KR100819037B1 - Method of time synchronization for wireless sensor network and sensor node - Google Patents

Abstract

A time synchronization method for a wireless sensor network and a sensor node are provided to synchronize a node in a propagation distance of a packet regardless of a network topology, thereby reducing a trouble of the node. A sensor node comprises a node synchronization part and a packet transformation part. When a packet having a time stamp is transmitted from a first node which is time-synchronized with a reference node(S100), the node synchronization part determines whether time-synchronization is performed or not based on a synchronization period in the transmitted packet, and performs the time synchronization by extracting the time stamp of the transmitted packet(S200). The packet transformation part produces a time stamp and transmits the produced time stamp and the packet having the synchronization period to a second node in a propagation distance(S300).

Description

Method of Time Synchronization for Wireless Sensor Network and sensor node

1 is a block diagram of a wireless sensor network according to an embodiment of the present invention;

2 is a schematic diagram of a sensor node according to an embodiment of the present invention;

3 is a flowchart illustrating a time synchronization method according to an embodiment of the present invention;

4 is a flowchart illustrating a time synchronization method according to an embodiment of the present invention incorporating a synchronization step in the time synchronization method of FIG. 3;

5 and 6 are diagrams illustrating a time synchronization method according to an embodiment of the present invention.

The present invention relates to the field of wireless sensor networks, and more particularly, to a method and a sensor node of the time synchronization of the nodes constituting the sensor network.

Nodes in wireless sensor networks are battery-powered, low-power, low-cost systems that have a limited lifespan with current battery technology and cannot be recharged or replaced even if the nodes stop running because of battery drain. Therefore, wireless sensor networks must be able to efficiently manage power usage, thereby maximizing network life.

In addition, the nodes constituting the sensor network have their own time information, and in order to express specific information, the time information of some or all nodes needs to be synchronized.

Therefore, in a wireless mobile communication system, GPS is generally installed in a system for time synchronization. However, in a low power-oriented network such as a sensor network, many functions including energy are limited as compared to the nodes of a conventional wireless network. In most cases, there are many limitations to using existing synchronization methods that use expensive, high-power equipment such as GPS.

Therefore, a number of synchronization methods suitable for sensor networks have been studied, and representative synchronization methods include time stamps such as Reference Broadcasting Synchronization (RPS), Timing-Sync Protocol for Sensor Networks (TPSN), and The Flooding Time Synchronization Protocol (FTSP). There is a time synchronization method through exchange.

RBS (J. Elson, L. Groid, D. Esttrin, "Fine-Grained Network Time Synchronization using Reference Broadcasts", 2002) is a representative scheme with a receiver-receiver synchronization structure. When one sender broadcasts a beacon, a plurality of receivers record each time the beacon is received and compare them with each other for time synchronization. In this method, as the number of nodes constituting the network increases, packet overhead increases rapidly. In case of multi-hop, the time synchronization must be converted through the connection node using the time-routing method.

TPSN (S.Ganeriwal, R. Kumar, M. Srivastava, "Timing-Sync Protocol for Sensor Networks", 2003) is a representative scheme with a sender-receiver synchronization structure. A sender uses a traditional hierarchical two-way synchronous delivery method that delivers motivation to the next receiver. Therefore, the greater the number of children of a node, the greater the number of sync packet transmissions that a parent node must send and receive. In addition, it has a disadvantage in that it cannot dynamically respond to changes in topology of the network.

FTSP (Miklos Maroti, Branislav Kusy, Gyula Simon, Akos Ledeczi, "The Flooding Time Synchronization Protocol", 2004) is also a method with a sender-receiver synchronization structure. This method uses synchronous sequences to prevent redundant propagation of synchronization packets and is inefficient by using multiple timestamps in a packet.

Therefore, there is a need for an efficient time synchronization method that can solve the disadvantages of the conventional methods as described above and dynamically cope with changing network configurations.

An object of the present invention is to provide a sensor node capable of time synchronization regardless of the network configuration at low power.

Another technical problem to be achieved by the present invention is to provide a method for efficiently synchronizing time information of some or all nodes at low power regardless of network configuration in a sensor network sensitive to power consumption.

Another technical problem to be solved by the present invention is to provide a computer readable recording medium having recorded thereon a program for executing on a computer a method of efficiently synchronizing time information of some or all nodes in a sensor network sensitive to power consumption. There is.

