JP2006191339A - Gateway apparatus for transferring message between buses - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a gateway apparatus capable of grasping a congestion stat of a message between networks interconnected by the gateway apparatus. <P>SOLUTION: The gateway apparatus provided between a plurality of the network BUS-1, 2 to which a plurality of device ECUs are respectively connected and transferring the messages MES between the networks includes: a transfer control means TCON for sequentially transmitting the message MES received from the sender network BUS-1 to a transmission section network BUS-2; and a transfer history storage means LOG for storing transfer history information related to reception and transmission of the message. Then the transfer history information stored in the transfer history storage means LOG can externally be read. Since the gateway apparatus is configured to store the transfer history information of the message and to be externally readable, the gateway apparatus can grasp the congestion state of messages between the networks thereby exploring practical ways to improve relaxing of the congestion. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

The present invention relates to a gateway device that performs message transfer between buses, and more particularly, to a gateway device that can store a delay history of message transfer between CAN (Controller Area Network) buses.

CAN is a serial communication protocol internationally standardized by ISO, and is mainly an in-vehicle LAN that connects ECUs (Electric Control Units) that control the hardware of automobiles.
It is widely used as a communication protocol. Recently, because of the high performance and reliability of CAN, its applications have expanded from automobiles to factory automation, ships, medical equipment, industrial equipment, and so on. For example, CAN is described in Non-Patent Documents 1 and 2 below. Further, a vehicle control system using an in-vehicle LAN is described in, for example, the following Patent Documents 1 and 2.
“CAN Introduction” www.renesas.com “What is CAN?” Www.toyo.co.jp/car/CAN/CAN_General.htm JP-A-2004-136816 Japanese Patent Application Laid-Open No. 2004-17676 describes a network system in a vehicle in which a plurality of ECUs are connected by a CAN and a plurality of CANs are connected via a gateway device. For example, an engine CAN that connects an ECU that controls an engine and a transmission and an information CAN that connects an ECU that controls a VICS navigation system are connected via a gateway device, and responds to a message from the information system. Therefore, it can be reflected in engine system control. Further, in Non-Patent Document 1, a failure diagnosis system CAN and an engine system CAN are connected via a gateway device. For example, from a failure diagnosis system CAN diagnosis tool to an engine control ECU of an engine system CAN. On the other hand, it is possible to enable control for diagnosis.

  In this way, the gateway device translates protocols between different CAN buses, and transfers a message transmitted from the ECU in one CAN bus to the ECU in the other CAN bus. According to the CAN specification, as will be described later, an identifier indicating the priority order of transmission rights in the bus is added to the message, and transmission rights are arbitrated between simultaneously transmitted messages. On the other hand, when transmitting a message received from one bus to the other bus, the gateway device transmits the message to the destination bus in the order received in principle. In addition, the received message is transmitted to the destination bus at the transmission cycle corresponding to the identifier that defines the priority order. In this way, important and urgent messages are distinguished from other messages by identifiers, and the messages are transferred with an urgency level corresponding to each message.

  In recent years, a network system that relays a plurality of CAN buses with a gateway device tends to enable higher-level control by transferring more messages. Thus, as more messages are transferred between the buses via the gateway device, it is expected that the transfer delay in the gateway device will seriously affect the performance of the network system. The analysis of the transfer delay in the gateway device is indispensable for grasping the congestion status of messages in the network, but the conventional gateway device does not have a function for grasping such a congestion status. .

  Accordingly, an object of the present invention is to provide a gateway device that makes it possible to grasp the congestion status of messages between networks connected by the gateway device.

  In order to achieve the above object, according to a first aspect of the present invention, in a gateway device that is provided between a plurality of networks to which a plurality of devices are respectively connected and transfers a message between the networks, Transfer history information stored in the transfer history storage means, including transfer control means for sequentially transmitting received messages to a destination network, and transfer history storage means for storing transfer history information related to reception and transmission of the messages. Is readable from the outside.

  In the first aspect described above, according to a preferred embodiment, an identifier defining a priority of transmission rights in the network is added to the message, and the transfer control means transmits a transmission period corresponding to the identifier. The transfer history storage means stores the transfer history information for each identifier, and the transfer history information corresponding to the identifier can be read out to the outside. Features.

