Improved DV-Hop positioning method based on nearest neighbor anchor node path connection relation

Abstract

The invention discloses an improved DV-Hop positioning method based on a nearest neighbor anchor node path connection relation, and belongs to the technical field of wireless sensor networks, and the method comprises the following steps: network nodes are initialized, and each anchor node calculates the average hop distance of the shortest path from other anchor nodes recorded in S1; according to the method, the path connection relation between the nearest neighbor anchor node of the unknown node and the other anchor nodes in the network is approximate to the path connection relation between the unknown node and the anchor nodes in the network, and the three anchor nodes closest to the linear path connection are screened through the average hop distance value to participate in the positioning of the unknown node. The method for estimating the distance between the unknown node and the anchor node participating in positioning is optimized, and the distance estimation precision is improved; besides, the multi-target particle swarm algorithm is used for replacing the original least square algorithm to realize the positioning of the unknown node, the distance constraint condition is fully utilized, the target function setting is more reasonable, and the overall positioning precision is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of wireless sensor networks, and in particular relates to an improved DV-Hop positioning method based on a nearest neighbor anchor node path connection relationship. Background Art

[0002] Wireless sensor network is a network technology that integrates sensors, wireless communication, microcomputer electronic technology, etc. for real-time monitoring and perception. Wireless sensor networks can be used to monitor temperature, sound, vibration, pressure, movement or pollutants within a certain range. They are widely used in military and civilian fields, such as environmental monitoring, industrial process monitoring, medical treatment, road traffic control, smart homes and cities, office automation, etc. Node positioning accuracy is increasingly becoming one of the core technical issues affecting the efficiency of wireless sensor networks, because the information uploaded by sensor nodes is only valuable when the location is accurate.

[0003] Wireless sensor networks are composed of two types of sensor nodes: anchor nodes and unknown nodes. Anchor nodes determine their location information by equipping GPS modules and other means. Due to cost constraints, anchor nodes only account for a small part of the total number of nodes; the remaining nodes are called unknown nodes and rely on anchor nodes and positioning algorithms to determine their own locations. Existing node positioning models can be divided into two categories according to whether it is necessary to directly measure the distance between nodes through hardware configuration: models based on direct distance measurement and models without direct distance measurement.

[0004] DV-Hop is a classic node positioning model that does not require direct ranging. It realizes node positioning based on the multi-hop connection relationship between nodes. It has the advantages of saving network resources and being easy to implement. However, its positioning error is large when facing anisotropic networks, which makes it unable to meet the changing application demand scenarios.

[0005] The main sources of error in the DV-Hop model include: (1) Hop count error: The number of hops between nodes is determined by the communication radius. The number of hops between a node and the other nodes within a communication radius is counted as 1 hop. If the communication radius is exceeded, the minimum hop count is counted with the help of multi-hop forwarding by neighbor nodes. Such a hop count statistical method obviously has errors; (2) Estimated distance error: The distance between two points is determined according to the straight-line distance determination algorithm, which is obtained by multiplying the average hop distance by the number of hops. However, in practical applications, sensor nodes are randomly distributed, and the minimum hop path between two nodes is not a strict straight line, which will cause errors in the calculated distance; (3) Error in the calculation method of unknown node position: The least squares algorithm used in the original model is an algorithm with simple principle and small computational complexity, but it is sensitive to distance error and the graphical relationship between anchor nodes. The positioning result is relatively accurate only when the distance estimation accuracy is high and the anchor nodes are not approximately collinear. If the estimated distance is inaccurate or the anchor node positions are approximately collinear, it will lead to large positioning errors. In addition, DV-Hop selects three or more anchor nodes with the closest hop distance to participate in the positioning calculation of unknown nodes, which cannot ensure the accuracy of the distance between the anchor nodes and the unknown nodes and reasonable graphic conditions. This is also an important factor that leads to the large positioning error of the DV-Hop model. In order to solve this problem, an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes is urgently needed. Summary of the invention

[0006] The purpose of the present invention is to solve the above-mentioned problem and to propose an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes.

