JP2001177487A - Data transmission method and its system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a compression data multiplexer that can decrease a transmission rate of compressed data in the case of multiplexing a plurality of transport streams and can process data at a low data rate. SOLUTION: In the data transmission where a transmitter side multiplexes a plurality of transport streams and the multiplexed streams are transmitted and a receiver side demultiplexes the received multiplexed transport streams and recovers the streams, when a plurality of the transport streams is multiplexed, a plurality of the transport streams is multiplexed in the unit of a prescribed data length, a synchronous code incidence sequence of each transport streams is detected from the received multiplexed transport streams so as to demultiplex and recover a plurality of the transport streams on the basis of the incidence.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improved method and apparatus for transmitting and receiving compressed video data and transmitting and receiving compressed video data.

[0002]

2. Description of the Related Art In recent years, image processing techniques, especially compression techniques for image information and audio information, have been remarkably developed, and high-quality moving images and audio can be transmitted even on a transmission line having a small capacity. became. By the way, in the above system, a group of data flowing on a transmission path is called a transport stream. A transport stream is made up of a fixed amount of data. A certain amount of data constituting the transport stream is called a packet, and is a minimum unit of the transport stream. The transport stream is composed of a plurality of packets. Since all packets have a fixed data amount, the packet length is constant. Also, it is not allowed that there is a gap between the packets, and the packets are always in contact before and after the packets, and the packet positions do not overlap. The packet always has a sync code at the beginning. The position of the sync code is used as a clue to find the packet.
Analyze internal data.

Next, the structure of a transport stream and a packet will be described with reference to FIG. The transport stream 30 includes a plurality of packets 31. Since all the packets 31 have a fixed data amount, the packet length is constant. The packet 31 is composed of two areas, a header area 32 and a compressed data area 33. Most of the area of the packet 31 is occupied by the compressed data area 33. The compressed data area 33 stores data obtained by compressing images and sounds. At the head of the packet 31, there is a header area 32 storing data indicating information of the packet and data indicating information of the compressed data area. The head data of the header area 32 is called a synchronization code 34, and has the same value in all packets. The synchronization code 34 serves as a mark for detecting the head of the packet 31 from the transport stream 30. Data having the same value as the synchronization code 34 may exist at random in the compressed data constituting the packet 31. Therefore, in order to determine whether or not the detected data is the real synchronization code 34, it is necessary to detect that the data which is considered to be the synchronization code 34 appears several times continuously at the same interval as the packet length. is there. That is, the synchronization code 34 is continuously transmitted several times at intervals of the packet length.
When the data having the same value as the above is detected, the current data is determined to be the synchronization code 34, and the data inside the packet 31 is analyzed based on this position.

The transport stream data is synchronized with a transmission clock, and one data is transmitted in one clock. When multiplexing multiple transport streams,
Cut out and multiplex in packet units. The order in which the packets are multiplexed is not particularly defined, but is generally arranged evenly. For example, in a case where two transport streams are multiplexed, it is easiest to arrange two types of transport stream packets alternately in order to realize an even distribution of the packets. However, as mentioned above, between transport stream packets,
There must be no gaps or overlap. Originally, in order to multiplex another packet into a transport stream having no gap between packets and to make one transport stream having a continuous packet structure again, another packet is inserted between packets having no gap. It is necessary to make a gap to insert. Therefore, when transmitting a multiplexed transport stream, a gap is created by increasing the data rate and narrowing the interval between individual data. In other words, since the transmission data is synchronized with the transmission clock, increasing the data rate means increasing the frequency of the transmission clock.

Here, as an example, in order to consider a case where two transport streams are multiplexed, a time chart of a transport stream before and after data rate conversion is shown in FIG. 3 and described. As shown in FIG. 3, the transmission clock 3 after the data rate conversion
The one clock width of 8 is half the one clock width of the transmission clock 35 before the data rate conversion. That is, in order to multiplex and transmit two transport streams, the transmission clock 38 after the data rate conversion is used.
Must be twice the frequency of the transmission clock 35 in the transport stream before the data rate conversion. Therefore, the data rate of the transport stream is doubled. By repeating such a data rate conversion process on a packet basis, other packets are inserted into gaps between packets created by the data rate conversion process to multiplex the transport stream. FIG. 4 shows a state in which a gap is created between packets by this data rate conversion process, and another packet is inserted into the gap. Transport stream 40
The packet is cut out in packet units, and data rate conversion processing is performed on the packet. As a result, the length of the packet is apparently halved, and a gap is generated between the next packet. The transport stream 41 is another transport stream to be multiplexed with the transport stream 40 and transmitted. The transport stream 41 is also the transport stream 40
Similarly, the data rate conversion process is performed. Thereby, the length of the packet becomes the transport stream 4
It is halved in the same way as the processing of 0. Here, the transport stream 41 has the same data rate as the transport stream 40, and the data length of one packet is equal. The packet length of the transport stream 41 after the data rate conversion processing is the transport stream 40
The length of the gap generated in the data rate conversion process. Therefore, by inserting the packets of the transport stream 41 after the data rate conversion processing into the gaps of the transport stream 40 after the data rate conversion processing, the multiplexed transport stream 42 with no gap between the packets is inserted. Can be generated.

