CN111669767B - Sensor network dynamic deployment method - Google Patents

Abstract

The invention proposes a dynamic deployment method of a sensor network. First, locate the sensors in the non-key monitoring area and select them as mobile nodes. Secondly, the concentric circle method is used to find the target nodes in the key monitoring areas. Finally, the cascade moving path to move the mobile node to the target node is calculated according to the Floyd algorithm, thus realizing the dynamic deployment of the water quality sensor network based on the regional characteristic model. The present invention realizes the dynamic deployment of key monitoring areas of the sensor network in a shorter period of time through the cascading movement of the sensor nodes on the premise of ensuring the network life cycle, so as to achieve better coverage performance.

Description

A dynamic deployment method for sensor networks

Technical Field

The present invention relates to the field of environmental monitoring and sensor networks, and in particular to a study of a sensor network dynamic deployment method.

Background Art

With the development of electronic communication technology, sensor networks are increasingly used in environmental monitoring, smart home, medical, military and other fields. From the perspective of rational use of resources and ensuring the accuracy of monitoring data, the characteristics of sensors that can be moved are used. Under the condition of a limited number of sensors, sensors in non-key monitoring areas are moved to key monitoring areas for real-time monitoring. Complete network coverage of the monitoring area is particularly critical for the deployment of wireless sensors, so more effective coverage of key monitoring areas must be guaranteed.

When coverage holes appear in the monitoring area, the node communication capability, information calculation and processing capability, and network life cycle of the current sensor network will be affected accordingly. Therefore, it is necessary to properly repair the coverage holes in the sensor nodes to ensure that the network's perception and monitoring capabilities of information will not be affected accordingly. Furthermore, the appearance of coverage holes will cause the failure of its adjacent nodes, damage the reliability of the entire network, and then cause a large number of nodes in the network to not be fully utilized. Therefore, in order to ensure the coverage area, service quality, and sufficient allocation of network resources in the monitoring area, it is necessary to promptly and effectively repair the holes formed in the network.

Therefore, the effective deployment of sensor networks can be achieved through the mobility strategy of sensor nodes, which can provide a substantial theoretical basis for accurate environmental monitoring.

Summary of the invention

The purpose of the present invention is to propose a sensor network dynamic deployment method, which can provide a theoretical basis for the deployment of sensor networks and can be widely used in environmental monitoring, smart home and other fields.

To achieve the above object, the present invention proposes a sensor network dynamic deployment method, which specifically includes four basic steps: establishing a monitoring area model, determining mobile node coordinates, determining target node coordinates, and cascading movement of sensor nodes.

Step 1: In one embodiment of the present invention, the step of establishing a monitoring area model further comprises:

The monitoring area includes key monitoring areas and non-key monitoring areas. The monitoring area is discretized into M grid points, where the coordinates of any grid point p _j are (x _j , y _j ). A group of sensor nodes with the same sensing radius r are uniformly deployed in the monitoring area. Let s = {s ₁ , s ₂ , s ₃ …s _n } represent the set of sensor nodes, where the coordinates of any sensor node _si are ( _xi , _yi ); the Euclidean distance from _si to point p _j is defined as:

Then the situation where a grid point _pj in the monitoring area is covered by a sensor node is:

P(s _i ,p _j )=1 indicates that the grid point can be covered by the sensor node.

Step 2: In one embodiment of the present invention, determining the coordinates of the mobile node further includes:

The coordinates of the mobile node are the coordinates of the sensor nodes in the non-key monitoring area. Since the range of the non-key monitoring area is known, it is only necessary to find the node coordinates of the corresponding sensor in the non-key monitoring area.

Step 3, in one embodiment of the present invention, determining the target node coordinates further includes:

The target node coordinates are the coordinates of the uncovered grid points in the key monitoring area. The specific steps are as follows:

(1) Determine the number of mobile nodes N according to the previous step;

(2) Find the area not covered by the sensor in the key monitoring area, traverse all the grid nodes of the area not covered by the sensor, take all the nodes as the center of the circle, and use the width of the grid as the initial radius. Then, make concentric circles outward with the grid width as the increasing radius. The maximum radius of the concentric circles is the sensing radius of the sensor.

(3) When the largest ring of the concentric circle overlaps with the original sensor coverage area, stop increasing the radius of the concentric circle and record the center position and radius size at this time;

(4) Let Q = {q ₁ ,q ₂ ,...q _m } be the set of concentric circles of all grid nodes in the area not covered by the sensor. Find the circle with the largest radius in the set. If the maximum radius is the same, take the circle with the smaller sum of the horizontal and vertical coordinates of the center position. Record the size of the node and radius of the circle at this time, and merge the area where the circle is located into the sensor coverage area.

