KR100918841B1 - A Head Node Election Method for Clustering in Wireless Seneor Network and thereof Wireless Seneor Network - Google Patents

Abstract

A head node selection method for clustering a wireless sensor network according to the present invention is a head node selection method for clustering a wireless sensor network including a sync node and a plurality of sensor nodes, and each of the plurality of sensor nodes determines an effective head node state. A first step of making; A second step of, according to the determination result, sensor nodes which are in an effective head node state broadcasting a head node selection request message including a distance ratio (EDR) of an energy to a sink node within a predetermined range; Receiving the second stage broadcast and comparing the EDRs of the sensor nodes in the other valid node states with their EDRs based on the received broadcast to determine whether to maintain the valid head node state. Three steps; And a fourth step of selecting the sensor nodes that maintain the valid head node state determined in the third step, as the head node.

Description

A Head Node Election Method for Clustering in Wireless Seneor Network and YUL Wireless Seneor Network

The present invention relates to a method for selecting a head node for clustering a wireless sensor network and a wireless sensor network thereof. In particular, when the wireless sensor network is clustered, energy consumption-intensive head nodes in a data transmission step are possible for a long time. A method for selecting a head node for clustering a wireless sensor network to operate during the same and a wireless sensor network therefor.

The present invention is derived from the research conducted as part of the IT growth engine technology development project of the Ministry of Information and Communication and the Ministry of Information and Communication Research and Development. [Task Management Number: 2006-S-026-02, Title: URC Server Framework for Active Services] Development of the URC Server Framework for Proactive Robotic Services].

Recently, due to the development of wireless communication and electronics technology, a network between low-cost and ultra-small sensors is enabled, and a network composed of these sensors is called a wireless sensor network.

Wireless sensor network is used for ecological environment monitoring, intelligent environment monitoring, location recognition service, intelligent medical system, and intelligent robot system, and is developing as a core technology of ubiquitous computing environment.

The wireless sensor network detects, measures, and floods physical data such as light, sound, temperature, and motion in the physical space, and a central primary node (sink node, It is a network composed of sink-nodes.

Sensor nodes typically include one or more sensors, actuators, microcontrollers, tens of kilobytes of EEPROM, several kilobytes of SRAM, hundreds of kilobytes of flash memory, analog-to-digital converters (ADCs), short-range wireless communication modules, and the like. Because of the limited resources of the sensor node, the life of the wireless sensor network depends entirely on the life of the sensor node because it consists of a power supply (energy source) for powering the components.

As such, sensor nodes in a wireless sensor network have limited energy, and since they are usually located in areas where battery replacement or management is impossible, a method for reducing energy consumption of the sensor node is required to extend the life of the sensor network.

Recently, clustering has emerged as a method of minimizing energy consumed in a wireless sensor network. Clustering is to divide a wireless sensor network into a plurality of clusters and to share the roles of nodes existing in the divided cluster.

As a conventional method for reducing energy consumption of a sensor node in a wireless sensor network, a cluster is composed of one head node (HN) and a plurality of member nodes (MN), and the head node is a member node. It consumes a lot of energy by repeating the role of transmitting the data collected from the data to the sink node (SN).

For example, representative examples of the clustering method that have already been proposed include the Least ID clustering method, the Linked clustering method, and the TopDisc (Topology Discovery) method. The Least ID clustering method is a method of performing clustering based on IDs of nodes, and the linked clustering method is a method in which unique slots are allocated to nodes in a high ID order. The Least ID clustering method and the linked clustering method have the advantage of being relatively simple, but the energy consumption of the wireless sensor network is increased because HN, which is a high energy consumption node, is generated.

