JPH08213995A - Data unit receiving method and communication interface device using the method - Google Patents

Abstract

PURPOSE: To reduce the probability that the data unit not inserted erroneously is disposed of by deciding the distance from the present data unit to the next data unit and retaining the present data unit in the case where the distance is shown to be normal. CONSTITUTION: It is decided whether the order of the preceding data unit and the present data unit is normal or not (S101); and if it is normal, the present data unit is retained (S107). If it is abnormal, it is decided that data units existing between the preceding data unit and the present data unit are lost (S103) and the present data unit is retained (S107) in the case that the distance is shorter than a decision distance. However, there is probability that the present data unit exists by erroneous insertion. The order of the present data unit and the next data unit is checked (S105). If the order of the present data unit and the next data unit is normal, it is decided that data units existing between the preceding data unit and the present data unit are lost (S105), and the present data unit is retained (S107).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data unit receiving method, and more particularly to a data unit receiving method which is advantageously applied to a communication interface device.

[0002]

2. Description of the Related Art In packet communication in which data is divided into packets each having a fixed block length and consecutive sequence numbers are given to a receiving side, bit errors occur in a header portion due to a network failure or a transmission error. Occurs and may be sent to recipients with different destinations. As described above, if an illegal packet that should not be received originally is reproduced, the communication quality is deteriorated. Therefore, the illegal packet is discarded using an error detection code or the like. However, this method cannot completely detect an illegal packet.

For example, when an illegal packet is received as a legitimate packet, consecutive sequence numbers become discontinuous due to this erroneous insertion. Therefore, the pseudo packet is inserted into the missing sequence number. However, if a packet with a missing sequence number is received after an incorrect packet, a communication delay of approximately one cycle occurs, and a pseudo packet is inserted between them, resulting in deterioration of communication quality. Therefore, for example, Japanese Patent Laid-Open No. 1-1922
No. 96 or JP-A-1-192298,
A conventional technique for preventing the occurrence of such communication delay is disclosed. The conventional techniques disclosed in these publications will be described below. Although the above publication describes the voice packet reception system, the ATM communication including voice packets will be described here as an example.

Generally, ATM (Asynchronou)
s Transfer Mode communication AAL1 (A
TM Adaptation Layer type
1) The protocol interface device recognizes the order of data by the data order recognition algorithm shown in FIG.

In the ATM communication, the communication data is 53
Communication is performed by dividing the ATM cell into a unit of byte length.

FIG. 3 shows the format of an ATM cell. In the ATM cell header area of the 1st to 5th bytes of the ATM cell, the address information of the communication destination and the like are stored.
Communication data and the like are stored in the ATM cell payload area of the byte. AT in the AAL1 protocol
48 bytes long SAR-PD in M cell payload area
U (Segmentation And Rease)
mbly-Protocol Data Unit; hereinafter referred to as a data unit) is stored.

FIG. 4 shows a data unit (SAR-PD
The format of U) is shown. In the SAR-PDU header area of the first byte of the data unit, as shown in FIG. 5, SCF (Sequence Count Field) storing SN (Sequence Number) is stored.
d), etc., and communication data is stored in the SAR-PDU payload area of the 2nd to 48th bytes.

FIG. 6 shows a schematic diagram of the data unit. FIG. 7 shows the relationship between communication data and data units using the schematic diagram of the data unit of FIG. However, in the SAR-PDU header area of the data unit, "0" to "7"
Only the SNs of are shown.

As shown in FIG. 7, communication data is divided into a plurality of data units for communication. At this time, SNs are stored in the SCF in order from 0. The maximum value of the SN is 7 since the SCF is 3 bits, and after 7 returns to 0 again. That is, SN is 0, 1, 2, 3, 4, 5,
Repeat 6, 7, 0, 1, 2, 3, ....

FIG. 8 shows a schematic block diagram of an interface device of the AAL1 protocol.

In FIG. 8, the transmitting side has the following functions. The data decomposition unit 30 decomposes the input transmission data into a 47-byte SAR-PDU payload length,
The AR-PDU header is added, converted into a data unit, and output to the SN insertion unit 32. On the other hand, the SN generation unit 36
Then, the SNs of “0” to “7” are generated, and the SN insertion unit 32
Output to. In the SN insertion unit 32, the SN generation unit 3 is added to the SCF of the data unit input from the data decomposition unit 30.
The SN input from 6 is inserted, and the data unit is output to the data unit transmission unit 34. The data unit transmission unit 34 adds an ATM cell header to the data unit input from the SN insertion unit 32, converts it into an ATM cell, and outputs it to the ATM network 38.

On the other hand, the receiving side has the following functions. In the data unit receiving section 40, the ATM network 38
In order to extract the data unit from the ATM cell input from, the ATM cell header is deleted and the data unit is output to the SN extraction unit 42. The SN extraction unit 42 extracts the SN from the SCF of the input data unit,
The data unit is output to the SN inspection unit 46 and the data unit is output to the data assembling unit 44. The SN inspection unit 46 inspects the input SN and executes order recognition of the data units.

When the received data unit is normal or abnormal as a result of order recognition, and if abnormal, it is judged whether it is due to loss of the data unit or due to erroneous insertion of the data unit, and it is judged as erroneous insertion. Only, the discard instruction signal of the corresponding data unit is output to the data assembling unit 44. In order to extract data from the input data unit, the data assembling unit 44 deletes the SAR-PDU header, assembles 47-byte long data, and outputs the received data. However, when the data unit discard instruction signal is input from the SN inspection unit 46, the corresponding data unit is discarded.

As described above, the SN inspection section 46 recognizes the order of data units by the data unit order recognition algorithm shown in FIG. That is, on the receiving side, it is judged whether it is normal or abnormal depending on the distance of the received data unit, and if it is abnormal, it is judged whether it is due to loss or incorrect insertion. Discard the unit, otherwise save and output the corresponding data unit.

In FIG. 2, the current data unit indicates a data unit which is currently a sequence recognition target, and the previous data unit is a previous sequence recognition target,
Indicates a stored data unit. In addition, the distance from the previous data unit to the current data unit is a concept of what number the current data unit corresponds to, calculated from the previous data unit when no loss or erroneous insertion error occurs.

For example, if the SN of the previous data unit is "3" and the SN of the current data unit is "7", and if no abnormality occurs, the current data unit is the fourth one calculated from the previous data unit. The distance from the previous data unit to the current data unit is "4"
Becomes Further, for example, if the SN of the previous data unit is 6 and the SN of the current data unit is 4, if no abnormality occurs, the current data unit corresponds to the sixth calculated from the previous data unit. The distance from the previous data unit to the current data unit is "6".

