JP2008219389A - Communication control method, communication protocol, node, computer program, and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a communication control method capable of efficiently performing data communication even in the case of a high load despite a distributed communication network, a communication protocol, a node, a computer program and a storage medium with a program stored therein. <P>SOLUTION: The communication control method of the distributed communication system network NET including a plurality of nodes 1 to 6 has a case of performing communication by a hybrid mode that repeatingly forms a super frame SF including a reservation period SP in which reserved nodes SN which have respectively autonomously calculated a transmission order transmit data according to the transmission order among the nodes 1 to 6, and a contention period CP in which an unreserved node NN which has not calculated a transmission order transmits data, and respective unreserved nodes NN try to transmit data while allowing that data transmission by the unreserved nodes NN sometimes collides with each other when there are a plurality of unreserved nodes NN. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a communication control method, a communication protocol, a node used for communication, a computer program, and a storage medium storing the program in a distributed communication network.

As a method for controlling communication between nodes (communication terminals) in a communication network, for example, a control station is provided as in the PCF system defined in IEEE 802.11, and each node communicates in accordance with instructions from the control station. A centralized communication network. Thus, according to the instruction of the control station, each node performs communication in synchronization, thereby preventing a collision with data transmission by another node and allowing each node to use a transmission path (media).
However, with this PCF method, communication can be performed even when the communication network is heavily loaded (when the amount of communication is large) or on the other hand, even if the load is low (when the amount of communication is small), communication via the control station is always performed. In such a case, there is a problem that the use efficiency of the media decreases.

On the other hand, for example, a distributed communication network in which each node communicates asynchronously and directly without providing a specific control station has been proposed, as in the DCF system defined in IEEE 802.11. Since this distributed network does not require a control station, the communication network is simplified, and there is an advantage that communication can be efficiently performed when the communication network has a low load.
However, when the load on the communication network increases, the communication efficiency decreases. Transmission delay increases due to collision of data transmission by each node. It is difficult to cope with burst communication (when communication requests are intensively generated), and there are drawbacks such as extreme reduction in communication efficiency or inability to communicate.

On the other hand, an HCF (Hybrid Coordination Function) method combining a DCF method suitable for a low load and a PCF method capable of dealing with a high load has been proposed.
In addition, for example, a non-patent document 1 has been proposed as an improvement plan for the DCF method.

Y. Kwon, Y. Fang and H. Latchman, A novel MAC protocalwith fast collision resolution for wireless LANs, "IEEE INFOCOM'03, Apr. 2003.

However, the above-described HCF method requires a control station and is difficult to apply to a distributed communication network. In addition, since the control station communicates with each node and adjusts transmission / reception on the network, there is a problem that overhead increases and communication efficiency decreases.
In addition, the method described in Non-Patent Document 1 described above is based on the DCF method in which each node communicates asynchronously even when the network is heavily loaded. Communication efficiency is not improved sufficiently.

  The present invention has been made in view of such a problem, and in particular, a communication control capable of efficiently performing data communication even in the case of a high load while being a distributed communication network in which no device as a control station is arranged. It is an object to provide a method, a communication protocol, a node used for such communication, a computer program, and a storage medium storing the program.

The solution is a communication control method of a distributed communication network including a plurality of nodes that transmit and receive data to each other, wherein the one or more nodes that join the distributed communication network and wait for data transmission A reserved period in which each of the reserved nodes that autonomously calculate the transmission order transmits data according to the transmission order, and a non-reserved node that does not calculate the transmission order among the nodes transmits data When there are a plurality of non-reserved nodes, each non-reserved node transmits data by allowing the data transmitted by the non-reserved nodes to collide with each other. A communication control method including a case of performing communication in a hybrid mode in which a super frame including a competition period to be repeatedly formed is repeatedly formed.

The communication control method of the present invention includes a case of performing communication in a hybrid mode in which a super frame is repeatedly formed. In communication in this hybrid mode, a reservation period and a competition period are included in the superframe.
Of these, during the reservation period, each reserved node that autonomously calculates the transmission order performs transmission according to the transmission order. Therefore, for reserved nodes, data transmission opportunities are ensured during the reservation period, so that data transmission can be performed reliably. In addition, since each reserved node autonomously calculates the transmission order, there is no need to place a control station or the like in the communication network that gives instructions to each reserved node such as the transmission order. Can be maintained.
On the other hand, during the competition period, the non-reserved node transmits data. In this competition period, when there are a plurality of non-reserved nodes, the transmission of data by the non-reserved nodes is allowed to collide. Therefore, even if each non-reserved node attempts to transmit data, the data transmission is successful (refers to a case where transmission to the partner node is successful, for example, a case where an ACK is returned from the partner node) Not exclusively. However, there may be a case where data transmission by a non-reserved node is successful.

Thus, according to this communication control method, data transmission is attempted for the non-reserved node during the competition period in the superframe, so that data transmission can be performed as much as successful transmission. In addition, in the reservation period, data transmission can be reliably performed for the reserved node.
For this reason, even though it is a distributed communication network, even when the load is high, data communication can be performed relatively efficiently as a whole as compared with the DCF method or the like without causing an extreme decrease in communication efficiency or inability to communicate. Can be done.

  As a distributed communication network, nodes to which it is subscribed are limited in advance. In addition, a node such as a mobile terminal (mobile phone) is moved or a device or a communication function unit is turned on. Some of them may join. For example, an ad hoc network can be mentioned. Examples of the distributed communication network include a network based on wired communication, a network based on wireless communication using radio waves or light (visible light, infrared rays, ultraviolet rays, etc.), and a network using these in combination. Therefore, the transmission path (media) includes a communication line, an optical line, a power line, a space for transmitting radio waves and light, etc. used in these networks.

  Further, the nodes included in the distributed communication network may be any data processing device (communication terminal) that can perform data processing through communication with other nodes. Therefore, for example, various devices having a communication function that can join this distributed communication network, personal computers, mobile phones (portable terminals), PDAs, and various home appliances (refrigerators, air conditioners, water heaters, lighting fixtures, audio devices) , Interphones and other security devices), various office equipment (printers, FAX machines, etc.), and industrial equipment used in factories (robots, devices operated by automatic control, etc.).

  In addition, as a method for the reserved node to autonomously calculate its transmission order, for example, the number of reserved nodes that succeeded in data transmission is counted before the data transmission in the previous reservation period. Based on this, there is a method of calculating the transmission order of itself in the next reservation period. On the other hand, there is a method of counting the number of reserved nodes that have succeeded in data transmission after transmitting data themselves, and calculating their transmission order in the next reservation period based on this.

  Non-reserved nodes attempt to transmit data during the competition period, for example, if each non-reserved node transmits data and collides with data transmissions of other non-reserved nodes (transmission is not successful). In this case, there is a method of waiting for idle slots by the number of random values within a predetermined range selected by each other and attempting data transmission again.

  Furthermore, the communication control method of the present invention only needs to have a case of performing communication in the hybrid mode, and in addition to this, may have a case of performing communication in another mode.

  Furthermore, in the communication control method described above, any one of the nodes may be a superordinate node, and the supervised node may be a communication control method that transmits a signal that defines the superframe.

In the communication control method of the present invention, the supervising node sends a signal that defines a superframe. Thus, since the supervising node defines a super frame, it is possible to easily and autonomously form a super frame repeatedly without setting a control station.
In addition, other nodes participating in the distributed communication network may perform predetermined operations in the superframe defined by the supervising node depending on whether they are reserved nodes or non-reserved nodes. As a result, the operation in the hybrid mode can be performed autonomously.

Examples of the signal that defines the super frame include a head beacon that defines the start of the super frame or an end beacon that defines the end of the super frame.
When transmitting the head beacon, if the position of the reservation period is set immediately after the start of the super frame, the head beacon can be used together as a signal that defines the start period of the reservation period.

  Further, the supervising node only needs to transmit a signal that defines a superframe, for example, the above-described head beacon, and is not limited to which timing of transmitting its own data. Therefore, the supervising node may be a reserved node or a non-reserved node.

Furthermore, in the communication control method described above, it is preferable that the supervising node is a communication control method that is also the reserved node.
This is because if the supervising node behaves as a reserved node, it can first reliably transmit its own data during the reservation period.

  Furthermore, in any of the communication control methods described above, any one of the reserved nodes may be a communication control method that transmits a signal that defines the start of the reservation period.

  In the communication control method of the present invention, one of the reserved nodes sends a signal that defines the start of the reservation period. Therefore, other nodes (other reserved nodes and non-reserved nodes) can detect the start of the reservation period in this superframe by this signal. For this reason, the reserved node can transmit autonomously according to the transmission order. In addition, since the non-reserved node can stop the transmission attempt after the reservation period starts, it is possible to autonomously avoid a transmission collision between the reserved node and the non-reserved node. It is possible to reliably perform data transmission.