  In the first aspect described above, according to a further preferred embodiment, the transfer history information includes the reception time and transmission time of the message, the time from reception to transmission, the number of messages waiting for transmission after reception, and the message waiting for transmission after reception. It includes at least one of staying information indicating whether the number exceeds a predetermined number or transmission abnormality information indicating whether the number is transmitted a plurality of times after reception.

  According to the first aspect described above, the gateway device stores the message transfer history information and is configured to be readable from the outside. Therefore, it is possible to grasp the congestion status of the messages between the networks, and to check the congestion level. It can be used for mitigating improvements.

  Further, according to the preferred embodiment, since transfer history information is stored for each identifier, it is possible to easily grasp which identifier message has a retention at the time of transfer, which is useful for grasping the congestion status of messages. be able to.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the technical scope of the present invention is not limited to these embodiments, but extends to the matters described in the claims and equivalents thereof.

  FIG. 1 is a diagram showing a configuration of a gateway device and a message transfer operation in the present embodiment. In FIG. 1, a gateway unit 10 for transferring messages between buses is provided between a first bus BUS-1 and a second bus BUS-2 which are CAN. Two devices ECU-1 and ECU-2 and a gateway unit 10 are connected to the first bus BUS-1, and two devices ECU-3 and ECOU-4 are connected to the second bus BUS-2. , Gateway unit 10 is connected. The first and second buses BUS-1 and BUS2 each constitute an individual network, and messages are exchanged between devices in the network using respective protocols. Each device ECU monitors and controls the state of hardware (not shown). The gateway unit 10 translates protocols and transfers messages between different networks.

The gateway unit 10 includes a reception buffer R-BUF for transferring a message from the first bus BUS-1 to the second bus BUS-2, a transfer controller TCON,
A transfer queue buffer T-QUE for storing a received message to be transferred. The transfer queue buffer T-QUE is a queue that stores received message groups to be transferred in the order of transfer. Similarly, it has a similar configuration for transferring messages in the reverse direction. Then, the message MES1 transmitted from the device ECU connected to one bus is transferred by the gateway unit 10 to the device ECU connected to the other bus.

  The features of the CAN protocol are (1) a multi-master system in which all devices (ECU) can send messages if the bus is free, and (2) all devices should receive and receive messages simultaneously. (3) No address is assigned to a device in the network, and the content or type of the message is defined by a unique identifier (ID) in the network associated with the message. , This ID is added to the transmission message, and the flexibility of the network that spontaneously receives the message of the ID that the receiving device should receive (therefore, the transmission source and the destination device are specified by this ID), and (4) The ID added to the message is the priority of the transmission right on the network This is a collision avoidance method in which when a plurality of devices try to send a message at the same time, a message with a higher priority ID is given priority and a transmission right is acquired, and a message with a lower priority ID is stopped. .

  Therefore, hereinafter, the transmission operation will be described by taking as an example the case where the device ECU-1 of the first bus BUS-1 transmits the message MES1 to the device ECU-3 of the second bus BUS-2. First, the device ECU-1 transmits a message MES1 to the bus when the first bus BUS-1 is in an idle state. In simple terms, the message MES1 is composed of a frame including an identifier ID indicating the type of message and including several bytes of data DATA. The transmission right on the first bus BUS-1 is first given to the device that has output the message frame to the idle bus first, and secondly, when the message frame is output simultaneously, It is given to the device that outputs the message having the higher priority given to the identifier ID added to the message. That is, all identifier IDs have different priorities. When the devices ECU-1 and ECU-2 output the messages MES1 and MES2 to the bus at the same time, the transmission right is determined according to the priorities of the identifier IDs. Transmission of the message MES1 having an identifier ID having a higher rank is continued, and transmission of the message MES2 having an identifier ID having a lower priority is stopped.

  The transmission right determination algorithm based on the identifier ID is performed by monitoring the state of the bus when the identifier ID is output in the CAN bus. In other words, the bus level has a dominant level (priority level) and a recessive level (acceptance level). In the case of a logically wired-and-bus, the logical value of the dominant level is “0”. , The logical value of the recessive level is “1”. If even one device outputs the dominant level “0”, the bus becomes the dominant level “0”. If all devices do not output the recessive level “1”, the bus does not become the recessive level “1”. Furthermore, “000” to “FFF” are assigned to the identifier ID, for example, in hexadecimal. As a result, the identifier “000” has the highest priority, and “FFF” has the lowest priority.