In the SN inserting section 32 on the transmitting side, 0, 1, 2, 3, 4, 5, 6, 7,
Since the SNs are inserted in the order of 0, 1, 2, 3, ..., If there is no abnormality in the SN of the data unit received on the receiving side, 0, 1, 2, 3, 4, 5, 6, 7, 0, 1,
It should be in the order of 2, 3, ... In FIG. 8, the transmission side and the reception side are set to one-to-one for simplification, but many devices are connected to an actual ATM network and there is a many-to-many relationship. Therefore, loss due to contention of ATM cells in the ATM network or erroneous insertion due to erroneous distribution occurs, and the SN of the data unit received by the receiving side is 0,
There are cases where the order is not 1, 2, 3, 4, 5, 6, 7, 0, 1, 2, 3, ....

Therefore, when the distance from the previous data unit to the current data unit is 1, it can be determined that the order of the previous data unit and the current data unit is normal, but the distance from the previous data unit to the current data unit is If is different from 1, the order of the previous data unit and the current data unit is abnormal, and the data unit existing between the previous data unit and the current data unit is lost or the current data unit is erroneously inserted. Can be judged.

When the order of the previous data unit and the current data unit is abnormal, if the loss of the data unit existing between the previous data unit and the current data unit occurs, the current data unit is saved, If an incorrect insertion of the current data unit occurs, the current data unit shall be discarded.

To determine whether the data unit existing between the previous data unit and the current data unit has been lost or the current data unit has been erroneously inserted, the determination distance is defined, and the current data unit from the previous data unit is defined. This is done by comparing the distance to the unit with this judgment distance. That is, if the distance from the previous data unit to the current data unit is greater than 1 and less than the determination distance, it is determined that the data unit existing between the previous data unit and the current data unit has been lost, and the previous data unit is lost. To the current data unit is greater than or equal to the decision distance,
It is determined that the incorrect insertion of the current data unit has occurred.

The rationale for this determination was that, on the assumption that the frequency of loss is not so high, a plurality of data units are rarely continuously lost. That is, when the distance from the previous data unit to the current data unit is small, it can be estimated that the data unit existing between the previous data unit and the current data unit has been lost, but If the distance is large, it is unlikely that the data unit that existed between the previous data unit and the current data unit is lost, and the current data unit can be presumed to have been misplaced due to incorrect insertion. .

The determination distance is such that SN is "0" to "7".
In the case of, it is possible to appropriately select from an integer of 2 to 8. Hereinafter, a specific example will be described with reference to FIG.
For example, if the judgment distance is 4 and the SN of the previous data unit is 2, if the SN of the current data unit is 3, then
The distance from the previous data unit to the current data unit is 1
Therefore, the order of the previous data unit and the current data unit is normal (201), and the current data unit is saved (205). If the SN of the current data unit is not 3,
The distance from the previous data unit to the current data unit is 1
Therefore, the order of the previous data unit and the current data unit is determined to be abnormal (201). Even if the SN of the current data unit is 4 or 5, since the distance from the previous data unit to the current data unit is less than the determination distance of 4, the data unit existing between the previous data unit and the current data unit. Was judged to have been lost (20
3) Save the current data unit (205).

However, the SN of the current data unit is 6,
If it is 7, 0, 1, or 2, the distance from the previous data unit to the current data unit is equal to or greater than the determination distance of 4, and therefore it exists between the previous data unit and the current data unit. It is unlikely that the data unit is lost, and it is determined that the current data unit has been misplaced due to incorrect insertion (203), and the current data unit is discarded (207).

[0024]

However, the above-described prior art data unit order recognition algorithm in the prior art assumes that the frequency of loss is not so high.
It was based on the fact that multiple data units are rarely lost in succession. That is, if the distance from the previous data unit to the current data unit is greater than 1 and less than the determination distance, it was determined that the data unit existing between the previous data unit and the current data unit was lost. If the distance from the unit to the current data unit is greater than or equal to the judgment distance, it is unlikely that the data unit existing between the previous data unit and the current data unit has been lost, and the current data unit was mistakenly inserted. I was making a decision.

On the other hand, in practice, a plurality of data units are rarely continuously lost, but this does not always occur. If multiple data units are lost consecutively, even if the distance from the previous data unit to the current data unit is greater than or equal to the judgment distance, if the current data unit is not misinserted,
The current data unit needs to be saved. However, in the conventional technique, such a data unit that needs to be stored is also discarded. Such discarding of the data unit means a loss of communication data, and there is a problem that communication quality is deteriorated.

It is an object of the present invention to solve the above-mentioned drawbacks of the prior art and to provide a data receiving method in which the probability of discarding a data unit that is not erroneously inserted is extremely small, and a communication interface device using the data receiving method. To do.

[0027]

In order to solve the above-mentioned problems, the present invention receives a data unit to which a sequence number is added in the order of transmission, and receives the sequence number of the received current data unit immediately before it. The data unit reception method that determines the distance between the previous data unit and the current data unit by comparing it with the sequence number of the previous data unit and selects saving or discarding of the current data unit from this distance The current data unit is saved if this distance is normal, and if the determined distance is abnormal, check whether the value is greater than the predetermined judgment distance. If it is smaller than the judgment distance, the current data unit is saved. To determine the distance to the data unit, the distance is to save the current data unit if indicates a normal, discards the current data unit if indicates an abnormal.

Further, by receiving the data units to which the sequence numbers are added in the order of transmission and comparing the sequence number of the received current data unit with the sequence number of the previous data unit received immediately before, the current data unit and the current data unit are compared. A communication interface device according to the present invention, which obtains a distance from a data unit and selects storage or discarding of a current data unit from this distance, comprises a first judging means, a second judging means, and a third judging means. Have. The first determining means determines the distance from the previous data unit to the current data unit, and if the distance is normal, saves the current data unit. When the distance determined by the first determining means indicates an abnormality, the second determining means determines whether or not the value is greater than a predetermined determination distance, and when the value is less than the determination distance. Save the current data unit. Further, the third judging means judges the distance from the current data unit to the next data unit when the distance judged by the second judging means is equal to or larger than the judging distance, and if the distance indicates normal. The current data unit is saved, and if there is an error, the current data unit is discarded.

[0029]

According to the present invention, the distance from the previous data unit to the current data unit is determined, and if the distance is normal, the current data unit is saved. If the determined distance indicates an abnormality, it is determined whether or not the value is larger than a predetermined determination distance. If the value is smaller than this determination distance, the current data unit is saved. If the distance is equal to or more than the judgment distance, the distance from the current data unit to the next data unit is further judged. If the distance is normal, the current data unit is saved. If it is abnormal, the current data unit is saved. Discard.

[0030]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A data receiving method and a communication interface device using the same according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows a data receiving method according to the present invention.
3 is a data unit sequence recognition algorithm showing the first embodiment when applied to an AAL1 protocol interface device for ATM communication.

