JPH02108348A - Packet transmission method - Google Patents

Abstract

PURPOSE:To change the data packing format of a packet without affecting a terminal during communication by disassembling the data part of the packet based on the data packing format designated with format information applied into the received packet. CONSTITUTION:By format information 8 included in a packet 6, a packet multiplexing device 1b on a receiving side knows that the data packing format is on a B side, and disassembles the packet 6 based on the data packing format on the B side. Consequently, even when a packet multiplexing device 1a on a transmitting side dynamically alters the packing format, the packet multiplexing device 1b on the receiving side can suitably disassemble the packet 6 following to the alteration. Accordingly, even when the packing format of the packet 6 is altered by the alteration of a communication pattern due to the generation or disappearance of a call, the receiving side is completely synchronized with the alteration, the disassembling operation of the data can be altered, and the terminal during the communication can continuously communicate without the effect.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a packet transmission method in a high-speed packet switching network in which all information such as audio, two images, and data is packetized and transmitted and exchanged.

[Conventional technology]

In conventional packet switching networks, data from one terminal is collected, a header containing information such as a destination address is added to the data, and a packet is assembled. Here, if the data length of the packet is short, the header occupies a large proportion of the packet, making it impossible to use the line effectively. It is generally known that increasing the length of the data portion of a packet improves the efficiency of line usage.

On the other hand, since the packet switching system allows flexible use of line bandwidth and is suitable for transmitting and exchanging multiple traffic, it has been considered to realize single line switching using the principle of packet switching. however.

Although packet switching is more efficient in using lines, it has the drawback of large delays. This is largely due to the data accumulation time, which is caused by accumulating data to a size that allows the packet to be assembled and then sending it out when assembling the packet. Circuit switching is characterized by transmitting information without changing the form of information (with short delay time), and in order to realize circuit switching using the principle of packet switching, a major issue is how to shorten the delay time. One way to reduce the delay time in packet switching is to reduce the amount of data stored during packet assembly, but this shortens the length of the data portion of the packet and reduces line utilization for the reasons mentioned above. Efficiency decreases.

As a packet transmission method to solve this problem,
Of data from multiple devices such as terminals.

A method of collecting data sent to the same switching device sequentially, assembling it into a single packet, and transmitting this packet.
That is, an effective method is to combine data from multiple devices having the same destination switching device into one packet. FIG. 3 is an explanatory diagram showing an example of such a carpooling system, and is similar to FIG.

(1a) and (1b) are packet multiplexing devices with switching functions, (2) are transmission lines, (3m) to (3c),
(3d) to (3f) are terminals accommodated in the packet multiplexing devices (la) and (lb), respectively, and (4a)
), (4b), and (4c) are the terminals (3a
), (3b).

(3c) is the transmission data, (5) is a header consisting of a destination address, etc., and (6) is a packet consisting of this header (5) and the above transmission data (3a) to (3c).

In FIG. 3, transmission data (4&) from terminal (3a) is transmitted to terminal (3d), transmission data (4b) from terminal (3b) is transmitted to terminal (3e), and data (4c) from terminal (3c) is transmitted to terminal (3d).
3f).

The packet multiplexer (1m) is the accommodation terminal (3a),
(3b).

Each transmission data in (3c) is packetized and transmitted, and here, the terminal (3d) receiving the transmission data of the accommodation terminal (3a), (3b), (3c) of the packet multiplexing device (1a) , (3e), (:H) is a single packet multiplexing bag f! f (la), the packet multiplexer (1a) receives one transmitted packet from the packet multiplexer (I
b) It is possible to combine multiple pieces of data transmitted to the terminals accommodated in the terminal. That is, the header of packet (6) (
5) includes information that identifies the packet multiplexing device that is the destination of the packet.

Since the destination packet multiplexing device for data (4a), (4b), and (4c) is the same, the same packet (6
). Therefore, the packet multiplexing device (fall, each terminal (3a), (3
b) When packetizing the transmission data from (3c), do not packetize the transmission data individually for each terminal,
Each terminal (3a), (3b).

Data from (3C) (4a), (4b), (
4c) and then the header (5) to assemble one packet (6).

This is sent to the destination packet multiplexer (lb). By combining the data of multiple terminals into the data portion of one packet in this way, the data length of one packet (6) can be increased without having to accumulate a large amount of data from two individual terminals until it reaches a predetermined data length. It would be possible to suppress the reduction in transmission efficiency due to the overhead of the header (5) and at the same time reduce the delay time due to data accumulation when assembling the packet (6).

