Dynamic topology ad hoc network node redistribution method and device and storage medium

Abstract

The invention discloses a dynamic topology ad hoc network node redistribution method and device and a storage medium, and belongs to the technical field of wireless sensor networks, and the method comprises the following steps: periodically obtaining environment parameters of sensor node deployment positions, and constructing node environment vectors based on the environment parameters; according to the node environment vector, a node topological graph is constructed based on the physical distance between the sensor nodes and the environment state similarity; based on a propagation and coding model, node features in the node topological graph are propagated and coded layer by layer by using a graph convolutional network, and a network coverage prediction index is obtained through calculation; constructing a node redistribution objective function; performing global search on the candidate position set of the sensor nodes to obtain a global optimal position of each sensor node; and moving each sensor to a corresponding global optimal position, and synchronously updating the node topological graph to complete dynamic topology ad hoc network node redistribution. According to the invention, the problem of poor adaptive capacity of a wireless sensor network environment is solved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wireless sensor networks, and particularly relates to a dynamic topology ad hoc network node rebalancing method, device and storage medium. BACKGROUND

[0002] With the rapid development of Internet of Things technology, wireless sensor networks have been widely applied in environmental monitoring, industrial control, disaster rescue and intelligent transportation fields. Wireless sensor networks are usually deployed in a self-organizing manner by a large number of nodes, and the perception and monitoring of a specific area are achieved through mutual communication. However, once the traditional sensor network topology structure is deployed, it is difficult to adjust in real time, and it cannot effectively cope with the reduction of communication quality and the increase of monitoring blind area caused by the dynamic changes of environmental parameters in the deployment area, such as temperature changes, obstacle increases or position changes.

[0003] In the prior art, the positions of sensor nodes are usually laid out based on static or semi-static topology planning methods, which are difficult to perceive environmental changes in real time and adjust node positions in time, resulting in insufficient network coverage, unbalanced energy consumption and low information transmission efficiency. Although some studies have proposed using traditional optimization algorithms or simple feedback mechanisms to adjust node positions, these methods often have the disadvantages of insufficient environmental perception ability, low optimization efficiency, high computational complexity, etc., and it is difficult to realize environmental adaptive dynamic topology optimization. SUMMARY

[0004] In view of the above deficiencies in the prior art, the present application provides a dynamic topology ad hoc network node rebalancing method, device and storage medium, which combines network communication state, perceives environmental parameters of the deployment area in real time, and dynamically and intelligently adjusts node positions to improve the overall performance and robustness of the network, solving the problem of poor environmental adaptability of wireless sensor networks.

[0005] In order to achieve the above application purposes, the technical solution adopted by the present application is as follows: In a first aspect, the present application provides a dynamic topology ad hoc network node rebalancing method, comprising the following steps: S1, periodically acquiring environmental parameters of the deployment position of the sensor node, and constructing a node environment vector based on the environmental parameters; S2, constructing a node topology graph based on the physical distance between the sensor nodes and the environmental state similarity according to the node environment vector; S3, based on the propagation and coding model, propagating and coding the node features in the node topology graph layer by layer using a graph convolution network, and calculating a network coverage prediction index; S4, constructing a node rebalancing objective function based on the network coverage prediction index, communication energy consumption and node displacement cost; S5. Based on the redistribution objective function, the particle swarm optimization algorithm is used to perform a global search on the candidate position set of sensor nodes to obtain the global optimal position of each sensor node. S6. Use motion actuators to move each sensor to its corresponding global optimal position and update the node topology graph simultaneously to complete the dynamic topology self-organizing network node redistribution.

