JP5601254B2 - Sensor data transmission method and sensor node - Google Patents

Description

  The present invention relates to a method for transmitting sensor data from a sensor node in a sensor network and a sensor node.

  A typical ad hoc network as a sensor network has the characteristics of "no fixed infrastructure required", "dynamic network", and "multi-hop capable", and is composed of a core relay (GW: Gate Way) and multiple nodes. It is configured. The node is, for example, a small device including a wireless communication function, a sensor function, a power supply unit, and a computer. This node autonomously forms a network while communicating with adjacent nodes.

  Sensors at each node in the network constantly measure data, and when the value exceeds a certain threshold (this is referred to as “detection”), a detection message (hereinafter referred to as “detection message”). To the server. The detection message includes a detection notification informing that the node has detected and a detection time indicating the time when the node has detected. This detection message hops to the adjacent node and propagates to the server. The detection process is performed at regular intervals, and the data is recorded. The detection message is notified every time a predetermined time has elapsed using the detection timer.

The following conventional techniques are known as a mechanism for notifying a server of a detection message.
First, a conventional technique for reducing the amount of traffic without affecting the packet arrival rate is known. The receiving unit receives a signal notified from the transmission-source wireless device and including a packet signal indicating the position of the wireless device. The relay processing unit executes the relay processing after a delay time set in proportion to the distance detected by the distance detecting unit has elapsed after receiving the signal notified from the wireless device by the receiving unit. When the same packet signal as the packet signal received by the receiving unit is received before the delay time elapses, the relay processing unit stops the relay processing after the delay time elapses (for example, patent document). 1).

  In addition, there is known a conventional technique for reducing network traffic and shortening a waiting time for transmitting a message in a wireless ad hoc network. A message that arrives at a receiving node of the network is received, and a schedule time for resending the message is determined. At this time, the schedule time for retransmission is set to be proportional to the signal strength at which the message is received by the receiving node. When the same message is received from a different node in the network before the scheduled time for retransmission, the retransmission of the message is canceled (for example, Patent Document 2).

  Further, a conventional technique for providing a frame processing device capable of reducing the load on a processor related to processing of an unnecessary message is known. A message received for the first time is used as a reference message for comparison, and a message match comparison unit that performs a match comparison with a reference message for messages that have been received and classified thereafter, and that message is received for a matched message. A redundant message determination unit that determines whether or not the message has been received, and the redundant message determination unit is configured to receive the message if the received elapsed time from the time when the reference message is received is within the range of the determination reference value. Is discarded, and the reference message is cleared when the elapsed reception time exceeds the determination reference value (for example, Patent Document 3).

JP 2008-47984 A JP 2007-13961 A JP 2007-43382 A

  However, with the above prior art method alone, it is difficult to completely avoid the concentration of packets to the communication node connected to the gateway when a detection message is notified, resulting in data delay and packet loss. It had the problem that.

  Further, each sensor node collects only data when a certain threshold is exceeded (at the time of detection) as measurement data, and has a problem that measurement data cannot be collected finely.

  Accordingly, an object of the present invention is to avoid the concentration of packets to communication nodes connected to a gateway.

  In an example of the aspect, at a node, sensor data is measured by a sensor in the node and sequentially recorded. When the node detects that the value of the measured sensor data satisfies a predetermined condition, When the detection message is not received, the detection time is held, and the detection message including the detection notification and the detection time is sequentially broadcast to the adjacent nodes and transmitted, and the timing of the predetermined time is started. The sensor that does not transmit the detection message when the detection message is received at the node, and that is measured and recorded on the server when the data acquisition request message is received from the server within the predetermined time at the node The data is transmitted, and the node is configured to cancel the state where the detection message is received at the end of the predetermined time.

  In another example of the aspect, if the node does not hold the detection time when receiving the detection message from the other node, the node holds the detection time included in the received detection message and also receives the detection message. Is transmitted to an adjacent node other than the other node, is transmitted and propagated, starts measuring a predetermined time, and the node holds the detection time when receiving the detection message, If the detection time included in the received detection message is earlier than the detection time held, update the received detection time by replacing it with the detection time included in the received detection message. The detected message is broadcast and transmitted to adjacent nodes other than other nodes, and when the detection message is received at the node, the detection time A case of holding, when the detection time included in the detected message received is of to have detected time after holding is configured to discard the probe message received.

  When the node enters the detection protection period, it waits for a data acquisition request message from the server, and the propagation of unnecessary detection messages is discarded. Traffic concentration (congestion) can be avoided.

It is a figure which shows the structural example of an ad hoc network. It is a figure which shows the hardware structural example of a node. It is a figure which shows the flowchart which shows the flow of the process in a node when it detects with a general method. It is a figure which shows the sequence at the time of the detection by a general method. It is a figure which shows description of the subject of an ad hoc network. It is a functional block diagram of a node in an embodiment. It is principle explanatory drawing which shows the state transition of the node in this embodiment. It is explanatory drawing of the detection Message notification process by this embodiment. It is explanatory drawing of the detection message notification process by the general method contrasted with this embodiment. It is explanatory drawing of the sensor data acquisition process by this embodiment. It is explanatory drawing of the sensor data acquisition process by the general method contrasted with this embodiment. It is a figure which shows the flowchart which shows the control processing in a node when it detects in this embodiment. It is a figure which shows the flowchart which shows the control processing in the node at the time of detection message reception in this embodiment. It is a figure which shows the sequence of the operation example of case 1 of this embodiment. It is a figure which shows the sequence of the operation example of case 2 of this embodiment. It is a figure which shows the sequence of the operation example of case 3 of this embodiment. It is a figure which shows the sequence of the operation example of case 4 of this embodiment.

Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram illustrating a configuration example of an ad hoc network.
The ad hoc network 103 includes nodes 101 numbered from # 1 to # 12 in the figure and a backbone relay 102 (GW: Gate Way).

  When the ad hoc network 103 is applied to an environment monitoring system for performing energy saving management by air conditioning control of a server data center, for example, the node 101 includes sensors such as temperature sensors and wind speed sensors installed at various locations in the data center. It is a communication device to be installed. Further, when the ad hoc network 103 is applied to, for example, a facility management / control system for performing energy saving and security management of a building, the node 101 includes, for example, an illuminance sensor, a temperature sensor, and a human sensor that detect a room state. It is a communication device equipped with a sensor such as. Alternatively, when the ad hoc network 103 is applied to, for example, an inspection business system for performing optimization management of power generation / distribution in an electric power company, the node 101 is, for example, a communication device equipped with a sensor of a meter-reading meter. Furthermore, when the ad hoc network 103 is applied to a structure monitoring system for confirming the soundness of aging due to a bridge, for example, the node 101 includes, for example, an acceleration sensor that detects the state of the structure, a strain It is a communication device equipped with sensors such as sensors and surveillance cameras.

  FIG. 1 shows a configuration in which 12 nodes 101 from # 1 to # 12 are connected to each other. The nodes 101 communicate with each other autonomously according to a communication method of a wired or wireless ad hoc routing protocol.

  When one node 101 and another node 101 are directly connected by a cable, the two nodes 101 are “adjacent”. Another node 101 adjacent to one node 101 may be referred to as an “adjacent node”. The node 101 itself may be referred to as “own node”.

  A core relay (GW) 102 is a gateway (Gate Way) that relays sensor information collected by each node 101 to a monitoring / control server 104 connected via a network 105 such as a LAN (local area network). ) Functions as a device. Also, various control information notified from the server 104 to each node 101 or itself is relayed to each node 101 or itself. In other words, the trunk relay 102 performs conversion between a LAN communication protocol and a wired ad hoc routing protocol.

  The server 104 monitors the sensor information collected by each node 101, displays the sensor information, or detects an abnormality in the sensor information and displays an alarm. The server 104 also controls each node 101.

  In the ad hoc network 103, for example, sensor information collected by each node 101 connected by wire is collected by the server 104 via the trunk relay 102 and is centrally monitored or controlled. This makes it possible to efficiently collect and monitor sensor information.

  FIG. 2 is a diagram illustrating a hardware configuration example of the node 101 in FIG. The node 101 includes a battery 208, a sensor 202, a timer 207, a memory 203, an MPU 201, a PHY 204, an RF unit 205, or a wired connector unit 206. The battery 208 supplies power for the entire node 101. The sensor 202 is a sensor that measures the various physical quantities described above. The timer 207 is a timer group that counts a detection period, a data acquisition period, a detection protection period, and the like, which will be described later. The memory 203 stores a routing table regarding measured sensor data, address information of adjacent nodes, and the like. An MPU (Micro Processing Unit) 201 controls the operation of the entire node 101. The PHY 204 is a conversion unit that converts a logical signal and a physical signal that are transmitted and received between adjacent nodes in the node 101. When the ad hoc network 103 is a wired network, a wired connector unit 206 is connected to the PHY 204. The wired connector unit 206 accommodates a wired communication line. When the ad hoc network 103 is a wireless network, an RF (Radio Frequency) unit 205 that is a radio frequency processing unit is connected to the PHY 204. The RF unit 205 performs wireless communication. Each node 101 has an RF unit 204 or a wired connector unit 205 having three or more ports, and the node 101 can form a network in a daisy chain, mesh, or tree form. The data and detection messages from each node 101 are referred to the routing table stored in the memory 203, and based on a routing method (protocol), the server 104 via the trunk relay 102 in FIG. Sent to.

  In the following description, the ad hoc network 103 having the configuration shown in FIG. 1 will be described as an example. First, a generally considered operation for the sensor of each node 101 on the ad hoc network 103 to perform sensing will be described. Thereafter, the sensing operation of the present embodiment will be described.

  Here, first, an example of a generally considered operation when detected by the node 101 of # 3 in FIG. 1 will be described below. In the following description, “node 101 of # 3” is simply expressed as “node # 3”. The same applies to nodes 101 of other numbers. That is, the node 101 of # 1 is “node # 1”, and the node 101 of # 12 is “node # 12”.

  At this time, it is assumed that each node 101 detects nodes # 3, # 2, # 4, and # 5 in this order, and that node # 1 is not detected. Note that “detection” refers to a state where the value exceeds a certain threshold in the state where the sensor of each node 101 in the ad hoc network 103 is constantly measuring data, as described above.

  In FIG. 1, the number of the node 101 of interest is # 1, # 2, # 3, # 4, # 5, and a flowchart showing a flow of processing of generally considered control operation when each node 101 detects. Is shown in FIG. 3, and a sequence diagram is shown in FIG. The flowchart in FIG. 3 and the sequence in FIG. 4 are control operations in which the MPU 201 in FIG. 2 executes a control program stored in the memory 203, for example.

In a generally considered control operation, each node 101 in the network operates in the order of steps S1 to S9 in FIGS.
First, in each node 101, for example, after establishing a wired link to connect to an adjacent node (step S1 in FIG. 3), data measurement is started by a sensor in the own node (step S2 in FIG. 3).

