An intelligent and simple wireless network and its operation method

Abstract

The present invention relates to an intelligent wireless network and an operation method thereof, the method comprising: step 1, according to a user request time T request The network maintenance time T of the intelligent wireless network maintain The minimum time interval T value in the network is used to obtain the overall network status information state of the intelligent wireless network. all , and initialize the slice state NS of the wireless communication cloud computing platform init ; Step 2, the wireless communication edge computing platform performs preliminary processing on the target state of the wireless communication edge computing platform and uploads it to the wireless communication cloud computing platform; Step 3, the wireless communication cloud computing platform obtains the pre-processed target information state′ user , etc. Compared with the prior art, the present invention can reduce the computing burden and storage pressure of the wireless communication cloud computing platform and solve the problem of incompleteness of data on the wireless communication edge computing platform through the cloud-edge adaptive collaboration of the wireless communication cloud computing platform and the wireless communication edge computing platform, thereby improving the efficiency and flexibly constructing a simple wireless network.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless networking, and in particular to an intelligent wireless network and an operation method thereof. Background Art

[0002] With the continuous development of communication technology, various terminal devices have begun to access wireless networks, and various differentiated business needs have also emerged. In order to meet various different business needs, it is necessary to construct corresponding wireless networks in a targeted manner. In recent years, the widespread application of cloud computing has shown many advantages such as high resource utilization, high scalability, high availability, and low operation and management costs. At the same time, the introduction of edge computing has brought data computing and storage closer to the data source. These advantages provide many technical supports and conveniences for the construction of wireless networks.

[0003] In the prior art, technicians usually determine the parameter information for constructing the network based on the wireless communication cloud computing platform according to different business needs, so as to use the wireless network construction equipment to construct a wireless network that can meet different business needs. However, the future wireless network will bring unprecedented data volume. When the wireless communication cloud computing platform collects and analyzes the large amount of data, it will lead to problems such as excessive computing and storage load, excessive energy consumption, high business latency, and large network bandwidth occupation. The data collected by the wireless communication edge computing platform has local characteristics, cannot be analyzed and optimized globally, and has weak computing power. In addition, since the wireless network construction is completed by technicians according to different business needs, when the number and type of business needs are large, the efficiency of wireless network construction is reduced. Due to the complex wireless network conditions and large changes in business needs, the wireless network constructed based on the prior art lacks flexibility and cannot adapt to changes in network conditions and business needs.

[0004] In the existing patent technology, for example, Chinese invention patent application number CN201910963276.2 discloses a cloud computing of PEMS, RDE and NRE, including a cloud computing platform, a user login module, a basic calculation input module, an upload test data module, an output report module, a basic remote vehicle terminal, a mobile terminal and a wireless communication module. The user login module, the output report module, the basic calculation input module and the upload test data module are arranged inside the cloud computing platform, and the remote vehicle terminal and the mobile terminal are both connected to the cloud computing platform through the wireless communication module. The present invention also provides a method for using cloud computing of PEMS, RDE and NRE. The present invention establishes a unified, standardized and efficient PEMS data post-processing unified platform, which can process PEMS device data, ECU data, mobile phone APP data and data recorded by the remote vehicle terminal. According to laws and regulations, the calculation is accurate, and it can also interact with the mobile phone APP through 4G in real time for calculation, and send a report to the APP after the calculation.

[0005] For another example, Chinese Patent Application No. CN202080074341.7 provides a method for establishing a connection with an edge application server. A user equipment (UE) in a wireless communication network establishes a connection with an edge application server to offload data processing of an application program executed on the UE to the edge application server. The UE transmits a key performance indicator (KPI) associated with the application program to an edge data network. The KPI indicates the resources used by the application program to process data. In response, the UE receives edge application server parameters that meet or exceed the KPI from multiple servers in the edge data network. The parameters include computing, graphics computing, memory, and storage parameters with various specificity levels. The UE selects one application server from the edge application servers based on the parameters to process data instead of the application program.

[0006] For yet another example, Chinese Patent Publication No. CN114600437A discloses a method for establishing a connection with an edge application server. A user equipment (UE) in a wireless communication network establishes a connection with an edge application server to offload data processing of an application program executed on the UE to the edge application server. The UE transmits a key performance indicator (KPI) associated with the application program to an edge data network. The KPI indicates the resources used by the application program to process data. In response, the UE receives edge application server parameters that meet or exceed the KPI from multiple servers in the edge data network. The parameters include computing, graphics computing, memory, and storage parameters with various specificity levels. The UE selects one application server from the edge application servers based on the parameters to process data instead of the application program.

[0007] The above-mentioned patent technology configurations all have the problems that when the wireless network conditions are complex and the business requirements change greatly, the constructed wireless network has local characteristics, cannot be analyzed and optimized globally, and has weak computing power and cannot adapt to changes in network conditions and business requirements. Summary of the Invention

[0008] The object of the present invention is to provide an intelligent and simplified wireless network and its operation method.

