JP2011103581A - Communication system and communication device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the failure of communication without reducing the efficiency of communication in a master-slave communication system which employs a schedule. <P>SOLUTION: If it is determined that none of data blocks to be transmitted has been updated from the previous time (YES in S130), control proceeds to the next slot without transmitting any data block. On the other hand, if it is determined that one or more of the data blocks to be transmitted have been updated from the previous time (NO in S130), a data block having a lowest update frequency and a data block or blocks having an update frequency higher than that of that data block, of the updated data block or blocks, and number-of-data-blocks information indicating the number of data blocks to be transmitted, are transmitted (S135). Thus, by transmitting data other than data which has not been updated, a period of time for transmission and reception of event data is increased, thereby reducing the probability that the communication failure occurs. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a communication system based on a master / slave system using a schedule and a communication apparatus belonging to the communication system.

  Various communication protocols for communicating with each other by a plurality of communication devices have been developed, and each has its own characteristics. One type of such communication protocol is a master-slave communication protocol using a schedule.

  This type of communication protocol will be described by taking LIN 2.1 (Non-Patent Document 1) as an example. In LIN2.1, a master node communicates with a slave node according to a schedule. This schedule is constituted by an array of a plurality of periods (time slots). The time slot includes a periodic slot that is a period for periodically transmitting and receiving periodic data, and an event data that is irregularly executed (that is, when an event occurs). It is classified as an event slot that is a period. By such a schedule, the master node manages communication with the slave node.

LIN 2.1 Specification (LIN Specification Package Revision 2.1), [online], [searched on August 28, 2009], Internet <URL: http://www.lin-subbus.org>

  The problem of the prior art described above is that if more events than the number of event slots in one cycle occur in one cycle, data that cannot be transmitted / received in that cycle is generated (communication leakage). In the first place, there is variation in how many events occur in one cycle. Therefore, in order to prevent such communication leakage, it is preferable to increase the number of event slots as much as possible.

  However, if the number of event slots is increased, there arises a problem that communication efficiency deteriorates. That is, the portion of the event slot that is less than the number of events that occurred during one cycle becomes a useless period in which nothing is transmitted / received. It will get worse.

  Such a problem does not occur only in LIN2.1, but may occur in a communication system using a master / slave system using a schedule as described in [Background Art]. Accordingly, an object of the present invention is to reduce communication leakage without deteriorating communication efficiency in a master / slave communication system using a schedule.

  In order to achieve this object, the invention according to claim 1 is configured so that the master node can communicate with a plurality of slave nodes, and the master node is a period for periodically transmitting and receiving periodic data. In a communication system that sequentially communicates with slave nodes one by one according to a schedule in which regular communication periods that are non-periodic and event communication periods that are irregularly executed event data are arranged It is a communication apparatus that functions as.

  And when transmitting regular data, in order to shorten the period required for the transmission, there is provided a master transmission means for transmitting only the updated part from the previous cycle. Then, event data is transmitted or received during the period generated by the shortening.

  According to this communication apparatus, communication leakage can be reduced without deteriorating communication efficiency. In other words, when transmitting regular data, this communication device transmits only the updated portion, thereby shortening the period required for transmission, and in the period generated by this shortening, transmission of event data or I try to receive. Therefore, the number of event data that can be actually transmitted / received can be increased according to the amount of the unupdated portion of the regular data, rather than the number of event data that can be transmitted / received in the prepared event communication period, and the above-described effect can be obtained.

  If there is no update part in the regular data, there is no transmission target, so transmission itself may be omitted. Similarly to the event communication period, the “period generated by shortening” is a period that can be used for transmission / reception of event data when the event data occurs. If there is no event data, transmission / reception is not performed. I will not.

  By the way, as a method for determining whether or not the regular data is updated, for example, a method of performing regular data in units of data blocks obtained by dividing the regular data into a plurality of data can be considered. That is, it is determined whether or not each data block is a transmission target. In this case, the slave node needs to grasp which of a plurality of data blocks has been transmitted. For this purpose, for example, a unique identifier may be attached to each data block. However, the data amount of each data block is undesirably large. Therefore, the following should be done.

  In the communication apparatus according to the second aspect, the regular data is divided into a plurality of data blocks. And a master transmission means transmits the transmission object information which shows which of a some data block is transmitted with the data block updated from one period ago.

According to this communication apparatus, without attaching an identifier or the like to each data block, the slave node can grasp which data block in each periodical data has been transmitted. For example, when the transmission target information indicates whether or not each data block is being transmitted, the same number of bits as the number of data blocks is required. Therefore, if the number of data blocks is 2 ⁿ (n: integer greater than or equal to 2), the data amount of the transmission target information is 2 ⁿ bits. Note that 2 ⁿ is used for easy comparison with an example described later.

