JP2003108193A - Encoder for real-time mpeg2 and multiplexing control method for encoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent an MPEG2 encoder from stopping operating even if a fault occurs to an ES encoding part. SOLUTION: In a real-time MPEG2 voice encoder having an ES encoding part which generates an ES (elementary stream) from a base-band voice signal and a multiplex processing part which generates a PES (packetized elementary stream) and a TSP (transport stream packet) from the ES and further performs multiplex processing together with the TSP of a PSI (program specific information) such as program information to output a TS (transport stream), the multiplex processing part while always monitoring a buffer occupation quantity value added to a frame as an encoding unit of the ES determines a multiplex position of the TSP including the frame head by using the value at need and multiplexes the TSP including the frame head at the determined multiplexing position when the ES of a fixed-bit-rate voice is multiplexed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a real-time MPEG2 audio encoder used in a digital broadcasting system or the like.

[0002]

2. Description of the Related Art A real-time MPEG2 audio encoder uses an input baseband signal as an ES (element) signal.
nary stream; elementary stream) is generated, and PES (packetize) is generated from this ES.
ed elemental stream; packetized elementary stream (PSI) is generated, and PSI (program specific) such as program information is generated.
information; program specific information (TSP)
This is an encoder that performs multiplex processing together with t stream packet (transport stream packet), and forms a TS (transport stream) packet of a predetermined length for output.

That is, the real-time MPEG2 audio encoder is an ES coding unit for coding an input audio baseband signal to generate an ES, and a multiplexing process for the generated ES and other signals. And a multiplex processing unit for converting into TS packets. by the way,
Real-time M used in digital broadcasting systems
As an encoder for PEG2 audio, it is important not to stop the operation as much as possible even when a failure occurs.

However, conventional real-time MPE
In the G2 audio encoder, the fixed bit rate E
When S is sent to the multiplex processing unit, which is a component of the final stage of the own device, and is multiplexed, the amount of ES to be multiplexed in a certain period of time is calculated and determined from the bit rate of ES, and thus the ES encoding unit is impaired. If the number of frames is correct and an error occurs in the generated code amount managed between the ES coding unit and the multiplexing processing unit due to the occurrence of error, etc., the multiplexing in the multiplexing processing unit is performed at that point. Processing could stop.

[0005]

As a real-time MPEG2 audio encoder used in a digital broadcasting system or the like, it is necessary to use a system that does not stop its operation as much as possible even when a failure occurs. In the case of E, the baseband signal is encoded and processed.
If a failure occurs in the ES coding unit that generates S,
Although the number of frames is correct, this ES encoder and this E encoder
If an error occurs in the generated code amount that is managed between the S encoding unit and the multiplexing processing unit that performs data multiplexing processing and TS packetizes, the multiplexing is performed at that point. There is a possibility that the multiple processing in the processing unit may stop, and there was a problem in terms of reliability.

Therefore, the object of the present invention is to
Real-time MPE that improves reliability by making it difficult to stop operation even when a failure occurs in the ES encoding unit
It is to provide a G2 encoder and a multiplexing control method in the encoder.

[0007]

In order to achieve the above object, the present invention is constructed as follows. That is, an ES (elementary str) is converted from a baseband audio signal.
and an ES encoding unit that generates the
(Packetized elemental st
ream), TSP (transport stream)
m.packet), and further PS such as program information
I (program specific Infoma
(TS) of TS (tra) and TS (tra)
In a real-time MPEG2 audio encoder having a multiplex processing unit for outputting an audio stream, the multiplex processing unit is an ES for audio of a fixed bit rate.
, The buffer occupancy value added to the frame that is the ES encoding unit is constantly monitored, and when necessary, the value is used to redetermine the TSP multiplexing position including the beginning of the frame, The TSP including the frame head is multiplexed at the re-determined multiplexing position.

According to this structure, the fixed bit rate E
In the case where S is multiplexed, the multiplexing process is performed by constantly monitoring the buffer occupancy value added to the frame, which is the ES encoding unit, and using that value when necessary to start the frame. The multiplexing position of the TSP including the frame is re-determined, and the TSP including the frame head is multiplexed at the determined multiplexing position.

