JPH0358582A - Packet transmission system, packet transmitter and packet receiver for television signal - Google Patents

Abstract

PURPOSE:To simplify constitution on a reception side by separating a video signal and a sound signal from a signal which is received through a packet network, executing the variable length decoding of the video signal and decoding the sound signal. CONSTITUTION:In order to execute the variable length decoding of the video signal by a variable length decoding circuit 207 on the reception side, when the video signals are successively read out from a common buffer memory 203, where the video signal and sound signal are written, at first, the block of the sound signal appears when the video signals are read out for a fixed picture element corresponding to a unit period. Accordingly, the sound signal can be read out and decoded hereafter from the buffer memory 203 by a fixed clock only for the sample of the sound signal known is advance. Thus, since the buffer memory 203 is made common for the video signal and sound signal, the buffer memory 203 and a control circuit 204 are simplified and the constitution of a receiver is simplified.

Description

[Detailed description of the invention] [Eleventh aspect of the invention] (Field of professional use) This invention provides a wideband ISDN asynchronous transfer mode (AT).
M) Relates to a transmission system for broadcasting television signals by transmitting packets used in networks, etc.

(Conventional Technique) When transmitting a television signal, that is, a video signal and the accompanying audio (No. 6), the encoding signal is
Depending on the formula and transmission method, the image quality at the decoding output end is
The relative delay between the signal and the sound (,; g. , the discrepancy between the image and the audio may cause a sense of incongruity, so a correction method such as fir is required.In addition, this... delay 11! j interval is 7:) encoding system and transmission It became a hole depending on the method (take the direct).

Conventionally, when transmitting a video signal and an audio signal accompanying the same, a four-wire mode transmission network is generally used in which each user is assigned a fixed ray channel. This is so-called synchronous time-division multiplex transmission, and an example of a multiplexing method for television signals in a line mode used for reconnaissance of broadcast television signals that requires a particularly high transmission rate will be described. FIG. 4 shows a block diagram of the transmitting side when using the line mode. The human-powered video signal and audio signal are each sent to an encoder 41 with a fixed output rate.
．． , 42, and the audio signal is further subjected to time-base compression processing by a time-base compression circuit 43 so that it can be multiplexed, for example, during the retrace period of the video signal (see Figure 1-5).

In addition, in general, the signal processing interval ■j7 required for video encoding and the signal processing interval ljt required for audio signal encoding is 5"4,
When video encoding requires more time,
As shown in FIG. 4, a delay correction circuit 44 is used for the purpose of correcting the delay time difference between the encoded video signal and the encoded audio signal. This delay correction circuit 44 may include not only the 10-pair delay time difference on the encoding side, but also extraction of the i-th pair delay time difference occurring on the decoding side, but in any case, as will be described later, It is sufficient to provide a fixed delay between {circle around (2)} and r.

Coded video f. The encoded audio signal from which the difference between the ; signal and the relative delay 111j has been extracted is sent to the fifth
Multiplexed as shown. In the video/audio multiplexed signal shown in FIG. 5, the bright TV area where the video signal V is transmitted and the bright area TA where the audio signal A is transmitted are periodic and constant.

This video/sound 7'! The one multiplexed signal is further multiplexed with signals output from other users within the line network, as shown in FIG. In Figure 6, dl+ d2+ a
3... are the video/audio multiplexed signals shown in FIG. 5, respectively. In line mode, each user is assigned a fixed rate channel. This means that a certain number of time slots are secured within the time division multiplex frame Fn (n-1.2,=-m> in FIG. 6).

6th V(b) shows how d is multiplexed into the second time slot of each jI division multiplex frame Fn.

In the multiplexing process (multiplexing) from (a) to (b) and the reverse processing (demultiplexing) from (b) to (a) in FIG. and voice 1.

There is no factor that changes the difference between the relative lengths of the numbers. In other words, there is no phase-sealing delay time difference between the video signal and the audio signal in the same wire network. From this, the delay sleeve W same path 44 in FIG. 4 encodes (and decodes) the video signal 71'.
and audio signal? It is only necessary to supplement the process 11 of the encoding (and decoding) process 11, and it is sufficient to use a predetermined fixed delay circuit.

On the other hand, ijI'l length as a video encoding method? :) By combining the encoding (machijo rate) and the smoothing buffer, the fixed rate encoder 41 of FIG.
In recent years, methods to increase encoding efficiency have been considered. In this method, the delay time of the delay correction circuit 44 may be determined in consideration of the uniform fixed delay time of variable length encoding.

