JP5235535B2 - Node time synchronization method and sensor network system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To synchronize, with high accuracy, time of a plurality of sensor nodes performing intermittent operation of repeating an operation state and a sleeping state at fixed time intervals for electric power saving. <P>SOLUTION: A first slave node acquires first sensor data in a first operation state to be transmitted to a master node. A second slave node acquires second sensor data in a second operation state to be transmitted to the master node. The master node sets a timing of transmitting a time synchronizing signal based on operation intervals of the first and the second slave nodes to be transmitted. Based on the timing of transmitting the time synchronizing signal, the first slave node shifts from a first sleeping state to a third operation state, and the second slave node shifts from a second sleeping state to a fourth operating state. The master node broadcast-transmits the time synchronizing signal at the timing of transmitting the time synchronizing signal. The first and the second slave nodes adjust their own times to the time included in the time synchronizing signal. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

  The present invention relates to a sensor node utilization system that performs intermittent operation in which an operation state and a sleep state are repeated at regular time intervals for power saving, and in particular, the entire sensor node in the system or a part of a plurality of sensor nodes. It is related with the technology which synchronizes time between.

  In recent years, a network system (hereinafter referred to as a sensor network system) in which a small electronic circuit having a wireless communication function is added to a sensor and various information in the real world is taken into an information processing device in real time has been studied.

  Further, development of a sensor network system including a small wireless sensor node (hereinafter referred to as a sensor node) equipped with a sensor function, a relay station, a base station, and a sensor network management server (hereinafter referred to as a management server) is being promoted. The sensor node acquires data (sensor data) related to the state of a person or a place, etc., relays the acquired sensor data in a multi-hop by a relay, and transmits it to the management server via the base station.

  The sensor network system collects sensor data of a plurality of sensor nodes, collectively manages them, performs data processing (analysis and display) according to the application, and realizes various services. For example, if acceleration sensor nodes are installed at multiple locations on a certain structure at a certain distance, the line, surface, and three-dimensional behavior of the structure corresponding to the shaking such as an earthquake will be analyzed from the time variation of each acceleration. it can.

  In addition, a sensor node based on battery driving is required to operate with low power consumption. For this reason, the sensor node normally performs intermittent operation. This is an operating state that drives the necessary hardware only when performing tasks such as sensing and data transmission, and a sleep state that sleeps the microcomputer, wireless communication function, etc. in low-power mode when there are no tasks to be performed. It is an operation to repeat. By performing the intermittent operation, the sensor node can operate for a long time under a limited battery.

As a method of adjusting the time of each sensor node, there is a method of performing time synchronization on a one-to-one basis between a master station and a slave station (see, for example, Patent Document 1).
In addition, there is a method in which a plurality of child nodes are time-synchronized simultaneously with the same time-synchronization signal transmitted from the parent node, in contrast to the above-described method of time-synchronizing the sensor nodes individually. Generally, it is classified as RBS (Reference Broadcast Synchronization), and high-precision synchronization accuracy is possible in a short time.

JP 2007-005991 A

  Install acceleration sensor nodes at multiple locations on a structure at a certain distance, and analyze the line, surface, or three-dimensional behavior of the structure in response to shaking such as an earthquake from the time variation of each acceleration. In this case, since the acceleration value with respect to shaking changes rapidly with respect to time, sensing time accuracy is very important. For example, when each of a plurality of acceleration sensor nodes is sampling acceleration at 100 Hz, it is necessary to maintain the operation so that the time between the sampling data of the plurality of sensor nodes is maintained at a deviation within at least 1/200 second.

  Here, within 1/200 second (5 milliseconds), sampling is performed at the same time when the behavior is accurately analyzed on the line, surface, or three dimensions from acceleration data at multiple sensor nodes. This is a restriction for obtaining an accuracy capable of associating a certain thing.