In FIG. 1, the current data unit indicates a data unit which is the current order recognition target, the previous data unit indicates the previously recognized data unit which is the order recognition target, and the next data unit. Indicates a data unit to be the next order recognition target. Also, the distance from the previous data unit to the current data unit is a concept of how many times the current data unit corresponds to, calculated from the previous data unit when no loss or incorrect insertion error occurs. The distance from the current data unit to the next data unit is a concept of what number the next data unit corresponds to when calculated from the current data unit when no loss or misinsertion abnormality occurs.

For example, if the SN of the previous data unit is 3, the SN of the current data unit is 7, and the SN of the next data unit is 2, if an abnormality does not occur, calculation is performed from the previous data unit to obtain the current data unit. Corresponds to the fourth data unit, the distance from the previous data unit to the current data unit is 4, and the next data unit calculated from the current data unit corresponds to the third data unit. The distance to is 3. Furthermore, for example, if the SN of the previous data unit is 6, the SN of the current data unit is 4, and the SN of the next data unit is 0, if an abnormality does not occur, calculation is performed from the previous data unit, and the current data unit is , Because it corresponds to the sixth,
The distance from the previous data unit to the current data unit is
6, the next data unit corresponds to the fourth data unit calculated from the current data unit, so the distance from the current data unit to the next data unit is 4.

In the SN inserting section on the transmitting side, 0, 1, 2, 3, 4, 5, 6, 7, 0,
Since the SNs are inserted in the order of 1, 2, 3, ..., The SNs of the data units received on the receiving side are also 0, 1, 2, 3 ,.
It should be in the order of 4, 5, 6, 7, 0, 1, 2, 3, ... In FIG. 8, the transmission side and the reception side are set to one-to-one for simplification, but many devices are connected to an actual ATM network and there is a many-to-many relationship. Therefore, loss due to contention of ATM cells in the ATM network or erroneous insertion due to erroneous distribution may occur, and the SN of the data unit received at the receiving side is 0, 1, 2,
In some cases, the order is not 3, 4, 5, 6, 7, 0, 1, 2, 3, ...

Therefore, when the distance from the previous data unit to the current data unit is 1, it can be judged that the order of the previous data unit and the current data unit is normal, but the distance from the previous data unit to the current data unit is If is different from 1, the order of the previous data unit and the current data unit is abnormal, and the data unit existing between the previous data unit and the current data unit is lost or the current data unit is erroneously inserted. Can be judged.

If the order of the previous data unit and the current data unit is abnormal, and if the loss of the data unit existing between the previous data unit and the current data unit occurs, the current data unit is saved, If an incorrect insertion of the current data unit occurs, the current data unit shall be discarded.

To determine whether the data unit existing between the previous data unit and the current data unit has been lost or the current data unit has been erroneously inserted, the determination distance is defined and the determination is made from the previous data unit to the current data unit. Is compared with the judgment distance, and the distance from the current data unit to the next data unit is inspected.

That is, when the distance from the previous data unit to the current data unit is greater than 1 and less than the determination distance,
Or, if the distance from the previous data unit to the current data unit is greater than or equal to the judgment distance and the distance from the current data unit to the next data unit is 1, the loss of the data unit existing between the previous data unit and the current data unit is It is determined that it has occurred. If the distance from the previous data unit to the current data unit is greater than or equal to the determination distance and the distance from the current data unit to the next data unit is different from 1, it is determined that the current data unit has been erroneously inserted.

As a basis for this judgment, the frequency of loss is
It is based on the fact that multiple data units are rarely lost in succession on the assumption that they are not too expensive, but this is not impossible. That is, when the distance from the previous data unit to the current data unit is small, it can be estimated that the data unit existing between the previous data unit and the current data unit is lost. In addition, when the distance from the previous data unit to the current data unit is large, it is unlikely that the data unit existing between the previous data unit and the current data unit is lost.

However, the current data unit may have been lost and may have been erroneously inserted due to incorrect insertion. Therefore, the order of the current data unit and the next data unit is checked, and if the order of the current data unit and the next data unit is normal (the distance from the current data unit to the next data unit is “1”), the previous data unit is checked. It is estimated that the data unit existing between the current data units has been lost, and if the order of the current data unit and the next data unit is abnormal (the distance from the current data unit to the next data unit is different from 1), The data unit is presumed to have been misplaced due to incorrect insertion.

As described above, in the present embodiment, it is focused on that there is almost no probability that a data unit having continuous SNs is erroneously inserted, and even if there is a loss of a data unit exceeding the judgment distance, that data unit is lost. A means for relieving is provided. The determination distance can be selected from an integer of 2 to 8 in this embodiment.

Next, the operation of this embodiment will be described with reference to FIG. If the determination distance is 4 and the SN of the previous data unit is 2, if the SN of the current data unit is 3, the distance from the previous data unit to the current data unit is 1, so the order of the previous data unit and the current data unit is It is normal (101) and the current data unit is saved (107). If the SN of the current data unit is not 3, the distance from the previous data unit to the current data unit is different from 1, so the order of the previous data unit and the current data unit is abnormal (101).

Even if the SN of the current data unit is 4 or 5, since the distance from the previous data unit to the current data unit is less than the judgment distance of 4, it exists between the previous data unit and the current data unit. The determined data unit is lost (103), and the current data unit is saved (107).

However, the SN of the current data unit is 6,
If it is 7, 0, 1, or 2, it is unlikely that the data unit existing between the previous data unit and the current data unit has been lost, but there is a possibility that the current data unit is erroneously inserted. There is a possibility that it was confused by. Therefore, the order of the current data unit and the next data unit is checked (105).

Specifically, the SN of the current data unit is 6
And the SN of the next data unit is 7, or the SN of the current data unit is 7, and the SN of the next data unit is 0, or the SN of the current data unit is 0, and the SN of the next data unit is 1, or SN of current data unit is 1
If the SN of the next data unit is 2, or the SN of the current data unit is 2, and the SN of the next data unit is 3, the distance from the current data unit to the next data unit is 1 is there. Therefore, the order of the current data unit and the next data unit is normal,
It is determined that the data unit existing between the previous data unit and the current data unit is lost (105), and the current data unit is saved (107).

On the other hand, in the order of the current data unit and the next data unit, the SN of the current data unit is 6, the SN of the next data unit is 7, or the SN of the current data unit.
Is 7, and the SN of the next data unit is 0, or the SN of the current data unit is 0 and the SN of the next data unit is 1, or the SN of the current data unit is 1 and the SN of the next data unit is 2, Or the SN of the current data unit
2 and the SN of the next data unit does not match either 3, the distance from the current data unit to the next data unit is different from 1. Therefore, in this case, the order of the current data unit and the next data unit is abnormal, and it is determined that the current data unit has been misplaced due to erroneous insertion (105), and the current data unit is discarded (109).

As described in detail above, the first aspect of the present invention
According to this embodiment, there are the following effects. That is, it is unlikely that multiple data units will be lost in succession, but this is not possible, and if multiple data units are lost in succession, even if the previous data unit becomes the current data unit. Even if the distance to is greater than or equal to the determination distance, the current data unit must be saved if it is not misaligned due to incorrect insertion of the current data unit.