Next, we will discuss problems when transmitting television signals in packets. Packet networks allow for more flexible and efficient use of available capacity transmission channels. Conventionally, packet networks have rarely been used in the transmission of high-speed broadcasting television signals, etc., because the connection protocols and packet exchange between the packet network and transmitting/receiving terminals are difficult. However, in recent years, as seen in the proposals for asynchronous transfer mode (ATM) for broadband ISDN, the flexibility of packet transmission has been improved by simplifying the above protocols, packet and soto exchange, and using fixed length packets. At the cost of some loss of merit, high-speed processing (exchange/connection) has become possible.

FIG. 7 shows a block diagram of the transmitting side when a packet network is used for transmitting television signals.

The video signal and the accompanying audio signal are respectively encoded by encoders 71 and 72, and further made into packets of an appropriate length by packetization buffers 73 and 74, and then time-packed by a packet multiplexing circuit 75. Divided and multiplexed.

In FIG. 7, even if the input video signal is encoded at a tableable rate, it does not pose a problem in packet transfer, so the encoder 71 performs more efficient nJ variable rate n encoding (
(2) Variable-length encoding). Regarding audio signals, in the case of broadcast television signal transmission, the ratio between the amount of video information and the amount of audio information is originally large, and high quality decoded audio is required, so the encoder 72 uses a fixed rate. encoding is commonly used.

FIG. 8 is a diagram showing an example of multiplexing of video/audio signals in a packetized transmission system, in which fixed periods of input video signals are respectively variable-length encoded, resulting in encoded signals with variable output rates. This indicates that the audio signal remains at a fixed output rate. Therefore, if the packet insertion position of the audio signal is maintained periodically and constant, the delay time difference between the decoded output of the video signal and the audio signal will be the same as that of synchronous time division multiplexing using line mode. Become.

However, within the packet network, the packet position is not preserved, which generally causes fluctuations in the t-objective delay time difference.

FIG. 9 is a diagram illustrating the occurrence of fluctuations within a packet network. In the same figure (a), PP2. . . . are packets generated from a certain user at regular intervals. These packets P,,P2,... are shown in the same figure (b
), it is time-division multiplexed with packets output from other users within the packet network. At this time, the same figure (b)
As is clear from P l + P 2 +...
The order of is preserved, but the pushing positions are not periodic due to asynchronous time-division multiplexing, nor are they at regular intervals.

FIG. 10 is a timing chart showing the arrival of video packets and audio packets at the receiving end when transmitted over a packet network, and it can be seen that fluctuations occur.

In this way, in a packet transmission system, the timing of received packets fluctuates regardless of the encoding method. For this reason, when transmitting television signals in packets, in order to cancel or sufficiently reduce the t0 vs. delay time difference between the video signal and the audio signal, it is necessary to absorb the fluctuations n.
A J-variable D1 noise compensation circuit is required.

Figure 1 shows a diagram of the receiving side when a television signal is transmitted using the conventional transmission system.

In the figure, the video packet transmission system and the audio packet transmission system are independent systems. Therefore, on the transmitting side, video packets and audio packets are multiplexed and sent to bucket 1/network at random timings.

In FIG. 11, a packet network is connected to the receiving end 1, and the arrival of video packets and audio packets is detected by a detection path 2.3. When a packet is detected,
The contents are written to the video and audio buffer memories 4.5, respectively.

Each buffer memory 4.5 is controlled by timing control circuits 6, 7 so as to absorb temporal fluctuations occurring in the packet transmission system (network).

However, is the video signal variable length? } Therefore, the timing control circuit 6 controls the reading of the video buffer memory 4 while identifying the length of each code based on the decoding result of the video r+1 variable length decoding circuit 9.

Image and sound i'! In order to correct the relative delay time difference between the two encoded outputs, a reference timing signal multiplexed in advance into the video signal and the audio signal is used. These reference timing signals are detected by detection circuits 11 and 12, respectively, and a time difference between both timing signals is determined by a delay detection circuit 13. Then, the delay detection circuit 13 controls the delay of the variable delay circuit 8 provided between the audio buffer memory 5 and the audio fixed-length decoding circuit 0, and the video decoding circuit 8 is output to the output terminal 14.16. The difference between the relative delays (2) and r between the output and the audio decoded output is compensated for.

The configuration shown in FIG. 11 requires buffer memories 4 and 5 for video and audio signals, and timing control circuits 6 and 7 to control them, and furthermore, it is necessary to provide buffer memories 4 and 5 for video and audio signals, and timing control circuits 6 and 7 to control them. A capsular delay circuit 8 for correcting including food intake due to fluctuation, a reference timing signal detection circuit 11 and 12 for controlling this, and a delay detection circuit 1
3, the circuit scale becomes extremely large.