In Patent Literature 1, a one-to-one time synchronization procedure is taken every time communication is performed between a master station and a slave station. In this way, when trying to synchronize each slave station individually, there is fluctuation in the required time from when the master station generates a signal until the slave station receives the signal and adjusts the time.
For example, in transmission processing, CSMA / CA (Carrier Sense Multiple Access with Collision Avoidance) processing to avoid radio wave collision and retransmission processing to improve reliability (retransmit when ack signal is not received) are required. However, an indeterminate (random number) delay time occurs as a standard in each process. Therefore, it may be appropriate when the required accuracy of time synchronization between multiple slave stations is low, for example, when realizing the time synchronization required for communication between the master station and the slave station, the time between slave stations It is difficult to increase the synchronization accuracy. Therefore, the required accuracy of time synchronization between sensor nodes in the case where acceleration is sampled with high frequency as described above cannot be satisfied.

  In addition, the method of time-synchronizing a plurality of child nodes at the same time with the same time synchronization signal is based on the premise that the child nodes can always receive, and the power consumption of the sensor node becomes a big problem. It is necessary to consider so that the same time synchronization signal can be reliably received between a plurality of sensor nodes that are intermittently operated.

  Therefore, an object of the present invention is to highly accurately synchronize the time between a plurality of sensor nodes that repeats an operation state and a stop state at regular time intervals in order to save power.

  Of the inventions disclosed in this application, the outline of typical ones will be briefly described as follows.

  The first child node repeats the first operation state and the first sleep state, acquires first sensor data in the first operation state, and transmits the first sensor data to the parent node. The second child node repeats the second operation state and the second sleep state, acquires the second sensor data in the second operation state, and transmits it to the parent node. The parent node sets the timing for transmitting the time synchronization signal based on the operation interval between the first and second child nodes, and indicates the timing for receiving the first and second sensor data and transmitting the time synchronization signal. A signal is transmitted to the first and second child nodes. The first child node shifts from the first sleep state to the third operation state based on the timing of transmitting the time synchronization signal. The second child node shifts from the second sleep state to the fourth operation state based on the timing of transmitting the time synchronization signal. The parent node broadcasts the time synchronization signal at the timing of transmitting the time synchronization signal. The first child node adjusts its own time to the time included in the time synchronization signal received in the third operation state, and the second child node uses the time synchronization signal received in the fourth operation state. Adjust own time to included time.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to perform highly accurate time synchronization between the data (for example, 3-axis acceleration by an acceleration sensor) acquired at high frequency in each of several sensor nodes.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a configuration example of a multipoint observation system using a plurality of sensor nodes. It comprises a plurality of sensor nodes 210 to 250, a radio base station 110, a radio repeater (or simply, repeater) 160, a management server 30, and a client terminal 40. The plurality of sensor nodes 210 to 250, the radio base station 110, and the radio repeater 160 are capable of radio communication, and the radio base station 110, the management server 30, and the client terminal 40 are connected via the network 20. The number of sensor nodes is not limited to five, and the number of wireless base stations 110 and wireless repeaters 160 is not limited to one each.

  The plurality of sensor nodes are intermittently operated by battery drive, and repeat the operation state and the pause state at intervals set for each sensor node. At the time of startup, a physical quantity is observed at a sampling frequency specified in advance by a sensor (temperature sensor, acceleration sensor, illuminance sensor, etc.) mounted, and the result is transmitted to the radio base station 110 by radio communication. The observed physical quantity is transferred from the radio base station 110 to the management server 30 via the network 20.

  A sensor node is connected to the radio base station 110 or the radio repeater 160, and the radio base station 110 or the radio repeater 160 is associated with the communication ID of the connected sensor node and the communication ID. A means for knowing the operation interval of the intermittent operation for each sensor node is provided. Specifically, the wireless base station 110 or the wireless repeater 160 may manage information (operation intervals, etc.) of sensor nodes connected by itself, or the management server 30 may collect information on sensor nodes. Management may be performed so that the radio base station 110 or the radio repeater 160 inquires the management server 30 about sensor node information when necessary.

  The management server 30 includes a transmission / reception unit, a storage unit, and a control unit (not shown). The transmission / reception unit transmits and receives data to and from the base station. The storage unit is configured by a nonvolatile storage device such as a hard disk or flash memory, and various system configuration information (operation parameters such as sensor node operation intervals, types of mounted sensor devices, various identifiers of each node, etc.) ), And a program executed by the control unit. The control unit executes data analysis and setting or updating of configuration information by an administrator by executing a program stored in the storage unit.