[0006] The beneficial effects of this invention are as follows: This invention provides a dynamic topology self-organizing network node redistribution method. By periodically acquiring environmental parameters to construct node environment vectors, it can accurately perceive changes in the network's environment. This invention constructs a node topology graph based on physical distance and environmental state similarity, making the network structure more closely resemble the actual scenario. This invention utilizes graph convolutional networks to propagate and encode node features and calculate network coverage prediction indicators, enabling early prediction of network coverage. The redistribution objective function constructed based on this comprehensively considers network coverage, communication energy consumption, and node displacement costs, achieving multi-objective optimization. This invention employs a particle swarm optimization algorithm to globally search for the optimal location of sensor nodes, ensuring the global optimality of the solution. Finally, a motion execution device is used to complete node redistribution and update the topology graph, effectively improving network coverage quality, reducing communication energy consumption, reducing node movement costs, and enhancing the adaptability and stability of the dynamic topology self-organizing network. This invention addresses the rigid topology structure and poor environmental adaptability of wireless sensor networks in complex environments. By combining environmental modeling and graph neural networks, it achieves efficient node redistribution and topology optimization, significantly improving network coverage, robustness, and lifespan.

[0007] Furthermore, the environmental parameters include: temperature, humidity, obstacle density, and current signal-to-noise ratio; The expression for calculating the node environment vector is as follows: , in, Indicates the current time Next The node environment vector of each sensor node. Indicates the current time Next Each sensor node in The temperature at that location Indicates the current time Next Each sensor node in Humidity at the location Indicates the current time Next Each sensor node in Obstacle density at the location, Indicates the current time Next Each sensor node in The current signal-to-noise ratio at the location, where, Indicates the first The horizontal coordinate position of each sensor node Indicates the first i The vertical coordinate position of each sensor node.

[0008] The beneficial effects of adopting the above-mentioned further solutions are as follows: The node environment vector provided by the present invention comprehensively considers the temperature, humidity, obstacle density of the environment in which the sensor node is located, as well as the current signal-to-noise ratio of the sensor, which can more realistically reflect the actual environment in which the sensor node is located, thereby constructing a node topology map that is more in line with the actual scenario, making the node redistribution strategy more scientific and reasonable, and effectively improving the communication quality and stability of the network.

[0009] Further, step S2 includes the following steps: S21. Based on the node environment vector, calculate the physical distance and environmental state similarity between sensor nodes respectively; S22. Use the environmental state similarity as the weight of the edges in the node topology graph, and combine it with the physical distance between sensor nodes to construct the node topology graph; The calculation expression for the node topology graph is as follows: , in, Represents a node topology graph. Represents a set of nodes. Denotes the set of edges. This represents the weight matrix of the edges.

[0010] The beneficial effects of adopting the above-mentioned further scheme are as follows: The present invention calculates the physical distance and environmental state similarity between sensor nodes based on the node environment vector, providing a comprehensive and detailed basis for constructing an accurate node topology map. The physical distance can intuitively reflect the actual spatial interval of nodes and is a fundamental element for determining the basic network connection architecture. At the same time, the environmental state similarity measures the degree of similarity of the environment in which the sensor nodes are located from the comprehensive perspective of environmental factors such as temperature, humidity, and obstacle density. This helps to deeply understand the potential correlation between signal propagation and communication quality between nodes. By using the environmental state similarity as the weight of the edge in the node topology map and combining it with the physical distance to construct the topology map, the constructed node topology map can not only reflect the spatial distribution relationship of nodes, but also accurately reflect the differences in communication characteristics between nodes caused by environmental differences.

[0011] Furthermore, the calculation expression for the physical distance between sensor nodes in S21 is as follows: in, Indicates the physical distance between sensor nodes. Indicates the first The horizontal coordinate position of each sensor node Indicates the first The vertical coordinate position of each sensor node.

[0012] The beneficial effects of adopting the above-mentioned further solutions are as follows: the present invention provides a method for calculating the physical distance between sensor nodes, which provides a basis for constructing a node topology map based on the accurate physical distance between sensor nodes.

[0013] Furthermore, the expression for calculating the environmental state similarity in S21 is as follows: , in, Indicates the first The sensor node and the first Environmental state similarity between sensor nodes Indicates An exponential function with a base of 1. Indicates the first The node environment vector of each sensor node. Indicates the first The node environment vector of each sensor node. This represents the adjustment coefficient.

