5G network resource dynamic adjustment method

By analyzing the time series requirements of 5G network resources, establishing a prediction model, separating the control layer and forwarding layer, and generating dynamic adjustment strategies, the problem of resource scheduling complexity in base station co-construction and sharing and Massive MIMO technology is solved, and the resource scheduling complexity of 5G networks is realized. Dynamic optimization and efficiency improvement of network resources.

CN118524546BActive Publication Date: 2024-11-22BEIJING HONGSHU TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410755329.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-12
Publication Date
2024-11-22
Estimated Expiration
2044-06-12

AI Technical Summary

Technical Problem

5G networks face complex resource scheduling and interference characteristics in base station co-construction and sharing and the application of Massive MIMO technology, resulting in increased network optimization difficulty, and it is difficult for existing technology to realize dynamic resource adjustment to optimize network performance.

Method used

By analyzing the time series demand trends of spectrum and base station resources, establishing a network resource demand prediction model, separating the control layer and forwarding layer, using the prediction model to generate dynamic adjustment strategies, and optimizing the allocation and use of network resources.

Benefits of technology

It has achieved dynamic optimization of 5G network resource scheduling, improved network efficiency, more in line with practical application scenarios, and improved network resource utilization efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118524546B_ABST
    Figure CN118524546B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for dynamically adjusting 5G network resources, and relates to the technical field of 5G network resource adjustment. The method comprises: taking the allocated spectrum resources and existing base stations as the main body of network resources, analyzing the demand change trend of the main body of network resources in time series, and establishing a network resource demand prediction model; separating the control layer from the forwarding layer, and the control layer uses the network resource demand prediction model to obtain the prediction result of the network resource demand; the control layer generates and executes the network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction result of the network resource demand; the forwarding layer forwards data based on the adjusted network resources. The beneficial effects achieved by the present invention are as follows: optimizing the scheduling process of 5G network resources, adding dynamic scheduling means, being more in line with the actual application scenario, and improving network efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of 5G network resource adjustment, and in particular to a method for dynamically adjusting 5G network resources. Background Art

[0002] Dynamic adjustment of 5G network resources refers to the real-time adjustment of the allocation and use of various resources in the network (such as spectrum, power, base stations, etc.) according to the actual operation of the network, so as to optimize network performance, improve network efficiency and meet user needs. At the same time, the optimization of 5G networks also faces some challenges, such as the complexity brought by the co-construction and sharing of base stations. This sharing includes two modes: independent carrier and shared carrier, each of which has its advantages and disadvantages. In the independent carrier mode, each operator manages its own network resources, while the shared carrier mode requires more complex resource allocation and scheduling strategies. In addition, the application of 5G MassiveMIMO (massive multi-antenna) technology also increases the difficulty of network optimization, because this technology brings more complex interference characteristics and overlapping coverage problems. Summary of the invention

[0003] The present invention provides a method for dynamically adjusting 5G network resources, comprising:

[0004] Step 1: Take the allocated spectrum resources and existing base stations as the main network resources, analyze the demand change trend of the main network resources in time series, and establish a network resource demand prediction model;

[0005] Step 2: Separate the control layer from the forwarding layer. The control layer uses the network resource demand prediction model to obtain the prediction results of network resource demand.

[0006] Step 3: The control layer generates and executes a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction results of network resource demand;

[0007] Step 4: The forwarding layer forwards data based on the adjusted network resources.

[0008] A 5G network resource dynamic adjustment method as described above, wherein the demand change trend of the network resource subject in the time series is analyzed and a network resource demand prediction model is established, specifically including the following sub-steps:

[0009] The spectrum usage and base station load conditions in different time periods are collected and organized into data set P1, and the resource demand impact characteristics in the traffic data are extracted and organized into data set P2;

[0010] Take dataset P2 as input set and dataset P1 as output set to train and build network resource demand prediction model;

[0011] Optimize the parameters of the network resource demand prediction model to ensure the accuracy of the model.

[0012] In the above-mentioned method for dynamically adjusting 5G network resources, the control layer uses a network resource demand prediction model to obtain a prediction result of the network resource demand, which is specifically divided into the following sub-steps:

[0013] Set up monitoring points at the base station to obtain real-time traffic data generated by users within the operating range and transmit it to the control layer in the form of message queues;

[0014] The control layer extracts resource demand characteristics from traffic data;

[0015] Input resource demand characteristics into the network resource demand prediction model and output the prediction results of network resource demand.