When the TSP multiplexing position including the frame head is re-determined, the re-determinable multiplexing position (TS
"TSP_pos" is the position at which packets can be multiplexed
Then, this "TSP_pos" is (However, PTS (unit is [s / 27000000])
Is the decoding time of an audio frame having the beginning of a frame (A (n)) of a TS packet, and Delay (unit is [s]) is the delay DelayA of frame A (n).
(N) (unit is [s]), pES (unit is [byte]
e]) is the ES size of A (n-1) that exists immediately after the TS header or immediately after the PES header to the beginning of the frame A (n) if there is a PES header, and is CBR.
(Unit is [bps]) is the bit rate of ES,
ptr (unit: [s]) is determined from the offset value in buffering of the TS header or PES header).

As described above, according to the present invention, in the multiplex processing unit, the buffer occupancy value added to the frame as the ES encoding unit is constantly monitored, and when necessary,
By re-determining the multiplexing position of the TSP (TS packet) including the head of the frame, a failure occurs in the ES coding unit, and the number of frames is correct, but it is managed between the ES coding unit and the multiplexing processing unit. When an error in the generated code amount occurs, the generated code amount managed by the ES encoding unit and the generated code amount managed by the multiplex processing unit are made as close as possible to each other, resulting in multiplex processing. T does not stop, and T including the beginning of the frame can be used to improve reliability.
Since the SP multiplexing position is re-determined and the TSP including the frame head is multiplexed at the re-determinable multiplexing position, there is no concern that the multiplexing process will stop, and the reliability that the multiplexing process can always be performed normally High real-time MP
An EG2 encoder and a multiplexing control method for the encoder can be provided.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a system configuration example of an encoder for real-time MPEG2 audio according to the present invention. In FIG. 1, 11 is an ES encoding unit, 12 is a PES conversion unit, 13 is a TS packetization unit, 14
Is the multiplex part.

Of these, the ES encoding unit 11 performs an encoding process from the input baseband signal of the voice to obtain E.
S (elementary stream) is generated, and the PES conversion unit 1
2 is P from the ES output from the ES encoding unit 11.
ES (packetized elements)
stream; a packetized elementary stream) is generated and output, and the TS packetization unit 13 outputs this PES to a TS packet (transport).
t stream packet; transport stream packet), and the multiplexing unit 1
Reference numeral 4 denotes a voice-type TS packet packetized by the TS packetization unit 13 and a PSI (program specific inf) such as program information input from another.
information; TSP (Transport st) of Program Specific Information
stream packet; transport stream packet) is multiplexed with TS (t) of a predetermined length.
transport stream; transport stream) packetized and output.

In the present system having such a configuration, the ES coding section 11 generates an ES (elementary stream) from the input baseband signal, and the PES conversion section 12
And the PES conversion unit 12 generates a PES (packetized elementary stream) from this ES,
It is passed to the TS packetization unit 13. The TS packetization unit 13 converts this PES into a TS packet. And this TS
Pass the packet to the TS multiplexer. In the TS multiplexing unit 14, T
The TS packet packetized by the S-packetization unit 13 and a TS packet (transport stream packet) of PSI (program specific information) such as program information input from the other side are subjected to a multiplexing process, and predetermined processing is performed. It is output as a long TS packet.

Specifically, for example, real-time MPE
In the G2 audio encoder, one packet length of a PES (packetized elementary stream) packet is configured to be an integral multiple of a frame which is a coding unit of an audio ES (elementary stream). Therefore, the total length is the PES packet length.

In the MPEG2 system, this is 188
The TS packet having a [byte] length is transmitted. In this case, the PES conversion unit 12 shown in FIG.
The constituent bit string of the PES packet obtained by S conversion is
In the TS packetization unit 13, the capacity of the TS packet (TSP) is sequentially added to the TS packet from the beginning to form the TS packet. Then, the TS multiplexing unit 14 multiplexes this with other TS packets such as PSI information.

[0016] Here, in the present system as described above, when necessary, such as when the TS multiplexing unit 14 starts multiplexing and when the ES encoding unit 11 cannot perform a certain amount of multiplexing for a certain period of time. By carrying out the control as shown in (1), the multiplexing position is determined and multiplexing is performed.

Now, on the TS, the position where the TS packet including the beginning of a certain voice frame can be multiplexed is TSP_po.
Let s. TSP_pos is a PCR described in PMT, which is a packet indicating a program configuration of a program including the audio.
(Program clock reference:
It is expressed as a count value of 27 [MHz] that can be calculated from the TSP on which the reference clock of 27 [MHz] is superimposed. The calculation method is specified in ISO13818-1.