  The client terminal 40 displays the observation data and the result of analyzing it based on the user's request. That is, the user can view the observation data and the result of analyzing it through the client terminal 40.

In addition, a remote monitoring service can be realized by making the network 20 a wide area network. By placing the management server 30 together with a large-capacity storage in a data center or the like, it is possible to conduct a service business including data accumulation, analysis, provision, and the like.
In FIG. 1, a plurality of sensor nodes 210 to 250 are installed in a certain structure 10, and at least an acceleration sensor is mounted on the plurality of sensor nodes. When some vibration (for example, earthquake) is applied to the structure 10, each point generates a shake, and the acceleration sensors of the sensor nodes 210 to 250 can detect the shake. By collecting the shaking results in the management server 30, it is possible to analyze what kind of shaking has occurred and transmitted to the structure.

  However, in order to perform such an analysis, it is necessary that time information (or time stamp) of sensor data to be collected is time synchronized with high accuracy. For example, when the sampling frequency of the acceleration sensor of the sensor node is 100 Hz, it is necessary to synchronize the time between the plurality of sensor nodes with accuracy that can correctly distinguish the sampling period, that is, 10 msec.

  2A shows a hardware configuration example of the radio base station 110 and the sensor node 210, and FIG. 2B shows a hardware configuration example of the repeater 160.

  The radio base station 110 includes an antenna 1101, an RF circuit 1102, a power source 1103, an MC (microcontroller) 1104, an RTC (Real-Time Clock) 1105, a non-volatile memory 1106, and an upper communication unit 1107, and operates according to a program on the MC 1104. To control.

  The sensor node 210 includes an antenna 2101, an RF circuit 2102, a power supply 2103, an MC 2104, an RTC 2105, a non-volatile memory 2106, and a sensor 2107, and the operation is controlled by a program on the MC (microcomputer or microcontroller) 2104. Reference numerals 2110 to 2119 denote functional blocks of the program. In the hibernation state, the power supply control unit 2118 cuts off the power supply other than the RTC 2105 to suppress power consumption. The MC 2104 is driven by the timer interrupt of the RTC 2105, and the task control unit 2110 controls the operation according to the program. The operation of the task control unit 2110 started after the timer interruption is roughly as follows. First, the power supply control unit 2118 operates the power supply circuit switching I / F (register) 2119 to supply power to each functional block on the sensor node. Next, the sensor I / F (register) is operated via the sensor control unit 2114, and as a result, acceleration data is acquired. The acquired acceleration data is stored in the nonvolatile memory 2106 by operating the nonvolatile memory I / F (register) via the data management unit 2116. At the same time, data for the upper system is created. When several samplings of acceleration data are accumulated, the task control unit 2110 operates the RF circuit I / F (register) via the wireless communication unit 2112 to accumulate and transmit the data wirelessly toward the host system. Such a series of sampling operations is executed with the signal of the timer interrupt register 2111 as a trigger.

  The repeater 160 includes an antenna 1601, an RF circuit 1602, a power source 1603, an MC (microcontroller) 1604, an RTC 1605, and a nonvolatile memory 1606. The repeater receives the sensor data via the antenna, accumulates it in the nonvolatile memory 1606, and then transfers it to the base station 110. These operations are controlled by a program on the MC.

  Here, there are roughly two types of conventional means for realizing time synchronization of sensor nodes. One is a method in which a parent node performs data synchronization with a plurality of child nodes individually and executes time synchronization processing, and the other is a method in which a plurality of child nodes simultaneously receive signals transmitted from the parent node, and time synchronization is performed. This is a method for executing processing.

  FIG. 3 is a sequence diagram showing an example of a conventional technique for realizing time synchronization, and illustrates a case where a parent node and a child node are time-synchronized on a one-to-one basis. In FIG. 3, a node that provides time information is a parent node, and a node that receives time information is a child node. In the sequence diagram of FIG. 3, the operation sequence of the parent node and the child node is divided into an application unit (Appli) and a radio unit (RF). The application unit is a part that controls operation of the node, and the wireless unit is a module that executes wireless transmission / reception processing in response to an instruction from the application unit. The wireless unit represents a MAC layer (Media Access Control Layer) having a network hierarchical structure. Here, the application part (Appli) and the radio part (RF) are divided into one general way.

