KR101800133B1 - Path selection method for relay node in disjoint networks - Google Patents

Abstract

A method for determining a moving path of a relay node for a divided network comprises the steps of: a computer device modeling an area, in which a plurality of segments are located, as a grid comprising a plurality of cells; when a relay node is located on the cell, the computer device using information on communicable segments to determine one reference cell for each segment among the plurality of cells; the computer device using the reference cell and grouping the plurality of segments into k numbers of temporal clusters; and the computer device grouping the plurality of segments into k numbers of final clusters on the basis of energy consumed by the relay node which moves the temporal clusters and transmits information. The segment corresponds to one connected network, and a path connecting a plurality of reference cells contained in each cluster in the final clusters is a path where the relay node moves. The present invention uses a limited numbers of relay nodes to transmit information between divided networks, and makes energy used by each relay node be uniform to the most, such that the relay node can perform a relay service for a long time.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a relay node,

The techniques described below relate to techniques for determining the path of a moving relay node.

In a wireless sensor network (WSN), sensor nodes are arranged adjacent to each other within a communication distance to form one connected network. A wireless sensor network transmits information collected by a source node to a target node via intermediate nodes using a wireless communication network between adjacent sensor nodes. Therefore, in the wireless sensor network, the connectivity between the sensor nodes must be guaranteed.

B. Hao, H. Tang, and G. Xue, "Fault-tolerant relay node placement in wireless sensor networks: formulation and approximation," Proc. the Workshop on High Perf. Switching and Routing, Phoenix, AZ, 2004 J. Tang, B. Hao, and A. Sen, "Relay Node Placement in Large Scale Wireless Sensor Networks", Computer Comm., Special issue on wireless sensor networks, Vol. 29. pp. 490-501, 2006.

The technique described below is intended to provide a technique for transferring information using a relay node moving in a divided network.

A method for determining a travel path of a relay node for a divided network includes the steps of modeling a region where a plurality of segments are located in a grid form having a plurality of cells, Determining a reference cell for each segment of the plurality of cells using information about a possible segment and setting the total number of determined reference cells to be the smallest; Grouping k segments into k temporary clusters, and grouping the plurality of segments into k final clusters based on the energy consumed by the relay node that conveys the information while moving the temporary clusters and clusters . The segment corresponds to one connected network, and a path connecting a plurality of reference cells included in each cluster in the final cluster is a path through which the relay node moves.

The techniques described below can transfer information between divided networks using a limited number of relay nodes. Also, the technique described below makes the energy used by each relay node as equal as possible, and the relay node performs the relay service for a long time.

1 is an example of transmitting information using relay nodes in a plurality of networks.
2 shows an example of a region of interest in which sensor nodes are arranged.
Figure 3 is an example of a temporary cluster formed in a region of interest.
4 is an example of expanding a cluster according to a greedy algorithm.
5 is an example of expanding a cluster according to a greedy algorithm.
FIG. 6 is an example in which an expanded cluster is optimized according to a greedy algorithm.
FIG. 7 is an example of a flowchart of a method of determining a movement path of a relay node for a divided network.

The following description is intended to illustrate and describe specific embodiments in the drawings, since various changes may be made and the embodiments may have various embodiments. However, it should be understood that the following description does not limit the specific embodiments, but includes all changes, equivalents, and alternatives falling within the spirit and scope of the following description.

The terms first, second, A, B, etc., may be used to describe various components, but the components are not limited by the terms, but may be used to distinguish one component from another . For example, without departing from the scope of the following description, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

As used herein, the singular " include "should be understood to include a plurality of representations unless the context clearly dictates otherwise, and the terms" comprises & , Parts or combinations thereof, and does not preclude the presence or addition of one or more other features, integers, steps, components, components, or combinations thereof.

Before describing the drawings in detail, it is to be clarified that the division of constituent parts in this specification is merely a division by main functions of each constituent part. That is, two or more constituent parts to be described below may be combined into one constituent part, or one constituent part may be divided into two or more functions according to functions that are more subdivided. In addition, each of the constituent units described below may additionally perform some or all of the functions of other constituent units in addition to the main functions of the constituent units themselves, and that some of the main functions, And may be carried out in a dedicated manner.

Also, in performing a method or an operation method, each of the processes constituting the method may take place differently from the stated order unless clearly specified in the context. That is, each process may occur in the same order as described, may be performed substantially concurrently, or may be performed in the opposite order.

The technique described below is for connecting (recovering) a network in which a source node delivers information to a target node through a plurality of relay nodes. The source node may include a client device, a sensor device, an IoT device, and the like. The target node may include a user terminal, a PC, a server, a gateway, an access point (AP), and the like. For example, the techniques described below can be used to recover a wireless sensor network. Generally, in a wireless sensor network, a sensor node collects information and acts as a relay for transmitting information. For convenience of explanation, the following description will be made on the basis of a wireless sensor network.