Therefore, TSP_pos (unit is [s / 2
7000000]) can be calculated as follows.

[0019] However, PTS [s / 27000000] is 27 [MH
z] is the decoding time of the audio frame represented by the count value, and X (the unit is [bits]) is ISO1.
3818-7 minimum_decode
The buffer occupancy immediately before the audio frame is decoded in the buffer on the virtual decoder called r_nput_buffer. Where adts_buffer_fullness
Is the adtsb described in the header of the audio frame.
The value of the "ufferf_fullness" field, and "channel_num" is the minimum of the voice.
um_decoder_input_buffer Number of channels used in size calculation, frame
_Length is a frame_length field value described in the header of the audio frame. In addition, pES (bits) is a TS packet including the beginning of the audio frame and the ES of the frame before the frame.
Is superimposed, then the superimposed ES size, br (bits / s) is the bit rate of the audio ES, and ptr (s) is sent earlier considering the delay of subsequent devices. Is an offset value such as.

If the multiplex position is determined by this calculation method, the position as a count value of 27 [MHz] on the TS is calculated for each TSP (each unit TS packet), and the TS is calculated.
Compared with P_pos, the position on TS is always TSP_
When pos is exceeded, multiplexing of TSP (TS packet) including the head of the audio frame can be started from there. Other than that, a certain amount of multiplexing is performed for a certain period according to the bit rate. Such control is performed by the TS multiplexing unit 14 in this system.

A more detailed description will be given. In the following, the audio ES is referred to as AAC (advance) of ISO13818.7.
The d audio coding method is assumed.

The code amount of each audio frame of the audio ES of the fixed bit rate is the MPEG standard ISO13818-.
It is necessary to control so that overflow or underflow of the input buffer does not occur when ES is input at a fixed bit rate to the input buffer of the decoder defined by 7.

For example, as shown in FIG. 2, when a certain audio ES is input to the input buffer of the audio decoder specified by the MPEG standard, the buffer occupancy when the code amount of each audio frame satisfies the standard. The transition example of is shown in this figure.

Here, the slope of the change in the accumulated amount of the buffer is
It is the bit rate of ES, and is assumed to be instantaneously decoded at the decoding time of the audio frame and removed from the input buffer.

StA (1), StA (2), StA
(3), StA (4) ... At the timing, buffer input of each audio frame is started, and DtA (1), Dt
Decoding of each audio frame is started at the timing of A (2), DtA (3), DtA (4), .... When the audio frame buffer input is started, the data occupancy in the buffer increases at a constant slope in time with the pace, and data is taken out from the buffer and decoded before the overflow occurs. There is. The amount fetched from the buffer for decoding is A (1), A
(2), A (3), A (4) ... Are voice frames. As the data is taken out of the buffer, the data of the next audio frame is accumulated in the buffer from the reduced position. The time position at which “0” is reached is the estimated time of the buffer input end time for the audio frame before the currently accumulated frame.

In the case of the example of FIG. 2, the buffer occupancy does not exceed the upper and lower limits of the buffer. When the data stored in the buffer is read-controlled as in the present invention and multiplexed by the TS multiplexing unit 14, a fixed bit rate is input to the input buffer of the decoder defined by MPEG standard ISO13818-7. It can be seen that control can be performed so that overflow or underflow of the input buffer does not occur when the voice ES is input.

Here, the buffer input start time StA (n) and the decoding time DtA (n) of the audio frame A (n) are set.
The positional relationship of is expressed by the following formula. Here, the above-mentioned ads_buffer_fullne
ss (32bitword / Channel_num)
Is ads_buf described in the header of the audio frame An described in MPEG standard ISO13818.7.
value of the fer_fullness field, Chann
el_num is the minimum_deco of the audio frame An described in the MPEG standard ISO13818.7.
der_input_buffer number of channels used in size calculation (Channel_configu
can be calculated from the relation field value), frame
e_length (unit is [byte]) is MPEG
The frame_length field value and CBR (unit: [bps]) described in the header of the audio frame An described in the standard ISO13818.7 are the bit rate of ES.

From the rear of the audio frame A (n), L
The time when [byte] is input to the buffer is Is expressed as

Based on this, how to perform the TS multiplex timing control will be specifically described.

[TS Multiplexing Method] When ES is converted into PES and then converted into TS packets, and TS packets obtained from this are further multiplexed, MPEG standard ISO1381 is used.
Follow 8.1. In multiplexing TS packets, similar to the case of ES, T-STD (T
transport Stream system
It is necessary to multiplex a buffer called a get decoder, which is attached to a decoder, while controlling so as not to overflow / underflow.

