JP2004080139A - Method for controlling sequence of packet in multi-link communication system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable an unarrived packet and an erased packet to be distinguished without monitoring the waiting states of all the packets under the sequence control of the packets and to enable a packet transfer delay due to waiting for an erased packet to be reduced. <P>SOLUTION: The sequence number of the packet reached at present time point, the sequence number (expected<SB>-</SB>seqno) expected to reach next, the minimum sequence number of the packet existing in a buffer, the sequence number (arrived<SB>-</SB>seqno [i]) of the packet reached at last at a port of a transmission line i, and the minimum sequence number of the arrived<SB>-</SB>seqno [i] are managed. A sequence control algorithm judges erasure of the unarrived packet when the packets of a sequence number larger than the expected<SB>-</SB>seqno arrived from all the transmission lines A, B and C, and a transfer process of the packet in the buffer is advanced. If the packet exists in the buffer, a time-out algorithm is performed at each predetermined time. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a packet order control method in a multilink communication system, and in particular, in a packet order control in a multilink communication system, lost packets due to unarrived packets and transmission errors without monitoring the wait state of all packets. The present invention relates to a packet order control method in a multilink communication system, which can distinguish a packet from a packet and reduce a packet transfer delay caused by waiting for a lost packet.
[0002]
[Prior art]
In packet communication, when a plurality of transmission paths can be used in parallel for communication between nodes, a packet is distributed to each transmission path while observing the availability of each of the plurality of transmission paths to perform load distribution. This can improve the throughput.
[0003]
At this time, there is no guarantee that the packets sequentially generated at the transmitting node will arrive at the receiving node in the same order, and the order of the packets arriving at the receiving node will normally be reversed. This is because even packets generated at the same time in the transmitting node have different arrival times at the receiving node due to the speed difference between a plurality of transmission paths and the size of the transmitted packet.
[0004]
As a countermeasure, when distributing packets to each transmission path at the transmitting node, a sequence number indicating the generation order of each packet is added to the packet and the packets are sequentially transmitted, and the sequence number assigned to each packet at the receiving node is transmitted. It is generally performed to control the order by using the above.
[0005]
For this order control, a buffer for order control is prepared in the receiving node. The sequence number of the arriving packet is checked, and if the sequence number is in the order, the packet is transferred to the upper layer (IP layer) as it is, otherwise, the packet is temporarily fetched into the buffer. When a packet is present in the buffer, the contents of the buffer are checked each time a packet arrives, the packets are arranged in the order of consecutive sequence numbers, and the packets are transferred to an upper layer. By performing the order control in this way, the upper layer can receive packets in consecutive sequence number order.
[0006]
As described above, when performing packet communication using a plurality of transmission paths in parallel between nodes as described above, as a method of controlling the order of the packets, a sequence number indicating the generation order of each packet is assigned to the packet at the transmitting node. In general, a method of performing sequence control based on a sequence number at a receiving node is used, and various methods have been proposed to efficiently perform sequence control.
[0007]
Japanese Unexamined Patent Publication No. Hei 5-110600 discloses that each receiving-side communication control device is connected to each other by using a broadcast bus to connect the receiving-side communication control devices with each other and to notify each other of a sequence number of a packet to be captured next. In order to reduce the time required for capturing packets, a method for controlling the order of received packets has been proposed.
[0008]
Japanese Patent Application Laid-Open No. Hei 5-103011 discloses a technique for detecting a failure of a partner communication control processor at an early stage, managing the control state of all packets, and efficiently transferring packets between the communication control processors. A packet transfer control method between communication control processors has been proposed.
[0009]
[Problems to be solved by the invention]
However, the method for controlling the order of received packets proposed in Japanese Patent Application Laid-Open No. 5-110600 does not consider the response to packet loss, and thus introduces this method into a wireless section where the communication environment of the system is unstable. In this case, the delay time when a packet is lost increases. The conventional sequence control method, including the proposed method of controlling the order of received packets, is premised on use in a wired network with a low packet loss rate, and thus it is difficult to cope with packet loss. There is.
[0010]
Further, the packet transfer control method between communication control processors proposed in Japanese Patent Application Laid-Open No. Hei 5-103011 manages the arrival of all packets. It is not suitable for systems requiring high-speed processing.