On the other hand, when an identifier is attached to each data block, the total data amount of the identifier is n2 ⁿ bits. This is because the number of identifiers assigned to each data block requires the same number of bits as the number of data blocks (2 ⁿ bits), and the number of identifiers needs to be the same as the number of data blocks (n). Because.

Therefore, the total data amount of the identifier> the data amount of the transmission target information (n2 ⁿ > 2 ⁿ ), and the data amount can be reduced by using the transmission target information.
By the way, it is desirable that the data amount of the transmission target information is smaller. Therefore, the following should be done.

  In the communication device according to the third aspect, the data blocks are arranged based on the update frequency. Then, the master transmission means has updated the data block having the lowest update frequency (hereinafter referred to as “representative data block”) and the data block having the update frequency higher than that of the representative data block among the updated data blocks. In addition to transmitting as data blocks, information indicating the number of data blocks to be transmitted is transmitted as transmission target information.

According to this communication device, the data amount of the transmission target information can be reduced. If the number of data blocks is 2 ^n, the transmission target information in this claim only needs to be expressed according to the number of data blocks (for example, 4 for 4), so n bits (for example, 4). This is because 2 bits). Since 2 ⁿ > n, the communication apparatus according to the present invention can reduce the amount of data compared to the case where the data block is transmitted or not (2 ⁿ bits).

  Also, the data block received by the slave node is identified by << 1 >> representative data block and a data block having a higher update frequency than the representative data block being transmitted, and << 2 >> an arrangement of data blocks based on the update frequency , << 3 >> the number of data blocks to be transmitted. Since << 1 >> and << 2 >> can be shared in advance by the master node and the slave node, the master node transmits the transmission target information as << 3 >>, so that the slave node receives the received data block. Can be specified.

  In the communication device according to the third aspect, a data block having a higher update frequency than the representative data block is transmitted regardless of whether or not it has been updated. However, since the representative data block with a lower update frequency is actually updated, the data block with the higher update frequency than the representative data block is often updated, which may result in useless transmission. The nature is low.

By the way, even in the above communication apparatus, communication leakage is not completely eliminated. Therefore, considering the case where communication leakage occurs, the following is preferable.
According to a fourth aspect of the present invention, communication is performed according to a schedule in which periodic communication periods are arranged so as to be transmitted / received from periodic data having higher importance.

According to this communication apparatus, since communication with a low importance level is a target of communication leakage, it is possible to reduce the influence of communication leakage.
Furthermore, considering the case where communication leakage occurs, the following is preferable.

  The communication apparatus according to claim 5 changes the arrangement of the schedule in order to execute transmission / reception of event data with priority over transmission / reception of regular data. Then, when there is regular data that could not be transmitted one cycle before, the master transmission means transmits and receives the regular data as event data.

According to this communication apparatus, when a communication leak occurs, the leaked communication can be preferentially executed in the next cycle, so that the influence of the communication leak can be kept low.
By the way, in the middle of communication, a slave node may be added to the communication system or may leave due to a failure. For example, if a slave node that is the target of communication in a certain periodic communication period leaves the communication, the periodic communication period in the schedule becomes a useless period in which communication is not performed, so communication efficiency deteriorates. become. Further, when a slave node to be communicated is added with a regular communication period, it is considered that the slave node cannot participate in communication unless the schedule is changed. Therefore, the following should be done.

  A communication apparatus according to a sixth aspect includes a specifying unit and a changing unit. The specifying means inquires of each of the plurality of slave nodes whether or not communication in the regular communication period can be executed, and obtains a reply to the inquiry, thereby specifying a slave node that can execute communication in the regular communication period.

  If there is a difference between the slave node specified by the specifying means that the communication in the regular communication period is executable and the slave node to which the communication in the regular communication period is assigned, the changing unit eliminates the difference. Change the schedule.

  According to this communication apparatus, even if a slave node is added or removed during communication, the addition or withdrawal can be handled. Therefore, it is possible to prevent the communication efficiency from deteriorating.

  Note that if there is a slave node that has not been identified as being able to communicate during the regular communication period even though communication during the regular communication period has been assigned, the schedule change will be deleted from the schedule. If there is a slave node that is identified as being able to communicate in the regular communication period even though communication in the regular communication period is not assigned, the regular communication period may be added to the schedule. For example.