Subsequently, when the sensor data does not exceed the threshold value (the determination value in step S3 is 0), the process returns to the sensor data measurement state in step S2.
When the sensor data value exceeds the threshold at a certain time and the detection state is reached (the determination value in step S3 is 1), the node 101 transmits a detection message for notifying the detection to the server, Start the data recording timer. These timers are initially set to individual predetermined values at the start, and the values decrease sequentially. At this time, the detection message is propagated through the ad hoc network 103 by hopping each node 101 one after another, such as an adjacent node and an adjacent node of the adjacent node, and is transmitted from the backbone relay 102 to the server 104. The

  The detected node 101 records data by the sensor of its own node and accumulates the data at a certain period, although not shown in FIG. 3, while the value of the detection timer is greater than 0 after the detection time. (The determination value in step S5 in FIG. 3 is 1).

  When the value of the data acquisition timer becomes 0 (the determination value in step S5 in FIG. 3 is 0), the recorded sensor data is collectively notified to the server, and the data acquisition timer is restarted (step S5 in FIG. 3). → S6).

Until it is determined in step S7 that the value of the detection timer has become 0, recording / notification of sensor data from steps S5 to S7 is performed (the determination value in step S7 in FIG. 3 is 1).
If it is determined in step S7 that the value of the detection timer has become 0 (the determination value in step S7 in FIG. 3 is 0), the sensor data recording and transmission process ends (step S7 → S8), and the data in step S2 Return to the measurement start process.

  Based on the above control operation, as shown in the sequence diagram of FIG. 4, for example, each of the nodes # 2, # 3, # 4, and # 5 starts recording sensor data at the detection timing (FIG. 4). 2 corresponds to S4). Then, each node 101 records and accumulates sensor data for a certain period counted by the data acquisition timer of the node, and collectively transmits it to the server 104 (corresponding to S5 and S6 in FIG. 2). As described with reference to FIG. 2, the above-described sensor data recording / transmission operation is executed for a predetermined time counted by the detection timer from the detection time, and then returns to the state of monitoring the detection again.

In such a generally considered technique, each time each node 101 performs detection, sensor data is recorded and transmitted together at each timing.
In the above operation, for example, the nodes # 2, # 3, # 4, and # 5 in FIG. 4 perform detection independently. In this case, there is a possibility that the respective detection timings are close or the timings indicated by 301 in FIG. More specifically, in the configuration example of the ad hoc network 103 in FIG. 2, consider a case in which a detection message and subsequent sensor data are transmitted from each node 101 as indicated by, for example, a dashed arrow in FIG. 5. In such a case, the data of all the nodes 101 is concentrated on the connection port (the RF unit 204 or the wired connector unit 205 in FIG. 2) of the node 101 (node # 1 in FIG. 5) connected to the trunk relay 102. As a result, there is a high possibility of data delay and packet loss.

  In addition, in the above-described generally considered sensing method, a node that does not exceed a certain threshold in the ad hoc network 103 (is not detected) in order to prevent an increase in the amount of data and perform efficient data collection. The data 101 is not acquired. For example, as indicated by 302 in FIG. 4, since the node # 1 has not detected, the sensor data of this node 101 is not transmitted to the server 104 (FIG. 1). However, for example, in the network configuration example shown in FIG. 5, the data of the node # 1 is also stored in the server 104 because of the relationship with the detected nodes 101 of # 2, # 3, # 4, and # 5. There may be situations where you want to observe. In such a case, with the generally considered sensing method described above, the data of node # 1 cannot be acquired until detection is performed at node # 1.

  Therefore, in the embodiment described below, the concentration of packets to the node 101 (node # 1 in the configuration example of FIG. 1) connected to the trunk relay 102 is avoided, and detailed measurement is also performed from a node that does not detect. It is a system that can collect sensor data.

  FIG. 6 is a functional block diagram of the node 101 (FIGS. 1 and 2) in the present embodiment. The functional blocks shown in FIG. 6 are implemented as control processing of the MPU 201 in the hardware configuration of FIG. More specifically, each functional block shown as 601 to 609 in FIG. 6 is implemented as a process in which the MPU 201 executes a control program corresponding to the processes of the flowcharts shown in FIGS. This control program is stored in, for example, the memory 203 in FIG.

In FIG. 6, a sensor data recording unit 601 measures sensor data with the sensor 202 (FIG. 2) in the node 101, and sequentially records the measurement results in the memory 203 (FIG. 2).
When the first detection state starting unit 602 detects that the value of the measured sensor data satisfies the predetermined condition, if the detection message is not received from another node, the first detection state start unit 602 stores the detection time in the memory 203. Alternatively, it is held in a register or the like inside the MPU 201 (FIG. 2). In addition, the first detection state start unit 602 generates a detection message including a detection notification and a detection time, and then sends the detection message to the server 104 (FIG. 1) and sends it to the transmission side PHY 204s of FIG. 6 (part of the PHY 204 of FIG. 2). ) And broadcast to the adjacent nodes. Further, the first detection state starting unit 602 controls the timer 207 to start measuring a predetermined time. Here, “the value of the measured sensor data satisfies a predetermined condition” means, for example, a case where a value larger than a certain value is observed in the sensor 202 and the value of the sensor data exceeds a predetermined threshold. In addition, “when the detection message is not received from another node” more specifically refers to a normal state that is not a detection state. Then, the state of the own node transitions from the normal state to the detection state by the start of timing of the predetermined time.

FIG. 7 is an explanatory diagram of the principle showing the state transition of the node in the present embodiment. In the present embodiment, the node 101 can take two states, a normal state and a detection state.
The normal state is a state in which only reading of sensor data from the sensor 202 (FIG. 2) and recording in the memory 203 are performed, and notification to the server 104 (FIG. 1) is not required. In the normal state, the node 101 only records the sensor data in the memory 203 when the sensor data is read from the sensor 202, as indicated by an event E1 in FIG. This control is executed by the sensor data recording unit 601 in FIG.