The reserved node that transmits a signal that defines the start of the reservation period may be any of the reserved nodes, but it is preferable that the reserved node having the first transmission order transmit this signal. .
If data is transmitted at a time close to the transmission of the signal that defines the start of the reservation period, the possibility of collision with transmission from another node, for example, a newly joined node, can be reduced, and its own data can be surely received. Can send. In addition, the position of the reserved node that transmits a signal that defines the start of the reservation period can be stably maintained.

  Further, in the communication control method according to any one of the above-described items, any one of the reserved nodes may be a communication control method that transmits a signal that defines the end of the reservation period.

  In the communication control method of the present invention, one of the reserved nodes sends a signal that defines the end of the reservation period. Therefore, other nodes (other reserved nodes and non-reserved nodes) can detect the end of the reservation period in this superframe by this signal. For this reason, the reserved node stops data transmission thereafter. Therefore, the non-reserved node may attempt to transmit data during the competition period from the end of the reservation period until the next reservation period starts without fear of a transmission collision with the reserved node. it can.

  In addition, when the super frame is composed of a reservation period and a competition period, a signal that defines the end of the reservation period can be used as a signal that defines the start of the competition period. For this reason, the start of the competition period can be clarified, and transmission attempts by non-reserved nodes can be reliably performed.

Note that the reserved node that transmits the signal that defines the end of the reservation period may be any of the reserved nodes, but the reserved node with the latest transmission order preferably transmits this signal.
The reserved node with the latest transmission order can transmit its own data reliably by transmitting a signal that defines the end of the reservation period following its transmission.

  Alternatively, the communication control method according to claim 1, wherein the superframe includes the reservation period starting from a start of the superframe and the competition period starting after the reservation period, and the reserved node Among them, the supervising node having the first transmission order may be a communication control method in which the superframe and a signal that defines the start of the reservation period are transmitted.

In the communication control method of the present invention, the superframe has a reservation period that starts with this start and a competition period after that. Further, the supervising node that transmits the reserved node first in the transmission order, and therefore transmits the first in the superframe, sends a signal that defines the start of the superframe and the reservation period.
Thus, since the supervising node sends a signal that defines the start of the superframe, it is possible to easily and autonomously form the superframe repeatedly without setting a control station.
In addition, other nodes participating in the distributed communication network may perform predetermined operations in the superframe defined by the supervising node depending on whether they are reserved nodes or non-reserved nodes. As a result, the operation in the hybrid mode can be performed autonomously.

  Moreover, the signal sent out by the supervising node is also a signal that defines the start of the reservation period. Therefore, other nodes (other reserved nodes and non-reserved nodes) can detect the start of the reservation period in this superframe by this signal. For this reason, each reserved node can autonomously perform transmission according to the transmission order. In addition, since the non-reserved node can stop the transmission attempt after the reservation period starts, it is possible to autonomously avoid a transmission collision between the reserved node and the non-reserved node. The data transmission by can be surely performed.

  In addition, the supervising node transmits a signal that defines the start of the superframe and the reservation period, and then transmits data to the first in the subsequent reservation period, and therefore to the first in the superframe. As a result, the possibility of collision with transmission from another node, for example, a newly joining node, can be reduced, and data transmission of the own node (supervising node) can be performed reliably. In addition, since the controlling node can be stably maintained, the hybrid mode can be stably maintained on the network.

  Further, in the communication control method according to any one of the above-described ones, a reserved non-reserved node that successfully transmits data during the competition period among the non-reserved nodes is a reserved node. A communication control method having a nodeization step is preferable.

  The communication control method of the present invention includes a reserved node conversion step in which a non-reserved successful node among non-reserved nodes is set as the reserved node. Therefore, the number of reserved nodes is gradually increased as non-reserved nodes succeed in transmission during the competition period. As a result, even for a node that has newly become a reserved node, data transmission can be reliably performed during the reservation period, so that collision of data transmission can be reduced when viewed as a whole network, and communication Efficiency and throughput can be improved.

  Further, in the communication control method described above, in the case where there are a plurality of non-reserved successful nodes, the reserved node conversion step is arranged in each of the non-reserved successful nodes in the reverse order of the successful transmission of the data. A communication control method that gives the order of transmission is preferable.

In the communication control method of the present invention, when there are a plurality of non-reservation successful nodes, the non-reservation success nodes are given the transmission order in the order opposite to the order of successful data transmission.
Specifically, for example, consider a case where three non-reserved nodes A, B, and C (non-reserved successful nodes) succeed in data transmission in the order of A, B, and C during a certain competition period. In this case, an order opposite to the order of successful transmission, that is, the order of transmission of C, B, A is given. For example, the fifth transmission order is given to C, the sixth transmission order to B, and the seventh transmission order to A.
By doing in this way, when the non-reserved node that succeeded in transmission (non-reserved successful node) obtains the transmission order, by counting the number of successful transmissions from the success of itself until the end of the competition period, This is because, among a plurality of non-reserved successful nodes, the ranking of itself can be easily known.

Furthermore, in the communication control method according to any one of the above-described ones, the reserved node transmits successfully from another reserved node that has occurred between the start of the reservation period and the transmission of its own data. In addition to this, a communication control method for counting idle slot and transmission collisions and determining the transmission timing of its own data based on the transmission order may be used.

In the communication control method of the present invention, the reserved node determines the data transmission timing based on the transmission order by counting not only successful transmissions by other reserved nodes but also idle slots and transmission collisions.
For this reason, for example, when there is no data to be transmitted, other reserved nodes may be left without causing any operation when their own transmission order is reached. If one idle slot is generated as a result of the neglect, the other reserved nodes can count the idle slot and appropriately determine the transmission order. That is, if idle slots are counted in this way, a reserved node that has no data to transmit can pause its transmission without affecting other reserved nodes.
In addition, since idle slots are counted, when the idle slot is detected, the reserved node in the next transmission order immediately transmits, so the unnecessary time generated due to the presence of the reserved node that suspended transmission is extremely short. Can be suppressed.
It also counts when transmissions of other nodes collide. In this case, transmission of another reserved node was performed, but for some reason, there is a high possibility that it collided with the transmission of another node, and there was a high possibility that transmission of another reserved node was performed. In that respect, it is appropriate to count in relation to the transmission order.

  Furthermore, in the communication control method according to any one of the above-described ones, the reserved node transmits successfully from another reserved node that has occurred between the start of the reservation period and the transmission of its own data. The communication control method may be the transmission order in the next reservation period.

In the communication control method of the present invention, the reserved node determines the number of successful transmissions by other reserved nodes that occurred between the start of the reservation period and the transmission of its own data as the transmission order in the next reservation period. To do. Thereby, each reserved node can autonomously and easily obtain the transmission order in the next reservation period. Also, in the reservation period, reserved nodes whose transmission collided for any reason and reserved nodes that did not transmit data (pause transmission) are excluded from the transmission order, that is, reserved nodes. The reserved nodes are narrowed down to the nodes that are expected to succeed in the next transmission, leaving only the reserved nodes that have been successfully transmitted, and can be transmitted in the next reservation period.
In addition, about the reserved node which transmission collided, after that, it is good to try transmission in a competition period as a non-reserved node.

  Furthermore, in the communication control method according to any one of the above, when one or a plurality of nodes waiting for transmission is a competition mode in which data is transmitted and there are a plurality of the nodes, the nodes It is preferable that the communication control method has a case in which each of the nodes performs communication in a competition mode in which data transmission by the above is allowed to collide, and each of the nodes tries to transmit data.

  In the communication control method of the present invention, in addition to the case of performing communication in the hybrid mode, the case of performing communication in the competitive mode having no portion corresponding to the reservation period in the hybrid mode. Therefore, the two modes can be properly used according to the load of the network.

  Communication in competitive mode is efficient when the network is lightly loaded, so communication is performed in competitive mode when the network is lightly loaded, and communication in hybrid mode is performed when the network load is high. It is preferable to migrate.

  Furthermore, in the communication control method according to any one of the above-described ones, when each of the nodes performs communication in the competition mode, the throughput of data transmission in any one of the nodes It is preferable to adopt a communication control method for shifting to the hybrid mode when the voltage drops.

In the communication control method of the present invention, when each node performs communication in the competition mode, when the throughput of data transmission is reduced in one of these nodes, the competition mode is switched to the hybrid mode. To do.
When the throughput of data transmission in a node is reduced, because many communications are performed by other nodes, the waiting time until data transmission becomes long, and a large amount of transmission data is retained. Even when data is transmitted from a node, there may be a situation in which data transmission cannot be performed properly (transmission is not successful) due to collision with data transmission of another node. Such a state is considered to occur when the load is increased in the entire network.
On the other hand, in the communication control method of the present invention, by shifting to the hybrid mode, it is possible to provide a reservation period during which the reserved node can reliably transmit data, so that the throughput of data transmission in the entire network is maintained. Can do.