Other objects and advantages of the present invention can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. It will also be readily appreciated that the objects and advantages of the invention may be realized by the means and combinations thereof indicated in the claims.

In order to achieve the above technical problem, an embodiment of a sensor node according to the present invention includes: a packet transmitter for generating a time stamp and transmitting a packet including the time stamp to another node within a propagation radius; And a node synchronization unit for determining whether to perform time synchronization from a packet received from another node, extracting a time stamp, and time synchronizing.

In order to achieve the above technical problem, an embodiment of a time synchronization method of a wireless sensor network according to the present invention includes a method in which a reference node generates a time stamp and transmits a packet including the time stamp to nodes within a propagation radius. step; Determining whether each node within a propagation radius of the reference node performs time synchronization from a packet received from the reference node and time-synchronizing based on a timestamp in the received packet; And each time-synchronized node generates a new timestamp of its own and transmits the packet including the newly generated timestamp to nodes within its respective propagation radius.

In order to achieve the above technical problem, the present invention provides a computer-readable recording medium having recorded thereon a program for executing the time synchronization method of the sensor network of the present invention on a computer.

Hereinafter, with reference to the accompanying drawings will be described in detail with respect to the present invention.

1 is a block diagram of a wireless sensor network according to an embodiment of the present invention.

Referring to FIG. 1, the wireless sensor network of the present invention includes a plurality of sensor nodes, each node including its own reference node R has its own propagation radius, and includes a plurality of nodes within its propagation radius. . Since the wireless sensor network of the present invention is not limited to a specific network configuration, for example, formation of a mesh or tree structure is not required.

FIG. 2 schematically illustrates one sensor node according to one embodiment of the invention in the wireless sensor network of FIG. 1.

Referring to FIG. 2, the sensor node 100 includes a packet transmitter 102 and a node synchronizer 104 for time synchronization, and a temperature and pressure at a point where a node is installed through a sensing unit (not shown). Acquire surrounding environment information such as vehicle speed. The obtained information is processed and converted into data suitable for communication with other nodes. In order for the converted data to be properly communicated with each other, it is important that synchronization is performed between sensor nodes.

The packet transmitter 102 generates its own timestamp and inserts the timestamp into a packet to transmit a synchronization message to other nodes within its propagation radius. The generated time stamp is preferably inserted into a PHY frame of a packet to reduce synchronization error in transmission. The packet transmitter 102 may additionally set a synchronization period and insert the packet with the time stamp in the packet to be transmitted.

The node synchronizer 104 extracts a timestamp from a packet including a timestamp received from another node in the vicinity and processes the timestamp to synchronize its internal time, and synchronizes the timestamp with the timestamp of the received packet. An analysis unit 106 for analyzing the conditions for determining whether or not to include.

The analyzer 106 determines whether the packet has been received from the other node for the first time, whether or not the packet has been received from the other node after a time elapsed by a synchronization period from the latest synchronized time, and the latest synchronization. It is determined whether the packet received from the other node includes a new shortened synchronization period before the synchronization period elapses from the time.

When the packet is transmitted from the other node by the analyzing unit 106 for the first time, when the packet is received from the other node after a lapse of a synchronization period from the latest synchronized time, the latest synchronization If a packet is received from the other node before the synchronization period has elapsed from one time point, but it is determined that the received packet includes a shortened new synchronization period, the node synchronizer 104 receives the packet from the other node. It synchronizes its internal time based on the time stamp of the received packet.

As a method of time synchronization using the time stamp extracted from the packet, the following method of synchronizing based on one or more time stamps extracted from one or more packets may be used. In addition, a known method that may be used for time synchronization may be used. Several methods may be considered:

i) A fast time synchronization method of setting the received time stamp as it is, but synchronizing by compensating for the time delay if known.

ii) after storing a certain number of recently received sync messages,

-A method of synchronizing with the average of time difference from the received synchronization timestamp compared to the internal clock of the user (Simple Average).

In the equation y = ax + b, x is a time when the packet is received and y is a reference time of the received timestamp to derive the a and b values, and the values are repeated to calculate the average of a and b values. How to sync (Linear Regression).

