Equipment ad hoc network method and system based on contour coefficient

Abstract

The invention relates to the technical field of computers, and discloses a device ad hoc network method and system based on contour coefficients, and the method comprises the steps: collecting the device information of all devices, and dividing all devices into different subnets according to the device information; calculating contour coefficients of all devices, and adjusting each subnet according to each contour coefficient; and controlling each subnet to carry out networking according to the equipment information of the equipment contained in the subnet. According to the invention, after subnet division is carried out on the networking system, the contour coefficients of all the devices are calculated, whether subnet division is reasonable is judged through the contour coefficients, unreasonable parts are adjusted according to the contour coefficients, and networking is carried out on each subnet after adjustment is completed, so that the rationality of subnet division is ensured, and the networking efficiency is improved. The network structure is optimized better, the networking efficiency is ensured, the influence of subnet division on the networking effect is reduced as much as possible, and the final networking effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a device self-organizing network method, system, terminal and readable storage medium based on outline coefficient. Background Art

[0002] With the popularization of IoT devices, the collaborative work of different devices in various application scenarios has become increasingly important. Due to the large number of devices and their different functions, how to divide the various devices and then network them has become a technical challenge.

[0003] Traditional networking methods mostly rely on static configuration and cannot adapt to rapidly changing application requirements and device status. For this reason, the existing technology has proposed a variety of self-organizing networking methods, such as methods of self-organizing devices according to network conditions or device performance. This not only saves manpower configuration at the beginning of networking, but also can be adjusted according to the self-organizing networking method when application requirements and device status change.

[0004] When these prior art self-organizing networking methods are applied to large-scale complex networks, as the number and scale of devices increase, the solvable space of the network also grows explosively, resulting in slow or even impossible calculation of the optimal solution for the network in the solvable space. To this end, a feasible method is to partition the devices according to their geographical locations, thereby reducing the solvable space of each sub-partition and increasing the speed of the network to meet the real-time nature of the network.

[0005] However, when dividing subnets according to regions, the algorithm sometimes makes incorrect divisions, resulting in unreasonable subnet divisions, which in turn affects the effect of subsequent networking. Summary of the invention

[0006] The purpose of the present invention is to provide a device self-networking method and system based on silhouette coefficient, which aims to solve the problem of unreasonable division when the entire networking system is divided into multiple subnets for networking separately in the prior art.

[0007] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0008] The present invention provides a device self-organizing network method based on a contour coefficient, and the device self-organizing network method based on a contour coefficient comprises the steps of:

[0009] Collect device information of all devices, and divide all the devices into different subnets according to the device information;

[0010] Calculating the silhouette coefficients of all devices, and adjusting each of the subnets according to each of the silhouette coefficients;

[0011] Control each subnet to form a network according to the device information of the devices included in the subnet.

[0012] Furthermore, dividing all the devices into different subnets according to the device information specifically includes:

[0013] Divide into multiple geographical areas and select a cluster center for each geographical area;

[0014] All the devices are clustered according to all the cluster centers and the device information, and all the devices are divided into different subnets.

[0015] Further, the calculating of the silhouette coefficients of all devices and adjusting each of the subnets according to each of the silhouette coefficients specifically includes:

[0016] Calculate the silhouette coefficient S(i) of the i-th device:

[0017]

[0018] Where a(i) represents the average distance from the ith device to other devices in the same subnet, and b(i) represents the average distance from the ith device to devices outside the same subnet.

[0019] The subnet division effect is evaluated according to the silhouette coefficients of all devices, and each subnet is adjusted according to the evaluation result.

[0020] Furthermore, the subnet division effect is evaluated according to the silhouette coefficients of all devices, and each subnet is adjusted according to the evaluation result, specifically including:

[0021] Determine whether subnet division needs to be performed again based on the contour coefficients of all devices. If necessary, perform subnet division again.

[0022] Control each subnet and adjust the corresponding subnet according to the profile coefficient of the devices contained in the subnet.

[0023] Furthermore, it is determined whether it is necessary to re-subnet the network based on the profile coefficients of all devices. If necessary, the subnet is re-divided, which specifically includes:

[0024] Calculate the average of the silhouette coefficients of all devices to obtain the overall average silhouette coefficient;

[0025] It is determined whether the overall average silhouette coefficient is less than a set overall threshold. If so, subnet division and re-clustering are performed.