The technology described below assumes a state in which a failure occurs in the wireless sensor network and is divided into a plurality of individual networks. In a wireless sensor network, one node may be disconnected from the network due to energy shortage or the like and may be disconnected. In a wireless sensor network, neighboring multiple nodes may lose their communication functions at the same time because of physical forces such as natural disasters and explosions. In this case, one wireless sensor network is divided into a plurality of individual networks. In the graphical data structure, a connected graph is divided into a plurality of segments (disjoint segments).

The technique described below is for connecting a plurality of segments to a relay node to recover a disconnected wireless sensor network. The techniques described below use a mobile relay node to recover the network. The technique described below restores the network using less mobile relay nodes than the number of fixed relay nodes needed to recover the network. The mobile relay node may be implemented in the form of an automobile, a robot, a drone, or a UAV. Hereinafter, the relay node assumes a mobile relay node.

The technique described below is a technique for determining the location and path of the relay node for restoring the network. Hereinafter, it is assumed that the computer apparatus calculates the position and path of the relay node. The computer device may be a server connected to a network, any node constituting the network, a central system for disaster relief, a personal PC, and the like.

1 is an example of transmitting information using relay nodes in a plurality of networks. Figure 1 shows five separate networks (A, B, C, D and E). Individual networks can not transmit information to each other. An individual network may be an independent network, with one original network being a failure or disaster. Or the individual networks may be essentially different networks. Individual networks can relay information to each other using relay nodes. In FIG. 1, an individual network A transmits information to an individual network C using a relay node. Circle shown around the relay node means the communication radius of the relay node. It is assumed that each individual network uses relay nodes to convey information to other individual networks. For convenience of explanation, it is assumed that each individual network is divided into a plurality of networks.

A plurality of relay nodes move and transfer information between segmented segments. A relay node acts as a kind of gateway between segments.

The n segments constitute the set S _T. S _T = {S ₁ , S ₂ , ..., S _n }. Where n> 2. The relay node set is M = {M ₁ , M ₂ , ..., M _k }. It is assumed that the number of relay nodes is k. Computer apparatus clustering the segments into a plurality of clusters C _i. Where i = 1, ..., k. As will be described later, the entire cluster has a star topology. And the other cluster is connected around the central cluster.

Any one of the relay node M _i is responsible for any one of a cluster C _i. The relay node M _i moves on a constant path TR (M _i ) visiting a plurality of points in the cluster C _i . For example, TR ( _Mi ) may be a plurality of coordinate lists. The relay node M _i receives information from an adjacent sensor node at a specified point constituting TR ( _Mi ) or transmits information to an adjacent sensor node. Thus, the relay node M _i transmits information between the segments belonging to the cluster C _i .

It is assumed that relay node M _i carries information from segment S _S to S _D. In this case, a certain delay may occur. Equation (1) below is a calculation of the delay D (S _s , S _d ) generated in this process.

In Equation (1), D _T is the trip latency occurring when the relay node moves from segment S _S to S _D. D _T is determined by Trip _{s, d} , which is the path distance between the segments S _S and S _{D ,} and the traveling speed V _M of the relay node. For convenience of explanation, it is assumed that the traveling speed V _M of the following relay node is the same.

D _C is the transmission latency that occurs when the relay node receives and transmits information. D _C is determined according to Data _{s, d} and the transmission rate B _M , which are the amounts of information transmitted in S _S and S _D. [sigma] is a constant value for cases in which information is transmitted over one or more segments. For example, if S _S and S _D belong to the same cluster, σ may be 1. As described above, if one of S _S and S _D belongs to the central cluster on the premise of the star topology, σ may be 2. If S _S and S _D are not central clusters but different clusters, σ can be 3.

S _S is in a first cluster C _1, and _D S is assumed to belong to a second cluster C _2. The relay node M ₁ that is responsible for the C ₁ and store information transmitted to the S _S delivered to a specific point in C _1. Here, the specific point may be any sensor node or a separate storage device. The specific point stores the information relayed by the relay node M ₁ . The relay node M ₂ responsible for C ₂ then receives the information stored at the specific point and delivers the information to S _D. Thus, if S _S and S _D do not belong to a cluster, a relay delay may occur in the information transmission process. D _R is the relay latency that occurs during the information transfer between clusters. ht _{s, d} is the time between the point at which a specific point (buffer) receives information from M ₁ and the point at which information is transmitted to M ₂ in the process of transmitting information from S _S to S _D.

Also, the relay node M _i consumes a certain amount of energy in the process of transmitting the information from the segments S _S to S _D. Equation (2) below is a calculation of the energy E (M _i ) consumed by the relay node M _i .

In Equation (2), E _M is the energy consumed by the relay node M _i while traveling through TR (M _i ). E _M is the energy consumed to move the path belonging to the segment C _i once. E _C is the energy that relay node M _i consumes in communication. E _C is the amount of information and the number of times of receiving (downloading) or transmitting (uploading) the information.