[0014] The beneficial effects of adopting the above-mentioned further scheme are as follows: The present invention provides a method for calculating the environmental state similarity between sensor nodes based on the node environment vector of sensor nodes. It fully considers environmental factors such as temperature, humidity, and obstacle density, which greatly improves the ability of the node topology graph to represent the actual network environment. It provides a more accurate and reliable model basis for subsequent operations such as network coverage prediction and node redistribution optimization based on the node topology graph, and helps to improve the performance and stability of the entire dynamic topology ad hoc network.

[0015] Furthermore, the calculation expression for the propagation and coding model in S3 is as follows: , , , in, The graph convolutional network represents the first... Sensor nodes in the layer eigenvectors, Represents a non-linear activation function. Represents sensor nodes The set of neighboring nodes, Represents sensor nodes The degree, Represents sensor nodes The degree, Indicates the first Adjacent sensor nodes in the layer eigenvectors, The graph convolutional network represents the first... The learnable weight matrix of the layer, Represents sensor nodes The total number of neighboring nodes, Represents sensor nodes The total number of neighboring nodes, Represents sensor nodes The set of neighboring nodes, where, , denoted as the total number of layers in the graph convolutional network.

[0016] The beneficial effects of adopting the above-mentioned further scheme are as follows: This invention provides a calculation method for the propagation and encoding model. By setting a symmetric normalization term, it can effectively prevent gradient explosion or vanishing. By aggregating the features of neighboring nodes, the local neighborhood information of the sensor node is integrated into its feature representation. The final output node feature is a low-dimensional dense vector, which is the node embedding vector. The node embedding vector simultaneously encodes node attributes and graph structure, and neighboring nodes... Feature information is aggregated to the central node through weighted summation. The node feature propagation is achieved. The aggregated feature vector is multiplied by a learnable weight matrix and then a new node feature vector, i.e., the encoding result, is obtained through an activation function. Based on the location coordinates of the sensor nodes in the encoding result, a basis can be provided for predicting network coverage in advance.

[0017] Furthermore, the calculation expression for the network coverage prediction index in S3 is as follows: , in, Indicates network coverage. Represents the union, Indicated by sensor node With center radius The coverage area Indicates the area of ​​the target region. Indicates the first The horizontal coordinates propagated and encoded from each sensor node Indicates the first The vertical coordinates are propagated and encoded from each sensor node.

[0018] The beneficial effects of adopting the above-mentioned further scheme are as follows: This invention provides a method for calculating network coverage prediction indicators, which accurately depicts the area covered by all sensor nodes under a given communication radius, and divides it by the area of ​​the entire target area. This reflects the degree of network coverage of the target area in a direct and quantitative way in the form of a ratio. It fully considers the actual impact of node location and communication range on network coverage, and provides an objective and accurate standard for evaluating the coverage performance of dynamic topology self-organizing networks. This helps to make accurate decisions and verify the effects in optimization processes such as node redistribution.

[0019] Furthermore, the calculation expression for the node redistribution objective function in S4 is as follows: , , , , in, This indicates taking the maximum value. This indicates the target value for node location redistribution. This represents the network coverage adjustment coefficient. This represents the communication energy consumption adjustment coefficient. This represents the node migration cost adjustment coefficient. Indicates communication energy consumption. Indicates the cost of node migration. Indicates the first The sensor node sends to the first Energy consumption per unit distance when each sensor node transmits data Represents the second-order norm operation. Indicates the path loss factor. For sensor node pairs The set of edges formed, where, .

[0020] The beneficial effects of adopting the above-mentioned further scheme are as follows: This invention provides a method for calculating the objective function of node redistribution. By introducing network coverage adjustment coefficient, communication energy consumption adjustment coefficient, and node migration cost adjustment coefficient, it can flexibly balance the relationship between the three key factors of network coverage, communication energy consumption, and node migration cost. In terms of network coverage, combined with the calculation of network coverage, it ensures that the redistributed network can effectively cover the target area and meet communication requirements. In terms of communication energy consumption, it considers the energy consumption per unit distance and the path loss factor, which can truly reflect the energy consumption of data transmission between sensor nodes, helping to reduce the overall energy consumption of the network and extend the service life of sensor nodes. The node migration cost considers the actual cost of node movement, avoiding frequent node movement due to excessive pursuit of network coverage and low energy consumption, thus increasing unnecessary costs.