Therefore, by checking not only the order of the previous data unit and the current data unit but also the order of the current data unit and the next data unit, a plurality of data units existing between the previous data unit and the current data unit can be detected. , It is determined whether the data unit has been lost continuously, or the current data unit has been erroneously inserted due to incorrect insertion, and multiple data units existing between the previous data unit and the current data unit have been continuously lost. If it is determined that the current data unit can be saved, it is possible to reduce the discard of the data unit, that is, the loss of communication data, and improve the communication quality.

Hereinafter, the prior art and the present invention will be compared with reference to specific examples.

The judgment distance is selected as 4, the SN of the first received data unit is 3, the SN of the second received data unit is 4, the SN of the third received data unit is 5, the 4th. The SN of the received data unit is 6, the SN of the 5th received data unit is 7, the SN of the 6th received data unit is 0, the SN of the 7th received data unit is the 1st, 8th received The SN of the data unit is 3rd and 9th, and the SN of the data unit is 4th and 10th.
N is 5, the SN of the 11th data unit received is 6, and the SN of the 12th data unit is 7, 1
The SN of the third received data unit is 0, the SN of the 14th received data unit is 1, the SN of the 15th received data unit is 2, the SN of the 16th received data unit is the 3rd, 17th The SN of the data unit received at 6th and 18th is the SN of the data unit received at 7
Is 0, the SN of the 20th received data unit is 1,
The SN of the 21st received data unit is 7, the SN of the 22nd received data unit is 2, the SN of the 23rd received data unit is 3, the SN of the 24th received data unit is the 4th, 25th The SN of the data unit received at 5, the SN of the 26th received data unit is 6, and the S of the 27th received data unit is S.
N is 3, 28, the SN of the data unit received at the 4th, SN of the data unit received at the 29th is 5, 3
The SN of the 0th received data unit is 6, the SN of the 31st received data unit is 7, the SN of the 32nd received data unit is 0, the SN of the 33rd received data unit is the 1st, 34th The SN of the data unit received on the 2nd, the SN of the data unit received on the 35th, and the SN of the data unit received on the 36th
, The SN of the 37th received data unit is 5,
The case where the SN of the 38th received data unit is 6 and the SN of the 39th received data unit is 7 will be described.

According to the conventional technique, the first to seventh data units are stored because the distance between two consecutive data units is 1, so that the sequence of two consecutive data units is normal.

Since the SN of the 7th received data unit is 1 and the SN of the 8th received data unit is 3, the distance from the 7th received data unit to the 8th received data unit is Since it is 2 and different from 1, the order of the 7th received data unit and the 8th received data unit is abnormal. Therefore, in order to determine whether the loss of the data unit existing between the seventh received data unit and the eighth received data unit has occurred, or the erroneous insertion of the eighth received data unit has occurred, The distance from the 7th received data unit to the 8th received data unit is compared with the determination distance. As a result of the comparison,
Since the distance from the 7th received data unit to the 8th received data unit is less than the judgment distance, the data units existing between the 7th received data unit and the 8th received data unit are It is determined that a loss has occurred, and the eighth received data unit is saved.

The data units received from the 9th to 16th data units are stored because the distance between two consecutive data units is 1, so the order of two consecutive data units is normal.

SN of the 16th received data unit
Since the SN of the 3rd and 17th received data unit is 6, the distance from the 16th received data unit to the 17th received data unit is 3, which is different from 1, so The order of the received data unit and the 17th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 16th received data unit and the 17th received data unit has been lost or whether the 17th received data unit has been erroneously inserted, The distance from the 16th received data unit to the 17th received data unit is compared with the determination distance. As a result of the comparison, the distance from the 16th received data unit to the 17th received data unit is less than the determination distance, so that the distance exists between the 16th received data unit and the 17th received data unit. It is determined that the data unit loss has occurred and the 17th received data unit is saved.

The data units received from the 18th to 20th have a distance of 1 between two consecutive data units.
Therefore, the order of two consecutive data units is normal, so the data unit is saved.

SN of the 20th received data unit
Since the SN of the 1st and 21st received data units is 7, the distance from the 20th received data unit to the 21st received data unit is 6, which is different from 1 and therefore The order of the received data unit and the 21st received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 20th received data unit and the 21st received data unit has been lost or whether the 21st received data unit has been erroneously inserted, The distance from the 20th received data unit to the 21st received data unit is compared with the determination distance. As a result of the comparison, the distance from the 20th received data unit to the 21st received data unit is greater than or equal to the determination distance, so it is determined that the 21st received data unit is erroneously inserted. Discard the second received data unit.

SN of the 20th received data unit
Since the SN of the 1st and 22nd received data units is 2, the distance from the 20th received data unit to the 22nd received data unit is 1 and the 20th received data unit Since the order of the 22nd received data unit is normal, the 22nd received data unit is saved.

The data units received from the 23rd to the 26th have a distance of 1 between two consecutive data units.
Therefore, the order of two consecutive data units is normal, so the data unit is saved.

SN of the 26th received data unit
Since the SN of the 6th and 27th received data units is 3, the distance from the 26th received data unit to the 27th received data unit is 5, which is different from 1, so The order of the received data unit and the 27th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 26th received data unit and the 27th received data unit has been lost or the 27th received data unit has been erroneously inserted, The distance from the 26th received data unit to the 27th received data unit is compared with the determination distance. As a result of the comparison, the distance from the 26th received data unit to the 27th received data unit is greater than or equal to the determination distance, so it is determined that the 27th received data unit has been erroneously inserted. Discard the second received data unit.

SN of the 26th received data unit
Since the SN of the 6th and 28th received data unit is 4, the distance from the 26th received data unit to the 28th received data unit is 6 and is different from 1, so The order of the received data unit and the 28th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 26th received data unit and the 28th received data unit is lost or whether the 28th received data unit is erroneously inserted, The distance from the 26th received data unit to the 28th received data unit is compared with the determination distance. As a result of the comparison, the distance from the 26th received data unit to the 28th received data unit is greater than or equal to the determination distance, so it is determined that the 28th received data unit has been erroneously inserted. Discard the second received data unit.

SN of 26th received data unit
Since the SN of the 6th and 29th received data units is 5, the distance from the 26th received data unit to the 29th received data unit is 7, which is different from 1, so The order of the received data unit and the 29th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 26th received data unit and the 29th received data unit is lost or whether the 29th received data unit is erroneously inserted, The distance from the 26th received data unit to the 29th received data unit is compared with the determination distance. As a result of the comparison, the distance from the 26th received data unit to the 29th received data unit is greater than or equal to the determination distance, so it is determined that the erroneous insertion of the 29th received data unit has occurred. Discard the second received data unit.