The wide-area ISDN ATM network is a simplified packet network in order to handle high-speed signals. Is there any consideration being given to a system that uses this ATM network to distribute television signals to each home?As mentioned above, if the size of the receiving side becomes large, the financial burden on the user will be heavy, and it will not become popular. It also becomes a hindrance.

(Problems to be Solved by the Invention) As mentioned above, in high-speed packet transmission for broadcasting television F5, etc., conventional video packets and audio packets are regarded as independent systems. If the transmission method is adopted, it is necessary to provide a buffer memory and four control circuits for each image signal and audio signal on the receiving side, and furthermore, it is necessary to set up a buffer memory and four control circuits for each image signal and audio signal, and also to reduce the delay time difference between the video signal and the audio signal due to temporal fluctuations. Since a variable delay circuit for including and correcting the delay circuit and a circuit for controlling it are required, there is a problem in that the scale of the circuit becomes extremely large.

SUMMARY OF THE INVENTION An object of the present invention is to provide a television signal packet transmission system that can simplify the configuration of a receiving side, and a packet transmitter and a packet receiver in this transmission system.

[Structure of the Invention] (Means for Resolving the Problem) In order to achieve the above-mentioned problem, the television signal packet transmission system by Masaaki Moto transmits a video signal for a predetermined unit period on the transmitting side and The audio for the accompanying li period (m codes are each = 1 variable-length encoded and encoded, divided into multiple video packets and audio packets, and then these multiple video packets and audio packets are combined into a continuous stream) is sent to the packet network, and received via the packet network on the receiving FCI side (≦ signal to video 1.; After separating the signal and audio signal, variable length decoding is performed on the video signal, and decoding is performed on the audio signal. It is characterized by the fact that

The packet transmitting device of the present invention also includes means for variable-length encoding a video signal for a predetermined 111th bright interval and dividing it into a plurality of video packets, and means for encoding an audio signal for the rat interval and dividing it into a plurality of video packets. It is characterized by comprising means for dividing the plurality of audio packets into audio packets, and means for combining the plurality of video packets and the plurality of audio packets and continuously sending them out to a packet network.

Furthermore, Masaaki Moto's packet receiving device receives a video signal obtained by dividing a variable-length n coded video signal for a predetermined unit period into carrier number packets, and encoded sound h for a unit period.
Common means for receiving a set of audio signals obtained by dividing the signal into packets of a number of stages via a packet network, and for storing the video signal and audio signal received by the 1,000 stages for at least a period of about 11j. a buffer memory, a means for separating a video signal and a received f≦ signal from an output signal of the buffer memory, a means for variable length decoding of the video signal separated by the separating means, and a means for decoding the video signal separated by the separating means, It is characterized by comprising an f stage for decoding the audio signal.

(Operation) In the tree invention, variable length encoded video f. ．． , and the coded audio signal are combined into blocks, packet-multiplexed, and sent to the packet network. As a result, the order of each packet within a block at the receiving end is the same as that at the transmitting end.

In order to perform variable length post-coding of the video signal on the receiving side, the video and audio signals are stored in a common buffer memory.
First, when the video signal is sequentially read out from the buffer memory, the time point when a certain number of pixels corresponding to a unit period have been read out becomes an audio signal section. From then on, the audio signal can be read out from the buffer memory using a fixed clock and decoded by the number of samples of the audio signal known in advance.

In this way, by using a common buffer memory for video signals and audio signals, the buffer memory and its control circuit become simpler. Further, temporal fluctuations in the video signal and audio signal that occur in the packet network, including fluctuations in relative delay time differences, are absorbed in the process of reading from the buffer memory. This eliminates the need for complex correction using a variable delay circuit, and the relative delay l'lJ between the video signal and the audio signal
Even when supplementing the difference between j, it is necessary to consider only the fixed delay time difference in the same way as in four-wire mode transmission.

(Example) Hereinafter, length embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a major embodiment of a packet transmission device in a television signal packet transmission system according to the present invention.

In FIG. 1, a digital video signal (for example, 8 bits/pixel) input to a video input terminal 101 is a video nI
The variable length encoder 103 performs variable length encoding at a predetermined bit rate or less (for example, 1 bit/pixel).