1 is a diagram illustrating a head node selection in a wireless sensor network according to the related art. As shown in FIG. 1, the TopDisc method is a clustering method of maximizing the distance between the HN 1 and the MN 2 to minimize the number of HN 1 nodes, which are energy-consuming nodes. This is based on the assumption that propagation delay time is proportional to the distance between nodes. Specifically, in the TopDisc method, the HN 1 broadcasts a Topology Discovery message to the nodes 2 and 4 within the transmittable range, and measures the transmission delay time of the response message received from each node. The MN 2 considers the node having the longest transmission delay time as the farthest node and selects it as the HN 1. The selected HN 1 selects HN 3 and MN 5 of neighboring clusters in the same manner as above. This process is then spread to the entire wireless sensor network.

That is, in the conventional head node (HN) selection method, when the head node transmits data to the sink node through wireless communication, the energy consumption of the head node is increased by the square of the distance of the sink node. The head node may select a random or high energy node among the sensor nodes as the head node.

However, when the head node is randomly selected, there is a problem that a node with very low energy is selected as the head. When selecting a node with a high energy, the node is located far from the sink node even if the node has a lot of energy. There is a problem that the head execution time cannot last longer than other head target nodes. In addition, there is a disadvantage in that the lifetime of the wireless sensor network is shortened due to the above problem.

In order to solve the above problems, the present invention calculates and calculates the energy ratio of the sensor node and the distance ratio between the sensor node and the sink node, that is, the energy-to-sink node distance ratio (EDR) of the node from two factors. Provided are a method for selecting a head node for clustering a wireless sensor network in which a sensor node having a large EDR is selected as a head node, and a wireless sensor network.

The method for selecting a head node during clustering of a wireless sensor network system according to an embodiment of the present invention is a method for selecting a head node during clustering of a wireless sensor network system including a sync node and a plurality of sensor nodes, wherein each of the plurality of sensor nodes is an effective head. Determining a node state; A second step of, according to the determination result, sensor nodes which are in an effective head node state broadcasting a head node selection request message including a distance ratio (EDR) of an energy to a sink node within a predetermined range; Receiving the second stage broadcast and comparing the EDRs of the sensor nodes in the other valid node states with their EDRs based on the received broadcast to determine whether to maintain the valid head node state. Three steps; And a fourth step of selecting the sensor nodes that maintain the valid head node state determined in the third step, as the head node.

In the first step of the present invention, it is preferable that each of the plurality of sensor nodes generate a random number having a percentage, and compare the random number with the cluster head ratio HDR to determine the effective head node state.

According to a second step of the present invention, steps 2-1 in which each node determined as an effective head node among a plurality of sensor nodes according to the determination result in the first step designates a node state as an effective head node state; Generating a head selection request message in which each of the sensor nodes designated as valid head node states includes a distance to energy ratio EDR; And step 2-3 of each of the sensor nodes designated as valid head node states to broadcast the head selection request message generated in step 2-2 within a predetermined communication radius.

The third step of the present invention includes a third step in which each of the sensor nodes designated as valid head node states receives a head selection request message from different sensor nodes; And if each of the sensor nodes designated as valid head node states based on the received head selection request message compares an EDR for another sensor node with its own EDR, and designates a non-head node state if it is smaller than an EDR for another sensor node. When each of the sensor nodes designated as valid head node states based on the received head selection request message compares an EDR for another sensor node with its own EDR, the valid head node state is determined. It is preferable to include; 3-2 step to maintain.

The second step of the present invention includes setting a head selection time value, and the third step includes determining whether a head selection time value has elapsed for each of the sensor nodes designated as valid head node states. Step 3-1 of receiving a head selection request message from different sensor nodes when the value has not elapsed; And if each of the sensor nodes designated as valid head node states based on the received head selection request message compares an EDR for another sensor node with its own EDR, and designates a non-head node state if it is smaller than an EDR for another sensor node. When each of the sensor nodes designated as valid head node states based on the received head selection request message compares an EDR for another sensor node with its own EDR, the valid head node state is determined. It is preferable to include; 3-2 step to maintain.

In step 3-2 of the present invention, it is preferable that each of the sensor nodes designated as valid head node states repeats the comparison of the EDR with respect to the other sensor node and their EDR based on the received head selection request message.