[0009] The intelligent and simplified wireless network includes: a wireless communication edge computing platform and a wireless communication cloud computing platform that are interconnected;

[0010] The wireless communication edge computing platform includes: a first information module, an edge computing module, an execution module, and a first storage module. Among them, the first information module, the edge computing module, and the execution module are interconnected, and the first information module, the edge computing module, and the execution module are respectively connected to the first storage module. The first information module is used to obtain target data information and collect wireless network measurement data. The edge computing module is used to preprocess the collected wireless network measurement data. The first storage module is used to store the target data information obtained by the first information module and the collected wireless network measurement data, and store the preprocessed wireless network measurement data of the edge computing module. The execution module implements the construction plan of the wireless network to be constructed;

[0011] The wireless communication cloud computing platform includes: a second information module and a performance evaluation module respectively connected to a second storage module, and a network construction module connected to the second storage module. The second information module is used to receive the target data information sent by the wireless communication edge computing platform and its preprocessed wireless network measurement data. The second storage module is used to store the received target data information and preprocessed wireless network measurement data, as well as the mapping relationship between the preset data information and the preset slice paradigm, the mapping relationship between the preset network information and the preset construction plan of the hardware layer of the network slice, and the mapping relationship between the preset slice paradigm and the preset construction plan of the software layer of the network slice. The network construction module is used to determine the slice paradigm of the wireless network to be constructed according to the target data information and based on the preset slice paradigm, determine the construction plan of the network slice of the wireless network to be constructed according to the slice paradigm and the resources of the wireless network, and send a construction instruction of the wireless network to be constructed carrying the construction plan to the wireless communication edge computing platform. The performance evaluation module is used to determine the evaluation performance value of the network slice according to the preprocessed wireless network measurement data, calculate the ratio between the number of target network slices and the total number of network slices, and decide whether to perform the reconstruction operation of the wireless network according to the relationship between the ratio and the preset threshold.

[0012] Further, the wireless communication cloud computing platform is set at the network control center position in the area covered by the intelligent and simple wireless network.

[0013] Further, the wireless communication cloud computing platform runs on one or more electronic devices and is used to obtain the target data information sent by the wireless communication edge computing platform.

[0014] Further, the wireless communication edge computing platform has computing and storage capabilities, is set in a single base station and is exclusive to the single base station, or is set in multiple base stations and is shared by multiple base stations, and can also be set independently of the base station.

[0015] The present invention further provides an operation method for an intelligent and simple wireless network, including the following steps:

[0016] Step 1, according to the user request time T request and the maintenance network time T of the intelligent and simplified wireless network maintain set the value of the preset time T with the minimum time interval, obtain the overall network status information state of the intelligent and simplified wireless network all , and initialize the slice status NS of the wireless communication cloud computing platform init ;

[0017] Step 2, the wireless communication edge computing platform performs preliminary processing on the target status of the wireless communication edge computing platform and uploads it to the wireless communication cloud computing platform;

[0018] Step 3, the wireless communication cloud computing platform obtains the preprocessed target information state′ user , the initially determined mapping relationship NS between the intelligent and simplified wireless network and the slice paradigm edge and the wireless communication edge computing platform information state edge , perform network analysis and slice orchestration through the machine learning and intelligent and simplified wireless network flexible adaptation method, complete the mapping relationship NS of the complete slice paradigm with the target information and the network status cloud , determine the complete network constructor NF cloud (α);

[0019] Step 4, the wireless communication edge computing platform, according to the slice paradigm mapping relationship NS edge and the complete network constructor NF cloud (α), uses the execution module and the management network element to call the software layer and hardware layer resources, determines the software layer construction plan of the network slice of the wireless network to be constructed, and completes the construction of the intelligent and simplified wireless network;

[0020] Step 5, the intelligent and simplified wireless network collects the target device evaluation performance value and the overall evaluation performance value of the intelligent and simplified wireless network, and maintains or reconstructs the constructed intelligent and simplified wireless network.

[0021] Furthermore, in Step 1, according to the user request time T request and the maintenance network time T of the intelligent and simplified wireless network maintain with the minimum time interval T value, obtain the overall network status information state of the intelligent and simplified wireless network all , and initialize the slice status NS of the wireless communication cloud computing platform init , specifically including:

[0022] Step 1.1, obtain the status information stat of the wireless communication edge computing platform at time interval T eedge and the status information state of the wireless communication cloud computing platform cloud , state edgewith state cloud including the platform location, platform load, and in-platform resource information, and connecting the wireless communication cloud computing platform with the network topology, link information, and network connection status information state link to synthesize the overall network state information state of the intelligent and simple wireless network RAN ;

[0023] Step 1.2, at time interval T, the wireless communication edge computing platform obtains target user information stat euser , the target user information state user including: target user location, target access status, target service type, target intent information;

[0024] Step 1.3, introduce the flexible adaptation network environment method of the intelligent and simple wireless network and build the intelligent and simple wireless network slice paradigm library, and in the wireless communication edge computing platform, according to the obtained wireless communication cloud computing platform, wireless communication edge computing platform, and target user information state user initialize the slice paradigm NS of the wireless communication cloud computing platform and the wireless communication edge computing platform init .

[0025] Furthermore, in Step 1, the overall network state information state all includes access network information state RAN and target user information state user , and are respectively stored in the intelligent and simple wireless network cloud computing platform and the intelligent and simple wireless network edge computing platform.