  On the other hand, the detection state is a state within a period called a detection protection period, in which a detection message and sensor data are notified to the server 104. As shown in event E2 in FIG. 7, when one node 101 detects that the value of sensor data from the sensor 202 in its own node exceeds the threshold, the node 101 holds the detection time. Then, a detection message including detection notification and detection time is broadcast. Then, the node 101 controls the timer 207 to start measuring a predetermined time, thereby transitioning from the normal state to the detection state as indicated by event E2 in FIG. The first detection state starting unit 602 in FIG. 6 executes this control. The sensor data at this time is also recorded in the memory 203 by the sensor data recording unit 601 in FIG.

  In the node 101 that has entered the detection state by transmitting the detection message, the sensor data recording unit 601 in FIG. 6 reads the sensor value from the sensor 202 as in the normal state, as indicated by event E4 in FIG. When the data is issued, the sensor data is recorded in the memory 203.

  At the same time, the data transmission unit 605 in the first detection state in FIG. 6 receives a data acquisition request message (hereinafter referred to as “data acquisition request message”) from the server 104 during the detection state (within a predetermined time). It will be in a state to wait. Then, as shown in event E5 in FIG. 7, when the data acquisition request Message is received, the first detection state data transmission unit 605 transmits the sensor data recorded in the memory 203 to the server 104. To do. The reception process of the data acquisition request Message is executed via the reception side PHY 204r in FIG. 6 (part of the PHY 204 in FIG. 2). The sensor data transmission process is executed as a process of transmitting to the adjacent node via the transmission side PHY 204s of FIG. 6 (part of the PHY 204 of FIG. 2).

  As described above, the node 101 that has performed the detection transmits the sensor data after receiving the data acquisition request message from the server 104. Therefore, the sensor 101 can transmit the sensor data under the control of the server 104. Become.

In addition to the above operations, the nodes 101 other than the node 101 that has transmitted the detection message perform the following characteristic operations.
As shown as event E3 in FIG. 7, when a detection message is received from another node 101 in a normal state in which the detection time is not held, the second detection state start unit 606 in FIG. The detection time included in is held. This reception operation is executed via the reception side PHY 204r in FIG. 6 (part of the PHY 204 in FIG. 2). The detection time is held in a register or the like in the memory 203 (FIG. 2) or the MPU 201 (FIG. 2). In addition, the second detection state starting unit 606 broadcasts the received detection message to an adjacency other than the receiving node 101 and propagates it. This transmission process is executed via the transmission side PHY 204s in FIG. 6 (part of the PHY 204 in FIG. 2). Then, the second detection state start unit 606 controls the timer 207 to start measuring a predetermined time, thereby transitioning from the normal state to the detection state as indicated by event E3 in FIG. That is, when another node 101 detects, the other node 101 that has received the detection message broadcast from the node 101 also shifts to the detection state.

  As described above, when the detection message is received and transition to the detection state, the detection message transmission suppression unit 604 in the detection state in FIG. 6 enters a state where the value of the sensor data in the own node exceeds the threshold and should be detected. However, the detection is suppressed and the detection message is not transmitted to the server 104. As a result, as described in FIG. 5 and the like, detection messages and subsequent sensor data are simultaneously transmitted from a plurality of nodes 101, and congestion occurs in the nodes 101 connected to the trunk relay 102. It becomes possible to suppress the possibility of doing.

  Here, the detection state detection message transfer unit 607 and the detection state detection message discard unit 608 of FIG. 6 detect from other nodes 101 when the detection time is already held, that is, when the own node is in the detection state. When Message is received, the following control processing is executed. This reception process is executed via the reception-side PHY 204r in FIG. 6 (part of the PHY 204 in FIG. 2). In addition, the detection time is determined when the first detection state start unit 602 or the second detection state start unit 606 in FIG. 6 executes a transition from the normal state to the detection state. It is held in a register or the like inside the MPU 201 (FIG. 2).

  First, as shown in event E6 in FIG. 7, the detection message transfer unit 607 in the detection state holds when the detection time in the received detection message is earlier than the detection time held. Replace the detected detection time with the received detection time and update it. Then, the detection message transfer unit 607 in the detection state broadcasts and transmits the received detection message to an adjacent node other than the node 101 that is the reception source. This transmission process is executed via the PHY 204s on the transmission side in FIG. 6 (part of the PHY 204 in FIG. 2).

  On the other hand, as also shown in event E6 of FIG. 7, the detection message discarding unit 608 in the detection state receives the reception when the detection time included in the received detection message is after the detection time held. Discard the detected message.

  As described above, the node 101 operates so that only the detection message detected and transmitted earliest in the current detection state is propagated to the server 104. Thereby, it is possible to suppress the occurrence of congestion due to the plurality of detection messages transmitted from the plurality of nodes 101 almost simultaneously.

  In the node 101 that has received the detection message and transitioned to the detection state, transmission of sensor data is suppressed by the detection message transmission suppression unit 604 in the detection state of FIG. However, such a node 101 can transmit the sensor data recorded in the node to the server 104 by the following means. That is, when the second detection state data transmission unit 609 in FIG. 6 receives sensor data from another node 101 in the detection state (within a predetermined time), the control operation shown in step E7 in FIG. 7 is executed. To do. That is, the second detection state data transmission unit 609 adds the sensor data measured by the sensor 202 in the own node and recorded in the memory 203 to the sensor data received from the other node 101, It is transmitted to an adjacent node other than the node 101 and propagated. The process of receiving sensor data from the other nodes 101 is executed via the PHY 204r on the receiving side in FIG. 6 (part of the PHY 204 in FIG. 2). Further, the transmission process by adding the sensor data is executed via the transmission-side PHY 204s in FIG. 6 (part of the PHY 204 in FIG. 2).