That is, when a plurality of devices output message frames at the same time, the data string of the identifier ID of the arbitration field located at the head of the frame is simultaneously output to the bus. Therefore, the bus level is controlled to the level of the data string of the identifier ID having the highest priority. Therefore, each device can detect whether or not the transmission right has been acquired by monitoring the bus level in each bit of the arbitration field. For example, in the case of identifier “000” having the highest priority, the bus can be set to a dominant level (logic “0”) in all 16 bits. It is detected that the state of is different from the state of the identifier output by itself. As a result, the device can detect that the transmission right cannot be acquired, and can stop the transmission of the message. On the other hand, when it is detected that the transmission right cannot be acquired in all 16 bits, the device has acquired the communication right and continues to transmit the message. As a result, data fields of a plurality of message frames are prevented from being output to the bus at the same time, and collision can be avoided.

  Returning to FIG. 1, when the device ECU-1 tries to output the message MES1, even if the device ECU-2 outputs the message MES2 at the same time, if the identifier ID of the message MES1 has a higher priority, The device ECU-2 stops outputting the message MES2, and eventually only the device ECU-1 can output the message MES1 to the bus BUS-1. Of course, when the device ECU-1 first starts outputting the message MES1, the other device ECU-2 does not start outputting the message.

  The message MES1 output to the bus BUS-1 is stored in the reception buffer R-BUF of the gateway unit 10. In the gateway unit 10, an identifier to be transferred to the second bus BUS-2 is set in advance, and only the message of the set identifier is stored in the reception buffer R-BUF.

  The transfer controller TCON in the gateway unit 10 stores the received message MES1 at the tail end of the transfer queue buffer T-QUE and the received message stored in the transfer queue buffer T-QUE at every predetermined transmission cycle. Are output to the second bus BUS-2 in the order of reception. Also in the second bus BUS-2, the transmission right is arbitrated in the same manner as the arbitration method in the first bus BUS-1. The message MES1 transmitted to the second bus BUS-2 is received by the device ECU-3 that should receive the message. That is, the device ECU-3 and ECU-4 connected to the second bus BUS-2 are set in advance as to which identifier message should be received, and the message MES1 with the set identifier is transmitted on the bus. It is determined from the message output to, and if applicable, it is received.

  Similarly to the above, the transmission of a message from the device connected to the second bus BUS-2 to the device connected to the first bus BUS-1 is also performed on the bus and transferred by the gateway unit 10. By the control.

  In this way, the normal message is stored in the transfer queue buffer T-QUE in the order received in the gateway unit 10 and transferred to the bus of the destination network in the order received.

  FIG. 2 is a diagram showing a detailed configuration with the gateway unit in the present embodiment. FIG. 3 shows, as an example, message identifier lists L1 and L2 transmitted by the devices ECU-1 and ECU-2, and message identifier lists L3 and L4 received by the device ECU-3 and ECU-4. . In this way, each device has a transmission identification list and a reception identification list set in advance. For example, when the device ECU-1 transmits a message to which the identifier “001” is added, the device ECU-3 receives the message via the gateway unit 10. Similarly, when the device ECU-2 transmits a message to which the identifier “002” is added, the device ECU-3 receives the message via the gateway unit 10. In this way, the message to be transmitted is given an identifier that identifies the transmission source and the transmission destination, and indicates the priority of transmission rights in the bus, not the transmission destination address.

Further, the gateway unit 10 is preset with an identifier of a message to be received in the reception buffer R-Buf, monitors the identifier of the message sent to the first bus BUS-1, and receives the message to be received in the reception buffer. Receive to R-Buf. The message received in the reception buffer R-Buf is temporarily stored in the message storage area in the shared memory 12. The transfer controller TCON checks the identifier ID of the received message, and among the transmission queue buffers Q1 to Qn provided individually for each identifier ID in the transmission queue buffer T-QUE, the corresponding transmission queue buffers Q1 to Qn. The received message is stored at the end of.