Repeat steps (2), (3), and (4) in sequence. When the number of circles determined is equal to N, the process ends. At this time, the center positions of the circles determined in sequence are the positions of the target nodes.

Step 4: In one embodiment of the present invention, the cascade movement of the sensor nodes further comprises:

After determining the coordinates of the mobile node and the target node, it is necessary to decide how to move the mobile node to the target node; introduce the Floyd algorithm into the cascade movement of sensors, take the sensor nodes in the non-key monitoring area and the target node in the key monitoring area as the starting point and the end point, and take the remaining sensors as the path algorithm nodes;

Set up a weighted directed graph G = (V, E, C), where the sensor node set V, the node connection line set E, and the distance adjacency matrix C; the elements v ₀ ,v ₁ ,…,v _l in the sensor node V represent l sensor nodes; the node connection line set E is formed by connecting the nodes in V, where the element e _k = [v _i v _j ], and the node _vi is connected to the node v _j ; C is the distance adjacency matrix corresponding to the graph G, and its elements c _ij are shown as follows:

表示传感器网络模型中两个相连节点v_i和v_j之间的距离，用下式表示，By constructing a directed network graph to simulate the sensor network, the shortest path between the determined mobile node and the target node is planned by the Floyd algorithm. The basic idea of the Floyd algorithm is to construct v matrices D ⁽¹⁾ , D ⁽²⁾ , ..., D ^(v) in sequence by inserting item points in the weighted adjacency matrix of the graph, so that the final matrix D ^(v) becomes the distance matrix of the graph, and the insertion point matrix is also calculated to obtain the shortest path between two points. Among them, the elements of D ⁽⁰⁾

It represents the distance between two connected nodes _vi and _vj in the sensor network model, which is expressed as follows:

The elements in D ^(k) are calculated using the following formula:

The elements in R ^(k) are calculated using the following formula:

Using the distance adjacency matrix C of graph G and the Floyd algorithm, we get the shortest distance matrix D ⁽ⁿ⁾ and the shortest path matrix R ⁽ⁿ⁾ . R ⁽ⁿ⁾ only stores the shortest path between the mobile node and the target node. On this basis, we select the shortest cascade path. Finally, we achieve effective deployment of key monitoring areas while ensuring the network life cycle, thereby improving the network's monitoring capabilities.

Figure 2 shows the monitoring area model. Sensor nodes are evenly deployed in the monitoring area. The area surrounded by black boxes represents the key monitoring area, and the area surrounded by light gray boxes represents the non-key monitoring area. The network deployment diagram after cascade movement is shown in Figure 3. It can be seen that the sensors in the non-key monitoring area are moved to the key monitoring area, which greatly increases the coverage of the key monitoring area. Figure 4 is a comparison of the total moving distance of sensor nodes using direct movement and cascade movement. Direct movement means directly moving the nodes in the non-key monitoring area to the key monitoring area without passing through the intermediate nodes. It can be seen from the figure that when the sensor nodes reach a new balance, the total moving distance of cascade movement is slightly higher than that of direct movement. This is related to the movement of intermediate nodes involved in cascade movement. However, compared with the time required for the network to reach equilibrium, cascade movement takes 23.963s and direct movement takes 35.012s. Therefore, compared with direct movement, the time for cascade movement to reach the equilibrium point is much lower than that of direct movement. Using cascade movement can effectively reduce the time for network adjustment. Figure 5 is a comparison of the residual energy of cascaded and direct mobile nodes. It can be seen from Figure 5 that although direct mobile can achieve key coverage of the area with fewer nodes, the residual energy of the nodes after the move is low and cannot continue to perform the water quality monitoring task well. Cascade mobile can enable more nodes to participate and share the energy consumption during the movement of sensor nodes. When the network reaches a new balance, the residual energy of the nodes is relatively balanced, which can effectively extend the network life.

The present invention provides a sensor network dynamic deployment method, which can realize effective monitoring of the monitoring area and provide a substantial theoretical basis for effective monitoring and comprehensive management of the environment.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of a sensor network dynamic deployment method according to an embodiment of the present invention;

FIG2 is a monitoring area model according to an embodiment of the present invention;

FIG3 is a diagram of a network deployment after cascade movement according to an embodiment of the present invention;

FIG4 is a curve showing movement distance versus time according to an embodiment of the present invention;

FIG5 is a diagram comparing the residual energy of nodes according to an embodiment of the present invention.

DETAILED DESCRIPTION

Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar meanings. The embodiments described below are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

The present invention proposes a sensor network dynamic deployment method for complex monitoring environments during environmental monitoring.

In order to have a clearer understanding of the present invention, a brief description is given here. The present invention includes four basic steps: step one, establishing a monitoring area model; step two, determining the coordinates of the mobile node; step three, determining the coordinates of the target node; step four, cascade movement of the sensor node.