  This additional sensor data transmission may be executed whenever sensor data is received from another node 101, or only when designated by the server 104. May be. In any case, the sensor data of not only the node 101 that performed the detection but also the other node 101 on the route between the node 101 and the core relay 102 is triggered by the data acquisition request message from the server 104 as a trigger. It becomes possible to collect finely. In addition, in that case, since no extra data transmission occurs from each node 101, it is possible to avoid the congestion state of the ad hoc network 103 (FIG. 1) in the node 101 connected to the trunk relay 102.

  As shown in event E8 of FIG. 7, in the detection state, when the detection state output unit 105 of FIG. 6 detects that the timer 207 of FIG. The state of the own node is changed from the detection state to the normal state. As a result, the state in which the detection message is received is canceled, and the normal state in which the sensor data need not be notified to the server 104 is restored.

  As described above, in this embodiment, when any node in the ad hoc network is the node #X and the node #Y, when the node #X detects, a detection message is transmitted by broadcast to all the nodes in the network. Then, the node #X transitions from the normal state to the detection state so as to have a data acquisition request message transmitted from the server. In addition, the node #Y (not detected) that received the detection message also has a detection state (detection protection period) in which the detection message is not detected by transitioning from the normal state to the detection state, and the detection message is not broadcast to other nodes. . As a result, it is possible to avoid congestion at the node 101 connected to the trunk relay 102, for example, node # 1 in the network configuration example of FIG. Note that, like the node #X, the node #Y also has a data acquisition request message transmitted from the server when the detection protection period is reached.

  As a data acquisition method, a server that has received a detection message transmits a data acquisition request message to a terminal node in the network. The sensor data recorded in the received end node is notified to the server by propagating through the adjacent nodes. And when propagating, the sensor data recorded at the intermediate node is also added and propagated. Thereby, the server 104 can collect the sensor data of the node 101 related to the detected node 101 collectively. As a result, congestion at the node 101 connected to the trunk relay 102 can be avoided, and at the same time, sensor data of the node 101 that has not been detected can be collected finely.

  The server may determine the request timing of the data acquisition request Message appropriately. In the present embodiment, it is assumed that time synchronization in each node 101 within the ad hoc network 103 (FIG. 1) is performed in advance. As a method of time synchronization, a generally used method such as a time synchronization protocol (NTP: Network Time Protocol) may be used.

FIG. 8 is an explanatory diagram of detection message notification processing according to this embodiment based on the functional configuration diagram of the node of FIG. 6 and the state transition diagram of FIG.
First, the link of each node in the network is established. In FIG. 8, it is assumed that the node #C detects first, and a detection message is transmitted by broadcast to all the nodes in the network.

  There are two conditions for entering the detection state (detection protection period). One is a case where an arbitrary node, in the example of FIG. 8, is detected by the node #C (S1 in FIG. 8). The other is when the detection message is received from the adjacent node, and in the example of FIG. 8, the node #B (#E) receives the detection message from the node #C (#D) (S2 in FIG. 8). ). At the same time as the detection protection period, data recording starts.

  Since the node #B (#E) during the detection protection period is in the detection inhibition state, the detection message is not transmitted even if the detection should be performed. Further, it waits for a data acquisition request message from the server (S3 in FIG. 8).

  Since nodes #C and #D are detecting, they are in the detection protection period. Node #C receives the detection message from node #D. However, since the detection time in the received detection message is later than the detection time originally held by the node #C, the received detection message is discarded (S4 in FIG. 8).

  Node #D receives the detection message from node #C. In this case, since the detection time included in the detection message from the node #C is earlier (older) than the detection time originally held by the node #D, the node #D updates the detection time of its own node and is adjacent. Propagate to node #E (step S5 in FIG. 8).

  At this time, each node constantly measures data, and past sensor data (not exceeding the threshold value) can be read for a few seconds before the detection time (time when the sensing value exceeds the threshold value). (S6 in FIG. 8). Thus, for example, the server can analyze the temperature change immediately before exceeding the threshold by requesting and acquiring the data immediately before exceeding the threshold from each node. The end of the detection state (detection protection period) is controlled by a detection protection period timer.

Finally, the server receives the detection message only from the node #C (S7 in FIG. 8).
The detection message received by the node #E is not transmitted to the server but is propagated to notify the detection message to other nodes in the network.

FIG. 9 is an explanatory diagram of detection message notification processing by a general method compared with the present embodiment of FIG.
In FIG. 9, when each node detects, it independently transmits a detection message to the server (S9 in FIG. 9).

The trunk relay and the server receive and process all the detected messages of the detected nodes (S10 in FIG. 9).
As a result, in the general method, depending on the proximity of each detection timing or the adjacent relationship of routing, the data of all nodes is connected to the connection port of the node (node # 1 in the example of FIG. 5) connected to the trunk relay. Can end up concentrating. As a result, the possibility of occurrence of congestion increases.

  In a general technique, the server may want to collect not only the nodes #B, #C, #D, and #E but also sensor data of the node #A. However, in this case, since the node #A has not detected, the sensor data cannot be acquired.

  On the other hand, in the present embodiment shown in FIG. 8, each node receives a detection message and transitions to a detection state (detection protection period), so that a node that has not been detected (node #A in the example of FIG. 8). ) Data acquisition is now possible. That is, the server can perform comprehensive data analysis including sensing states of nodes around the detected node. This is because the detection message is broadcast and the detection message (detection protection period) is entered upon reception of the detection message.

  The detection protection period starts at the detection time and ends at the set predetermined time. On the other hand, the data collection period (the period until the data acquisition request message is received) may be shorter than the detection protection period.

FIG. 10 is an explanatory diagram of sensor data acquisition processing according to the present embodiment.
Nodes #A, #B, #C, #D, and #E are in the detection state (detection protection period), and the node #E is the terminal node. Since the server has the configuration information of the ad hoc network, it is possible to determine the end node.