By the way, even in event data, there may be no update from the previous time. Transmission of such event data is also a useless period. Therefore, the following is recommended.
When transmitting the event data, the master transmission means included in the communication device according to claim 7 transmits only the updated portion from the previous time in order to shorten the period required for the transmission.

According to this communication apparatus, transmission of a part not updated for event data is also omitted, so that communication leakage can be further reduced.
By the way, as described so far, it is also conceivable that not a communication device as a master node but a slave node is provided with a configuration for reducing communication leakage without deteriorating communication efficiency.

  In the invention according to claim 8, the master node is configured to be able to communicate with a plurality of slave nodes, and the master node is provided with a regular communication period prepared for regular data that is periodically transmitted and received, and irregular The communication system sequentially communicates with the slave nodes one by one according to a schedule in which event communication periods prepared for event data transmitted and received are arranged.

  The slave node is provided with slave transmission means for transmitting only the updated portion from one cycle before in order to shorten the period required for the transmission when transmitting regular data as a reply to the master node. Further, the master node transmits or receives event data during the period generated by the shortening.

  According to this communication system, communication leakage can be reduced without deteriorating communication efficiency. This is because the slave node of this communication system includes slave transmission means similar to the master transmission means included in the communication device as the master node. Therefore, the same effect as the above communication device can be obtained.

The block block diagram of a communication system. Explanatory drawing of a schedule and a communication frame. The flowchart which shows a 1st master process. The figure which shows the change of a schedule. The flowchart which shows a slave process. The flowchart which shows a 2nd master process.

[Example 1]
[1. hardware]
Examples of the present invention will be described. FIG. 1 is a block diagram of an in-vehicle communication system 1 to which the present invention is applied. The in-vehicle communication system 1 is mounted on an automobile and includes a master ECU (Electronic Control Unit) 10 and a plurality of slave ECUs (Electronic Control Devices) 20, 20,. Each of the master ECU 10 and each slave ECU 20 is a device that controls a vehicle-mounted device (a door lock device, a power window device, a side mirror angle adjustment device, or the like) that is to be controlled. The master ECU 10 is configured to be able to communicate with the slave ECUs 20 one by one in order via the bus 99. This communication is performed according to a communication protocol improved from LIN2.1.

  The improvement of the communication protocol is a feature of the present invention. Although the specific contents will be described later, at least the “master / slave method”, “schedule”, “periodic slot”, “event” described in [Background Art] Since “slot” is the same as that in LIN 2.1, description thereof will be omitted.

  The master ECU 10 includes a microcomputer (hereinafter referred to as “microcomputer”) 11 and a transceiver 19 as illustrated. Further, the microcomputer 11 includes a control unit 12, a communication controller 13, and a transmission / reception buffer 14.

  The control unit 12 controls the control target described above and transmits / receives a signal for the control to / from the slave ECU 20 via the communication controller 13. The communication controller 13 performs communication according to the communication protocol by executing software (program) stored therein, and inputs / outputs the control signal to / from the control unit 12.

  The transmission / reception buffer 14 is connected between the communication controller 13 and the transceiver 19 and temporarily stores signals to be transmitted / received. The transceiver 19 mediates signals between the bus 99 and the microcomputer 11 (specifically, the transmission / reception buffer 14). Since the internal configuration of the slave ECU 20 is the same as that of the master ECU 10, illustration and description are omitted.

[2. Schedule and data structure]
FIG. 2A is a diagram for explaining “schedule A” that is a schedule used in the present embodiment and a communication frame transmitted by the master ECU 10 in the regular slot 1. First, schedule A will be described. As shown in FIG. 2A, the schedule A is divided into six time slots. Of these time slots, the first to fifth slots are regular slots, and the sixth slot is an event slot. In the following description, the first regular slot is called “periodic slot 1”, the second regular slot is called “periodic slot 2”, and so on. The sixth event slot arranged is called “event slot α”.

  In the schedule A, the regular slots for communication with higher importance are arranged on the left side of the figure so that they are transmitted earlier. For this reason, communication with higher importance is less likely to cause communication leakage.

  A communication instruction is assigned to each regular slot in advance. The communication instruction is information that defines communication contents to be executed by the master ECU 10. Specifically, a “data address” for identifying a data type to be transmitted / received, a “slave address” for identifying any one slave ECU 20 as a communication partner, and a read of data write It is information that defines what is.

  The data type indicates the type of data, such as information indicating door lock on / off, power window position information, side mirror angle information, and the like. Further, the data write is to transmit data to the slave ECU 20 and write the data. On the other hand, the data read is to make a data return request to the slave ECU 20 and acquire the data.