E _M is determined according to the distance traveled by the relay node as shown in Equation (3) below. In Equation (3), mu has a size between 0.1 and 1 (J / m).

릴레이 노드 M_i가 전달하는 정보의 양이다. 한편 관련된 선행 연구 중 정보 전달에 소비되는 에너지를 연산한 것이 있다. 이에 따르면 α = 100nJ, β = 0.1

이고

= 2이다. 물론 이와 다른 기준을 사용하여 E_C를 결정할 수도 있을 것이다.E _C is determined according to the energy used to transmit one bit through wireless communication as shown in Equation (4) below. In Equation 4,

The amount of information relayed by relay node M _i . On the other hand, among related studies, there is a calculation of the energy consumed in information transmission. According to this, α = 100 nJ, β = 0.1

ego

= 2. Of course, you can also use other criteria to determine E _C.

The computer device determines the areas of the plurality of clusters and determines the paths of the plurality of relay nodes. The computing device determines the clusters so that the average energy consumed by the relay nodes in each cluster is at most similar, while at the same time the delay time to transfer information between the segments is minimized. Hereinafter, a process of determining a plurality of clusters by a computer apparatus will be described. In the following description, it is assumed that all the relay nodes have the same communication radius R. The communication radius R may be the same as the communication radius r of the sensor node.

A computer device tries to recover a network divided into n segments by using k relay nodes. The computer device determines k clusters corresponding to the moving region or path of k relay nodes. The computing device determines the cluster such that it has a central cluster C _k and the remaining clusters have a topology connected via C _k .

The process by which a computer device determines a cluster includes two processes. In the first step, (1) the computer device reconstructs the region of interest in which the sensor node is disposed into a grid having cells of the same size, and determines cells near the center of the ROI including all of the segments . (2) The computer device determines k-1 clusters connected to the central cluster and the central cluster based on the proximity of the cells included in the cluster. The k clusters determined in the first step are temporary clusters. Hereinafter referred to as a temporary cluster. In the second step, the computing device optimizes the cluster while balancing energy among the plurality of clusters. A specific procedure will be described below.

First step: Modeling the damage area and determining a temporary cluster

The computer device models the corrupted network topology into a grid area. Then, the computing device determines k-1 VC _i , which are partially overlapped with the center clusters VC _k and VC _k, and located in the periphery. Where VC _k and VC _i are temporary clusters.

(1) modeling of the damaged area

2 shows an example of a region of interest in which sensor nodes are arranged. 2 (a) shows an example in which the ROI is divided into 18 segments. Figure 2 (a) shows each segment as a circle and identifies it as a different number. In FIG. 2 (a), G denotes the center of the region of interest.

크기를 갖는 그리드 형태로 관심 영역을 모델링한다. 따라서, 이동 릴레이는 어느 하나의 셀의 중심에서 주변의 8개의 셀과 통신이 가능하다. 도 2(b)에서 셀에 표시된 숫자는 이동 릴레이가 셀의 중심에서 통신 가능한 세그먼트를 나타낸다. L(c_i)는 셀 c_i에서 통신 가능한 세그먼트에 대한 정보를 나타낸다. c(소문자)는 셀을 의미한다. 예컨대, c(6,5)에 위치한 이동 릴레이는 세그먼트 S₅ 및 S₁₈와 통신(접속)할 수 있다. FIG. 2 (b) is an example of modeling the region of interest shown in FIG. 2 (a) with a grid of a certain size. 2 (b) is an example of modeling an area of 1700 x 1100 m ² as 17 x 11 cells. The computer device constantly models the region of interest in which the sensor node is disposed based on the communication radius R of the relay node or the sensor node. In Fig. 2 (b), the communication radius is indicated by the dotted line at the upper left corner. The computer device has one side of a square cell

And models the region of interest in the form of a grid of dimensions. Therefore, the mobile relay can communicate with eight surrounding cells at the center of one cell. The number displayed in the cell in Fig. 2 (b) indicates a segment through which the mobile relay can communicate at the center of the cell. L (c _i ) represents information on segments that can be communicated in the cell c _i . c (lower case) means cell. For example, a mobile relay located at c ( _{6, 5} ) can communicate (connect) with segments S ₅ and S ₁₈ .

이다. NBR(c_i)는 c_i의 이웃 셀의 집합이다. |NBR(c_i)|≤8이다. prox(c_i)는 셀 c_i와 관심 영역의 중심점 G 사이의 근접성을 나타낸다. prox(c_i)는 c_i와 중심점 G 사이에 위치하는 셀의 개수 일 수 있다. 예컨대, 셀 c_i(x_i,y_i)과 c_j(x_j,y_j) 사이에 있는 셀의 개수를 CellDist(c_i,c_j)라고 하면, CellDist(c_i,c_j)는 max(|x_i-x_j|,|y_i-y_j|)로 결정될 수 있다. 즉 prox(c_i)는 max(|x_i-x_g|,|y_i-y_g|)이다. 여기서 G는 c_g(c_g,y_g)이다.The computer device groups (clusters) the segments using information about the cells. The information on the cell includes an identifier of the cell, and information on a segment to which the relay node located in the cell can communicate. The computer device uses three criteria to cluster the segments. Computer devices use a standard called _{access (c i), 1hop (} c i) and prox (c _i) and clustering the segments. access (c _i ) is the number of segments that can be accessed (communicated) in cell c _i . That is, access (c _i ) is | L (c _i ) |. 1hop (c _i ) is the number of segments accessible via one neighbor cell in cell c _i . That is, 1hop (c _i )