[0016] A 5G network resource dynamic adjustment method as described above, wherein a network resource dynamic adjustment strategy is generated based on a network resource dynamic adjustment function and a prediction result of network resource demand, and is specifically divided into the following sub-steps:

[0017] Taking spectrum and base station as resource adjustment items, a network resource dynamic adjustment function is constructed;

[0018] The prediction result of network resource demand is passed as an input parameter to the network resource dynamic adjustment function, and the network resource dynamic adjustment strategy is returned.

[0019] The present invention also provides a 5G network resource dynamic adjustment system, including: a monitoring layer, a control layer and a forwarding layer;

[0020] The monitoring layer includes monitoring points deployed on all base stations within the operating range;

[0021] The control layer includes a network resource demand prediction module and a network resource dynamic adjustment module;

[0022] The forwarding layer forwards data based on the network resources adjusted by the control layer.

[0023] In a 5G network resource dynamic adjustment system as described above, the monitoring layer obtains the traffic data generated by users in real time, transmits it to the control layer in the form of a message queue, and monitors the forwarding efficiency of network resources. If it is lower than a preset threshold, dynamic adjustment is initiated to the control layer.

[0024] A 5G network resource dynamic adjustment system as described above, wherein the network resource demand prediction module is used to analyze the demand change trend of spectrum resource and base station resource usage in a time series, establish a network resource demand prediction model based on the analysis results, and use the network resource demand prediction model to obtain the prediction result of network resource demand.

[0025] A 5G network resource dynamic adjustment system as described above, wherein the network resource dynamic adjustment module is used to generate and execute a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the predicted results of the network resource demand.

[0026] The beneficial effects achieved by the present invention are as follows: optimizing the scheduling process of 5G network resources, adding dynamic scheduling means, being more in line with actual application scenarios, and improving network efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0028] Figure 1 This is a flow chart of a method for dynamically adjusting 5G network resources provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0029] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0030] Embodiment 1

[0031] like Figure 1 As shown, Embodiment 1 of the present invention provides a method for dynamically adjusting 5G network resources, including:

[0032] Step S10: using the allocated spectrum resources and existing base stations as the main network resources, analyzing the demand change trend of the main network resources in time series, and establishing a network resource demand prediction model;

[0033] Spectrum resources and base stations are the main body of 5G network resources. In different time periods, the demand for network resources will change due to factors such as the number of users and data traffic. Based on the demand change trend of network resources in time series, network resources are dynamically adjusted and scheduled to improve network efficiency. The demand change trend of network resources in time series is analyzed and a network resource demand prediction model is established, which specifically includes the following sub-steps:

[0034] Step S11: Collect spectrum usage and base station load conditions in different time periods and organize them into a data set P1, and extract resource demand impact features in traffic data and organize them into a data set P2;

[0035] The data of spectrum usage and base station load are organized into a data set P1, which is expressed as: P1 = {(a1, b1), (a2, b2), (a3, b3), …, (a n ,b n )},a1~a n are the spectrum usage in different time periods, b1~b n are the base station loads in different time periods, and n is the number of acquisitions;

[0036] The factors affecting resource demand in the traffic data are extracted, including traffic patterns, user distribution, special events (such as holidays, large-scale events), etc., and the resource demand influencing features are formed after normalization. After serialization, they are organized into a data set P2, which is expressed as: P2 = {R1, R2, R3, ..., R n}, R1~R n are the resource demand impact feature sets in different time periods, n is the number of collections, R i ={r1,r2,r3,…,r m}, r1~r m are different resource impact characteristics respectively, and m is the number of resource demand impact characteristic items.

[0037] Step S12: using the data set P2 as the input set and the data set P1 as the output set, training and constructing a network resource demand prediction model;

[0038] The network resource demand prediction model includes two parts: spectrum demand prediction model and base station load prediction model. is the spectrum demand prediction model, and the output result is recorded as y1. is the base station load prediction model, and the output result is recorded as y2, ε1, ε2,…,ε z is the coefficient of the autoregressive term, is the autoregressive term, z is the order of the autoregressive term, d is the number of differences, x j is the resource demand impact feature of the jth input, j ranges from 1 to m, m is the number of resource demand impact feature items, σ1,σ2,…,σ q is the coefficient of the moving average term, γ j The error term of the j-th resource demand influencing term, q is the order of the moving average term, γ1,γ2,…,γ q is the error term of the moving average term.

[0039] Step S13: Optimizing the parameters of the network resource demand prediction model to ensure the accuracy of the model;

[0040] The training data set is divided into k subsets of equal size. In each iteration, one subset is selected as the validation set and the rest as the training set. This process is repeated k times, and a different subset is selected as the validation set each time. The optimal model parameters are determined based on the loss function.