SN of the 26th received data unit
Since the SN of the 6th and 30th received data units is 6, the distance from the 26th received data unit to the 30th received data unit is 8, which is different from 1 and therefore The order of the received data unit and the 30th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 26th received data unit and the 30th received data unit is lost or whether the 30th received data unit is erroneously inserted, The distance from the 26th received data unit to the 30th received data unit is compared with the determination distance. As a result of the comparison, the distance from the 26th received data unit to the 30th received data unit is greater than or equal to the determination distance, so it is determined that the erroneous insertion of the 30th received data unit has occurred. Discard the second received data unit.

SN of the 26th received data unit
Is 6, the SN of the 31st received data unit is 7, so the distance from the 26th received data unit to the 31st received data unit is 1, and the 26th received data unit Since the order of the 31st received data unit is normal, the 31st received data unit is saved.

The data units received from the 32nd to the 39th have a distance of 1 between two consecutive data units.
Therefore, the order of two consecutive data units is normal, so the data unit is saved.

On the other hand, according to the first embodiment, since the distance between two consecutive data units is 1 in the first to seventh received data units, the order of two consecutive data units is normal. Save because there is.

Since the SN of the 7th received data unit is 1 and the SN of the 8th received data unit is 3, the distance from the 7th received data unit to the 8th received data unit is Since it is 2 and different from 1, the order of the 7th received data unit and the 8th received data unit is abnormal. Therefore, in order to determine whether the loss of the data unit existing between the seventh received data unit and the eighth received data unit has occurred, or the erroneous insertion of the eighth received data unit has occurred, The distance from the 7th received data unit to the 8th received data unit is compared with the determination distance. As a result of the comparison,
Since the distance from the 7th received data unit to the 8th received data unit is less than the judgment distance, the data units existing between the 7th received data unit and the 8th received data unit are It is determined that a loss has occurred, and the eighth received data unit is saved.

Since the distance between two consecutive data units is 1, the ninth to sixteenth data units are stored because the order of the two consecutive data units is normal.

SN of the 16th received data unit
Since the SN of the 3rd and 17th received data unit is 6, the distance from the 16th received data unit to the 17th received data unit is 3, which is different from 1, so The order of the received data unit and the 17th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 16th received data unit and the 17th received data unit has been lost or whether the 17th received data unit has been erroneously inserted, The distance from the 16th received data unit to the 17th received data unit is compared with the determination distance. As a result of the comparison, the distance from the 16th received data unit to the 17th received data unit is less than the determination distance, so that the distance exists between the 16th received data unit and the 17th received data unit. It is determined that the data unit loss has occurred and the 17th received data unit is saved.

The eighteenth to twentieth data units received have a distance of 1 between two consecutive data units.
Therefore, the order of two consecutive data units is normal, so the data unit is saved.

SN of the 20th received data unit
Since the SN of the 1st and 21st received data units is 7, the distance from the 20th received data unit to the 21st received data unit is 6, which is different from 1 and therefore The order of the received data unit and the 21st received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 20th received data unit and the 21st received data unit has been lost or whether the 21st received data unit has been erroneously inserted, A comparison between the distance from the 20th received data unit to the 21st received data unit and the judgment distance, and 2 from the 21st received data unit
The distance to the second received data unit is checked. As a result of the comparison, the distance from the 20th received data unit to the 21st received data unit is
Since the distance is equal to or larger than the determination distance and the result of the inspection is that the distance from the 21st received data unit to the 22nd received data unit is different from 1, it is determined that the 21st received data unit is erroneously inserted. It is determined and the 21st received data unit is discarded.

SN of the 20th received data unit
Since the SN of the 1st and 22nd received data units is 2, the distance from the 20th received data unit to the 22nd received data unit is 1 and the 20th received data unit Since the order of the 22nd received data unit is normal, the 22nd received data unit is saved.

The data units received from the 23rd to the 26th have a distance of 1 between two consecutive data units.
Therefore, the order of two consecutive data units is normal, so the data unit is saved.

SN of the 26th received data unit
Since the SN of the 6th and 27th received data units is 3, the distance from the 26th received data unit to the 27th received data unit is 5, which is different from 1, so The order of the received data unit and the 27th received data unit is abnormal. Therefore, in order to determine whether the data unit existing between the 26th received data unit and the 27th received data unit has been lost or the 27th received data unit has been erroneously inserted, The distance from the 26th received data unit to the 27th received data unit is compared with the judgment distance, and 2 from the 27th received data unit.
The distance to the 8th received data unit is checked. As a result of the comparison, the distance from the 26th received data unit to the 27th received data unit is
As a result of the inspection, the distance from the 27th received data unit to the 28th received data unit is 1, so the data received from the 26th received data unit to the 27th received data unit It is determined that the data unit existing between the units has been lost, and the 27th received data unit is stored.

The data units received from the 28th to 39th data units have a distance of 1 between two consecutive data units.
Therefore, the order of two consecutive data units is normal, so the data unit is saved.

Table 1 compares the data unit discarded in the prior art with the data unit discarded in the first embodiment.

[0076]

[Table 1]

Differences between the prior art and the first embodiment will be described below.

In the prior art, as shown in (a) of Table 1, it is determined that the 27th to 30th received data units have been erroneously inserted, and the 27th to 30th received data units are discarded. Was. On the other hand, in the first embodiment, as shown in (b) of Table 1, it is determined that the data unit existing between the 26th received data unit and the 27th received data unit is lost. , The 27th to 30th received data units are stored.

As described above, according to the first embodiment, the effects of discarding the data unit, that is, reducing the loss of the communication data and improving the communication quality can be expected.

Next, a second embodiment of the data receiving method will be described. The second embodiment corresponds to a superordinate concept of the first embodiment and substantially includes the first embodiment.

In the description, D is a division of communication data such as a data unit or packet data, and indicates fixed length or variable length data and S storage area having an arbitrary length. . S indicates a code such as SN which indicates the order of D by making a cycle. In the description, it is assumed that S cycles through integer values from 0 to 7.

FIG. 9 is a schematic block diagram of a communication interface device to which the second embodiment of the data receiving method is applied. In FIG. 9, the transmitting side has the following functions. The data decomposition unit 50 decomposes the input transmission data into a certain data length and outputs it to the S insertion unit 52. S generation unit 56
Then, S is generated and output to the S insertion unit 52. The S insertion unit 52 inserts the S input from the S generation unit 56 into the S storage area of D input from the data decomposition unit 50, and outputs D to the D transmission unit 54. The D transmission unit 54 outputs the D input from the S insertion unit 52 to the communication network 58.

On the other hand, the receiving side has the following functions. D
The receiving unit 60 outputs the D input from the communication network 58 to the S extracting unit 62. The S extraction unit 62 extracts S from the input S storage area of D, and the S inspection unit 66
And D to the data assembling unit 64. The S inspection unit 66 inspects the input S and recognizes the order of D. As a result of order recognition, the received D is normal or abnormal, and if abnormal, it is judged whether it is due to the loss of D or due to the incorrect insertion of D. The discard instruction signal is output to the data assembling unit 64. In order to extract data from the input D, the data assembling unit 64 assembles data of a certain data length and outputs the received data. However, when the D discard instruction signal is input from the S inspection unit 66, the corresponding D
Discard.