  In addition, as a technique for determining a decrease in throughput, the number of transmission data arranged in the transmission queue in the node, and the case where collisions (transmission unsuccessful) occurred between the data transmission of other nodes occur continuously. For example, the number of consecutive times.

That is, in the communication control method described above, in any one of the above nodes, it occurs when the number of transmission data arranged in the transmission queue exceeds the first threshold and between data transmissions of other nodes. It is preferable to adopt a communication control method for determining that the throughput has decreased when at least one of the occurrences of continuous collisions exceeding the second threshold when the collisions occur. .
Thus, if the number of transmission data and the number of collisions increase in the transmission queue, it is considered that the load on the network is high.

  Further, in the communication control method described above, the transition from the competitive mode to the hybrid mode is performed, and any one of the nodes becomes a central node that transmits a signal that defines the superframe, and other nodes However, the communication control method is preferably performed by operating according to the superframe defined by the supervising node.

In the communication control method according to the present invention, the transition to the hybrid mode is performed by either of the nodes serving as a central node and transmitting a signal defining a superframe, and the other node transmitting a signal transmitted from the central node. Based on the superframe defined by the supervising node.
Thus, since each node autonomously behaves as a supervising node or a node following it, it is possible to appropriately shift to the hybrid mode without setting a control station.

It is preferable that the supervising node is a node that detects that it is preferable to shift to the hybrid mode. This is because the distributed communication network can be quickly shifted to the work hybrid mode.
Specifically, it is preferable that the node that has detected that the load of the distributed communication network has increased, particularly the node that has quickly detected this. More specifically, it is preferable that the node detects that the throughput of data transmission is reduced. Furthermore, when the number of transmission data queued in the transmission queue exceeds the first threshold value, and when a collision that occurs with data transmission of another node occurs continuously, the number of consecutive times is second. It is preferable that a node in which at least one of the threshold values exceeds occurs.
The signals defined by the superframe and the supervising node have been described above.

  Another solution is a communication protocol used for a distributed communication network including a plurality of nodes that transmit and receive data to each other, and joins the distributed communication network and waits for data transmission. Among the nodes, each reserved node that has autonomously calculated the transmission order has a reservation period during which data is transmitted according to the transmission order, and among the nodes, non-reserved nodes that have not calculated the transmission order have data When there is a plurality of non-reserved nodes in a competition period in which transmission is performed, each non-reserved node transmits data, allowing data transmission by the non-reserved nodes to collide with each other. A communication protocol having a case of performing communication in a hybrid mode in which a superframe including a competition period is repeatedly formed.

According to the communication protocol of the present invention, the case of performing communication in the hybrid mode in which the super frame is repeatedly formed is included. In the communication in the hybrid mode, the reservation period and the competition period are included in the super frame.
Of these, during the reservation period, each reserved node that autonomously calculates the transmission order performs transmission according to the transmission order. Therefore, for reserved nodes, data transmission opportunities are ensured during the reservation period, so that data transmission can be performed reliably. In addition, since each reserved node autonomously calculates the transmission order, it is necessary to arrange a control station or the like in the communication network that gives instructions such as the transmission order to each reserved node in order to maintain the communication network. There is no need to transmit / receive control signals to / from the control station.
On the other hand, during the competition period, the non-reserved node transmits data. In this competition period, when there are a plurality of non-reserved nodes, the transmission of data by the non-reserved nodes is allowed to collide. Therefore, even if each non-reserved node attempts to transmit data, the data transmission is successful (refers to a case where transmission to the partner node is successful, for example, a case where an ACK is returned from the partner node) Not exclusively. However, there may be a case where data transmission by a non-reserved node is successful.

Thus, according to this communication protocol, data transmission is attempted for the non-reserved node during the competition period in the superframe, so that data transmission can be performed as much as successful transmission. In addition, in the reservation period, data transmission can be reliably performed for the reserved node.
For this reason, although it is a distributed communication network, even when the load is high, data communication can be performed relatively efficiently as a whole without causing an extreme decrease in communication efficiency or inability to communicate.

  Still another solution is a distributed communication network including a node having a data transmission control unit that controls data transmission, and includes a plurality of nodes that transmit and receive data to and from each other. Among the one or more nodes waiting to transmit data, a reserved node that autonomously calculates the transmission order is a reservation period for transmitting data according to the transmission order, and among the nodes, The non-reserved node for which the transmission order is not calculated is a competition period in which data transmission is performed, and when there are a plurality of the non-reserved nodes, the transmission of data by the non-reserved nodes may collide with each other. And a hybrid mode that repeatedly forms a superframe including a competition period in which each of the non-reserved nodes attempts to transmit data. In a distributed communications network having a case of the signal, the control of the data transmission control unit is configured to become the node to function as any of the reserved node and the non-reserved node.

The node of the present invention is configured to function as both a reserved node and the non-reserved node under the control of the data transmission control unit. For this reason, in the above-mentioned distributed communication network, it becomes a node capable of appropriately transmitting data.
Note that the nodes included in the network have been described above.

  Further, when one or more nodes waiting for the transmission are in the competition mode in which data is transmitted under the control of the data transmission control unit and there are a plurality of the nodes, the nodes It is preferable that each of the nodes is configured to be able to function as a node that performs communication in a competition mode in which each of the above nodes attempts to transmit data by allowing the transmission of data according to each other to collide with each other. .

  The node of the present invention is configured to function as a node that performs communication in the competition mode. Therefore, the network can appropriately transmit data in the network including the competition mode.

  Still another solution is described in a computer-readable format so that communication control used in a distributed communication network including a plurality of nodes that transmit and receive data to and from each other is executed in the computer system of each of the above nodes. Among the one or more nodes waiting for data transmission, a reserved node, each of which is autonomously calculating a transmission order, joins the distributed communication network, and each of the reserved nodes A reservation period for transmission, and a non-reserved node that does not calculate the transmission order among the nodes is a competition period for transmitting data, and when there are a plurality of the non-reserved nodes, data by the non-reserved nodes Each of the above non-reserved nodes will attempt to transmit data, allowing the transmissions of And between, the case of performing communication by hybrid mode for repeatedly forming a superframe comprising providing, the node is a computer plug ram described as may be allowed to function as any of the above reserved node and the non-reserved node.

  According to the computer program of the present invention, by executing this program in the computer system of each node, each node functions as a reserved node and a non-reserved node depending on the situation, and communication in the hybrid mode is appropriately performed. Can be done.

  Furthermore, the above-described computer program may be a storage medium that is stored in a computer-readable manner.

Embodiments of the present invention will be described with reference to the drawings. Each of the nodes 1, 2, 3, 4, 5, 6 constituting the distributed communication network (hereinafter simply referred to as a network) NET having no control station shown in FIG. The communication terminal operates in accordance with the communication protocol. All of these nodes 1 to 6 communicate with each other by wireless communication in which space is a medium (transmission path) MD.
However, some or all of the nodes 1 to 6 can also communicate by wire using a communication line or the like.

  Although not shown, these nodes 1 to 6 constitute a known computer system including a CPU, a ROM, a RAM, and the like. Further, the computer system stores an executable computer program in a medium such as a ROM or a RAM. The stored computer program includes a computer program related to communication processing according to the present invention. Therefore, when the power of each node is turned on, the stored computer program is loaded into the CPU, and each of the nodes 1 to 6 becomes a communication terminal having the configuration shown in FIG. Each of the nodes 1 to 6 includes a data processing unit 18 that processes various types of data, a data transmission / reception control unit 11 that controls transmission / reception of data with other nodes, and an antenna 19. The data transmission / reception control unit 11 includes a transmission data holding unit 16 that holds transmission data requested to be transmitted from the data processing unit 11, a transmission control unit 14 that controls transmission, and an instruction from the transmission control unit 14 to transmit transmission data. A transmission unit 12 is provided that transmits the signal toward the antenna 19 and transmits the signal from the antenna to another node. In addition, a receiving unit 13 that receives data received from the antenna 19 and a reception control unit 15 that controls the received data and delivers the data to the data processing unit 18 are provided.

  In each node 1, etc., as shown in the upper part of FIG. 12A, after transmitting transmission data (packet), after waiting for SIFS time, an ACK returned from the partner node Is successfully received (successful transmission ST: Successful Transmission). Accordingly, the transmission control unit 14 and the reception control unit 15 operate by exchanging information with each other.

