JPH11262001A - Method for calculating time stamp value in encoding transmission system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To dispense with a memory and a related circuit and to facilitate a transmission system by calculating a first time stamp TS value, based on delay time information of a video stream and calculating the TS value based on the obtained TS value and the period of a video signal afterward. SOLUTION: A current presentation time stamp PTS obtained by the prescribed calculation of the TS value by a reset signal 202, based on delay time information of the video stream, is adopted as the reference PTS of a succeeding processing, and then a picture interval N until a picture to be obtained as the TS value is obtained by setting the number of pictures included in a present packet. The present PTS is obtained through the use of a last time PTS and N in a succeeding packet processing, and the same processing is repeated after thereafter. Calculation is executed based on the obtained last time TS value and a picture period 201 inputted to an arithmetic part 200, so that the delay time information is unnecessitated after the first TS value and a large capacity memory and a data transfer control part with high precision are dispensed with.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method of calculating a time stamp value required for encoding and transmitting an audio signal and a video signal.

[0002]

2. Description of the Related Art Generally, in an encoding and transmitting apparatus for multiplexing and transmitting an audio signal and a video signal, a decoding processing section for each of audio and video synchronizes the two.
Need to output. This is to synchronize the image and audio, for example, the video decoding processing unit,
When the face of the person generating the sound "a" is being output, the output of the audio decoding processing unit must output the sound "a". As described above, in order to enable synchronization between images and audio, MPEG2 (Moving Picture Imaging), which is an international standard encoding method, is used.
Age Coding Experts Group Phase 2), MPEG1 and the like use output timing information called a time stamp (hereinafter referred to as TS).
The TS includes a presentation time stamp (prese
ntation time stamp: hereinafter, referred to as PTS) and a decoding time stamp (hereinafter, referred to as PTS).
Basically, synchronization can be achieved by outputting an audio signal or a video signal that has been subjected to a decoding process in accordance with the time designated by the PTS. Generally, a stream obtained by multiplexing data in which an audio signal is encoded (hereinafter, referred to as an audio stream) and data in which a video signal is encoded (hereinafter, referred to as a video stream) is referred to as a system stream. Audio and video TSs are added to the system stream. Note that these TSs are sync words (codes at certain words in the audio stream) in the case of audio signals, and picture start codes (picture start codes in the video stream in the case of video signals. It is added on the basis of a code) part indicating a picture delimiter.

FIG. 2 is a schematic block diagram of a conventional coded transmission apparatus. The coded transmission device is roughly divided into
It is composed of a transmitting side, a transmission path, and a receiving side. In FIG. 2, on the transmission side, an input audio signal 50 is encoded by an audio encoding processing unit 51 to become an audio stream 52 and output to a multiplexer unit 56.
The input video signal 53 is supplied to a video encoding processing unit 54.
, And becomes a video stream 55, which is output to the multiplexer unit 56. The audio stream 52 and the video stream 5 input to the multiplexer unit 56
5 are multiplexed into a system stream 61 and output to a transmission line 62. The multiplexer section 56 that performs multiplexing will be described. The multiplexer section 56 includes a system time clock (hereinafter referred to as S).
TC) generating unit 57, audio packetizing unit 5
9, a video packetizing unit 60 and a switch unit 105. STC stands for MPEG2, MPEG
1 and the like, and is like a common clock in a signal processing system as shown in FIG.

[0004] The audio packetizing section 59 divides the input audio stream 52 into fixed lengths, forms packets, and outputs the data 103 to the switch section 105. Then, it is selected by the switch unit 105 and multiplexed as a system stream 61 in packet units. At this time, if there is a sync word in the data length of the audio stream 52 to be packetized, a TS is added to the header portion of the packet. Note that this TS is obtained based on the STC data 58 of the STC generator 57. On the other hand, the video packetizing unit 60 also divides the input video stream 55 into fixed lengths, forms packets, and outputs the data 104 to the switch unit 105. Then, it is selected by the switch unit 105 and multiplexed as a system stream 61 in packet units. At this time, if the picture start code is included in the data length of the packetized video stream 55, a TS is added to the header of the packet. Note that this TS is the STC data 58 of the STC generator 57.
It is based on. Then, the STC data 58 is periodically selected by the switch unit 105 so that the STC can be reproduced on the receiving side, and the system stream 61 is reproduced.
Multiplexed as

On the receiving side, the system stream 63 transmitted through the transmission line 62 is separated by the demultiplexer 64, and the audio stream 67 and the audio TS 112, and the video stream 70 and the video TS 1
13. STC data 110 is output. Further, the demultiplexer unit 64 for performing this separation will be described.
6, STC playback unit 111, audio packet analysis unit 6
5, a video packet analysis unit 66. The demultiplexer 64 uses the switch 106 to convert audio packet data 107 to the audio packet analyzer 65, video packet data 109 to the video packet analyzer 66, and STC from the input system stream 63. The data 108 is output to the STC reproducing unit 111. Audio packet analyzer 65
Analyzes the input audio packet data 107, and analyzes the audio stream 67 and the audio TS 112.
And outputs it to the audio decoding processing unit 68. On the other hand, the video packet analysis unit 66 analyzes the input video packet data 109, separates the video stream into a video stream 70 and a video TS 113, and outputs the video stream 70 and the video TS 113 to the video decoding processing unit 71.