[0026] Further, the controlling each subnet to adjust the subnet according to the profile coefficient of the device included in the subnet specifically includes:

[0027] Control each subnet to calculate the average value of the silhouette coefficients of the devices contained in the subnet, and obtain the average silhouette coefficient of each subnet;

[0028] It is determined whether the average silhouette coefficient of each subnet is less than a set subnet threshold. If so, the cluster center of the corresponding subnet is adjusted and clustering is performed again.

[0029] Further, the controlling each subnet adjusts the subnet according to the profile coefficient of the device included in the subnet, and then further includes:

[0030] Determine whether the profile coefficient of the device is lower than the set device threshold. If so, adjust the subnet to which the device belongs.

[0031] In addition, to achieve the above-mentioned purpose, the present invention further provides a device self-organizing network system based on a contour coefficient, and the device self-organizing network system based on a contour coefficient comprises:

[0032] A subnet division module is used to collect device information of all devices and divide all the devices into different subnets according to the device information;

[0033] A subnet adjustment module, used to calculate the profile coefficients of all devices and adjust each subnet according to each profile coefficient;

[0034] The subnet networking module is used to control each subnet to be networked according to the device information of the devices included in the subnet.

[0035] In addition, to achieve the above-mentioned purpose, the present invention also provides a terminal, which includes: a memory, a processor, and a device self-organizing network program based on a contour coefficient stored in the memory and executable on the processor, wherein the device self-organizing network program based on a contour coefficient, when executed by the processor, controls the terminal to implement the steps of the device self-organizing network method based on a contour coefficient as described above.

[0036] In addition, to achieve the above-mentioned purpose, the present invention also provides a readable storage medium, which stores a device self-organizing networking program based on a contour coefficient, and when the device self-organizing networking program based on a contour coefficient is executed by a processor, the steps of the device self-organizing networking method based on a contour coefficient as described above are implemented.

[0037] The present invention adopts the above technical solution to achieve the following effects:

[0038] The present invention calculates the contour coefficients of all devices after dividing the networking system into subnets, determines whether the subnet division is reasonable by the contour coefficients, and adjusts the unreasonable parts according to the contour coefficients. After the adjustment is completed, each subnet is networked separately, thereby ensuring the rationality of the subnet division, so as to better optimize the network structure, and while ensuring the networking efficiency, reduce the influence of the subnet division on the networking effect as much as possible, thereby improving the final networking effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a flow chart of the steps of a device self-organizing network method based on a contour coefficient in a preferred embodiment of the present invention;

[0040] Figure 2 It is a structural schematic diagram of a device self-organizing network system based on a contour coefficient in a preferred embodiment of the present invention;

[0041] Figure 3 It is a schematic diagram of the operating environment of a preferred embodiment of the terminal of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention. All other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0043] Embodiment 1

[0044] See also Figure 1 , Embodiment 1 of the present application is a device self-organizing networking method based on contour coefficient, which can be applied to management terminals of various types of Internet of Things systems, such as management terminals of edge computing systems, management terminals of smart home systems, or management terminals of other types of Internet of Things systems. In addition, a device correlation self-organizing networking method of this embodiment can also be run outside the Internet of Things system to provide networking support services for the Internet of Things system. It can be run in a server, which can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud storage, big data and artificial intelligence platforms.

[0045] In this embodiment, the device self-organizing network method based on the silhouette coefficient includes the following steps:

[0046] S1. Collect device information of all devices, and divide all the devices into different subnets according to the device information.

[0047] Specifically, in this embodiment, each uploaded device data is collected, and the device data includes the device's location information, hardware information, function information, environment information, and operating status information, wherein the location information refers to information describing the location of the device, including coordinate information of the device's coordinates, and height information describing the height of the device considering that the device may be on different floors.

[0048] In this embodiment, the coordinate information is obtained through the GPS (Global Positioning System) integrated in the device. The hardware information refers to the information describing the unchanged hardware performance of the device, including central processing unit (CPU) information, memory information, storage information and sensor type information. The functional information refers to the information describing the software functions supported by the device, including protocol information describing the communication protocols supported by the device and application information describing the special software functions of the device, such as graphics processing function. The environmental information is the information of the environment in which the device is located, which is used to describe the operating environment of the device, including temperature information, humidity information and light information. The operating status information is the operating status of the device under the current task, including central processing unit load, memory utilization, storage utilization and network delay.