  Now, the operation of such nodes 1 to 6 will be described with reference to the flowcharts of FIGS. First, it is assumed that a node having a transmission request (for example, the node 5) is turned on, or a node 5 newly joined in the network NET due to the movement of the node 5.

In the present embodiment, communication control is performed in two modes of the competition mode and the hybrid mode in the network NET (each node 1 to 6).
Of these, the competition mode is a mode in which the node 1 that waits for transmission performs data transmission, as will be described in detail later. However, when there are a plurality of nodes that wait for transmission, the transmission of data performed by each node 1 or the like is performed. This is a mode in which each node 1 etc. tries to transmit data by allowing the nodes 1 and 2 to collide (thus, transmission may not be successful). In the competition mode, there is no super frame in the hybrid mode described below.

On the other hand, in this embodiment, for example, as shown in FIGS. 12A to 12C, the hybrid mode includes a reservation period SP and a subsequent competition period CP within the time defined by the superframe SF. It has two periods.
Of these, the reservation period SP is determined by the reserved nodes (for example, nodes 1, 2, 3, and 4 in FIG. 12A) that have autonomously calculated the transmission order, respectively. ) Is a period in which data is transmitted according to the transmission order of the nodes 1, 4, 2, and 3).
In the present embodiment, each reserved node performs control so that transmission is performed at a shorter interval according to the transmission order (with no idle slot IS between transmissions).
In contrast, the competition period is determined by the non-reserved node when there are a plurality of non-reserved nodes (for example, nodes 5 and 6 in FIG. 12A) in the competition mode. This is a period in which each non-reserved node attempts to transmit data, allowing data transmissions to collide.

  At the stage when a certain node newly joins the network NET, it is unclear to this node whether the network NET is currently controlled for communication in the competitive mode or the hybrid mode.

Therefore, the newly joined node first checks in step S1 in FIG. 3 whether the media MD is idle, that is, whether communication by another node is not being performed. Here, if Yes, that is, if the media MD is idle, the process moves to step S6 to perform data transmission.
On the other hand, if No in step S1, that is, if the medium is busy, data transmission is not performed and the process proceeds to step S2. Since other nodes are communicating, in order to avoid transmission collision, it waits for a while until several idle slots IS appear.

Therefore, in step S2, the back-off counter BF is set to an initial value. Specifically, a random number having an appropriate range (for example, 1 to 32) is generated, and the generated random value (for example, 17) is set as an initial value of the back-off counter BF.
In step S3, it is determined whether the medium is idle. When the idle slot IS is detected (Yes), the process proceeds to step S4. On the other hand, when the media MD is busy, the process returns to step S3 again. That is, step S3 is repeated until an idle slot IS is detected.
In step S4, the value of the back-off counter BF is decremented by 1 every time step S4 is passed, and thus every time the idle slot IS is detected.

Subsequently, in step S5, it is determined whether the back-off counter BF has become 0 or the media MD is busy. Here, when it is neither (No), it returns to step S3.
On the other hand, if the back-off counter BF = 0, the process proceeds to step S6 to transmit data.
On the other hand, if the media MD is busy, the process waits until the media MD becomes idle in step S11, and proceeds to step S12 to determine whether the network NET is in the hybrid mode. Specifically, it is determined from step S1 to step S12 whether a signal characteristic of the hybrid mode, that is, the head beacon HB and the end beacon EB (see FIG. 12) is detected.

If it is No in step S12, that is, if it is not yet detected that the vehicle is in the hybrid mode, the process returns to step S3.
On the other hand, in the case of Yes, the process proceeds to step S61 (see FIG. 8), and thereafter, the operation of this node (for example, node 5) is switched to the hybrid mode.

  On the other hand, when data is transmitted in step S6, it is determined in step S7 whether or not the transmission is successful. Specifically, it is determined whether or not an ACK from the counterpart node corresponding to the transmission has been received (see FIG. 12A). If Yes, that is, if the transmission is successful, the process proceeds to the next transmission process (step S8). Specifically, the collision number counter CC described later is cleared (step S9), and it is further determined whether a head flag described later is set (step S10). If the head flag is not set (No), the process returns to step S1. Thereby, the transmission of the next transmission data is tried again according to the processing after step S1.

  If YES in step S10, that is, if the head flag is set, the process proceeds to step S16 (see FIG. 4). Thereafter, this node itself (for example, the node 5) tries to be the head node HN. As will be described below, since the head flag is set when the load on the network NET increases, the node that detects this becomes the head node HN, so that the network NET can be quickly shifted to the hybrid mode. Because it can.

  If YES in step S7, that is, if the transmission is not successful, it is considered that a collision with the transmission of another node has occurred. Therefore, in step S13, the collision number counter CC is incremented by one. The initial value of the collision number counter CC is 0, and as described above, it is cleared (CC = 0) when transmission is successful (see step S9). In step S13, the back-off counter BF is set to an initial value. Note that the use of a random number within a predetermined range as the initial value is the same as in step S2 described above. By using a random number as an initial value instead of a predetermined number, the possibility that the transmission collides again with the node where the previous transmission collided is reduced.

In step S14, the collision counter CC is greater than the first threshold value TH1, or the number of transmission data (transmission queue number) DN in the transmission queue in the transmission data holding unit 16 is second. It is determined whether or not the threshold value TH2 is greater. Here, in the case of No, the process returns to step S3 (the values of the back-off counter BF and the collision number counter CC are maintained). On the other hand, if Yes in step S14, the head flag is set in step S15, and the process returns to step S3.
When data transmission is difficult due to a collision (when CC becomes a large value) and when transmission data waiting for transmission increases (when DN becomes a large value), transmission requests by each node increase. This is probably because the load on the network NET is increasing, and for this reason, collisions and transmission stagnation are increasing. Therefore, a head flag is set in step S15 so that the node that detects this becomes a head node HN in order to try in step S16 (see FIG. 4).

Next, step S16 and subsequent steps will be described with reference to FIG. In step S16, after successful data transmission in step S6 (ACK reception from the partner node), a temporary end beacon is transmitted from this node after waiting for SIFS time. By transmitting this temporary end beacon, it is declared that it will become the head node HN.
Thereafter, in step S17, the value of the slot counter SC is set to CPIS (the number of idle slots IS included in the competition period CP) (SC = CPIS). As this CPIS value, a fixed value such as 16, 32 may be used. However, as described below, the head node HN designates the CPIS value according to the network situation, and the operation is performed. May be changed. Further, every time the idle slot IS is detected, the value of the slot counter SC is decremented by 1 (SC = SC-1). This is repeated until SC = 0. That is, it waits until the number of detected idle slots IS becomes CPIS.

Next, in step S18, it is determined whether or not a head beacon HB by another node has been detected after the provisional end beacon is transmitted. If YES, that is, if the head beacon HB by another node is detected, the process proceeds to step S60, and the operation is switched to the operation in the hybrid mode according to the received head beacon HB by the other node. That is, it abandons the attempt to become the head node HN, becomes itself a non-reserved node, and performs the operations after step S61 (see FIG. 8). This is because other nodes have declared that they have become head nodes HN before themselves.
On the other hand, if the head beacon HB has not been detected in step S18, itself becomes the head node HN. Thereby, the network NET shifts to the hybrid mode. Specifically, the process proceeds to step S21 (see FIG. 5), and NNSP (number of reserved nodes included in the reservation period SP), NMSP (number of temporary reservation nodes included in the temporary reservation period), CPIS (idle slot included in the competition period CP) A head beacon HB including a value of (number of IS) is created and transmitted. This is because the head node HN has been declared ahead of other nodes.

  The operation of the head node HN will be described with reference to FIG. In step S21, after sending the head beacon HB, it is investigated whether or not the media MD is cleared only for the collision check time CCS (see FIG. 12) corresponding to one slot. That is, it is detected whether other nodes are communicating. At almost the same time, the risk of declaring that two or more nodes become head nodes HN is taken into consideration, and the movement of other nodes is detected. Here, No, that is, immediately after the head beacon HB is transmitted, if it is determined that another node is outputting any signal (for example, a head beacon), the process proceeds to step S25, and a random number of 0 or 1 RND is generated. In step S26, it is determined whether the random number RND is 0 (RND = 0?). If the random number RND = 0 (Yes), the process immediately returns to step S21, and the head beacon HB is transmitted again.

  On the other hand, if No in step S26, that is, if RND = 1, one slot is waited (step S27). Thereafter, in step S28, the detection of the head beacon HB by another node is confirmed. Here, if Yes, that is, if the head beacon HB by another node can be confirmed, the attempt to become the head node HN is abandoned, and the node itself becomes a non-reserved node, and step S61 (see FIG. 8). Perform the following operations. This is because, prior to itself, another node declares again that it has become the head node HN. In this way, even when two or more nodes compete and declare to be the head node HN, the head beacon is retransmitted autonomously by using the random number RND. It is possible to autonomously determine which of the two nodes becomes the head node HN by the collision of the head beacons HB several times.