FIG. 5 shows an example of a configuration of a conventional technology of a transport stream demultiplexing circuit when two transport streams are multiplexed and transmitted. The transport stream demultiplexing circuit includes a transport stream demultiplexing unit Tx 'forming the transmitting side and a transport stream demultiplexing unit Rx' forming the receiving side. The transport stream multiplexing unit Tx ′ has an input side to which encoders 50 and 51 are connected and a transmission clock 5.
8 are supplied. A transmission path 69 is connected to the output side, and the multiplexed transport stream 68
Then, the transmission clock 58 is output. The transport stream separation unit Rx ′ has an input side connected to the transmission line 69 and an output side connected to the decoders 72 and 73.

First, the transport stream multiplexing section T
The configuration and operation of x 'will be described. The transport stream multiplexing unit Tx 'includes encoders 50 and 51, a clock divider 59, a packet length counter 61, a packet selection signal generator 63, data accumulators 66 and 67, and synchronization code detectors 54 and 56. The clock frequency divider 59 divides the frequency of the transmission clock 58 input from outside the transport stream multiplexing unit Tx ′ to generate a frequency-divided clock 60 having a lower frequency, and generates the encoders 50 and 51 and the data accumulator 66. , 67. For example, when the two transport streams 52 and 53 are multiplexed, the transmission clock 58 is frequency-divided by a clock frequency divider 59 to generate a clock signal having a half frequency and output as a frequency-divided clock 60. The packet length counter 61 operates an internal counter in synchronization with a transmission clock 58 input from outside. When the internal counter counts the number equal to the packet length, it outputs an overflow signal 62, and starts counting a new number equal to the packet length.
By repeating this operation endlessly, the overflow signal 62 is continuously output at intervals equal to the packet length. The packet selection signal generator 63 outputs an overflow signal 62 output from the packet length counter 61.
And outputs a packet selection signal. The number of outputs of this packet selection signal is equal to the number of data accumulators of the transport stream multiplexing unit. For example, when multiplexing two transport streams, two packet selection signals 64 for the data storage 66 and a packet selection signal 65 for the data storage 67 are output. Further, each packet selection signal is switched every time the overflow signal 62 is input. Since the interval at which the overflow signal 62 is output is equal to the packet length, the interval at which the packet selection signal switches is also equal to the packet length. The order in which the packet selection signals are switched is such that when one packet selection signal becomes valid, all other packet selection signals become invalid, and only one packet selection signal becomes valid. Also, the ratios at which the individual packet selection signals are valid for all the packet selection signals are equal. Therefore, when two transport streams are multiplexed, two packet selection signals are output, but in order to make them effective at the same rate, it is effective to alternately enable the two to achieve the same rate. The easiest way to do it.

[0008] In the synchronous code detector 54, the encoder 5
0 analyzes the transport stream 52 output in synchronization with the divided clock 60 from the clock divider 59,
The synchronization code in the transport stream 52 is detected. Then, the head of the packet is found by detecting the synchronization code. Then, when the head of the packet is found, the write permission signal 55 is switched from invalid to valid and output to the data storage 66 so that the transport stream 52 is accumulated from the head of the packet. Thereafter, the state of the write permission signal 55 remains valid. In the same manner as the synchronous code detector 54, the synchronous code detector 56 outputs the divided clock 6 from the clock frequency divider 59.
The transport stream 53 output in synchronization with 0 is analyzed, and the head of the packet is found by detecting the synchronization code in the transport stream 53.
When the head of the packet is found, the write enable signal 57 is switched from invalid to valid, output to the data storage 67, and the transport stream 5 is transferred from the head of the packet.
3 is accumulated. Thereafter, the state of the write permission signal 57 remains valid. The data accumulator 66 writes and accumulates the transport stream 52 output from the encoder 50 in synchronization with the divided clock 60 from the clock divider 59. At this time, the transport stream 52 is not written unless the write permission signal 55 output from the synchronization code detector 54 is in a valid state.
That is, by writing after the write permission signal 55 becomes valid, the head of the accumulated data coincides with the head position of the packet.