[0049] Afterwards, a networking model is established based on the device information. In this embodiment, the networking model is a super-object model used to describe each device involved in the networking. The super-object model is an abstract model used to describe and manage device capabilities, covering multiple aspects such as the device's hardware resources, software functions, current status, and environmental information. This can not only help manage a single device, but also serve as a unified representation of a collection of virtual devices.

[0050] Therefore, a unified model is used to describe the spatial attributes, hardware characteristics, software functions, real-time operating status and other information of each device, and the devices are classified and managed according to different functional levels, such as perception layer (sensor equipment), processing layer (edge ​​computing nodes), communication layer (network equipment), etc. A corresponding super object model is established for each layer, and finally a hierarchical super object model system is formed.

[0051] After completing the construction of the networking model, subnet division can be performed based on the device location information, user hardware information and user function information to divide the entire system into multiple different areas. In this embodiment, subnet division is performed using the K-Means clustering method.

[0052] First, the devices are preliminarily divided according to their location information. Specifically, the entire system is divided into multiple different areas. Specifically, relatively independent geographical areas are divided into a single area. The number of divided areas can be determined based on the distribution density of the network to ensure that each area has a network of a certain scale, while not having a network that is too large. The network in each area can meet the needs of work while not causing the subnet networking time of the area to be too long. The specific division method can be artificial division or an algorithm can be used to divide according to coordinate information, such as using a graph algorithm for division.

[0053] Afterwards, a location is selected in each area as the cluster center to serve as the basis for subsequent clustering. Specifically, the center point of each area can be calculated as the initial cluster center, or the cluster center can be calculated according to other set rules. For example, a device that interacts most frequently with other devices in the area can be selected as the cluster center, that is, the cluster center.

[0054] Afterwards, the distance between each device and each cluster center is calculated to cluster each device and divide each device into different clusters. The devices in each cluster form a subnet.

[0055] In this embodiment, the distance adopts a multi-dimensional distance measurement method, including spatial distance and functional distance, wherein the spatial distance adopts the Euclidean distance, and the Euclidean distance between each device and each cluster center is calculated according to the coordinate information of each device. The functional distance is calculated using the Manhattan distance, which firstly needs to vectorize the functional information and hardware information of each device, wherein for the hardware information, the central processing unit information, memory information, storage information and sensor type information are converted using a set ratio value, and each becomes an element of the distance vector, and for the functional information, if the function is present, the corresponding element is set to 1, and if the function is not present, the corresponding element is set to 0, and then the converted distance vector is used to calculate the Manhattan distance between each device and each cluster center, and finally clustering is performed according to the Manhattan distance and the Euclidean distance.

[0056] It is worth noting that in this embodiment, fine-tuning is performed based on the functional distance of the device in order to ensure that devices with complementary functions can be in the same cluster. Such a choice can improve the efficiency of networking. Similarly, in an optional embodiment, the distance can also include the load distance, which is also calculated using the Manhattan distance and weighted calculation, so as to further optimize the clustering effect. It is worth noting that since the distance is a measure of the similarity between two devices, and in this embodiment, it is necessary to achieve complementarity between different devices, the weighted value of the Manhattan distance in this embodiment is a negative number.

[0057] S2. Calculate the silhouette coefficients of all devices, and adjust each of the subnets according to each of the silhouette coefficients.

[0058] Specifically, the silhouette coefficient is first calculated for each device, where the silhouette coefficient S(i) of the i-th device is:

[0059]

[0060] Among them, a(i) represents the average distance from the i-th device to other devices in the same subnet, which is used to measure the closeness between the i-th device and the devices in the subnet; b(i) represents the average distance from the i-th device to devices outside the same subnet, which is used to measure the separation between the i-th device and other subnets.

[0061] Afterwards, the rationality of the subnet division is evaluated based on the contour coefficients of all devices, and the unreasonable parts are adjusted. Specifically, the evaluation of the rationality of the subnet division includes device evaluation, subnet internal evaluation and overall networking evaluation.

[0062] Among them, the overall networking evaluation is to calculate the average value of the silhouette coefficients of all devices to obtain the overall average silhouette coefficient to evaluate the effect of subnet division in the entire network. If the overall average silhouette coefficient is greater than the set overall threshold, it indicates that the subnet division structure is reasonable and only minor adjustments are required. If the overall average silhouette coefficient is less than the set overall threshold, the clustering of the devices may need to be readjusted.