FIG. 10 shows a D order recognition algorithm in the communication interface device in the second embodiment.

In FIG. 10, the current D indicates the current order recognition target D, the previous D indicates the previous order recognition target and is saved, and the next D is The D that will be the next sequence recognition target is shown. Further, the distance from the front D to the current D means that when no loss or misinsertion abnormality occurs,
It is a concept of what number the current D corresponds to from the previous D, and the distance from the current D to the next D is calculated from the current D when no loss or abnormal insertion error occurs. , The next D corresponds to the concept.

For example, if the S of the previous D is 3, the S of the current D is 7, and the S of the next D is 2, when the abnormality does not occur, the current D is calculated as the 4th, and calculated from the previous D. Since it corresponds, the distance from the previous D to the current D is 4, and the distance from the current D to the next D is calculated.
Corresponds to the third, so the distance from the current D to the next D is 3. Further, for example, if the S of the previous D is 6, the S of the current D is 4, and the S of the next D is 0, when no abnormality occurs,
Since the current D corresponds to the sixth calculated from the previous D, the distance from the previous D to the current D is 6, and the next D calculated from the current D corresponds to the fourth. The distance from D to the next D is 4.

In the S insertion section of the transmission section, 0, 1, 2, 3, 4, 5, 6, 7, 0, 1, 2,
Since S is inserted in the order of 3, ..., S of the received D is also 0, 1, 2, 3, 4, 5, 6, 7, 0, 1,
It should be in the order of 2, 3, ... However, loss of D or erroneous insertion may occur, and S of D received on the receiving side is 0, 1, 2, 3, 4, 5, 6, 7, 0,
In some cases, the order is not 1, 2, 3, ....

Therefore, the distance from the previous D to the present D is 1
In the case of, the order of the front D and the current D can be judged to be normal, but if the distance from the front D to the current D is different from 1, the order of the front D and the current D is abnormal, It can be determined that the loss of D existing between D and the current D or the erroneous insertion of the current D occurs.

When the order of the previous D and the current D is abnormal, if the loss of the D existing between the previous D and the current D occurs, the current D is saved and the current D is erroneously inserted. If so, the current D must be discarded.

The determination of whether the loss of D existing between the previous D and the current D has occurred or the erroneous insertion of the current D has occurred is defined by a determination distance. This is performed by comparing the judgment distances and inspecting the distance from the current D to the next D. That is, when the distance from the front D to the current D is greater than 1 and less than the determination distance, or the distance from the front D to the current D is greater than or equal to the determination distance and the distance from the current D to the next D is
In the case of 1, it is determined that the loss of D existing between the previous D and the current D has occurred, the distance from the previous D to the current D is equal to or more than the determination distance, and the distance from the current D to the next D is If it is different from 1, it is determined that the current D has been erroneously inserted.

The basis of this judgment is that the frequency of loss is
It is based on the fact that multiple D's are not likely to be lost in succession on the assumption that they are not too high, but this is not something that cannot occur. That is, when the distance from the previous D to the current D is small, the D existing between the previous D and the current D is present.
However, it can be inferred that it has been lost. If the distance from the previous D to the current D is large, it is unlikely that the D existing between the previous D and the current D is lost, but there is a possibility that the current D is erroneously inserted. , It may be mixed up, so the order of the current D and the next D is checked, and the current D and the next D
If the order of is normal (the distance from the current D to the next D is 1), it can be estimated that the D existing between the previous D and the current D is lost, and the order of the current D and the next D is In the case of abnormality (the distance from the current D to the next D is different from 1), the current D can be presumed to have been misplaced due to erroneous insertion. The determination distance can be selected from integers from 2 to 8.

A specific example will be given below, and the second example will be described with reference to FIG.
The operation of this embodiment will be described. If the determination distance is 4, the S of the previous D is 2, and the S of the current D is 3, the distance from the previous D to the current D is 1, so the order of the previous D and the current D is normal (9
01), the current D is saved (907). If S of the current D is not 3, the distance from the previous D to the current D is different from 1, so
The order of the previous D and the current D is abnormal (901). Even if the S of the current D is 4 or 5, the distance from the previous D to the current D is less than the determination distance of 4, so the D existing between the previous D and the current D is lost. (903), current D
Is saved (907).

However, when the S of the current D is 6, 7, 0, 1, 2, it is unlikely that the D existing between the previous D and the current D is lost, but it is possible. There is also a nature, current D
May have been misplaced due to incorrect insertion. Therefore, the order of the current D and the next D is checked, and S of the current D is 6 and S of the next D is 7, or S of the current D is 7 and S of the next D is 0,
Or S of the current D is 0 and S of the next D is 1, or the current D
S of 1 is S of the next D is 2, or S of the current D is 2,
If the S of the next D is either 3, the distance from the current D to the next D is 1, so the order of the current D and the next D is normal, and the order between the previous D and the current D is between. It is determined that the D existing in the node has been lost (905), and the current D is saved (90).
7).

The order of the current D and the next D is 6 when the S of the current D is 6.
And S of next D is 7, or S of current D is 7 and S of next D
Is 0, or S of the current D is 0, S of the next D is 1, or S of the current D is 1, and S of the next D is 2, or S of the current D
Is 2 and the S of the next D does not match either 3, the distance from the current D to the next D is different from 1, so the current D
The order of the next D and the next D is abnormal.
It is determined that it is mixed (905), and the current D is discarded (909).

As described above, according to the second embodiment, a plurality of D's are rarely continuously lost, but this is not impossible, and a plurality of D's are continuously lost. In this case, even if the distance from the previous D to the current D is equal to or larger than the determination distance, the current D needs to be saved if the current D is not misplaced due to incorrect insertion.

Therefore, in addition to the order of the previous D and the current D,
By checking the order of the current D and the next D, the previous D to the current D
The multiple D's that existed between the
It is determined whether the current D is misplaced due to incorrect insertion, and the previous D
If it is determined that a plurality of Ds existing between the current D and the current D are continuously lost, the current D can be saved, and the D is discarded, that is, the loss of communication data is reduced. However, improvement in communication quality can be expected.

In the following, the prior art and the second embodiment will be compared by a concrete example.