The reading of the stored transport stream 52 is performed in synchronization with the transmission clock 58. Here, when the packet selection signal 64 output from the packet selection signal generator 63 is invalid, the transport stream 52 is not output while being stored. Then, when the packet selection signal 64 becomes valid,
The data storage 66 outputs the stored transport stream 52. As described above, the switching interval of the packet selection signal 64 is equal to the packet length. Therefore, the switching interval of the stored data output is performed in units of the minimum packet length. The data storage 67 is the same as the data storage 66, and
The transport stream 53 output from 1 is written and stored in synchronization with the divided clock 60 from the clock divider 59. Here, unless the write permission signal 57 output from the synchronization code detector 56 is in a valid state, the transport stream 53 is not written, and the write is performed after the write permission signal 57 becomes valid. Also,
Reading of the stored transport stream 53 is performed in synchronization with the transmission clock 58. here,
When the packet selection signal 65 output from the packet selection signal generator 63 is invalid, the transport stream 53 is not output while being stored. Then, when the packet selection signal 65 becomes valid, the data storage 67 outputs the stored transport stream 53. In this manner, the transport stream 68 alternately output from the data storage 66 and the data storage 67 is
The signal is transmitted to the transmission line 69 together with the transmission clock 58 and transmitted. Here, the output data rate of the transport stream 68 is twice as high as the input data rate of the transport stream data in all data accumulators due to the data rate conversion process. That is, the transport stream 68 multiplexed to the transmission path 69 is output at a data rate twice as high as the transport stream data output from each encoder.

Next, the transport stream separation unit R
The configuration and operation of x 'will be described. The transport stream separating unit Rx ′ receives the transport stream 70 from the transmission path 69 and outputs the transport stream 70 to the decoders 72 and 73 as it is. The decoders 72 and 73 detect a synchronization code from the transport stream 70 and take in only necessary packets to expand the data. As described above, since the transport stream data of the encoders 50 and 51 are alternately multiplexed in the transport stream multiplexing unit Tx ′ in packet units, only the transport stream 52 output from the encoder 50 is focused. Then, after continuously outputting the same length as the packet length, it repeats that the transport stream 52 does not come again by the same length. That is, when the transport stream 52 generated by the encoder 50 is viewed from the decoder side of the transport stream separation unit Rx ′, the same data rate as the transmission path 69, which is twice the data rate of the transport stream 52 of the encoder 50, After the transport stream 52 has been continuously input in step (1), the transport stream 52 disappears for a while, and the transport stream 52 appears to be input. This is because the transmission rate is doubled by the data rate conversion processing. Therefore, the decoders 72, 73
0 is the speed at which the transport stream 52 is generated.
Captures the transport stream 70 at twice the speed,
Processing for detecting the synchronization code must be performed.

Similarly, when multiplexing three or more transport streams, data rate conversion processing is performed.
The difference from the method of multiplexing two transport streams is that the data rate must be tripled when multiplexing three transport streams, and the data rate must be quadrupled when multiplexing four transport streams. That is, it is impossible to generate a transport stream having no gap between them. Therefore, in this data rate conversion processing, the length of one packet is set to 3
One transport stream needs to be reduced to one third, and four transport streams need to be reduced to one fourth. For example, when a transport stream generated by a certain encoder has a data rate synchronized with a clock of 15 MHz, transport streams having the same data rate as this transport stream are collected and collected.
If one transport stream is to be multiplexed and transmitted, it must be transmitted in synchronization with a transmission clock of 60 MHz, which is four times 15 MHz. That is, all the decoders of the transport stream separation unit need to fetch and process the transport stream data synchronized with the 60 MHz transmission clock in order to detect the head of the packet. However, a generally used decoder takes in the transport stream data output by the transport stream multiplexing unit at a data rate four times that of the transport stream data generated by each encoder, and only detects the beginning of a packet. You may not be able to use it because you do not have the ability. This is because the connection form between the normal encoder and the decoder is assumed to be one-to-one, and the decoder does not require data processing capability at such a data rate. Therefore, when multiplexing a plurality of transport streams, a special decoder having higher processing capability than a generally used decoder is required.