[0063] Specifically, when adjusting, the overall networking evaluation is first determined. If the overall networking evaluation fails, the step of dividing all the devices into different subnets is performed again, that is, the division of the areas is adjusted, including the division of the number of areas and the division of the locations of the areas.

[0064] After the overall networking evaluation is passed, each subnet is evaluated internally, and the subnets that fail are adjusted. In this embodiment, this is done by adjusting the location of the cluster center of the subnet.

[0065] The internal evaluation of the subnet is to calculate the average value of the silhouette coefficients of all devices in each subnet to obtain the average silhouette coefficient of the subnet. If the average silhouette coefficient of a subnet is low, for example, lower than the set subnet threshold, or lower than a certain proportion of the overall average silhouette coefficient, it means that the devices in the subnet are not dense enough or the separation from other subnets is not enough, there may be unreasonable device attribution, and it is necessary to adjust the subnet boundary or the location of the cluster center corresponding to the subnet.

[0066] Device evaluation is to determine whether the device's profile coefficient is lower than the set device threshold. If so, the device may need to be adjusted to adjust the subnet to which the device belongs.

[0067] Afterwards, for each device, device evaluation is performed separately, and for the device whose silhouette coefficient is lower than the set device threshold, the subnet to which it belongs is adjusted.

[0068] S3. Control each subnet to form a network according to the device information of the devices included in the subnet.

[0069] Specifically, in this embodiment, the core device of each subnet is selected according to the networking model. In this embodiment, the core device is the device that interacts most with other devices. The core device serves as the central node of the subnet and is responsible for communication coordination and data processing within the subnet.

[0070] After the core device is selected, the function of each device in the subnet is allocated according to the networking model. In this embodiment, the Internet of Things system is specifically used in the field of augmented reality, so the function allocation is specifically allocated to edge nodes, data processing nodes and sensor nodes. Among them, the edge nodes are mainly responsible for data preprocessing and transmission, the data processing nodes are responsible for complex calculations, and the sensor nodes are responsible for environmental data collection.

[0071] After dividing the roles of all nodes, determine the specific network structure based on the networking model. Specifically, define the data flow and fault handling mechanism of each device based on the communication protocols supported by each device, the role of each device, and the performance it has.

[0072] For example, based on the performance of the downstream nodes that each node task needs to occupy, the performance constraints of the network are defined, that is, the constraints are used to ensure that each node can meet the tasks of the upstream nodes, and then the specific networking structure is defined according to the indicators. For example, the total delay of the system is calculated based on the delay between each node and other nodes, and a solution model is established with the total delay as the target, and then the solution model is solved to obtain the specific structure of the subnet.

[0073] After the networking is completed, in addition, in this embodiment, the edge computing node regularly updates the device information to the super object model to ensure that the model reflects the latest status and capabilities of the device, and still uses the device capability description and status information provided by the super object model to dynamically allocate computing resources and network bandwidth, and realize unified management of the virtual resource pool through soft bus technology, and allocate resources to each device in the network to achieve load balancing. During operation, the load status of each device is monitored in real time, and resource allocation is dynamically adjusted to avoid overloading of certain devices or waste of resources.

[0074] In addition, as the dynamic state, resource utilization, and spatial location information of the equipment change in real time during operation, the equipment's silhouette coefficient, the average silhouette coefficient of each subnet, and the overall average silhouette coefficient will also change accordingly. At this time, real-time adjustments can be made based on actual conditions.

[0075] Specifically, if the real-time profile coefficient of a device is lower than the set device threshold, the subnet to which the device belongs is adjusted. If the subnet average profile coefficient of a subnet is low, or even the overall average profile coefficient is lower than the set overall threshold, the network is formed after adjustment. The adjustment method can refer to the aforementioned adjustment method.

[0076] Embodiment 2

[0077] See also Figure 2 Based on the above method, the present invention also provides a device self-organizing network system based on contour coefficient, and the device self-organizing network system based on contour coefficient includes:

[0078] A subnet division module 51 is used to collect device information of all devices and divide all the devices into different subnets according to the device information;

[0079] A subnet adjustment module 52, used to calculate the profile coefficients of all devices and adjust each subnet according to each profile coefficient;

[0080] The subnet networking module 53 is used to control each subnet to be networked according to the device information of the devices included in the subnet.