The STC reproducing unit 111 reproduces the STC data 110 from the input STC data 108 so as to always output the same output as the STC data 58 output from the STC generating unit 57 on the transmission side, and performs audio decoding. Output to the decoding processing unit 68 and the video decoding processing unit 71. Further, the audio decoding processing unit 68 converts the input audio stream 67 into the audio TS 112 and the STC
The decoding is performed with reference to the data 110, and the output audio signal 69 is output at a time according to the TS value. The video decoding processing unit 71 decodes the input video stream 70 with reference to the video TS 113 and the STC data 110, and outputs an output video signal 72 at a time according to the TS value. Next, the output operation by the TS in each decoding processing unit will be described more specifically. The output operation by the TS is performed by the audio decoding processing unit 68 and the video decoding processing unit 71 by using the T
This is realized by outputting the corresponding data portion when the value of the PTS of S matches the value of the STC data.
For example, if the value of the PTS corresponding to a certain sync word An of the audio stream 67 is PTS (An),
When the value of the STC data 110 becomes the same as PTS (An), the audio decoding processing unit 68 sets the sync word A
Output the n-part decoded data.

Similarly, the value of the PTS corresponding to a certain picture start code Vn of the video stream 70 is
If S (Vn), the value of STC data 110 is PTS
When it becomes the same as (Vn), the video decoding processing unit 71
Outputs the decoded data of the picture start code Vn. Here, in the encoding transmission apparatus shown in FIG. 2, after the input audio signal 50 is input to the audio encoding processing unit 51 on the transmission side, the output audio signal 69 is output from the audio decoding processing unit 68 on the receiving side. It is assumed that the time until the start is ta. Similarly, it is assumed that the time from when the input video signal 53 is input to the video encoding processing unit 54 on the transmitting side to when the output video signal 72 is output from the video decoding processing unit 71 on the receiving side is tv. I do. In this case, in order to synchronize audio and video, the time of each transmission processing system is set so that the time from the input to the output becomes ta = tv, and the respective PTSs are set based on the time. What is necessary is to set the value of.

Next, an audio packetizing section 59 and a video packetizing section 60 constituting the multiplexer 56 are described.
Will be described. First, it is assumed that, of the time ta from the input to the output of the audio signal, the time from the input to the multiplexer unit 56 is ta1, and the remaining time is ta2. Here, the multiplexer unit 5
6, there is a certain sync word An in the data length to be packetized of the audio stream 52 input to S6, and when the TS is added to the header portion of the packet, S
If the value of the TC data 58 is STC2 (An),
PTS (An) can be obtained as follows. PTS (An) = STC2 (An) + ta2 Similarly, for video, of the time tv from the input to the output of the video signal, the time until the video signal is input to the multiplexer 56 is tv1, and the remaining time is tv2. Suppose there is. Here, there is a certain picture start code Vn in the data length to be packetized of the video stream 55 input to the multiplexer section 56, and the value of the STC data 58 when adding TS to the header portion of the packet is STC2 ( Vn), PTS (Vn) could be determined as follows. PTS (Vn) = STC2 (Vn) + tv2 However, in the case of video, tv1 varies depending on the encoding state of the video signal, so that tv2 also varies. Because tv is always fixed, tv2 =
tv-tv1.

Therefore, there is a problem that tv2 to be added to STC2 (Vn) in the above equation cannot be set in advance, and it is impossible to obtain PTS (Vn) as in the case of audio. The main causes of the change of tv1 are the following two. (1) Fluctuation due to change in picture unit (2) Fluctuation due to variable length coding In addition, since there is a change in picture unit due to the above cause (1), DTS also needs to be obtained.

Here, the replacement in units of pictures will be described with reference to the schematic diagram of FIG. In FIG. 3, the delay time related to the replacement in units of pictures is considered, but other processing times and the like are omitted. 3A shows an input video signal 53 input to a video encoding processing unit 54, and FIG.
The video stream 55 (or the video stream 70 output from the video packet analysis unit 66) and (c) show the output video signal 72 output from the video decoding processing unit 71. Symbols I, P, and B described in any of these signals are MPE
It shows a picture coding type (picture coding type) defined in G2, MPEG1, and the like.

In the example of FIG. 3, the picture interval of the I or P picture is three pictures, and the picture interval of the I or P picture is generally called an M value (in the case of FIG. 3, M value = 3). The numbers in parentheses at the top of each signal indicate the temporal reference (t
emporal reference), which is the input video signal 5
3 shows the order of pictures. The replacement in picture units is performed as follows. First, in the video encoding processing unit 54, the input video signal 53 shown in (a) is sequentially delayed by inserting only a B picture after a subsequent I or P picture, inserted and the order of pictures is changed, and (b) The state of the video stream 55 shown in FIG. Note that this exchange state does not change even in the video stream 70. Next, the video decoding processing unit 71
, On the contrary, the I and P pictures of the video stream 70 are sequentially delayed and inserted after the succeeding consecutive B pictures, and are returned to the original order, and the video output signal 72 shown in FIG.
And the original order. Returning the order by the video decoding processing unit 71 in this way is called reordering.

