JPH02200040A - Stuff system for picture signal - Google Patents

Abstract

PURPOSE:To reduce the size of a stuff memory by defining the least common multiple of bit cycles between an input picture signal and a transmission signal as a unit time to first store the bit information equal to the difference of bits between both signals into a memory and then performing the input and output operations at one time and in the unit time. CONSTITUTION:A write clock fv and a read clock ft are produced via a PLL circuit 4, and the least common multiple of bit cycles between an input picture signal and a transmission signal is defined as a unit time. In such a case, the input bet number is referred to as xv1 with the output bit umber as xt respectively. A switch 32 is set at the (b) side to store the information into a stuff memory 2. At the same time, a stuff dummy data generating circuit 31 outputs the dymmy data synchronous with the clock ft. Then the switch 32 is set at the (a) side to perform the input and output operations synchronously with both clocks fv and ft. This action is repeated by xv times. Thus, the size of the memory 2 can be reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a stuffing method, and particularly to an image signal stuffing method suitable for transmitting image signals through various hierarchical digital transmission paths.

[Conventional technology]

A configuration diagram of a device related to the conventional image stuffing method is shown in FIG. 2, and a conceptual diagram is shown in FIG.
The method was to store information on the product of the difference in the number of input and output bits per unit time and the number of input bits in a stuff memory, and then perform input and output simultaneously to empty the memory. This conventional method will be explained below. The writing clock of the input image signal to the stuff memory is fv, and the reading clock from the memory is fc6. Let the size of the stuff memory be M, the number of input bits per unit time be Xv, and the number of output bits be Xt.

In FIG. 3, if the time for writing M-bit information into the stuff memory is τ1, then it is expressed by equation (1).

τ1=M/Xv・
(1) During τ1, the output from the memory is interrupted, and Bo stuffing dummy data are output in synchronization with ft. sBo is expressed by equation (2).

Bo = τ1m Xt = M umbrella x c / x v・= (2) fv between 12
The memory becomes empty by inputting the Sv bit and outputting the St bit in synchronization with .

This τlHSv+st is expressed by equations (3) to (5).

τz=M/ (xt xv) −(
3) S v "X v-τ2 = M umbrella xv/ (xt-Xv)...-(4)
S t = x in fist τ2 =M* xt/ (xt xv) ・=
(5) In order to control the above operations, BoH8vH
If 8t is an integer, then M is Xv (xt xn
) is realized if it is an integer multiple of ). Also, if the image has a unique block configuration, and its size is Xs,
M is an integer multiple of X$*Xv Liao (xt xn).

As a specific example, f −=392.736M b/s
, f t =397.2M b/s, then Xv =8182 bits.

xt” is 8275 bits, and the memory size M is 8275 bits.
182ψ(8275-8182) = 760926
Becomes a bit. Also, if the image has a unique block structure of 364 bits, M: 364 out of 8182 (8
275-8182) = 276977φ64 bits.

[Problem to be solved by the invention]

In the above conventional technology, stuffing of image signals is performed using pre-calculated control signals, but no consideration is given to the size of the stuffing memory;
There is a problem that the capacity becomes large depending on the value of v.

An object of the present invention is to reduce the size of the staff memory from the conventional
v” (X t X v).

[Means to solve the problem]

In order to achieve the above objective, the least common multiple of the bit period of the input image signal and the bit period of the transmission signal is taken as a unit time, and first, the difference in the number of bits between them, Cx@-xw) bit information, is stored in a memory, After that, this can be achieved by performing simultaneous input and output operations in unit time. Therefore, stuffing is repeated multiple times in the time it takes to complete one stuffing in the conventional example.

[Effect]

The basic configuration of the stuffing device, whose basic conceptual diagram of the present invention is shown in FIG. 1, is the same as that of the conventional example shown in FIG. In FIG. 2, 1 indicates a stuffing device, and a write clock fv is generated using a PLL circuit 4 together with a read clock f.