[0081] Furthermore, dividing all the devices into different subnets according to the device information specifically includes:

[0082] Divide into multiple geographical areas and select a cluster center for each geographical area;

[0083] All the devices are clustered according to all the cluster centers and the device information, and all the devices are divided into different subnets.

[0084] Further, the calculating of the silhouette coefficients of all devices and adjusting each of the subnets according to each of the silhouette coefficients specifically includes:

[0085] Calculate the silhouette coefficient S(i) of the i-th device:

[0086]

[0087] Where a(i) represents the average distance from the ith device to other devices in the same subnet, and b(i) represents the average distance from the ith device to devices outside the same subnet.

[0088] The subnet division effect is evaluated according to the silhouette coefficients of all devices, and each subnet is adjusted according to the evaluation result.

[0089] Furthermore, the subnet division effect is evaluated according to the silhouette coefficients of all devices, and each subnet is adjusted according to the evaluation result, specifically including:

[0090] Determine whether subnet division needs to be performed again based on the contour coefficients of all devices. If necessary, perform subnet division again.

[0091] Control each subnet and adjust the corresponding subnet according to the profile coefficient of the devices contained in the subnet.

[0092] Furthermore, it is determined whether it is necessary to re-subnet the network based on the profile coefficients of all devices. If necessary, the subnet is re-divided, which specifically includes:

[0093] Calculate the average of the silhouette coefficients of all devices to obtain the overall average silhouette coefficient;

[0094] It is determined whether the overall average silhouette coefficient is less than a set overall threshold. If so, subnet division and re-clustering are performed.

[0095] Further, the controlling each subnet to adjust the subnet according to the profile coefficient of the device included in the subnet specifically includes:

[0096] Control each subnet to calculate the average value of the silhouette coefficients of the devices contained in the subnet, and obtain the average silhouette coefficient of each subnet;

[0097] It is determined whether the average silhouette coefficient of each subnet is less than a set subnet threshold. If so, the cluster center of the corresponding subnet is adjusted and clustering is performed again.

[0098] Further, the controlling each subnet adjusts the subnet according to the profile coefficient of the device included in the subnet, and then further includes:

[0099] Determine whether the profile coefficient of the device is lower than the set device threshold. If so, adjust the subnet to which the device belongs.

[0100] Embodiment 3

[0101] See also Figure 3 Based on the above method, the present invention also provides a terminal, which includes a processor 10, a memory 20 and a display 30. Figure 3 Only some components of the terminal are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0102] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal. In other embodiments, the memory 20 may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the terminal. Further, the memory 20 may also include both an internal storage unit of the terminal and an external storage device. The memory 20 is used to store application software and various types of data installed in the terminal, such as the program code of the installation terminal, etc. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, a device self-organizing network program 40 based on a contour coefficient is stored on the memory 20, and the device self-organizing network program 40 based on a contour coefficient can be executed by the processor 10.

[0103] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run the program code or process data stored in the memory 20, such as executing related programs of the device self-organizing networking method based on the contour coefficient.

[0104] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other via a system bus.

[0105] In one embodiment, when the processor 10 executes the device self-organizing network program 40 based on the silhouette coefficient in the memory 20, the steps of the device self-organizing network method based on the silhouette coefficient as described above are implemented.

[0106] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal. In other embodiments, the memory 20 may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. equipped on the terminal. Further, the memory 20 may also include both an internal storage unit of the terminal and an external storage device. The memory 20 is used to store application software and various types of data installed in the terminal, such as the program code of the installation terminal. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, a device self-organizing network program based on a contour coefficient is stored on the memory 20, and the device self-organizing network program based on a contour coefficient can be executed by the processor 10, thereby realizing the device self-organizing network method based on a contour coefficient in the present application.

[0107] In some embodiments, the processor 10 may be a central processing unit, a microprocessor or other data processing chip, used to run the program code or process data stored in the memory 20, such as executing the device self-organizing network method based on the contour coefficient.

[0108] Embodiment 4

[0109] This embodiment provides a storage medium, wherein the readable storage medium stores a device self-organizing network program based on a silhouette coefficient, and when the device self-organizing network program based on a silhouette coefficient is executed by a processor, the steps of the device self-organizing network method based on a silhouette coefficient are implemented.

