Self-adaptive group building method and system

Abstract

The invention provides a self-adaptive group building method and system, and belongs to the technical field of communication, and the method comprises the steps: obtaining an initial group building condition, and carrying out the screening based on the initial group building condition, and obtaining candidate equipment; determining significant features according to the historical communication data of the candidate device, inputting target communication data corresponding to the current candidate device into a random forest regression model according to the significant features, and predicting a device-environment adaptation degree; target equipment is screened from the candidate equipment based on the equipment-environment adaptation degree, dominant adaptation equipment is determined through threshold value screening, and a channel evaluation area is determined according to the position of the dominant adaptation equipment; acquiring test information of a preset channel in the channel evaluation area; calculating a channel-scene adaptation degree according to the test information, and generating a channel selection decision; and selecting group members according to the target equipment, and establishing a group according to a channel selection decision. According to the invention, minute-level rapid, high-reliability and self-adaptive temporary establishment in a complex dynamic environment is realized, and full-link communication guarantee is ensured.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, specifically relating to an adaptive group building method and system. Background Technology

[0002] In critical tasks such as emergency rescue, security for large-scale events, and routine patrols, dedicated network walkie-talkies are core communication tools, often requiring the rapid formation of temporary communication teams across departments and regions. Traditional methods heavily rely on manual operation: manually entering members, assigning fixed channels, and configuring static permissions. As mission scenarios become increasingly complex and dynamic, this approach reveals serious problems such as low efficiency, poor adaptability, and insufficient reliability, failing to meet the core requirements of modern emergency command: "rapid response, precise coordination, and full coverage." Summary of the Invention

[0003] The purpose of this invention is to provide an adaptive grouping method and system to improve the aforementioned problems. To achieve this purpose, the technical solution adopted by this invention is as follows:

[0004] Firstly, this application provides an adaptive group building method, including:

[0005] Obtain the initial group formation conditions, and filter candidate devices based on the initial group formation conditions;

[0006] Based on the historical communication data of candidate devices, mutual information entropy is used to filter the historical communication data to obtain significant features;

[0007] Based on the salient features, find the target communication data corresponding to the salient features within a preset time period of the candidate device, and input the target communication data into a pre-trained random forest regression model to predict the device-environment fit.

[0008] The target device is selected from the candidate devices based on the device-environment compatibility.

[0009] Threshold filtering is performed on the device-environment adaptability of all the target devices to obtain the dominant adapting device, and the channel evaluation area is determined according to the location of the dominant adapting device;

[0010] In the channel evaluation area, test information of a preset channel is obtained, and the test information includes at least one of communication quality information, interference information, transmission performance information, and resource usage information.

[0011] Calculate the channel-scenario fit based on the test information, and generate a channel selection decision.

[0012] Group members are selected based on the target device, and the group is created based on the channel selection decision.

[0013] Secondly, this application provides an adaptive group creation system, comprising:

[0014] The first module is used to obtain the initial group formation conditions and filter candidate devices based on the initial group formation conditions.

[0015] The second module is used to perform feature filtering on the historical communication data of candidate devices using mutual information entropy to obtain significant features;

[0016] The third module searches for target communication data corresponding to the salient features within a preset time period for candidate devices based on the salient features, and inputs the target communication data into a pre-trained random forest regression model to predict device-environment fit.

[0017] The fourth module is used to select the target device from the candidate devices based on the device-environment compatibility.

[0018] The fifth module is used to perform threshold filtering on the device-environment adaptability of all the target devices, obtain the dominant adapting device, and determine the channel evaluation area based on the location of the dominant adapting device.

[0019] The sixth module is used to acquire test information of a preset channel in the channel evaluation area. The test information includes at least one of communication quality information, interference information, transmission performance information, and resource usage information.

[0020] The seventh module is used to calculate the channel-scenario adaptability based on the test information and generate a channel selection decision;

[0021] The eighth module is used to select group members based on the target device and to create a group based on the channel selection decision.

[0022] The beneficial effects of this invention are as follows:

[0023] This invention integrates the physical state of the device with external environmental data, uses a random forest regression model to screen target devices that are suitable for the scenario, constructs a channel-scenario matching entity and performs dynamic scoring, and builds a twin framework by combining geographical and spectrum information to perform multi-dimensional risk simulation on candidate solutions, thereby achieving optimal matching of device, channel and parameters, effectively avoiding problems such as communication interruption, congestion and excessive energy consumption, improving the stability, adaptability and energy efficiency of group communication, and adapting to complex dynamic scenarios.