Next, the relationship between PTS and DTS in reordering in the video decoding processing section 71 will be described with reference to FIG. The video stream 70 input to the video decoding processing unit 71 is stored in a video decoding buffer 73 to absorb a variation on the time axis due to the variable length encoding, and then converted to a video decoding circuit as a video stream 74. 75
Output to The video decoding circuit 75 decodes the input video stream 74 into data 76. Here, the B picture portion of the data 76 is output as the output video signal 72 as it is. On the other hand, the I or P picture portion of the data 76 is delayed by the video reorder buffer 77 to become data 78, and is output as the video output signal 72. This is switched and selected by the switch unit 79. Also, the video reorder buffer 77
Is (M value × picture cycle). By the means of the video decoding processing section 71, the order of the picture unit can be restored.

In FIG. 4, PTS indicates the output time of the output video signal 72 output from the video decoding processing unit 71. DTS indicates the output time of the video stream 74 output from the video decoding buffer 73. It indicates the time that the output has. In MPEG2 and MPEG1,
In the definition of TS, the processing time of the video decoding circuit 75 is assumed to be zero. Therefore, in the case of a B picture,
DTS = PTS. That is, I and P pictures: PTS and DTS are added. B picture: only PTS is added. (DTS = PT
The above is related to the cause (1) of variation of tv1.
As described above, the amount of the change can be obtained if the M value, the picture coding type, and the picture period are known.

FIG. 6 shows an example of a configuration in which the picture start code, temporal reference, and picture coding type used in the above description are added to the video stream 55 in FIG. This is an example in MPEG2. As shown in FIG. 6, the input video signal 53
The code data portion corresponding to each frame or each field in FIG. 2 is called a picture layer and starts with a picture start code which is a fixed value of 32 bits. Following this code, there is a 10-bit temporal reference, and there is also a 3-bit picture coding type. After that, 16-bit video delay time control information (video buffering verif
ier delay: vbv_delay), and the actual code data follows the header. Here, the video delay time control information is the above-mentioned tv1
(2). A buffer in the video decoding processing unit necessary for receiving and decoding variable-length coded data, that is, partially variable-length coded data at a constant rate as its average rate.
This is information indicating a delay time in the (video decoding buffer 73 in FIG. 4).

Therefore, by controlling the reading of the buffer based on the video delay time control information,
Buffer underflow and overflow can be avoided and decoding can be performed continuously. That is, the video delay time control information is a time for absorbing the fluctuation of tv1 that fluctuates due to the variable length coding on the decoding side. Note that the video delay time control information is generated and added inside the video encoding processing unit 54. At this time, in general, each picture unit is obtained from the capacity of a buffer for the purpose in the video decoding processing unit 71 on the receiving side, the code amount as a result of compression for each picture, and the average rate. Is calculated.
In MPEG2, the video delay time control information is set to vbv_delay
And the buffer is called a VBV buffer. Since this information depends on the code length of the variable-length coding, it differs for each picture.

Generally, the video delay time control information is effective when the video stream 55 at a constant rate shown in FIG. 2 is directly decoded by a video decoding processing unit. If the stream is converted to a stream at a different rate from the stream or converted to a stream shifted in the time axis direction due to time division multiplexing, accuracy is lost according to the content of the conversion.
Therefore, at the stage of the multiplexed system stream, the TS is used instead of the video delay time control information. That is, if reading control of the buffer (video decoding buffer 73) in the video decoding processing unit is performed based on the TS, it is guaranteed that the buffer will not underflow or overflow, and continuous decoding can be performed. Will be possible. As described above, the TS has not only a function of synchronizing signals to be multiplexed but also a function of avoiding buffer underflow and overflow. For this reason, a TS is required even when the system stream is a single signal (for example, video only).

Here, a description will be given of a method of calculating a TS value in the prior art. In the prior art, as described above, since there is a change in tv1, instead of using STC2 (Vn), STC1 (Vn) shown in FIG. 2 is used. This is because the time corresponding to the part corresponding to the picture start code part of the video stream 55, ie, the head data of each picture,
This is because PTS (Vn) can be obtained by adding TC1 (Vn) to the time tv from input to output. As described above, tv is always fixed. However, the video encoding processing unit 54 requires a certain amount of processing time (several tens of milliseconds) for signal processing. Therefore, until the STC1 (Vn) value is actually used for calculating the TS value, Must be stored. Furthermore, since the video coding processing unit 54 has the above-described replacement in units of pictures, it is necessary to take this into account when using the stored STC1 (Vn). Further, in the control of the memory, a write control unit that operates at a fixed cycle depending on the input of the head data (frame) of a picture of the video encoding processing unit 54 and a video stream 55 that has been Depending on the input of the picture start code,
The read control units that operate not at a fixed cycle,
Because of the asynchronous relationship, means for avoiding conflict between the occurrence of read and write timing is always necessary. Therefore, the processing of the video packetizing unit 60 is complicated.