[0026] Furthermore, in Step 2, the wireless communication edge computing platform performs preliminary processing on the target state of the wireless communication edge computing platform and uploads it to the wireless communication cloud computing platform, specifically including:

[0027] Step 2.1, the wireless communication edge computing platform preprocesses the obtained target information for the purpose of reducing information granularity and extracts target information features, and transforms the state user analysis into state′ user , and the analysis transformation process includes extracting and analyzing user intent information, user slice requirements, transforming into machine language, and tagging user features;

[0028] Step 2.2, the computing unit of the wireless communication edge computing platform uses machine learning methods to determine partial network construction parameters according to partial target information;

[0029] Step 2.3, the wireless communication edge computing platform preliminarily determines the mapping relationship NS between the slice and the intelligent and simple wireless network according to the preset or machine learning-obtained slice paradigm sample libraryedge ;

[0030] Step 2.4, the wireless network edge computing platform sends the preprocessed target information state′ user , the initially determined network and slice paradigm mapping relationship NS edge and the status information state of the wireless communication edge computing platform edge to the wireless network cloud computing platform.

[0031] Furthermore, in step 2.2, some network construction parameters NF edge (α) are determined, including radio frequency control variables, some link layer protocols, network layer cache resources, and computing resources.

[0032] Furthermore, in step 3, the wireless network cloud computing platform obtains the preprocessed target information state′ user , the initially determined network and slice paradigm mapping relationship NS edge and the wireless network edge computing platform information state edge , and performs network analysis and slice orchestration through the machine learning and intelligent and simplified network flexible adaptation method to complete the mapping relationship NS cloud of the complete slice paradigm with the target information and the network state, and determines the complete network constructor NF cloud (α), specifically including:

[0033] Step 3.1, the wireless communication cloud computing platform obtains the preprocessed target information and the initially determined network and slice paradigm mapping relationship transmitted by the wireless network edge computing platform, adjusts the expressions of the environment function and the reward function and the weight values of the services, and matches the mapping relationship with the slice paradigm;

[0034] Step 3.2, the wireless communication cloud computing platform calls the computing module and the storage module, and performs network analysis and slice orchestration through the machine learning and intelligent and simplified network flexible adaptation method to determine the slice paradigm mapping relationship NS RAN and the complete network constructor NF cloud (α), and distributes them to the wireless communication edge computing platform;

[0035] Step 3.3, the wireless communication cloud computing platform records the result of the current mapping relationship construction, and records it in the flexible adaptation slice paradigm unit in the storage module to provide paradigm support for the next intelligent and simplified wireless network construction process.

[0036] Furthermore, in step 3, the network constructor NF cloud (α) includes radio frequency control variables, link layer protocols, network layer cache resources, content caching, interface relationships, network topologies, computing resources, link relationships, and slice orchestration functions.

[0037] Further, in step 4, the wireless communication edge computing platform determines according to the slice paradigm mapping relationship NS RAN and the complete network constructor NF cloud (α), uses the execution module and the management network element to call the software layer and hardware layer resources of the intelligent and simplified wireless network, determines the software layer construction plan of the network slice of the intelligent and simplified wireless network to be constructed, and completes the construction of the intelligent and simplified wireless network, specifically including:

[0038] Step 4.1, determine the hardware layer construction plan of the network slice of the intelligent and simplified wireless network to be constructed according to the target information characteristics;

[0039] Step 4.2, based on the software layer and hardware layer resources of the intelligent and simplified wireless network, determine the scheduling plan of the hardware layer resources of each network slice of the intelligent and simplified wireless network. The scheduling plan includes time-frequency resource scheduling, cache resource scheduling, and power resource scheduling;

[0040] Step 4.3, according to the intelligent and simplified network slice paradigm, determine the software layer construction plan of the intelligent and simplified wireless network slice to be constructed, including the physical layer protocol, link layer protocol, network layer protocol, topology structure, and network element function orchestration of the connection state, and complete the construction of the intelligent and simplified wireless network.

[0041] Further, in step 4, if the corresponding mapping relationship is not found in the preset network slice mapping sample library, machine learning or other intelligent learning will be used to determine and record the new slice paradigm and slice mapping relationship.

[0042] Further, in step 5, the intelligent and simplified wireless network collects the target device evaluation performance value and the network overall evaluation performance value, and maintains or reconstructs the constructed intelligent and simplified wireless network, specifically including:

[0043] Step 5.1, the wireless communication edge computing platform obtains the evaluation performance value value of the target device performance , when there is an evaluation performance value value performance that does not meet the preset expected performance value, the wireless communication edge computing platform obtains the measurement data of the intelligent and simplified wireless network at intervals of time T, and performs data cleaning on the obtained measurement data. The wireless communication edge computing platform performs data dimensionality reduction processing or data standardization processing on the measurement data of the intelligent and simplified wireless network after cleaning, and sends the processed data to the wireless communication cloud computing platform. The wireless communication cloud computing platform uses the preset statistical analysis algorithm to perform statistical analysis on the measurement data of the intelligent and simplified wireless network after processing, and takes the analysis result as the evaluation performance value of the target device, and calculates the ratio between the number of network slices of the target device and the total number of network slices;

[0044] Step 5.2: Determine whether the calculated ratio is less than a preset threshold to decide whether to select to adjust the network structure, network virtual parameters, and topological structure or reconstruct the intelligent and simplified wireless network as needed. If the calculated ratio is less than the preset threshold, execute Step 3; if it is greater than the preset threshold, execute Step 1.