[0012]

In the case where transport stream data output from a plurality of encoders is multiplexed by a transport stream multiplexing unit and output to a transmission path, the data is connected to a transport stream separating unit on the receiving side. The decoder must detect the head of the packet for data input at a data rate several times the data rate of the transport stream output from one encoder. Cannot handle such a high data rate transport stream. The present invention eliminates these drawbacks, and can reduce the transmission speed of compressed data handled by each encoder / decoder even when multiplexing a plurality of transport streams, and cannot process a high data rate. It is an object of the present invention to realize a compressed data multiplexing apparatus that can cope with the above.

[0013]

According to the present invention, in order to achieve the above object, a plurality of transport streams are multiplexed and transmitted on a transmitting side, and the plurality of transport streams are received from a multiplexed transport stream received on a receiving side. In data transmission that separates and reproduces the transport stream of
When multiplexing the plurality of transport streams,
The plurality of transport streams are multiplexed in units of a predetermined data length. Also,
Based on the synchronization codes included in the plurality of input transport streams, the synchronization codes and the data of the plurality of transport streams are arranged and multiplexed in units of a predetermined data length according to the multiplexing order, and transmitted. It is something to do. Further, from the multiplexed transport stream received on the receiving side, the synchronization code appearance order of each transport stream is detected, and based on this, the plurality of transport streams are separated and reproduced. is there. That is, when multiplexing the transport stream data output from a plurality of encoders, the data is divided and multiplexed with one data as a unit. Speed is lower than the transmission clock,
Make it look like a low data rate. As a result, a decoder that can process only a data rate lower than the data rate on the actual transmission path can be connected. That is, the transmitting side alternately multiplexes the data of each channel for each data and transmits the data, and the receiving side extracts the compressed data of each channel from the transmission data using a clock obtained by dividing the transmission clock. As a result, the transmission speed of the compressed data handled by each encoder and decoder becomes slow, and the system can cope with a system that cannot handle a high data rate.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The configuration of an embodiment of a compressed data multiplexing apparatus according to the present invention is shown in FIG. 1 and will be described in detail below. This is a configuration example of a transport stream demultiplexing circuit in a case where two transport streams are multiplexed and transmitted. The transport stream demultiplexing circuit includes a transport stream multiplexing unit Tx forming the transmitting side and a transport stream demultiplexing unit Rx forming the receiving side. Encoders 1 and 2 are connected to the input side of the transport stream multiplexing unit Tx, and a transmission clock 6 is input thereto. The transmission line 15 is connected to the output side, and the transport stream 24 and the transmission clock 6 are output. To the transport stream separation unit Rx, the transmission line 15 is connected to the input side, and the decoders 17 and 18 are connected to the output side.

First, the transport stream multiplexing unit T
The configuration and operation of x will be described. The transport stream multiplexing unit Tx includes encoders 1 and 2, a clock divider 7, a data selection signal generator 10, a data storage 9, a synchronization code detector 5, and a data selector 14. The clock frequency divider 7 divides the frequency of the transmission clock 6 input from outside the transport stream multiplexing unit Tx to generate a frequency-divided clock 8 having a lower frequency.
Output to the data storage 9. For example, when multiplexing two transport streams, the transmission clock 6
Is divided by 2 to output a frequency-divided clock 8 having a half frequency. The data selection signal generator 10 receives the transmission clock 6 from outside the transport stream multiplexing unit Tx and outputs a data selection signal 12 to the data selector 14. The data selection signal 12 switches in synchronization with the transmission clock 6. This is equivalent to switching in units corresponding to one data constituting the transport stream. The data selector 14 selects only one transport stream and outputs it to the transmission path 15 by switching the transport streams input from the plurality of encoders (1, 2) according to the data selection signal 12. Here, the rate at which individual transport streams are selected is set to be equal for all transport streams. That is, when two transport streams are multiplexed, selecting the two transport streams alternately in synchronization with the transmission clock 6 in order to select them at the same rate realizes switching at the same rate. The easiest way to do that.