  If the head beacon HB by another node cannot be confirmed in step S28, the process returns to step S21 and the head beacon HB is transmitted again. After retransmitting the head beacon HB, it is checked again in step S22 whether or not the media MD is clear only for the collision check time CCS corresponding to one slot.

If the media MD is clear, the process proceeds to step S23. As a result, the device itself becomes the only head node HN in the network. In step S23, it is confirmed whether there is untransmitted data in the head node HN. If No, that is, if the transmission data is exhausted, the process proceeds to step S52, and the transmission is stopped. That is, the head node HN abandons the position as the head node and the reserved node at the end of the role of sending the head beacon HB (step S21).
In this case, the other reserved node SN, specifically, the reserved node having the second transmission order autonomously takes over the position of the head node HN. It will be described later.
As described above, in this embodiment, even when there is no transmission data in the head node HN (transmission request is lost), the head node HN simply transmits itself without performing any notification or other procedures to other nodes. It is possible to withdraw from a group of nodes having a transmission request simply by stopping. For this reason, the procedure at the time of withdrawal can be performed very easily.

  On the other hand, if YES in step S23, that is, if there is untransmitted data, in step S24, the transmission data is transmitted to the first of the reservation period SP.

In the present embodiment, as shown in FIG. 12, the head beacon HB transmitted from the head node HN defines a super frame SF including a reservation period SP and a contention period CP. Specifically, the signal defines the start of the super frame SF. Further, the head beacon HB also functions as a signal that defines the start period SP1 of the reservation period SP. The head node HN is also one of the reserved nodes SN.
In the present embodiment, the head node HN always transmits data first in the reservation period SP. If data transmission is performed at a time close to the transmission of the head beacon HB that defines the start period SP1 of the reservation period SP, the possibility of collision with transmission from another node, for example, a newly joined node can be reduced. This is because it can transmit its own data. In addition, since the position of the head node HN that transmits the head beacon HB that defines the superframe SF and the start period SP1 of the reservation period SP can be stably maintained, the hybrid mode can be stably maintained on the network NET. Because.

In addition, since the head node (supervising node) HN sends out a head beacon HB that defines the start SP1 of the superframe SF, the head node HN makes it easy even if no control station is set in the network NET. In addition, the super frame SF can be repeatedly formed autonomously.
In addition, in the superframe SF defined by the head node HN, the other nodes joining the network NET depend on whether they are reserved nodes SN or non-reserved nodes NN, as will be described later. Since a predetermined operation is required, the operation in the hybrid mode can be performed autonomously.

  Moreover, the head beacon HB transmitted from the head node HN is also a signal that defines the start SP1 of the reservation period SP. Therefore, other nodes (other reserved node SN and non-reserved node NN) can detect the start of the reservation period SP in the superframe SF by this head beacon HB. For this reason, each reserved node SN can autonomously transmit according to the transmission order. Further, since the non-reserved node NN can stop the transmission attempt after the start of the reservation period SP, it is possible to autonomously avoid a transmission collision between the reserved node SN and the non-reserved node NN. The data transmission by the reserved node SN can be surely performed.

  Next, the process proceeds to step S29 to determine whether NNSP is 2 or more. Here, if No, that is, NNSP = 1, there is no node other than the head node HN in the reservation period SP, so the process proceeds to step S36 (see FIG. 6), and the end beacon EB is detected. to decide. However, in this case, since there is no node that transmits the end beacon EB other than the head node HN, it is determined No, and in step S37, the head node HN transmits the end beacon EB, and the process proceeds to step S38. move on.

  On the other hand, if it is determined in step S29 that NNSP ≧ 2 (Yes), the reserved node SN other than the head node HN should transmit data within the reservation period SP. Therefore, in step S30, the reserved node counter NC is set to NNSP (NC = NNSP). Further, in step S31, the process waits until a data transmission success, collision, or idle slot IS is detected by other reserved nodes (for example, nodes 2, 3, and 4 in FIG. 12A). If these are detected (Yes), the reserved node counter NC is decremented by 1 in step S32 (NC = NC-1). Further, in step S33, it is determined whether the detected communication is a successful communication (ST), that is, a successful data transmission. If No, that is, if it is a collision or an idle slot IS, the process skips step S34 and proceeds to step S35. On the other hand, if the communication is successful (Yes), the number of successful communication nodes CNNSP is counted up (CNNSP = CNNSP + 1) in step S34. Note that the initial value of the number of successful communication nodes CNNSP is CNNSP = 1.

  In step S35, it is determined whether or not the reserved node counter NC has become 1 (NC = 1). Here, if No, that is, NC> 1, the process returns to step S31. On the other hand, if NC = 1 (Yes), the process proceeds to step S36 (see FIG. 6). Thus, in the reservation period SP, the number of reserved nodes SN successfully communicated can be obtained as CNNSP.

  Next, in step S36, the detection of the end beacon EB is determined. As will be described later, in this embodiment, among the reserved nodes, the node that performs the latest data transmission is supposed to send the end beacon EB (for example, node 3 in FIG. 12A). Therefore, the head node HN determines whether another reserved node SN has transmitted the end beacon EB. If the end beacon EB cannot be detected (No), there is a possibility that some failure has occurred in the reservation node to which the end beacon EB should be sent. Therefore, in step S35, the head node HN sends an end beacon EB instead. As a result, the reservation period SP ends and the competition period CP starts. That is, in the present embodiment, the end beacon EB serves as a signal that defines the end SP2 of the reservation period SP and also serves as a signal that defines the start of the competition period CP.

In the present embodiment, non-reserved nodes (non-reserved successful nodes, for example, nodes 5 and 6 in FIG. 12A) that have made a non-reserved node attempt transmission and perform successful communication ST in the competition period CP In the superframe SF (see FIG. 12B), it is treated as a temporary reservation node HSN, and prior to other non-reserved nodes, data is transmitted to this temporary reservation node HSN in the initial temporary reservation period HSP of the competition period CP. Send a transmission.
Further, when the temporary reservation node HSN (for example, the nodes 5 and 6) performs the successful communication ST in the temporary reservation period HSP, this is replaced with the reservation period in the next superframe SF (see FIG. 12C). It is added to a member of a reserved node that performs transmission at the SP.

Therefore, in step S38, first, the temporary reservation node counter HC is set to NMSP (HC = NMSP). A method for obtaining the value of NMSP will be described later.
Further, in step S39, the successful communication ST, collision, or idle slot IS of the temporary reservation node HSN (for example, nodes 5 and 6 in FIG. 12B) is detected at the beginning of the competition period CP (provisional reservation period HSP). Wait until you do. If these are detected (Yes), the temporary reservation node counter HC is decremented by 1 (HC = NH-1) in step S41. Further, in step S42, it is determined whether the detected communication is a successful communication (ST), that is, a successful data transmission. If No, that is, if it is a collision or an idle slot IS, step S43 is skipped and the process proceeds to step S44. On the other hand, if the communication is successful (Yes), in step S43, the number of successful communication temporary reserved nodes CNMSP is counted up (CNMSP = CNMSP + 1). Note that the initial value of the number of successful communication temporary reservation nodes CNMSP is CNMSP = 0.

  In step S44, it is determined whether or not the temporary reservation node counter HC has become 0 (HC = 0). If No, that is, HC ≧ 1, the process returns to step S39. on the other hand. If HC = 0 (Yes), the process proceeds to step S45. Thus, in the temporary reservation period HSP, the number of temporary reserved nodes HSN successfully communicated can be obtained as CNMSP.

  Further, in step S45, the NNSP transmitted by the next head beacon HB is updated. Specifically, NNSP = CNNNSP + CNMSP. That is, the sum of the number of successful communication nodes CNNSP and the number of successful communication temporary reservation nodes CNMSP is used as a new NNSP (number of nodes newly entering during the reservation period). This means that transmission can be performed as the reserved node SN in the reservation period SP in the next superframe SF in the reserved node (successful communication ST in the reservation period in the previous superframe SF ( For example, only the temporary reservation nodes (for example, nodes 5 and 6 in FIG. 12B) that have performed successful communication ST in the previous temporary reservation period HSP in the nodes 1, 2, 3) in FIG. 12B. It is shown that.

