JPH02112327A - Packet satellite communication system - Google Patents

Abstract

PURPOSE:To reduce a packet header and to improve time slot efficiency by setting the packet number of a first segmented packet at N-1, and setting the packet number of an i-th segmented packet out of third to N-th segmented packets at N-i+1. CONSTITUTION:An original packet 1 is made into the N pieces of packets 4-1 to 4-N segmented to a time slot length on a communication satellite 3 in an earth station 2, and the first packet number of packet headers PH of the divided packets 4-1 to 4-N is set at N-1. Thus, when the total number of the segmented packets can be easily identified by setting the first packet number at +1, and since the total number of the packets is unnecessary to be added to the packet header PH, the packet header PH can be reduced by that portion, and the time slot efficiency can be improved.

Description

[Detailed Description of the Invention] [Summary] Regarding the packet communication system that performs bucket communication using a random access method, the purpose of this invention is to improve time slot efficiency by reducing the packet header of divided packets. In the packet satellite communication system that adopts this method, a ring packet is divided into N divided packets by the length of the time slot on the communication satellite at the earth station, and the packet of the first divided packet is divided into N divided packets. Number N-1
, the packet number of the second divided packet is N-1 or N, and the packet number of the i-th divided packet of the third to Nth divided packets thereafter is N-i+.
1, it is added to the packet header of a fragmented packet and sent, and when the receiving earth station detects a fragmented packet that has been lost due to a collision etc., it adds the packet number of the fragmented packet and requests retransmission. It was configured as follows.

[Industrial application field]

The present invention relates to a bucket satellite communication method that performs bucket communication using a random access method.

In the bucket satellite communication system, as a random access method, -2 is a pure Aloha method in which buckets are sent completely randomly, and in communication satellite -1, divided buckets are divided into 7 timeslots and synchronized with the time slot. A slotted random access method (slotted random access method) is known.

The bucket terminal equipment is used as a very small earth station (VSAT).
ry Small, I A,, perl:ure Te
rminal) and based on the bucket switching network (1
- In a communication system that connects to an earth station and performs bucket communication between a reference earth station and a micro earth station via a communication satellite, control is relatively simple and a relatively high throughput can be obtained. The slotted Aloha method is mainly adopted, and it is desired to facilitate bucket retransmission processing. [Prior art] For example, as shown in Figure 7,
Bucket terminal device (PT) 41 and terminal station device (OPT) 4
2, and also the bucket exchanger 4 of Baku/Toy446.
5 to the reference station device (HUB) 44. communication satellite 4
In the slot Aloha method, communication is carried out via the time slot TS of the communication satellite 43-to.
Divide one node of buff into the length of the time slot ′r
The divided buckets are sent out at timing S.

In that case, the original packet 4 from the packet terminal device 41
In the terminal equipment 42, the time slots TS(7j) on the communication station 43 are divided into 3 times in the ffllt office equipment η44, based on the source information from the exchange 45. For example, those original buckets are divided into parts as shown in FIG. 8, and a bucket header PH is added to each divided bucket. M, C, total number of buckets P A, , , bucket number PN, user control signal υ
If the ring packet is divided into four parts, the total number of buckets PA in the bucket header PH of each divided bucket is
is "4", the packet number PN of the first divided bucket is "1", and the packet numbers PN of subsequent divided buckets are sequentially increased by 1. Therefore, the packet number PN of the final divided bucket is 4. ” becomes.

Communication satellite 43 at terminal station device 42 or reference station device 44
The divided packets received via the packet controller are combined into an original packet and sent to the packet terminal device 41 or the packet switch 45. Furthermore, for fragmented packets that cannot be identified due to collision or the like, a retransmission request is made.

In this case, even if it is impossible to identify the final divided bucket due to a collision or the like, it is possible to identify whether the final divided bucket is missing based on the total number of buckets PA)c.

[Problem to be solved by the invention]

In the slotted Aloha method, by shortening the length of the time slot, the probability of collision is reduced and the divided packets for retransmission processing can be shortened. --.

time slot efficiency is low)
There are drawbacks to doing so.

In addition, if the divided packet 71 is lost due to a collision, etc., it is possible to identify the number of the divided packet from the bucket numbers of the divided buckets before and after it, but if the total number of buckets PA is not added, the final divided bucket This means that the omission cannot be identified. Therefore, adding the total number of buckets PA,
It is necessary to identify which divided bucket of the total number 1" A is missing. Therefore, if the original bucket is long and the total number of buckets PA becomes large, the baggage
This has the drawback that the number of bits in the total number PA and the number PN becomes large, resulting in a decrease in time slot efficiency.

An object of the present invention is to improve time slot efficiency by reducing the size of bucket headers of divided packets.

[Means to solve the problem]

The bucket satellite communication system of the present invention uses a rung, 14 access system, adds a packet number to the packet header of divided packets, but does not add the total number of packets, thereby reducing the size of the packet header.
This will be explained with reference to the figures.