[0007] In order to achieve the above object, the present invention adopts the following technical solution: an improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship, comprising the following steps:

[0008] S1, network node initialization;

[0009] S2, each anchor node calculates the average hop distance of the shortest path to other anchor nodes recorded in S1;

[0010] S3, each unknown node selects the nearest neighbor anchor node, and performs positioning calculation according to the path information recorded by the nearest neighbor anchor node in S1 and S2. In this step, the unknown node completes the selection of the anchor node set participating in its own positioning calculation and the distance estimation;

[0011] S4, locate the unknown nodes whose anchor nodes have been selected and distances estimated in S3;

[0012] S5. Obtain the final coordinates (x, y) of u.

[0013] As a further description of the above technical solution:

[0014] In S1, the network nodes are initialized, and a number of network nodes are randomly deployed in the set monitoring area. The number of network nodes includes anchor nodes and unknown nodes. Specifically, sensor nodes are randomly deployed in the specified area, among which a certain proportion of nodes can know their own positions through measurement or carrying positioning devices such as GPS, which are called anchor nodes, and the remaining nodes to be located are called unknown nodes.

[0015] As a further description of the above technical solution:

[0016] In S3, each unknown node selects the nearest neighbor anchor node, and performs positioning calculation according to the path information recorded by the nearest neighbor anchor node in S1 and S2. In this step, the unknown node completes the selection and distance estimation of the anchor node set participating in its own positioning calculation. According to the above idea, step S3 is specifically described as: for a certain unknown node, select the anchor node with the smallest number of hops, that is, the nearest neighbor anchor node, sort the average hop distances recorded by the nearest neighbor anchor node to the remaining anchor nodes in order from large to small, select the remaining anchor nodes corresponding to the top 3 average hop distances to participate in the positioning of the unknown node, and add them to the positioning anchor node set of the unknown node.

[0017] As a further description of the above technical solution:

[0018] In S3, if there is more than one nearest neighbor anchor node for the unknown node, the average hop distances recorded by all nearest neighbor anchor nodes to the remaining anchor nodes are sorted in descending order, and then the remaining anchor nodes corresponding to the top 3 average hop distances are selected and added to the positioning anchor node set of the unknown node, and then, based on the existing nearest neighbor anchor nodes and the anchor nodes in the positioning anchor node set.

[0019] As a further description of the above technical solution:

[0020] In S4, the unknown node for which the positioning anchor node selection and distance estimation have been completed in S3 is estimated using the Min-Max algorithm to estimate an initial position rectangular range of the unknown node.

[0021] As a further description of the above technical solution:

[0022] In S5, a multi-objective particle swarm optimization algorithm is used to obtain the final coordinates (x, y) of u.

[0023] As a further description of the above technical solution:

[0024] In S5, the positioning problem is modeled as a problem of finding the optimal solution of a nonlinear system of equations. The nonlinear system of equations is listed according to the Euclidean distance relationship between the unknown node and the three anchor nodes in the positioning anchor node combination. The nonlinear system of equations is solved using a multi-objective particle swarm optimization algorithm. Three objective functions are determined based on equations established based on the three Euclidean distances and coordinates in the system of equations.

[0025] As a further description of the above technical solution:

[0026] In S5, the initial values ​​of the x and y components of each particle are set to random values ​​within the rectangular position range obtained in S4, and the initial value of the di component is the respective distance values ​​obtained in the distance estimation step in S3; the Pareto optimal solution set is obtained by iterative calculation, and the global optimal solution of the particle is randomly obtained, which is the positioning value of the unknown node.

[0027] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0028] In the present invention, in the method, the path connection relationship between the unknown node and the anchor nodes in the network is approximated by the path connection relationship between the nearest neighbor anchor node of the unknown node and the remaining anchor nodes in the network, and the three anchor nodes closest to the straight path connection are selected by the average hop distance value to participate in the positioning of the unknown node, so that the distance estimation method between the unknown node and the anchor nodes participating in the positioning is optimized, and the distance estimation accuracy is improved; in addition, a multi-objective particle swarm algorithm is used instead of the original least squares algorithm to realize the positioning of the unknown node, which makes full use of the distance constraint conditions, and the objective function setting is more reasonable, thereby improving the overall positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The flowchart of an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes.

[0030] Figure 2 This is a diagram illustrating the algorithm for selecting the nearest neighbor anchor nodes and the corresponding set of anchor nodes participating in the positioning calculation in an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes.

[0031] Figure 3 This is a comparison chart of the impact of anchor node ratio on distance estimation accuracy in an improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship.