According to another exemplary embodiment of the present invention, a method for selecting a head node during clustering of a wireless sensor network system may include: a sink node receiving data through wireless communication; A plurality of sensor nodes for transmitting the sensed data to the sink node; And a plurality of sensor nodes are located in a sensor field divided into a plurality of clusters, and collect the sensed data transmitted from member nodes and member nodes that transmit sensed data included in each of the plurality of clusters to sink nodes. As a method of selecting a head node during clustering of a wireless sensor network system including a transmitting head node, it is preferable to select a head node based on a distance ratio (EDR) of an energy to a sink node among a plurality of sensor nodes.

The head node of the present invention is preferably selected from among a plurality of sensor nodes located in a cluster which is reconfigured at regular intervals.

The head node of the present invention is selected based on the distance-to-energy ratio (EDR) of the plurality of sensor nodes, but preferably has the highest EDR compared to the member nodes.

A wireless sensor network according to an embodiment of the present invention, comprising: a sink node receiving data through wireless communication; A plurality of sensor nodes for transmitting the sensed data to the sink node; And a plurality of sensor nodes are located in a sensor field divided into a plurality of clusters, and collect the sensed data transmitted from member nodes and member nodes that transmit sensed data included in each of the plurality of clusters to sink nodes. The head node includes a head node to transmit, and the head node is selected based on a distance ratio (EDR) of an energy to a sink node among a plurality of sensor nodes.

The head node of the present invention is preferably selected from among a plurality of sensor nodes located in a cluster which is reconfigured at regular intervals.

The head node of the present invention is selected based on the distance-to-energy ratio (EDR) of the plurality of sensor nodes, but preferably has the highest EDR compared to the member nodes.

According to the present invention, if the sensor node calculates the EDR and determines whether to operate as a head node, the head nodes determined to operate as the head node broadcast a head node selection request message, and then the corresponding sensor node itself has the largest EDR. In this case, by selecting the sensor node having the largest EDR as the head node, the head node operates for the longest time in the data transmission step.

In addition, according to the present invention, since the node having the largest EDR becomes the head node among the effective head nodes, it can be performed as the head node for the longest time regardless of the distance from the sink node, and the energy of the sensor node based on the EDR. In addition, the member node communicates with the long-lived head node, which has an excellent effect of extending the life of the wireless sensor network as a whole.

1 is an exemplary diagram illustrating head node selection in a wireless sensor network according to the related art.

2 is an exemplary diagram illustrating a wireless sensor network in accordance with the present invention.

3 is a cluster configuration diagram of a wireless sensor network based on head node selection of the present invention.

4 is a graph illustrating the EDRs of sensor nodes N1 and N5 of FIG.

5 is a graph illustrating the EDRs of sensor nodes N2 and N6 of FIG.

6 is a flowchart illustrating a method of selecting a head node of a wireless sensor network according to the present invention.

Explanation of symbols on the main parts of the drawings

10: head node 20: member node

100: sink node 200: sensor field

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, a wireless sensor network can be sensed by a sensor and includes a processor for processing the collected information, and a sensor node, which is a small wireless transceiver transmitting the same, and collecting the same and sending them out to the outside. A network composed of sink nodes will be described as a basic description, and a description of components common to the prior art will be omitted.

2 is a cluster configuration diagram of a wireless sensor network based on head node selection according to the present invention. 2, a wireless sensor network to which a method for selecting a head node for clustering a wireless sensor network according to the present invention is applied is disposed in a sensor field 200 to wirelessly communicate with a sink node 100 and each other. It includes a plurality of sensor nodes (10, 20).

Here, a sensor node (SN) is a kind of small device having collected sensing and communication functions for observing physical phenomena, and a smart dust is a basic element of a wireless sensor network. Yes.

Here, the sensor field 200 is an area in which a plurality of sensor nodes are distributed and sensed, and may configure sensor nodes in a predetermined range by clustering.

In addition, the sensor nodes 10 and 20 disposed in the sensor field 200 may be configured to operate as the member node 20 and the head node 10 for each cluster.