The synchronous code detector 5 includes a transport stream 3 output from the encoder 1 in synchronization with the frequency-divided clock 8 from the clock frequency divider 7, and the encoder 2 outputs the frequency-divided clock 8 from the clock frequency divider 7. The transport stream 4 output in synchronization with the transport stream 4 is analyzed, and a synchronization code is detected from both transport streams. When the synchronization code is detected by the synchronization code detector 5, a read permission signal 11 for outputting the synchronization code position of the transport stream 4 based on the synchronization code position of the transport stream 3 is output to the data storage 9. I do. For details of the read permission signal 11,
This will be described together with the description of the data storage 9. The data accumulator 9 writes and accumulates the transport stream 4 output from the encoder 2 in synchronization with the divided clock 8 from the clock divider 7. Further, the data storage 9 stores the transport stream 4 stored therein until the read permission signal 11 is output from the synchronization code detector 5.
Is not output. When the read permission signal 11 becomes valid,
The transport stream 13 stored in the data storage 9 starts to be output. When the output timing of the read permission signal 11 is viewed on the basis of the transmission clock 6, the synchronization code of the transport stream 3 output from the encoder 1 is the next clock input to the data selector 14. The read permission signal 11 is output so that the timing at which the synchronization code of No. 4 is input to the data selector 14. That is, the transport stream 4 is stored in the data storage 9 by the number of data from when the synchronization code of the transport stream 4 is output from the encoder 2 to when the synchronization code of the transport stream 3 is output from the encoder 1. That is, the synchronization code output timing of the transport stream 4 is delayed. As a result, the data selector 14 switches the plurality of input transport streams and outputs them to the transmission path 15 in accordance with the data selection signal 12 from the data selection signal generator 10. For example, when multiplexing two transport streams,
The two transport streams are alternately switched according to the data selection signal 12 and output to the transmission path 15. As described above, when multiplexing two transport streams, the simplest method is to alternately select every one data in synchronization with the transmission clock 6, and the transport stream 24 output to the transmission path 15 is selected. Means that transport streams 3 and 4 are alternately multiplexed for each data.

Next, a case where two transport streams of transport streams 3 and 4 are multiplexed by such a transport stream multiplexing unit Tx will be described with reference to FIG. The transport stream 3 is a transport stream that constitutes one packet with a total of n pieces including the synchronization code 81 and (n-1) pieces of data from A1 to A (n-1). Similarly, the transport stream 4 has the synchronization code 83 and B1
This is a transport stream that forms one packet with a total of n pieces of (n-1) pieces of data from (n) to B (n-1). These two transport streams are output from each encoder in synchronization with a divided clock 8 generated by dividing the transmission clock 6 by 2 by the clock divider 7. It becomes one data. The data selection signal 12 output from the data selector 14 is used for alternately multiplexing the transport stream 3 and the transport stream 4 for each data as described in the configuration example of the transport stream multiplexing unit x. Then, it changes every clock of the transmission clock 6. For example, when the data selector 14 operates to select the transport stream 3 when the data selection signal 12 is at the high (H) level and to select the transport stream 4 when the data selection signal 12 is at the low (L) level, the multiplexed data is used. The contents of the transport stream 24 include a synchronization code 81, a synchronization code 83, data A1, B1, A2, B2,.
-1) and B (n-1) in order, resulting in a total of 2n data strings. At this time, the arrangement order of the synchronization code portion is that the synchronization code 81 of the transport stream 4 output from the encoder 2 must follow the synchronization code 81 of the transport stream 3 output from the encoder 1. . This order is important when the transport stream separation unit Rx separates the transport stream 16.

Next, the transport stream separating unit R
The configuration and operation of x will be described. The transport stream separation unit Rx includes a synchronization code detector 19, a clock frequency divider 20, and decoders 17 and 18. The synchronization code detector 19 has the same function as the synchronization code detector 5 of the transport stream multiplexing unit Tx, and multiplexes the transport stream 1 input from the transmission path 15.
6 is analyzed, and a synchronization code in the transport stream 16 is detected. As described above, in the transport stream 24 multiplexed by the transport stream multiplexing unit Tx, if the synchronization codes of a plurality of transport streams are continuous and it is determined in which order the synchronization codes are arranged, it is possible to determine which encoder Can be recognized as a synchronization code of the transport stream output by the. When detecting the continuous synchronization code, the synchronization code detector 19 outputs a reset signal 25 at a predetermined position. The clock divider 20 is connected to the transmission path 15
Divides the frequency of the transmission clock 21 input from the controller, and generates a clock signal having a lower frequency. Here, when multiplexing two transport streams, the transmission clock 2
1 is divided by 2 to generate a divided clock signal having a half frequency, and the divided clock 22 is output to the decoder 17 and the divided clock 23 is output to the decoder 18. Clock divider 20
When the reset signal 25 is input from the sync code detector 19, the decoder 17 outputs the frequency-divided clock 22 so that the decoder 17 can take in the sync code 81 in the multiplexed transport stream 16. Similarly, when the reset signal 25 is input, the frequency-divided clock 23 is output so that the timing at which the decoder 18 can capture the synchronization code 83 in the transport stream 16 is obtained.