식과 Y_j=aX+b 식에서, X에는 패킷을 받은 시간, Yj에는 받은 타임스탬프의 기준 시간을 넣고 E를 최소화하는 a,b를 도출하여 이를 기반으로 Y_j=aX+b를 하여 동기화하는 방법(Least Square).-

In the equation and Y _j = aX + b, the method of synchronizing Y _j = aX + b based on the result of deriving a, b to minimize the E and inserting the reference time of the received time stamp in X and the time stamp received in Yj (Least Square).

Hereinafter, a method of time synchronization using the time synchronization device will be described in detail.

3 is a flowchart illustrating a time synchronization method according to the present invention, and FIG. 4 is a flowchart illustrating the synchronization step of the time synchronization method according to the present invention in FIG. 3 in detail.

Referring to the drawings, in the time synchronization method according to the present invention, the reference node R generates a timestamp and transmits the packet including the timestamp to nodes within a propagation radius (S100). Time synchronization of each node based on the time stamp in the packet received from the reference node R in the propagation radius of S200 and each synchronized node generates its own timestamp in its propagation radius. And transmitting to the nodes (S300).

In step S100, the reference node R generates a timestamp for the reference time and transmits it in a packet to other nodes within the propagation radius. The transmitted packet may further include a synchronization period P in addition to a time stamp providing a reference time. Hereinafter, a packet including a time stamp (and synchronization period) is also called a synchronization message.

In general, the synchronization error increases as time passes from the synchronization point. Therefore, when the maximum synchronization error is determined by the requirements of a specific application (for example, the speed and direction of an object moving at 5 km / h are known within a specific error range). Nodes in the network must be synchronized to within a certain margin of error), and synchronization cycles rather than FTSPs using synchronization round numbers as a synchronization redundancy condition enable effective synchronization without additional information. There is an advantage.

Therefore, when the synchronization period is used, for example, when the node knows the maximum error such as delay or drift due to the system characteristics of the node itself, the synchronization period does not exceed the maximum synchronization error. It is also possible to use the appropriate configuration method for the system (see “Cluster-based hierarchical time synchronization for multi-hop wireless sensor networks, IEEE AINA 2006, Kim Hyunhak, etc.”).

The time stamp is preferably inserted into a PHY frame of a packet in order to reduce synchronization delay error and increase synchronization performance. When using a MAC technology of the same type as CSMA / CA, the probability of packet collision increases as the back of the packet increases, so inserting a timestamp into the PHY frame of the packet can reduce packet overhead by reducing the loss of timestamp due to packet collision. have.

In step S200, each node within the propagation radius of the reference node R that has received the packet from the reference node R extracts a timestamp from the received packet and modifies and synchronizes the internal time of the node after processing.

Hereinafter, the synchronization step S200 will be described in detail with reference to FIG. 4.

The node receiving the packet including the time stamp generated by the reference node R and the synchronization period P determines whether the node is received for the first time through the analyzer 106 (S210). When the packet is first received, the receiving node performs time synchronization (S250) by using the time stamp in the received packet by the node synchronization unit 104.

If it is not the first packet reception, the receiving node determines whether the reception time has been received after the synchronization period P has elapsed from its most recently synchronized time A through the analysis unit 106. Determine (S220). If it is determined that the time point of receiving the packet transmitted from the reference node R again after the time of the synchronization period P has elapsed from the time of the most recently synchronized time A, the node synchronization unit 104 determines that the received packet has been received. Extracting the time stamp from it synchronizes the internal time therefrom (S250).

If the reception time is not the first packet reception and the reception time is before the synchronization period P elapses from the most recently synchronized time A, the receiving node receives the reference received through the analysis unit 106. It is determined whether the packet from the node includes the shortened synchronization period P '(S230). If the determination result includes a shortened synchronization period P ′, the node synchronization unit 104 extracts a time stamp from the received packet and synchronizes an internal time therefrom (S250). Otherwise, the packet is ignored and not synchronized.

The synchronization according to the shortened synchronization period P ′ may be synchronized regardless of the time elapsed from the latest synchronized time, and the packet is received after the shortened synchronization period P ′ has elapsed from the latest synchronized time. You can also synchronize only in one case.

Through this process, the synchronization message of the same period can be synchronized and repeated many times without the need for a node in the period, thereby preventing propagation of synchronization to other nodes.