[0045] The intelligent and simplified wireless network and its operation method according to the present invention have the following beneficial effects:

[0046] 1. By applying the intelligent and simplified wireless network according to the present invention, the wireless communication edge computing platform is used to collect target data information and construct a wireless network according to the construction scheme of the network slice determined by the wireless communication cloud computing platform. The wireless communication cloud computing platform is used to determine the construction scheme of the network slice according to the target data information sent by the wireless communication edge computing platform. Compared with the prior art, through the cloud-edge adaptive collaboration of the wireless communication cloud computing platform and the wireless communication edge computing platform, the computing burden and storage pressure of the wireless communication cloud computing platform can be reduced, and the problem of data incompleteness of the wireless communication edge computing platform can be solved, so that a wireless network can be constructed efficiently and flexibly.

[0047] 2. For the intelligent and simplified wireless network according to the present invention, since the wireless communication cloud computing platform determines the construction scheme of the network slice of the wireless network to be constructed and sends a construction instruction of the wireless network to be constructed carrying the construction scheme to the wireless communication edge computing platform, so that the wireless communication edge computing platform constructs the wireless network according to the above construction instruction. Compared with the prior art, since it is not necessary for technicians to determine the construction scheme of the network slice according to different service requirements and construct the wireless network based on the determined construction scheme, the construction efficiency of the wireless network is improved.

[0048] 3. For the intelligent and simplified wireless network according to the present invention, since the slice paradigm of the wireless network to be constructed is determined according to the target data information, and the construction scheme of the network slice of the wireless network to be constructed is determined according to the determined slice paradigm and the resources of the wireless network. And since the target data information includes the technical scenario information of the target area, the service attribute information of the wireless network to be constructed, and the expected performance values of each target device, the above target data information more accurately reflects the service requirements of each target device. The slice paradigm determined according to the above target data information can better meet the service requirements of each target device, so that the wireless network constructed based on the determined slice paradigm can fully meet the service requirements of users.

[0049] 4. For the intelligent and simplified wireless network of the present invention, since the evaluation performance values of each network slice reflect the performance values during the actual operation of the constructed wireless network, and the expected performance values of each network slice can reflect the service requirements of users. If the evaluation performance value of a network slice does not meet the expected performance value, it means that the service requirements of users are not satisfied under the constructed wireless network. Therefore, when there is a network slice whose evaluation performance value does not meet the expected performance value, reconstructing the constructed wireless network can enable the finally constructed wireless network to fully meet the service requirements of each user.

[0050] 5. For the operation method of the intelligent and simplified wireless network of the present invention, when the quantity and types of service requirements are large, it can ensure the high-efficiency operation of the wireless network, adapt to the requirements of complex wireless network conditions and large service changes, and improve the situation that the wireless network constructed based on the existing technology lacks a certain degree of flexibility and cannot adapt to the changes of network conditions and service requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0052] Figure 1 It is a schematic diagram of the network structure of the intelligent and simplified wireless network of the present invention;

[0053] Figure 2 It is a schematic diagram of the structure of the wireless communication cloud computing platform in the intelligent and simplified wireless network of the present invention;

[0054] Figure 3 It is a schematic diagram of the structure of the wireless communication edge computing platform in the intelligent and simplified wireless network of the present invention;

[0055] Figure 4 It is a schematic diagram of the flow of the operation method of the intelligent and simplified wireless network of the present invention.

[0056] Figure 5 It is a schematic diagram of the construction scheme of the hardware layer of the network slice of the present invention.

[0057] Figure 6 It is a schematic diagram of the construction scheme of the software layer of the network slice of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] The following will clearly and completely introduce the detailed embodiments of the intelligent and simplified wireless network and its operation method of the present invention in conjunction with the Figures 1 to 6 description of the accompanying drawings.

[0059] Such as Figure 1 、 Figure 2As shown in the figure, the intelligent and simplified wireless network of the present invention includes: a wireless communication cloud computing platform 101 and a wireless communication edge computing platform 102. The wireless communication edge computing platform 102 has computing power and storage capacity. The wireless communication edge computing platform 102 can be set on a single base station or multiple base stations, or can be set on one electronic device or multiple electronic devices to obtain target data information and send the target data information to the wireless communication cloud computing platform 101, for obtaining construction instructions sent by the wireless communication cloud computing platform 101, and implementing the construction plan of the to-be-constructed intelligent and simplified wireless network based on the construction instructions and for preprocessing the acquired measurement data of the intelligent and simplified wireless network, and sending the preprocessed data to the wireless communication cloud computing platform 101;

[0060] As Figure 2 shown in the figure, the wireless communication edge computing platform 102 is composed of a first information module, an edge computing module, and an execution module that are interconnected. The first information module, the edge computing module, and the execution module are respectively connected to the first storage module. The first information module is used to obtain target data information and collect wireless network measurement data. The edge computing module is used to preprocess the collected wireless network measurement data. The first storage module is used to store the target data information obtained by the first information module and the collected wireless network measurement data, and store the preprocessed measurement data of the intelligent and simplified wireless network by the edge computing module. The execution module implements the construction plan of the to-be-constructed intelligent and simplified wireless network based on the obtained construction instructions;