In FIG. 10, in order to acquire data, the server that has received the detection message transmits a data acquisition request message to the terminal node #E (S1 in FIG. 10).
The sensor data of the node #E that has received the data acquisition request message is propagated by hopping from node # D → node # C → node # B → node #A (S2 in FIG. 10). At this time, the sensor data of the intermediate node is added to the sensor data of the node #E, and these sensor data groups are acquired by the server. Although all sensor data of nodes in the detection state (detection protection period) in the network can be acquired, data of all nodes in the network cannot be acquired with one detection message. For this reason, there is a node from which data is not acquired by routing. However, the number of packets can be significantly reduced compared to sending a data acquisition request message to each node. Further, when the data acquisition route is covered, it is determined by the node, and it is possible to operate so as not to add the data.

  FIG. 11 is an explanatory diagram of sensor data acquisition processing by a general method compared with the present embodiment of FIG. The sensor data is independently transmitted at the timing detected by each node and received by the server (S3 in FIG. 11).

  As a result, as in the case of the detection message, in the general method, the sensor data of all nodes is concentrated on the connection ports of the nodes connected to the trunk relay, and the possibility of occurrence of congestion increases. Further, in a general method, the server cannot collect sensor data of the node #A.

  On the other hand, in this embodiment, the server can also collect sensor data of nodes related to the detected node collectively. As a result, congestion at nodes connected to the trunk relay can be avoided, and at the same time, sensor data of nodes not detected can be collected finely. FIG. 12 is a flowchart showing a control process in the node 101 (FIG. 1) when it is detected in the present embodiment. This control process includes a sensor data recording unit 601, a first detection state start unit 602, a detection state end unit 603, a detection message transmission suppression unit 604, and a first detection state data transmission unit 605 in FIG. Realize the function. This control process is, for example, a part of the operation in which the MPU 201 executes the control program stored in the memory 203 in the hardware configuration of FIG.

A link is established within the ad hoc network 103 (FIG. 1) (step S1 in FIG. 12).
Measurement and recording of sensor data is started by the sensor 202 (FIGS. 2 and 6) (step S2 in FIG. 12). This processing corresponds to the function of the sensor data recording unit 601 in FIG.

If the measured data does not exceed the threshold value, the sensor data measurement / recording process is repeated (the determination value in step S3 in FIG. 12 is 0 → step S2).
When the measured data is detected exceeding the threshold, it is determined whether or not the detected message is received (the determination value in step S3 in FIG. 12 is 1 → step S4). This process is a part of the function of the first detection state start unit 602 in FIG. Here, it may be determined by providing an internal register that is set upon reception of the detection message and reset at the end of the detection state (detection protection period), or when the value of the detection protection period timer is not 0, the detection message It may be a process for determining that the state is being received.

  If the detection message is already received, the detection message is not transmitted because it is in the detection state (detection protection period) (the determination value in step S4 in FIG. 12 is 1 → step S5). This process is a part of the function of the first detection state start unit 602 in FIG.

  When the detection message is not received, the timer 207 (FIGS. 2 and 6) is controlled to start the detection protection period timer. Then, the detection time is recorded in the memory 203 (FIGS. 2 and 6) or a register in the MPU 201, and the detection message is notified to all nodes by broadcast (the determination value in step S4 in FIG. 12 is 0 → step S6). ). This process is a part of the function of the first detection state start unit 602 in FIG.

  While waiting for reception of the data acquisition request message transmitted from the server 104, it is determined whether or not the value of the detection protection period timer in the timer 207 has become 0 (determination of step S7 → step S8 in FIG. 12). Value is 1 → repetition processing of step S7). This processing is a part of the function of the first detection state data transmission unit 605 and the detection state end unit 603 of FIG.

  When the data acquisition request message is received in the detection state, the sensor data is transmitted to the server 104 (the determination value in step S8 in FIG. 12 is 0 → step S9). This process is a part of the function of the data transmission unit 605 in the first detection state in FIG.

  When the detection protection period timer becomes 0 after a predetermined time has elapsed, the detection protection period timer is reset, and the detection protection period (detection state) ends (the determination value in step S7 in FIG. 12 is 0 → step S10). Then, it returns to step S2.

  FIG. 13 is a flowchart showing the control process in the node when receiving the detection message in this embodiment. This control process realizes the functions of the second detection state start unit 606, the detection state detection message transfer unit 607, the detection state detection message discard unit 608, and the second detection state data transmission unit 609 in FIG. . This control process is, for example, a part of the operation in which the MPU 201 executes the control program stored in the memory 203 in the hardware configuration of FIG. The node 101 executes detection message reception processing (step S1 in FIG. 13). This process is a function of the PHY 204r in FIG. 6 (part of the PHY 204 in FIG. 2).

  It is determined whether the node 101 receiving the detection message holds the detection time. The detection time is held in, for example, a register inside the memory 203 (FIG. 2) or the MPU 201 (FIG. 2) in step S6 of FIG. 12 or step S3 of FIG. This process is a part of the function of the second detection state start unit 606 in FIG.

  When the detection time is not held, the timer 207 is controlled to start the detection protection period timer, and the detection time is recorded in a register inside the memory 203 (FIG. 2) or MPU 201 (FIG. 2). The detection protection period (detection state) is reached (the determination value in step S2 in FIG. 13 is 0 → step S3). Thereafter, the detection message is propagated by hopping to an adjacent node (steps S3 → S7 in FIG. 13). This process is a part of the function of the second detection state starting unit 606 in FIG.

  When the detection time is held, the detection time included in the received detection message is compared with the held detection time (the determination value in step S2 in FIG. 13 is 1 → step S4). This processing is a part of the functions of the detection state detection message transfer unit 607 and the detection state detection message discard unit 608 in FIG.

