CN-121985302-A - Self-adaptive grouping method and system

CN121985302ACN 121985302 ACN121985302 ACN 121985302ACN-121985302-A

Abstract

The application provides a self-adaptive construction method and system, which belong to the technical field of communication and comprise the steps of obtaining initial construction conditions, screening and obtaining candidate devices based on the initial construction conditions, determining significant features according to historical communication data of the candidate devices, inputting target communication data corresponding to current candidate devices into a random forest regression model according to the significant features, predicting device-environment adaptation degree, screening and obtaining target devices from the candidate devices based on the device-environment adaptation degree, determining dominant adaptation devices through threshold screening, determining channel evaluation areas according to the positions of the dominant adaptation devices, obtaining test information of preset channels in the channel evaluation areas, calculating channel-scene adaptation degree according to the test information, generating channel selection decisions, selecting group members according to the target devices, and constructing according to the channel selection decisions. The application realizes the temporary construction of minute level fast, high reliability and self-adaption under the complex dynamic environment, and ensures the full-link communication guarantee.

Inventors

TIAN WO
GUO MEIHUA
LIN HAIXIN
Shuai Guanhui
ZHANG WEI
HUANG JUNJIE
SONG YANG
SUN RONG
XU QING
TAN QIUYU
FANG YUAN
GU ZHONGJIAN
WANG XIAOFAN
HU JINTAO

Assignees

中铁工程设计咨询集团有限公司
海能达通信股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260209

Claims (10)

1. A method of adaptive networking, comprising: Acquiring initial building conditions, and screening to obtain candidate equipment based on the initial building conditions; According to the historical communication data of the candidate equipment, adopting mutual information entropy to perform feature screening on the historical communication data to obtain significant features; Searching target communication data corresponding to the salient features in a preset period of candidate equipment according to the salient features, inputting the target communication data into a pre-trained random forest regression model, and predicting equipment-environment adaptation degree; screening target equipment from the candidate equipment based on the equipment-environment adaptation degree; Threshold screening is carried out on the equipment-environment adaptation degrees of all the target equipment, leading adaptation equipment is obtained, and a channel evaluation area is determined according to the position of the leading adaptation equipment; in the channel evaluation area, test information of a preset channel is acquired, wherein the test information comprises at least one of communication quality information, interference information, transmission performance information and resource occupation information; Calculating the 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 constructing a group according to the channel selection decision.
2. The method of adaptive networking of claim 1, wherein the feature screening the historical communication data with mutual information entropy according to the historical communication data of the candidate device, to obtain salient features, comprises: Inputting the historical equipment information, the historical environment information, the historical network information and the historical task information of the candidate equipment into an edge node, extracting basic characteristics and acquiring a corresponding communication result label; calculating the correlation between each basic feature and the communication result label by using mutual information entropy; And screening the basic features according to the correlation to obtain the remarkable features.
3. The method of adaptive networking of claim 1, wherein the determining a channel assessment area based on the location of the dominant adaptation device comprises: Clustering based on the geographic position coordinates of the dominant adaptive equipment by adopting a spatial clustering algorithm based on density, and dividing at least one equipment cluster; In each device cluster, selecting one device as the central device of the cluster according to the device-environment adaptation degree, the residual electric quantity and the intra-cluster centrality of each device; calculating the effective communication radius of each center device according to the transmitting power of each center device and the geographical environment of the position of each center device, and generating a plurality of coverage circles; And taking the union of all the coverage circles to obtain a final channel evaluation area.
4. The method of adaptive networking of claim 1, wherein calculating a channel-scene adaptation based on the test information comprises: Acquiring at least one available channel; for each channel, calculating corresponding evaluation indexes according to the communication quality information, the interference information, the transmission performance information and the resource occupation information; constructing corresponding judgment matrixes for different preset service scenes by using a 1-9 scale method; Calculating the weight of each type of test information according to the judgment matrix; And calculating the comprehensive value of each channel based on the weight and the evaluation index of each type of test information to obtain the channel-scene adaptation degree.
5. The method of adaptive grouping according to claim 1, wherein said grouping according to the channel selection decision comprises: Constructing a twin frame according to the basic geographic information, the frequency spectrum map and the state of the synchronous equipment; inputting the member list, the channel and the candidate communication parameters into a twin framework for risk deduction, and obtaining the communication risk trend of each building scheme according to the risk deduction; and selecting candidate communication parameters according to the communication risk trend to obtain target communication parameters.
6. The method of adaptive networking of claim 5, wherein deriving the communication risk trend for each networking scenario based on risk deduction comprises: Determining a device planning track according to the member list, simulating a link budget by combining basic geographic information, and generating a connectivity trend curve; determining the position of equipment according to the member list, and predicting the change of signal to noise ratio by simulating the propagation and superposition of equipment signals in a frequency spectrum map to generate an interference trend curve; generating virtual service data by adopting a service model, and generating a capacity trend curve by simulating queuing and transmission of the virtual service data in a channel; And calculating the energy consumption rate by adopting an electrochemical model according to the queuing and transmission simulation result, and generating a continuous voyage trend curve.
7. A system for adaptive networking, comprising: the first module is used for acquiring initial building set conditions and screening candidate equipment based on the initial building set conditions; The second module is used for screening the characteristics of the historical communication data by adopting mutual information entropy according to the historical communication data of the candidate equipment to obtain obvious characteristics; The third module searches target communication data corresponding to the salient features in a preset period of candidate equipment according to the salient features, inputs the target communication data into a pre-trained random forest regression model, and predicts equipment-environment adaptation degree; A fourth module, configured to screen the target device from the candidate devices based on the device-environment adaptation degree; A fifth module, configured to perform threshold screening on device-environment adaptation degrees of all the target devices, obtain a dominant adaptation device, and determine a channel estimation area according to a location of the dominant adaptation device; a sixth module, configured to obtain, in the channel estimation area, test information of a preset channel, where the test information includes at least one of communication quality information, interference information, transmission performance information, and resource occupation information; a seventh module, configured to calculate a channel-scene adaptation degree according to the test information, and generate a channel selection decision; And an eighth module, configured to select a group member according to the target device, and perform group establishment according to the channel selection decision.
8. The adaptive networking system of claim 1, wherein the second module comprises: the first unit is used for inputting the edge node according to the historical equipment information, the historical environment information, the historical network information and the historical task information of the candidate equipment, extracting basic characteristics and obtaining a corresponding communication result label; The second unit is used for calculating the correlation between each basic feature and the communication result label by using mutual information entropy; and the third unit is used for screening the basic features according to the correlation to obtain the obvious features.
9. The adaptive networking system of claim 1, wherein the fifth module comprises: A fourth unit, configured to perform clustering based on the geographic location coordinates of the dominant adaptive device by using a spatial clustering algorithm based on density, and divide at least one device cluster; a fifth unit for selecting, in each device cluster, a device as a central device of the cluster according to the device-environment adaptation degree, the remaining power, and the intra-cluster centrality of each device; a sixth unit, configured to calculate an effective communication radius of each center device according to the transmitting power of each center device and the geographical environment where the center device is located, and generate a plurality of coverage circles; And a seventh unit, configured to obtain a final channel estimation area by taking the union of all the coverage circles.
10. The adaptive networking system of claim 1, wherein the seventh module comprises: An eighth unit, configured to acquire at least one available channel; A ninth unit, configured to calculate, for each channel, a corresponding evaluation index according to the communication quality information, the interference information, the transmission performance information, and the resource occupation information; Tenth unit, is used for using 1-9 scale method to construct the corresponding judgement matrix for different preset business scenes; an eleventh unit for calculating the weight of each type of test information according to the judgment matrix; And a twelfth unit for calculating the comprehensive value of each channel based on the weight and the evaluation index of each type of test information to obtain the channel-scene adaptation degree.