[0024] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. Attached Figure Description

[0025] 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.

[0026] Figure 1 This is a flowchart of the adaptive grouping method in an embodiment of this application;

[0027] Figure 2 This is a schematic diagram of the device structure for the adaptive grouping method in an embodiment of this application.

[0028] Symbol explanation: 800 - Device for adaptive grouping; 801 - Processor; 802 - Memory; 803 - Multimedia component; 804 - I / O interface; 805 - Communication component. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. 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.

[0030] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0031] Existing walkie-talkie group creation schemes are essentially based on a "person finds device, person matches network" model using predefined rules and static configurations. Their main drawback lies in the system's lack of awareness and intelligent response capabilities to dynamic environments and real-time device status. Member selection relies solely on static identity tags such as job title and department, typically neglecting crucial factors like real-time walkie-talkie battery level, signal strength, and the surrounding terrain and electromagnetic environment. This often results in "selected people, but devices unable to communicate." Furthermore, temporary group configuration parameters such as channel, bandwidth, and permissions are pre-configured and fixed, unable to be dynamically adjusted based on real-time task requirements, network congestion, and heterogeneous device capabilities, leading to resource waste or performance bottlenecks.

[0032] Example 1:

[0033] See Figure 1 In order to solve the problems of the prior art, this embodiment provides an adaptive group building method, including steps S100, S200, S300, S400, S500, S600, S700 and S800.

[0034] S100: Obtain the initial grouping conditions and filter candidate devices based on the initial grouping conditions;

[0035] Administrators can initiate group creation requests from the command center or via terminals, setting initial screening criteria such as target area, personnel tags (department, job level), and scenario type (e.g., "mountain rescue").

[0036] Select the devices that meet the criteria from the list of all devices to obtain a list of candidate devices;

[0037] S200. Based on the historical communication data of the candidate devices, mutual information entropy is used to perform feature filtering on the historical communication data to obtain significant features; the historical communication data includes device information, environmental information, network information, and task information.

[0038] Equipment information includes: remaining battery power (%) and discharge curve, received signal strength (RSSI, dBm), GPS / BeiDou location, moving speed, equipment model, capability attenuation rate, RF aging index, etc.

[0039] Environmental information includes: real-time spectrum data (SNR, interference intensity), 3D terrain data (elevation, obstructions), building penetration loss (based on GIS data), rain attenuation, temperature and humidity, multipath fading intensity, dynamic occlusion probability, etc. This environmental information is collected through edge nodes.

[0040] Network information includes: channel occupancy, end-to-end latency (average of the last 10 times), packet loss rate statistics, neighbor node density, relay capability score, etc.

[0041] Task information includes: scene type code (mountainous area=1, city=2, sea surface=3, mining area=4...), task type (routine patrol, emergency rescue, large-scale event, etc.), and expected task duration.

[0042] The above information is preprocessed, including synchronizing the data through an NTP time server to ensure spatiotemporal consistency; and applying the 3σ criterion to each data item to eliminate transient anomalies, such as GPS drift and sudden drops in RSSI caused by sudden noise.

[0043] The preprocessed device information, environmental information, network information, and task information are input into the edge node to extract basic features and obtain the corresponding communication result labels.

[0044] Extracting basic features involves directly obtaining data values ​​of continuous information and extracting the codes of discrete information to obtain multiple basic features;

[0045] The correlation between each basic feature and the communication result label is calculated using mutual information entropy; the communication result, i.e., whether the communication was successful or failed, is a binary label.

[0046] Mutual information entropy is the calculation of the joint probability distribution of each basic feature and the communication result label. It is achieved by statistically analyzing the frequency of occurrence of the combination of feature value and label value of each basic feature. The calculation result is the correlation between the basic feature and the communication result label.

[0047] Based on the correlation, the basic features are filtered to obtain significant features. In this embodiment, only features with an importance of ≥0.2 are retained. Filtering significant features means finding the data types that have the greatest impact on communication performance. Subsequently, the data values ​​corresponding to these important data types are used to predict device-environment adaptability.