Similarly, when three or more transport streams are multiplexed, the corresponding decoder is determined from the position of the continuous synchronization code, and a frequency-divided clock having a timing such that the decoder can capture the synchronization code is output. Since all frequency-divided clock outputs are repeated at regular intervals, data of one transport stream multiplexed at the same regular intervals can be continuously captured by one corresponding decoder. Paying attention to one of these divided clocks, the clock is output from the clock divider at regular intervals, and from the viewpoint of the decoder to which the divided clock is connected, the multiplexed transport stream is Transport stream data to be taken out also appears at regular intervals. That is, the data rate is the same as the transport stream data output by the encoder before being multiplexed by the transport stream multiplexing unit Tx on the transmission side. Therefore, decoder 1
The transport stream capture of 7, 18 does not require the ability to process high data rate data such as data on a transmission path. Further, in this configuration, there is no need to have a data storage for data rate conversion on the transmission side and the reception side, and only one data storage for adjusting the position of the synchronization code is required. Becomes easier.

[0020]

As described above, when the structure of the present invention is used, even if a plurality of transport streams are multiplexed, it is not necessary to process a high data rate by a decoder unlike the conventional transport stream multiplexing. A compressed data multiplexing apparatus having a simple configuration can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment for realizing a compressed data multiplexing device of the present invention.

FIG. 2 is a diagram showing a structure of a general transport stream and a packet;

FIG. 3 is a diagram showing before and after data rate conversion of a transport stream.

FIG. 4 is a diagram showing insertion of a packet by a data rate conversion process;

FIG. 5 is a block diagram showing a configuration of a conventional compressed data multiplexing device.

FIG. 6 is a diagram showing two multiplexed transport streams.

[Explanation of symbols]

1, 2: Encoder, 3, 4, 16, 24: Transport stream, 5, 19: Synchronous code detector, 6, 2
1: transmission clock, 7, 20: clock divider, 8, 2
2, 23: frequency-divided clock, 9: data accumulator, 10: data selection signal generator, 11: read enable signal, 12:
Data selection signal, 13: stored transport stream, 14: data selector, 15: transmission line, 17, 18:
Decoder, 25: reset signal.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04N 7/24

Claims (6)

[Claims] 1. A data transmission method for multiplexing and transmitting a plurality of transport streams on a transmission side and separating and reproducing the plurality of transport streams from the multiplexed transport stream received on a reception side. A method of multiplexing the plurality of transport streams, multiplexing the plurality of transport streams in units of a predetermined data length. 2. The data transmission method according to claim 1, wherein each synchronization code and each data of the plurality of transport streams are multiplexed based on each synchronization code included in the plurality of input transport streams. A data transmission method characterized by performing array multiplexing in units of a predetermined data length according to an order and transmitting the data. 3. The data transmission method according to claim 1, wherein an appearance order of synchronization codes of each of the transport streams is detected from the multiplexed transport stream received at a receiving side, and based on the detected order. A data transmission method comprising separating and reproducing a plurality of transport streams. 4. A data transmission apparatus for multiplexing and transmitting a plurality of transport streams on a transmitting side and separating and reproducing the plurality of transport streams from the multiplexed transport stream received on a receiving side. And a data transmission unit on the transmitting side for multiplexing the plurality of transport streams in units of data of a predetermined length when multiplexing the plurality of transport streams. 5. The data transmission apparatus according to claim 4, wherein said data multiplexing means includes means for detecting each synchronization code included in the plurality of input transport streams, and said plurality of transformers at said detection timing. A data transmission device comprising means for array-multiplexing each synchronization code and each data of a port stream in units of a predetermined data length according to a multiplexing order. 6. The data transmission apparatus according to claim 4, wherein the receiving side detects, from the received multiplexed transport stream, an appearance order of synchronization codes of the respective transport streams, and A data transmission device comprising: means for separating and reproducing the plurality of transport streams based on a detection result.