  In the parent node, a time synchronization request (Time-Sync-Request 301) is generated at a certain timing. Thereafter, the application section of the child node (Child Node) inquires about the presence / absence of data addressed to itself via the wireless section (Data-Polling 701). When receiving the data request message (Data-Request-CMD 601) from the child node by wireless communication, the wireless unit (RF) of the parent node immediately returns a confirmation signal (ACK 602) by wireless communication. When the data request message (Data-Request-CMD 302) is transferred from the wireless unit, the application unit (Appli) of the parent node acquires the current time T1 from the RTC held by itself (Get Time 401), The radio unit is instructed to transmit the time information T1 to the child node (Time-Sync-CMD (T1) 303). Then, the wireless unit of the parent node transmits a time synchronization command (Time-Sync-CMD (T1) 603) to the child node by wireless communication. The radio unit of the child node that has received the time synchronization command transfers the time synchronization command (Time-Sync-CMD (T1) 303) to the application unit, and the application unit sets its own time to T1 (Set Time (T1 801). In addition, the radio unit of the child node that has received the time synchronization command transmits a confirmation message (ACK602) to the parent node, and the radio unit of the parent node that has received the message has transmitted the time synchronization command to the application unit. A signal indicating that (Time-Sync-CMD-Sent 304) is transmitted.

  Here, when the child node attempts to transmit a data request message wirelessly and when the parent node attempts to wirelessly transmit a message including time information, the wireless transmission processing takes an uncertain time. This causes fluctuations in time synchronization accuracy. This uncertain duration is due to CSMA / CA processing.

  The outline of message transmission using CSMA / CA processing is as follows. A device attempting to transmit data first checks whether communication is being performed on the channel. When communication is detected, after waiting for a random length of time, starting again and checking for communication is repeated several times. If the channel is free, a message is transmitted. The number of repetitions is determined by the standard or implementation-dependent, so it is not uniquely determined. This number of repetitions is one of the causes of fluctuations in time required for time adjustment, that is, time synchronization accuracy.

  When one device tries to notify the other device of the current time, the waiting time is repeated by encoding the time in the application section of the transmission source and then actually detecting the free channel in the wireless section. It is a process that enters until it is sent out. There is no way of knowing this required time on the message receiving side, and correction on the receiving side is impossible. The time required for fluctuation is, for example, 0 seconds to several tens of milliseconds.

  Note that it is possible to encode the transmission time into a message after confirming that the channel is free by CSMA / CA processing, but this may cause other devices to use the channel while encoding. The original purpose of CSMA / CA cannot be achieved.

  The uncertain required time includes a procedure for confirming the use status of a radio channel called CCA (Clear Channel Assessment) in CSMA / CA. CCA is an operation procedure defined by wireless communication standards such as IEEE 802.11 and IEEE 802.15.4. The RF unit performs CCA when attempting to transmit data wirelessly (Check CCA 501), and if the wireless channel is not used as a result, wirelessly transmits data to be transmitted immediately after that, but the wireless channel is used. If this is the case, CCA is executed again after waiting for a random time. CCA is repeatedly executed until an unused state of a radio channel is detected or until a predetermined number of times is reached.

  In the upper half of FIG. 3, when the wireless transmission from the child node is performed, the CCA is used only once to detect the unused state of the wireless channel, and when the wireless transmission from the parent node is performed, the CCA is executed three times and the wireless channel is not used. The example which detected the use condition is shown. That is, how many times CCA is executed is not constant depending on the state of the radio channel space at that time. Also, there is usually no means by which this number can be determined on the receiving side. In the lower half of FIG. 3, the confirmation message (ACK 602) for the time synchronization command (Time-Sync-CMD (T2) 603) from the parent node cannot be confirmed by the parent node, and the parent node waits for the ACK signal. This shows a case where the same time synchronization command is retransmitted after the time is up (ACK Time-out 502).