Thus, the wireless sensor network may include a plurality of clusters 201 configured as such. In this case, the cluster 201 is reconfigured for each round having a certain period, and includes configuring a cluster for each round and transmitting data from the sensor node to the sink node.

That is, in the cluster 201 configured through clustering, data collected by the member nodes 20 is transmitted to the head node 10 in the step of transmitting data of each round, and the head node 10 transmits the data to the sink node 100. To send).

In addition, the head node 10 repeats the data collection of the neighboring member nodes 20 and transmitting the collected data to the sink node 100 every round. In other words, the energy consumption of the head node 10 is relatively large compared to the member node 20, and the head node 10 is particularly energy intensive since it is located remote from the sink node 100.

Thus, the wireless sensor network according to the present invention calculates the head to node ratio in the cluster configuration step according to the following equation: residual energy to sink node (EDR; Energy to Distance Ratio, EDR). A sensor node having a large calculated EDR value is selected as a head node to operate.

EDR = node energy / (sink node distance) ²

In other words, when configuring a cluster, the wireless sensor network according to the present invention determines that each sensor node included in the cluster calculates an EDR, determines whether to operate as a head node, and determines to operate as a head node. The effective head nodes broadcast (broadcast) the head node selection request message to the surroundings, and then repeat the process of determining whether the EDR is large among the sensor nodes that receive the broadcast. Will be selected.

3 is an exemplary diagram illustrating a wireless sensor network according to the present invention. Referring to FIG. 3, in the wireless sensor network according to the present invention, as described above, the sensor nodes N1 and N2 having a large EDR are selected as the head nodes 10 in FIG. 2 to configure the wireless sensor network by clustering. do. Hereinafter, a description will be given as an example in which only a part of the plurality of sensor nodes are selected to select the head node 10 and the member node 20 in FIG. 2.

In detail, the sensor node N1 has a cluster radius 211, a sensor node N2 has a cluster radius 221, a sensor node N5 has a cluster radius 210, and a sensor node N6 has a cluster radius 220 as a communication range. The selection request message will be broadcast.

That is, the sensor nodes N1, N5, N2, and N6 determine to be head nodes (set as valid head nodes), and the sensor nodes broadcast a head node selection request message within each cluster radius. In this case, the head node selection request message may include the calculated value of the EDR.

Here, the setting of the effective head nodes N1, N5, N2, and N6 may be set based on the ratio of the head node to the member node in the cluster radius 211.

At this time, the valid head nodes N5 and N6 which have received the broadcast messages of the near valid head nodes N1 and N2 compare with their own EDR values and release their effective head nodes when their EDR values are small. , N6).

In addition, the valid head nodes N1 and N2 that have received the broadcast messages of the near valid head nodes N5 and N6 maintain the effective head node when their EDR value is large compared to their EDR value, and thus the head node N5, N6).

In this way, two clusters C1 and C2 are formed, the cluster C1 is composed of the head node N1 and the member nodes N5, N7, and the cluster C2 is composed of the head node N2 and the member nodes N4, N6, N3.

As described above, the clustering of the wireless sensor network is configured by selecting a head node and a member node by repeating a process of comparing EDR values between sensor nodes.

4 is a graph illustrating EDRs of sensor nodes N1 and N5 of FIG. 3. Referring to FIG. 4, the energy relationship of the sensor node is N1 >> N5, that is, the energy of the sensor node N1 is much higher than that of the sensor node N5, and as shown in FIG. 3, the sensor node N1 may be connected to the sink node than the sensor node N5. Although it is located far away, the energy of N1 is much higher, so the EDR of sensor node N1 is larger than that of sensor node N5, so that N1 is selected as the head node.

FIG. 5 is a graph illustrating EDRs of sensor nodes N2 and N6 of FIG. 3. Referring to FIG. 5, the energy relationship of the sensor node is N2 <N6, that is, the energy of the sensor node N2 is somewhat higher than the energy of the sensor node N6, and as shown in FIG. Since it is located close to the sink node, the sensor node N2 is selected as the head node because the EDR of the sensor node N2 is larger than that of the sensor node N6.