Let the least common multiple of the bit period of the input image signal and the bit period of the transmission signal be the unit time, and let the number of input bits during that time be Xt. First, the switch 32 is turned to the b side to store information in the staff memory 2. During this period, the staff dummy data generation circuit 31 generates fi
Dummy data is output in synchronization with , then the switch 32 is turned to the a side, and input and output are performed simultaneously in synchronization with fv and ft.

This operation takes the time required to complete one staff operation in the conventional example
This is repeated v times, and the operation of the stuff is controlled by the stuff control circuit 3.

Figure 1 shows the input and output data flow of this operation.Since the number of bits stored in memory in one period is M"xt
The number of bits of the dummy data to be output is 81 and 82 as shown in equations (6a) and (6b).
There are two types.

B t = I N T (M / x v fist x t
) −(6a) Bz=INT (M/Xvxt) +1 ・=(6b) Xv times If the number of times Bl and B2 bit dummy data are output during stuffing is α times and β times, respectively, the output is Since the total number of bits of the dummy data is M x t bits, the following equation holds true.

By solving equation (7), 2
The number of times dummy data of a type of size is output can be obtained. Furthermore, during the period of 12, Xv bits are inputted in synchronization with fv, and Xt bits are outputted in synchronization with ft.

FIG. 4 shows the transition of information I & 1m stored in the stuff memory in the present invention and the conventional example. In this example, as shown in FIG. 5, 82-bit dummy data is first stored β times. Repeat, then set B1 bit dummy data to α
Shows what has been repeated twice.

〔Example〕

Examples of the present invention will be shown below.

(Example 1) Figure 5 shows the data flow when first stuffing using dummy data of Bz bits is performed β times, and then stuffing using 81-bit dummy data is performed α times. The information stored in the staff memory is shown in FIG. 4, and the loading speed is f v "392.
736M b/s read speed f t”397.
In the case of 2Mb/s (5th group of optical digital hierarchy in Japan), if the least common multiple of the input/output bit period is the unit time, the number of input/output bits during that time is
v = 8182 bits, X t = 8275 bits. Furthermore, M=93 bits are stored during the first τ1.

From formula (6) a, b and formula (7), the dummy data size and each number of times are: Bz = 94 bits α = 7715 times,
Bx=95 bits becomes β=467 times.

As shown in FIG. 5, when the dummy data size is 82, the output time is longer than τ1, and as expressed by equation (8), more σ bits than M bits are stored in the stuff memory.

a = (B2/ xt - t 1) skewer xv
...(8) Point A in Figures 4 and 5, that is, B
When 2 is repeated 467 times, the capacity stored in the memory becomes maximum, and this value is 436+93=529 bits. This combination of α and β is an example that uses the most stuff memory, but the 76092
Much smaller than 6 bits. After that, B1 is repeated 7715 times to complete the stuffing. Also, between 12 and 12, f
Input 8182 bits in synchronization with v, and input 8 in synchronization with ft.
Outputs 275 bits.

(Example 2) f v = 392.744 M b / s, f t =
An example is shown for the case of 397.2M b/s, where Xv = 49093 bits, xt ”49650
bit, M=577 bits, Bz:=:563 bits a=33402 times, Bz=564 bits β=15
If you use the method shown in Figure 5, which performs staffing 691 times, A
When time point B1 is repeated 15691 times, the maximum amount of 557+10557 bits is stored in the stuffing memory. This is considerably larger than the basic size of 557 bits, so it can be reduced by using the method shown in FIG. In Figure 6, first Bt is performed 28 times, then Bl is performed 33 times, then BZ is performed 23 times, and L is 49 times.
Repeat the set 681 times to complete the staff.

When the dummy data size is 81, the output time is τ1
The number of bits stored is smaller than M, which is the number of bits initially stored, by δ shown in Equation 9.

δ= (ft Bz/xt)xv −(9)
In the method shown in FIG. 6, the information stored in the staff memory becomes maximum at time A, that is, when B2 has been performed 28 times, Bz has been performed 33 times, and B2 has been performed 23 times.

From equations (8) and (9), this value becomes 557+24 bits, which allows the stuff memory size to be reduced. Furthermore, during the period 12, 49093 bits are input in synchronization with fv, and 49650 bits are output in synchronization with f.