On the other hand, the digital audio signal input to the audio input terminal 102 is fixed-length encoded by the audio fixed-length encoding circuit 204, and then converted into a video signal by the time axis j compression circuit 105 for a fixed unit period, for example, for each field. The patent axis is compressed to fall within the vertical planking period of . Here, the field + 1t processing and the special axis compression rate (in the above example, it means the compression rate that allows the audio signal to be inserted into the vertical blanking period (c) of the video signal) can be easily explained.
It was used to make it 1. However, it is not necessarily necessary to do it this way.

The variable-length encoded video signal output from the variable-length encoder circuit 103 and the time-axis compressed fixed-length 7:) encoded audio signal output from the time-axis compression circuit 105 are respectively input to the video packetization circuit 106. After being divided and horizontally divided into a plurality of packets in field units by the audio packetization circuit 107, the signals are time-division multiplexed by the selector 108,
The data is sent from the transmitting end 109 to a packet network (not shown).

At this time, at a certain operating clock, the output of the video (゛℃packetization circuit 1 ()6) or the audio packet is
-C with a gap in the vertical planking light so that people can do it
If there is, it is difficult to multiplex video and audio signals,
This can be realized using a simple selector 108 as in this example.

Figure 2 shows Bakeso l, the television color issue by Masaaki Moto.
Packet receiver IZ in transmission system! This is a block diagram showing the i, I, and 7 configurations of one embodiment of the 4", and has a configuration corresponding to the 1. side of the first garden.

In the second country, received f. The end 201 is connected to a packet network. The reception signal received by the receiving end 201 is input to the packet detection channel 202, and one of the aspiration packets is detected. When a desired packet is detected, the video signal and audio signal of the desired packet are sequentially written into the buffer memory 203.

The video signal and audio signal read from the puffer memory 203 are sent to the video/audio separation path 205 to be divided into two images according to the video/audio identification code added to the video packet and audio packet in advance. It is divided into '' and Otono 4i signals. Among the outputs of the video/audio separation circuit 205, the video signal output is manually input to the video 11 Onnachoichi number circuit 207 and routed.

Similar to the same length decoding circuit, long-length encoding 11j that supports video
Also in path 207, the throughput at pixel 111 in human output is the same. Therefore, 1 from the buffer memory 203
The time required to output data for field pixels is one field bright interval. This bathofa memory 203
The reading control from the image 1 image nJ variable length decoding circuit 207
This is done by a feedback loop from the buffer memory 203 to the buffer memory 203 via the huffer memory timing control path 204. In other words, the buffer memory timing control circuit 204 identifies the length of each symbol 7a and then identifies the length of each symbol 7a and 1b''. Controls the dispensing operation.

When the direct blanking interval arrives when the video signal is being signaled, an audio signal starts to be output from the buffer memory 203 in place of the video signal. Since the audio signal is fixed length encoded,
Unlike video signals, there is no need for special control when reading from the buffer memory 203, and it is sufficient to simply read using a clock signal with a fixed frequency.

The audio signal thus read out from the buffer memory 203 is separated from the video signal by the video/audio separation circuit 205, further passes through the time expansion circuit 20b, and then decoded by the audio factor decoding circuit 208.

Figure 3 shows the operation of the two consecutive transmitting and receiving sides! This is a timing diagram.

(a) shows digital video input to the video input terminal 101, sampled at clock frequency f5, and 8
Assume that it is quantized into bits.

(1)) shows the digital voice input to the voice input terminal 102, and for the sake of simplicity, it is assumed that f is sampled at the clock frequency fA, which is the same as the field reference.

(C) shows that the digital voice in (b) is fixed-length coded and then time-axis compressed at field f11.
It shows l child.

(d) shows a field sound in which the video input in (a) is band-compressed to a bit rate of 1 bit/pixel or less at field jl1. The sound of this place is Ei 1! Human power 1 bit No. 4 to 8
・If you packetize it with a clock called fs (for simplicity,
(excluding the packet header ``.9''), it is always empty for the vertical planking period. uf = means that during this vertical planking period, the audio packet is processed as shown in (C). Multiplexing can be performed.

(1″) is the + of the receiving Ig bucket 1 at the receiving end 201.
One child is shown. As described above, in a packet network, the order of packets at the time of transmission 1 is preserved, or the difference between 0 and lj between each packet is not preserved, and fluctuations occur between arrivals 11j. Out of these packets, only the necessary packets are extracted by the packet detection circuit 202, written to the buffer memory 203, and read from the buffer memory 203 by the buffer memory timing control circuit 204 as described above. The output of 203 is the third d (
As shown in g), only the 0.1j fluctuation of the audio signal is absorbed, resulting in burst data synchronized with the field frequency. However, in this case, it is assumed that the Huffer memory 203 has sufficient capacity to absorb the particular fluctuation.