Accordingly, since the wireless sensor network according to the present invention selects a sensor node having a large EDR as a head node from among a plurality of sensor nodes constituting the cluster, the head node operates for the longest time in the data transmission step, thereby making the entire wireless sensor network. It will increase the energy efficiency of.

In addition, since the node with the largest EDR becomes the cluster head node among the effective head nodes, it is possible to perform the head node for the longest time regardless of the distance from the sink node, and each member node in the cluster has a long life head node. Communication with the STM extends the life of the wireless sensor network as a whole.

Hereinafter, a method of selecting a head node of a wireless sensor network according to the present invention will be described with reference to the drawings. In FIG. 1, the same reference numerals as used in FIG. 5 refer to performing the same function. Hereinafter, a description will be given as an example in which only a part of the plurality of sensor nodes are selected to select the head node 10 and the member node 20 in FIG. 2.

6 is a flowchart illustrating a method of selecting a head node of a wireless sensor network according to the present invention. Referring to FIG. 6, a method of selecting a head node of a wireless sensor network according to the present invention may determine whether an effective head node state of each of a plurality of sensor nodes 10 and 20 existing in a sensor field 200 is determined. In the first step (S300 ~ S305), according to the determination result of the first step, the sensor node 10 in the effective head node state broadcasts the head node selection request message to the sensor node 20 in the other valid head node state within a certain range. On the basis of the second stage (S306 to S307) and the broadcast of the second stage, each of the sensor nodes in the effective head node state compares the distance ratio (EDR) of the node energy to the sink node to determine whether to maintain the effective head node state. The third step (S308 to S311) of determining, and the fourth step (S312 to S313) of designating the sensor node 10 which is an effective head node state as the head node.

In the wireless sensor network to which the method for selecting the head node of the wireless sensor network of the present invention is applied, the sensor nodes N1, N3, N5, and N7 represent a variable E for indicating the remaining energy amount and a distance to the sink node. Has a variable (D) and a variable (ST) to indicate the state of the sensor nodes (N1, N3, N5, N7), the head selection timer (HEAD_ELECTION_REQ) for limiting the head selection time, the head selection request message broadcast range It is assumed that a communication radius value (COM_RADIUS) for specifying a value and a value (HDR) for specifying a head ratio in a cluster can be arbitrarily designated.

Here, the communication radius value COM_RADIUS designates a communication radius value that designates a message broadcast range, which is specified when initializing the sensor nodes N1, N3, N5, and N7 to form a cluster in the sensor field 200. According to the wireless communication capability of the sensor nodes (N1, N3, N5, N7), for example, can be randomly specified, such as 10, 50, 100 meters.

Here, the value HDR for designating the cluster head ratio may designate the ratio of the head node among the sensor nodes N1, N3, N5, and N7 in the cluster. For example, when HDR is 0.05 (5%), it means that the average number of head nodes is 5 out of 100 sensor nodes N1, N3, N5, and N7.

In addition, according to the RN value generated by each sensor node (N1, N3, N5, N7) at the time of head selection time, even if HDR is 5%, the sensor node (N1, N3, When the N5 and N7 are in the sensor field 200, the sensor node N1, which is in the total effective head node state, is located in the area corresponding to the communication radius of the sensor nodes N1, N3, N5 and N7 in the effective head state. Only sensor nodes N1, N3, N5, N7 in fewer effective head node states than N3, N5, N7 may be included. Therefore, among the sensor nodes N1, N3, N5, and N7 in the effective head state belonging to the subregion, the sensor node N1 having a large EDR can be selected as the head node.

In this case, the sensor field 200 refers to the entire area where the sensor nodes N1, N3, N5, and N7 are arranged, and the communication radius of the effective head node (for example, x COM_RADIUS2) is used in the sensor field 200. ) Is part of the entire area of the sensor field 200.