[0048] S300. Based on the significant features, find the target communication data of the candidate device within a preset time period that corresponds to the significant features, and input the target communication data into a pre-trained random forest regression model to predict the device-environment fit.

[0049] Obtain communication data of candidate devices in the current time period, such as communication data in the last 12 hours; extract data corresponding to significant features from them, such as significant features selected in step S200 including received signal strength, GPS / BeiDou location, scene type code, task type, spectrum data, and building penetration loss, that is, find the corresponding data values ​​(target communication data) in the last 12 hours based on these features, and use them as input to the random forest regression model;

[0050] Input the target communication data obtained into the random forest regression model, and the model outputs the device-environment fit score (0-10 points).

[0051] When training a random forest regression model, the device-environment fit labels of the training samples are calculated based on the communication quality of each communication. The main factors for measuring communication quality include signal strength, bit error rate, transmission delay, and sound quality.

[0052] S400: Based on the device-environment compatibility, select the target device from the candidate devices;

[0053] In this embodiment, devices are classified according to their device-environment compatibility scores, as detailed below:

[0054] High compatibility (8-10 points): Selected target devices, green label.

[0055] Medium compatibility (6-8 points): Selected target devices, marked in yellow;

[0056] Low compatibility (<6 points): Automatically excluded, marked in red;

[0057] Each device supports receiving and executing dynamic configuration commands and has a built-in emergency self-organizing protocol, enabling it to autonomously form a network even without a network connection. In the event of a network outage, the highly adaptable device broadcasts a network establishment beacon via an emergency frequency band, and surrounding terminals automatically join based on the self-organizing Mesh protocol, forming a multi-hop relay network. If the highly adaptable device malfunctions or is poorly positioned, the mid-adaptive device will act as the master node for network formation.

[0058] S500: Perform threshold filtering on the device-environment adaptability of all the target devices to obtain the dominant adapting device, and determine the channel evaluation area based on the location of the dominant adapting device;

[0059] Pre-set a high threshold (e.g., the fit is in the top 10% or the fit is >0.85), and select the dominant fitting devices that meet the threshold conditions;

[0060] S510. A density-based spatial clustering algorithm is used to cluster devices based on the geographical coordinates of the dominant adaptation devices, and at least one device cluster is formed.

[0061] In order to group devices that are close in location and have high compatibility together to form multiple clusters, this step uses the DBSCAN clustering algorithm to automatically identify "core points" and "noise points" and does not require pre-specifying the number of groups, making it suitable for irregular geographical distributions.

[0062] Set a distance threshold ,For example For 500 meters or the minimum effective propagation distance of radio waves, if the distance between the two dominant adaptors is less than... If they belong to the same potential cluster, then they are considered to belong to the same potential cluster, eventually forming K clusters.

[0063] S520. In each device cluster, select a device as the cluster center device based on each device's device-environment adaptability, remaining power, and cluster centrality.

[0064] The device-environment compatibility, remaining power, and cluster centrality are weighted and summed. The device with the highest calculated value is selected as the central device and represents the region.

[0065] S530: Calculate the effective communication radius of each central device based on its transmission power and geographical location, and generate multiple coverage circles.

[0066] Each central device calculates an effective radius based on its own transmission power and environmental attenuation model;

[0067] Environmental attenuation models describe the attenuation characteristics of signals under different environmental conditions, thereby predicting the signal propagation effect. These models are pre-built based on the task environment and the characteristics of the device itself. For example, a free-space attenuation model is used in an unobstructed open environment. When in use, the corresponding model is called based on the device's signal strength and location. When the signal attenuates to a preset value, the corresponding propagation distance is the effective communication radius.

[0068] S540. Take the union of all the coverage circles to obtain the final channel evaluation area.

[0069] This method focuses on evaluating channels in areas where central devices are highly adaptable to the environment, reducing computational overhead. Furthermore, selecting channels based on the location of the central devices ensures more stable communication during subsequent task execution, thus improving scenario adaptability.