Description

Self-adaptive grouping method and system Technical Field The invention belongs to the technical field of communication, and particularly relates to a self-adaptive building method and system. Background In critical tasks such as emergency rescue, large-scale activity security, daily patrol and the like, the private network interphone is a core communication tool, and temporary communication groups are often required to be quickly built across departments and areas. The traditional mode is highly dependent on manual operations of manually inputting members, designating fixed channels and configuring static rights. Along with the increasingly complex and dynamic task scenes, the mode exposes serious problems of low efficiency, poor adaptability, insufficient reliability and the like, and can not meet the core requirements of 'quick response, accurate coordination and global coverage' of modern emergency command. Disclosure of Invention The present invention is directed to a method and system for adaptive construction, which improves the above-mentioned problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: In a first aspect, the present application provides a method of adaptive networking, comprising: Acquiring initial building conditions, and screening to obtain candidate equipment based on the initial building conditions; according to the historical communication data of the candidate equipment, adopting mutual information entropy to perform feature screening on the historical communication data to obtain significant features; Searching target communication data corresponding to the salient features in a preset period of candidate equipment according to the salient features, inputting the target communication data into a pre-trained random forest regression model, and predicting equipment-environment adaptation degree; screening target equipment from the candidate equipment based on the equipment-environment adaptation degree; Threshold screening is carried out on the equipment-environment adaptation degrees of all the target equipment, leading adaptation equipment is obtained, and a channel evaluation area is determined according to the position of the leading adaptation equipment; in the channel evaluation area, test information of a preset channel is acquired, wherein the test information comprises at least one of communication quality information, interference information, transmission performance information and resource occupation information; Calculating the 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 constructing a group according to the channel selection decision. In a second aspect, the present application provides a system for adaptive networking, comprising: the first module is used for acquiring initial building set conditions and screening candidate equipment based on the initial building set conditions; The second module is used for screening the characteristics of the historical communication data by adopting mutual information entropy according to the historical communication data of the candidate equipment to obtain obvious characteristics; The third module searches target communication data corresponding to the salient features in a preset period of candidate equipment according to the salient features, inputs the target communication data into a pre-trained random forest regression model, and predicts equipment-environment adaptation degree; A fourth module, configured to screen out a target device from the candidate devices based on the device-environment adaptation degree; A fifth module, configured to perform threshold screening on device-environment adaptation degrees of all the target devices, obtain a dominant adaptation device, and determine a channel estimation area according to a location of the dominant adaptation device; A sixth module for acquiring test information of a preset channel in the channel estimation area, the test information including at least one of communication quality information, interference information, transmission performance information, and resource occupation information, A seventh module, configured to calculate a channel-scene adaptation degree according to the test information, and generate a channel selection decision; And an eighth module, configured to select a group member according to the target device, and perform group establishment according to the channel selection decision. The beneficial effects of the invention are as follows: According to the method, the physical state of the equipment and external environment data are fused, a random forest regression model is used for screening target equipment adapting to a scene, a channel-scene matching body is constructed, dynamic scoring is carried out, a twin frame is built by combining geography and spectrum