Next, the nodes synchronized through the above steps respectively generate their own time stamps and transmit a packet including the same to the nodes within their respective propagation radius (S300).

Steps S200 and S300 are repeated for the entire network or for a part of the network, so that synchronization messages are propagated and synchronization is performed. Therefore, since the present invention propagates and synchronizes packets to nodes within a propagation radius of any node, synchronization is possible regardless of the network configuration.

5 and 6 are simplified diagrams illustrating a method of time synchronization of nodes in a wireless sensor network in accordance with the present invention. Hereinafter, an embodiment of a time synchronization method of nodes in a wireless sensor network will be described with reference to the drawings. In this embodiment, it is assumed that the reference node R starts synchronizing for the first time, and among the plurality of nodes of FIGS. 5 and 6, R, A1, A2, A3, A4, B1, and B2 are representative of the nodes of FIGS. Only B3, B4, B5, and B6 will be described as an example.

First, the reference node R generates a timestamp and a synchronization period (for example, 5 minutes) for the reference time, and puts it in a packet to other nodes (for example, A1, A2, A3, and A4) within a propagation radius. send.

Since the nodes A1, A2, A3 and A4 receive the transmitted packet for the first time, the nodes A1, A2, A3 and A4 modify their internal time by processing the time stamp in the packet. Synchronize. In this case, as described above, various methods may be used.

Next, each of the synchronized nodes A1, A2, A3, and A4 generates their own timestamp and synchronization period (5 minutes) for the nodes that have not yet received the packet including the reference time to generate a packet including the same. Transmit to each other node within its own radius of propagation. Node A1 is at node R, node A2 is at node R, A3, B4, B5, B6, node A3 is at node R, A2, A4, B3, B6, node (A4) transmits a packet to the nodes R, A3, B1, B2, and B3, respectively.

Since the nodes R, A2, A3, and A4 receive the same synchronization message within the same synchronization period (5 minutes) from the previous synchronization time, they ignore the received packet again and do not synchronize again. Since nodes B1 to B6 receive the packet for the first time and receive a synchronization message, the nodes B1 to B6 extract the time stamp in the packet and synchronize their internal time based on this after appropriate processing. In this case, as described above, various methods may be used.

The synchronized nodes B1 to B6 likewise generate their time stamps and synchronization periods (5 minutes) after internal time synchronization and transmit packets containing them to other nodes within their respective propagation radius.

Synchronization is repeated with this timestamp generation and transmission step so that synchronization is propagated throughout the network or in part.

Then, when the reference node R wants to synchronize newly, the synchronization starts after the synchronization period (5 minutes) from the latest synchronization start time, or the synchronization period shortened within the synchronization period (5 minutes) from the latest synchronization start time (for example, , 3 minutes).

Nodes within a propagation radius that have received a synchronization message after the second most recent synchronization period (5 minutes) from the reference node R synchronize using a timestamp from the second transmitted packet.

If nodes within the propagation radius have received a synchronization message with a shortened synchronization period (3 minutes) within a synchronization period (5 minutes) from the latest synchronization start time, the timestamp from the received packet is not ignored without ignoring the received packet. Synchronize using. At this time, the receiving nodes extract time stamps from the shortened synchronization period packets only when the shortened synchronization period (3 minutes) is received after the shortened synchronization period (3 minutes) has elapsed from the time of the latest synchronization. Synchronizing or receiving a packet of the shortened synchronization period regardless of whether a shortened synchronization period (3 minutes) has elapsed from a recently synchronized time may extract and synchronize a time stamp therefrom. In this case, as described above, various methods may be used.

On the other hand, if the nodes are aware of the maximum error such as delay or drift due to the above-described system characteristics, use a method of properly setting the synchronization period within the range not exceeding the maximum synchronization error. If the node changes itself to an appropriate synchronization period (2 minutes), as described above, if a synchronization message of a synchronization period (3 minutes) is received after its synchronization period (2 minutes), or if its own synchronization period (2 When a synchronization message of a shortened synchronization period (1 minute) is received within minutes), the received packet is synchronized using a time stamp from the received packet without ignoring the received packet. At this time, the synchronization period generated again may be a synchronization period (3 minutes or 1 minute) of the received synchronization message or its own synchronization period (2 minutes).