  For example, in the regular slot 1, as shown in FIG. As shown in FIG. 2 (a), the communication instruction of each regular slot indicates only whether it is a write or read, and the data address and the slave address are not related to the point of the invention. Illustration and description of what instructions are specifically assigned are omitted.

  Next, the communication frame will be described taking the regular slot 1 as an example. In FIG. 2A, the regular slot 1 is divided into “master” and “slave” because the regular slot 1 is a period for the master ECU 10 to transmit a communication frame, and the slave ECU 20 transmits a response. It is shown that the period is divided into 2A shows a part of the communication frame transmitted by the master ECU 10 below the schedule A. This part is data block number information and four data blocks (the actual communication frame includes other parts such as a header and a footer, but the description thereof is omitted).

  The data block is obtained by dividing the data to be transmitted into four, and each has a storage area of 2 bytes. The reason for such division is to use a data block as a unit for determining whether or not to update and whether or not to make a transmission target. That is, the in-vehicle communication system 1 is characterized in that only updated data is transmitted by first master processing and slave processing to be described later to reduce communication leakage, but “transmission of updated data only”. Since it is difficult to strictly execute the above, it is determined for each data block whether to update or not to be transmitted (described later in the first master process or the like).

  However, it does not necessarily mean that “only data blocks that have been updated are transmission targets, and data blocks that have not been updated are not transmission targets”. Specifically, data blocks with the lowest update frequency (hereinafter referred to as “representative data block”) among data blocks that have been updated and data blocks with a higher update frequency than the representative data block are transmission targets. . Therefore, a data block having a higher update frequency than the representative data block is transmitted even if it is not updated. The update frequency is stored in advance by the master ECU 10 and the slave ECU 20.

  The data block number information is information indicating the number of data blocks to be transmitted. Specifically, it is 2-bit information such as “00” if four data blocks are to be transmitted, “01” if three, and so on.

  Note that the arrangement of the data blocks is in the order of update frequency (the lower the update frequency, the more on the left side in the figure). Therefore, the representative data block and the right data block in the drawing with respect to the representative data block are transmission targets. That is, if the first data block is updated, all data blocks are to be transmitted, if the second data block is updated, the second to fourth data blocks are to be transmitted. That's it. Therefore, the slave ECU 20 can grasp which data block is transmitted based on the data block number information.

  If information indicating whether or not each data block is updated is used instead of the data block number information, only the updated data block can be transmitted, and wasteful data transmission can be reduced more reliably. . However, 4 bits are required to indicate whether or not the four data blocks are updated. On the other hand, 2 bits may be used to indicate the number of data blocks, and the storage area is saved, so data block number information is used.

  Thus, the data amount information is 66 bits for each data block number information and each data block. Although the above explanation has been given by taking the regular slot 1 as an example, this data structure is a regular slot or event slot, whether the master ECU 10 transmits in the write slot, Regardless of whether the slave ECU 20 transmits in the slot, this embodiment is common.

[3. software]
FIG. 3 is a flowchart showing the first master process. The first master process is a process in which the master ECU 10 is the main body (specifically, the microcomputer 11 is the main body, and more specifically, the communication controller 13 is the main body), and is for communication with the slave ECU 20. The first master process is a process that is repeatedly executed when the master ECU 10 is powered on.

  When the process is started, first, a schedule is set (S110). Here, the above-described schedule A is set. Note that the schedule set in S110 is changed in the subsequent processing. For example, when an event with a high communication request occurs, moving an event slot to immediately execute communication for the event is also performed in this embodiment as defined in LIN2.1. Furthermore, in the first master process, the length of the regular slot or event slot is shortened according to the update status of the data block. Therefore, the occurrence of an event and the update of data are divided into cases, and what kind of change is added to the schedule will be described together with each step of the process.

[3-1. When all data blocks in all scheduled slots are updated and no event has occurred (see FIG. 4A)]
First, a case where all data blocks in all regular slots are updated and no event occurs will be described. In this case, the communication is executed as set without changing the schedule.

  Next to S110, it is determined whether the cycle has ended (S115). Specifically, it is determined whether or not the time elapsed from the start of the schedule (time on the schedule) has reached a time corresponding to the schedule period set in S110. Since the cycle has not ended immediately after S110, if it is determined that the cycle has not ended (NO in S115), is there any unexecuted communication that should be executed in the regular slot or event slot? Is determined (S120). Since the communication instructions for the regular slots 1 to 5 are not executed, if it is determined that there is an unexecuted communication (YES in S120), it is determined whether the current slot is a write or a read (S125).