This buffer is composed of the above ES buffer and a buffer for buffering other PES packets and TS packets. Basically, the ES superimposed on the payload of the TS packet is fixed as much as possible. Evenly multiplex the bit rate, and
MPEG to maintain the relationship between ta (n) and DtA (n)
The buffer conditions can be met by stamping the time stamps specified in the system standard.

Therefore, in the present system, with respect to the TS packet including the beginning of the audio frame, the position where the TS packet can be multiplexed (multiplexable position) is determined as follows, and multiplexing is performed at that position. .

First, regarding the TS packet including the beginning of the audio frame, the position where the TS packet can be multiplexed is defined as "TSP_pos".

The "TSP_pos" is a PMT (program.
PCR described in the map table)
mclock reference; 27 [MHz]
It is expressed as a count value of 27 [MHz] that can be calculated from the TS packet on which the (reference clock of) is superimposed. The calculation method complies with the MPEG standard ISO13818.1. As a result, the TS packet multiplexable position "TSP_p"
os ”(unit is [s / 27000000]) can be calculated as follows.

[0035] However, PTS (the unit is [s / 27000000]) is
Decoding time of an audio frame having the head of a frame of a TS packet (DtA (n) in terms of ES)
Time (corresponding to [s]), Delay (unit is [s]) is DelayA (n) (unit is [s]) described above, and pES (unit is [byte]) is TS.
If it is immediately after the header or if there is a PES header, it is the ES size of A (n-1) existing immediately after the PES header to the beginning of the audio frame A (n), and CBR (unit: [bps]) is the audio ES. Is the bit rate of
r (unit: [s]) is an offset value for buffering the TS header or PES header, or when sending earlier in consideration of delay of subsequent devices.

If the multiplex position is determined by this method, the position as the count value of 27 [MHz] on the TS is set to 1TS.
It is calculated for each P (unit TS packet) and is constantly compared with TSP_pos. When the position on the TS exceeds TSP_pos, the multiplexing start is possible from there.

For other TS packets, ES
Multiplex at a rate higher than the bit rate of. Note that if multiplexing is performed at a bit rate of ES or more, the buffer standard attached to the T-STD of ISO13818.1 may overflow. Therefore, while monitoring the buffer occupancy at all times, if overflow is likely to occur, T
The process of temporarily stopping the S packet multiplexing is performed so that the standard is satisfied.

The rate above the ES bit rate may be basically any number, but if the rate is too high, the multiplexing of TS packets is not smoothed and becomes bursty, so that it is further increased in subsequent devices. Since there is a risk that the multiplexing position of other streams will be affected in the case of multiplexing with TS packets such as video and audio, it is necessary to carry out this within an appropriate range.
If the code amount from the ES coding unit is too large,
The closer to the ES bit rate, the longer it takes to restore the delayed multiplex position. Also, ptr
[S] is also a value of "0" or more, but it is desirable to determine it by taking into account the minimum delay in the trailing outliner and the like.
If you make it too big, the buffer will overflow, so be careful.

TS for which multiple processing is performed by applying the above
The processing flow in the multiplexing unit 14 is shown in FIG. The TS multiplexing unit 14 acquires the information of the audio TS packet to be multiplexed next (S1 in FIG. 3). Then, it is checked whether or not the TS packet includes the head of the audio frame (S2 in FIG. 3). As a result, if the audio TS packet to be multiplexed next is the TS packet including the head of the audio frame, this TS packet is At the current timing, it is checked whether or not it is a position where multiplexing is possible (S3 in FIG. 3).

As a result, if it is not at the position where multiplexing is possible, it waits until it comes to the position where multiplexing is possible (S3 in FIG. 3).

If there is a position where multiplexing is possible, TS packets are multiplexed at a predetermined rate (S4 in FIG. 3). Then, the process returns to step S1 and the processes after step S1 are repeated.

On the other hand, as a result of the check in step S2, if the next audio TS packet to be multiplexed is not a TS packet including the beginning of an audio frame, this TS packet is multiplexed at a predetermined rate (see FIG. 3). S4).
Then, the process returns to step S1 and the processes after step S1 are repeated. Such control is performed by the TS multiplexing unit 14
Let's do it in.

By carrying out such control, the following effects can be secured.