[0021] In a second aspect, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it executes any one of the dynamic topology self-organizing network node redistribution methods in the first aspect.

[0022] Thirdly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the dynamic topology self-organizing network node redistribution methods in the first aspect.

[0023] Other advantages of the present invention will be analyzed in more detail in the following embodiments. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating the steps of a dynamic topology self-organizing network node redistribution method based on an environment-aware graph neural network according to an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0027] like Figure 1 As shown, in one embodiment of the present invention, the present invention provides a method for redistributing nodes in a dynamic topology ad hoc network, comprising the following steps: S1. Periodically acquire environmental parameters of the sensor node deployment location, and construct node environment vectors based on the environmental parameters; The environmental parameters include: temperature, humidity, obstacle density, and current signal-to-noise ratio; The expression for calculating the node environment vector is as follows: , in, Indicates the current time Next The node environment vector of each sensor node. Indicates the current time Next Each sensor node in The temperature at that location Indicates the current time Next Each sensor node in Humidity at the location Indicates the current time Next Each sensor node in Obstacle density at the location, Indicates the current time Next Each sensor node in The current signal-to-noise ratio at the location, where, Indicates the first The horizontal coordinate position of each sensor node Indicates the first i The vertical coordinate position of each sensor node.

[0028] In this embodiment, the acquired environmental parameters were first standardized, and then the corresponding node environment vectors were constructed.

[0029] S2. Based on the node environment vector, construct a node topology graph based on the physical distance between sensor nodes and the similarity of environmental states; S2 includes the following steps: S21. Based on the node environment vector, calculate the physical distance and environmental state similarity between sensor nodes respectively; The formula for calculating the physical distance between sensor nodes in S21 is as follows: in, Indicates the physical distance between sensor nodes. Indicates the first The horizontal coordinate position of each sensor node Indicates the first The vertical coordinate position of each sensor node.

[0030] The formula for calculating the environmental state similarity in S21 is as follows: , in, Indicates the first The sensor node and the first Environmental state similarity between sensor nodes Indicates An exponential function with a base of 1. Indicates the first The node environment vector of each sensor node. Indicates the first The node environment vector of each sensor node. This represents the adjustment coefficient.

[0031] S22. Use the environmental state similarity as the weight of the edges in the node topology graph, and combine it with the physical distance between sensor nodes to construct the node topology graph; The calculation expression for the node topology graph is as follows: , in, Represents a node topology graph. Represents a set of nodes. Denotes the set of edges. This represents the weight matrix of the edges.

[0032] S3. Based on the propagation and coding model, the node features in the node topology graph are propagated and coded layer by layer using graph convolutional networks, and the network coverage prediction index is calculated. The calculation expression for the propagation and coding model in S3 is as follows: , , , in, The graph convolutional network represents the first... Sensor nodes in the layer eigenvectors, Represents a non-linear activation function. Represents sensor nodes The set of neighboring nodes, Represents sensor nodes The degree, Represents sensor nodes The degree, Indicates the first Adjacent sensor nodes in the layer eigenvectors, The graph convolutional network represents the first... The learnable weight matrix of the layer, Represents sensor nodes The total number of neighboring nodes, Represents sensor nodes The total number of neighboring nodes, Represents sensor nodes The set of neighboring nodes, where, , denoted as the total number of layers in the graph convolutional network.

[0033] In this plan, As a symmetric normalization term, it effectively prevents gradient explosion or vanishing, especially in graphs with large differences in node degree. Graph convolutional networks integrate the local neighborhood information of sensor nodes into their feature representations by aggregating the features of neighboring nodes. The final output node features are low-dimensional dense vectors, namely node embedding vectors. These vectors encode both node attributes and graph structure. Node attributes refer to the node properties formed by mapping sensor node features to a higher-order semantic space through nonlinear transformations of multi-layer graph convolutional networks. Graph structure refers to the position of the sensor node in the node topology graph, implicitly encoded through neighbor aggregation. Adjacent nodes. Feature information is aggregated to the central node through weighted summation. To achieve node feature propagation, the aggregated feature vector is multiplied by a learnable weight matrix and then passed through an activation function to obtain a new node feature vector, i.e., the encoding result.