  As shown in FIG. 4A, since the regular slot 1 is a write, if it is determined to be a write (S125 write), each data block of the same regular slot one period before (in this case, the regular slot 1) is compared. Then, it is determined whether all data blocks to be transmitted are updated (S130). In the first period, there is no data of the previous period and comparison is not possible, but it is assumed that all data blocks are updated. Therefore, if it is determined that there is an update (NO in S130), a communication frame including all the data blocks assigned to the regular slot 1 and the data block number information (in this case, “00” indicating the number of four data blocks) is obtained. The write request is transmitted to the slave ECU 20 defined in the communication instruction of the regular slot 1 (S135). Then, a response (see S230 of the slave process in FIG. 5 described later) is received from the transmission destination slave ECU 20 (S140), and the process returns to S115.

  Turning now to FIG. 5, slave processing will be described. The slave process is a process executed mainly by the slave ECU 20, and is a process for returning the above-described response (information received by the master ECU 10 in the first master process). Then, it is triggered by receiving a write request or read request from the master ECU 10. In addition, this slave process omits transmission of data blocks that have not been updated, similar to the first master process.

  First, it is determined whether the received request is a write request or a read request (S210). If the write request is determined (S210 write), the received data block is stored (S220). This storage is performed after referring to the data block number information received together with the data blocks and specifying which data block is received (see FIG. 2). That is, if the number indicated in the data block number information is m (1 ≦ m ≦ 4), the first received data block is originally the (5-m) th data block and is received second. The data block originally is the (6-m) th data block, and so on.

  Then, a response indicating the completion of writing is returned to the master ECU 10 (S230), and the process is terminated. The response sent in this way is a response to S135 of the first master process.

  Returning to FIG. 3, the description of the first master process is resumed. When S140 in the regular slot 1 is completed, the communication instruction for the regular slot 2 is executed. The communication instruction for the regular slot 2 is also a light, and communication is performed in the same steps as for the regular slot 1, so that the description is omitted. The same applies to the regular slot 3 next to the regular slot 2.

  When S140 in the regular slot 3 ends, it is determined that the cycle has not ended (S115 NO), it is determined that there is an unexecuted communication (S120 YES), and the regular slot 4 is determined as shown in FIG. Since it is a lead, it is determined as a lead (S125 lead). Then, a read request is made to the slave ECU 20 (S145). Thereafter, a response to the request (see S250 / S260 of slave processing in FIG. 5 described later) is received (S140).

  Note that the data update determination (S130) is performed in the case of writing, but is not performed in the case of reading because the data is acquired from the slave. Such data determination is performed in the slave processing. (S240 described later).

  Returning to FIG. 5, the step of returning a response to the read request in the slave process will be described. If it is determined that the received request is a read request (S210 read), it is determined whether all requested data blocks are updated (S240). If it is determined that there is no update (S240 YES), a response indicating no update is returned to the master ECU 10 (S250), and the slave process is terminated. The response sent in this way is the response to S145 of the first master process described above.

  On the other hand, if it is determined that at least one data block is updated (NO in S240), the data block number information and the data blocks after the representative data block are returned to the master ECU 10 as a response, as in S135 of the first master process described above. (S260), and the slave processing ends. The response sent in this way is also a response to S145 of the first master process described above.

  Since the regular slot 5 is also a lead and is the same as the regular slot 4, the description is omitted. Then, when S140 in the regular slot 5 ends, the event slot α remains, so it is determined that the cycle has not ended (S115 NO). Since it is assumed in 3-1 that an event does not occur, it is determined that there is no communication to be executed in the event slot α and there is no unexecuted communication (NO in S120). If S120NO is determined, the process returns to S115, and S115NO and S120NO are repeated during the event slot α.

  If it is determined that the cycle has ended (S115 YES), it is determined whether there is a communication leak (S150). Since there was no communication leak in the first cycle, it is determined that there is no communication leak (S150 NO), the time on the schedule is initialized (S160), and the second cycle is executed from S115.

  The above description is based on the specific example of the first cycle, but the same contents are obtained regardless of the cycle as long as all data blocks are updated and no event occurs.

[3-2. When all data blocks in all slots are updated and one event occurs (see FIG. 4B)]
A case will be described in which all data blocks in all slots (periodic slots and event slots) are updated and one event occurs. The event is a lead and occurs during the regular slot 2 (see FIG. 4B).

  When an event occurs during the regular slot 2 as described above, the regular slots 1 and 2 are executed without being affected by the occurrence of the event as shown in FIG. After the end of step 2, the schedule is changed so that event slot α is moved immediately after the occurrence of an event with a high communication request, and the subsequent periodic slot is moved accordingly. Is done. Therefore, the description starts after the end of the regular slot 2.