Hereinafter, the first steps S300 to S305 of determining whether each of the plurality of sensor nodes N1, N3, N5, and N7 are valid head nodes will be described.

First, each of the plurality of sensor nodes N1, N3, N5, and N7 designates a node state as a non-head node state (S300). That is, the head node state ST may include a non-head node that designates a member node and an effective head node that indicates a head node possible state.

At this time, a time limit value for selecting the head node is set (S301). That is, the head selection timer HEAD_ELECTION_REQ specifies a head selection time limit. The head selection timer HEAD_ELECTION_REQ specifies a head selection time value, and the sensor nodes N1, N3, and N5 form a cluster in a new round. , When N7) is initialized, the time value is enough time (e.g., 10, 50, 100 msec (milliseconds), etc.) for head node selection message transmission, message processing and head node determination. The node (N1, N3, N5, N7) can be randomly specified according to the processing performance.

 Each of the plurality of sensor nodes N1, N3, N5, and N7 calculates a distance ratio EDR of the node energy to the sink node (S302). That is, as described above, in the cluster configuration step, the head node 10 may calculate an energy to distance ratio (EDR) of the residual energy to the sink node of the corresponding node according to the following equation.

EDR = node energy / (sink node distance) ²

 Each of the plurality of sensor nodes N1, N3, N5, and N7 generates a random number RN having a percentage (S303). That is, each sensor node N1, N3, N5, N7 generates a random number RN having a range of percentages (0-1).

At this time, the sensor node compares the random number RN with the cluster head ratio HDR to determine whether to operate as an effective head node (S304), and the node is in the effective head node state only when the random number RN is smaller than HDR. State ST is specified (S305).

Here, the sensor nodes N1, N3, N5, and N7 may arbitrarily specify a value HDR for specifying the cluster head ratio, and the value HDR for specifying the cluster head ratio is a sensor node N1 in the cluster. , N3, N5, N7), the ratio of the head node can be specified. For example, when HDR is 0.05 (5%), it means that the average number of head nodes is 5 out of 100 sensor nodes N1, N3, N5, and N7.

The sensor nodes N3, N5, and N7, which are not determined to operate as the head node in step S305, operate as the member node 20 because the node state is set to the non-head ST_NONE_HEAD state.

Hereinafter, according to the determination result of the first step, a second step (S306 to S307) in which the sensor node in the effective head node state broadcasts the head node selection request message to the sensor node in the other valid head node state within a predetermined range.

In step S305, the sensor nodes N1, N3, N5, N7 designate the node state ST as a valid head node state (for example, ST_VALID_HEAD, etc.), and then specify the sensor nodes N1, N3, N5 and N7 generate a head selection request message (S306). That is, the sensor nodes N1, N3, N5, and N7 generate a head selection request message HEAD_ELECTION_REQ composed of their node identifier ID, residual energy to sink node distance ratio EDR field.

Subsequently, the sensor nodes N1, N3, N5, and N7 broadcast the head selection request message at a communication radius that designates a message broadcasting range (S307). That is, the sensor nodes N1, N3, N5, and N7 broadcast the head selection request message HEAD_ELECTION_REQ to the communication radius COM_RADIUS that designates the message broadcast range.

Here, the communication radius value COM_RADIUS that designates the head selection request message broadcast range designates a communication radius value that designates the message broadcast range, which is specified when the sensor nodes N1, N3, N5, and N7 are initialized. According to the wireless communication capability of the node, it can be randomly designated, for example, 10, 50, 100 meters, or the like.

Hereinafter, a third step (S308 to S311) of determining whether to maintain the effective head node state by comparing the distance ratio EDR between the node energy and the sink node, respectively, based on the broadcast of the second step, the sensor nodes in the effective head node state are compared. ).

After step S307, each of the sensor nodes N1, N3, N5, and N7 designated as an effective head node state determines whether the head selection time has elapsed (S308), and when the head selection time has not elapsed, from another sensor node. The head selection request message HEAD_ELECTION_REQ is received (S309).