[0011]
SUMMARY OF THE INVENTION An object of the present invention is to make it possible to distinguish between a packet that has not arrived and a packet that has been lost due to a transmission error without monitoring the waiting state of all the packets in packet order control, and to transfer packets by waiting for a lost packet. An object of the present invention is to provide a packet order control method in a multi-link communication system that can reduce delay.
[0012]
[Means for Solving the Problems]
In order to solve the above problems, the present invention provides a method for performing communication between nodes using a plurality of transmission paths in parallel, using a buffer in a receiving node, and assigning the order control of arrived packets to each packet. In the method of controlling the order of packets in a multi-link communication system based on the sequence number of a packet, the transmitting node assigns a sequence number to each packet in the order of packet generation, and the receiving node Sequence number of the packet to be raised to the next higher layer, the minimum sequence number of the packet existing in the buffer, the sequence number of the last packet arriving for each transmission path, the last arriving for each transmission path Of the sequence number of the minimum value of the There, using these sequence numbers, there is first characterized in that executing the order control algorithm of sequence control gave up the arrival of the packet at the time it is determined that the packet non-arrival disappeared.
[0013]
Further, the present invention has a second feature in that when a packet is present in the buffer, a timeout algorithm is executed at regular intervals.
[0014]
Further, the present invention provides the above-mentioned sequence control algorithm, wherein when all of the sequence numbers of the packets arriving from the plurality of transmission paths exceed the sequence numbers of the next scheduled packets, A packet having a sequence number equal to or smaller than the minimum value of the sequence number of the arriving packet is determined as a lost packet, and a packet having a sequence number equal to or smaller than the minimum value existing in the buffer is determined without waiting for the arrival of the packet. There is a third feature in that it includes a process of transferring to
[0015]
Also, the present invention provides the timeout algorithm, wherein when a packet present in the buffer times out, the timeout algorithm does not wait for an unarrived packet having a sequence number equal to or less than the sequence number of the timed out packet. A fourth feature is that the process includes a process of transferring a packet having a sequence number equal to or smaller than the sequence number of the packet and existing in the buffer to an upper layer.
[0016]
Further, the present invention is characterized in that the timeout time in the timeout algorithm is a transmission delay time when a maximum-size packet is transmitted on a transmission line with the lowest transmission speed among transmission lines waiting for the arrival of a packet. Has a fifth feature.
[0017]
Further, the present invention has a sixth feature in that the timeout time in the timeout algorithm is dynamically changed according to the transmission speed of the transmission path.
[0018]
According to the first feature, by managing the five sequence numbers and executing an order control algorithm, unarrived packets and packets lost due to transmission errors can be identified without monitoring the wait state of all packets. It is possible to make a distinction and reduce the packet transfer delay due to waiting for lost packets.
[0019]
Further, according to the second and fourth characteristics, when a packet existing in the buffer times out, the process can proceed without waiting for the arrival of an unarrived packet whose sequence number is equal to or less than the sequence number of the timed out packet. It is possible to reduce the packet transfer delay.
[0020]
Further, according to the third feature, when all the sequence numbers of the packets arriving from the plurality of transmission paths exceed the sequence numbers of the next arriving packet, the sequence number of the last arriving packet is Since it is possible to determine that a packet that has not arrived below the minimum value has been lost, the process can proceed without waiting for the arrival of such a packet that has not arrived.
[0021]
Further, according to the fifth feature, the time-out period can be appropriately set.
[0022]
Further, according to the sixth feature, even if the transmission speed of the transmission line dynamically changes, the packet transfer delay can be reduced by changing the timeout period to a value corresponding to the transmission speed of the transmission line. It can be reduced according to the operation.
[0023]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a conceptual diagram showing an order control method according to the present invention. Here, as an example, multilink communication in which a transmitting node and a receiving node are connected by three parallel transmission lines A, B, and C. Assume the system. The transmitting node distributes the generated packet to three transmission paths A, B, and C. At this time, a sequence number is assigned to each packet according to the generation order of the packets. The packets distributed to each of the transmission paths A, B, and C are buffered according to the respective transmission speeds and transmitted to the receiving node.
[0024]
In the conventional sequence control, one continuous number is managed without considering the correspondence between the sequence number and each transmission path. However, in the present invention, a packet arriving for each transmission path is received at the receiving node. , And executes a sequence control algorithm and a timeout algorithm using these sequence numbers. This makes it possible to distinguish between a packet that has not arrived and a packet that has been lost due to a transmission error, and to reduce a packet transfer delay caused by waiting for a packet that has not arrived.