In such a conventional coded transmission apparatus, the multiplexer on the transmitting side holds the value of the STC (system time clock) and holds the held STC (system time clock).
Since a delay time of the order of several tens of milliseconds occurs before the data is used for calculating a TS (time stamp) value, a memory for STC is required. In addition, this memory must be accessed in a complicated manner in consideration of the switching of picture units. In the international standard MPEG2 and MPEG1, the bit length of the STC used for the TS is used. Is 33 bits, so that the number of memories increases, and in addition, since the relationship between writing and reading with respect to the memory is asynchronous, there is a problem that an adjustment circuit for competing for those timings becomes indispensable. .

As a second conventional technique for solving the above-mentioned problem, TS based on video delay time control information (vbv_delay) in a video stream output by a video encoding processing unit is used.
By calculating the value, a memory having a wide bit width of 33 bits, a related address generation unit, and the like, which are conventionally required, and an adjustment circuit when writing and reading occur simultaneously are not required. There is a method of calculating the TS value. The time stamp value calculation method in the second conventional coded transmission system compresses and encodes at least one signal including at least a video signal, divides the compressed coded signal into a predetermined data length, and In an encoding transmission system in which a header such as a stamp value is added to generate and transmit one stream, a time stamp value of the video signal added to the header is converted to video decoding added to the stream. The calculation is performed based on the video delay time control information corresponding to the buffer delay time of the processing unit. Further, a time stamp value calculating method according to a second prior art is a signal processing method in which the time stamp value of the video signal added in the header is converted into the one stream in which the compression-encoded video signal is generated. Time information input to the video stream, video delay time control information corresponding to a buffer delay time of a video decoding processing unit added to the video stream, a predetermined divided data length of the compression-encoded signal, and the video The calculation is performed by adding to fixed time information determined according to the signal encoding mode.

The coded transmission apparatus according to the second prior art will be described with reference to FIG. 2 using FIG. FIG.
3 shows a configuration of a part relating to a calculation of a TS value in a video packetizing unit 60 of a multiplexer unit 56 included in the encoding transmission device shown in FIG. The STC data 58 output from the STC generator 57 is input to the TS calculator 116. A video stream 55 in which the input video signal 53 is data encoded by the video encoding processing unit 54 is provided with a picture start code detection unit 86, a picture coding type detection unit 87, and a video delay time control information (vbv_delay) detection unit. 115. The picture stream code validity signal 82 output from the video encoding processing unit 54 is also input to the picture start code detection unit 86. On the other hand, data 89 which is an output of the picture start code detection unit 86 and data 9 which is an output of the picture coding type detection unit 87
3. Data 117 output from the video delay time control information detection unit 115 is input to the TS calculation unit 116, respectively. The data 89 output from the picture start code detection unit 86 is also input to the picture coding type detection unit 87 and the video delay time control information detection unit 115. Further, the M-value parameter 99, which is information indicating the encoding mode, and the delay parameter 100, which is information on the fixed offset delay amount, are input to the TS calculation unit 116. Then, the data 97 and the control signal 98 are output from the TS operation unit 116 to the buffer 101. From the buffer 101, data 104
Is output to the switch unit 105 (see FIG. 2).

Next, this operation will be described. In the conventional encoded transmission apparatus, the video encoding processing unit 5
The picture start code in the data length to be packetized of the video stream 55 output from the video stream 55 is input to the multiplexer 56 at the time (STC2).
(Vn) value) is used for calculating the TS value. However, since tv2 shown in FIG. 2 fluctuates, information called video delay time control information constituting the video stream 55 is used. Next, FIG. 7 shows the relationship between the video delay time control information and the processing time of the coded transmission device. Tv shown in FIG.
As described above, after the input video signal 53 is input to the video encoding processing unit 54, the video decoding processing unit 71
The time until the output video signal 72 is output. This tv is fixed. In this tv, the time from when the input video signal 53 is processed by the video encoding processing unit 54 to when it is input to the multiplexer unit 118 is represented by tv1
The time until the video is output from the remaining video decoding unit 71 is tv2. The times tv1 and tv2 vary as described above. However, tv = tv1 + tv2 becomes
Always fixed. Therefore, the two causes of the aforementioned tv1 variation are absorbed at tv2. Here, the time of tv2 is further divided into tv3, tv4 and tv5. t
v3 is the time from input to the multiplexer 118 to input to the video decoding processor 71, tv4
Is the delay time in the video decoding buffer 73 in the video decoding processing unit 71, and tv5 is the video decoding processing unit 71
Is the delay time in the video reorder buffer 77 in FIG.