[0032] Figure 4 This is a comparison chart of the impact of the total number of nodes on the distance estimation accuracy in an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes.

[0033] Figure 5This is a comparison chart of the impact of communication radius on distance estimation accuracy in an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes.

[0034] Figure 6 This is a comparison chart of the impact of the anchor node ratio on positioning accuracy in an improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship.

[0035] Figure 7 This is a comparison chart of the impact of the total number of nodes on positioning accuracy in an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes.

[0036] Figure 8 This is a comparison chart of the impact of communication radius on positioning accuracy in an improved DV-Hop positioning method based on the path connection relationship of the nearest neighbor anchor nodes. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] See also Figure 1-8 The present invention provides a technical solution: an improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship, comprising the following steps:

[0039] S1: Network node initialization;

[0040] Specifically, several network nodes are randomly deployed in the set monitoring area, and the several network nodes include anchor nodes and unknown nodes. Specifically, sensor nodes are randomly deployed in the specified area, among which a certain proportion of nodes can know their own positions through measurement or carrying positioning devices such as GPS, which are called anchor nodes, and the remaining nodes to be located are called unknown nodes;

[0041] For example, the monitoring area is a square two-dimensional plane of 100 (m) × 100 (m); sensor nodes are randomly arranged in the monitoring area, all nodes are isomorphic, and all nodes have the same communication radius; in order to obtain objective and accurate experimental results, this example sets up three experimental scenarios: Scenario 1 is that 100 nodes are randomly distributed in the area, the node communication radius is set to 20m, and the proportion of anchor nodes is changed (10% to 40%); Scenario 2 is to set the communication radius to 20m, the proportion of anchor nodes is fixed to 20%, and the total number of nodes in the area is changed (100 to 400); Scenario 3 is to set the total number of nodes to 100, the proportion of anchor nodes is set to 20%, and the communication radius of the nodes is changed (20m to 40m);

[0042] Table 1 Scene parameter settings

[0043]

[0044] The distance vector exchange protocol algorithm is first implemented in three simulation experiment scenarios. Each node records the multi-hop shortest path information (including anchor node number and minimum hop number) from the anchor node in the network.

[0045] S2: Each anchor node calculates the average hop distance of the shortest path to other anchor nodes recorded in S1;

[0046] S3: Each unknown node selects the nearest neighbor anchor node and performs positioning calculation based on the path information recorded by the nearest neighbor anchor node in S1 and S2. In this step, the unknown node completes the selection of the anchor node set participating in its own positioning calculation and the distance estimation;

[0047] According to the above idea, step S3 is specifically described as follows: for a certain unknown node, select the anchor node with the smallest number of hops, that is, the nearest neighbor anchor node, sort the average hop distances recorded by the nearest neighbor anchor node to the remaining anchor nodes in order from large to small, select the remaining anchor nodes corresponding to the top 3 average hop distances to participate in the positioning of the unknown node, and add them to the positioning anchor node set of the unknown node; if there is more than one nearest neighbor anchor node of the unknown node, sort the average hop distances recorded by all nearest neighbor anchor nodes to the remaining anchor nodes in order from large to small, and then select the remaining anchor nodes corresponding to the top 3 average hop distances to add them to the positioning anchor node set of the unknown node; then, calculate the distance between the unknown node and the three positioning anchor nodes according to the average hop distance of the shortest path between the existing nearest neighbor anchor node and the anchor nodes in the positioning anchor node set;

[0048] by Figure 2Take as an example to specifically explain step S3, for example: suppose that two nearest neighbor anchor nodes of unknown node u are detected, namely A1 and A2, which are 1 hop away; the anchor nodes connected to A1 and A2 are sorted from large to small according to the average hop distance of the path, and the paths with the top three average hop distances are selected as A6-A1, A3-A2, and A4-A2, so {A3, A4, A6} become the positioning anchor node set of unknown node u; the average hop distance of the A6-A1 path is regarded as the average hop distance between the A6-u path, thereby estimating the distance from A6 to u; in this way, the distance between u and the remaining two beacons A3 and A4 in the positioning anchor node combination is estimated; so far, the calculation conditions for u to realize positioning are complete;