[0025]
Also, even if the transmission line speed dynamically changes, by changing the timeout time in the timeout algorithm to a value corresponding to the transmission line speed, the packet transfer delay should be reduced in accordance with the actual operation. Can be.
[0026]
The order control method of the present invention will be specifically described.
The transmitting node distributes traffic passing through the link to each transmission path in packet units according to the transmission speed of each transmission path. For example, if the transmission speeds of the transmission lines A, B, and C are 3: 2: 1, the packets to be distributed are also 3: 2: 1. At this time, a unique sequence number is added to the header of each packet in accordance with the order of generation of the packets. On the other hand, the receiving node buffers the arriving packet, arranges the packet using an order control algorithm, and transfers the packet to an upper layer.
[0027]
The order control algorithm manages the following five parameters (1) to (5) centered on the sequence number seqno of the packet that has arrived at the present time, and performs order control. In addition, a sequence control with a short waiting time is realized.
(1) seqno: sequence number of the packet that has arrived at the current time
(2) expected_seqno: sequence number expected to arrive next
(3) min_buffer_seqno: minimum sequence number existing in the reception buffer
(4) arrived_seqno [i]: sequence number of the last packet arriving at the port of transmission line i (= maximum seqno packet arriving at the port of transmission line i)
(5) arrived_minimum: smallest sequence number in arrived_seqno [i]
[0028]
FIG. 2 shows, as an example, the behavior of the sequence number at the receiving node when packets with sequence numbers 1 to 6 are sequentially transmitted. In this example, first, the first packet arrives at the port (Port) 0 of the transmission path A, and the second packet arrives at the port 2 of the transmission path C while the second packet does not arrive at the port 1 of the transmission path B. It is assumed that the third packet arrives, and subsequently the fourth, fifth, and sixth packets arrive at ports 0, 1, and 2, respectively.
[0029]
The first packet to port 0 is transferred to the upper layer (IP layer) upon arrival, and the third packet to port 2 is stored in the reception buffer because the second packet has not arrived. You.
[0030]
When the fourth, fifth, and sixth packets arrive at ports 0, 1, and 2, respectively, seqno = 6, expected_seqno = 2, and min_buffer_seqno = 3. Also, arrived_seqno [i] is arrived_seqno [A] = 4, arrived_seqno [B] = 5, arrived_seqno [C] = 6, and arrived_minimum = arrived_seqno [A] = 4, respectively.
[0031]
The order control algorithm manages these sequence numbers and controls the order. FIG. 3 is a flowchart showing an example of the order control algorithm of the present invention, which is executed when a packet arrives. Hereinafter, description will be made along this flow.
[0032]
First, when a seqno packet arrives at the port of the transmission line i, the seqno is set to arrived_seqno [i] (S1). Next, it is determined whether or not seqno is equal to or less than expected_seqno (S2). Note that seqno may be smaller than expected_seqno due to a timeout algorithm or an error described below, but this is rare. If it is determined in S2 that seqno is not less than expected_seqno, the arriving packet, that is, the seqno packet is stored in the reception buffer, and a time stamp is set to the packet (S3). This time stamp is used in a timeout algorithm described later.
[0033]
Next, it is determined whether or not arrived_seqno [i] of all the transmission lines i is greater than expected_seqno (S4). If not, the process returns to a packet reception waiting state (S5). Further, in S4, if it is determined that arrived_seqno [i] of all the transmission lines i is greater than expected_seqno, the arrived_seqno [i] of all the transmission lines i is set to arrived_minimum (S6), and then all the packets below the uppermost of the arrived_minimum are set. Transfer to the layer (IP layer) (S7). Subsequently, expected_seqno is set to arrived_minimum + 1.
[0034]
The above steps S4 to S8 are processing [A] in which unarrived packets equal to or less than arrived_minimum are discarded as they have been lost, and their arrival is abandoned and all packets equal to or less than arrived_minimum present in the buffer are transferred to the upper layer. The reason for giving up the arrival of unarrived packets equal to or less than "arrived_minimum" is based on the fact that the order of packets does not reverse in one transmission path. Subsequent to S8, a process [B] of transferring the remaining packets in the reception buffer to the upper layer in order is executed.
