as shown in FIG.
The signal has a width of wb.

This F. The trailing edge of the F output pulse 6 becomes a signal synchronized with the falling edge of the burst pulse 2, and a pulse corresponding to the phase of one burst pulse is formed. The problem here is that when the phase difference Ta, tb between HDl and the burst pulse 2 changes depending on the input signal conditions as described above, for example, the pulse width T of the MV output signal 4.

Even if the MV output signal 4 is constant, there are cases where the trailing edge of the MV output signal 4 lags behind the falling edge of B1 or precedes the falling edge of B1. The edges will be extracted, and B2 and B2, which were extracted until then, will be extracted.
If the falling edge of the combination ′ is not extracted (for example, B2
and B3', or (if a combination of B1 and B is extracted) are generated. However, each time the input signal condition changes, the output pulse width regulator VR of MV3 must be varied to adjust the width of TO.
, B;) may not be extracted. Also, between HDl and burst pulse 2 there is a ±l/4SC period (
(SC means subcarrier) If there is a jitter equivalent to Br or above, even if the output pulse width adjuster VR of MV3 is set to the best point, the trailing edge of M output signal 4 will be Br as in the previous case. Sometimes it lags behind the falling edge of B, and sometimes it precedes the falling edge of B.
It becomes impossible to extract the falling edge of the burst pulse of S). Therefore, the purpose of the present invention is to
and to stably form horizontal reference pulses from burst pulses. Another object is to make it possible to stably form a horizontal reference pulse even if there is jitter between the HD and burst pulses, as long as the amount of jitter corresponds to within ±1/2 SC period. To achieve the above objective, we measure the phase of the HD and burst pulses, and vary the MV pulse width according to the amount so that the desired falling edge of the burst pulse can always be extracted, and at the same time change the MV pulse width. The allowable amount of jitter between the HD and the burst pulse can be reduced by modulating the width by an amount corresponding to ±l/4SC period using a pulse width variable signal that changes at a period twice as long as the H generated by the HD. This allows it to be expanded to an amount equivalent to 1/2 SC cycle.

A block diagram of an example of the proposed method is shown in FIG. 4, and a waveform diagram thereof is shown in FIG.

In the figure, the same parts and the same signals as in Figs. 1, 2, and 3 are given the same symbols, 7 is a variable pulse width signal generator, 8 is a variable pulse width signal, and 9 is an addition signal. , 10 is a pulse width error signal, and 11 is a pulse width detector.

To explain the operation, HDl triggers MV3 and at the same time becomes an input to a variable width signal generator 7 composed of a flip-flop η, and generates a variable pulse width signal 8.

This pulse width variable signal 8 changes the pulse width of MV3 by a certain amount Td every 1H (in this embodiment, 1/4SC cycle key 70
The voltage +Vd or -Vd is such that the voltage changes by nS) in the positive or negative direction.

MV3 is a pulse width error voltage 10, which will be described later, and the average value of the pulse width is T. It is set so that Therefore, the MV output signal 4 triggered by HDl is the fifth
TO+T, or T as shown in FIG. -T, which is a signal that changes every 1H. In other words, MV3 is HDl (
It can be thought of as a delay means that delays the leading edge of the burst pulse) to the position of the burst pulse.

On the other hand, F. F5 is set at the trailing edge of the MV output signal 4 and is set at the trailing edge of the burst pulse 2 following it, so that the F5 as shown in FIG. Generate F output pulse 6.

This F. The pulse width Twa or Twb of the F output pulse 6 is sequentially measured by the next pulse width detector 11, and the average value is always a certain constant value (in this embodiment, 1/2 SC period - 1
40 nS) is generated to control MV3. Therefore, the average value of the pulse width of the MV output signal 4 controlled by this pulse width error signal 10 is TO. For example, T. T when becomes wider. +T
d or T. -Td also becomes wider and Twa or TWb becomes narrower, so the pulse width error signal 10 becomes T. Generates a voltage that narrows T. The regular width T. It is starting to return to .
T again.

Even if the width becomes narrow, the width returns to the normal width by the reverse operation.
On the other hand, when the phase difference Ta, tb between HDl and burst pulse 2 changes, initially TO+Td or T. -Td does not change, but Twa and tWb change, so the pulse width error signal 10 becomes a positive or negative voltage depending on the amount of change. Change TWaC! ．． It operates so that the average value of the width of TWb becomes a constant value. This means that the phase difference Ta or Tb between D1 and burst pulse 2 is determined by F.
This means that the widths TWa and tWb of the F output pulse 6 are measured instead. Therefore F. F output pulse 6
The average value of the widths TWa and twb is always a constant value, and HD
Even if the phase difference Ta or Tb between burst pulse 2 and burst pulse 2 changes, if it is within 1/2 SC period = 140 nS, F. The trailing edge of the F output signal 6 (FIG. 4, 6) is always the falling edge of the desired burst pulse 2 (B2.B in this embodiment).
: ), and the falling edge of one burst pulse has been extracted. In this embodiment, the phase difference Ta or Tb between HDl and burst pulse 2 is determined by F. The width of the F output pulse 6 was measured and used as a substitute using the feedback loop, but the phase difference between the trailing edge of the MV output signal 4 and the subsequent falling edge of the burst pulse 2 was measured and the width was used as a substitute using the feedback loop. How to keep the phase difference always at a constant value, and the phase difference Tw between HDl and burst pulse 2
A possible method is to measure a and tWb and control M3 using an open loop. As described above, according to the invention of the present application, it is possible to form a horizontal reference pulse stably and accurately even when the phase difference between the HD and burst pulses changes or when jitter is included.

[Brief explanation of drawings]

FIG. 1 shows a block diagram of an example of a conventional method. FIG. 2 shows an input signal waveform diagram, and FIG. 3 shows a waveform diagram for explaining the operation of FIG. Further, FIG. 4 shows a block diagram of an embodiment of the present invention, and FIG. 5 shows an explanatory waveform diagram thereof. 1 is HDl2 is burst pulse, 3 is MVl4 is M
V output signal, 5 is F. Fl6 is F. F output pulse, 7 a pulse width variable signal generator, 8 a pulse width variable signal, 9 an addition signal, 10 a pulse width error signal, 11 a pulse width detector.

Claims (1)

[Claims] 1. In a horizontal reference pulse forming method in which a horizontal reference pulse is formed by an input horizontal synchronization signal and an input color burst signal, a delay means for delaying the input horizontal synchronization signal;
means for creating a horizontal reference pulse having a pulse width corresponding to the phase difference between the delayed horizontal synchronizing signal output from the delay means and the input color burst signal; means for producing a phase difference signal between the input color burst signals; means for obtaining a variable pulse width signal having a period twice that of the horizontal synchronization signal based on the input horizontal synchronization signal; A horizontal reference pulse forming method in which the pulse width of the horizontal reference pulse is made constant by adding variable pulse width signals and controlling the delay amount of the delay means using the added signal.