[0049] S4: For the unknown node whose positioning anchor node selection and distance estimation have been completed in S3, the Min-Max algorithm is used to estimate an initial position rectangular range of the unknown node;

[0050] S5: Use multi-objective particle swarm optimization algorithm to obtain the final coordinates (x, y) of u;

[0051] The positioning problem is modeled as a problem of finding the optimal solution of a nonlinear equation set. According to the Euclidean distance relationship between the unknown node and the three anchor nodes in the positioning anchor node combination, the following nonlinear equation set is listed:

[0052]

[0053] In the above formula, the unknown node coordinates (x, y) are the unknowns in the equation system, the known coordinates of the three anchor nodes are expressed as (X1, Y1), (X2, Y2) and (X3, Y3), and the distances between the unknown node u and the three anchor nodes are recorded as d1, d2, and d3 respectively;

[0054] The multi-objective particle swarm optimization algorithm is used to solve the nonlinear equations. According to the equations established by the three Euclidean distances and coordinates in the equations, three objective functions are determined:

[0055]

[0056] Set the initial values ​​of the x and y components of each particle to random values ​​within the rectangular position range obtained in S4, and the initial value of the di component to the distance values ​​obtained in the distance estimation step in S3; iteratively calculate the Pareto optimal solution set, and randomly obtain the global optimal solution of the particle, which is the positioning value of the unknown node.

[0057] In this embodiment, the path connection relationship between the unknown node and the anchor nodes in the network is approximated by the path connection relationship between the nearest neighbor anchor node of the unknown node and the remaining anchor nodes in the network, and the three anchor nodes closest to the straight path connection are selected to participate in the positioning of the unknown node through the average hop distance value, so that the distance estimation method between the unknown node and the anchor nodes participating in the positioning is optimized, and the distance estimation accuracy is improved; in addition, the multi-objective particle swarm algorithm is used instead of the original least squares algorithm to realize the positioning of the unknown node, which makes full use of the distance constraint conditions, and the objective function setting is more reasonable, thereby improving the overall positioning accuracy.

[0058] In addition to the embodiments, it should be noted that:

[0059] The following comparative experiments are conducted on the algorithms in steps S3 and S5 to prove that the improvement of the present application in the key step of node positioning has obvious effects.

[0060] In the DV-Hop improved positioning strategy proposed in this invention, the distance between the unknown node and the anchor node is a necessary calculation condition for implementing the unknown node positioning, so distance estimation is a necessary step to achieve the unknown node positioning, and the distance estimation accuracy is a decisive factor affecting the final positioning accuracy and an important indicator for evaluating the algorithm performance. In this example, the distance estimation accuracy of the algorithm is measured by the error mean of the estimated distances of all participating unknown node positioning, and is normalized as a percentage of the communication radius.

[0061] In the distance estimation step S3 experiment, the algorithm of this example is compared with the classic DV-Hop algorithm. In experimental scenario 1, the comparison of the impact of the change in anchor node ratio on the distance estimation accuracy between the algorithm of this example and the classic DV-Hop algorithm is shown in Figure 1. Figure 3 As shown in the figure, the algorithm in this example and the classic DV-Hop algorithm are compared in experimental scenario 2 in terms of the impact of the change in the total number of nodes on the distance estimation accuracy. Figure 4 As shown in the figure, the algorithm in this example is compared with the classic DV-Hop algorithm in experimental scenario 3, in terms of the impact of the change in communication radius on the distance estimation accuracy. Figure 5 The experimental results show that: as the proportion of anchor nodes changes, the algorithm can reduce the distance estimation error by about 4.5% compared with the original method; as the density of network node deployment changes, the distance estimation error can be reduced by about 5.6%; as the size of the communication radius changes, the distance estimation error can be reduced by about 8.2%.

[0062] The present invention proposes to make targeted improvements to the DV-Hop model, with the goal of obtaining a positioning model that maintains high accuracy and stable performance under a variable network topology environment. The final positioning accuracy is a key indicator for evaluating the performance of the algorithm. In this example, the positioning accuracy is measured by the mean of the Euclidean distance between the positions of all unknown nodes obtained through positioning calculation and the true position, and is normalized as a percentage of the communication radius.