By the way, in such a system, the multiplexer on the receiving side of the packet (6) needs to know how the data of a plurality of terminals are multiplexed in the data portion of the received packet (6).

When circuit switching is implemented using packet switching, the data that is output from the terminal at a constant rate is packetized, but by packetizing the data at a constant cycle and selecting the packetization cycle appropriately.

The data length for each terminal included in the packet (6) can be made constant for each packet. Therefore.

Since the padding format of the data portion of the packet (6) transmitted and received between a pair of multiplexing devices is fixed, the multiplexing device on the receiving side uses the packet (6) based on the predetermined padding format. All you have to do is decompose it. However, in the case of single-circuit switching, as mentioned above, the data length for each terminal changes for each packet (6), and although the data packing format does not change dynamically, calls occur and disappear. This is because the number of terminals communicating between a pair of multiplexing devices varies.

It is also necessary to change the data packing format from the perspective of line utilization efficiency. Due to the occurrence and disappearance of such calls, the multiplexer (1a) on the transmitting side of packet (6)
If you change the padding format of the data part of .

The multiplexer (1b) on the receiving side must synchronize with this and change the way it decomposes the data portion of the packet (6). In particular, in order to change the packing format of the data portion of packet (6) without affecting the two terminals in communication, it is necessary to completely synchronize the change on the sending and receiving sides. As a method of synchronizing the packing format change on the sending side and the receiving side, it is necessary to check how the data of multiple terminals is packed in the data part of each packet (6) in its header (5) sequentially. One possible method would be to include information indicating whether or not there is a child. This means that even if the data stuffing format changes dynamically, the receiving side can decompose the data part in perfect synchronization, but the information representing the data stuffing format increases the length of the header and reduces transmission efficiency. This is not preferable because it lowers the value. As another method, a method using time is proposed in Technical Report 5E84-39 of the Institute of Information, Electronics and Communication Engineers of Japan Switching System Study Group. In this method, the sending and receiving sides agree in advance on the packing format switching time using exchange control packets, and after this time,
This is to change the data stuffing format of the packet (5) transmitted by the transmitting side to the same data stuffing format when the packet (6) is disassembled on the receiving side. However, even with this method, if the transmission delay time of the packet (6) is large, it is difficult to achieve complete synchronization between the assembling side of the packet (6) and the side of disassembling the packet (6).

[Problem to be solved by the invention]

As explained above, when circuit switching is realized by packet switching, in order to reduce delay time, data from multiple devices having the same destination packet multiplexing device must be
An effective method is to combine two packets (6), but in this method, if the number of terminals communicating between a pair of packet multiplexers changes due to the occurrence and disappearance of calls, this will affect the two terminals currently communicating. There is a problem in that it is difficult to completely synchronize changes in the data filling format of the data portion of a packet without causing any changes.

This invention was made to solve these problems, and it is possible to efficiently realize circuit switching with low delay time by combining W, number data into one packet, and to efficiently realize line switching with low delay time by packet switching. The purpose is to obtain a highly reliable packet transmission system by completely synchronizing changes in the data filling format of the data portion of the packet on the receiving side and on the receiving side.

[Means to solve the problem]

The packet transmission method according to the present invention sets a plurality of types of data stuffing formats in common for the packet multiplexing device on the packet transmitting side and the receiving side when assembling the data portion of one packet from data from multiple devices, On the packet sending side, data from multiple devices is assembled as a data part of one packet based on one of the plurality of types of data stuffing formats mentioned above, and format information indicating the data stuffing format is added and transmitted. , among the multiple types of data stuffing formats listed above on the packet receiving side.

The data portion of the received packet is decomposed based on the data stuffing format specified by the format information in the received packet.

[For production]

In this invention, the data portion of a received packet is decomposed based on the data stuffing format specified by the format information in the received packet from among a plurality of data stuffing formats preset on the receiving side of the packet. Therefore, by adding format information that simply specifies the data stuffing format used to assemble the data section to the transmitted packet, the sending side's data Changes in the data stuffing format on the receiver side are completely synchronized with changes in the stuffing format.

The data stuffing format can be changed without affecting the call in progress.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings. 1st
2 are explanatory diagrams of a packet transmission method according to an embodiment of the present invention, and only the receiving side is shown. In FIGS. 1 and 2, (11 to (6)) are similar to the conventional example described above. (7) shows multiple types of stuffing formats of each transmission data in the data part of packet l-(61). table, and is a packet multiplexer on the sending and receiving sides (
Ia) and (lb) are set to the same value.