[Effect of application of the present invention] As an application example, consider the following example. Bit rate of TS and its audio E
This is the case where the voice ES is converted into TSP and multiplexed for 5 packets for a certain fixed time calculated from the bit rate of S.

FIG. 4 shows an example of a steady state where no problem occurs. (A) shows a conventional method and (b) shows an example to which the present invention is applied. An area in which dots are scattered indicates an area in which TSP (TS packets) including the head of a certain voice frame is multiplexed, and a shaded area indicates the remaining TSP (the remaining portion of the TS packet) of the voice frame. The multiplexed area is shown, and the solid area is the T other than the voice ES.
An area in which SPs (TS packets) are multiplexed is shown.

As shown in FIG. 4, there is no difference between the conventional system and the system of the present invention during normal operation.

Next, let us compare the abnormal times. FIG. 5 is an example of the case where the code amount is reduced, and the headers of the second frame and the third frame are preserved due to a failure of the ES encoding unit, etc., but the code amount becomes 1/3. Is an example of. Figure 5
In the conventional method shown in (a), since a certain amount is multiplexed in the period of T, three frames are multiplexed in the first period of T.
Therefore, since there is no code in the next T period, it is not possible to determine how to multiplex thereafter, which causes a problem.

On the other hand, in the invention system of FIG. 5B, since the multiplex position is recalculated for the frame and the multiplex is started based on the calculated position, the operation can be continued without any error.

On the other hand, FIG. 6 shows an example in which the code amount is increased. In this case, if it is assumed in the above that the code amount of the second frame has increased 4/3 times due to the occurrence of 11 failures in the ES encoding unit, in this case, in the conventional method of FIG.
Since a fixed amount is multiplexed in the period of, the multiplexing of the third frame and thereafter remains delayed.

On the other hand, in the method of the present invention shown in FIG. 6B, since the output position of the TS packet is behind, the next frame head can be multiplexed immediately and the multiplexing is started immediately. Therefore, the multiplex position returns to the original position at the fourth frame.

As described above, according to the present invention, for the TS packet including the beginning of the voice frame of the voice ES input at a fixed rate, the position (multiplexable position) "TSP_pos" at which the TS packet can be multiplexed is set to TSP_pos.
= PTS- (Delay + pES / CBR + ptr) *
(1/27000000), and multiplexing is performed at that position.

In the present invention, in the multiplexing unit which is a constituent element of the encoder, the buffer occupancy value added to the frame which is the ES encoding unit is constantly monitored, and the frame head is added when necessary. By re-determining the multiplexing position of the TSP that contains it and starting multiplexing from that position, a failure occurs in the ES encoding unit, and the number of frames is correct, but it is managed between the ES encoding unit and the multiplexing unit. When an error in the generated code amount occurs, the generated code amount managed by the ES coding unit and the generated code amount managed by the multiplexing unit can be brought closer to each other as a result. The encoder can be operated so that the multiprocessing does not stop. Therefore, according to the present invention, it is possible to prevent the multiprocessing from being stopped and ensure the reliability.

In addition to the above-mentioned audio, the present invention provides MPE such as video.
When the occupancy of the decoder buffer is described for each frame in the G encoded data, the I
When using the SO13818 / 1 system standard,
All can be applied by calculating the above-mentioned Dolay according to each standard.

The present invention is not limited to the above-described embodiment, but can be implemented by being modified in various ways, and the present invention can be applied and applied to not only audio but also MPEG2 encoded data such as video. is there. Further, in the present invention, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some of the constituent elements are deleted from all the constituent elements shown in the embodiment, at least one of the problems described in the section of the problem to be solved by the invention can be solved and described in the section of the effect of the invention. When at least one of the effects described above is obtained, a configuration in which this constituent element is deleted can be extracted as an invention.

The method described in the embodiments of the present invention is a program that can be executed by a computer, such as a magnetic disk (flexible disk, hard disk, etc.), optical disk (CD-ROM, CD).
-R, CD-RW, DVD, MO, etc.), and can be distributed by being stored in a recording medium such as a semiconductor memory.
It can also be distributed by transmission via a network.

[0056]

As described above in detail, even when a failure occurs in the ES coding unit, it is difficult to stop the operation and the reliability of the encoder for real-time MPEG2 and the multiplexing control in the encoder are improved. A method can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of a main part of a system of the present invention.

FIG. 2 is a diagram for explaining the present invention.