to be. NBR (c _i ) is the set of neighbor cells of c _i . | NBR (c _i ) | 8. prox (c _i ) represents the proximity between the cell c _i and the center point G of the ROI. prox (c _i ) may be the number of cells located between c _i and the center point G. For example, the cell c _i (x _i, y _i) and c _j (x _j, y _j) Speaking CellDist (c _i, c _j) the number of cells between, CellDist (c _i, c _j) is max (| x _i -x _j |, | y _i -y _j |). That is, prox (c _i ) is max (| x _i -x _g |, | y _i -y _g |). Where G is _g c (c _g, y _g).

The computer device determines the minimum number of cells CS _T covering all segments S _? Using access (c _i ), 1hop (c _i ) and prox (c _i ). The computer device attempts to find a path with less energy used by the relay node while reducing the delay due to information transfer. The computing device determines a central cluster that includes the center G of the region of interest, and determines the other cluster to be associated with the central cluster. To this end, the computer device selects a cell based on the _{access (c i), 1hop (} c i) and prox (c _i) for each segment. One cell selected for each segment is referred to as a reference cell. CS _T corresponds to a set of selected reference cells.

The computer device selects the reference cell by judging for each segment in the following order. (i) Selects c _i that can connect to the largest segment within the distance R. That is, a cell having a large access (c _i ) is selected. (ii) Select c _i that can connect to the largest number of segments within a distance 2R. That is, a cell with 1hop (c _i ) is selected. (iii) Select c _i closest to the center G of the region of interest. That is, prox (c _i ) selects a small cell. That is to select the reference cell as the reference sequence of _{access (c i), 1hop (} c i) and prox (c _i). For example, if a reference cell is not selected for a specific segment with access (c _i ), the reference cell is selected using the next reference.

For example, in FIG. 2 (b), c (4, 1) is a reference cell selected for segments S ₈ and S ₉ based on access (c _i ). c (3, 3) is the reference cell selected for segment S ₁₀ based on 1hop (c _i ). c (9,14) is the reference cell selected for S ₇ based on prox (c _i ). FIG. 2 (b) shows an example in which a reference cell is selected for 18 segments using the reference. In Fig. 2 (b), the reference cell is shown as shaded. There are overlapping reference cells for a plurality of segments. FIG. 2 (b) is an example in which all 14 reference cells are selected.

(2) Temporary cluster formation based on proximity to G

The computer device groups all segments S _{? Into} k clusters. Each cluster corresponds to the area covered by the relay node. The computer device forms a temporary cluster such that all cells are connected within two clusters. This can minimize the information transmission delay between two segments belonging to different clusters. The computer apparatus attempts to minimize the delay in information transfer between the two cells while at the same time equalizing the energy consumed by each relay node.

The computer apparatus first determines a temporary center cluster VC _k that includes the center G of the region of interest. At this time, the temporary center cluster VC _k includes only the cell in which the center G is located. In addition, the computer device determines the clusters around the center of the temporary k-1 cluster VC _i. A cluster located around a temporary central cluster is called a temporary peripheral cluster. The cluster has a star-shaped topology centered on the central cluster.

The computer device initially allows each segment to be assigned to a different temporary perimeter cluster including the center G. The computer device then merges the clusters using a greedy method that reduces the length of the travel path. For example, J. L.V.M. Stanislaus, and M. Younis, " Mobile Relays Based Federation Multiple Wireless Sensor Network Segments with Reduced-latency, " Proc. of Int'l Conf. on Comm. (ICC'13), Budapest, Hungary, June 2013, etc., the computer device can merge clusters and determine k-1 temporal neighbor clusters. The computer unit allocates the reference cell c for each segment to one cluster VC and merges two neighboring VCs having the minimum travel distance to visit all cells belonging to the cluster including the center G. [ The computer device repeats the merging process for the two VCs until the VC becomes k-1.

Figure 3 is an example of a temporary cluster formed in a region of interest. 3 shows an example in which 14 reference cells CS _T are grouped into five temporary clusters. Each temporary cluster has an identifier (VCID) for identifying the cluster. Figure 3 shows four temporary perimeter clusters containing the regions of the temporary center clusters VC ₅ and VC ₅ comprising the center (cell) of the region of interest.