  When S140 in the regular slot 2 ends, it is determined that the cycle has not ended (S115 NO), and it is determined that there is an unexecuted communication (S120 YES). Since the generated event is a read, the process proceeds to S125 read, and a read request is made according to the communication instruction of the event (S145). Then, the response to the read request is received and the event slot α ends.

  Thereafter, when periodic slots 3 to 5 are executed, it is determined that the cycle has ended (S115 YES), it is determined that there is no communication leak (S150 NO), the time on the schedule is initialized (S160), and the next cycle starts.

[3-3. When all data blocks in all slots are updated and two events occur (see FIGS. 4C and 4D)]
A case will be described where all data blocks in all slots are updated and two events occur. The first event is a lead and occurs during the regular slot 2. The second event is also a lead and occurs during the event slot α that has moved after the regular slot 2 (see FIG. 4C). The process until event slot α is executed is the same as that described in 3-2.

  When S140 in the event slot α is finished, the event slot β is inserted immediately after the event slot α. The event slot β is not defined in the schedule A, but is added when an event occurs. In this event slot β, a communication instruction for the event that has occurred second is executed. After that, when the communication instruction for the regular slots 3 and 4 is executed, it is determined that the period has ended although the communication instruction for the regular slot 5 has not been executed (S115 YES). Since the regular slot 5 has a communication leak, it is determined that there is a communication leak (S150 YES). Then, it is assumed that an event corresponding to the communication instruction of the slot that has lost communication (in this case, the regular slot 5) has occurred, and is stored in the transmission / reception buffer 14 (S155). Then, the time on the schedule is initialized (S160), and the next cycle starts.

  When a regular slot that has lost communication is set as an event as shown in FIG. 4D, the event is executed at the beginning of the next cycle. It will shift later.

[3-4. When some data blocks are not updated and two events occur (see FIG. 4E)]
A case will be described in which some data blocks are not updated and two events occur. In the specific example described here, all the data blocks in the regular slots 2 and 4 are not updated, and a part of the regular slot 3 (the first to third data blocks) is not updated. Other than that, there is an update. Assume that two events occur as in the case of FIG.

Since all data blocks in the regular slot 1 are updated and the same as described above, the explanation starts after the regular slot 1 ends.
When the regular slot 1 is completed and the process proceeds to S115, it is determined that the cycle has not ended (S115 NO), it is determined that there is an unexecuted communication (S120 YES), and the periodic slot 2 is write, so it is determined as write ( (S125 write), the process proceeds to S130. Since all the data blocks in the regular slot 2 are not updated, it is determined that there is no data update (S130 YES), and the process returns to S115. That is, the regular slot 2 is skipped.

  Then, when the process returns to S115, it is determined that the cycle is not over (S115 NO), it is determined that there is an unexecuted communication (S120 YES), and since the regular slot 3 is a write, it is determined as a write (S125 write), move on. Since the regular slot 3 has the fourth representative data block as described above, the data block number information storing “11” and the fourth data block (18 bits in total: see FIG. 2B) are transmitted. (S135). Then, a response is received (S140). As a result, as shown in FIG. 4E, the actual time length of the regular slot 3 is shorter than the time length defined in the schedule A.

  Then, when returning to S115, it is determined that the cycle is not over (S115 NO), and it is determined that there is an unexecuted communication (S120 YES). As described above, since a read event occurs during the second slot as in the case of 3-3, it is determined as a read (S125 read), and a read request is made in the event slot α. (S145). And the response from slave ECU20 is received (S140), and it returns to S115.

  When returning to S115, as described above, a read event occurs during the event slot α as in the case of 3-3. Therefore, similarly to the event slot α, S115 to S140 are executed, and the procedure returns to S115. .

  Next is the regular slot 4. When returning to S115, it is determined that the cycle is not over (NO in S115), and it is determined that there is an unexecuted communication (YES in S120). Since the regular slot 4 is a read, a read request is made (S145). Then, the response is received (S140), and the process returns to S115. At this time, as described above, since all the data blocks in the regular slot 4 are not updated, a response indicating no update is returned (see S260 in FIG. 5). This response has a data amount smaller than that of one data block. As a result, as shown in FIG. 4 (e), the actual length of the regular slot 4 is the length specified in the schedule A (the length when the transmission of all data blocks is executed without being omitted). ).