[0070] S600. In the channel evaluation area, acquire test information of a preset channel, wherein the test information includes at least one of communication quality information, interference information, transmission performance information, and resource usage information;

[0071] S700. Calculate the channel-scene adaptability based on the test information and generate a channel selection decision; specifically as follows:

[0072] S710. Obtain at least one available channel, namely the preset channel mentioned above;

[0073] S720. For each channel, calculate the corresponding evaluation index based on communication quality information, interference information, transmission performance information, and resource usage information.

[0074] Evaluation metrics for communication quality include: signal-to-noise ratio (SNR), bit error rate (BER), received signal strength (RSSI), and modulation order stability.

[0075] The evaluation indicators for interference status include: co-channel interference intensity, adjacent channel interference intensity, probability of sudden interference, and type of interference source (interference types such as WiFi / Bluetooth / microwave are identified through a spectrum fingerprint database).

[0076] Evaluation metrics for resource utilization (channel load) include: channel utilization rate, number of competing nodes, average utilization duration, and service type mixing degree (mixing of different services such as voice / data / video).

[0077] Evaluation metrics for transmission performance include: end-to-end latency, jitter, packet loss rate, and retransmission rate.

[0078] This method pre-constructs an indicator rating table, and finds the corresponding rating in the rating table based on the indicator value to obtain the rating result.

[0079] S730. Use the 1-9 scaling method to construct corresponding judgment matrices for different preset business scenarios; the judgment matrix is ​​used to measure the importance of data in different dimensions; the 1-9 scaling method is used to compare two indicators pairwise.

[0080] 1. Both indicators are equally important;

[0081] 3: The former is slightly more important than the latter;

[0082] 5: The former is significantly more important than the latter;

[0083] 7: The former is more strongly important than the latter;

[0084] 9: The former is extremely important than the latter;

[0085] 2, 4, 6, 8 are the intermediate values;

[0086] As an example, in emergency rescue scenarios, communication quality and transmission performance are the core factors, followed by interference, while resource consumption has a relatively low weight. The corresponding judgment matrix elements are shown in Table 1:

[0087] Table 1

[0088]

[0089] In comparison, the first row in the column is considered; therefore, communication quality is slightly more important than transmission performance, communication quality is significantly more important than interference conditions, transmission performance is slightly more important than interference conditions, and so on.

[0090] S740. Calculate the weight of each type of test information based on the judgment matrix;

[0091] Each type of test information represents the test information for each dimension. First, the judgment matrix is ​​normalized by calculating the sum of each row. Then, each matrix element is divided by the sum of its row to obtain the normalized matrix. The average value of each column of the normalized matrix is ​​calculated to obtain the eigenvector W, which represents the weights of each dimension.

[0092] S750: Based on the weight and evaluation index of each type of test information, calculate the comprehensive value of each channel to obtain the channel-scenario adaptability.

[0093] The weight of each type of test information is multiplied by the score of the indicator under that type of test information, and then the weighted total score of each indicator is calculated to obtain the comprehensive value of each channel, which is used as the channel-scenario adaptability.

[0094] S800: Select group members according to the target device, and create a group according to the channel selection decision.

[0095] This step requires shifting the focus of communication risks from "post-event discovery" to "pre-event avoidance." Therefore, before issuing commands to physical devices, a rapid simulation and risk assessment of the group setup scheme is conducted in a virtual digital twin environment using real-time data to determine the optimal communication parameters; specifically as follows:

[0096] S810. Construct a twin framework based on basic geographic information, spectrum map, and synchronization device status;

[0097] S820. Input the member list, channel, and candidate communication parameters into the twin framework to perform risk simulation; obtain the communication risk trend of each group formation scheme based on the risk simulation; candidate communication parameters include bandwidth and permissions;

[0098] Based on the member list, the planned trajectory of the equipment is determined, and the link budget is simulated by combining basic geographic information to generate a connectivity trend curve;

[0099] Specifically, the planned movement trajectory of each device is extracted based on the member list. Combined with the basic geographic information (terrain, building distribution, etc.) in the twin framework, link budget simulation is carried out for each device to calculate the signal propagation loss, received power and other indicators between devices at different time nodes. Based on the simulation results, it is determined whether there is a line-of-sight direct path or severe obstruction between devices. A connectivity trend curve is generated with time as the horizontal axis and connectivity compliance rate as the vertical axis.