Specifically, FIG1 is a flow chart of a sensor network dynamic deployment method according to an embodiment of the present invention, comprising the following steps:

Step S101, establishing a monitoring area model.

In one embodiment of the present invention, the monitoring area includes a key monitoring area and a non-key monitoring area. The monitoring area is discretized into M grid points, wherein the coordinates of any grid point p _j are (x _j , y _j ). A group of sensor nodes with the same sensing radius r are uniformly deployed in the monitoring area. Let s = {s ₁ , s ₂ , s ₃ …s _n } represent the set of sensor nodes, wherein the coordinates of any sensor node _si are ( _xi , _yi ). The Euclidean distance from _si to point p _j is defined as:

Then the situation where a grid point _pj in the monitoring area is covered by a sensor node is:

P(s _i ,p _j )＝1 indicates that the grid point can be covered by the sensor node;

Step S102, determining the coordinates of the mobile node.

In one embodiment of the present invention, the mobile node coordinates are the sensor node coordinates in the non-key monitoring area. Since the range of the non-key monitoring area is known, it is only necessary to find the node coordinates of the corresponding sensor in the non-key monitoring area.

Step S103, determining the target node coordinates.

In one embodiment of the present invention, the target node coordinates are the coordinates of the uncovered grid points in the key monitoring area, and the specific steps are as follows:

(1) Determine the number of mobile nodes N according to the previous step;

(2) Find the area not covered by the sensor in the key monitoring area, traverse all the grid nodes of the area not covered by the sensor, take all the nodes as the center of the circle, and use the width of the grid as the initial radius. Then, make concentric circles outward with the grid width as the increasing radius. The maximum radius of the concentric circles is the sensing radius of the sensor.

(3) When the largest ring of the concentric circle overlaps with the original sensor coverage area, stop increasing the radius of the concentric circle and record the center position and radius size at this time;

(4) Let Q = {q ₁ ,q ₂ ,...q _m } be the set of concentric circles of all grid nodes in the area not covered by the sensor. Find the circle with the largest radius in the set. If the maximum radius is the same, take the circle with the smaller sum of the horizontal and vertical coordinates of the center position. Record the size of the node and radius of the circle at this time, and merge the area where the circle is located into the sensor coverage area.

Repeat steps (2), (3), and (4) in sequence. When the number of circles determined is equal to N, the process ends. At this time, the center positions of the circles determined in sequence are the positions of the target nodes.

Step S104: cascade movement of sensor nodes.

In one embodiment of the present invention, after determining the coordinates of the mobile node and the target node, it is necessary to decide how to move the mobile node to the target node; the Floyd algorithm is introduced into the cascade movement of the sensor, the sensor nodes in the non-key monitoring area and the target nodes in the key monitoring area are used as the starting point and the end point, and the remaining sensors are used as the path algorithm nodes.

Suppose a weighted directed graph G = (V, E, C), where the sensor node set V, the node connection line set E, and the distance adjacency matrix C; the elements v ₀ ,v ₁ ,…,v _l in the sensor node V represent l sensor nodes; the node connection line set E is formed by connecting the nodes in V, where the element e _k = [v _i v _j ], and the node _vi is connected to the node v _j ; C is the distance adjacency matrix corresponding to the graph G, and its elements c _ij are shown in formula (3):

表示传感器网络模型中两个相连节点v_i和v_j之间的距离，用公式(4)表示，D^(k)、R^(k)的计算为公式(5)，公式(6)：By constructing a directed network graph to simulate the sensor network, the shortest path between the determined mobile node and the target node is planned by the Floyd algorithm. The basic idea of the Floyd algorithm is to construct v matrices D ⁽¹⁾ , D ⁽²⁾ , ..., D ^(v) in sequence by inserting points in the weighted adjacency matrix of the graph, so that the final matrix D ^(v) becomes the distance matrix of the graph, and the insertion point matrix is also calculated to obtain the shortest path between two points. Among them, the elements of D ⁽⁰⁾

The distance between two connected nodes v _i and v _j in the sensor network model is expressed by formula (4). The calculation of D ^(k) and R ^(k) is as follows:

Using the distance adjacency matrix C of graph G and the Floyd algorithm, we get the shortest distance matrix D ⁽ⁿ⁾ and the shortest path matrix R ⁽ⁿ⁾ . R ⁽ⁿ⁾ only stores the shortest path between the mobile node and the target node. On this basis, we select the shortest cascade path. Finally, we achieve effective deployment of key monitoring areas while ensuring the network life cycle, thereby improving the network's monitoring capabilities.

The sensor network dynamic deployment method proposed in the present invention can realize the optimized deployment of the sensor network and provide a substantial theoretical basis for the effective monitoring and comprehensive management of the environment.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some of the technical features therein by equivalents, and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. The scope of the present invention is defined by the attached claims and their equivalents.