  As in the cases of the upper half and the lower half of FIG. 3, the time synchronization command (Time-Sync-CMD) is sent from the parent node because the number of CCA processing executions is different or there is a difference in the presence or absence of command retransmission processing. 603) to send the time information to the child node (delay time Time Error) cannot be constant. Also, there is no way for the child node to know the delay time (Time Error) in either case of the upper half or the lower half of FIG. Therefore, when a plurality of sensor nodes individually try to synchronize time with the parent node (radio base station) by this procedure, at least an error due to variation in delay time is inevitable. The cycle of repeating the CCA is, for example, several milliseconds to several tens of milliseconds. For example, when the acceleration is sampled at a high frequency such as 100 Hz, this error exceeds the allowable range, and is required as a time synchronization means. Does not meet accuracy. In the case where the temperature is sampled at a frequency lower than 1 Hz, the conventional method shown in FIG. 3 can be used.

  Next, the problem of another conventional method will be described. A method in which a plurality of child nodes simultaneously receive a signal transmitted from a parent node and a plurality of child nodes execute time synchronization processing at the same time is because the child nodes perform time synchronization processing simultaneously with the same signal. Variations in the delay time may not be a problem. However, there is no effect if all the target child nodes are not in a receivable state when the parent node transmits a signal. However, in a sensor node that operates intermittently, it is difficult to constantly maintain radio reception from the viewpoint of power consumption, and the above method cannot be simply applied.

  FIG. 4 is a sequence diagram illustrating an example of a time synchronization method according to the present embodiment in which one parent node 110 simultaneously synchronizes two child nodes 210 and 220 at the same time. . Here, the parent node corresponds to the radio base station 110 or the repeater 160 of FIG. 1, and the child nodes correspond to the sensor nodes 210 to 250.

  In FIG. 4, since the sequence diagram is a layer one level higher than the second layer (MAC layer: Media Access Control layer) of the network hierarchy, the application part (Appli) and the application part (Appli) The radio part (RF) distinction is not expressed, and the Acknowledge signal is not described. This is omitted because there is no variation in delay time between child nodes even when CCA processing is performed.

  Furthermore, in both the parent node and the child node, the wireless unit (RF) performs encoding into a wireless communication frame at the time of transmission and decoding processing from the wireless communication frame at the time of reception.

  The child node operates intermittently at intervals determined for each child node. In one operation, the sensor is operated after activation to observe acceleration (sensing 802), and the radio unit is operated (RF- working 803), immediately after the observation data is transmitted to the parent node (Data-Sending 604), a data request message (Data-Req-CMD 601) is subsequently transmitted to the parent node.

  On the other hand, when a time synchronization request (Time-Sync-Request 301) is generated at a certain timing, the parent node 110 transmits the next timing for broadcasting the time synchronization signal (timing when the child node shifts to the next operation state) to the child node 110. Predetermined based on the operation interval of the node (Decide Time To Sync, 402). This time synchronization request is periodically generated by a timer, or is generated in response to a time synchronization instruction from a higher system (for example, the management server 30 in FIG. 1). The parent node 110 receives a data request message (Data-) from one child node (Child Node # 1) 210 in a state where a time synchronization request is generated (that is, the timing for broadcasting the time synchronization signal is determined). Req-CMD 601) is received, the time synchronization timing is calculated as a relative time (T1) from that point (Calculate Time 403), and the time (T1) notification message (Wakeup-later-CMD (T1)) 605) to the child node # 1.

  The child node 210 that has received the relative time (T1) sets a timer so as to be activated after the time (T1) has elapsed, and shifts to a dormant state.

  Similarly, when another child node (Child Node # 2) 220 transmits a data request message (Data-Req-CMD 601) after transmitting observation data (Data Sending 604), the parent node 110 broadcasts a time synchronization signal. A message (Wakeup-later-CMD (T2) 606) notifying the transmission timing as a relative time (T2) from that point is transmitted to the child node 220, and the child node 220 is activated after the time (T2) has elapsed. In this way, the timer is set to enter the sleep state.