[0035]
If it is determined in S2 that seqno is equal to or less than expected_seqno, the arriving packet, that is, the seqno packet is transferred to the upper layer (S9). Note that, here, the packet of seqno smaller than expected_seqno is transferred to the upper layer without discarding, and is left to be processed by the upper layer. Next, it is determined whether or not seqno is equal to expected_seqno (S10). If it is determined that seqno is not equal to expected_seqno, the process returns to a packet reception waiting state (S5). If it is determined in step S10 that seqno is equal to expected_seqno, the process executes a process [B] of transferring the packets in the reception buffer to the upper layer according to the order, after replacing expected_seqno + 1 with a new expected_seqno (S11).
[0036]
In the process [B], it is determined whether or not an expected_seqno packet exists in the reception buffer (S12), and if it exists, the packet is transferred to an upper layer (S13) and the expected_seqno + 1 is replaced with a new expected_seqno and new expected_seqno. The update process (S14) is repeatedly executed, and when the expected_seqno packet does not exist in the reception buffer, the process returns to the packet reception waiting state (S5).
[0037]
As described above, the order control algorithm of FIG. 3 utilizes the fact that packet order inversion does not occur in one transmission path, and when the sequence number of a packet arriving from all transmission paths becomes larger than expected_seqno, It gives up the unarrived packets equal to or less than arrived_minimum and transfers the packets equal to or less than arrived_minimum in the reception buffer ([A] in FIG. 3). Also, the remaining packets in the reception buffer are transferred in a controlled order (part [B] in FIG. 3).
[0038]
FIG. 4 is a flowchart illustrating an example of the timeout algorithm according to the present invention, which is executed at regular intervals when a packet exists in the reception buffer. In accordance with this timeout algorithm, if there is a packet exceeding the specified allowable waiting time in the reception buffer, the sequence number of the packet having the longest elapsed time from the timeout is set to "expected_seqno", and then the "expected_seqno" existing in the reception buffer is set. The following packets are transferred to the upper layer, and successive packets from expected_seqno are sequentially transferred to the upper layer (IP layer).
[0039]
Hereinafter, description will be given along the flow of FIG. When there is a packet in the reception buffer, an interrupt process is started at regular intervals, and it is determined whether or not a packet exceeding the specified allowable waiting time exists in the reception buffer (S15). Whether or not the specified allowable waiting time has been exceeded can be determined based on the time stamp set in S3.
[0040]
The allowable waiting time may be a transmission delay time when a maximum-size packet is transmitted on the transmission line with the lowest transmission speed among the transmission lines waiting for the arrival of the packet. This transmission delay time, that is, the time when it is determined that a packet does not arrive on the transmission line even after waiting longer than that time is set as a timer value, and the determination in S15 can be made based on this timer value.
[0041]
In S15, if it is determined that there is no packet exceeding the specified allowable waiting time in the reception buffer, the process returns to the packet reception waiting state (S5). The _seqno of the long packet is set as expected_seqno (S16).
[0042]
Next, it is determined whether or not min_buffer_seqno is equal to or less than expected_seqno set in S16 (S17). If it is determined that min_buffer_seqno is equal to or less than expected_seqno, the process [C] is executed. After setting expected_seqno + 1 as new expected_seqno (S18), the process [D] is executed.
[0043]
The process [C] is a process of giving up the arrival of unarrived packets equal to or less than the expected_seqno. The loop of S17, S19, and S20 transfers the packet of the min_buffer_seqno to the upper layer (S19) and updates the min_buffer_seqno (S20). The packet is repeatedly executed to sequentially transfer packets below expected_seqno existing in the reception buffer to the upper layer.
[0044]
The process [D] is a process of sequentially transferring the packets in the reception buffer from the next of the expected_seqno replaced in S16 to the upper layer. First, there is a packet of the expected_seqno (= expected_seqno + 1) in the reception buffer. It is determined whether or not (S21). If it is determined that there is no expected_seqno packet in the reception buffer, the process returns to the packet reception wait state (S5). If it is determined that the packet does not exist, the upper layer of the packet is removed by the loop of S21, S22, and S23 until the packet no longer exists. (S22) and the update process (S23) in which expected_seqno + 1 is set as new expected_seqno.