[0034] The calculation expression for the network coverage prediction index in S3 is as follows: , in, Indicates network coverage. Represents the union, Indicated by sensor node With center radius The coverage area Indicates the area of ​​the target region. Indicates the first The horizontal coordinates propagated and encoded from each sensor node Indicates the first The vertical coordinates are propagated and encoded from each sensor node.

[0035] In this scheme, the target area is the region that can minimize communication energy consumption and node migration costs.

[0036] S4. Construct a node redistribution objective function based on network coverage prediction indicators, communication energy consumption, and node displacement costs; The calculation expression for the node redistribution objective function in S4 is as follows: , , , , in, This indicates taking the maximum value. This indicates the target value for node location redistribution. This represents the network coverage adjustment coefficient. This represents the communication energy consumption adjustment coefficient. This represents the node migration cost adjustment coefficient. Indicates communication energy consumption. Indicates the cost of node migration. Indicates the first The sensor node sends to the first Energy consumption per unit distance when each sensor node transmits data Represents the second-order norm operation. Indicates the path loss factor. For sensor node pairs The set of edges formed, where, In this embodiment, the sensor node pair The first one with a communication connection The sensor node sends to the first It consists of several sensor nodes.

[0037] S5. Based on the redistribution objective function, the particle swarm optimization algorithm is used to perform a global search on the candidate position set of sensor nodes to obtain the global optimal position of each sensor node. S6. Use motion actuators to move each sensor to its corresponding global optimal position and update the node topology graph simultaneously to complete the dynamic topology self-organizing network node redistribution.

[0038] It should be understood that, although attached Figure 1 The steps in the flowchart are shown sequentially according to the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order requirement for the execution of these steps, and they can be executed in other orders. Furthermore, [the following is a list of steps]. Figure 1 At least some of the steps in the process may include multiple sub-steps or sub-stages. These sub-steps or stages are not necessarily executed at the same time, but may be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0039] In one embodiment, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the above-described dynamic topology self-organizing network node redistribution method.

[0040] The computer device includes a memory, a processor, and one or more computer programs, wherein the one or more computer programs may be stored in the memory and configured to be executed by one or more processors, and the one or more application programs are configured to perform the above-described dynamic topology self-organizing network node redistribution method.

[0041] A processor may include one or more processing cores. The processor connects to various parts of the computer device using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory, and by calling data stored in memory. Optionally, the processor may be implemented using at least one hardware form of digital signal processing, field-programmable gate arrays, or programmable logic arrays. The processor may integrate one or a combination of several of the following: a central processing unit (CPU) data reporting and verification unit, and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also be implemented separately as a communication chip, without being integrated into the processor.

[0042] The memory may include random access memory (RAM) or read-only memory (ROM). The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the various method embodiments described above, etc. The data storage area may also store data created by the terminal device during use.

[0043] In one embodiment, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described dynamic topology self-organizing network node redistribution method.

[0044] The computer-readable storage medium may be an electronic storage device such as flash memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, hard disk, or ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable medium. The computer-readable storage medium has storage space for program code that performs any of the method steps described above. This program code can be read from or written to one or more computer program products, and the program code may be compressed in an appropriate form.

[0045] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for redistributing nodes in a dynamic topology ad hoc network, characterized in that, Includes the following steps: S1. Periodically acquire environmental parameters of the sensor node deployment location, and construct node environment vectors based on the environmental parameters; S2. Based on the node environment vector, construct a node topology graph based on the physical distance between sensor nodes and the similarity of environmental states; S3. Based on the propagation and coding model, the node features in the node topology graph are propagated and coded layer by layer using graph convolutional networks, and the network coverage prediction index is calculated. S4. Construct a node redistribution objective function based on network coverage prediction indicators, communication energy consumption, and node displacement costs; S5. Based on the redistribution objective function, the particle swarm optimization algorithm is used to perform a global search on the candidate position set of sensor nodes to obtain the global optimal position of each sensor node. S6. Use motion actuators to move each sensor to its corresponding global optimal position and update the node topology graph simultaneously to complete the dynamic topology self-organizing network node redistribution.