Step 2: Form a final cluster based on the energy consumption of the relay node

The computer device forms the final cluster using the temporary cluster VC generated in the first step. The second step is to maximize the energy consumed for movement and data transmission / reception while moving within the cluster to which each relay node is allocated, assuming that the initial energy of the relay nodes is the same. The computer device changes the cell belonging to each temporary cluster so that the energy consumed by the relay node is appropriately distributed.

The computer device is first placed a temporary center cluster _k VC to the center cluster C _k, and sets the relay node M _k C _k that is responsible for a fixed RS. The computer device then changes the location of M _k based on the energy consumption of the relay node or extends the central cluster C _k to include other cells. That is, the final cluster is completed by extending the cluster starting with the temporary cluster.

Prior to the description, the terms used in each step are explained.

VCID (c) denotes an identifier (ID) of a temporary cluster including c ∈ CS _T.

ID (c) denotes an identifier of a cluster including c ∈ CS _T.

Anchor (C _i ) is a point at which a mobile relay carrying a cluster C _i visits and transmits information to a central cluster C _k . Anchor (C _i ) may be a point at which a fixed relay node M _k exists, It may also be a point where there is a storage device for temporarily storing information. Anchor (C _i ) is the point at which the relay node M _i connects with the central cluster C _k . Anchor (C _i ) corresponds to a cell in which a fixed M _k is placed or included in C _k . The computer device allocates one Anchor (C _i ) point for each cluster. Anchor (C _i ) may be the closest point in any cell belonging to cluster C _i among the reference cells belonging to C _k . Anchor (C _i ) can be changed while C _k is expanded or changed in the process of forming the final cluster.

TR (M _i ) means the shortest path through which the relay node M _i visits the segments belonging to the cluster C _i . That is, TR (M _i ) corresponds to the path where M _i starts from Anchor (C _i ) and visits all the reference cells belonging to C _i and returns. It may use daang a heuristic algorithm to determine the TR (M _i). For example, A. Kashyap, S. Khuller, and M. Shayman, "Relay placement for higher order connectivity in wireless sensor networks," Proc. of IEEE INFOCOM, Barcelona Spain, April 2006, and the like. Therefore, a detailed description of the process of determining TR ( _Mi ) will be omitted.

The second method is repeated in a rounded fashion, where E ^r _i is the energy consumed by the relay node M _i in one round r. The process of forming the final clusters are repeated while the relay node M TR _i (M _i) moving one time. One round is to move the relay node M TR _i (M _i) at a time, it refers to the process of the computer device to optimize the enlarged cluster. E ^r _i are the segments that belong to the relay node M _i, the cluster C _i segments and _{Anchor (C i) E C (} M i) of the energy uploading / downloading of information from and relay node M _i, the cluster C _i belongs to the And the energy E _M (M _i ) consumed for the motion to visit. E _M (M _i ) and E _C (M _i ) are calculated using Equations (3) and (4), respectively.

The process of forming the final cluster involves a greedy-expansion and optimization process according to the greedy algorithm. The process of forming the final cluster is performed by repeating the experience of the relay node moving and exchanging information. Hereinafter, a process of forming a final cluster will be described.

(1) Cluster expansion according to greedy algorithm

The computer device extends the cluster to equalize the energy consumed by the relay node starting from the temporary cluster VC. The computer device expands or changes clusters on the basis of the energy consumed by the relay node responsible for each cluster each time the visited path TR ( _Mi ) moves. Computer devices expand the cluster with the lowest energy consumption per round. The computer device repeats each round until the energy consumed by the relay node is no longer distributed. Or the computer device may repeat each round until the standard deviation of the energy of all the relay nodes is within a specified reference value.

In the first round (r = 1) the computer unit sets a temporary center cluster _k VC to the center cluster C _k. In the first round (r = 1), the computer device configures the surrounding cluster C _i to include the cell c _g where the relay node M _k is located in the central cluster C _k and the reference cell closest to the temporary surrounding cluster. The most recently included cell in cluster C _i is called RC (C _i ). For example, in the first round RC (C _k ) is c _g .

The relay node performs the process of moving the path and transmitting the information in each set cluster. The computer device then determines E ¹ _i based on the cluster set in the first round. The relay node M _k responsible for the central cluster C _k consumes energy only for transmitting and receiving information because it assumes a fixed state. Thus E ¹ _k is E _C (M _k ).

After the second round ( ^r ≥ ² ), the expansion process is performed for the cluster with the lowest energy in E ^r-1 _i (hereinafter referred to as C _least ). There are two possible scenarios for expansion:

(i) When C _least is _Ck : In this case there are two possible actions to control _Ck .

1) Relocate M _k to the cell closest to c _g (c _to ∈ CS _T ). That is, the position of the central cluster is changed. Thus, the surrounding clusters are also changed to include the new c _to .