[0063] In the positioning step S5 experiment, the algorithm of this example is compared with the classic DV-Hop algorithm. In experimental scenario 1, the comparison of the impact of anchor node ratio change on positioning accuracy between the algorithm of this example and the classic DV-Hop algorithm is as follows: Figure 6 As shown in the figure, the algorithm in this example and the classic DV-Hop algorithm are compared in experimental scenario 2 in terms of the impact of the change in the total number of nodes on positioning accuracy. Figure 7 As shown in the figure, the algorithm in this example is compared with the classic DV-Hop algorithm in experimental scenario 3, in terms of the impact of the change in communication radius on positioning accuracy. Figure 8 The experimental results show that: as the proportion of anchor nodes changes, the algorithm can reduce the positioning error by about 18.8% compared with the original method; as the density of network node deployment changes, the positioning error can be reduced by about 15.9%; as the size of the communication radius changes, the positioning error can be reduced by about 14.8%.

[0064] In general, the improved DV-Hop positioning model based on the nearest neighbor anchor node path connection relationship proposed in the present invention improves the distance estimation accuracy and the final positioning accuracy better than the original classic DV-Hop model while relatively increasing the amount of calculation, indicating the feasibility of this method.

[0065] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship, characterized in that: The steps include: S1, network node initialization; S2, each anchor node calculates the average hop distance of the shortest path to other anchor nodes recorded in S1; S3, each unknown node selects the nearest neighbor anchor node, and performs positioning calculation according to the path information recorded by the nearest neighbor anchor node in S1 and S2. In this step, the unknown node completes the selection of the anchor node set participating in its own positioning calculation and the distance estimation; S4, locate the unknown nodes whose anchor nodes have been selected and distances estimated in S3; S5. Obtain the final coordinates (x, y) of u.

2. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S1, the network nodes are initialized, and a number of network nodes are randomly deployed in the set monitoring area. The number of network nodes includes anchor nodes and unknown nodes. Specifically, sensor nodes are randomly deployed in the specified area, among which a certain proportion of nodes can know their own positions through measurement or carrying positioning devices such as GPS, which are called anchor nodes, and the remaining nodes to be located are called unknown nodes.

3. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S3, each unknown node selects the nearest neighbor anchor node, and performs positioning calculation according to the path information recorded by the nearest neighbor anchor node in S1 and S2. In this step, the unknown node completes the selection and distance estimation of the anchor node set participating in its own positioning calculation. According to the above idea, step S3 is specifically described as: for a certain unknown node, select the anchor node with the smallest number of hops, that is, the nearest neighbor anchor node, sort the average hop distances recorded by the nearest neighbor anchor node to the remaining anchor nodes in order from large to small, select the remaining anchor nodes corresponding to the top 3 average hop distances to participate in the positioning of the unknown node, and add them to the positioning anchor node set of the unknown node.

4. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S3, if there is more than one nearest neighbor anchor node for the unknown node, the average hop distances recorded by all nearest neighbor anchor nodes to the remaining anchor nodes are sorted in descending order, and then the remaining anchor nodes corresponding to the top 3 average hop distances are selected and added to the positioning anchor node set of the unknown node, and then, based on the existing nearest neighbor anchor nodes and the anchor nodes in the positioning anchor node set.

5. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S4, the unknown node for which the positioning anchor node selection and distance estimation have been completed in S3 is estimated using the Min-Max algorithm to estimate an initial position rectangular range of the unknown node.

6. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S5, a multi-objective particle swarm optimization algorithm is used to obtain the final coordinates (x, y) of u.

7. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S5, the positioning problem is modeled as a problem of finding the optimal solution of a nonlinear system of equations. The nonlinear system of equations is listed according to the Euclidean distance relationship between the unknown node and the three anchor nodes in the positioning anchor node combination. The nonlinear system of equations is solved using a multi-objective particle swarm optimization algorithm. Three objective functions are determined based on equations established based on the three Euclidean distances and coordinates in the system of equations.

8. The improved DV-Hop positioning method based on the nearest neighbor anchor node path connection relationship according to claim 1, characterized in that: In S5, the initial values ​​of the x and y components of each particle are set to random values ​​within the rectangular position range obtained in S4, and the initial value of the di component is the respective distance values ​​obtained in the distance estimation step in S3; the Pareto optimal solution set is obtained by iterative calculation, and the global optimal solution of the particle is randomly obtained, which is the positioning value of the unknown node.

Images