The video delay time control information (vbv_buffer) is
Video decoding buffer 73 in video decoding processing section 71
Tv4, which is the delay time at
Reference numeral 4 denotes a fluctuation, which is a fluctuation for absorbing the cause (2) of the fluctuation of tv1. Further, tv5 also fluctuates, and this fluctuation is a fluctuation for absorbing the fluctuation cause (1) of tv1 and corresponds to the above-described reorder time. As described above, the fluctuating tv5 depends on the M value, the picture period, and the picture coding type, but the M value and the picture period are fixed as initialization parameters of the system. Therefore, dynamic fluctuations during encoding depend on the picture coding type. The remaining tv3 depends on factors such as the multiplexed packet length and the transmission path coding, and is a fixed value. Therefore, in each picture, the picture coding type, the video delay time control information, and the time (ST
The TS value can be calculated based on C2 (Vn).

Here, a specific operation of the TS value calculation will be described with reference to FIG. First, the picture start code detection unit 86 recognizes that the video stream 55 has become valid from the input video stream valid signal 82 and detects a picture start code in the valid video stream 55. When the picture start code is detected, the data 89 is output to the TS calculation unit 116, the picture coding type detection unit 87, and the video delay time control information detection unit 115. Triggered by the detection of the picture start code, the picture coding type detection section 87 and the video delay time control information detection section 115 detect the picture coding type and video delay time control information,
3. The data 117 is output to the TS calculation unit 116, respectively. Here, upon detection of the picture start code based on the data 89, the TS calculation unit 116 holds the STC data 58 from the STC generation unit 57, and
3. The TS is calculated based on the data 117, the M value parameter 99, and the delay parameter 100.

First, the picture coding type is determined based on the data 93, and then the TS is calculated for each picture coding type as follows. [In case of I or P picture] PTS = STC data 58 + delay parameter 100 + data 117+ (M value parameter 99 × picture cycle) DTS = STC data 58 + delay parameter 100 + data 117 [In case of B picture] PTS = STC data 58 + delay parameter 100 + data 117 In the above equation, the STC data 58 is the value held at the time of detection of the picture start code,
C2 (Vn), the delay parameter 100 is a fixed value corresponding to tv3 shown in FIG.
The value of the video delay time control information, which corresponds to tv4 shown in FIG. Also, (M value parameter 99 × picture cycle)
Corresponds to the delay time (tv5) by the video reorder buffer 77 shown in FIG. tv5 is zero in the case of a B picture because it does not pass through the video reorder buffer.

Normally, in the initial value setting of the system, the M value and the picture period are set to fixed values.

The above equation is for a video stream containing B pictures (M> 1), but when no B picture is contained (M = 1), video reorder processing is performed for I and P pictures. Becomes unnecessary. Therefore, the delay time tv5 due to the video reorder buffer 77 in the video decoding processing unit 71 is zero (PTS = DTS). Therefore, in this case, the TS is calculated by the following equation. [In the case of an I or P picture] PTS = STC data 58 + delay parameter 100 + data 117
The data is written to the buffer 101 by the control signal 98.
Thereafter, at the TS output timing of the video packet header, the TS value written in the buffer 101 is output as data 104. Therefore, in each picture, the TS value can be calculated based on the picture coding type, the video delay time control information, and the time (STC2 (Vn)) input to the multiplexer 118. As described above, when the picture start code is input to the multiplexer unit 118, the STC data 58 (S
(TC2 (Vn) value), and the TS value can be calculated immediately based on the picture coding type and the video delay time control information. Therefore, the memory for STC can be eliminated. In addition, a complicated circuit for memory control can be omitted.

In the second conventional example, the TS is basically calculated for all the pictures based on the STC value at the time when the picture start code is inputted to the multiplexer section and the video delay time control information. Has become.
However, this implementation has a severe restriction that it must be input to the multiplexer section at an accurate average rate. The video delay time control information is information that is significant when data is continuously transferred at an accurate average rate.
That is, the video delay time control information is calculated from the code amount and the average rate, which are the result of compression for each picture. If the picture start code is not input to the multiplexer section at the correct average rate, the original relationship between the STC value at the time when the picture start code is input to the multiplexer section and the video delay time control information is disturbed.
An error occurs in the continuity of values. Specifically, for example, the difference between the TS values for two consecutive picture start codes must naturally match the picture period. That is, assuming that TS values are added to all consecutive picture start codes, those TS values should change exactly in picture cycle units. However, if the picture start code is not input to the multiplexer at an accurate average rate, the original relationship between the STC value at the time when the picture start code is input to the multiplexer and the video delay time control information is disturbed, and the relationship between the TS value of the operation result is Is
It does not change exactly for each picture cycle unit.
If this error becomes large, the buffer on the receiving side that performs output control based on the TS value breaks down, leading to a fatal error as a system.

The average rate generally corresponds to a target rate for controlling the amount of code resulting from the variable-length code to a certain rate by rate control in the video encoding processing unit. Since this rate control usually has a large time constant, the target rate can be achieved by averaging the code amounts of several pictures. That is, the code amount of each picture increases or decreases with respect to the target rate.
Therefore, in order to input code data to the multiplexer unit at an accurate average rate, code data for several pictures is accumulated, and then output at an accurate average rate that completely matches the rate targeted by the rate control. Such a configuration (buffer) is required for the video encoding processing unit. For this reason, a large-capacity buffer memory capable of storing code data of several pictures is required for the video encoding processing unit.
In addition, a high-precision data transfer control unit that inputs data from the buffer memory to the multiplexer at an accurate rate is required.