Claims (1)

1. A dynamic deployment method of a sensor network is characterized in that: the method comprises the following steps of establishing a monitoring area model, determining a mobile node coordinate, determining a target node coordinate and cascading movement of a sensor node;

the establishing the monitoring area model comprises the following steps: the monitoring area comprises a key monitoring area and a non-key monitoring area, discretization processing is carried out on the monitoring area, and the monitoring area is discretized into M grid points, wherein any grid point p _j Is (x) _j ，y _j ) Uniformly deploying a group of sensor nodes with the same sensing radius r in a monitoring area, and setting s= { s ₁ ，s ₂ ，s ₃ …s _n And represents the set of sensor nodes, any one of which is s _i Is (x) _i ，y _i ) The method comprises the steps of carrying out a first treatment on the surface of the Calculation s _i To point p _j The euclidean distance of (2) is defined as:

then a certain grid point p in the area is monitored _j The cases covered by the sensor nodes are:

P(s _i ,p _j ) =1 indicates that the grid point can be covered by a sensor node;

the determining mobile node coordinates includes: the mobile node coordinates are sensor node coordinates in the non-key monitoring area, and as the range of the non-key monitoring area is known, the node coordinates of the corresponding sensor are only needed to be searched in the non-key monitoring area;

the determining the target node coordinates includes: the target node coordinates are grid point coordinates which are not covered in the key monitoring area; the method specifically comprises the following steps:

(1) Determining the number N of mobile nodes;

(2) Finding out areas which are not covered by the sensor in the key monitoring areas, traversing all grid nodes of the areas which are not covered by the sensor, taking all nodes as circle centers, taking the width of the grid as an initial radius, and taking the width of the grid as an incremental radius to make concentric circles outwards, wherein the maximum radius of the concentric circles is the sensing radius of the sensor;

(3) Stopping the increase of the radius of the concentric circle when the maximum ring of the concentric circle is overlapped with the coverage area of the original sensor, and recording the position of the circle center and the radius at the moment;

(4) Let q= { Q ₁ ,q ₂ ,...q _m The method comprises finding out the circle with the largest radius from the set of concentric circles of all grid nodes not covered by the sensor, if the largest radius is the same, taking the circle with smaller sum of the horizontal and vertical coordinates of the circle center position,

recording the node and radius of the circle at the moment, and merging the area where the circle is located into a sensor coverage area;

(5) Sequentially repeating the steps (2), (3) and (4), ending the process when the number of circles to be determined is equal to N, and sequentially determining the positions of the centers of circles to be determined to be the positions of the target nodes;

the cascading movement of the sensor nodes includes: after determining the coordinates of the mobile node and the target node, a decision is made as to how to move the mobile node over the target node; introducing a Floyd algorithm into cascade movement of the sensors, taking sensor nodes of non-key monitoring areas and target nodes in key monitoring areas as starting points and end points, and taking the rest sensors as path algorithm nodes;

setting a weighted directed graph g= (V, E, C), wherein the set of sensor nodes V, the set of node communication lines E, and the distance adjacency matrix C; element V in sensor node V ₀ ,v ₁ ,…,v _l Representing l sensor nodes; the node communication line set E is formed by connecting nodes in V, wherein the element E _k ＝[v _i v _j ]Node v _i And node v _j Are connected with each other; c is the distance adjacency matrix corresponding to the graph G, and the element C thereof _ij As shown in formula (3):

simulating a sensor network by constructing a directed network diagram, and planning the determined shortest path between the mobile node and the target node by using a Fluedel algorithm; the basic idea of the Floyd algorithm is to construct v matrices D in turn by inserting entry points directly in weighted adjacency matrices of the graph ⁽¹⁾ 、D ⁽²⁾ 、…、D ^(v) So that the resulting matrix D ^(v) Forming a distance matrix of the graph, and simultaneously, solving an insertion point matrix to obtain the shortest path between two points; wherein D is ⁽⁰⁾ Elements of (2)

Representing two connected nodes v in a sensor network model _i And v _j The distance between them is expressed by the formula (4), D ^(k) 、R ^(k) The calculation of the element in the formula (5), the formula (6):

obtaining a shortest distance matrix D by using a distance adjacent matrix C of the graph G and through a Floride algorithm ⁽ⁿ⁾ And a shortest path matrix R ⁽ⁿ⁾ ，R ⁽ⁿ⁾ Only the shortest path between the mobile node and the target node is saved, and on the basis, the shortest cascade path is selected; finally, on the premise of ensuring the life cycle of the network, the effective deployment of the heavy point monitoring area is realized, and the network is improvedMonitoring the capacity.

Images