At the timing when the parent node broadcasts the time synchronization signal (Time-Sync-CMD (T) 607), all the child nodes (Child Node # 1 210 and Child Node # 2 220 in FIG. 4) are in the operating state. The child nodes that have shifted to and are ready for wireless reception and have received the time synchronization signal adjust their time to time T all at once (Set Time (T)).
Although an example in which the parent node calculates the relative time from the time when the data request message is received from the child node has been described, the relative time from the time when the sensor data is received may be calculated. That is, when the parent node receives the sensor data and transmits the relative time, the child node can perform time adjustment without transmitting the data request message. Thereby, further reduction of power consumption can be aimed at.

In this way, a timing for performing time synchronization in the parent node is determined, and a message for notifying each child node of the timing is introduced in advance. Then, the time synchronization signal is broadcasted simultaneously to a plurality of child nodes performing intermittent operation. As a result, low power consumption of the child nodes can be maintained, and there is no variation in delay time between the child nodes, so that highly accurate time synchronization can be performed among a plurality of child nodes.
In addition, assuming that each child node's microcomputer operates at an operating clock of several MHz, there are several differences in the timing of actually adjusting the child node's own time after receiving the time synchronization signal. Since it is about a cycle (several microseconds), it can be sufficiently suppressed within 1/200 second (5 milliseconds).

  In the message (Wakeup-later-CMD 605 and 606) in which the parent node notifies the relative time to the child node, the time (T) at which the time synchronization signal is broadcasted may be notified in advance. By doing so, it is possible to perform the time information decoding process in advance when receiving the time synchronization signal (Time-Sync-CMD 607), and to reduce the delay with the parent node when setting the time. Is possible.

  In addition, there is usually an individual difference in the clock module (electronic component consisting of a crystal resonator and an oscillation circuit) that determines the operation timing of the RTC that ticks the time, so the above time synchronization processing is performed when a predetermined interval has elapsed. It is desirable to carry out. The predetermined interval here is different from the sensing interval performed by the child node, and when the sampling frequency of the acceleration sensor is 100 Hz, it is possible to maintain the accuracy with which the sampling period of 10 msec can be correctly distinguished among multiple sensor nodes. It is an interval.

  The interval for executing this time synchronization processing is determined in advance based on the accuracy of the clock module of the child node, and parameters are incorporated into the system. For example, from the specifications related to the accuracy of the clock module used as a part by the sensor node (usually described in the data sheet of the product), the time when the timer deviation can be up to 10 msec is obtained, and an appropriate time shorter than that is obtained. It is possible to generate Time-Sync-Request at time intervals. Generally, this interval is several hours, and it is considered that time synchronization processing is required several times a day.

  In FIG. 4, the number of child nodes is two in order to simplify the description. Needless to say, the number is not limited to two.

  Next, a method for determining the time T at which the parent node broadcasts the time synchronization signal (Time-Sync-CMD (T) 607) will be described in detail. As described with reference to FIG. 1, (1) a parent node manages a child node connected to itself with a communication ID (included in a communication frame). (2) Each child node is intermittently operated at intervals determined for each child node. (3) The parent node knows the interval (operation cycle) of each intermittent operation of all the child nodes connected to itself (as described above, the interval of intermittent operation is a management having a configuration management function. The server may be inquired, or the parent node may manage it).

  At this time, the time T is determined by the following procedure. When a time synchronization request occurs, the parent node obtains the maximum value of the operation cycle of all the child nodes connected to itself, and after the elapse of the time obtained by adding an appropriate margin time to the maximum time from the current time. The timing is set as the transmission timing T of the time synchronization signal (Time-Sync-CMD 607).

  In addition, when the sensing process by the sensor and the timing of receiving the time synchronization signal overlap on the child node side, it is preferable to prioritize the time synchronization process because the value of the sensing data is lowered when there is a time lag.

  Next, a second embodiment in which a time synchronization verification step is added to the first embodiment will be described.

FIG. 5 is a sequence diagram illustrating the operation of the second embodiment. The difference from FIG. 4 is that the child nodes # 1 and # 2 have adjusted the time by the time synchronization signal immediately after receiving the time synchronization signal (Time-Sync-CMD (T) 607) from the parent node. The time adjustment notice shown is transmitted. Specifically, the adjusted time T and the time lag (Δt1 and Δt)
t2) (Data-sending (T, Δt1) 608, Data-sending (T, Δt2) 609).