[0029]

In the conventional coded transmission apparatus, the multiplexer on the transmitting side holds the value of the STC and uses the held STC for calculating the TS (time stamp) value. Since a delay time of the order of tens of milliseconds occurs, a memory for STC is required. In addition, this memory must be accessed in a complicated manner in consideration of the replacement of the picture unit.
In MPEG1 or the like, the bit length of the STC used for the TS is 33 bits, so that the number of memories is large. Further, since the relationship between writing and reading with respect to this memory is asynchronous, there is no conflict between the timings thereof. There were various problems such as the necessity of an adjustment circuit. Further, in the second conventional example, there is a severe restriction that code data must be input to the multiplexer unit at an accurate average rate. Therefore, in order to input the code data to the multiplexer unit at an accurate average rate, it is necessary for the video encoding processing unit to accumulate code data for several pictures and then output the data at the average rate. Thus, a large-capacity buffer memory that can store code data of several pictures is required for the video encoding processing unit. Also,
There have been various problems such as the necessity of a high-accuracy data transfer control unit for inputting data from the buffer memory to the multiplexer at an accurate rate. The present invention eliminates these drawbacks, eliminates the need for a conventionally required memory having a wide bit width of 33 bits and an associated address generation section, and has a large-capacity memory for a video encoding processing section, It is an object of the present invention to provide a method for calculating a TS value in an encoded transmission system, which does not require a highly accurate data transfer control unit.

[0030]

In order to achieve the above object, a method for calculating a time stamp value in an encoding transmission system according to the present invention comprises the steps of: compressing and encoding at least one signal including at least a video signal; In a coded transmission system in which a signal is divided into a predetermined data length and a header such as a time stamp value is added to generate and transmit one stream, a predetermined time stamp of the video signal added to the header Value is calculated based on video delay time control information corresponding to the buffer delay time of the video decoding processing unit added in the stream, and the subsequent time stamp value is calculated by comparing the previously obtained time stamp value with the time stamp value of the video signal. The calculation is based on the cycle. In addition, the time stamp value calculation method in the encoded transmission system of the present invention,
A predetermined time stamp value of the video signal added to the header is converted into time information input to a signal processing unit in which the compression-encoded video signal is generated in the one stream, and , Video delay time control information corresponding to the buffer delay time of the video decoding processing unit, and a fixed time determined according to a predetermined divided data length of the compressed coded signal and a coding mode of the video signal. The calculation is performed by adding the information and the subsequent time stamp value is calculated based on the time stamp value obtained last time and the cycle of the video signal. Further, in the time stamp value calculating method in the coded transmission method according to the present invention, the signal generated in the one stream includes at least the compression-coded video signal and the compression-coded audio signal.

[0031]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The configuration of an embodiment of the coded transmission apparatus according to the present invention will be described below with reference to FIG. FIG. 1 shows FIG.
Multiplexer unit 56 constituting the coded transmission device shown in FIG.
3 shows a configuration of a portion relating to the calculation of the TS value in the video packetizing section 60. 5 is different from FIG. 5 showing the prior art in that the TS calculation unit 200 is different. STC generator 5
7 is output from the TS calculation unit 200
Has been entered. The video stream 55, which is data obtained by encoding the input video signal 53 in the video encoding processing unit 54 (FIG. 2), includes a picture start code detection unit 86, a picture coding type detection unit 87, and video delay time control information ( vbv_delay) detection unit 115. The picture stream code validity signal 82 output from the video encoding processing unit 54 is also input to the picture start code detection unit 86. On the other hand, the data 8 which is the output of the picture start code
9, the data 93 output from the picture coding type detection unit 87 and the data 117 output from the video delay time control information detection unit 115 are transmitted to the TS operation unit 200, respectively.
Has been entered. The data 89 output from the picture start code detection unit 86 includes a picture coding type detection unit 87 and a video delay time control information detection unit 1.
15 is also input. Further, the TS operation unit 200 includes an M-value parameter 99 which is information indicating the encoding mode.
And a delay parameter 100 which is information of a fixed offset delay amount. Then, the TS calculation unit 20
From 0, the data 97 to the buffer 101 and the control signal 9
8 is output. Also, from the buffer 101,
Data 104 is output to the switch unit 105 (see FIG. 2). Further, the picture period 201 of the input video signal and the reset signal 202 are also input to the TS calculation unit 200.