② is recorded to the nearest cell c ∈ _T Add the CS to extend the C _k and the nearest cell c to RC (C _k). When C _k is expanded, the anchor (C _i ) of the surrounding cluster C _i can be changed. The new anchor (C _i ) is basically chosen so that the path through which M _i is responsible for C _i is minimized. Therefore, the nearest reference cell and the C _k in the extended C _i is changed so that the Anchor (C _i). If C _k may be extended to include a reference cell belonging to the prior art in accordance with C _i. In this case, it is added to the conventional C _i that are not included in the cell of the closest prior art reference cell in the cluster C _i. Also, in the conventional C _i , the closest reference cell among the extended C _k is changed to be anchor (C _i ).

(ii) when C _least is not C _k : that is, C _least is the surrounding cluster C _i . The computer device adds a reference cell that is closest to RC (C _i ) among the reference cells belonging to the temporary cluster VC _i and not included in another cluster to C _i . If all of the reference cells belonging to VC _i are included in another cluster, a reference cell that is closest to RC (C _i ) among the reference cells belonging to the temporary cluster neighboring VC _i and not belonging to the cluster is added to C _i . If there are no more cells to add, the expansion of the selected cluster C _least is terminated.

4 is an example of expanding a cluster according to a greedy algorithm. FIG. 4 is an example of expanding a cluster based on the temporary cluster of FIG.

4 (a) shows an example in which the cluster is set in the first round (r = 1). The computer device sets the temporary central cluster VC ₅ to the central cluster C ₅ . Therefore, C ₅ includes c (5,8). In addition, the computer apparatus selects a reference cell closest to the central cluster C ₅ (or the center G) in each temporary surrounding cluster to set the surrounding cluster. C ₁ includes c (2, 17) and c (5, 8). C ₂ includes c (6, 5) and c (5, 8). C ₃ includes c (6, 11) and c (5, 8). C ₄ includes c (5, 11) and c (5, 8).

After the second round, the computing device measures the energy consumed by relay nodes in each cluster to transmit information every round. The energy consumed by a relay node is the sum of the energy consumed to travel and the energy consumed to send and receive information in the cluster. For example, after the first round, the computer device measures the energy consumed by the relay node, assuming that the energy consumed in the central cluster C ₅ is the smallest. As described above, two scenarios are possible in this case. 4 (b) shows an example in which the position of the center cluster C ₅ is changed at r = 2. The center cluster C ₅ was changed to c (3, 9). As a result, E (M _k ) increases by the energy for data transmission / reception of segments covered by c (3, 9), that is, EC ( _Mi ). The computer apparatus changes each surrounding cluster to include a new reference cell c (3, 9) instead of c (5, 8) so that the surrounding cluster is connected to the changed central cluster.

After r = 2, the computing device measures the energy of the relay nodes consumed in each cluster. At this time, it is assumed that the central cluster C ₅ again consumes the minimum energy. 4 (c) is an example of r = 3. Computer apparatus added for the C ₅ closest to the reference cell, c (5,11) in the RC (C ₅₎ of reference cells that are not included in the current cluster to C _5. On the other hand, the selected c (5,11) was a cell belonging to C ₄ . Thus, the computing device extends C ₄ by adding c (5, 12) to C ₄ . Anchor (C _i ) can also be changed when the central cluster is expanded. Referring to Figure 4 (c) it can be seen that the C ₃ Anchor (C ₃₎ is altered to be c (5,11).

After r = 3, the computing device measures the energy of relay nodes consumed in each cluster. It is assumed that C ₃ and C ₄ consume the minimum energy. 4 (d) is an example of r = 4. The computer device adds to cells C ₃ and C _{4 the} cells closest to RC (C ₃ ) and RC (C ₄ ) of the reference cells not included in the current cluster for C ₃ and C _4, respectively. The added cells are recorded in RC (C ₃ ) and RC (C ₄ ), respectively.

After r = 4, the computing device measures the energy of relay nodes consumed in each cluster. It is assumed that C ₄ and C ₅ consume minimal energy. 4 (e) is an example of r = 5. Computer apparatus adds the nearest reference cell, c (1,11) in the RC (C ₄₎ of the reference cell to C _4. In addition, the computer device will add the nearest reference cell, c (6,5) in the C ₅ to RC (C ₅₎ of the reference cell. In this case c (6,5) is added to the new cell c (9,5) for a cell so belonged to C _2, a cluster C _2, and changing the Anchor (C ₂₎ to c (6,5). And it records the cell c (9,5) to RC (C _2).

After r = 5, the computing device measures the energy of the relay nodes consumed in each cluster. It is assumed that C ₂ consumes the minimum energy. Fig. 4 (f) is an example of r = 6. Computer apparatus adds the nearest reference cell, c (9,4) in which the C ₂ c (9,5) of the reference cell RC (C _2).

Fig. 5 is an example of expanding a cluster following Fig. 4 (f). After r = 6, the computing device measures the energy of relay nodes consumed in each cluster. At this time, it is assumed that C ₂ consumes minimum energy. 5 (a) shows an example of r = 7. Computer apparatus adds the nearest reference cell, c (4,1) in which the C ₂ c (9,4) of the reference cell RC (C _2). Then, the cell c (4, ₁ ) is recorded as RC (C ₂ ).