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

So far, the present invention has been described with reference to preferred embodiments. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

Since the present invention only needs to transmit n packets for n nodes within the propagation radius, it consumes packets rapidly in proportion to the number of nodes in the synchronization beacon reach (e.g., n ² when the number of nodes in range is n ^2). Increasing the number of synchronization packets that a parent node must respond and transmit (for example, 2n + 1 packets need for n child nodes) in proportion to the number of RBSs or child nodes it needs to send It is possible to achieve wide time synchronization more efficiently than FTSP which uses TPSN or synchronization cycle as a synchronization condition and timestamps one packet several times.

In addition, since the present invention can provide time synchronization regardless of a specific network topology, it is not necessary to establish a synchronization path in advance like TPSN, so that time synchronization can be provided in any structure such as a mesh or a tree.

In addition, the present invention is resistant to node failure because it synchronizes nodes within the propagation distance of a packet regardless of the network configuration.

In addition, since the present invention uses a synchronization cycle, the nodes that propagate the time synchronization do not overlap in repetitive time synchronization packet retransmission, so that time synchronization is effectively propagated throughout the network and customized synchronization according to the synchronization requirements can be effectively performed. do.

In addition, time stamps can be inserted into PHY frames to reduce synchronization delay errors for improved synchronization performance.

Claims (14)

Receiving a packet including a timestamp from a first node time-synchronized to a reference node, and determining whether to perform time synchronization based on a synchronization period included in the received packet to avoid duplicate synchronization with the reference node, A node synchronizer configured to time-synchronize the timestamp of the received packet; And And a packet transmitter for generating a timestamp and transmitting a packet including the generated timestamp and the synchronization period to a second node within a propagation radius. delete The method of claim 1, wherein the node synchronization unit, Whether the packet has been received for the first time from the other node, whether the packet has been received from the other node after a time elapsed from the latest synchronization time, and the time for the synchronization period from the latest time of synchronization. And a analyzing unit for determining whether a packet received from the other node includes a shortened new synchronization period before the elapsed time. The method of claim 1, wherein the node synchronization unit, And synchronize based on one or more timestamps extracted from one or more packets. The method of claim 1, wherein the time stamp, Sensor node, characterized in that inserted into the PHY frame of the packet. Generating, by a reference node, a timestamp and transmitting a packet including the timestamp to nodes within a propagation radius; It is determined whether each node within the propagation radius of the reference node performs time synchronization based on the synchronization period included in the packet transmitted from the reference node to avoid duplication synchronization with the reference node, and time stamp in the received packet. Time synchronizing based on the time synchronization; And Wherein each time-synchronized node generates a new timestamp of its own and transmits the packet including the newly generated timestamp and the synchronization period to nodes within its respective propagation radius; How to synchronize time on your network. delete The method of claim 6, wherein the synchronizing step, And when each node within a propagation radius of the reference node receives a packet from the reference node for the first time, synchronizing time based on a time stamp in the packet received from the reference node. Time synchronization method. The method of claim 6, wherein the synchronizing step, When each node within the propagation radius of the reference node is transmitted after a time interval of a synchronization period has elapsed from the time when the packet was recently synchronized from the reference node, the time is based on the timestamp in the packet received from the reference node. And synchronizing further comprising: synchronizing. The method of claim 6, wherein the synchronizing step, When each node within the propagation radius of the reference node is transmitted from the reference node before the time of the synchronization period elapses from the time when the packet was recently synchronized, the packet received from the reference node includes a shortened synchronization period. And synchronizing the time based on the time stamp in the received packet. The method of claim 6, wherein the synchronizing step, Synchronizing based on one or more timestamps extracted from one or more packets. The method of claim 6, wherein the time stamp, Time synchronization method of a wireless sensor network, characterized in that inserted into the PHY frame of the packet. A computer-readable recording medium having recorded thereon a program for executing the method of any one of claims 6 to 12. Receiving a packet including a timestamp from a first node synchronized with a reference node, and determining whether to perform time synchronization based on a synchronization period included in the received packet to avoid duplex synchronization with the reference node, Extracting time stamps in the received packets and time synchronizing them; And Generating a new self timestamp after the time synchronization and transmitting a packet including the newly generated timestamp and the synchronization period to second nodes within a propagation radius of the node in the wireless sensor network. Time synchronization method.

Images