Next, this operation will be described. In the coded transmission device according to the present invention, the reset signal 202
Is calculated using the delay time control information in the same manner as described with reference to FIG. However, in the conventional example of FIG.
Subsequent TS values also use the delay time control information. For this reason, in order for the delay time control information to have a meaningful state, it is necessary to transfer the code data at an accurate average rate, which requires a large-capacity buffer memory in the video coding processing unit or a high-precision A data transfer control unit was required. Therefore, in the present invention, after the first TS value, calculation is performed based on the previously obtained TS value and the picture period 201 input to the TS calculation unit 200. As a result, since the delay time control information is not used after the first TS value, the transfer rate of the code data does not necessarily have to be an accurate average rate, and a large-capacity memory or a highly accurate data transfer control unit is not required. Can be eliminated. Specifically, the code data is transferred at a rate higher than the average rate. As a result, the capacity of the buffer memory can be reduced, and there is no problem if the transfer rate is anyway faster than the average rate. However, since the data is transferred faster than the average rate, the buffer memory may become empty. Therefore, it is necessary to perform control to temporarily stop the transfer when the buffer memory is empty. However, since this is controlled only by the empty flag of the buffer memory, it is very easy to realize. In addition,
In order to reduce the memory capacity as much as possible, it is possible to transfer the data at a higher rate.

Next, the TS after the first TS value calculation
An example of a value calculation is shown below. First, the aforementioned M value is
In the mode of M = 1, the following calculation is performed based on the previous PTS and the picture cycle. (1) When M = 1 PTS (PIC_cur) = PTS (PIC_ref) + PIC_TIME × N where PIC_cur: current picture, PIC_r
ef: PTS reference picture, PIC_
TIME: picture period, N: number of pictures between PIC_cur and PIC_ref, PTS (PIC_cur): PTS of PIC_cur, PTS
(PIC_ref): PTS of PIC_ref. In this mode,
As described above, only the PTS is obtained without considering the reordering of the pictures. For N, how many picture start codes are included in the packet currently being processed can be obtained by counting the data 89. As a result, the number of pictures between the previous reference picture and the picture in the packet currently being processed is obtained. Therefore, as described above, the PTS of the target picture can be calculated by adding the picture period N times to the PTS used as a reference (previously obtained). As described above, since the calculation is performed based on the previous TS value without using the STC value and the video delay time control information, a strict restriction on the input rate to the multiplexer unit can be unnecessary.

Next, the case of the mode of M> 1 will be described and described below. In this case, if the previous PTS is referred to, the reordering of pictures must be taken into consideration, so that the processing becomes complicated. Therefore, the following calculation is performed based on the previous DTS and the picture cycle.
In the multiplexer input portion, the DTS basically increases in the order of pictures in units of a picture cycle. (2) If M> 1 DTS (PIC_cur) = DTS (PIC_ref) + PIC_TIME × N Here, if the picture coding type is I or P
Then, from DTS (PIC_cur), PTS (PI
C_cur). PTS (PIC_cur) = DTS (PIC_cur) + PIC_TIME × N If the picture coding type is B, as shown below, DTS (PIC_cur) is replaced with PTS (PI_cur).
C_cur). This is because this B picture does not reorder, so that PTS = DTS. PTS (PIC_cur) = DTS (PIC_cur) where PIC_cur: current picture, PIC_r
ef: Picture used as DTS reference, PIC_
TIME: picture period, N: number of pictures between PIC_cur and PIC_ref, PTS (PIC_cur): PTS of PIC_cur, DTS
(PIC_cur): DTS of PIC_cur, DTS (PIC_ref): PIC
_ref DTS. As described above, even in the case of M> 1, since the TS value can be calculated based on the previous TS without using the STC value and the video delay time control information, a strict restriction on the input rate to the multiplexer is required. Can be eliminated.

Hereinafter, a specific calculation procedure of the present invention will be described with reference to FIGS. 8 and 9 show the operation flow of the time stamp value calculation in the time stamp calculation unit 200. This specific implementation method can be implemented by hardware, software, or a combination of both. 8 and 9 are each assumed to start from the start in packet units and return by return. Further, when a plurality of pictures (picture start codes) exist in a packet, a TS value is calculated for the first picture. FIG.
Is a time stamp value calculation operation flow when M = 1. First, the process starts from step 301 to process the corresponding packet. Then, in step 302, it is determined whether it is the first time stamp calculation after reset, and if it is the first time stamp processing, the process of step 303 is performed. This processing step 303 is the same as in the conventional example described with reference to FIG.

The next processing step 304 is for the next PTS calculation processing. That is, PTS (PIC_cur) = PTS (PIC_cur) in order to use the PTS (PIC_cur) obtained in step 303 as a reference for the next process. Then, the picture interval N to the picture for which the next time stamp value is to be obtained is determined by the number of pictures (the number of picture start codes) included in the current packet.
By setting. Thus, the first time stamp calculation process is completed, and the process exits from step 307. To process the next packet, repeat step 3
The processing is started from 01. This time is not the first timestamp calculation, so the process of step 305 is performed. Here, the previous PTS (PIC_r
ef) and N are used to determine PTS (PIC_cur). Step 306 is for the next arithmetic processing, similar to step 304 described above. Then, the process exits at step 307. The calculation of the time stamp value for the subsequent packet starts from step 301 and repeats the processing of step 305 and step 306 as described above.