[0045]
Next, a specific example of the operation of the order control algorithm will be described with reference to FIGS. FIG. 5 shows an example of the flow of packet arrival at a certain point in time. In this example, the first and second packets arrive in order from the transmission paths A and B, the fifth and seventh packets arrive from the transmission path A, and the ninth packet arrives from the transmission path C. It is assumed that the packet arrives, the 10th packet arrives from the transmission path B, and the 3rd, 4th, 6th, and 8th packets have not arrived.
[0046]
The first and second packets arriving from the transmission lines A and B are in order and are immediately transferred to the upper layer (S9). The fifth and seventh packets arriving from the transmission path A and the ninth packet arriving from the transmission path C have the possibility that an unarrived third packet may still arrive from the transmission path B. Is stored in
[0047]
When the tenth packet arrives from the transmission line B, seqno = 10, expected_seqno = 3, and min_buffer_seqno = 5. Also, arrived_seqno [i] is arrived_seqno [A] = 7, arrived_seqno [B] = 10, arrived_seqno [C] = 9, and arrived_minimum = arrived_seqno [A] = 7, respectively.
[0048]
At this point, a packet having a sequence number larger than expected_seqno (= 3) has arrived at all the ports of the transmission paths A, B, and C, and the unarrived packets 3, 4, and 6 are delayed. Instead, it is determined that it has disappeared (S4).
[0049]
Then, according to the order control algorithm, packets up to arrived_min = arrived_seqno [A] = 7 existing in the reception buffer are sequentially transferred to the upper layer (S7). In this example, the fifth and seventh packets are transferred to the upper layer.
[0050]
FIG. 6 shows a state immediately after the fifth and seventh packets have been transferred. In this state, seqno = 10, expected_seqno = 8, and min_buffer_seqno = 9. Also, arrived_seqno [i] is the same as before the transfer, arrived_seqno [A] = 7, arrived_seqno [B] = 10, arrived_seqno [C] = 9, and arrived_minimum = arrived_seqno [A] = 7.
[0051]
As described above, the fifth and seventh packets are transferred to the upper layer, and the arrival of the third, fourth, and sixth packets is abandoned according to the process [A] in FIG. In this example, it is assumed that the eighth packet is also lost. However, in the order control algorithm, since it is determined that the eighth packet may still arrive from the transmission line A, the ninth packet is lost. The 10th packet waits in the reception buffer and waits for the arrival of a new packet.
[0052]
Next, a specific example of the operation of the timeout algorithm will be described with reference to FIGS. In this example, the first and second packets arrive in order from the transmission lines A and B, the fifth and seventh packets arrive from the transmission line A, and the eighth packet arrives from the transmission line B. It is assumed that the third, fourth, and sixth packets have not arrived.
[0053]
The first and second packets arriving from the transmission paths A and B are immediately transferred to the upper layer (S9). The fifth, seventh, and eighth packets are stored in the reception buffer because the third packet may still arrive from the transmission path C.
[0054]
When the eighth packet arrives from the transmission line B, seqno = 8, expected_seqno = 3, and min_buffer_seqno = 5. Also, arrived_seqno [i] is arrived_seqno [A] = 7, arrived_seqno [B] = 8, and arrived_minimum = arrived_seqno [A] = 7, respectively.
[0055]
Here, if the ninth packet arrives from the transmission line C, the third, fourth, and sixth packets are abandoned according to the above-described sequence control algorithm, and the fifth packet is transferred to the upper layer. (S7). However, assuming that the fifth packet times out before that (S15) and the timeout algorithm is executed, the fifth packet is transferred to the upper layer by the timeout algorithm (S19). In this example, only the fifth packet is transferred to the upper layer by the timeout algorithm of FIG.
[0056]
FIG. 8 shows a state immediately after the fifth packet has been transferred to the upper layer. In this state, seqno = 8, expected_seqno = 6, and min_buffer_seqno = 7. Also, arrived_seqno [i] is the same as before the transfer, arrived_seqno [A] = 7, arrived_seqno [B] = 8, and arrived_minimum = arrived_seqno [A] = 7, respectively.
[0057]
As described above, the fifth packet is transferred to the upper layer, and the arrival of the third and fourth packets is abandoned according to the process [C] in FIG. Also in this example, as in the case of the order control algorithm, it is assumed that the sixth packet is also lost. However, both the seventh and eighth packets existing in the reception buffer are not timed out yet. , 6th packet may still arrive from the transmission path C, so that the 7th and 8th packets wait in the reception buffer and wait for the arrival of a new packet. In this case, the seventh and eighth waiting times can be the time required for the completion of transmission when it is assumed that the packet of the maximum size has been transmitted to the transmission path C.