  In the example shown in FIG. 2, the message MES1 (identifier “001”, the transfer path is a broken line) transmitted by the device ECU-1 is sent to the first bus BUS-1, and received by the reception buffer R− of the gateway unit 10. The data is taken into the Buf, temporarily stored in the message area of the shared memory 12, and then stored at the end of the transmission queue buffer Q1 corresponding to the identifier ID. Similarly, the message MES2 (identifier “004”, the transfer path is a one-dot chain line) transmitted by the device ECU-2 is also sent to the first bus BUS-1, and taken into the reception buffer R-Buf of the gateway unit 10, It is temporarily stored in the message area of the shared memory 12, and then stored at the end of the transmission queue buffer Qn corresponding to the identifier “004” by the transfer controller TCON.

  Then, the transfer controller TCON sends the first message in the transmission queue buffer to the second bus BUS-2, which is the transmission destination network, in the transmission cycle corresponding to each identifier. Further, the arbitration of the transmission right in the second bus BUS-2 that is the transmission destination network is determined according to the priority order of the identifier added to the message. Further, the message transmitted to the second bus BUS-2 is detected and captured by the device ECUs 3 and 4 in the bus.

  As described above, in each bus, a message having a higher priority is preferentially sent out according to the priority of the identifier, and the gateway unit 10 transmits the transmission queue buffer Q1 corresponding to each identifier for each transmission cycle corresponding to the identifier. The head message of .about.Qn is sent to the destination bus. Therefore, messages having a higher priority are preferentially transferred within the bus or between buses. Therefore, in the gateway unit 10, transmission queue buffers Q1 to Qn are separately provided for each identifier, and received messages are stored in the respective transmission queue buffers Q1 to Qn. Then, at the transmission cycle corresponding to the identifier, the first message in the transmission queue buffer corresponding to the identifier is transmitted to the transmission destination bus.

  In the present embodiment, the shared memory 12 of the gateway unit 10 is provided with a transfer history area LOG for storing a message transfer history in addition to the message area. When the message is received in the reception buffer R-Buf, the transfer controller TCON stores the reception time in the transfer history area LOG. When the message is transmitted from the transmission queue buffer to the destination bus, the transmission controller TCON transmits the message. The time is stored in the transfer log area LOG. Alternatively, the transfer controller TCON stores the transfer time from reception to transmission in the transfer log area LOG for each message by identifier ID. Alternatively, as will be described later, the transfer controller TCON may store the number of received and untransmitted messages in the transfer history area LOG. Alternatively, the transfer controller TCON may store a stay flag in the transfer history area LOG when the number of received and untransmitted messages exceeds a predetermined number. Furthermore, the transfer controller TCON may store a transfer abnormality flag in the transfer history area LOG when a received message is unnecessarily sent to the destination bus a plurality of times. As described above, as the transfer history information, the simplest is information on the reception time and transmission time of all messages, information on the time required for transfer, or the number of messages staying in the gateway unit 10. Information on transfer or information on dwell time is stored in the transfer history area LOG, and information on transfer abnormalities when a received message is transmitted more than necessary.

The transfer history information in the transfer history area LOG is stored in the log interface means LOG.
It is configured to be readable from the outside of the gateway unit 10 via the G-IF. In other words, in a network system that transfers a plurality of network buses with a gateway device, if any failure occurs, the transfer status in the gateway unit 10 is read out to determine the transfer status of messages between networks, and the gateway unit Analyzing the status of message retention makes it easy to find the cause of a failure.

  Further, if it is found by checking this transfer history that a message with a specific identifier has accumulated significantly in the gateway unit 10, the transfer cycle of the message with this identifier to the destination bus is shortened. By taking measures such as resetting, reoccurrence of the failure can be avoided.

  The reading of the transfer history information in the transfer history area LOG from the outside is performed via the log interface LOG-IF shown in FIG. 2, but specifically, for example, the log interface LOG from an external failure diagnosis tool or the like. This can be realized by inputting an interrupt instruction to the transfer controller TCON via the -IF, reading the transfer history information in the transfer history area LOG from the transfer controller TCON, and outputting it to an external tool. In other words, a configuration that can be read from the outside can be achieved by the same configuration as the debugging tool of the microcomputer.