In this way, in the process of writing and reading video and audio signals to and from the common buffer memory 203, temporal fluctuations occurring in the packet network can be easily corrected.

Furthermore, regarding the delay time difference between the video signal and the audio signal, only the known fixed delay need be considered, as in the case of the line mode. Specifically, for example, the difference between the writing and reading timings of the digital memory forming the time axis compression circuit 105 in FIG. In addition, the difference between the writing and reading timings of the digital memory that constitutes the 0 and 'j axis expansion path 206 in FIG. The value may be selected to compensate for the difference in processing time of the encoding circuit 208. Of course, without using the delay of the time axis compression circuit 1υ5 or the time axis expansion port 206 to correct the relative delay time,
Another fixed delay circuit may be provided, and the video signal and audio 1. ;The difference between the relative delays njf of the signals is the same on the transmitting side or only on the receiving side.
! It is also possible to correct it to j.

[Effects of the Invention] According to the present invention, problems of delay time differences and temporal fluctuations between video and audio signals that required complex h[i TE processing in the past in transmitting television signals using a packet network] can be solved. is the same as in the conventional line mode, and the configuration of the receiving device can be extremely simplified. That is, by sharing the buffer memory on the receiving side and eliminating the need for a variable delay circuit and its control circuit, it is possible to significantly reduce the circuit scale. Therefore, broadband IS
It is possible to reduce the cost of receiving terminals, which is one of the most important items when constructing a television signal distribution system using DN, and has a great commercial effect.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of a packet transmitting device in a television signal packet transfer system according to the present invention, FIG. 2 is a block diagram showing an embodiment of a packet receiving fJ device, and the third country is the same. FIG. 4 is a block diagram of the transmitting side multiplexing circuit for video/audio f. ; Figure 6 is a timing diagram to explain multiplexing within the same wire network; Figure 7 is a timing diagram to explain multiplexing in the same wire network; Figure 7 is a diagram of the multiplexing circuit on the transmitting side of video and audio signals in a packet transmission system. 8 is a diagram showing an example of multiplexing of video and audio signals in the packet (TG transmission system), FIG. 9 is an explanatory diagram of multiplexing within the packet network, and FIG. 10 is a diagram showing video and audio signals output from the Bagut network. 11 is a block diagram of a conventional packet receiving circuit. 103...Video edible length encoding circuit, 104...Audio fixed length encoding circuit, 105...・Time axis compression circuit, 10
6... Video packetization circuit, 207... Audio packetization circuit, 108... Selector (video/audio packet multiplexing path), 202... Packet/soto detection (b) path,
203...Buffer memory, 204...Buffer memory/timing control path, 205...Video/audio separation path, 206...11, lj axis expansion circuit, 207
...Video 6■Variable length decoding path, 208...Audio fixed length decoding path.

Claims (3)

[Claims] (1) In a television signal packet transmission system that packet-transmits a television signal consisting of a variable-length coded video signal and an accompanying audio signal, the video signal for a predetermined unit period is variable-length coded. means for encoding the audio signal for the unit period and dividing it into a plurality of audio packets, and continuously combining the plurality of video packets and the plurality of audio packets into sets. means for transmitting to a packet network; means for separating a video signal and an audio signal from a signal received via the packet network; means for variable length decoding the video signal separated by the separating means; and the separating means. 1. A television signal packet transmission system, comprising means for decoding an audio signal separated by the means. (2) In a packet transmission device in a television signal packet transmission system that transmits a television signal consisting of a variable-length coded video signal and an accompanying audio signal in packets, the video signal for a predetermined unit period is transmitted. means for variable-length encoding and dividing into a plurality of video packets; means for encoding the audio signal for the unit period and dividing it into a plurality of audio packets; and combining the plurality of video packets and the plurality of audio packets. 1. A packet transmission apparatus for television signals, comprising means for continuously transmitting a television signal to a packet network. (3) In a packet receiving device in a television signal packet transmission system that transmits a television signal consisting of a variable length encoded video signal and an accompanying audio signal in packets, a predetermined variable length encoded unit. means for receiving, via a packet network, a set of a video signal formed by dividing a video signal for a period into a plurality of packets and an audio signal formed by dividing a coded audio signal for a unit period into a plurality of packets; , a common buffer memory for temporarily storing the video signal and audio signal received by the means for at least the unit period; means for separating the video signal and the received signal from the output signal of the buffer memory; and separation by the separating means. What is claimed is: 1. A packet receiving apparatus for a television signal, comprising: means for variable length decoding of the video signal separated by the separating means; and means for decoding the audio signal separated by the separating means.