At the earth station 2, the original packet node 1 is divided into N divided packets 4- to the length of the time slot on the communication satellite 3.
1 to 4-N, the packet number of the first divided packet 4-1 is N-1, the packet number of the second divided packet 4-2 is N-1 or N, and the subsequent divided packets 4 The packet numbers from −3 to 4-N are added to the packet header P) of divided packets 4-1 to 4-N, with the first packet set as N −i +1, and sent to the receiving earth station 2. In this system, when a fragmented packet that is lost due to a collision or the like is detected, the packet number of the fragmented packet is added and a retransmission request is made.

[Effect]

By setting the packet number of the packet header PH of divided packets 4-1 to 4-N as N-1 for the first packet, the total number of divided packets can be easily calculated by adding +1 to the first packet number. can be identified. Since the packet number of the final divided packet is 1, it is possible to easily identify the end of reception of the divided packets. If the original packet is too short to be divided, a packet number of N-1=00 will be added.

Also, when the first packet number and the second packet number are set to the same N-1, if the first divided packet is lost, divided packets with the same packet number cannot be continuously received. This allows for easy identification. Also, when the second packet number is N, if the packet numbers of the received divided packets are N, N-2,..., it means that the first divided packet with packet number N-1 has been lost. Can be easily identified.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 2 is a block diagram of a terminal device according to an embodiment of the present invention, and FIG.
The figure shows a block diagram of a reference station device according to an embodiment of the present invention,
11.21 is a packet buffer, 12.22 is a packet division unit, 13.23 is a transmission processing unit, 14.24 is a reception processing unit, 15.25 is a packet combining unit, 16.26 is a retransmission processing unit, 27-1 27-m is a terminal station corresponding section, and 28 is a line concentration control section, which corresponds to the earth station 2 in FIG.

The original packet from the packet terminal device is sent to 7 of the terminal device.
It is stored in the packet bus sofa 11, and the packet division unit 12
At this point, the data is divided according to the length of the time slot yl- on the communication satellite, and a packet number is added. The original packets from the packet switching network are stored in the packet buffer 21 of the reference station device, and are divided by the packet dividing section 22 according to the length of the time slot on the communication satellite, and a packet number is added.

The packet number is the first packet number when the original bucket is divided into N parts.
Set the packet number of the th divided packet to N1, and set the packet number of the second divided packet to the same N-1, or set the number of divisions to N, and set the packet number of the third divided packet to N-2. . Thereafter, the packet numbers are incremented in reverse order so that the packet number of the Nth divided packet becomes 1.

For example, as shown in FIG. 4, the original bucket is divided according to the length of the time slot (TS) on the communication satellite, and the packet header PH of the divided packet is the modem control signal MC.
, bucket number PN, , -L-the control signal UC, and the number of packets PA added in the conventional example is omitted.

As shown in the figure, when the original bucket is divided into N=4, the packet number PN of the first divided packet is
-t-N-1=3, the packet number PN of the second divided packet is N-1=3 or N=4, the packet number PN of the third divided packet is N-2=2, the fourth The packet number PN of the last divided packet is N-3=1
That is.

A packet header P with such a packet number PN
The divided packets to which H is added are transmitted from the transmission processing section 13 to the communication satellite via the antenna, and are also transmitted from the transmission processing section 23 from the antenna via the line concentration control section 28.

The divided buckets received via satellite communication are received (8) by the reception processing units 14 and 24, and are identified by the bucket synthesis unit 15.2.
In step 5, the divided packets are combined to reproduce the ring packet. In addition, by simultaneously transmitting signals from multiple terminal devices, reference station devices, etc. to the same time slot,
If fragmented packets collide, reception cannot be identified, resulting in a missing fragmented packet. When the reception processing unit 1.4.24 identifies this missing divided packet, it adds the packet number and issues a retransmission request. For example, the transmission processing unit 13 sends a retransmission request by adding the bucket number PN of the missing divided packet.

Also, if there is a retransmission request from the receiving side, the retransmission processing unit 16
The bucket buffer 11 is controlled from ↑, and the divided packet corresponding to the packet number of the verified divided packet is transmitted.

FIG. 5 is a bucket sending flowchart according to an embodiment of the present invention, showing a case where a ring packet from a bucket terminal device PT is sent from a terminal device, and this bucket terminal device P
Receive the original bucket from T, the number of divided packets (N)
Check■.

That is, it is determined whether the ring packet needs to be divided depending on whether the length of the ring packet accumulated in the bucket buffer 11 is longer than the length of one time stamp on the communication satellite.
If the ring packet is within the length of the time slot, there is no need to divide it, so "0" is added to the packet i number and the divided packet is transmitted.