  FIG. 3 is a diagram illustrating an example of a transfer history area in the present embodiment. In this example, in the transfer history area LOG, the time when the gateway unit 10 received the message with the identifier corresponding to the identifier ID and the time when the message was transmitted to the destination bus are recorded as the transfer history. Is done. In FIG. 3, transfer history information consisting of the reception time and transmission time of identifiers ID = 000 and ID = NNN is recorded. For a message with identifier ID = 000, it is recorded that three messages have been received and only one message has been transmitted. As for the message with the identifier ID = NNN, it is recorded that four messages have been received and all transmitted. Therefore, by reading this transfer history from the outside and analyzing it, it is possible to know how the message was transferred and how it stayed in the gateway unit corresponding to each identifier ID. Simply, the residence time of the message in the gateway unit can be determined by calculating the difference between the reception time and the transmission time. Further, it is possible to know in detail the state of message retention in the gateway unit from the reception time and transmission time.

FIG. 4 is a view showing a state of message retention in the gateway unit obtained by using the transfer history of FIG. The horizontal axis indicates time t, the vertical axis indicates the number of staying messages (dashed line is the number of staying messages = 0), (1) shows the state of staying when identifier ID = 000, and (2) shows the state of staying when identifier ID = NNN. Show. According to FIG. 4 (1), the message (reception 1) received at the reception time LL: LL: LL: LL is transmitted at the transmission time MM: MM: MM: MM (transmission 1), and the reception time NN: NN. The messages (reception 2 and 3) received by: NN: NN and PP: PP: PP: PP have not been transmitted yet. In other words, it is possible to know that two messages have stayed in the gateway unit at the time of reception 3 and the transfer time of the first message. Further, according to FIG. 4 (2), the message (reception 1) received at the reception time LL: LL: LL: LL is then transmitted to the transmission time MM: MM; MM: MM (transmission 1), and the reception time. NN: NN: NN: NN, PP: PP: PP: PP The two messages (reception 2 and 3) received in succession are transmitted at transmission time OO: OO: OO: OO (transmission 2). ), After receiving further messages at reception time RR: RR: RR: RR (reception 4), messages were transmitted continuously at transmission times QQ: QQ: QQ: QQ and SS: SS: SS: SS (Transmission 3, 4) can be analyzed. In the example of FIG. 4B, the message group with the identifier ID = NNN can know that a maximum of two messages have accumulated in the gateway unit, the transfer time of each message, and the like.

  In this way, as shown in FIG. 3, by recording the history of the reception time and transmission time of all messages corresponding to each identifier, the transfer history is read from the outside and analyzed, so that the identifier Message tends to stay in the gateway unit, and the message with the identifier can be analyzed smoothly.

  Each time preferably includes a date, hour, minute, second, and microsecond. However, since the capacity of the transfer history area LOG has a certain limit, the number of messages held as the transfer history needs to be limited to a predetermined number. In this case, the old transfer history is deleted.

  FIG. 5 is a diagram showing another example of the transfer history area in the present embodiment. The transfer history information shown in FIG. 3 has the reception time and transmission time of all messages, and requires a predetermined large-capacity storage area in the shared memory 12 in the gateway unit 10. The gateway unit 10 is constituted by a microcomputer having a CPU and a program ROM constituting the transfer controller TCON, a shared memory 12, a transmission queue buffer T-QUE, and the like. Therefore, the storage capacity of the shared memory 12 may not be so large. Therefore, in the example of FIG. 5, a stay counter for monitoring the number of messages staying in the gateway unit is provided, and a stay flag and a transfer error flag are stored in the transfer history area LOG as transfer history information. The

  As shown in FIG. 5, a stay counter is provided for each identifier ID (000 to FFF). This stay counter is decremented by 1 when a message is received and decremented by 1 when a message is transmitted. Therefore, the number of messages currently retained in the transmission queue buffer of the gateway unit is maintained in the retention counter. Therefore, when the stay counter exceeds a predetermined limit value, the stay flag indicating that the stay is excessive is changed to “1”. Further, when the stay counter becomes a negative value, the transfer abnormality flag indicating that transmission or more has occurred is changed to “1”.