[0058]
Although the embodiment in which the timeout algorithm is executed together with the order control algorithm has been described above, the packet transfer delay can be reduced by executing only the order control algorithm. In this case, the setting of the time stamp in S3 is unnecessary. .
[0059]
The present invention manages the sequence numbers of packets arriving for each transmission path, and executes a sequence control algorithm and a timeout algorithm using the sequence numbers, thereby reducing the transfer delay in consideration of packet loss. A wireless multi-link system having a relatively large packet loss rate, such as a wireless access system having a wireless multi-link configuration in which a wireless link connecting wireless nodes is configured by a plurality of wireless channels. The present invention is effective when applied to a fixed broadband wireless access system (FWA), but can also be applied to a system having a wired multilink configuration in which packet loss is considered.
[0060]
【The invention's effect】
As is apparent from the above description, according to the present invention, by managing the latest sequence number of a packet arriving for each transmission path, the waiting state of all packets is monitored when the order of the packets is controlled. Thus, it is possible to distinguish between a packet that has not arrived and a packet that has been lost due to a transmission error, and reduce packet transfer delay caused by waiting for a lost packet.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram showing a sequence control method according to the present invention.
FIG. 2 is an explanatory diagram of an example of a behavior of a sequence number in a receiving node.
FIG. 3 is a flowchart illustrating an example of an order control algorithm according to the present invention.
FIG. 4 is a flowchart illustrating an example of a timeout algorithm according to the present invention.
FIG. 5 is an explanatory diagram showing an example of a state of a packet by the order control according to the present invention;
FIG. 6 is an explanatory diagram showing another example of the state of the packet by the order control of the present invention.
FIG. 7 is an explanatory diagram showing an example of a state of a packet according to the timeout algorithm of the present invention.
FIG. 8 is an explanatory diagram showing another example of the state of the packet according to the timeout algorithm of the present invention.
[Explanation of symbols]
1-10: Packet, A, B, C: Transmission path

Claims (6)

Multilink communication in which communication between nodes is performed by using a plurality of transmission paths in parallel, and the receiving node uses a buffer to control the order of arrived packets based on the sequence number assigned to each packet. In a method for controlling the order of packets in a system,
At the transmitting node, a sequence number is assigned to each packet according to the packet generation order,
The receiving node manages the following sequence numbers (1) to (5),
(1) the sequence number of the packet that has arrived at the current time,
(2) the sequence number of the packet to be raised to the next higher layer,
(3) the minimum sequence number of the packet present in the buffer,
(4) the sequence number of the last packet arrived for each transmission path,
(5) The minimum value of the sequence number (4) of the last packet arrived for each transmission path,
A multi-link algorithm using the sequence numbers (1) to (5) and executing a sequence control algorithm of sequence control that gives up arrival of the unarrived packet when it is determined that the unarrived packet has been lost. A packet order control method in a communication system. The method according to claim 1, further comprising, when a packet exists in the buffer, executing a time-out algorithm at regular intervals. The sequence control algorithm is configured such that when all of the sequence numbers of the packets arriving from the plurality of transmission paths exceed the sequence numbers of the next scheduled packets, the sequence number of the last packet arriving for each transmission path , Determining that a packet that has not arrived below the minimum value is a lost packet and transferring a packet having a sequence number that is less than the minimum value and that is present in the buffer to the upper layer without waiting for the arrival of the packet. 3. The packet order control method in a multilink communication system according to claim 1, wherein: The time-out algorithm, when a packet present in the buffer times out, without waiting for the arrival of an unarrived packet that is not more than the sequence number of the timed-out packet, without waiting for the sequence number of the packet that has timed out. The method according to claim 2, further comprising a step of transferring a packet having a sequence number in the buffer to an upper layer. The time-out time in the time-out algorithm is a transmission delay time when a maximum-size packet is transmitted on a transmission line with the lowest transmission speed among transmission lines waiting for the arrival of a packet. Item 5. A method for controlling the order of packets in a multilink communication system according to item 2 or 4. 6. The packet sequence control in the multilink communication system according to claim 2, wherein a timeout time in said timeout algorithm is dynamically changed in accordance with a transmission speed of a transmission line. Method.

Images