[0110] In summary, the present invention calculates the contour coefficients of all devices after dividing the networking system into subnets, determines whether the subnet division is reasonable through the contour coefficients, and adjusts the unreasonable parts according to the contour coefficients. After the adjustment is completed, each subnet is networked separately, thereby ensuring the rationality of the subnet division, so as to better optimize the network structure, while ensuring the networking efficiency, reducing the impact of the subnet division on the networking effect as much as possible, and improving the final networking effect.

[0111] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or terminal including the element.

[0112] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer-readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The storage medium can be a memory, a disk, an optical disk, etc.

[0113] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A device self-organizing network method based on silhouette coefficient, characterized in that: The device self-organizing network method based on the silhouette coefficient comprises the steps of: Collect device information of all devices, and divide all the devices into different subnets according to the device information; Calculating the silhouette coefficients of all devices, and adjusting each of the subnets according to each of the silhouette coefficients; Control each subnet to form a network according to the device information of the devices included in the subnet.

2. The device self-organizing network method based on contour coefficient according to claim 1, characterized in that: The dividing all the devices into different subnets according to the device information specifically includes: Divide into multiple geographical areas and select a cluster center for each geographical area; All the devices are clustered according to all the cluster centers and the device information, and all the devices are divided into different subnets.

3. The device self-organizing network method based on contour coefficient according to claim 2, characterized in that: The calculating of the silhouette coefficients of all devices and adjusting each of the subnets according to each of the silhouette coefficients specifically includes: Calculate the silhouette coefficient S(i) of the i-th device: Where a(i) represents the average distance from the ith device to other devices in the same subnet, and b(i) represents the average distance from the ith device to devices outside the same subnet. The subnet division effect is evaluated according to the silhouette coefficients of all devices, and each subnet is adjusted according to the evaluation result.

4. The device self-organizing network method based on silhouette coefficient according to claim 3, characterized in that: The subnet division effect is evaluated according to the silhouette coefficients of all devices, and each subnet is adjusted according to the evaluation result, specifically including: Determine whether subnet division needs to be performed again based on the contour coefficients of all devices. If necessary, perform subnet division again. Control each subnet and adjust the corresponding subnet according to the profile coefficient of the devices contained in the subnet.

5. The device self-organizing network method based on silhouette coefficient according to claim 4, characterized in that: Determine whether subnet division needs to be performed again based on the profile coefficients of all devices. If necessary, perform subnet division again, including: Calculate the average of the silhouette coefficients of all devices to obtain the overall average silhouette coefficient; It is determined whether the overall average silhouette coefficient is less than a set overall threshold. If so, subnet division and re-clustering are performed.

6. The device self-organizing network method based on contour coefficient according to claim 4, characterized in that: The controlling of each subnet to adjust the subnet according to the profile coefficient of the device included in the subnet specifically includes: Control each subnet to calculate the average value of the silhouette coefficients of the devices contained in the subnet, and obtain the average silhouette coefficient of each subnet; It is determined whether the average silhouette coefficient of each subnet is less than a set subnet threshold. If so, the cluster center of the corresponding subnet is adjusted and clustering is performed again.

7. The device self-organizing network method based on silhouette coefficient according to claim 4, characterized in that: The controlling each subnet adjusts the subnet according to the profile coefficient of the devices included in the subnet, and then further includes: Determine whether the profile coefficient of the device is lower than the set device threshold. If so, adjust the subnet to which the device belongs.

8. A device self-organizing network system based on silhouette coefficient, characterized in that: The device self-organizing network system based on the silhouette coefficient includes: A subnet division module is used to collect device information of all devices and divide all the devices into different subnets according to the device information; A subnet adjustment module, used to calculate the profile coefficients of all devices and adjust each subnet according to each profile coefficient; The subnet networking module is used to control each subnet to be networked according to the device information of the devices included in the subnet.

9. A terminal, characterized in that: The terminal includes: a memory, a processor, and a device self-organizing network program based on a contour coefficient stored in the memory and executable on the processor. When the device self-organizing network program based on a contour coefficient is executed by the processor, the terminal is controlled to implement the steps of the device self-organizing network method based on a contour coefficient as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a device self-organizing network program based on a contour coefficient, and when the device self-organizing network program based on a contour coefficient is executed by a processor, the steps of the device self-organizing network method based on a contour coefficient are implemented as described in any one of claims 1-7.

Images