[0061] As Figure 3As shown in the figure, the wireless communication cloud computing platform 101 includes: a second information module, a network construction module, a performance evaluation module, and a second storage module. Among them, the second information module and the network construction module are respectively connected to the second storage module, and the performance evaluation module is connected to the second storage module. The second information module is used to receive the target data information sent by the wireless communication edge computing platform 102 and the intelligent and simplified wireless network measurement data after its preprocessing; the second storage module is used to store the received target data information and the preprocessed wireless network measurement data, as well as the mapping relationship between the preset data information and the preset slice paradigm, the mapping relationship between the preset network information and the preset construction plan of the hardware layer of the network slice, and the mapping relationship between the preset slice paradigm and the preset construction plan of the software layer of the network slice; the network construction module is used to determine the slice paradigm of the intelligent and simplified wireless network to be constructed according to the target data information and based on the preset slice paradigm, determine the construction plan of the network slice of the intelligent and simplified wireless network to be constructed according to the slice paradigm and the resources of the wireless network, and send a construction instruction of the intelligent and simplified wireless network to be constructed carrying the to-be-constructed plan to the wireless communication edge computing platform 102; the performance evaluation module is used to determine the evaluation performance value of the network slice according to the preprocessed intelligent and simplified wireless network measurement data, calculate the ratio between the number of target network slices and the total number of network slices, and decide whether to perform the reconstruction operation of the intelligent and simplified wireless network according to the relationship between the ratio and the preset threshold.

[0062] Further, the wireless communication cloud computing platform 101 is set at the network control center position in the area covered by the construction of the intelligent and simplified wireless network.

[0063] Further, the wireless communication cloud computing platform 101 runs on multiple electronic devices or one electronic device and is used to obtain the target data information sent by the wireless communication edge computing platform 102.

[0064] Further, the wireless communication edge computing platform 102 has computing and storage capabilities, is set in a single base station and is exclusive to the single base station, or is set in multiple base stations and is shared by multiple base stations, and can also be set independently of the base station.

[0065] As Figure 4 shown, the present invention further provides an operation method for an intelligent and simplified wireless network, including the following steps:

[0066] Step 1, according to the user request time T request and the minimum time interval T value in the network time T maintained by the intelligent and simplified wireless network, obtain the overall network state information state maintain of the intelligent and simplified wireless network, and initialize the slice state NS all of the wireless communication platform, where the overall network state information state init allContaining access network information state RAN And target user information state user And store them in the intelligent and simplified wireless network cloud computing platform and the intelligent and simplified wireless network edge computing platform respectively. Specifically, it includes:

[0067] Step 1.1, according to the time interval T, obtain the state information of the wireless communication edge computing platform edge And the state information of the wireless communication cloud computing platform cloud , state edge And state cloud Containing platform location, platform load, and platform internal resource information, and in the wireless communication cloud computing platform and the network connection status information state including network topology structure and link information link Are synthesized into the overall network state information of the intelligent and simplified wireless network RAN ;

[0068] Step 1.2, according to the time interval T, the wireless communication edge computing platform obtains the target user information user , the target user information user Includes target user location, target access status, target service type, and target intention information;

[0069] Step 1.3, introduce the flexible adaptation network environment method of the intelligent and simplified network and build an intelligent and simplified network slice paradigm library, and in the wireless communication edge computing platform, initialize the slice paradigm NS of the wireless communication cloud computing platform and the wireless communication edge computing platform according to the obtained wireless communication cloud computing platform, wireless communication network edge platform, and target information init ;

[0070] Step 2, the wireless communication edge computing platform performs preliminary processing on the edge computing platform and the target state, and uploads it to the intelligent and simplified wireless communication cloud computing platform. Specifically, it includes:

[0071] Step 2.1, the wireless communication edge computing platform preprocesses the obtained target information for the purpose of reducing information granularity, extracts the target information features, and converts state user Analysis and conversion into state' user , and the analysis and conversion process includes extracting and analyzing user intention information, user slice requirements, converting into machine language, and tagging user features;

[0072] Step 2.2, the computing unit of the wireless communication edge computing platform uses machine learning methods to determine partial network construction parameters according to partial target information;

[0073] Step 2.3, the wireless communication edge computing platform preliminarily determines the mapping relationship NS between the slice and the wireless network according to the preset or slice paradigm sample library obtained after machine learning edge ;

[0074] Step 2.4, the wireless network edge computing platform sends the preprocessed target information state′ user , the preliminarily determined mapping relationship NS between the network and the slice paradigm edge and the wireless network edge computing platform information state edge to the wireless network cloud computing platform;

[0075] Step 3, the wireless network cloud computing platform obtains the preprocessed target information state′ user , the preliminarily determined mapping relationship NS between the network and the slice paradigm edge and the wireless network edge computing platform information state edge , performs network analysis and slice orchestration through the machine learning and intelligent and simple network flexible adaptation method, completes the mapping relationship ND between the complete slice paradigm and the target information and the network state cloud , determines the complete network constructor ND cloud (α), including network functions such as radio frequency control variables, link layer protocols, network layer cache resources, content cache, interface relationships, network topologies, computing resources, link relationships, slice orchestration functions, etc.;

[0076] Step 3.1, the wireless network cloud computing platform obtains the preprocessed target information and the preliminarily determined mapping relationship between the network and the slice paradigm transmitted by the wireless network edge computing platform, adjusts the expressions of the environment function and the reward function and the weight values of the services, and matches the mapping relationship with the slice paradigm;

[0077] Step 3.2, the wireless network cloud computing platform calls the computing module and the storage module, performs network analysis and slice orchestration through the machine learning and intelligent and simple network flexible adaptation method, determines the slice paradigm mapping relationship NS RAN and the complete network constructor NF cloud (α), and distributes them to the wireless network edge computing platform;