Also, if the ring packet is longer than the length of the time slot, it needs to be divided, so the ring packet is divided into multiple parts, and the first
■ Add the same "N-]" to the packet numbers of the second and second packets. That is, in the case shown in FIG. 4, since the number of divided packets N=4, the first and second packet numbers are set to "3". Then, after that, a buff 1 soto number is given by -1 and 7, and the divided packets are transmitted. Then, it is determined whether the packet number has become 112 or not, and if it is the final divided bucket, the packet number is “
1'', the transmission ends with the transmission of this final divided bucket (■).

Then, when there is a retransmission request from the receiving side, the retransmission processing unit 16
Perform retransmission processing in ■. Retransmission processing is performed according to the packet number added to the retransmission request. In other words, it is determined whether it is the first packet or not, and when it does not indicate the first packet, only the divided packets with the specified packet number are retransmitted. The divided packets up to the packet number are retransmitted. 0° FIG. 6 shows a bucket reception flowchart of the present invention.
[Phase] When the reference station device receives the divided packets from the terminal device OPT, the reception processing unit 24 checks the packet number (2). Determine whether this packet number is “0” or not. If it is “0”, the number of divisions of the ring packet is 0.
Since this is a divided packet, reception of one divided packet results in a reception end of 0.

Also, if the packet number is not “0”, the first and second
Check the packet number with the packet number 0, and if it is the same, continue to receive the divided packets up to the packet number "1" @
1. It is determined whether there is a missing segmented packet on the way or not.
At , the bucket is reconfigured [phase] and the reception is completed [phase].

If the first and second packet numbers are different, the first detected packet number is added and the terminal device OP
@ Makes a retransmission request to T. In addition, when fragmented packets with packet numbers up to 'J' are received by continuous Vf, +-', if a fragmented packet in the middle is lost due to a collision, etc., the packet number of the missing fragmented packet is added and the packet is sent to the end. Station equipment OP
Makes a retransmission request to T [phase]. Therefore, at the time of a retransmission request, by indicating whether it is the first packet or not and adding a packet/node number, the transmitting side can indicate the first packet in step [phase] of FIG. “
N-1", the i-th divided packet from the beginning will be retransmitted. Also, if the leading bucket is not indicated, the packet number added to the retransmission request will be "N-1", for example.
”, the packet number of the first packet is “N-1”
Since the number of divided packets is the packet number of the first packet plus 1, the i+1th divided packet from the first packet will be retransmitted.

In step (2) in Figure 5 above, the first packet number is set to "N-1", and the second packet number is set to "N-1".
Accordingly, in the step [phase] of FIG. 6, when the first packet number is smaller than the second packet number by 1, the first and second It is determined that the reception and identification of the divided packet with
The number N of divided packets can be easily identified from ``1'', and since the final divided packet has a packet number of ``1'', it is possible to easily identify the end of transmitting and receiving the divided packets. It is also easy to detect when the information is missing.

〔Effect of the invention〕

As explained above, in the present invention, the original bucket 1 is divided into the length of the time slot on the communication satellite 3, into N divided packets 4-1 to 4-N, and the first packet number is N-1, the second packet number is N-1 or N, the first packet number after that is N-1+1, and a packet header PH with this packet number is added to the divided packet, Since it is transmitted from the earth station 2 such as a terminal station device or a reference station device, and the total number of packets is not added to the packet header, the packet header PH can be made smaller accordingly. Therefore, time slot efficiency can be improved.

In addition, the missing divided packets can be identified when the first packet cannot be identified by receiving the same packet number consecutively, or when the packet number with a value smaller by 1 cannot be identified first, and the last divided packet is missing. Since this is the case when the divided packet with the packet number "1" cannot be identified as being received, the missing divided packet can be easily detected as in the conventional example in which the total number of packets is added.

[Brief explanation of the drawing]

FIG. 1 is a diagram explaining the principle of the present invention, FIG. 2 is a block diagram of a terminal station device according to an embodiment of the present invention, FIG. 3 is a block diagram of a reference station device according to an embodiment of the present invention, and FIG. 4 is a block diagram of a terminal station device according to an embodiment of the present invention. FIG. 5 is a packet sending flowchart of the embodiment of the present invention, FIG. 6 is a packet reception flowchart of the embodiment of the present invention, FIG. 7 is an explanatory diagram of the satellite communication system, and FIG. The figure is an explanatory diagram of a frame structure of a conventional example. 1 is the original frame, 2 is the earth station, 3 is the communication satellite, 4-1~
4-N is a divided packet, and PH is a packet header.

Claims (1)

[Claims] In a packet satellite communication system that employs a random access method, an original packet (1) is sent to an earth station (2) by a communication satellite (
3) Divide the above time slot into N divided packets (4-1 to 4-N), and set the packet number of the first divided packet (4-1) to (N-1). , the packet number of the second divided packet (4-2) is (N-1)
Or (N), and the packet number of the i-th divided packet of the 3rd to Nth divided packets after that is (N-i+1), and the divided packets (4-1 to 4-N
) is added to the packet header (PH) of A packet satellite communication system that is characterized by the following.