[0100] Based on the member list, determine the device location, and predict the signal-to-noise ratio change by simulating the propagation and superposition of device signals in the spectrum map, thereby generating an interference trend curve.

[0101] Specifically, based on the member list, the real-time simulation position of each device is determined. Combined with the spectrum map in the twin framework (including background noise, interference source distribution, etc.), the propagation and superposition process of each device's signal in the selected channel is simulated. The signal-to-noise ratio change of the channel at different time points is predicted in real time, potential co-channel interference and adjacent channel interference risks are identified, and interference trend curves are generated (the horizontal axis is time, and the vertical axis is signal-to-noise ratio / interference intensity).

[0102] Virtual service data is generated using a service model, and a capacity trend curve is generated by simulating the queuing and transmission of virtual service data in the channel.

[0103] Specifically, based on the business requirements of the current group building scenario, virtual business data is generated using an adapted business model (such as a voice ON-OFF model or a data burst model); the queuing, competition, and transmission process of virtual business data in the channel is simulated in the twin framework, and the queue length, channel utilization, and packet loss at each time node are statistically analyzed in real time; combined with the bandwidth specifications in the candidate communication parameters, a capacity trend curve is generated (the horizontal axis is time, and the vertical axis is queue length / channel utilization / packet loss rate).

[0104] Based on queuing and transmission simulation results, an electrochemical model is used to calculate the energy consumption rate and generate a range trend curve.

[0105] Specifically, the timing simulation results of the queuing and transmission of the aforementioned virtual service data (including the proportion of device transmission / reception / idle time, transmission power and other load information) are extracted and input into the electrochemical model; the transient energy consumption rate and remaining power decay process of each device are calculated through model iteration, and the permission level in the candidate communication parameters (such as the service load characteristics of high-privilege devices) are combined to generate the battery life trend curve (the horizontal axis is time, and the vertical axis is the remaining power / SOC value).

[0106] S830. Select candidate communication parameters based on communication risk trends to obtain target communication parameters.

[0107] The optimal communication parameters should be selected by considering various risk trends, such as choosing a scheme with stable connectivity trends or a scheme with balanced performance across different aspects.

[0108] This invention is applicable to the field of private network communication technology, especially to scenarios that require rapid group setup and dynamic communication, such as emergency rescue, urban security, large-scale event coordination, and daily patrols based on walkie-talkies; it can be extended to fields such as collaborative networking of IoT devices and construction of industrial control field communication groups, and has broad technical application prospects.

[0109] Example 2:

[0110] An adaptive grouping system, comprising:

[0111] The first module is used to obtain the initial group formation conditions and filter candidate devices based on the initial group formation conditions.

[0112] The second module is used to perform feature filtering on the historical communication data of candidate devices using mutual information entropy to obtain significant features;

[0113] The third module searches for target communication data corresponding to the salient features within a preset time period for candidate devices based on the salient features, and inputs the target communication data into a pre-trained random forest regression model to predict device-environment fit.

[0114] The fourth module is used to filter the target device from the candidate devices based on device-environment compatibility.

[0115] The fifth module is used to perform threshold filtering on the device-environment adaptability of all the target devices, obtain the dominant adapting device, and determine the channel evaluation area based on the location of the dominant adapting device.

[0116] The sixth module is used to acquire test information of a preset channel in the channel evaluation area. The test information includes at least one of communication quality information, interference information, transmission performance information, and resource usage information.

[0117] The seventh module is used to calculate the channel-scenario adaptability based on the test information and generate a channel selection decision;

[0118] The eighth module is used to select group members based on the target device and to create a group based on the channel selection decision.

[0119] As an optional implementation, the second module includes:

[0120] The first unit is used to input the candidate device's historical device information, historical environment information, historical network information, and historical task information into the edge node, extract basic features, and obtain the corresponding communication result labels.

[0121] The second unit is used to calculate the correlation between the basic features and the communication result tags using mutual information entropy;

[0122] The third unit is used to filter the basic features based on the correlation to obtain significant features.

[0123] As an optional implementation, the fifth module includes:

[0124] The fourth unit is used to perform clustering based on the geographic location coordinates of the dominant adaptation device using a density-based spatial clustering algorithm, and to divide the device into at least one cluster.

[0125] The fifth unit is used to select a device as the central device of each device cluster based on each device's device-environment adaptability, remaining power, and cluster centrality.