[0078] Step 3.3, the wireless network cloud computing platform records the result of the current mapping relationship construction, records it in the flexible adaptation slice paradigm unit in the storage module, and provides paradigm support for the next network construction process;

[0079] Step 4, the wireless network edge computing platform according to the slice paradigm mapping relationship NS RAN and the complete network constructor NF cloud(α), using an execution module and a management network element, call the resources of the software layer and the hardware layer to determine the software layer construction plan for the network slice of the wireless network to be constructed, and complete the construction of the intelligent and simplified wireless network, specifically including:

[0080] Step 4.1, determine the hardware layer construction plan for the network slice of the wireless network to be constructed according to the target information characteristics;

[0081] Step 4.2, based on the wireless resources, determine the wireless resource scheduling plan for each network slice, including time-frequency resource scheduling, cache resource scheduling, and power resource scheduling;

[0082] Step 4.3, according to the slice paradigm, determine the software layer construction plan for the wireless network slice to be constructed, including physical layer protocol, link layer protocol, network layer protocol, topology structure, and network element function orchestration of the connection state;

[0083] Step 5, the wireless network collects the target device evaluation performance value and the network overall evaluation performance value, and maintains or reconstructs the constructed intelligent and simplified wireless network, specifically including:

[0084] Step 5.1, the wireless network edge computing platform obtains the evaluation performance value value of the target device performance , when there is an evaluation performance value that does not meet the preset expected performance value, and calculate the ratio between the number of network slices of the target device and the total number of network slices;

[0085] Step 5.2, determine whether the calculated ratio is less than the preset threshold, select whether to adjust the network structure, network virtual parameters, and topology structure or reconstruct the intelligent and simplified network as needed. If it is greater than the preset threshold, execute Step 1 to reconstruct the intelligent and simplified wireless network.

[0086] Furthermore, in Step 2.2, determine some network construction parameters NF edge (α), including radio frequency control variables, some link layer protocols, network layer cache resources, and computing resources.

[0087] Furthermore, in Step 4, if the corresponding mapping relationship is not found in the preset slice mapping sample library, use machine learning or other intelligent learning to determine and record the new slice paradigm and slice mapping relationship.

[0088] Furthermore, in Step 4.1, the hardware layer construction plan for the network slice of the wireless network to be constructed is as Figure 5 shown, Figure 5 each rectangular box in represents a hardware layer construction plan for a network slice. Specifically, the above-mentioned hardware layer construction plan for the network slice also includes the base station attributes and the topology structure of the network slice.

[0089] The first rectangular box from left to right on the upper side is: the hardware layer construction scheme of network slices matching the dense high-capacity slice paradigm. The configured infrastructure resources of the wireless network to be constructed in the rectangular box are: BBU pool, RRU. The rectangular box here adopts the tree topology of the BBU+RRU networking mode, deploys multiple RRUs close to the terminal, and accesses the BBU pool in the cloud through the fronthaul link.

[0090] The second rectangular box from left to right on the upper side is: the hardware layer construction scheme of the network slice that matches the wide coverage slice paradigm. The configured infrastructure resources of the wireless network to be constructed in the rectangular box are: HPN. The rectangular box adopts a mesh topology structure composed of deploying HPN for global coverage, the transmission power and switching parameters of HPN are determined by the optimal coverage distance, and adjacent HPNs are connected through X2 / S1 interfaces.

[0091] The third rectangular box from left to right on the upper side is: the hardware layer construction scheme of network slices that matches the large connection delay tolerance low power consumption slice paradigm. The configured infrastructure resources of the wireless network to be constructed in the rectangular box are: small base stations. The rectangular box adopts the tree topology structure of clustering terminals in the wireless network area into self-organizing networks, selecting nodes with strong processing capabilities as cluster heads, and the terminals in the cluster aggregate information to the cluster head directly or through multi-hop relays, and then access the small base station after aggregation processing by the cluster head, and each terminal forms a tree topology structure.

[0092] The first rectangular box from left to right on the lower side is: the hardware layer construction plan of the network slice that matches the large-capacity, low-latency, and high-reliability slicing paradigm. The configured infrastructure resources of the wireless network to be constructed in the rectangular box are: CU, DU, and RRU. The rectangular box here adopts a tree topology structure composed of a CU+DU+RRU network hierarchical structure for functional deployment, multiple RRUs accessing the DU through fronthaul links, and multiple distributed DUs accessing the centralized CU through midhaul links. Among them, the CU aggregates the low-real-time functions in the protocol stack L3 and L2, the DU concentrates the high-real-time functions in the protocol stack L2 and some L1 layer functions, and the RRU is deployed close to the terminal.

[0093] The second rectangular box from left to right on the lower side is: the hardware layer construction scheme of the network slice that matches the large connection low latency high reliability slice paradigm. The configured infrastructure resources of the wireless network to be constructed in the rectangular box are: CU, DU, RRU. The rectangular box adopts the network structure of CU+DU+RRU+terminal self-organizing network, in which the terminal performs clustered self-organizing network, and the terminal in the cluster uses direct or relay multi-hop methods to aggregate information to the cluster head, and accesses the RRU through the cluster head to increase the number of terminals accessing the network. Multiple RRUs access the DU through the fronthaul link, and multiple distributed DUs access the centralized CU through the midhaul link, forming a tree topology. The CU aggregates the low real-time functions of the protocol stack, and the DU concentrates the high real-time functions of the protocol stack for local processing. The RRU is deployed close to the terminal.