Subsequently, each of the sensor nodes N1, N3, N5, and N7 designated as valid head node states based on the received head selection request message compares the EDRs for other sensor nodes, and compares the sizes with the EDRs for other sensor nodes. (S310). If the comparison result is small, the non-head node state is specified (S320). That is, when the EDR of the sensor node itself is smaller than the EDR of the received message, the node state ST of the sensor node itself is designated as the non-head node ST_NONE_HEAD state.

If, based on the head selection request message received in step S310, each of the sensor nodes N1, N3, N5, and N7 designated as valid head node states compares the EDRs for other sensor nodes, and is greater than the EDRs for other sensor nodes. In this case, the valid head node state is maintained (S311), i.e., a message broadcast by another sensor node (in this case, when the node's own ID and the received message ID are not the same), the node's own EDR is smaller than that of the received message. If it is judged that it is not small, the node state ST of the sensor node itself is reassigned to the valid head node ST_VALID_HEAD state.

In addition, each of the sensor nodes N1, N3, N5, and N7 designated as valid head node states in steps S308 to S320 is designated as an effective head node state based on the received head selection request message until the head selection time has elapsed. Each of the sensor nodes N1, N3, N5, and N7 cyclically repeats the comparison of the EDRs for the other sensor nodes. That is, the node state of the sensor node N1 having a large EDR becomes a valid head node ST_VALID_HEAD state by a cyclical repetition of steps 309, 310, and 311 until the head selection time has elapsed. The node state of N5 and N7 becomes the non-head node ST_NONE_HEAD state.

Hereinafter, fourth steps S312 to S313 for designating a sensor node which is an effective head node state as a head node state will be described.

After repeating steps S308 to S320, it is determined whether each of the sensor nodes N1, N3, N5, and N7 is in a valid head node state (S312), and the determination result is a sensor node in a valid head node state ( N1) is designated as the head node 10 (S313). That is, it is determined whether the node state of the sensor node is a valid head node (ST_VALID_HEAD) state, and when the node state of the sensor node is a valid head node (ST_VALID_HEAD) state. , The node state is designated as the head node ST_HEAD.

If it is determined in step S312 whether each sensor node is in an effective head node state, and if the sensor node is not in an effective head node state, the node state is designated as a non-head state (S330). That is, when the node state of the sensor nodes N1, N3, N5, and N7 is not an effective head node ST_VALID_HEAD state, the node state is operated as the member node 20 by designating the node state to the non-head ST_NONE_HEAD state. will be.

Therefore, the present invention has the effect that the sensor node having the largest EDR among the effective head nodes is selected as the head node 10 in the cluster and the energy of the sensor nodes based on the EDR is efficiently consumed.

That is, the sensor node calculates the EDR and determines whether to operate as the head node, and the nodes determined to operate as the head node broadcast the head node selection request message, and then determine whether the node itself has the largest EDR, By selecting this largest node as the head node, there is an effect of selecting a head node for the head node to operate for the longest time in the data transmission step.

In addition, according to the present invention, since the node having the largest EDR becomes the cluster head node among the effective head nodes, the head node can be performed for the longest time regardless of the distance from the sink node, and the energy of the nodes based on the EDR is increased. It has the advantage of being efficiently consumed, and the member node communicates with the long-lived head node, thereby extending the life of the wireless sensor network as a whole.

While the technical idea of the embodiment as described above has been described with the accompanying drawings, it is intended to illustrate the best embodiment of the present invention by way of example and not to limit the invention. In addition, it will be apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit of the present invention.