FIG. 3 is a diagram showing an example of a control flow in the TS multiplexer 14 in the system of the present invention.

FIG. 4 is a diagram for explaining an effect of the present invention in comparison with a conventional technique.

FIG. 5 is a diagram for explaining an effect of the present invention in comparison with a conventional technique.

FIG. 6 is a diagram for explaining an effect of the present invention in comparison with a conventional technique.

[Explanation of symbols]

11 ... ES coding unit, 12 ... PES conversion unit, 13 ... TS packetization unit, 14 ... Multiplexing unit.

Claims (5)

[Claims] 1. From a baseband audio signal to an ES (elem)
An ES coding unit that generates an entry stream, and a PES (packetized element) from the ES.
entry stream), TSP (transp)
ort stream packet), and further PSI (program specif) such as program information.
In a real-time MPEG2 audio encoder having a multiplex processing unit that performs multiplex processing together with TSP of ic information) and outputs a TS (transport stream), the multiplex processing unit is configured to multiplex ES of fixed bit rate audio. While constantly monitoring the buffer occupancy value added to the frame, which is the encoding unit of TSP, the value is used when necessary to redetermine the TSP multiplexing position including the beginning of the frame, and the determined multiplexing capability An encoder for real-time MPEG2, characterized in that a TSP including the frame head at a position is multiplexed. 2. An ES (element) based on an input signal.
an ES encoding unit for generating an (ary stream), and E
S to PES (packetized element)
ary stream), the first type of TSP (tra
nstream streampacket),
Furthermore, multiple processing is performed together with other TSP, and TS (trans
In a real-time MPEG2 encoder having a multiplex processing unit for outputting a sport stream), the multiplex processing unit, when multiplexing the ES having a fixed bit rate, has a buffer occupation amount added to a frame which is a coding unit of the ES. While constantly monitoring the value, the value is used to redetermine the multiplexing possible position of the TSP that includes the frame head, and the TSP that includes the frame beginning is multiplexed at the redeterminable multiplexing position. An encoder device for real-time MPEG2 characterized by the following. 3. When re-determining the multiplex position of the TSP including the frame head, the multiplexable position to be re-determined is "T".
If SP_pos ”, then this“ TSP_pos ”is (However, PTS (unit is [s / 27000000])
Is the decoding time of an audio frame having the beginning of a frame (A (n)) of a TS packet, and Delay (unit is [s]) is the delay DelayA of frame A (n).
(N) (unit is [s]), pES (unit is [byte]
e]) is the ES size of A (n-1) that exists immediately after the TS header or immediately after the PES header to the beginning of the frame A (n) if there is a PES header, and is CBR.
(Unit is [bps]) is the bit rate of ES,
The encoder for real-time MPEG2 according to claim 1 or 2, wherein ptr (unit: [s]) is determined from an offset value in buffering of a TS header or a PES header. 4. An ES (elem) from a baseband audio signal.
An ES coding unit that generates an entry stream, and a PES (packetized element) from the ES.
entry stream), TSP (transp)
ort stream packet), and further PSI (program specif) such as program information.
In a multiplexing control method in a real-time MPEG2 audio encoder having a multiplex processing unit that performs multiplex processing with TSP of ic information and outputs TS (transport stream), the multiplex processing unit in the case of multiplexing ES with a fixed bit rate In the multiplexing of, while constantly monitoring the buffer occupancy value added to the frame which is the ES encoding unit, the value is used when necessary to redetermine the TSP multiplexing possible position including the frame head, A multiplexing control method in an encoder for real-time MPEG2, characterized in that TSP including the frame head is multiplexed at the redeterminable multiplexing position. 5. When re-determining the multiplexing position of the TSP including the frame head, the re-determinable multiplexing position is designated as "T".
If SP_pos ”, then this“ TSP_pos ”is (However, PTS (unit is [s / 27000000])
Is the decoding time of an audio frame having the beginning of a frame (A (n)) of a TS packet, and Delay (unit: [s]) is the delay DelayA of frame A (n).
(N) (unit is [s]), pES (unit is [byte]
e]) is the ES size of A (n-1) existing immediately after the TS header or immediately after the PES header to the beginning of the frame A (n) if there is a PES header, and is CBR.
(Unit is [bps]) is the bit rate of ES,
5. The real-time MPEG2 according to claim 4, wherein ptr (unit: [s]) is determined from offset value in buffering of TS header or PES header.
Control method in encoder for video.