  FIG. 6 is a diagram illustrating a state in which a message with a certain identifier remains. Similar to FIG. 4, the horizontal axis represents time, and the vertical axis represents the number of staying messages, that is, the staying counter value. In this example, a change in the count value of the stay counter indicating the number of stay messages is shown corresponding to reception and transmission of a message with a certain identifier. In other words, the stay counter is incremented by +1 if there is reception, and is decremented by -1 if there is transmission. Therefore, if the transmission frequency cannot catch up with the reception frequency, the count value is +2 to +7 as shown from reception 4 onward. Increase. Therefore, when the count value exceeds a predetermined reference value Vth (count value + 5), the transfer controller 12 sets the stay flag to “1”.

  Conversely, if the transmission frequency becomes higher than the reception frequency and the number of receptions exceeds the number of receptions, unnecessary messages are transmitted. In this case, although not shown, the stay counter takes a negative count value, and therefore the transfer controller 12 sets the transfer abnormality flag to “1” when the stay counter becomes a negative count value.

  In this way, the number of messages staying in the gateway unit is monitored, and if the reference value is exceeded, the retention flag is turned on. If it becomes negative, the transfer error flag is turned on, and only the flag information is recorded. To do. Then, by making it possible to read the stay flag and the transfer abnormality flag from the outside via the transfer controller 12, it is possible to detect whether a stay of the identifier message has occurred or whether a transfer error has occurred. Further, since the information to be stored is only the flag information for each identifier ID, the memory capacity of the transfer history area LOG can be kept small.

  In the example of FIG. 5, a stay counter is provided to monitor the number of staying messages. However, the staying number can be monitored, for example, by sequentially calculating the difference between the write pointer and the read pointer of the transmission queue buffer T-QUE having the FIFO configuration. That is, each time a message reception or transmission event occurs, the pointer is controlled for FIFO control. At that time, by calculating the pointer difference, the number of staying messages at that time can be obtained. Then, when the number exceeds a predetermined threshold value Vth, the stay flag may be turned on.

  It is desirable that the transfer controller TCON can be controlled to improve the message retention state in the gateway unit in accordance with the transfer history. For this purpose, the transfer controller TCON monitors, for example, an identifier whose stay flag is turned on, and controls to change the transmission cycle of the transmission queue buffer corresponding to the identifier. Alternatively, for a message group having an identifier for which the stay flag is turned on, control may be performed to forcibly discard a predetermined number of messages at the beginning.

  As described above, according to the present embodiment, since the message transfer history information in the gateway device 10 is stored, it can be used for failure diagnosis when any failure occurs. Can enhance the quality.

It is a figure which shows the structure of the gateway apparatus in this Embodiment, and the transfer operation | movement of a message. It is a figure which shows the detailed structure of the gateway unit in this Embodiment. It is a figure which shows an example of the transfer log | history area | region in this Embodiment. It is a figure which shows the mode of the retention of the message in the gateway unit obtained using the transfer log | history of FIG. It is a figure which shows the other example of the transfer log | history area | region in this Embodiment. It is a figure which shows the mode of the stay of the message of a certain identifier.

Explanation of symbols

ECU1-4: Device, BUS-1,2: CAN bus (network)
10: Gateway device (gateway unit)
12: Shared memory (message storage area MES, transfer history area LOG)
TCON: Transfer controller (transfer control means)
T-QUE: Transmission queue buffer MES1, 2: Message

Claims (3)

In a gateway device provided between a plurality of networks to which a plurality of devices are connected, respectively, for transferring messages between the networks,
Transfer control means for sequentially transmitting messages received from the source network to the destination network;
Transfer history storage means for storing transfer history information related to reception and transmission of the message,
A gateway apparatus characterized in that the transfer history information stored in the transfer history storage means can be read from the outside. In claim 1,
An identifier defining the priority of transmission rights in the network is added to the message, and the transfer control means transmits a message received at a transmission period corresponding to the identifier to the destination network, and transfers the transfer The gateway device, wherein the history storage means stores the transfer history information for each identifier, and the transfer history information corresponding to the identifier can be read out to the outside. In claim 1 or 2,
The transfer history information includes the reception time and transmission time of the message, the time from reception to transmission, the number of messages waiting for transmission after reception, the staying information indicating whether the number of messages waiting for transmission after reception exceeds a predetermined number, or after reception A gateway apparatus comprising at least one of transmission abnormality information indicating whether or not a plurality of transmissions have been made.

Images