  Thereby, the parent node can verify whether or not all the child nodes managed by the parent node have been time synchronized. Therefore, the reliability of the system is increased. For example, when the time adjustment notification (corresponding to 608 or 609) does not arrive from some of the child nodes (when the predetermined time set in advance is not received), the parent node broadcasts the time synchronization signal (607). You may reset the time and execute the procedure from the beginning. Thereby, the reliability of time synchronization can be improved.

In addition, the parent node notifies the management server 30 of the result, and the management server can manage these time synchronization situations. An example in which the management server manages the time synchronization status of the sensor node using the time synchronization state table will be described with reference to FIG. In FIG. 6, “Sync-Failed”, “Latest Sync-Time Reported”, and “Latest Time” are associated with the sensor node IDs.
"Error" is recorded, "ID" represents the identifier of the sensor node, "Sync-Failed" is the number of times each sensor node has failed in time synchronization, and "Latest Sync-Time Reported" is each sensor node “Latest Time Error” indicates the time difference when adjusting to “Latest Sync-Time Reported”.

  Next, the procedure for creating and updating the table in FIG. 6 will be described. When creating a table, entries corresponding to all sensor nodes included in the system are created as part of system configuration management. When a Time-Sync-Request 301 is generated during system operation, the “Sync-Failed” values of all target sensor nodes are incremented by +1. When the management server receives the sensor node time adjustment notification via the parent node, the value of “Sync-Failed” is reduced by 1 and “Latest Sync-Time Reported” and “Latest Time Error” are updated.

  The management server may manage the table of FIG. 6 as part of the system configuration management information. The management server can check the reliability of the time information added to the observation data collected from each sensor node by referring to the table shown in FIG. 6, and can determine whether the application can use the data. Can be used for For example, when Sync-Failed is larger than a predetermined value, it may be determined whether to use observation data with a time stamp after Last Sync-Time Reported as analysis data. Specifically, the management server checks the time synchronization accuracy as a request of the application, and when the value of “Sync-Falied” is 1 or more and the value of “Latest Time Error” is required When the value is larger, the management server analyzes without using observation data having a time stamp after “Last Sync-Time Reported”. This has the effect of guaranteeing data reliability.

  Furthermore, using the distribution of “Last Time Error” values in the time synchronization state table of FIG. 6, it is possible to optimize the interval of periodic time synchronization processing for maintaining the required time synchronization accuracy. Become. That is, after performing a test operation for a while while performing time synchronization processing periodically, check the value of “Time Error” of each sensor node, and if the value is larger than the time accuracy required by the system, It can be considered that the interval for executing the process is adjusted to be shortened, and conversely if the value of “Time Error” is smaller than the required time accuracy, the interval of the time synchronization process is adjusted to be increased.

  If the parent node has a function to synchronize with the standard time using, for example, a GPS (Global Positioning System) or a radio clock, the time synchronization accuracy that GPS or radio clock can provide can be synchronized with absolute time. Can be raised. By doing so, it is possible to improve time synchronization accuracy between sensor nodes in a wider range (for example, a multi-hop range).

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made, and it is possible to appropriately combine the above-described embodiments. It will be understood by the contractor.

  The present invention can be used to collect data for response analysis of a structure such as an earthquake. This is particularly effective when it is difficult to install a wired sensor.

Example of sensor network system configuration diagram Example of hardware configuration diagram of radio base station and sensor node Example of hardware configuration diagram of wireless repeater An example of a sequence diagram of a time synchronization process in which a parent node and a child node are one-to-one in the conventional example Example of sequence diagram of time synchronization processing of embodiment 1 Example of sequence diagram of time synchronization processing of embodiment 2 An example of a time synchronization state table of the second embodiment

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Structure 20 ... Network 30 ... Management server 40 ... Client terminal 110 ... Wireless base station 160 ... Wireless repeater 210, 220, 230, 240, 250 ... Sensor node

Claims (8)