Next, the operation after step S46 of the head node HN, specifically, the operation of the head node HN after the temporary reservation node HSN finishes transmission and the temporary reservation period HSP ends in the competition period CP will be described. .
In step S46, the slot counter SC is set to CPIS (SC = CPIS). Next, it waits until successful communication ST by the non-reserved node or idle slot IS is detected (in the case of No). If the idle slot IS is detected, the process proceeds to step S48, and the slot counter SC is decremented by one. On the other hand, when the successful communication ST by the non-reserved node is detected, the process proceeds to step 49 to increment the value of NMSP (NMSP = NMSP + 1). Note that the initial value of NMSP is 0. In any case, it is then determined in step S51 whether SC = 0 or not. If the slot counter SC is 1 or more (No), the process returns to step S47.
On the other hand, when the slot counter SC becomes 0 (Yes), the process returns to step S21. That is, the head beacon HB is transmitted.
Thus, the above operation is repeated until untransmitted data is exhausted in the head node HN (see step S23).

Next, the operation of the node (non-reserved node NN) after step S61, that is, when another node becomes the head node will be described with reference to FIGS.
As described above, the non-reserved node contends for transmission with other non-reserved nodes in the contention period CP whose start time is defined by the end beacon. As described above, the contention period CP is a variable-length variable that is terminated by sending the head beacon HB when the number of idle slots IS detected by the slot counter SC of the head node HN reaches CPIS (for example, 32). It is a period.
Further, as described above, among the non-reserved nodes, the non-reserved successful nodes NSN (for example, the nodes 5 and 6 in FIG. 12A) that have performed the successful communication ST in the competition period CP are the next superframe. In SF (for example, FIG. 12B), it is treated as a temporary reservation node HN. Furthermore, if the successful communication ST is also performed here, in the next superframe (for example, FIG. 12C), the reserved node SN is given a position capable of transmitting data during the reservation period SP. After that, if there is transmission data and the successful communication ST is performed every time, the transmission can be continued as the reserved node SN.
Note that if there is a transmission collision with the reserved node SN and the tentative reserved node HN for any reason, the node returns to the non-reserved node thereafter.

  Therefore, first, in step S61 (see FIG. 8), the detection of the end beacon EB, that is, the start of the competition period CP is awaited (No). When the end beacon EB is detected, in step S62, the slot counter SC is set to the sum of the CPIS transmitted by the head beacon HB (see step S21) and the allowable value α held in advance (SC = CPIS + α). This allowable value α is a value smaller than CPIS, such as 3 or 5.

  In step S63, each time the idle slot IS is detected, the back-off counter BF and the slot counter SC are decremented by one. Note that the value of the back-off counter BF is set in steps S2 and S13, and is decremented in step S4 (see FIG. 3) or set in step S67 described later.

  Further, in step S63, it is determined whether or not the back-off counter BF has become 0 (BF = 0?). Here, if Yes (BF = 0), the process proceeds to step S65 to perform data transmission. Subsequently, in step S66, it is confirmed whether or not the transmission is successful (ACK reception). When the transmission is successful, that is, when the successful communication ST is performed (Yes), the non-reserved node NN This is a successful node NSN (corresponding to nodes 5 and 6 in FIG. 12A). Therefore, the operation as the temporary reservation node HN after step S72 is performed.

  On the other hand, if the transmission is not successful in step S66, that is, if there is a transmission collision with another non-reserved node, the backoff counter BF is set to the initial value again in step S67, and step S61 is performed. Return to wait for the next competition period. Note that the use of a random value as the initial value of the back-off counter BF is the same as in steps S2 and S13. This is because the random number is used to autonomously shift the retransmission timing with other non-reserved nodes that have caused a collision, thereby reducing the possibility of a collision.

  On the other hand, if it is determined No (BF ≧ 1) in step S64, the process proceeds to step S68. If the slot counter SC is 1 or more (No), it is further determined in step S70 whether or not the head beacon HB has been detected. If the head beacon HB is detected, the process returns to step S61. This is because the competition period CP has ended and transmission in the next competition period CP is attempted. The process returns with the value of the back-off counter BF being maintained.

On the other hand, if it is No in step S70, that is, if the head beacon HB has not been detected yet, the process returns to step S63. Thereby, until the back-off counter BF becomes 0, the slot counter SC becomes 0, or the head beacon HB is detected, the idle slot IS is detected, and the back-off counter BF and the slot counter SC are detected. The decrement continues.
If it is determined in step S68 that the slot counter SC = 0 (Yes: if the head beacon HB cannot be detected while detecting (CPIS + α) idle slots IS in the competition period CP). Returning to the communication in the competition mode (step S3). This is because the head beacon HN cannot be detected even at the scheduled time, and it is considered that it is difficult to maintain the hybrid mode for some reason.

Next, the case where this non-reserved node NN has performed successful communication ST (after step S71) will be described.
First, in step S71, the temporary order counter OD is set to an initial value 0 (OD = 0). Next, in step S72, the other non-reserved node NN checks whether or not the communication is successful, and every time a successful communication is detected, the temporary order counter OD is incremented in step S73 (OD = OD + 1). This is repeated until the head beacon HB is detected, that is, until the competition period CP ends. As a result, the number of non-reserved nodes NN (non-reserved successful nodes) that have performed successful communication ST after their own successful communication ST can be obtained as the value of the temporary order OD.

  Therefore, in step S75, the value of the temporary order counter OD is set as the temporary order value ODJ. Thereafter, in step S76, the detection of the end beacon EB, that is, the end of the reservation period SP is awaited.

Next, in step S77 (see FIG. 9), the latest NNSP value included in the end beacon EB (the number of reserved nodes that have achieved successful communication ST during the reservation period SP) is recorded.
In step S78, the back-off counter BF is set to the temporary order value ODJ (BF = ODJ). Next, in step S79, it is checked whether BF = 0. If No (BF ≧ 1), the process waits until successful communication, collision, and idle slot IS of another temporary reservation node are detected in step S80. If these are detected, in step S81, the back-off counter BF is counted down by 1 (BF = BF-1), and the process proceeds to step S82. Here, it is checked whether the detected communication is a successful communication ST. If the communication is successful communication ST (Yes), the temporary success counter HD is incremented in step 83, and the process returns to step S79. Note that the initial value of the temporary success counter is set to HD = 0. On the other hand, if it is not successful communication ST (when collision or idle slot IS is detected, No), step S83 is skipped and the process returns to step S79. This is repeated until BF = 0.

  In step S79, when BF = 0 (Yes), the process proceeds to step S84 and data transmission is performed. Thereby, each temporary reservation node HN can transmit in order according to the temporary order value ODJ obtained autonomously in the temporary reservation period HSP.

  Further, in step S85, it is confirmed whether or not the transmission is successful. If the transmission is successful (Yes), the process proceeds to step S86, and the order value ODS is set. Specifically, ODS = NNSP + HD. That is, the sum of the number of reserved nodes NNSP that has achieved successful communication ST during the reservation period SP and the number HD of temporary reservation nodes HN that have performed successful communication earlier than itself in the temporary reservation period HSP is the order value ODS. Suppose. Thus, it was possible to autonomously obtain the order value ODS indicating the transmission order in the next reservation period SP for the temporary reservation node HN. Thereafter, as shown in step S87, communication is performed as the reserved node SN in the reservation mode, that is, in the reservation period SP.

  On the other hand, if the transmission is not successful in step S85 (a transmission collision occurs), as shown in step S88, an attempt to enter the reserved portion, that is, the reserved node SN becomes the reservation period SP. The abandonment of the attempt to transmit data to is performed, the backoff counter BF is reset to the initial value using a random number, and the process returns to step S61. In other words, the temporary reservation node HN shifts to the non-reservation node NN.

Next, the operation (step S90 and subsequent steps) when the node is a reserved node SN that is not the head node HN will be described with reference to FIGS.
First, in step S90, the detection of the head beacon HB, that is, the start of the super frame SF and the reservation period SP is awaited. When the head beacon HB is detected (Yes), the reservation period starts, and in step S91, the values of NNSP, NMSP, and CPIS included in the head beacon HB are acquired. Further, in step S92, it is confirmed whether or not the order value ODS indicating the transmission order of this node is NNSP + NMSP. In short, it confirms whether it is the last node of the reserved node SN to be transmitted during the reservation period. In the case of Yes, in step S93, the tail flag is set and the process returns. In step S94, it is checked whether or not the order value ODS is 2. As described above, the second node next to the head node HN in the transmission order is the position to take over the head node instead when there is no data to be transmitted to the head node HN and withdraws. This is to confirm.

  Therefore, if it is not ODS = 2 (No), the process proceeds to step S97. If Yes (ODS = 2), in step S95, the transmission part of the head node HN, that is, immediately after the collision check time CCS, It is checked whether or not an idle slot IS is detected. If the idle slot IS is detected at this timing, the current head node HN is considered to leave (step S52). Therefore, the head flag is set in step S96 so that the head node HN becomes a new head node HN. Proceed to step S97. On the other hand, when the idle slot IS is not detected (when the head node is transmitting), the process proceeds to step S97.