2. The dynamic topology self-organizing network node redistribution method according to claim 1, characterized in that, The environmental parameters include: temperature, humidity, obstacle density, and current signal-to-noise ratio; The expression for calculating the node environment vector is as follows: ， in, Indicates the current time Next The node environment vector of each sensor node. Indicates the current time Next Each sensor node in The temperature at that location Indicates the current time Next Each sensor node in Humidity at the location Indicates the current time Next Each sensor node in Obstacle density at the location, Indicates the current time Next Each sensor node in The current signal-to-noise ratio at the location, where, Indicates the first The horizontal coordinate position of each sensor node Indicates the first i The vertical coordinate position of each sensor node.

3. The dynamic topology self-organizing network node redistribution method according to claim 2, characterized in that, S2 includes the following steps: S21. Based on the node environment vector, calculate the physical distance and environmental state similarity between sensor nodes respectively; S22. Use the environmental state similarity as the weight of the edges in the node topology graph, and combine it with the physical distance between sensor nodes to construct the node topology graph; The calculation expression for the node topology graph is as follows: ， in, Represents a node topology graph. Represents a set of nodes. Denotes the set of edges. This represents the weight matrix of the edges.

4. The dynamic topology self-organizing network node redistribution method according to claim 3, characterized in that, The formula for calculating the physical distance between sensor nodes in S21 is as follows: in, Indicates the physical distance between sensor nodes. Indicates the first The horizontal coordinate position of each sensor node Indicates the first The vertical coordinate position of each sensor node.

5. The dynamic topology self-organizing network node redistribution method according to claim 3, characterized in that, The formula for calculating the environmental state similarity in S21 is as follows: ， in, Indicates the first The sensor node and the first Environmental state similarity between sensor nodes Indicated by An exponential function with a base of 1. Indicates the first The node environment vector of each sensor node. Indicates the first The node environment vector of each sensor node. This represents the adjustment coefficient.

6. The dynamic topology self-organizing network node redistribution method according to claim 5, characterized in that, The calculation expression for the propagation and coding model in S3 is as follows: ， ， ， in, This represents the graph convolutional network. Sensor nodes in the layer eigenvectors, Represents a non-linear activation function. Represents sensor nodes The set of neighboring nodes, Represents sensor nodes The degree, Represents sensor nodes The degree, Indicates the first Adjacent sensor nodes in the layer eigenvectors, This represents the graph convolutional network. The learnable weight matrix of the layer, Represents sensor nodes The total number of neighboring nodes, Represents sensor nodes The total number of neighboring nodes, Represents sensor nodes The set of neighboring nodes, where, , denoted as the total number of layers in the graph convolutional network.

7. The dynamic topology self-organizing network node redistribution method according to claim 6, characterized in that, The calculation expression for the network coverage prediction index in S3 is as follows: ， in, Indicates network coverage. Represents the union, Indicated by sensor node With center radius The coverage area Indicates the area of ​​the target region. Indicates the first The horizontal coordinates propagated and encoded from each sensor node Indicates the first The vertical coordinates are propagated and encoded from each sensor node.

8. The dynamic topology self-organizing network node redistribution method according to claim 7, characterized in that, The calculation expression for the node redistribution objective function in S4 is as follows: ， ， ， ， in, This indicates taking the maximum value. This indicates the target value for node location redistribution. This represents the network coverage adjustment coefficient. This represents the communication energy consumption adjustment coefficient. This represents the node migration cost adjustment coefficient. Indicates communication energy consumption. Indicates the cost of node migration. Indicates the first The sensor node sends to the first Energy consumption per unit distance when each sensor node transmits data Represents the second-order norm operation. Indicates the path loss factor. For sensor node pairs The set of edges formed, where, .

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it performs the dynamic topology self-organizing network node redistribution method according to any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the dynamic topology self-organizing network node redistribution method according to any one of claims 1-8.

Images