  Then, S115 to S140 are executed for the next regular slot 5, and when returning to S115, S115NO and S120NO are repeated until the cycle ends. If it is determined that the cycle is over (S115 YES), it is determined that there is no communication leak (S150 NO), the schedule is initialized (S160), and the next cycle is entered.

  Although the data block transmitted in the event slot is not illustrated when it is not updated, transmission of the data block without update is omitted as in the regular slot. In the case of an event slot, the comparison target is not the previous cycle, but the data block in the immediately preceding event slot that executed the same type of communication instruction is the comparison target.

[4. Effect of Example 1]
According to the in-vehicle communication system 1, communication leakage can be reduced without deteriorating communication efficiency. In 3-4, although there are more events than the number of event slots specified in the schedule, no communication leakage has occurred. This is because, because there is a data block that has not been updated, the period generated by omitting the transmission of the data block is allocated to transmission / reception of event data.

  In addition, when communication leakage occurs, as described in 3-3, the influence of communication leakage can be reduced by transmitting at the beginning of the next cycle. In addition, the data amount of newly added information is reduced by using the data block number information in order to specify the data block from which transmission is omitted.

[Example 2]
[5. Difference from Example 1]
The difference between the second embodiment and the first embodiment will be described. In the second embodiment, the second master process shown in FIG. 6 is executed instead of the first master process. In the first master process and the second master process, the steps from the start to the end of the schedule setting (S210 / S310) and the schedule cycle are the same (they are not shown because they are the same), but after the schedule cycle ends There is a difference. Therefore, the description starts after the schedule period ends.

  When determining that the cycle has ended (S315 YES), the master ECU 10 transmits a slave connection state inquiry signal (S320). The slave connection state inquiry signal is a connection state (which slave ECU 20 is communicably connected to the master ECU 10, and each slave is not in a sleep state (a state in which communication is suspended) but in an active state (communication). Is a signal to be broadcast in order to make an inquiry to each slave ECU 20. In addition, slave ECU20 which received the slave connection state inquiry signal returns the signal which shows that it is connected so that communication is possible to master ECU10. However, the slave ECU 20 newly connected to the in-vehicle communication system 1 also returns information indicating communication contents (data that can be transmitted, data that should be received, etc.) that can be executed by the slave ECU 20.

  Then, it is determined based on a reply from the slave ECU 20 whether there is a difference in connection state from the previous transmission (S330). If it is determined that there is no difference in the connection state (NO in S330), the time on the schedule is initialized and the process returns to S315.

  On the other hand, if it is determined that there is a difference in the connection state (YES in S330), information for resetting a schedule that matches the current connection state is created based on the reply from the slave ECU 20 (S350). Specifically, if there is a slave node that has not responded even though communication in the regular slot is assigned, the regular slot is deleted from the schedule, or communication in the regular slot is not assigned. Nevertheless, if there is a slave node that has returned a reply, if a regular slot for communication with the slave node is required, the regular slot is added to the schedule. Note that the period of the schedule is also changed in accordance with the addition or deletion of such regular slots. Then, returning to S310, the schedule is reset based on the information created in S350.

[6. Effect of Example 2]
As in the first embodiment, in addition to reducing communication leakage, in the second embodiment, even if the connection state of the slave ECU 20 changes during the communication, the communication is interrupted and no manual setting by the user is required. You can reset the schedule. Therefore, even if there is an addition by a user of a slave node or a detachment due to a failure, the addition or detachment can be handled without interruption of communication. Therefore, it is possible to prevent the communication efficiency from deteriorating.

[7. Correspondence with Claims]
The correspondence relationship between the claims and the embodiments will be described. S130 and S135 correspond to master transmission means, S250 or S260 corresponds to slave transmission means, S310 and S350 correspond to change means, and S320 corresponds to identification means software.

[8. Modified example]
A modification will be described. Instead of the data block number information, information indicating whether or not each data block has been updated may be transmitted. As described above, this information has a larger data amount than the data block number information, but can accurately indicate the updated portion. Therefore, it is effective when it is not reliable to estimate the updated data block based on the update frequency.

  Further, in the embodiment, when the data of the regular slot that has been leaked is transmitted / received in the next cycle, the data in the regular slot is transmitted in the event slot and the regular slot, so that it is transmitted twice in one cycle. Become. Therefore, a regular slot corresponding to the second transmission may be skipped so as not to be transmitted twice.

In the embodiment, the data blocks are arranged in ascending order of update frequency. However, the data blocks may be arranged in descending order, or other arrangements may be used.
In addition, in the schedule change according to the second embodiment, when the number of regular slots is reduced, the period need not be changed. In this case, the periodic slot that disappears may be replaced with an event slot.