A node time synchronization method in a sensor network system comprising a first child node, a second child node, and a parent node that performs wireless communication with the first and second child nodes,
The first child node, a first operating state and repeatedly to intermittent operation and the first resting state, transmits to obtain the first sensor data in the first operation state to said master node And
The second child node, a second operating state and repeatedly to intermittent operation and a second sleep state, transmitted in the second operating state to obtain the second sensor data to the parent node And
The parent node is
The timing of broadcast time synchronization signal based on the operation interval of the intermittent operation of the first and the second child node to determine,
When receiving the first and second sensor data from the first and second child node transmits a signal including the timing of broadcasting the time synchronization signal the determined to the first and second child node And
The first child node transitions from the first hibernation state to the third operation state based on the timing at which the parent node transmits the time synchronization signal,
The second child node transitions from the second hibernation state to the fourth operation state based on the timing at which the parent node transmits the time synchronization signal,
The parent node broadcasts the time synchronization signal at a timing to transmit the time synchronization signal,
The first child node adjusts its own time to a time included in the broadcast time synchronization signal received in the third operation state,
The node time synchronization method in which the second child node adjusts its own time to a time included in the broadcast time synchronization signal received in the fourth operation state. The node time synchronization method according to claim 1,
The timing of transmitting the time synchronization signal received by the first child node is a time indicating the difference between the time of receiving the first sensor data and the time of transition to the third operation state,
The timing at which the time synchronization signal received by the second child node is transmitted is a node time that indicates the difference between the time at which the second sensor data is received and the time at which the fourth operation state is shifted to. Synchronization method. The node time synchronization method according to claim 1,
The first child node transmits a first data request command following transmission of the first sensor data,
The second child node transmits a second data request command following transmission of the second sensor data,
The timing of transmitting the time synchronization signal received by the first child node is a time indicating a difference between the time of receiving the first data request command and the time of transition to the third operation state,
The timing at which the time synchronization signal received by the second child node is transmitted is a time indicating the difference between the time at which the second data request command is received and the time at which the fourth operation state is shifted to Time synchronization method. The node time synchronization method according to claim 1,
The parent node calculates the maximum value of the operation cycle of the intermittent operation of the first and second child nodes to be connected, and the time after the elapse of time obtained by adding a margin time to the maximum value from the current time is the time The node time synchronization method according to claim 1, wherein the synchronization time is transmitted . The node time synchronization method according to claim 1,
When the parent node has passed a predetermined interval based on the clock accuracy of the first and second child nodes and receives the first and second sensor data, A node time synchronization method for transmitting a signal indicating a timing for transmitting a synchronization signal. The node time synchronization method according to claim 1,
The node time synchronization method, wherein the parent node has a function of synchronizing its own time with a standard time. A first operating state and repeatedly to intermittent operation and a first dormant transmitted, a first sensor for acquiring the first sensor data in the first operating state, the first sensor data A first wireless node having a first child node,
The second operating state and repeatedly to intermittent operation and a second dormant transmitted, a second sensor for acquiring the second sensor data in the second operating state, the second sensor data A second radio node having a second radio unit,
A control unit for determining a timing of broadcast transmission of a time synchronization signal based on an operation interval of the intermittent operation of the first and second child nodes, and the first and second child nodes from the first and second child nodes . A parent node having a transmission / reception unit that transmits to the first and second child nodes a signal including a timing for transmitting the time synchronization signal determined when the sensor data of 2 is received,
The first child node transitions from the first hibernation state to the third operation state based on the timing at which the parent node transmits the time synchronization signal,
The second child node transitions from the second hibernation state to the fourth operation state based on the timing at which the parent node transmits the time synchronization signal,
The transmission / reception unit of the parent node broadcasts the time synchronization signal at the timing of transmitting the time synchronization signal,
The first child node adjusts its own time to a time included in the broadcast time synchronization signal received in the third operation state,
The sensor network system, wherein the second child node adjusts its own time to a time included in the broadcast time synchronization signal received in the fourth operation state. The sensor network system according to claim 7 ,
A sensor network system including a time included in the time synchronization signal in a signal indicating a timing of transmitting the time synchronization signal.

Images