[0126] The sixth unit is used to calculate the effective communication radius of each central device based on its transmission power and geographical location, and to generate multiple coverage circles.

[0127] The seventh unit is used to take the union of all the covered circles to obtain the final channel evaluation area.

[0128] As an optional implementation, the seventh module includes:

[0129] The eighth unit is used to acquire at least one available channel;

[0130] The ninth unit is used to calculate the corresponding evaluation index for each channel based on communication quality information, interference information, transmission performance information, and resource usage information.

[0131] Unit 10 is used to construct corresponding judgment matrices for different preset business scenarios using the 1-9 scale method;

[0132] Unit 11 is used to calculate the weight of each type of test information based on the judgment matrix;

[0133] The twelfth unit is used to calculate the comprehensive value of each channel based on the weight and evaluation index of each type of test information, and to obtain the channel-scenario fit.

[0134] Example 3:

[0135] Corresponding to the above method embodiments, this embodiment also provides an adaptive grouping device. The adaptive grouping device described below and the adaptive grouping method described above can be referred to each other.

[0136] Figure 2 This is a block diagram illustrating an adaptive grouping device 800 according to an exemplary embodiment. For example... Figure 2 As shown, the adaptive grouping device 800 includes a processor 801 and a memory 802. The adaptive grouping device 800 may also include one or more of a multimedia component 803, an input / output (I / O) interface 804, and a communication component 805. The processor 801 controls the overall operation of the adaptive grouping device 800 to complete all or part of the steps in the adaptive grouping method described above. The memory 802 stores various types of data to support the operation of the adaptive grouping device 800. This data may include, for example, commands for any application or method operating on the adaptive grouping device 800, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0137] Multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals.

[0138] The received audio signal can be further stored in memory 802 or transmitted via communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the adaptively grouped device 800 and other devices. Wireless communication includes, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof; therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0139] Example 4:

[0140] Corresponding to the above embodiments of the adaptive grouping method, this embodiment also provides a readable storage medium. The readable storage medium described below corresponds to the adaptive grouping method described above.

[0141] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described adaptive grouping method embodiment.

[0142] Specifically, the readable storage medium can be a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or any other readable storage medium capable of storing program code.

[0143] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0144] 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 variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An adaptive grouping method, characterized in that, include: Obtain the initial group formation conditions, and filter candidate devices based on the initial group formation conditions; Based on the historical communication data of the candidate devices, mutual information entropy is used to perform feature filtering on the historical communication data to obtain significant features; Based on the salient features, find the target communication data corresponding to the salient features within a preset time period of the candidate device, and input the target communication data into a pre-trained random forest regression model to predict the device-environment fit. The target device is selected from the candidate devices based on the device-environment compatibility. Threshold filtering is performed on the device-environment adaptability of all the target devices to obtain the dominant adapting device, and the channel evaluation area is determined according to the location of the dominant adapting device; In the channel evaluation area, test information of a preset channel is obtained, and the test information includes at least one of communication quality information, interference information, transmission performance information, and resource usage information. Calculate the channel-scenario fit based on the test information, and generate a channel selection decision. Group members are selected based on the target device, and the group is created based on the channel selection decision.

2. The adaptive grouping method according to claim 1, characterized in that, The step of using mutual information entropy to perform feature filtering on the historical communication data of the candidate devices to obtain significant features includes: Based on the historical device information, historical environment information, historical network information, and historical task information of the candidate devices, the edge nodes are input, basic features are extracted, and corresponding communication result labels are obtained. The correlation between the basic features and the communication result tags is calculated using mutual information entropy; The basic features are filtered based on the correlation to obtain significant features.

3. The adaptive grouping method according to claim 1, characterized in that, The step of determining the channel evaluation area based on the location of the dominant adaptation device includes: A density-based spatial clustering algorithm is used to cluster the dominant adaptation devices based on their geographic location coordinates, thereby dividing them into at least one device cluster. Within each device cluster, a device is selected as the cluster's central device based on each device's device-environment compatibility, remaining power, and cluster centrality. Based on the transmission power and geographical environment of each central device, the effective communication radius of each central device is calculated, and multiple coverage circles are generated. The final channel evaluation region is obtained by taking the union of all the coverage circles.