In this embodiment, the data filling format is set to two sides, side A and side 8. (8) is format information indicating which side A or B of the above two side data packing formats is used when assembling the data portion of the packet on the transmitting side.

It is given in the packet (6) in 5).

Figure 1 shows a terminal (3a) and a terminal (3d) similar to the conventional example above.
), the terminal (3b) and the terminal (3el), the terminal (3c) and the terminal (3f) communicate, and the packet multiplexing device (l) on the sending side communicates.
In a), the packet (
6) shows the case where the data section is assembled. At this time, the format information (8) indicates side A, so the receiving side packet multiplexer (1b) receives the data based on the data stuffing format of side A of the two preset data stuffing formats. Decomposes the data part of the packet and sends the data (4a)
, (4b), (4c) are the terminals (3dl, (3
e) will be sent to (3f).

FIG. 2 shows the operation after the communication of the terminal (3d) has been completed in the state shown in FIG. 1 and the filling format of the data portion of the packet (6) has been changed. Here, the terminal (3d
) is no longer sent, and the packet (4a) is no longer sent.
6) is data from two terminals (4b) and (4c) 2
Let's share and choose Tsunomi. Packet (6) in Figure 1 and packet (6) in Figure 2 have different data packing formats, but the packet multiplexer (1b) on the receiving side
Based on the format information (8) included in 6), it is known that the data stuffing format is side B, and the packet (6) is disassembled based on the data stuffing format of side B. Therefore.

Even if the packet multiplexing device (1a) on the sending side dynamically changes the packing format, the packet multiplexing device (]b) on the receiving side can follow the change and appropriately decompose the packet (6)2. For example, even if the packing format of packet (6) is changed due to a change in the communication pattern due to the generation or disappearance of a call, the receiving side can completely synchronize with the change and change the data disassembly operation. The terminal can continue to communicate without being affected. Also, according to this method, regarding the data filling format, A is added in the header (5).
Since the information only specifies either the side or the B side, the header (5) does not become large.

In the description of the above embodiment, there are only two data packing formats, side A and side B, but in general, there are many patterns of combining data from multiple terminals in a packet depending on the communication form. Therefore, if the data filling format for two sides is insufficient, it is necessary to prepare another data filling format. For example, if a communication channel for management is provided between the packet and the multi-electronic device, and another data stuffing format newly prepared on the sending side is loaded onto the receiving side through this channel, a large number of data stuffing can be performed. ' format can be used. That is, of the data stuffing formats on side A or B in the data stuffing format table on the receiving side, the side that is not being used at that time is rewritten to the data stuffing format sent from the sending side.

Once the sending side confirms that this loading is complete, it can use the new data stuffing format.

In addition, in the above embodiment, the case where three terminals are connected to the packet multiplexer and one of these terminals stops communication has been explained, but the present invention is not based on this, and the packet multiplexer has an arbitrary function. This also applies when a number of terminals are connected, and any number of terminals in parentheses stop or start communication.

〔Effect of the invention〕

As described above, according to the present invention, among the plurality of data stuffing formats commonly set on the sending side and the receiving side, the data stuffing format specified by the format information added to the received packet is selected. Since the data part of the packet is decomposed based on the data part of the packet, it is possible to completely synchronize changes in the data packing format on the sending and receiving sides without increasing the size of the packet header. This has the advantage that the data filling format of the packet can be changed without affecting the data.

[Brief explanation of drawings]

FIGS. 1 and 2 are explanatory diagrams showing an embodiment of the packet transmission system of the present invention, and FIG. 3 is an explanatory diagram showing a conventional packet transmission system. In the figure, (Ia) is a transmitting side packet multiplexing device. (1b) is the receiving side packet multiplexer # (3a) to (
3f) is the terminal, (4a) to (4c) are the transmission data, (
5) is a header, (6) is a packet, (7) is a data filling format table, and (8) is format information. Please note the figure. The same reference numerals in each figure indicate the same or corresponding parts.

Claims (1)

[Claims] When assembling the data portion of one packet from data from multiple devices, multiple types of data stuffing formats are set in common for the packet multiplexing devices on the packet sending and receiving sides, and the packet sending side uses the above multiple types of data. Assembling data from multiple devices as a data portion of one packet based on one of the data stuffing formats, adding format information indicating the data stuffing format, and transmitting it;
A packet transmission method characterized in that the packet receiving side decomposes the data part of the received packet based on the data stuffing format specified by the format information in the received packet from among the plurality of data stuffing formats. .