  In step S97, the reservation order counter OC is set to the order value ODS (OC = ODS). Also, the order value ODS is reset. Further, it waits until the successful communication ST, collision, and idle slot IS of the other reserved nodes SN are detected. If these are detected (Yes), the reservation order counter OC is counted down by 1 in step S99 ( OC = OC-1).

Further, it is checked whether or not the successful communication ST is detected in step S100. If the successful communication ST is (Yes), the sequence value ODS is incremented by 1 in step S101. Note that the initial value of the order value ODS is set to 0 (ODS = 0). On the other hand, if the detected successful communication ST is not, the process returns to step S98. Further, in step S102 (see FIG. 11), it is checked whether or not the reservation order counter OC has become 0. If OC ≧ 1 (No), the process returns to step S98.
If these steps are repeated until the reservation order counter OC = 0 (Yes in step S102), data transmission is performed in step S103.

Further, in step S104, it is checked whether or not the data transmission is successful (successful transmission ST). If the data transmission is successful (Yes), the presence or absence of the tail flag is confirmed in step S105 (see step S93). When this reserved node SN is the last node to transmit in the reservation period SP, the tail flag is set (Yes), so in step S106, the end beacon EB is transmitted.
The end beacon EB includes the NNSP value (see step S77). As the value of this NNSP, the order value ODS of this last reserved node SN is used (NNSP = ODS).

  On the other hand, if the tail flag is not set (No in step S105), step S106 is skipped and the process proceeds to step S108. In step S108, it is checked whether or not the head flag is set (see step S96). If the head flag is not set (No), the process proceeds to step S109, where it is confirmed whether there is untransmitted data in the reserved node SN (specifically, the transmission data holding unit 16), If it still exists, the process proceeds to step S110, where it is confirmed again whether the tail flag is set. If the tail flag is not set, that is, if this reserved node SN is not the last node to transmit in the reservation period SP (No), the process returns to step S90. Thereby, this reserved node SN can repeatedly perform data transmission in the order of transmission according to the order value ODS in the reservation period SP of the superframe SF as the reserved node SN that is not the head node HN.

  On the other hand, when this reserved node SN is the last node to transmit in the reservation period SP (Yes), the process proceeds to step S111. Here, the slot counter SC is set to CPIS + α. Next, in step S112, the presence or absence of detection of the head beacon HB is checked. If this is detected (Yes), the process returns to step 91. If not detected (No), it is checked in step S113 whether or not the idle slot IS is detected. If the idle slot IS is not detected, the process returns to step S112. On the other hand, when the idle slot IS is detected, the slot counter SC is decremented by 1 and the process proceeds to step S114. Here, it is checked whether or not SC = 0. If SC ≧ 1 (No), the process returns to step S112. If SC = 0, the process proceeds to step S115, and the head beacon HB is transmitted even though this node is not a head node.

  This is done for the following reason. After the transmission of the end beacon EB, the competition period CP is started, and when the CPIS number of idle slots IS is counted, the head beacon HB should be transmitted from the head node HN. However, even if (CPIS + α) idle slots IS are counted, if the head beacon HB cannot be detected, there is a possibility that some trouble that the head beacon HB cannot be transmitted has occurred in the head node HN. Therefore, the last node (end node) that performs transmission in the reservation period SP transmits the head beacon HB only once (step S116). As a result, each reserved node SN excluding the head node HN can again perform transmission according to the transmission order.

  On the other hand, if No in step S109, that is, if there is no data to be transmitted to the reserved node SN, the transmission is stopped in step S110, and the network NET is withdrawn from the node waiting for transmission. When the transmission is stopped, in the next reservation period, the node that has stopped the transmission enters a state in which the transmission is not performed when it is time to transmit.

  However, as described in step S98, each reserved node SN counts not only successful communication STs of other reserved nodes but also transmission collisions and idle slots IS in counting its transmission order. ing. For this reason, since there is a reserved node whose transmission has been stopped, the other reserved node SN detects one idle slot IS when the period during which no transmission is performed is one slot. As a result, each reserved node SN can autonomously know the existence of the reserved node whose transmission has been stopped without receiving any notification. In addition, in steps S100 and S101, the number of reserved nodes SN that have achieved successful communication ST before itself is counted, and the transmission order (order value ODS) in the next reservation period is known. In the reserved node SN, the node that has stopped such transmission can be excluded, and the problem that the reservation period SP is unnecessarily extended does not occur.

  In addition, since this idle slot IS is also counted in the order of transmission, the reserved node having this transmission order immediately starts transmitting to the reserved node whose transmission has been stopped. That is, transmission is not performed for a short period of time compared to data transmission corresponding to one idle slot IS, and thereafter, the reserved node performs transmission again in order.

  As described above, according to the method of the present embodiment, when the data to be transmitted is exhausted, the processing to be performed by the reserved node is very simple (simply stops transmission). On the other hand, other reserved nodes autonomously determine such withdrawal and appropriately adjust the transmission order in the reservation period SP. For this reason, the communication in the hybrid mode, specifically, the communication during the reservation period can be continued without being affected by such withdrawal.

In addition, if it is determined in step S108 that the head flag is set (Yes), as described above, this node becomes a new head node, so that it will operate as the head node HN from the next time. Therefore, it progresses to step S21.
Again, when the transmission data is exhausted in the head node HN (No in step S23), the head node HN simply stops data transmission. As a result, the other reserved nodes SN know the withdrawal of the head node HN (first reserved node), and similarly appropriately adjust the transmission order in the reservation period SP. Also, the second reserved node SN becomes a new head node HN in place of the withdrawn head node HN. Therefore, also in this case, communication in the reservation period and further communication in the hybrid mode including the competition period can be continued without being affected by the withdrawal of the head node HN.

  In step S104, if No, that is, if data transmission from the reserved node SN is unsuccessful (for example, if a transmission collision occurs for some reason), as shown in step S107, The node is moved from the reserved node SN position to the non-reserved node NN position, excluded from transmission in the reservation period SP, and thereafter transmitted in the competition period CP.

  In addition, if there is no reserved node (the only remaining head node stops transmission), the head beacon HB that defines the end beacon EB and the superframe SF is not transmitted, so the network NET naturally Return from hybrid mode to competitive mode (transition).

  As described above, the operation in the competition mode and the operation in the hybrid mode (operation as a non-reserved node, temporary reserved node, reserved node) for each node waiting for transmission in the network NET using FIGS. explained.

  As described above, in the communication control method according to the present embodiment, each node operates in the competition mode, and when the load on the network increases, each node autonomously shifts to the hybrid mode. In this hybrid mode, in addition to the competition period CP in which each non-reserved node NN tries to transmit in the superframe SF, a reserved period SP in which the reserved node SN performs data transmission is provided, which increases the load on the network NET. Even at least the reserved node SN, data transmission can be reliably performed.

If the network NET, and therefore each node is operating in the hybrid mode, provisional reservations are made sequentially from the non-reserved node NN (non-reserved successful node NSN) that has performed successful communication ST in the competition period CP. It becomes the node HN and further the reserved node SN, and communication is performed during the reservation period.
For this reason, when there is no node newly joining the network NET and no collision or the like occurs in transmission of the reserved node SN (when successful communication can be continued), as shown in FIG. 13, n belonging to the network NET Any of the two nodes is configured to perform transmission (saturated state) according to the transmission order in the reservation period SP as the reserved node SN.
In this case, in addition to the reservation period SP, there is a difference in that a competition period CP consisting of CPIS idle slots IS is added, but the communication of each node is controlled by the control station, The communication is almost the same. Also from this point, according to the communication control method of the present embodiment, the communication control method can perform communication while maintaining high communication efficiency even in the case of a high load even though the distributed communication network NET does not include a control station. It turns out that it is.

In order to confirm this point, assuming that there are eight nodes in the network NET, the relationship between the requested communication volume (Mbps) generated in each node and the total throughput (Mbps) in the entire network NET is simulated. investigated. The result is shown in FIG. What is described as SRAP is based on the communication control method of this embodiment.
For comparison, a similar simulation was performed when control was performed using the DCF method defined in IEEE 802.11. In addition, the case where the data rate was 11 Mbps and the frame length was 1000 bytes and 500 bytes were investigated.
Although it differs depending on the frame length, when this embodiment (SRAP) and IEEE802.11 (DCF method) are compared for the same frame length, the present embodiment (SRAP) has the required traffic ( (Mbps), the total throughput can be increased accordingly. For example, in the case of a frame length of 1000 bytes, the DCF system has a maximum total throughput of about 3.2 Mbps. In this embodiment (SRAP), it can be seen that the speed is increased to about 5 Mbps. Similarly, in the case of a frame length of 500 bytes, it can be seen that the maximum value of the total throughput can be increased in the present embodiment.