The judgment distance is selected to be 4, the S of D received first is 3, the S of D received second is 4, the S of D received third is 5, the D received fourth. S is 6, 5th received S of D is 7, 6th received S of D is 0, 7th received D is 1st, 8th received D
Of S is 3, the S of D received at the 9th is 4, the S of D received at the 10th is 5, the S of D received at the 11th is 6, 1
The second S received D is 7, the 13th received D S is 0, the 14th received D S is 1, the 15th received S is 2, 16th received D S is 3, 1
The 7th received S of D is 6, the 18th received S of D is 7, the 19th received S of D is 0, the 20th received S of D is 1, 21st received D S is 7, 2
The second received S of D is 2, the 23rd received S of D is 3, the 24th received S of D is 4, the 25th received S of D is 5, and the 26th received D S of 6, 2
7th received S of D is 3, 28th received S of D is 4, 29th received S of D is 5, 30th received S of D is 6, 31st received D S of 7, 3
S of D received second is 0, S of D received at 33rd is 1, S of D received at 34th is 2, S of D received at 35th is D received at 3rd, 36th S is 4, 3
A case will be described in which the 7th received D S is 5, the 38th received D S is 6, and the 39th received S is 7.

According to the conventional technique, the first to seventh received Ds are stored because the distance between two consecutive Ds is 1, so that the order of two consecutive Ds is normal.

Since the S of the 7th received D is 1 and the S of the 8th received D is 3, the 7th received D
The distance from the 8th to the 8th received D is 2, which is different from the 1st. Therefore, the order of the 7th received D and the 8th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 7th received D and the 8th received D has occurred, or the erroneous insertion of the 8th received D has occurred, the 7th received The distance from D to the eighth received D is compared with the determination distance. As a result of comparison, the distance from the 7th received D to the 8th received D is less than the determination distance, so the 7th received D
It is determined that the loss of D that was present between the 8th received D to 8th is generated, and the 8th received D is saved.

The 9th to 16th received Ds are stored because the distance between two consecutive Ds is 1, so the order of two consecutive Ds is normal.

Since the S of the 16th received D is 3 and the S of the 17th received D is 6, the distance from the 16th received D to the 17th received D is 3, Since it is different from 1, the order of the 16th received D and the 17th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 16th received D and the 17th received D has occurred or the 17th received D erroneous insertion has occurred, the 16th received The distance from D to the 17th received D is compared with the determination distance. As a result of the comparison, since the distance from the 16th received D to the 17th received D is less than the determination distance, the loss of D existing between the 16th received D and the 17th received D Is determined to have occurred and 1
Save the 7th received D.

The Ds received from the 18th to the 20th have two consecutive Ds because the distance between the two consecutive Ds is 1.
Save because the order of is normal.

Since the S of the 20th received D is 1 and the S of the 21st received D is 7, the distance from the 20th received D to the 21st received D is 6, Since it is different from 1, the order of the 20th received D and the 21st received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 20th received D to the 21st received D has occurred or the 21st received D erroneous insertion has occurred, the 20th received The distance from D to the 21st received D is compared with the determination distance. As a result of the comparison, the distance from the 20th received D to the 21st received D is greater than or equal to the determination distance, so it is determined that the 21st received D is erroneously inserted, and the 21st received Discard D.

Since the S of the 20th received D is 1 and the S of the 22nd received D is 2, the distance from the 20th received D to the 22nd received D is 1, Since the order of the 20th received D and the 22nd received D is normal, the 22nd received D is stored.

The Ds received from the 23rd to the 26th are two consecutive Ds because the distance between the two consecutive Ds is 1.
Save because the order of is normal.

Since the S of the 26th received D is 6 and the S of the 27th received D is 3, the distance from the 26th received D to the 27th received D is 5, Since it is different from 1, the order of the 26th received D and the 27th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 26th received D and the 27th received D has occurred or the 27th received D insertion error has occurred, the 26th received The distance from D to the 27th received D is compared with the determination distance. As a result of comparison, the distance from the 26th received D to the 27th received D is greater than or equal to the determination distance, so it is determined that the 27th received D is erroneously inserted, and the 27th received Discard D.

Since the S of the 26th received D is 6 and the S of the 28th received D is 4, the distance from the 26th received D to the 28th received D is 6, Since it is different from 1, the order of the 26th received D and the 28th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 26th received D and the 28th received D occurred or the 28th received D insertion error occurred, the 26th received The distance from D to the 28th received D is compared with the determination distance. As a result of the comparison, the distance from the 26th received D to the 28th received D is equal to or greater than the determination distance, so it is determined that the 28th received D is erroneously inserted, and the 28th received Discard D.

Since the S of the 26th received D is 6 and the S of the 29th received D is 5, the distance from the 26th received D to the 29th received D is 7, Since it is different from 1, the order of the 26th received D and the 29th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 26th received D and the 29th received D has occurred or the 29th received D insertion error has occurred, the 26th received The distance from D to the 29th received D is compared with the determination distance. As a result of the comparison, the distance from the 26th received D to the 29th received D is equal to or greater than the determination distance, so it is determined that the 29th received D is erroneously inserted, and the 29th received Discard D.

Since the S of the 26th received D is 6 and the S of the 30th received D is 6, the distance from the 26th received D to the 30th received D is 8, Since it is different from 1, the order of the 26th received D and the 30th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 26th received D and the 30th received D occurred or the 30th received D insertion error occurred, the 26th received The distance from D to the 30th received D is compared with the determination distance. As a result of comparison, the distance from the 26th received D to the 30th received D is greater than or equal to the determination distance, so it is determined that the 30th received D has been erroneously inserted, and the 30th received Discard D.

Since the S of the 26th received D is 6 and the S of the 31st received D is 7, the distance from the 26th received D to the 31st received D is 1, Since the order of the 26th received D and the 31st received D is normal, the 31st received D is stored.

The D's received from the 32nd to the 39th are two continuous D's because the distance between the two continuous D's is 1.
Save because the order of is normal.

On the other hand, according to the second embodiment, the first to seventh received Ds have a distance between two consecutive Ds of 1
Therefore, since the order of two consecutive Ds is normal, it is saved.

Since the S of the 7th received D is 1 and the S of the 8th received D is 3, the 7th received D
The distance from the 8th to the 8th received D is 2, which is different from the 1st. Therefore, the order of the 7th received D and the 8th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 7th received D and the 8th received D has occurred, or the erroneous insertion of the 8th received D has occurred, the 7th received The distance from D to the eighth received D is compared with the determination distance. As a result of comparison, the distance from the 7th received D to the 8th received D is less than the determination distance, so the 7th received D
It is determined that the loss of D that was present between the 8th received D to 8th is generated, and the 8th received D is saved.

The 9th to 16th received Ds are saved because the distance between two consecutive Ds is 1, so the order of two consecutive Ds is normal.

Since the S of the 16th received D is 3 and the S of the 17th received D is 6, the distance from the 16th received D to the 17th received D is 3, Since it is different from 1, the order of the 16th received D and the 17th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 16th received D and the 17th received D has occurred or the 17th received D erroneous insertion has occurred, the 16th received The distance from D to the 17th received D is compared with the determination distance. As a result of the comparison, since the distance from the 16th received D to the 17th received D is less than the determination distance, the loss of D existing between the 16th received D and the 17th received D Is determined to have occurred and 1
Save the 7th received D.