Claims (12)

A head node selection method for clustering a wireless sensor network composed of a sync node and a plurality of sensor nodes, A first step of each of the plurality of sensor nodes determining an effective head node state; A second step of the sensor nodes in an effective head node state broadcasting a head node selection request message including a distance ratio of energy to sink node (EDR) within a predetermined range according to the determination result; Receiving the broadcast of the second step and based on the received broadcast, each of the sensor nodes in the effective head node state compares the EDR of the sensor node in the other valid node state with its own EDR to determine whether to maintain the effective head node state Third step; And And a fourth step of selecting sensor nodes maintaining a valid head node state determined in the third step as a head node. The method of claim 1, The first step is, Each of the plurality of sensor nodes generates a random number having a percentage, and compares the random number with a cluster head ratio HDR to determine the effective head node state. . The method of claim 1, The second step, A second step in which each node determined as an effective head node among the plurality of sensor nodes designates a node state as an effective head node state according to the determination result in the first step; Generating a head selection request message in which each of the sensor nodes designated as the effective head node state includes a distance ratio (EDR) of an energy to a sink node; And Step 2-3 of each of the sensor nodes designated as valid head node states to broadcast the head selection request message generated in step 2-2 within a predetermined communication radius; Head Node Selection Method for Clustering. The method of claim 1, The third step, Step 3-1, in which each of the sensor nodes designated as valid head node states receives a head selection request message from different sensor nodes; And On the basis of the received head selection request message, each of the sensor nodes designated as the valid head node state returns to the non-head node state when the EDR for another sensor node is smaller than the EDR for another sensor node. Designate, If each of the sensor nodes designated as the valid head node state based on the received head selection request message is larger than the EDR for another sensor node by comparing the EDR for another sensor node with its own EDR, the valid head node state Step 3-2 of maintaining; The head node selection method for clustering of the wireless sensor network comprising a. The method of claim 1, The second step is Setting the head selection time value; The third step, 3-1 that each of the sensor nodes designated as the effective head node state determines whether the head selection time value has elapsed, and receives a head selection request message from different sensor nodes when the head selection time value has not elapsed. step; And On the basis of the received head selection request message, each of the sensor nodes designated as the valid head node state returns to the non-head node state when the EDR for another sensor node is smaller than the EDR for another sensor node. Designate, If each of the sensor nodes designated as the valid head node state based on the received head selection request message is larger than the EDR for another sensor node by comparing the EDR for another sensor node with its own EDR, the valid head node state Step 3-2 of maintaining; The head node selection method for clustering of the wireless sensor network comprising a. The method of claim 5, wherein The third step is, Based on the received head selection request message, each of the sensor nodes designated as the effective head node state cyclically repeats the comparison of the EDR with respect to the other sensor node and its own EDR. Node selection method. A sink node receiving data through wireless communication; A plurality of sensor nodes transmitting the sensed data to the sink node; Wherein the plurality of sensor nodes are located in a sensor field divided into a plurality of clusters, and collect member data transmitting the sensed data included in each of the plurality of clusters and sensed data transmitted from the member nodes. A method of selecting a head node of a wireless sensor network including a head node transmitting to the sink node, And selecting the head node based on a distance-to-energy ratio (EDR) from the plurality of sensor nodes. The method of claim 7, wherein The head node, The method for selecting a head node for clustering a wireless sensor network, characterized in that selecting from among the plurality of sensor nodes located in a cluster reconfigured at regular intervals. The method of claim 7, wherein The head node, A method for selecting a head node for clustering a wireless sensor network, wherein the sensor node is selected based on a distance-to-energy ratio (EDR) of the plurality of sensor nodes, but has the highest EDR compared to the member nodes. As a wireless sensor network, A sink node receiving data through wireless communication; A plurality of sensor nodes transmitting the sensed data to the sink node; Including, The plurality of sensor nodes are located in a sensor field divided into a plurality of clusters, and collect member data transmitting the sensed data included in each of the plurality of clusters and sensed data transmitted from the member nodes to the sink node. Includes a head node to transmit, And selecting the head node based on a distance to energy ratio (EDR) ratio among the plurality of sensor nodes. The method of claim 10, The head node, And selecting from among the plurality of sensor nodes located in a cluster that is reconfigured at regular intervals. The method of claim 10, The head node, A wireless sensor network, wherein the wireless sensor network is selected based on a distance-to-energy ratio (EDR) of the plurality of sensor nodes, but has the highest EDR compared to the member nodes.