Similarly, the case where there are 24 nodes in the network NET was also examined. The result is shown in FIG.
When this embodiment (SRAP) and IEEE802.11 (DCF method) are compared with each other for the same frame length, the requested communication amount (Mbps) of each node is greater in this embodiment (SRAP). ) Increases, the total throughput can be increased correspondingly. For example, in the case of a frame length of 1000 bytes, the DCF method has a maximum total throughput of about 2.8 Mbps, In this embodiment (SRAP), it can be seen that the speed is increased to about 4.1 Mbps. Similarly, in the case of a frame length of 500 bytes, it can be seen that the maximum value of the total throughput can be increased in the present embodiment.

In the above, the present invention has been described with reference to the embodiments. However, the present invention is not limited to the above embodiments, and it is needless to say that the present invention can be appropriately modified and applied without departing from the gist thereof.
For example, in the present embodiment, the nodes 1 to 6 perform wireless communication, but some or all of them may perform wired communication. However, each node is preferably applied to wireless communication using radio waves, light, or the like.

It is explanatory drawing which shows the mode of the distributed communication network containing a node (communication terminal). It is explanatory drawing which shows the structure of each node. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the node in competition mode among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the node in the process in which a certain node becomes a head node among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the head node in a reservation period among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the head node in a competition period among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the head node in a competition period among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the non-reserved node in hybrid mode among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the non-reservation success node in a hybrid mode among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the reserved node in a hybrid mode among these. It is a flowchart which shows operation | movement of the node which joined the distributed communication network, and mainly shows operation | movement of the reserved node in a hybrid mode among these. It is explanatory drawing which shows the operation example of the reserved node and the non-reserved node in hybrid mode. In hybrid mode, it is explanatory drawing which shows the operation example in the state (saturation state) in which all the nodes which have joined the distributed communication network became a reserved node. Simulation that evaluates the relationship between the traffic generated by each node and the throughput of the entire network, assuming a DCF type distributed communication network specified in this case and IEEE 802.11, in which 8 nodes are subscribed. It is a result. Simulation that evaluates the relationship between the traffic generated by each node and the throughput of the entire network, assuming a DCF distributed communication network specified in this case and IEEE 802.11, to which 24 nodes are subscribed It is a result.

Explanation of symbols

NET Distributed communication network MD Media 1, 2, 3, 4, 5, 6 Node 11 Data transmission / reception control unit 12 Transmission unit 13 Reception unit 14 Transmission control unit 15 Reception control unit 16 Transmission data holding unit 18 Data processing unit 19 Antenna SN Reserved node NN Non-reserved node NSN Non-reserved successful node HSN Temporary reserved node SP Reserved period SP1 Reserved period start SP2 Reserved period end CP Competition period HB Head beacon EB End beacon CCS Collision check slot IS Idle slot BF Backoff counter CC Number of collision counter DN Number of transmission queues (number of transmission data queued in transmission queue)
SC Slot counter NC Reserved node counter NNSP Number of nodes included in reservation period NMSP Number of nodes newly entering in reservation period CPIS Number of idle slots included in competition period CP CNNNSP Successful communication node number CNMSP Successful communication temporary reservation node number RND Random number HC Temporary reserved node counter OD Temporary order counter ODJ Temporary order value ODS Order value OC Reserved order counter HD Temporary success counter

Claims (17)

A communication control method for a distributed communication network including a plurality of nodes that transmit and receive data to and from each other,
A reservation period in which a reserved node that autonomously calculates a transmission order among the one or a plurality of nodes waiting for data transmission by joining the distributed communication network transmits data according to the transmission order; ,
Among the above nodes, a non-reserved node that has not calculated the transmission order is a competition period in which data is transmitted,
When there are a plurality of non-reserved nodes, the transmission of data by the non-reserved nodes is allowed to collide, and each non-reserved node tries to transmit data. A communication control method including a case of performing communication in a hybrid mode in which frames are repeatedly formed. The communication control method according to claim 1,
Any of the nodes is a supervisory node,
A communication control method in which the supervising node transmits a signal that defines the superframe. The communication control method according to claim 1 or 2,
A communication control method in which one of the reserved nodes transmits a signal that defines the start of the reservation period. The communication control method according to any one of claims 1 to 3,
A communication control method in which one of the reserved nodes transmits a signal that defines the end of the reservation period. The communication control method according to claim 1,
The superframe has the reservation period starting from the beginning of the superframe and the competition period starting after the reservation period;
Among the reserved nodes, the supervising node having the first transmission order is
A communication control method for transmitting a signal defining the start of the super frame and the reservation period. It is the communication control method of any one of Claims 1-4, Comprising:
A communication control method comprising a reserved node conversion step in which, among the non-reserved nodes, a non-reserved successful node that has successfully transmitted data during the competition period is the reserved node. The communication control method according to claim 6, comprising:
The reserved node conversion step includes:
A communication control method in which when there are a plurality of non-reserved successful nodes, each non-reserved successful node is assigned the transmission order in an order opposite to the order of successful data transmission. The communication control method according to any one of claims 1 to 7,
The reserved node counts idle slot and transmission collisions in addition to successful transmissions by other reserved nodes that occurred from the start of the reservation period to the transmission of its own data. The communication control method which determines the transmission timing of own data based on. The communication control method according to any one of claims 1 to 8,
The reserved node is a communication control method in which the number of successful transmissions by other reserved nodes occurring from the start of the reservation period to the transmission of its own data is set as the transmission order in the next reservation period. The communication control method according to any one of claims 1 to 9,
One or more nodes waiting for the transmission are in a competition mode for transmitting data,
When there are a plurality of nodes, a communication control method including a case where each node performs communication in a competition mode while allowing data transmission by the nodes to collide with each other and attempting to transmit data. . The communication control method according to claim 10, comprising:
In the case where each of the nodes performs communication in the competitive mode,
When the throughput of data transmission in any one of the above nodes decreases,
A communication control method for shifting to the hybrid mode. The communication control method according to claim 10 or 11,
Transition from the competitive mode to the hybrid mode,
One of the nodes becomes a central node that sends a signal that defines the superframe,
A communication control method performed by another node operating in accordance with a superframe defined by the supervising node. A communication protocol used in a distributed communication network including a plurality of nodes that transmit and receive data to and from each other,
A reservation period in which a reserved node that autonomously calculates a transmission order among the one or a plurality of nodes waiting for data transmission by joining the distributed communication network transmits data according to the transmission order; ,
Among the above nodes, a non-reserved node that has not calculated the transmission order is a competition period in which data is transmitted,
When there are a plurality of non-reserved nodes, the transmission of data by the non-reserved nodes is allowed to collide, and each non-reserved node tries to transmit data. A communication protocol having a case of performing communication in a hybrid mode in which frames are repeatedly formed. A node having a data transmission control unit for controlling transmission of data,
A distributed communication network including a plurality of nodes that transmit and receive data to and from each other,
A reservation period in which a reserved node that autonomously calculates a transmission order among the one or a plurality of nodes waiting for data transmission by joining the distributed communication network transmits data according to the transmission order; ,
Among the above nodes, a non-reserved node that has not calculated the transmission order is a competition period in which data is transmitted,
When there are a plurality of non-reserved nodes, the transmission of data by the non-reserved nodes is allowed to collide, and each non-reserved node tries to transmit data. In a distributed communication network having a case of performing communication in a hybrid mode in which frames are repeatedly formed,
A node configured to function as both the reserved node and the non-reserved node under the control of the data transmission control unit. 15. A node according to claim 14, comprising:
Under the control of the data transmission control unit,
One or more nodes waiting for the transmission are in a competition mode for transmitting data,
When there are multiple nodes, it is possible to allow data transmissions by the nodes to collide with each other, and each node can also function as a node that performs communication in a competition mode that attempts to transmit data. A node configured as follows. A computer program written in a computer-readable format so as to execute communication control used in a distributed communication network including a plurality of nodes that transmit and receive data to and from each other in the computer system of each of the nodes,
A reservation period in which a reserved node that autonomously calculates a transmission order among the one or a plurality of nodes waiting for data transmission by joining the distributed communication network transmits data according to the transmission order; ,
Among the above nodes, a non-reserved node that has not calculated the transmission order is a competition period in which data is transmitted,
When there are a plurality of non-reserved nodes, the transmission of data by the non-reserved nodes is allowed to collide, and each non-reserved node tries to transmit data. A case where communication in a hybrid mode in which a frame is repeatedly formed is provided,
A computer program in which the node is described so that it can function as both the reserved node and the non-reserved node.   A storage medium storing the computer program according to claim 16 in a computer-readable manner.

Images