After r = 7, the computing device measures the energy of the relay nodes consumed in each cluster. It is assumed that C ₃ consumes the minimum energy. 5 (b) is an example of r = 8. Computer apparatus adds the nearest reference cell, c (9,14) in the C ₃ to c (9,11) of the reference cell RC (C _3). Then, the cell c (9, ₁₄ ) is recorded as RC (C ₃ ).

After r = 8, the computing device measures the energy of relay nodes consumed in each cluster. It is assumed that C ₁ consumes the minimum energy. 5 (c) is an example of r = 9. The computer device adds c (3,3), which is the reference cell closest to c (1,5), RC (C ₁ ), to C ₁ . Since there are no more clusters to add, the process of determining the final cluster ends.

(2) Cluster optimization

Cluster expansion due to greedy algorithms can result in energy imbalance between clusters. Therefore, the computer system performs a process of energy consumption of the relay node is to match the energy balance between the lowest cluster C and _least energy consumption of the relay node number of the cluster C from the _most close to the cluster. The process of optimizing the cluster reduces the C _most and the C _least for each round. The optimization process is divided into the following two cases.

(i) When C _k is not C _most and C _least :

First, the computer device shrinks C _most . Selecting a computer device to the nearest cell to Anchor (C _most) of one reference cell in the _most C _in c and c except _in at _most C. That is, C _maxim is reduced. E (C _most ) is reduced by the sum of E _C (energy consumed in exchanging information between c _in and Anchor (C _most )) and E _M (energy consumed in path travel up to c _in ).

The computer device adds c _in to C _k . And a computer device extends the _least C to select the nearest cell, c _out to Anchor (C _least), and add c to the C _out at _least C _k. E (C _least ) is increased by the sum of E _C (energy consumed for information exchange between c _out and Anchor (C _least )) and E _M (energy consumed for path movement to c _out ).

C _k is added to c _in , and c _out is excluded. Therefore, E (C _k ) is also changed according to the added reference cell c _in and the excluded reference cell c _out . If in the state where C _k there is only one having a c _{out least} a C takes a c _out, the computer apparatus to move _in the c M c _k from _out repositions the C _k.

Furthermore, anchor (C _i ) of the surrounding cluster C _i can be changed while c _in is added to c _k and c _out is excluded. Basically, anchor (C _i ) is the closest cell of C _i among the reference cells of C _k .

(ii) if C _k is C _most or C _least :

The case where C _k is C _least is explained first. First, the computer device shrinks C _most . Selecting a computer device to the nearest cell to Anchor (C _most) of one reference cell in the _most C _in c and c except _in at _most C. And the computer device adds c _in to C _least C _k . This expanded C _k and C _max .

The case where C _k is C _most will be described. Computer apparatus in C _k, select the nearest cell, c _out in the Anchor (C _least), and, except for the c _out in C _k add c _out the C _least by extending the C _least, and reducing the C _k. Or more computer devices to the nearest reference cell in the Anchor (C _least) in C _k, and the _least C, except Anchor (C _least) in C _k.

Furthermore, while C _k is changed can be changed around the cluster C _i Anchor (C _i). Basically, anchor (C _i ) is the closest cell of C _i among the reference cells of C _k .

The computing device computes E ^r _i and repeats the cluster optimization process based on the results computed every round. The computer device may stop the optimization process if the standard deviation of the energy consumption of each cluster is below the reference value. Or the computing device may stop the optimization process if the standard deviation of the energy consumption in round r is not less than the standard deviation of the energy consumption in the immediately preceding round r-1. That is, the computer device may repeat the optimization process to the point where no further optimization is performed.

FIG. 6 is an example in which an expanded cluster is optimized according to a greedy algorithm. Fig. 6A is an example in which the final cluster shown in Fig. 5C is optimized according to an optimization algorithm. FIG. 6 (a) is an optimization result in the case of FIG. 5 (c) where the cluster C ₃ is the cluster C _least having the smallest energy consumption and C ₅ is the cluster C _most having the largest energy consumption. 6 (b), the final cluster is simplified, and the path TR ( _Mi ) in which the relay node moves in each cluster is indicated by a dotted line.

The foregoing is summarized. FIG. 7 is an example of a flowchart of a method 100 for determining a route of a relay node for a divided network. The computer device first models (110) an area where a plurality of segments are located, in a grid form composed of a plurality of cells. The computer device determines 120 a reference cell for each segment of the plurality of cells based on information indicating a communicable segment when a relay node is located for each cell. At this time, the total number of selected reference cells is selected to be minimum.

The computer device sets up a temporary central cluster including the center of the area, views the reference cells as one cluster, visits the reference cells included in the two clusters including the temporary central cluster, and merges the clusters having the shortest- (K-1) temporary neighboring clusters connected to the temporary central cluster are set (130).