(Example 3) If an image has a unique block configuration, for example, if a synchronization signal is added to each horizontal scanning line and signal processing is performed, in order not to destroy the configuration, bits proportional to the block configuration are added. First store the number in memory. That is, if the block size of the image is XS, then M = x s (X(Xv
). fv=392.736 Mb/s, f
t = 397.2 Mb/s, the number of pixels in the horizontal scanning line is 1456 bits, and when performing 4-phase processing, M = 9.
3X364=33852 bits. Bi=94x3
64. The number of times Bz =95×364 is the same as in Example 1, and if the method of FIG. 5 is used, the maximum amount stored in the stuff memory at time A is 192337 bits. This is 1M
bit field memory (e.g. commercially available 26
2264 x 4 bits) and smaller than the 276977φ64 bits shown in the conventional example.On the other hand, between 12 and 12, 36418182 bits are input in synchronization with fv,
Outputs 36488275 bits in synchronization with ft. Similar to Example 1, this is an example that uses the most staff memory.

(Example 4) An image has a block of 1 horizontal scanning line (364 bits), f = 392.744 M b / s, f t
= 397.2M b / s, M = 577X36
4 bits, B1=563X364 bits, Bz=564
Each number of X364 bits is the same as in Example 2. If the method shown in Fig. 5 is used, a memory with a capacity of 1M or more is required and is not effective.If the method shown in Fig. 6 is used, the maximum amount of memory stored in the staff memory at time A is 577 x 36.
4 + 8693 = 2H1721 bits, 1
It is smaller than M field memory. On the other hand, during τ2, a 64 x 49093 bits are input in synchronization with fv, and 364 x 49650 bits are output in synchronization with ft.

(Example 5) The image has blocks of 364 bits, f v = 3
85.648Mb/s, f t=397.2
When M/s.

x v = 24103 bits x , = 24825 bits, M = 3646722 = 262808 bits, which exceeds the capacity of the 1M field memory.In this case, there is a method to reduce the size of the image block by performing polyphase processing on one horizontal scanning line. Or, for example, set M to 364 skewers 3
It is clear that this can be dealt with by setting the number to 61 and processing it twice.

Throughout Examples 1 to 4, when it is necessary to have a difference of more than a certain value between the read address and the write address of memory 2 in Figure 2, the number of bits of more than a certain value is always stored in the memory before and after stuffing. It is also clear that by controlling in this manner, the same operation as in each embodiment can be performed.

(Effects of the Invention) According to the present invention, by repeatedly performing Xv stuffing in the time required to complete one stuffing in the conventional example, the size of the stuffing memory can be reduced by Xv-(xt
-xv) If the 6 images have their own block size, the effect is that it can be reduced by
This effect is particularly significant from the viewpoint of hardware (memory number jk) feasibility.

[Brief explanation of the drawing]

Fig. 1 is a diagram explaining the principle of the present invention, Fig. 2 is a block diagram showing the basic configuration of the stuffing device, Fig. 3 is a diagram explaining the principle of the conventional example, and Fig. 4 is the stuffing memory of the present invention and the conventional example. 5 and 6 are conceptual explanatory diagrams of the staff according to the embodiment. 1...Staff device, 2...Staff memory, 3
. . . stuff control circuit, 31 . . . dummy data generation circuit, 32 . . . switch, 4 . . . PLL circuit. Prison L Ro ¥ 3 Figure No. 1 Country J-Mori 1.

Claims (1)

[Claims] 1. In a stuffing device that converts the information speed of a digitized image signal to the transmission speed of a transmission path, the least common multiple of the bit periods of the image signal and the transmission signal is taken as a unit time, and the input between them is An image signal stuffing method characterized in that information about the difference in the number of output bits is first stored in a stuffing memory and then stuffing is performed. 2. In item 1 above, if the image signal has a unique block configuration, the information on the block size and the difference in the number of bits per unit time is first stored in the stuffing memory before stuffing is performed. The image signal stuffing method according to claim 1, characterized in that: