JP2000069064A - Method and device for transmitting data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device improved for transmitting data so as to surely detect abnormality according to a communication system based on a tossing protocol. SOLUTION: This data transmission equipment is provided with a transmission counter 13, reception counter 14 and duration timer 15 in addition to a transmission processing part 11 and a reception processing part 12 for transmitting data through a communication network. The transmission counter 13 is a counter for recording how much data are transmitted, and the reception counter 14 is a counter for recording how much data are received. The duration timer 15 is a timer for presenting the time to be waited for error detection at the time of transmission. When it is judged the present equipment has the priority at the time of packet collision, the transmission processing part 11 does not immediately try packet retransmission but performs the retransmission after the duration timer 15 is turned off. While utilizing this duration, the abnormality (such as the multi-body collision of packets, for example), on the network can be surely detected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data transmission method and a data transmission apparatus for exchanging data in packet format between a transmitting terminal and a receiving terminal via a transmission path such as a LAN. The present invention relates to a data transmission method and a data transmission apparatus of the Janken protocol type, in which a terminal on the transmission side adds data (Janken header) indicating priority at the time of collision to a packet and transmits the packet.

More specifically, the present invention relates to a data transmission method and a data transmission apparatus capable of reliably detecting an abnormality in a communication system according to the Janken protocol.

[0003]

2. Description of the Related Art Recently, the term "network computing" has been used in various media such as newspapers and magazines. Network computing, as its name implies, refers to an environment in which a plurality of computers and peripheral devices are connected via a network. The main purpose of network computing is to share computer resources and to share and distribute information.

A LAN (Local Area Network)
Ork) is a minimum unit of a network that covers only a relatively small area such as one campus, and has the property of being independently operated and managed by an individual organization such as a university or a research institution. LANs can be laid out easily without licenses or licenses related to the telecommunications business, and are supported by lower prices for telecommunications equipment and higher functionality of telecommunications software due to improvements in semiconductor technology. It has penetrated deeply into the home.

[0005] Terminal equipment (DTE: Data Termi)
nal Equipment)
The system uses a line termination device (DCE: Data Cir)
It is connected to a LAN via a LAN adapter as a “quit Terminal Equipment”. L
The terminals connected by AN can perform data communication by following a predetermined communication protocol. The standard model of the communication protocol is (OpenSystems Interco
(connection: open system interconnection) reference model. The OSI reference model includes seven layers: a physical layer, a data link layer, a network layer, a transport layer, a session layer, a presentation layer, and an application layer (well-known).

[0006] In 1976, Xerox Co., Ltd., in the United States, started to use LAN called Ethernet ("Ethernet") as one of the realizing elements of "paperless office".
It originated in the announcement of AN products. Ethernet has a bus-type network topology, and uses CSMA / CD (Ca) as a data link layer protocol.
rrier Sense Multiple Access
ss with Collision detecti
on) method is adopted. The CSMA / CD system is a communication system in which a terminal transmitting a packet monitors a transmission path to check for a signal (carrier), and confirms that there is no carrier before transmitting a packet. Unfortunately, when two or more terminals transmit packets at the same time and detect a collision, each transmitting terminal waits for a predetermined waiting time and then tries transmission again.

[0007] Recently, the use of optical fibers in LANs has been studied in order to increase the transmission speed. However, the CSMA / CD system, which is still the mainstream communication protocol, has a large waste of bandwidth due to packet collision and cannot fully enjoy the benefits of high-speed transmission. This is because the CSMA / CD protocol specifies that all transmitting terminals be backed up in the event of a collision.

[0008] Japanese Patent Application Laid-Open No. 5-25 assigned to the present applicant
No. 2170 (“Optical communication network and communication system”) discloses a CSM for an optical fiber network.
A communication protocol that can obtain better performance than the A / CD protocol is disclosed. The communication protocol is based on the CSMA / CD protocol, but when a packet collision occurs, one of the transmitting terminals can immediately retransmit without backing off.

The operation at the time of collision as disclosed in Japanese Patent Application Laid-Open No. 5-252170 is made possible by using an interconnectable star coupler for the LAN connection of each terminal. Also depends. The star coupler is disclosed in, for example, Japanese Patent Application Laid-Open
No. -3457 ("Star coupler and optical communication network"). This interconnectable star coupler has the property that an optical signal input from one port is distributed to ports other than its own port. Therefore, the terminal transmitting the packet does not receive the packet transmitted by itself. If any signal is received during transmission, it is the signal of the packet transmitted by another terminal. In other words, it means that a packet collision has occurred on the transmission path.

A collision between two packets (ie, "2
If "body collision"), each transmitting terminal can normally receive the packet of the other party. However, when three or more packets collide (that is, “many collisions”), interference occurs, and no transmitting terminal can receive a normal packet. In the case of multiple collisions, the transmitting terminal uses CRV (Code Rule Viola) at the physical layer.
collision) (i.e., an abnormality) can be detected by using the method. Abnormality detection is reported to the upper protocol layer. Note that terminals other than the sending terminal
In a packet collision, whether it is a two-body collision or a many-body collision, an interference signal is received.

In the communication protocol disclosed in Japanese Patent Laid-Open No. 5-252170, a bit string indicating the relative priority between packets is added to the header of the packet. The priority is determined by the value of the bit string.
In the case of a two-body collision, each transmitting terminal compares the bit sequence of its own packet with the bit sequence of the colliding partner terminal. Then, the terminal having its own value is given the priority of packet retransmission, and starts transmitting packets immediately after jamming. Conversely, the transmitting terminal with the smaller bit sequence enters a back-off operation. Also, if the values of the bit strings are the same,
Both terminals enter backoff. This protocol is hereinafter referred to as a “Janken protocol”, and the bit string indicating the priority is referred to as a “Janken header”.

According to the communication system disclosed in Japanese Patent Application Laid-Open No. 5-252170, even if a collision occurs, one of the packets can be transmitted immediately after jamming. In addition, since the retransmission is performed immediately after the transmission of the jamming packet, the other terminal that has collided is still in the carrier idle waiting state, so that the retransmitted packet is unlikely to collide with another packet again.

As described above, Japanese Patent Application Laid-Open No. 5-252170
In the case of the communication system disclosed in Japanese Patent Application Laid-Open Publication No. H10-264, in order to detect an interference signal in the physical layer, CRV (Code Rule) is used.
(Violation) method is assumed to be used. However, CRV
Since the device itself is highly functional, the abnormality detection device is expensive. Further, since this communication method uses the redundancy of codes, it is not always possible to reliably detect an abnormality on a transmission line.

[0014]

SUMMARY OF THE INVENTION It is an object of the present invention to provide an excellent data transmission method for exchanging packet data between a transmitting terminal and a receiving terminal via a transmission path such as a LAN. A data transmission device is provided.

A further object of the present invention is to provide an excellent data transmission method of the Janken protocol type, in which a transmitting terminal adds data indicating the priority at the time of collision (Janken header) to a packet and transmits the packet. And a data transmission device.

A further object of the present invention is to provide an excellent data transmission method and an excellent data transmission apparatus capable of reliably detecting an abnormality in a communication system according to the Janken protocol.

However, in this specification, the “abnormality” referred to below refers to a phenomenon in which two or more terminals have determined that they have the right to retransmit. When an abnormality occurs, two or more transmitting terminals determine that they have priority by comparing the Janken header due to a multi-body collision or an error on the transmission path, and retransmit the packet. Even if resent, a collision occurs naturally. At this repetition, the network system breaks down. If it is determined that only one terminal has priority, such repetition of collision does not occur.
I don't say "abnormal".

[0018]

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above problems, and a first aspect of the present invention is that a terminal on a transmitting side has a priority on a transmission line to which a plurality of terminals are connected. Is a type of data transmission method in which a packet is added with a Janken header indicating the packet transmission and a packet retransmission priority is given by comparing the value of the Janken header of each packet when a packet collision occurs. (A) measuring the amount of transmission data;
(B) a step of measuring the amount of received data; (c) a step of judging whether or not there is a priority at the time of a packet collision; and (d) a retransmission wait when judging that there is a priority. Measuring time,
(E) retransmitting the colliding packet after the retransmission waiting time has elapsed.

Here, in the step (d), the measurement of the waiting time is started when the transmission data amount reaches a value indicating the completion of the transmission of the Janken header, and the measurement is completed in a certain period. Is also good.

At least a certain period of time during which the measurement of the waiting time is completed includes at least the maximum round-trip propagation delay time, the time required for comparing the Janken header, the time required for transmitting the jam packet, and the inter-frame space (IFS). ) The time during which carrier sense is not performed, and the MAC (Media Access Control).
l) One clock time in the layer may be included.

Also, an inter-frame space (IF
S) The time during which carrier sense is not performed may not exceed the time required to transmit a jam packet.

The maximum round-trip propagation delay time may be a time required to transmit a packet having a maximum packet length.

The time required for the comparison of the Janken header may be one clock time.

According to a second aspect of the present invention, on a transmission line to which a plurality of terminals are connected, a Janken header indicating priority is added to a packet at the time of transmission, and the packet is transmitted at the time of a packet collision. A data transmission device of the type which gives priority to packet retransmission by comparing the value of a Janken header, wherein (a) a transmission amount measuring means for measuring the transmission data amount, and (b) a reception data amount measurement. (C) a waiting time measuring means for measuring a retransmission waiting time when it is determined that it has a priority, and (d) a transmission processing means for performing packet transmission processing according to the Janken protocol. , (E) reception processing means for performing packet reception processing according to the Janken protocol.

Here, the waiting time measuring means (c) comprises:
The measurement of the waiting time may be started when the transmission data amount reaches a value indicating the completion of the transmission of the Janken header, and the measurement may be ended in a certain period.

At least a certain period of time during which the measurement of the waiting time is completed is at least the maximum round-trip propagation delay time, the time required to compare the Janken header, the time required to transmit the jam packet, and the inter-frame space (IFS). ) The time during which carrier sense is not performed, and the MAC (Media Access Control).
l) One clock time in the layer may be included.

Further, the inter-frame space (IF
S) The time during which carrier sense is not performed may not exceed the time required to transmit a jam packet.

The maximum round-trip propagation delay time may be a time required to transmit a packet having a maximum packet length.

The time required for the comparison of the Janken header may be one clock time.

[0030]

The data transmission apparatus 100 according to the present invention has the configuration shown in FIG.
As shown in FIG. 1, the transmission processing unit 11 and the reception processing unit 12, which enable data transmission via a communication network, include a transmission counter 13, a reception counter 14, and a waiting time timer 15.

The transmission processing unit 11 and the reception processing unit 12 perform data communication according to the Janken protocol. Further, the communication device 100 preferably has a structure in which an output signal from a transmission signal line is input to a reception signal line of another transmission device.

During transmission of a packet, the transmission counter 13
This is a counter that records how much data has been transmitted. The counting unit may be a bit unit or a byte unit.

The reception counter 14 is a counter for recording how much data has been received during packet transmission. The counting unit may be a bit unit or a byte unit.

The waiting time timer 15 is a counter for displaying the time to wait for error detection during transmission.

In the event of a packet collision, if the transmission processing unit 11 determines that it has a priority, it does not immediately try to retransmit the packet, but retransmits after the waiting time timer 15 has expired. Utilizing this waiting time, it is possible to reliably detect an abnormality on the network (for example, multiple collision of packets). In addition, CRV (Code Rule Vi)
It is not necessary to use an expensive abnormality detection device having a high function such as the above-described operation.

Still other objects, features and advantages of the present invention are:
It will become apparent from the following more detailed description based on the embodiments of the present invention and the accompanying drawings.

[0037]

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

FIG. 2 shows a conceptual diagram of a communication network to which a plurality of data transmission devices in this embodiment are connected.

Many data transmission devices are connected to the network, and data can be exchanged via the network according to a predetermined communication protocol. Each data transmission device shall have the ability to transmit and receive using the Janken protocol (described above).

If the transmission medium of the network is an optical fiber, the gist of the present invention can be realized by using an interconnectable star coupler (described above). Also,
If the transmission medium is a twisted pair cable, the signal lines in the cable are divided into those for transmission and those for reception, and signals from the transmission signal lines of the own terminal are received by terminals other than the own terminal at the hub or switch. The present invention can be applied by changing the structure so that the signal is output to the use signal line. However, in the following, description will be made assuming that the transmission medium is an optical fiber.

As already described with reference to FIG. 1, the terminal device 100 includes a transmission counter 13 and a reception counter 14.
, A waiting time timer 15, a transmission processing unit 11, a reception processing unit 12, and a physical interface unit (not shown in FIG. 1).

The transmission counter 13 starts counting at the same time as the start of packet transmission. If the packet transmission ends normally or ends halfway, the counting ends there. The counting method is
The count may be either down or count up. If the count is down, the initial value is the packet size of the transmission packet, and if the count is up, the initial value is 0.

The reception counter 14 starts counting at the same time as the start of packet reception, and terminates counting when packet reception ends normally or halfway. The counting method is
The count may be either down or up. If the count is down, the initial value is the maximum packet size, and if the count is up, the initial value is 0.

The transmission processing section 11 is a section for performing transmission processing of the Janken protocol. The transmission processing procedure performed by the transmission processing unit 11 is basically a CSMA / CD (Car
rier Sense Multiple Acces
s with Collision Detection
n) It is almost the same as that according to the protocol (described above), but differs from the CSMA / CD protocol in the processing of adding a Janken header and the retransmission processing in the event of a collision.

The reception processing unit 12 is a part that performs transmission processing of the Janken protocol. The reception processing procedure performed by the reception processing unit 12 is almost the same as that of the CSMA / CD protocol. The difference is in the handling of packet headers. Unlike the packet format of the CSMA / CD protocol, a Janken header is added, so a process for removing it is necessary.

The waiting time timer 15 is a timer for measuring the waiting time.
Starts counting from the rising edge of the clock immediately after indicating the count number after transmission of the Janken header. The timing ends when a certain period of time described later is ended, or when the value of the Janken header of the packet of the own terminal is not larger than the value of the counterpart packet by comparing the Janken header.

At the time of transmission, the transmission processing unit 11 adds a random value to each packet as a Janken header. At the time of packet collision, the priority is determined by this value. Hereinafter, the transmission process at the time of collision will be described with reference to FIG.

When a transmission request is received from a protocol upper layer, a packet is first created (step S101). At this time, a Janken header is added to the packet. Then, the carrier on the transmission path is monitored (step S10).
2) Wait until the carrier disappears. After the carrier disappears, the transmission is started (step S104) after waiting for an interframe space (IFS) time to elapse (step S103). For example, IEEE802.3
If so, this IFS time is 96 bits time. This means the time required to transmit 96-bit data, and if the transmission speed is 100 Mbps, 0.96
Microseconds.

The inter-frame space time includes the first half Part 1 (IFS1 bit time) and the second half Part 1
2 (IFS 2 bit time). For example, I
In the case of EEE802.3, Part1 is 64 bits time and Part2 is 32 bits time.

During the period of the part 1, the presence or absence of a carrier is continuously monitored (step S106). If a carrier is detected, the transmission process is stopped, and the process waits until the carrier is exhausted again. On the other hand, even if a carrier is detected during Part2,
Transmission processing continues.

After the start of transmission, if transmission is completed without receiving a carrier, transmission ends normally (S106 → S
104 → S105). At this time, at the end of the packet,
(End of Stream Delimiter)
Is inserted and transmitted.

On the other hand, if a carrier is received during transmission, it can be determined that another terminal is also transmitting, that is, a collision has occurred. However, it is necessary to compare the Janken header of each other at the time of collision (see above), at least, janken header H ₁ of the terminal has a packet being transmitted
The transmission process is continued until all the messages have been transmitted (S106).
→ S107). Whether or not all the Janken headers have been transmitted can be determined by referring to the value of the transmission counter 13.

Further, it is necessary to receive all the Janken headers H ₂ of the packet being received. Whether it has been received can be determined by referring to the value of the reception counter 14. At the rising edge of the clock immediately after sending all the Janken headers of the own terminal, time measurement by the monitoring timer is started (step S108).

Janken header H of the packet being received
₂ at the rising edge of the clock immediately after the completion of reception (step S109), the two Janken headers H ₁ and H ₂
Are compared (steps S110, S111).

As a result of the comparison in decision block S111, the Janken header H of the packet transmitted by the own terminal is obtained.
₁ is smaller or two Janken headers H ₁
And when the value of H ₂ are equal, immediately switches to the transmission of the jam packet (packet length J bits) (step S112).

After that, according to the back-off algorithm defined in IEEE802.3 (step S11).
3). That is, when the transmission of the jam packet is completed, the backoff time is calculated, and the retransmission is suspended for the backoff time. If the number of collisions is 16 or more, the upper layer of the protocol is notified that the detection event is an error (S114 → S115).

On the other hand, if the number of collisions is less than 16, the process returns to step S101 to detect the carrier again and perform the transmission process again.

[0058] Two Janken header H ₁ and the results of the comparison of H ₂ at decision block S111, the larger the better the Janken header H ₁ packet during transmission, continues to transmit until the monitoring timer expires (S 111 → S11
6).

At the time when the monitoring timer expires, it is determined whether a carrier is being received (step S117).

If a carrier has been received, it is determined that a terminal other than the own terminal determines that the terminal has priority and continues to transmit packets. Since this is caused by many-body collision or some error, the same error processing as described above is performed. That is, a jam packet (packet length J bits) is immediately transmitted (step S11).
2) After that, it follows the back-off algorithm specified in IEEE802.3 (step S113).

On the other hand, if the carrier is not received at the time when the monitoring timer expires, it can be understood that the priority is properly determined. In this case, the transmission of the packet being transmitted is stopped, an IDLE signal for one clock is inserted, and the packet is immediately retransmitted from the beginning (S11).
7 → S101).

By transmitting the IDLE signal, the other terminal can recognize that the received data is a collision packet. Because, at the end of the transmission data, there is an ESD, and after that, receiving the IDLE is a normal end, and the ESD
The reason why IDLE is received without receiving is that it is possible only when a collision occurs.

Next, a method of calculating a certain time measured by the above-mentioned waiting time timer 15 will be described. This fixed time is determined so that an error can be detected even in the worst case. An error is a phenomenon in which two or more terminals determine that the own terminal has a retransmission right at the time of packet collision.
That is, as a result of comparing each other's Janken header,
Due to a multi-body collision or an error on the transmission path, two or more terminals determine that the Janken header of their own packet is large, and each terminal retransmits the packet. Even if resent, collision still occurs. After that, this repetition causes the system to fail. (If only one terminal determines that the packet's Janken header is large, such a situation does not occur and is not treated as an error in this specification.)

In order to prevent an error from occurring, a terminal that has determined that its own terminal has a retransmission right waits for a signal of another terminal until a certain period, and if a carrier is still received, determines that an error has occurred. You should be able to do it.

Here, the certain period refers to a period during which an error can be detected even in the worst situation. The worst case is when it takes the longest time until an error is found. In other words, this is a case where a collision with a packet of another terminal occurs as soon as possible, rather than immediately after transmission. This will be described below with reference to FIG.

In FIG. 4, the packet of the transmitting terminal (terminal B) collides with the packet transmitted by the terminal (terminal A) farthest from the terminal B on the network.
If the carrier is not received even after counting only the IFS 1-bit time, the transmission starts after the IFS 2-bit time. In the worst case, the terminal A
Transmission is started after receiving the FS2 bit. FIG. 4 illustrates just this case.

If there is no error, the terminal A receives the Janken header of the terminal B and compares the value of the Janken header with each other. Then, if it is determined that the value of the Janken header is small, a jam packet is transmitted. When the jam packet is received (the time is set to T _j
After that, an IDLE signal is transmitted.

However, if an error has occurred and terminal A determines that its own Janken header is larger,
Terminal A continues to transmit packets. Therefore, terminal B
It is, even after the time T _j, continue to receive the packet. This reveals the error.

Therefore, the start of the calculation of the time that the terminal B should wait is when the terminal A has finished transmitting the Janken header, and the end is when the terminal A has finished receiving the jam packet, that is, the time _Tj . Is represented by the following equation.

[0070]

(Equation 1)

In the above expression, one clock time includes a margin determined by the clock immediately after the completion of the reception of the jam packet.

The propagation time from terminal A to terminal B and terminal B
The value obtained by adding the propagation times from the terminal to the terminal A is the round-trip propagation delay time between the terminals when the distance between the terminals is the longest. Therefore, the waiting time is as follows.

[0073]

(Equation 2)

The maximum round-trip propagation delay time is often reflected in the minimum packet length (bit unit). That is, the minimum packet length corresponds to the maximum round-trip propagation delay time. Further, assuming that the time required for the comparison is C clock, if the apparatus transmits B bits in one clock,
When converted to bit time, it becomes CB bit time. Therefore, the above equation is as follows.

[0075]

(Equation 3)

The Janken header comparison must be at least 1
A clock is needed. In other words, C = 1 is the minimum value, and the optimal waiting time is expressed by the following equation.

[0077]

(Equation 4)

Here, assuming, for example, the parameters defined by IEEE 802.3, the minimum packet length is 512 bits, the IFS is 32 bits, and the jam packet length is 32 bits. In the case of a data transmission device that transmits 8 bits in one clock, the waiting time is as follows.

[0079]

(Equation 5)

In a system in which transmission data is decoded at 4B / 5B and transmitted, when transmitting 10 bits with one clock, the waiting time is as follows.

[0081]

(Equation 6)

Next, the manner in which an error can be detected by the waiting time timer 15 will be described with reference to FIGS. 5, 6, and 7. FIG. However, in the following, it is assumed that the value of the maximum round-trip propagation delay time is directly reflected on the minimum packet length. For example,
In a network with a transmission speed of 100 Mbps, the maximum round-trip propagation delay time is 5.12 microseconds and the minimum packet length is 5
This is the case when it is 12 bits. The maximum round-trip propagation delay time is represented by τ seconds for the following description.

FIGS. 5 to 7 show that the packets transmitted by the terminal A and the terminal B collide with each other, and as a result of comparing the Janken headers with each other, it is erroneously determined that both terminals have priority due to many-body collision or the like. This is shown in a time chart. 5 shows a case where terminal A and terminal B start transmission almost simultaneously, FIG. 6 shows a case where terminal A starts transmission at the same time that a packet of terminal B arrives at terminal A, and FIG. , Respectively, when transmission is started after receiving two bits of IFS.

In these figures, the part indicated as “header” indicates transmission from the preamble to the Janken header, and the part indicated as “data” indicates transmission of the data body. WT is the time counted by the waiting time counter.

When the terminal A and the terminal B start transmission almost simultaneously, as shown in FIG. 5, both the terminal A and the terminal B
After the WT has elapsed, the data of the other party is continuously received. For this reason,
Each terminal can detect the abnormality.

When terminal A starts transmission at the same time that a transmission packet from terminal B arrives at terminal A, as shown in FIG. 6, terminal B continues to receive data of terminal A after the WT has elapsed. Can be detected. The terminal B that has detected the abnormality switches from data transmission to jam packet transmission. One terminal A can detect the abnormality by receiving the last data transmitted by the terminal B before the switching.