[0094] Further, in step 4.3, the software layer construction scheme of the wireless network slice to be constructed is determined as follows: Figure 6 As shown, in Figure 6 In the figure, NSIA, NSIB, and NSIC represent network slices, NSSI1, NSSI2, and NSSI3 represent core network slices, and NSSI4 and NSSI5 represent radio access network slices. Figure 6 It is known that the NSIA network slice consists of the NSSI1 core network slice and the NSSI4 wireless access network slice, the NSIB network slice consists of the NSSI2 core network slice and the NSSI5 wireless access network slice, and the NSIC network slice consists of the NSSI3 core network slice and the NSSI5 wireless access network slice.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of each claim in the claims of the present invention.

Claims

1. A simple and intelligent wireless network, characterized in that, it includes: The simple and intelligent wireless network includes: a wireless communication edge computing platform and a wireless communication cloud computing platform: The wireless communication edge computing platform includes: a first information module, an edge computing module, an execution module, and a first storage module. The first information module, the edge computing module, and the execution module are interconnected, and the first information module, the edge computing module, and the execution module are respectively connected to the first storage module. The first information module is used to obtain target data information and collect wireless network measurement data. The edge computing module is used to preprocess the collected wireless network measurement data. The first storage module is used to store the target data information obtained by the first information module and the collected wireless network measurement data, and store the preprocessed wireless network measurement data by the edge computing module. The execution module implements the construction plan of the wireless network to be constructed; The wireless communication cloud computing platform includes: a second information module and a performance evaluation module that are respectively connected to the second storage module, and a network construction module that is connected to the second storage module. The second information module is used to receive the target data information and the preprocessed wireless network measurement data sent by the wireless communication edge computing platform. The second storage module is used to store the received target data information and the preprocessed wireless network measurement data, as well as the mapping relationship between the preset data information and the preset slice paradigm, the mapping relationship between the preset network information and the preset construction plan of the hardware layer of the network slice, and the mapping relationship between the preset slice paradigm and the preset construction plan of the software layer of the network slice. The network construction module is used to determine the slice paradigm of the wireless network to be constructed according to the target data information and based on the preset slice paradigm, determine the construction plan of the network slice of the wireless network to be constructed according to the slice paradigm and the resources of the wireless network, and send a construction instruction of the wireless network to be constructed carrying the construction plan to the wireless communication edge computing platform. The performance evaluation module is used to determine the evaluation performance value of the network slice according to the preprocessed wireless network measurement data, calculate the ratio between the number of target network slices and the total number of network slices, and decide whether to perform the reconstruction operation of the wireless network according to the relationship between the ratio and the preset threshold.

2. The simple and intelligent wireless network according to claim 1, characterized in that, The wireless communication cloud computing platform runs on one or more electronic devices and is used to obtain the target data information sent by the wireless communication edge computing platform.

3. The simple and intelligent wireless network according to claim 1, characterized in that, The wireless communication edge computing platform has computing and storage capabilities, is set in a single base station and is exclusive to the single base station, or is set in multiple base stations and is shared by multiple base stations, and can also be set independently of the base station.

4. A running method of a simple and intelligent wireless network, characterized in that, it includes the following steps: Step 1, according to the user request time T request and the minimum time interval T value in the maintenance network time T of the intelligent and simplified wireless network, obtain the overall network status information state of the intelligent and simplified wireless network maintain , and initialize the slice status NS of the wireless communication cloud computing platform all ; init ; Step 2, the wireless communication edge computing platform performs preliminary processing on the target state of the wireless communication edge computing platform and uploads it to the wireless communication cloud computing platform; Step 3, the wireless communication cloud computing platform obtains the preprocessed target information state′ user and the initially determined mapping relationship NS between the intelligent and simplified wireless network and the slice paradigm edge as well as the wireless communication edge computing platform information state edge , and through the machine learning and intelligent and simplified wireless network flexible adaptation method, network analysis and slice orchestration are carried out to complete the mapping relationship NS between the complete slice paradigm, target information and the network platform cloud , and determine the complete network constructor NF cloud (α): Step 3.1, the wireless communication cloud computing platform obtains the preprocessed target information and the initially determined mapping relationship between the network and the slice paradigm transmitted by the wireless network edge computing platform, adjusts the expressions of the environment function and the reward function and the weight values of the services, and matches the mapping relationship with the slice paradigm; Step 3.2, the wireless communication cloud computing platform calls the computing module and the storage module, performs network analysis and slice orchestration through the machine learning and intelligent and simple network flexible adaptation method, and determines the slice paradigm mapping relationship NS cloud and the complete network constructor NF cloud (α), and sends it to the wireless communication edge computing platform; Step 3.3, the wireless communication cloud computing platform records the result of this mapping relationship construction and stores it in the flexible adaptation slice paradigm unit in the storage module, providing paradigm support for the next intelligent and simplified wireless network construction process; Step 4, the wireless communication edge computing platform determines the software layer construction plan of the network slice of the wireless network to be constructed according to the slice paradigm mapping relationship NS cloud and the complete network constructor NF cloud (α), uses the execution module and the management network element to call the resources of the software layer and the hardware layer, determines the software layer construction plan of the network slice of the wireless network to be constructed, and completes the construction of the intelligent and simple wireless network; Step 5, the intelligent and simplified wireless network collects the target device evaluation performance value and the overall evaluation performance value of the intelligent and simplified wireless network, and maintains or reconstructs the constructed intelligent and simplified wireless network.