  Further, the place where the event slot is inserted may be changed based on the comparison of the priority of the event slot and the regular slot. Specifically, this is performed as follows. First, a priority indicating a degree to be preferentially executed is predetermined for each event slot and each regular slot. When an event occurs, the priority of the event slot corresponding to the event is compared with the priority of the next regular slot. If the priority of the event slot is higher than the priority of the regular slot, the event slot is interrupted and executed as in the embodiment.

  On the other hand, if the priority of the regular slot is higher than the priority of the event slot, the regular slot is executed. When the execution of the event slot is postponed in this way, it is further compared with the priority of the next regular slot. By repeating such a procedure, the insertion location of the event slot may be determined.

Further, when there is data that cannot be communicated due to a communication error in the regular slot or event slot, the data may be communicated through the event slot.
Further, in S320 (see FIG. 6) of the second master process, in order to reduce the reduction in communication efficiency due to the slave connection state inquiry signal, transmission of the slave connection state inquiry signal targeting all slave ECUs 20 may be thinned out. Good. For example, a method in which S320 is executed only once in 10 cycles (a signal is transmitted to all slave ECUs 20) can be considered. Alternatively, in the first cycle, a slave connection state inquiry signal is transmitted only to one slave ECU 20, and in the second cycle, a slave connection state inquiry signal is transmitted only to another one slave ECU 20. In other words, a method of transmitting a signal to only one slave ECU 20 at a time can be considered.

  Further, the communication protocol to be based is not limited to LIN2.1, but may be the same type. In addition, the present invention may be applied to a communication system for other purposes even if it is not mounted on a car.

DESCRIPTION OF SYMBOLS 1 ... Vehicle-mounted communication system, 10 ... Master ECU, 11 ... Microcomputer, 12 ... Control part, 13 ... Communication controller, 14 ... Transmission / reception buffer, 19 ... Transceiver, 20 ... Slave ECU, 99 ... Bus

Claims (8)

The master node is configured to be able to communicate with multiple slave nodes,
The master node is arranged with a periodic communication period that is a period for periodically transmitting and receiving periodic data and an event communication period that is a period for transmitting and receiving event data that is irregularly performed. In a communication system that sequentially communicates with the slave nodes one by one according to a schedule, a communication device that functions as the master node,
When transmitting the periodic data, in order to reduce the period required for the transmission, it comprises a master transmission means for transmitting only the updated part from one cycle before,
The communication apparatus transmits or receives the event data in a period caused by the shortening. The periodic data is divided into a plurality of data blocks,
The said master transmission means transmits the transmission object information which shows which of these data blocks is transmitted with the said data block updated from one period ago. Communication equipment. The data blocks are arranged based on the update frequency,
The master transmission means transmits a data block having the lowest update frequency among the updated data blocks and a data block having a higher update frequency than the data block as the updated data block, and The communication apparatus according to claim 2, wherein information indicating the number of data blocks to be transmitted is transmitted as the transmission target information. The communication according to any one of claims 1 to 3, wherein communication is performed according to a schedule in which the periodic communication periods are arranged so that data is transmitted / received from a higher importance of the periodic data. Communication equipment. In order to execute the transmission / reception of the event data in preference to the transmission / reception of the regular data, the arrangement of the schedule is changed,
The said master transmission means transmits / receives the regular data as said event data, when there exists the said regular data which was not able to be transmitted in one period ago. The one of Claims 1-4 characterized by the above-mentioned. Communication device. Inquiring to each of the plurality of slave nodes whether or not communication in the regular communication period can be performed, and obtaining a reply to the query, specifying means for identifying a slave node capable of performing communication in the regular communication period; ,
If there is a difference between the slave node specified by the specifying means that the communication in the regular communication period can be executed and the slave node to which the communication in the regular communication period is assigned, the schedule is set so that the difference is eliminated. The communication apparatus according to claim 1, further comprising: a changing unit that changes The said master transmission means transmits only the update part from the last time, in order to shorten the period required for the transmission, when transmitting the event data. The communication apparatus of any one. The master node is configured to be able to communicate with multiple slave nodes,
The master node follows a schedule in which a regular communication period prepared for periodic data transmitted and received periodically and an event communication period prepared for event data transmitted and received irregularly are arranged, A communication system that sequentially communicates with slave nodes one by one,
The slave node includes slave transmission means for transmitting only the updated portion from one cycle before in order to shorten the period required for transmission when transmitting the periodic data as a reply to the master node. ,
The master node transmits or receives the event data in a period generated by the shortening.

Images