The D's received from the 18th to the 20th have two consecutive D's because the distance between the two consecutive D's is 1.
Save because the order of is normal.

Since the S of the 20th received D is 1 and the S of the 21st received D is 7, the distance from the 20th received D to the 21st received D is 6, Since it is different from 1, the order of the 20th received D and the 21st received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 20th received D to the 21st received D has occurred or the 21st received D erroneous insertion has occurred, the 20th received The distance from the received D to the 21st received D is compared with the determination distance, and the distance from the 21st received D to the 22nd received D is checked. As a result of comparison, 20
The distance from the second received D to the 21st received D is greater than or equal to the determination distance, and as a result of the inspection, the distance from the 21st received D to the 22nd received D is different from 1, It is determined that the 21st received D is erroneously inserted, and the 21st received D is discarded.

Since the S of the 20th received D is 1 and the S of the 22nd received D is 2, the distance from the 20th received D to the 22nd received D is 1, Since the order of the 20th received D and the 22nd received D is normal, the 22nd received D is stored.

The D's received from the 23rd to the 26th are two consecutive D's because the distance between the two consecutive D's is 1.
Save because the order of is normal.

Since the S of the 26th received D is 6 and the S of the 27th received D is 3, the distance from the 26th received D to the 27th received D is 5, Since it is different from 1, the order of the 26th received D and the 27th received D is abnormal. Therefore, in order to determine whether the loss of D existing between the 26th received D and the 27th received D has occurred or the 27th received D insertion error has occurred, the 26th received The distance from the received D to the 27th received D is compared with the determination distance, and the distance from the 27th received D to the 28th received D is checked. As a result of comparison, 26
The distance from the Dth received at the 27th to the D received at the 27th is greater than or equal to the determination distance, but as a result of the inspection, the distance from the D received at the 27th to the D received at the 28th is 1 , It is determined that the loss of D existing between the 26th received D and the 27th received D occurs,
Save the 27th received D.

The Ds received from the 28th to the 39th have two consecutive Ds because the distance between the two consecutive Ds is 1.
Save because the order of is normal.

Differences between the prior art and the second embodiment will be described below with reference to Table 1.

In the prior art (a), it was determined that the 27th to 30th received Ds were erroneously inserted, and the 27th to 30th received Ds were discarded. In the example, it is determined that the loss of D existing between the 26th received D and the 27th received D occurs, and the 27th to 30th received D are stored.

As described above, according to the second embodiment, D
To reduce the loss of communication data units,
The effect of improving communication quality can be expected.

The conditions of the usage mode of the communication interface device applicable to this embodiment will be described below. 1. A communication data unit is divided into fixed-length or variable-length data units for communication, such as ATM cells and packet data.

2. The divided fixed-length or variable-length data unit is transmitted together with a code indicating the order of the divided fixed-length or variable-length data unit, such as SN.

3. The code indicating the order of the divided fixed-length or variable-length data unit may or may not be a numerical value, and it is cyclic, and the receiving side divides the fixed-length or variable-length data The order of the divided fixed-length or variable-length data units is recognized by a code indicating the order of the units.

In this embodiment, any interface device satisfying the above conditions can be applied.

[0130]

As described above, according to the data unit receiving method of the present invention, when the distance from the previous data unit to the current data unit is abnormal, it is conventionally determined whether to discard or save the current data unit. It becomes possible to carry out more strictly compared with the technology. Therefore, since the data unit to be discarded can be selected accurately, improvement in communication quality can be expected.

[Brief description of drawings]

FIG. 1 is a data unit order recognition algorithm showing an embodiment of a data unit receiving method according to the present invention.

FIG. 2 is a data unit order recognition algorithm in the prior art.

FIG. 3 is an ATM cell format.

FIG. 4 is a format of an ATM cell payload (data unit).

FIG. 5 is a SAR-PDU header format.

6 is a schematic diagram of the data unit shown in FIG. 4. FIG.

FIG. 7 is an explanatory diagram showing the relationship between communication data and data units.

FIG. 8 is a schematic block diagram of an interface device to which the first embodiment of the data receiving method according to the present invention is applied.

FIG. 9 is a schematic block diagram of an interface device to which a second embodiment of the data receiving method according to the present invention is applied.

FIG. 10 is a D-order recognition algorithm showing a second embodiment of the data receiving method according to the present invention.

[Explanation of symbols]

 30, 50 Data decomposition section 32 SN insertion section 34 Data unit transmission section 36 SN generation section 38, 58 Communication network 40 Data unit reception section 42 SN extraction section 44, 64 Data assembly section 46 SN inspection section 52 S insertion section 54 D transmission Section 56 S generation section 60 D reception section 62 S extraction section 66 S inspection section

Claims (6)

[Claims] 1. A data unit to which a sequence number is added in the order of transmission is received, and the sequence number of the received current data unit is compared with the sequence number of the previous data unit received immediately before the data unit to obtain the previous data unit. In a data unit receiving method of obtaining a distance from the current data unit and selecting saving or discarding of the current data unit from this distance, a distance from the previous data unit to the current data unit is determined, and this distance is normal. If it indicates that the current data unit is saved, if the determined distance is abnormal, it is determined whether the value is greater than a predetermined determination distance, if it is less than the determination distance The current data unit is stored in the A method for receiving a data unit, characterized in that the distance to the next data unit is determined, and if the distance indicates normal, the current data unit is saved, and if abnormal, the current data unit is discarded. 2. The data unit receiving method according to claim 1, wherein the data unit receiving method is normal when the distance from the previous data unit to the current data unit is “1”. 3. The data unit receiving method according to claim 1, wherein the data unit receiving method is normal when the distance from the current data unit to the next data unit is “1”. 4. The data unit receiving method according to claim 1, wherein the data unit is a fixed length data unit including an ATM cell or a variable length data unit including packet data. 5. A data unit to which a sequence number is added in the order of transmission is received, and the sequence number of the received current data unit is compared with the sequence number of the previous data unit received immediately before the data unit to obtain the previous data unit. In a communication interface device that determines the distance from the current data unit and selects storage or discard of the current data unit from this distance, the distance from the previous data unit to the current data unit is determined, and if this distance is normal, If it is indicated, the first judging means for storing the current data unit; and, if the judged distance indicates an abnormality, it is judged whether or not the value is larger than a predetermined judging distance, Second judging means for storing the current data unit if the judgment distance is smaller than the judgment distance; In this case, the distance from the current data unit to the next data unit is further determined, and if the distance is normal, the current data unit is saved, and if it is abnormal, the current data unit is discarded. And a third determination means for performing the communication interface device. 6. The communication interface device according to claim 5, wherein the first determination means and the third determination means are normal when the distance between the sequence numbers of the data units to be compared is “1”. A communication interface device characterized by the above.