The computer device determines the initial temporary central cluster as a central cluster based on the information about the temporary cluster, and determines k-1 surrounding clusters including the closest cell among the reference cells belonging to the central cluster for each surrounding temporary cluster (140). Then, the energy of the relay node consumed in each cluster is measured, and the cluster having the smallest energy consumption is expanded (150). The process of extending the cluster is repeated (160 and 150) until no more relay nodes exist to be added to the cluster. Furthermore, the computing device may optimize the cluster to maximize the energy of the relay nodes consuming the final cluster in each cluster (170).

The present embodiment and drawings attached hereto are only a part of the technical idea included in the above-described technology, and it is easy for a person skilled in the art to easily understand the technical idea included in the description of the above- It will be appreciated that variations that may be deduced and specific embodiments are included within the scope of the foregoing description.

Claims (12)

The computer device modeling an area where a plurality of segments are located in a grid form comprising a plurality of cells;
Wherein the computer apparatus determines one reference cell for each segment of the plurality of cells using information on a communicable segment when a relay node is located in the cell so that the total number of determined reference cells is minimized step;
Grouping the plurality of segments into k temporary clusters using the reference cell; And
Grouping the plurality of segments into k final clusters based on the energy consumed by the relay node that conveys information as the computer apparatus moves through the temporary clusters and clusters, Wherein the path connecting the plurality of reference cells included in each cluster in the final cluster is a path through which the relay node moves. The method according to claim 1,
The communication radius of the relay node is R, and the length of the two sides of the cell is

A method for determining a movement path of a relay node for a divided network. The method according to claim 1,
Wherein the computer device determines a first cell having the largest number of segments that can be communicated when the relay node is located among the plurality of cells as the reference cell for the communicable segment, . The method of claim 3,
The computer device according to claim 1, wherein when the number of the first cells for any one of the plurality of segments is plural, a second cell of the first cell, which has the largest number of segments communicable with the first cell, To the reference cell for any one of the segments. 5. The method of claim 4,
Wherein the computer device determines a cell closest to the center of the second cell among the second cells as the reference cell for the one segment if the second cell for the one segment is a plurality of segments, A method for determining the movement path of a relay node. The method according to claim 1,
Wherein the temporary cluster includes one central cluster and k-1 neighboring clusters respectively connected to the central cluster,
Wherein the computer apparatus determines a cell including the center of the region as the center cluster,
After each of the reference cells is set as one temporary cluster, until the provisional cluster becomes k-1, two of the plurality of temporary clusters including the center and the minimum travel distance passing through the cells belonging to the two clusters, And determining the k-1 neighboring clusters by merging the temporary clusters to determine the movement path of the relay node for the divided network. The method according to claim 1,
Wherein the computer device is configured to generate a final cluster by changing a cell belonging to the temporary cluster until the deviation of the energy consumed by the mobile relay moving in the k temporary clusters while moving the information is within a reference value, A method for determining the path of a relay node for a network. The method according to claim 1,
Wherein the temporary cluster includes one provisional central cluster including a center cell in which the center of the region is located and k-1 temporary neighboring clusters connected to the provisional central cluster,
Wherein grouping into the final cluster comprises:
The computer apparatus sets the temporary central cluster as a central cluster and sets k-1 neighboring clusters including one cell closest to the center cell among the reference cells belonging to the temporary surrounding cluster for each of the temporary surrounding clusters ; And
Wherein the computer apparatus measures energy consumed by the relay node in each of the central cluster and the neighboring cluster to transmit information and changes the cluster having the smallest energy consumed by the relay node among the central cluster and the neighboring cluster And generating the (k-1) final clusters. 9. The method of claim 8,
Wherein when the cluster having the smallest energy is the central cluster, the computer apparatus sets a new central cluster closest to the reference cluster in the central cluster, and the peripheral cluster includes a cell belonging to the new central cluster A method for determining the path of a relay node for a network. 9. The method of claim 8,
If the cluster having the smallest energy is the center cluster, the computer apparatus adds the reference cell closest to the center cluster among the reference cells not belonging to the center cluster and the surrounding cluster to the central cluster,
Wherein the computer apparatus sets a reference cell closest to the neighboring cluster among the reference cells belonging to the center cluster to each of the neighboring clusters as a point at which the mobile relay exchanges information, . 9. The method of claim 8,
Wherein when the cluster having the smallest energy is the peripheral cluster, the computer apparatus adds a cell closest to the peripheral cluster having the smallest energy among the reference cells not belonging to the central cluster and the peripheral cluster, Of the moving path. The method according to claim 1,
Wherein the computer device measures energy consumed by each relay node in the final cluster to transmit information, and the computer device calculates the energy consumed by the cluster and the relay node, which are consumed by the relay node in the final cluster, Determining a smallest cluster and adding any one of the reference cells to the cluster with the smallest energy consumption, excluding the one reference cell in the cluster with the highest energy consumption, A method for determining a movement path of a relay node for a relay node.

Images