  If the detection time included in the received detection message is later than or equal to the detection time held, the received detection message is discarded (the determination value in step S4 in FIG. 13 is 0 → S5). Thereafter, the detection message reception process in FIG. 13 is terminated. This process is a part of the function of the detection message discarding unit 608 in the detection state of FIG.

If the received detection time is earlier than the held detection time, the held detection time is updated with the received detection time (the determination value in step S4 in FIG. 13 is 1 → step S6). Thereafter, the detection message is propagated by hopping to an adjacent node (step S7 in FIG. 13). This processing is a part of the function of the detection state detection message transfer unit 607 in FIG.
As described above, the processing corresponding to the function of FIG. 6 according to the present embodiment is realized.

Hereinafter, operation examples in some cases in the present embodiment will be described.
FIG. 14 is a diagram illustrating a sequence of an operation example of case 1 of the present embodiment.

Consider a case where, in the ad hoc network 103 of FIG. 1, node # 3 is detected first and node # 5 is not detected.
FIG. 14 is a sequence diagram when nodes are detected focusing on the nodes # 1, # 2, # 3, # 4, and # 5.

The routing method is node # 5 → node # 4 → node # 3 → node # 1 → server 104.
Node # 3 detects and notifies all nodes 101 in the network by broadcast (S1 in FIG. 14). This is according to step S6 in FIG.

  Although the nodes # 2, # 4, and # 5 are in the detection state, the detection message is not transmitted because the detection message is received from the node # 3 (S2 in FIG. 14). This is due to step S4 in FIG.

The server 104 that has received the detection message from the node # 3 transmits a data acquisition request message to the end node # 5 (S3 in FIG. 14).
In response to the data acquisition request message, sensor data is transmitted from the node # 5. This process is in accordance with steps S8 → S9 in FIG. Sensor data recorded by the node 101 in the middle of receiving the sensor data is added to the sensor data transmitted from the node # 5 and propagated (S4 in FIG. 14).

Since the node # 5 which has not been detected has also received the detection message and has transitioned to the detection state, the sensor data can be acquired based on the data acquisition request message (S5 in FIG. 14).
The data of node # 2 is not acquired because of routing (acquired in the case of a route that propagates through node # 2) (S6 in FIG. 14).

FIG. 15 is a diagram illustrating a sequence of an operation example of case 2 of the present embodiment.
Consider a case where a certain node 101 equipped with a sensor 202 (FIG. 2) detects in the ad hoc network 103 of FIG.

The nodes are detected in the order of nodes # 3, # 2, # 4, and # 5, and node # 1 is not detected.
FIG. 15 is a sequence diagram when the nodes are detected focusing on the nodes # 1, # 2, # 3, # 4, and # 5.

The routing method is node # 5 → node # 4 → node # 3 → node # 1 → server.
The node # 3 detected first notifies the detection message to all nodes in the network by broadcast in order to notify the detection (S1 in FIG. 15). This is according to step S6 in FIG.

The node # 2 detected next to the node # 3 detects it before receiving the detection message from the node # 3 (S2 in FIG. 15). This is due to steps S4 → S6 in FIG.
The detected node # 2 notifies the detection message to all nodes in the network by broadcast. The detection message from the node # 2 to the node # 3 (detected) is compared with the detection time at the node # 3, and since the detection time at the node # 3 is earlier, it is discarded and not routed (S3 in FIG. 15). This is due to steps S4 → S5 in FIG.

  Next, the detected nodes # 4 and # 5 receive the detection message from the node # 3 and are in the detection state (detection protection period). It is not broadcast to the node (S4 in FIG. 15). This process is performed in steps S4 to S5 in FIG.

The server 104 that has received the detection message transmits a data acquisition request message to the node # 5 in order to acquire sensor data (S5 in FIG. 15).
The data of the node # 5 that has received the data acquisition request message is propagated by hopping from the node # 5 → # 4 → # 3 → # 1. This process is in accordance with steps S8 → S9 in FIG. At this time, the data of the nodes (# 4, # 3, # 1) in the middle are added to the data of the node # 5, and the sensor data is notified to the server 104 (S6 in FIG. 15).
When acquiring data, it is also possible to acquire data of node # 1 that has not been detected (S7 in FIG. 15).

FIG. 16 is a diagram illustrating a sequence of an operation example of case 3 of the present embodiment.
Consider a case where a certain node 101 equipped with a sensor 202 (FIG. 2) detects in the ad hoc network 103 of FIG.
FIG. 16 is a sequence diagram focusing on the nodes # 1, # 2, # 3, # 8, # 10, and # 11 assuming that all the nodes 101 in the ad hoc network 103 are in the detection state at the same time.

  Detection is performed by each node 101 and a detection message is notified to all nodes 101 by broadcast. This process is performed in steps S3 → S4 → S6 in FIG. The detection message broadcasted by the nodes # 2, # 3, # 8, # 10, and # 11 to the adjacent nodes is compared with the detection time of the received node and discarded (S1 in FIG. 16). This process is performed in steps S4 to S5 in FIG.

Only the detection message of the node # 1 is notified to the server 104, and the nodes # 2, # 3, # 8, # 10, and # 11 enter the detection state (detection protection period) and wait for the data acquisition request message (S2 in FIG. 16). ). This process is based on the loop of S7 → S8 → S7 in FIG.
The node # 11 that has received the data acquisition request Message hops through each node and propagates according to the routing method (S3 in FIG. 16).

FIG. 17 is a diagram illustrating a sequence of an operation example of case 4 of the present embodiment.
Consider a case where a node 101 equipped with a sensor 202 detects in the ad hoc network 103 of FIG.
In FIG. 17, it is assumed that the nodes # 2, # 3, and # 11 are simultaneously detected first, and the nodes # 1, # 10, and # 11 are detected second, and the nodes # 1, # 2, # 3, # 8, FIG. 10 is a sequence diagram focusing on # 10 and # 11.