FIG. 9 is a time stamp value calculation operation flow when M> 1. First, in order to process a corresponding packet, the process starts from step 401. Then, in step 402, it is determined whether or not the time stamp calculation is the first time stamp calculation after reset. If it is the first time stamp calculation, the picture coding type is further determined in step 403, and step 404 is performed.
(For I and P pictures) or step 405 (for B pictures). Steps 403 and 40
4, 405 is the same as the conventional example described with reference to FIG. In step 405, DTS (PIC_cur) = PTS (PIC_cur)
cur) in the next processing step 406 in the DT
This is for setting S (PIC_ref). Originally, PTS =
Since it is a DTS, there is no need to find the DTS.
Step 406 is for the next PTS calculation process. That is, step 404 or step 4
In order to use DTS (PIC_cur) obtained in step 05 as a reference for the next arithmetic processing, DTS (PIC_ref) = DTS (PIC_cur)
cur).

Then, the picture interval N to the picture for which the time stamp value is to be obtained is obtained by setting the number of pictures (the number of picture start codes) included in the current packet. Thus, the first time stamp calculation processing is completed, and the processing exits from step 412. To process the next packet, the process starts again from step 401. Since this is not the first time stamp calculation, the process of step 407 is performed. Here, the previous DT set in step 406
S (PIC_ref) and N are used to determine DTS (PIC_cur). Next, in step 408, the picture coding type is determined, and the processing of processing step 409 (for I and P pictures) or step 410 (for B pictures) is performed. In step 409, the DT obtained in step 407
Add picture reorder time to S (PIC_cur) and calculate PT
Find S (PIC_cur). In step 410, B
Since PTS = DTS for a picture, PTS (P
PTS is obtained from (IC_cur) = DTS (PIC_cur). The next processing step 411 is the same as step 406 described above.
This is for the next arithmetic processing. And Step 4
At 12, the process is exited. The calculation of the time stamp value for the subsequent packet starts from step 401 and repeats the processing of step 409 and step 410 as described above.

[0039]

According to the present invention, a video stream output by a video encoding processing unit is calculated based on video delay time information in a video stream only when the first TS value is calculated. By adding N times the picture period to the previously obtained TS value and obtaining the same, a memory having a wide bit width of 33 bits and an associated address generation unit, which are conventionally required, are required. In addition, it is possible to provide a method of calculating a TS value in an encoding transmission method, which does not require a large-capacity memory and a high-accuracy data transfer control unit in a video encoding processing unit.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a time stamp value calculation processing part of the present invention.

FIG. 2 is a block diagram of a general coded transmission device.

FIG. 3 is a diagram for explaining replacement of pictures in a video signal encoding process;

FIG. 4 is a block diagram of a video encoding processing unit.

FIG. 5 is a block diagram showing a configuration of a conventional time stamp value calculation processing part.

FIG. 6 is an explanatory diagram of a configuration of a video stream.

FIG. 7 is an explanatory diagram showing the relationship between video delay time control information and processing time in an encoding transmission device.

FIG. 8 is a view for explaining an example of a time stamp value calculation processing procedure according to the present invention;

FIG. 9 is a view for explaining an example of a time stamp value calculation processing procedure of the present invention.

[Explanation of symbols]

54: video encoding processing unit, 56, 118: multiplexer unit, 57: system time clock generation unit, 60:
Video packetizing section, 62: transmission path, 64: demultiplexer section, 66: video packet analyzing section, 71: video decoding processing section, 73: video decoding buffer, 75:
Video decoding circuit, 77: video reorder buffer, 8
6: picture start code detecting section, 87: picture coding type detecting section, 101: buffer, 111:
System time clock reproducing section, 115: video delay time control information detecting section, 200: time stamp calculating section, 2
01: picture cycle, 201: reset.

Claims (3)

[Claims] At least one signal including at least a video signal is compression-encoded, the compression-encoded signal is divided into a predetermined data length, and a header such as a time stamp value is added to generate a single stream. In a coded transmission system, a predetermined time stamp value of the video signal added in the header is set to a video delay time control corresponding to a buffer delay time of a video decoding processing unit added in the stream. A timestamp value calculating method in an encoding transmission system, wherein the method calculates based on information and calculates a subsequent timestamp value based on a previously obtained timestamp value and a cycle of the video signal. 2. The signal processing unit according to claim 1, wherein a predetermined time stamp value of the video signal added in the header is input to a signal processing unit in which the compression-encoded video signal is generated in the one stream. Time information, video delay time control information corresponding to a buffer delay time of a video decoding processing unit added to the video stream, a predetermined divided data length of the compression coded signal, and a code of the video signal. Encoding by calculating addition of fixed time information determined according to the mode of encoding, and calculating the subsequent time stamp value based on the previously obtained time stamp value and the cycle of the video signal. Timestamp value calculation method in the system. 3. The encoding according to claim 1, wherein the signal generated in the one stream includes at least the compression-encoded video signal and the compression-encoding audio signal. Time stamp value calculation method in the transmission method.