4. The adaptive grouping method according to claim 1, characterized in that, Calculate the channel-scene adaptability based on the test information, including: Obtain at least one available channel; For each channel, calculate the corresponding evaluation index based on communication quality information, interference information, transmission performance information, and resource usage information; The 1-9 scaling method is used to construct corresponding judgment matrices for different preset business scenarios; Calculate the weight of each type of test information based on the judgment matrix; Based on the weight and evaluation index of each type of test information, the comprehensive value of each channel is calculated to obtain the channel-scenario fit.

5. The adaptive grouping method according to claim 1, characterized in that, The process of creating groups based on the channel selection decision includes: A twin framework is constructed based on basic geographic information, spectrum maps, and the status of synchronization devices; Input the member list, channel, and candidate communication parameters into the twin framework to perform risk simulation; obtain the communication risk trend of each group formation scheme based on the risk simulation; Candidate communication parameters are selected based on communication risk trends to obtain the target communication parameters.

6. The adaptive grouping method according to claim 5, characterized in that, Based on risk projection, the communication risk trends for each group setup scheme are obtained, including: Based on the member list, the planned trajectory of the equipment is determined, and the link budget is simulated by combining basic geographic information to generate a connectivity trend curve; Based on the member list, determine the device location, and predict the signal-to-noise ratio change by simulating the propagation and superposition of device signals in the spectrum map, thereby generating an interference trend curve. Virtual service data is generated using a service model, and a capacity trend curve is generated by simulating the queuing and transmission of virtual service data in the channel. Based on the queuing and transmission simulation results, an electrochemical model is used to calculate the energy consumption rate and generate a range trend curve.

7. An adaptive grouping system, characterized in that, include: The first module is used to obtain the initial group formation conditions and filter candidate devices based on the initial group formation conditions. The second module is used to perform feature filtering on the historical communication data of candidate devices using mutual information entropy to obtain significant features; The third module searches for target communication data corresponding to the salient features within a preset time period for candidate devices based on the salient features, and inputs the target communication data into a pre-trained random forest regression model to predict device-environment fit. The fourth module is used to filter the target device from the candidate devices based on device-environment compatibility. The fifth module is used to perform threshold filtering on the device-environment adaptability of all the target devices, obtain the dominant adapting device, and determine the channel evaluation area based on the location of the dominant adapting device. The sixth module is used to acquire test information of a preset channel in the channel evaluation area. The test information includes at least one of communication quality information, interference information, transmission performance information, and resource usage information. The seventh module is used to calculate the channel-scenario adaptability based on the test information and generate a channel selection decision; The eighth module is used to select group members based on the target device and to create a group based on the channel selection decision.

8. The adaptive grouping system according to claim 1, characterized in that, The second module includes: The first unit is used to input the candidate device's historical device information, historical environment information, historical network information, and historical task information into the edge node, extract basic features, and obtain the corresponding communication result labels. The second unit is used to calculate the correlation between the basic features and the communication result tags using mutual information entropy; The third unit is used to filter the basic features based on the correlation to obtain significant features.

9. The adaptive grouping system according to claim 1, characterized in that, The fifth module includes: The fourth unit is used to perform clustering based on the geographic location coordinates of the dominant adaptation device using a density-based spatial clustering algorithm, and to divide the device into at least one cluster. The fifth unit is used to select a device as the central device of each device cluster based on each device's device-environment adaptability, remaining power, and cluster centrality. The sixth unit is used to calculate the effective communication radius of each central device based on its transmission power and geographical location, and to generate multiple coverage circles. The seventh unit is used to take the union of all the covered circles to obtain the final channel evaluation area.

10. The adaptive grouping system according to claim 1, characterized in that, The seventh module includes: The eighth unit is used to acquire at least one available channel; The ninth unit is used to calculate the corresponding evaluation index for each channel based on communication quality information, interference information, transmission performance information, and resource usage information. Unit 10 is used to construct corresponding judgment matrices for different preset business scenarios using the 1-9 scale method; Unit 11 is used to calculate the weight of each type of test information based on the judgment matrix; The twelfth unit is used to calculate the comprehensive value of each channel based on the weight and evaluation index of each type of test information, and to obtain the channel-scenario fit.

Images