The detected node 101 broadcasts a detection message to all the nodes 101 by broadcast.
Since the detection message from the nodes # 2, # 8, and # 11 detected first is the same as the detection time of each node 101, the detection message (Δ mark in FIG. 17) is discarded (S1 in FIG. 17). ).

Detection messages (circles in FIG. 17) from the nodes # 1, # 3, and # 10 are discarded because the first detection message has been received (S2 in FIG. 17).
In the node # 1 that receives the detection message from the nodes # 2 and # 3, the detection message with the later reception time is discarded (S3 in FIG. 17).

  In this case, only the detection message of node # 1 and node # 2 (or # 3) is notified to the server 104, and each node 101 enters the detection protection period and waits for a data acquisition request message from the server 104.

  As described above, in this embodiment, when a node enters the detection protection period, the server has a data acquisition request message from the server, and even if the node detects it, transmission of the detection message can be avoided. . Also, if a detection message (slow detection time) is sent to a node that has received a detection message (detection time is early), it will be discarded and the detection message will no longer be routed. This makes it possible to reduce traffic in the network and avoid traffic concentration (congestion) at ports connected to backbone relay.

  In this embodiment, when acquiring data measured at each node, by specifying the end node from which data is acquired from the server, by adding the data of the intermediate node that the end node is hopping and propagating and propagating Further traffic reduction is obtained. The end node is assumed to be known.

  In the conventional node device, only the data after detection by the node is recorded and notified to the server, but in this embodiment, not only data after detection but also data about several seconds before the detection start time can be acquired. It becomes like this.

  The conventional node device has a mechanism that does not acquire data of a node that has not been detected. However, in the present embodiment, it is also possible to acquire data of a node that has not been detected.

101 Node 102 Core Relay 102 (GW: Gate Way)
103 ad hoc network 104 server 105 network 201 MPU (Micro Processing Unit)
202 Sensor 203 Memory 204 PHY
205 RF unit 206 Wired connector unit 207 Timer 208 Battery 601 Sensor data recording unit 602 First detection state start unit 603 Detection state end unit 604 Detection state detection message transmission suppression unit 605 First detection state data transmission unit 606 Detection state start unit 607 Detected state detection message transfer unit 608 Detected state detection message discard unit 609 Second detection state data transmission unit

Claims (6)

In the node, sensor data is measured by the sensor in the node and sequentially recorded,
In the node, when it is detected that the value of the measured sensor data satisfies a predetermined condition, if the detection message is not received from another node, the detection time is held, and the detection notification And a detection message including the detection time is sequentially broadcast to adjacent nodes and transmitted, and a predetermined time is started.
In the node, when the detection message is received, the detection message is not transmitted,
In the node, when the data acquisition request message is received from the server within the predetermined time, the sensor data that is measured and recorded is transmitted to the server,
In the node, cancel the state in which the detection message is received at the end of the predetermined time.
Sensor data transmission method characterized by the above. In the node, when the detection time is not held when receiving the detection message from another node, the detection time included in the received detection message is held, and the received detection message is transferred to the other node. Broadcast to adjacent nodes other than, transmit and propagate, and start timing a predetermined time,
In the node, when the detection message is received, the detection time is held, and the detection time included in the received detection message is earlier than the held detection time. The update is performed by replacing the held detection time with the detection time included in the received detection message, and the received detection message is broadcast and transmitted to an adjacent node other than the other nodes. Let
In the node, when the detection time is received when the detection message is received, and the detection time included in the received detection message is after the held detection time , Discard the received detection message,
Sensor data transmission method characterized by the above. In the node, when sensor data is received from the other node within the predetermined time, the sensor data measured and recorded by the sensor in the node is added to the sensor data received from the other node. Transmit and propagate to adjacent nodes other than the other nodes;
The sensor data transmission method according to claim 1 or 2, wherein A sensor data recording unit that measures and sequentially records sensor data with a sensor in the node;
When it is detected that the value of the measured sensor data satisfies a predetermined condition, if the detection message is not received from another node, the detection time is held, and the detection notification and the detection time are set. A detection message including a first detection state start unit that sequentially broadcasts and transmits the detection message to adjacent nodes and starts measuring a predetermined time;
In the node, when the detection message has been received, the detection message is not transmitted, the detection message transmission suppression unit in the detection state,
In the node, when a data acquisition request message is received from the server within the predetermined time, a first detection state data transmission unit that transmits the measured and recorded sensor data to the server;
In the node, a detection state end unit that cancels the state in which the detection message is received at the end of the predetermined time;
A sensor node characterized by comprising: In the node, when the detection time is not held when receiving the detection message from another node, the detection time included in the received detection message is held, and the received detection message is transferred to the other node. A second detection state starter that broadcasts and transmits to adjacent nodes other than and propagates and starts measuring a predetermined time;
In the node, when the detection message is received, the detection time is held, and the detection time included in the received detection message is earlier than the held detection time. The update is performed by replacing the detected detection time with the detection time included in the received detection message, and broadcasting the received detection message to an adjacent node other than the other node. A detection message transfer unit in a detection state to be propagated; and
In the node, when the detection time is received when the detection message is received, and the detection time included in the received detection message is after the held detection time A detection state discarding message discarding unit for discarding the received detection message;
A sensor node characterized by that. In the node, when sensor data is received from the other node within the predetermined time, the sensor data measured and recorded by the sensor in the node is added to the sensor data received from the other node. A second detection state data transmitter for transmitting and propagating to an adjacent node other than the other node;
The sensor node according to claim 4, wherein the sensor node is a sensor node.