This will be described in more detail. Assuming that the time at which the last bit of the Janken header of terminal B is transmitted is t, the start of the waiting time timer 15 of terminal A is at time t + τ / 2. Then, it is time t + WT that terminal B starts transmission of the last bit that has detected an error and has stopped transmitting data, and that bit arrives at terminal A at time t + WT + τ / 2. On the other hand, the timer of terminal A expires at time t + τ / 2 + WT. Therefore,
The end of data will be received.

When the terminal A starts transmission after receiving the packet of the terminal B by 2 bits of IFS, as shown in FIG.
, So that an abnormality can be detected. Then, the terminal B that has detected the abnormality switches from data transmission to jam packet transmission. On the other hand, terminal A
Can detect an abnormality by receiving the jam packet of terminal B after the switching.

This will be described in more detail. Assuming that the start time of the waiting time timer 15 of the terminal B is t, in this case, the terminal A
The waiting time timer 15 starts counting time at time t +
τ / 2 + IFS 2 bit seconds. Then, the terminal B detects the error and starts transmitting the jam packet at time t + WT, and the end of the packet arrives at the terminal A at t + WT + τ / 2 + j (where j is J
Bit second). On the other hand, the timer 15 of the terminal A expires at time t + τ / 2 + ifs2 + WT (ifs2
Represents 2 seconds of IFS). Therefore, if j ≧ ifs2, that is, J ≧ IFS2, the last signal of the jam packet will be received when the timer expires. IEEE
According to the provisions of E802.33, J = 32, IFS
Since 2 = 32, detection is possible.

The transmission counter 13, reception counter 14, waiting time timer 15, transmission processing unit 1 shown in FIG.
1. Each component of the reception processing unit 11 is a single LSI (L
(Arrange Scale Integration) chip.

The physical layer interface (not shown) is an interface between the transmission medium and the terminal device. If the transmission medium is an optical fiber, the physical layer interface is a light emitting / receiving circuit.

The transmission medium may be an optical fiber or a twisted pair cable.

In the above-described embodiment, the IDLE signal is transmitted when data transmission is stopped in the event of a collision.
Even if an ESD is inserted and transmitted, the other terminals do not receive CSMA /
An error can be detected by using the same error detection mechanism as the CD protocol. For example, FCS
If a collision packet is detected by a frame error check by (Frame Check Sequence), a frame size check, a destination address check, etc., if any of them are abnormal, the collision can be detected. it can.

[Supplement] The present invention has been described in detail with reference to the specific embodiments. However, it is obvious that those skilled in the art can modify or substitute the embodiment without departing from the spirit of the present invention. That is, the present invention has been disclosed by way of example, and should not be construed as limiting. In order to determine the gist of the present invention, the claims described at the beginning should be considered.

[0095]

As described above in detail, according to the present invention,
It is possible to provide an excellent data transmission method and an excellent data transmission device for exchanging packet format data between a transmitting terminal and a receiving terminal via a transmission path such as a LAN.

Further, according to the present invention, an excellent data transmission method of the Janken protocol type, in which the transmitting terminal adds data indicating the priority at the time of collision (Janken header) to a packet and transmits the packet. And a data transmission device.

Further, according to the present invention, it is possible to provide an excellent data transmission method and an excellent data transmission apparatus capable of reliably detecting an abnormality in a communication system according to the Janken protocol.

In the communication network to which the present invention is applied,
An expensive abnormality detection means (for example, CRV (Code
Rule Violation))
MAC (Media Access Control)
Since abnormality can be detected in the layer, it is possible to reliably detect errors at low cost.

[Brief description of the drawings]

FIG. 1 is a diagram conceptually showing a configuration of a data communication device that realizes the present invention.

FIG. 2 is a diagram conceptually illustrating a communication network connecting data communication apparatuses according to the embodiment.

FIG. 3 is a diagram illustrating a transmission flowchart according to the embodiment of the present invention.

FIG. 4 is a time chart showing a configuration of a waiting time in the embodiment of the present invention.