5. The operation method of the intelligent and simplified wireless network according to claim 4, characterized in that In step 1, according to the user request time T request and the minimum time interval T value in the maintenance network time T of the intelligent and simplified wireless network maintain obtain the overall network status information state of the intelligent and simplified wireless network all and initialize the slice status NS of the wireless communication cloud computing platform init , including: Step 1.1, obtain the status information state of the wireless communication edge computing platform at time intervals T edge and the status information state of the wireless communication cloud computing platform cloud , state edge and state cloud include the platform location, platform load, and resource information within the platform, and combine the wireless communication cloud computing platform with the network connection status information statel including the network topology structure and link information ink into the overall network status information state of the intelligent and simplified wireless network all ; Step 1.2, according to the time interval T, the wireless communication edge computing platform obtains the target user information state user , the target user information state user includes: target user location, target access status, target service type, target intention information; Step 1.3, introduce the flexible adaptation network environment method of the intelligent and simple wireless network, build the intelligent and simple wireless network slice paradigm library, and in the wireless communication edge computing platform, according to the obtained state information of the wireless communication cloud computing platform, the state information of the wireless communication edge computing platform, and the target user information state user Initialize the slice state NS of the wireless communication cloud computing platform and the wireless communication edge computing platform init .

6. The operation method of the intelligent and simplified wireless network according to claim 4, characterized in that In step 2, the wireless communication edge computing platform performs preliminary processing on the target state of the wireless communication edge computing platform and uploads it to the wireless communication cloud computing platform, specifically including: Step 2.1, the wireless communication edge computing platform preprocesses the acquired target information and extracts the target information features for the purpose of reducing the information granularity, and converts state user analysis into state′ user , and the analysis and conversion process includes extracting and analyzing user intention information, user slice requirements, converting them into machine language, and tagging user characteristics; Step 2.2, the computing unit of the wireless communication edge computing platform uses a machine learning method to determine partial network construction parameters according to the target partial information; Step 2.3, the wireless communication edge computing platform preliminarily determines the mapping relationship NS between the slice and the intelligent and simplified wireless network according to the slice paradigm sample library obtained by preset or machine learning edge ; Step 2.4, the wireless network edge computing platform sends the preprocessed target information state′ user , the preliminarily determined mapping relationship NS between the network and the slice paradigm edge and the status information state of the wireless communication edge computing platform edge to the wireless network cloud computing platform.

7. The operation method of the intelligent and simplified wireless network according to claim 4, characterized in that In step 4, the wireless communication edge computing platform determines the software layer construction plan of the network slice of the intelligent and simplified wireless network by using the slice paradigm mapping relationship NS cloud and the complete network constructor NF cloud (α), and uses the execution module and the management network element to call the software layer and hardware layer resources of the intelligent and simplified wireless network to determine the software layer construction plan of the network slice of the intelligent and simplified wireless network to be constructed, so as to complete the construction of the intelligent and simplified wireless network, specifically including: Step 4.1, according to the target information characteristics, determine the hardware layer construction plan of the network slice of the to-be-constructed intelligent and simplified wireless network; Step 4.2, based on the software layer and hardware layer resources of the intelligent and simplified wireless network, determine the scheduling plan for the hardware layer resources of each network slice of the intelligent and simplified wireless network. The scheduling plan includes time-frequency resource scheduling, cache resource scheduling, and power resource scheduling; Step 4.3, according to the intelligent network slice paradigm, determine the software layer construction plan of the to-be-constructed intelligent and simplified wireless network slice, including physical layer protocol, link layer protocol, network layer protocol, topology structure, and network element function orchestration of connection status, and complete the construction of the intelligent and simplified wireless network.

8. The operation method of the intelligent and simplified wireless network according to claim 4, characterized in that In step 4, if the corresponding mapping relationship is not found in the preset network slice mapping sample library, machine learning will be used to determine and record the new slice paradigm and slice mapping relationship.

9. The operation method of the intelligent and simplified wireless network according to claim 4, characterized in that In step 5, the intelligent and simplified wireless network collects the target device evaluation performance value and the overall network evaluation performance value, and maintains or reconstructs the constructed intelligent and simplified wireless network, specifically including: Step 5.1, the wireless communication edge computing platform obtains the evaluation performance value value of the target device performance , when there is an evaluation performance value value performance that does not meet the preset expected performance value, the wireless communication edge computing platform obtains the measurement data of the intelligent and simplified wireless network at intervals of time T, and performs data cleaning on the obtained measurement data. The wireless communication edge computing platform performs data dimensionality reduction processing or data standardization processing on the measurement data of the intelligent and simplified wireless network after cleaning, and sends the processed data to the wireless communication cloud computing platform. The wireless communication cloud computing platform uses a preset statistical analysis algorithm to perform statistical analysis on the measurement data of the intelligent and simplified wireless network after processing, and uses the analysis result as the evaluation performance value of the target device, and calculates the ratio between the number of network slices of the target device and the total number of network slices; Step 5.2, determine whether the calculated ratio is less than the preset threshold to decide whether to select to adjust the network structure, network virtual parameters, and topology structure or reconstruct the intelligent and simplified wireless network as needed. If the calculated ratio is less than the preset threshold, execute step 3; if it is greater than the preset threshold, execute step 1.

Images