FIG. 5 is a time chart showing a collision pattern according to the embodiment of the present invention, and more specifically, a time chart in a case where terminals A and B start transmission almost simultaneously.

FIG. 6 is a time chart showing a collision pattern according to the embodiment of the present invention, and more specifically, a time chart in a case where terminal A starts transmission at the same time that a packet of terminal B arrives at terminal A;

FIG. 7 is a time chart showing a collision pattern in the embodiment of the present invention, and more specifically, a time chart in a case where terminal A starts transmitting after receiving a packet of terminal B by 2 bits of IFS. is there.

[Explanation of symbols]

 REFERENCE SIGNS LIST 11 transmission processing unit 12 reception processing unit 13 transmission counter 14 reception counter 15 waiting time timer 100 data communication device

Claims (12)

[Claims] A transmitting terminal adds a Janken header indicating a priority to a packet on a transmission line to which a plurality of terminals are connected, and transmits the packet. When a packet collision occurs, a value of the Janken header of each packet is added. A data transmission method of a type that gives priority of packet retransmission by comparing,
At the time of packet transmission, (a) a step of measuring the amount of transmission data, (b) a step of measuring the amount of reception data, (c) a step of determining whether or not there is a priority at the time of packet collision, d) measuring a retransmission waiting time when it is determined that the user has the priority, and (e) retransmitting a packet that collided after elapse of the retransmission waiting time. Data transmission method. 2. In the step (d), the measurement of the waiting time is started when the transmission data amount reaches a value indicating the completion of the transmission of the Janken header, and the measurement is completed in a predetermined period. The data transmission method according to claim 1, wherein: 3. A fixed period for ending the measurement of the waiting time includes at least a maximum round-trip propagation delay time, a time required for comparing a Janken header, a time required for transmitting a jam packet, and an inter-frame space (IF).
S) Time without carrier sense in time and MA
3. The method according to claim 2, further comprising one clock time in a C (Media Access Control) layer.
2. The data transmission method according to item 1. 4. The data transmission method according to claim 3, wherein a time during which no carrier sense is performed in an inter-frame space (IFS) time does not exceed a time required for transmitting a jam packet. 5. The data transmission method according to claim 3, wherein the maximum round-trip propagation delay time is a time required for transmitting a packet having a maximum packet length. 6. The data transmission method according to claim 3, wherein the time required for the comparison of the Janken header is one clock time. 7. On a transmission line to which a plurality of terminals are connected, a Janken header indicating priority is added to a packet at the time of transmission and transmitted, and at the time of packet collision, the value of the Janken header of each packet is compared. A data transmission device of the type that gives priority to packet retransmission in
(A) transmission amount measuring means for measuring a transmission data amount;
(B) a reception amount measuring means for measuring a reception data amount;
(C) a waiting time measuring means for measuring a retransmission waiting time when it is determined that it has a priority, (d) a transmission processing means for performing packet transmission processing according to the Janken protocol, and (e) a A data transmission device comprising: a reception processing unit that performs a reception process of a packet according to a protocol. 8. The waiting time measuring means (c) starts measuring the waiting time when the transmission data amount reaches a value indicating the completion of transmission of the Janken header, and ends the measurement in a certain period. The data transmission device according to claim 7, wherein: 9. A fixed period for ending the measurement of the waiting time includes at least a maximum round-trip propagation delay time, a time required for comparing a Janken header, a time required for transmitting a jam packet, and an inter-frame space (IF).
S) Time without carrier sense in time and MA
9. The method includes one clock time in a C (Media Access Control) layer.
A data transmission device according to claim 1. 10. The data transmission apparatus according to claim 9, wherein a time during which no carrier sense is performed in an inter-frame space (IFS) time does not exceed a time required for transmitting a jam packet. 11. The data transmission apparatus according to claim 9, wherein the maximum round-trip propagation delay time is a time required for transmitting a packet having a maximum packet length. 12. The data transmission apparatus according to claim 9, wherein the time required for comparing the Janken header is one clock time.