[0041] The loss function is expressed as: where y1 k is the spectrum demand predicted for the kth time, Δy1 k is the actual spectrum usage for the kth time, y2 k is the k-th predicted base station load, Δy2 k is the actual base station load for the kth time, k ranges from 1 to L, and L is the total number of training times.

[0042] Step S20: separating the control layer from the forwarding layer, and the control layer uses the network resource demand prediction model to obtain a prediction result of the network resource demand;

[0043] To achieve dynamic adjustment of network resources, it is necessary to control the allocation of network resources in a programmatic way, separate the control plane of the network from the data plane, and let the control layer complete complex decision-making work, while the forwarding layer is only responsible for simple forwarding functions. The control layer uses the network resource demand prediction model to obtain the prediction results of network resource demand, which is divided into the following sub-steps:

[0044] Step S21: Setting a monitoring point at the base station to obtain the traffic data generated by users within the operating range in real time and transmitting it to the control layer in the form of a message queue;

[0045] Step S22: The control layer extracts resource demand characteristics from the traffic data;

[0046] The flow data is transmitted to the feature extraction step using stream processing. Because the amount of flow data is huge, extracting it after it is transmitted will reduce the computing efficiency. Therefore, the stream processing method is adopted. The control layer extracts the resource demand characteristics in the data while receiving the data.

[0047] Step S23: input resource demand characteristics into the network resource demand prediction model, and output the prediction result of network resource demand;

[0048] Step S30: The control layer generates and executes a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction result of the network resource demand;

[0049] Generate a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction result of network resource demand, which is specifically divided into the following sub-steps:

[0050] Step S31: Taking spectrum and base station as resource adjustment items, constructing a network resource dynamic adjustment function;

[0051] The network resource dynamic adjustment function includes two functions: spectrum frequency adjustment function and base station load adjustment function. The spectrum adjustment function is expressed as L1(y1). The calculation process is: If it is true, it means that the current spectrum frequency is too large, then execute τ φ =max(τ1,τ2,…,τ Φ )-t to reduce the spectrum frequency, if If it is true, it means that the current spectrum frequency is too small, then execute τ φ =min(τ1,τ2,…,τ Φ )+t to increase the spectrum frequency, if If it is true, it means that the current spectrum frequency just meets the predicted demand, then the frequency adjustment of each segment of the spectrum τ1, τ2, ..., τ Φ , where y1 is the predicted spectrum demand, c φ is the current frequency of the φth segment of the spectrum, τ φ is the adjustment amount of the φth segment spectrum frequency, l1 φ is the working efficiency of the φth spectrum segment, φ ranges from 1 to Φ, Φ is the number of spectrum segments, τ1, τ2, …, τ Φ is the adjustment amount of each spectrum, and t is the adjustment step.

[0052] The base station load adjustment function is expressed as L2(y2), and the calculation process is: This indicates that the current base station load is too high, so execute o =max(υ1,υ1,…,υ Ψ )-t to reduce the base station load, if This indicates that the current base station load is too low, so execute o =min(υ1,υ1,…,υ Ψ )+t to increase the base station load. This means that the current base station load just meets the predicted demand of the base station, and the load adjustment of each base station is returned υ1,υ1,…,υ Ψ , where y2 is the predicted base station load, h o is the current load of the oth base station, υ o is the adjustment amount of o base stations, l2 o is the base station transmission distance, o takes values ​​from 1 to Ψ, Ψ is the total number of base stations, and t is the adjustment step size.

[0053] Step S32: passing the prediction result of the network resource demand as an input parameter into the network resource dynamic adjustment function, and returning the network resource dynamic adjustment strategy;

[0054] Step S40: the forwarding layer forwards data based on the adjusted network resources;

[0055] After the control layer executes the dynamic adjustment strategy of network resources, the forwarding layer will forward data again, and the data monitoring point of each base station is responsible for real-time monitoring of the efficiency of data forwarding. If it is lower than a certain threshold, it will initiate dynamic adjustment to the control layer again to improve the utilization efficiency of network resources.

[0056] Embodiment 2

[0057] Embodiment 2 of the present invention provides a 5G network resource dynamic adjustment system, including: a monitoring layer, a control layer and a forwarding layer;

[0058] (1) The monitoring layer includes monitoring points deployed on all base stations within the operating range. It is used to obtain traffic data generated by users in real time, transmit it to the control layer in the form of message queues, and monitor the forwarding efficiency of network resources. If it is lower than the preset threshold, it will initiate dynamic adjustments to the control layer.

[0059] (2) The control layer includes a network resource demand prediction module and a network resource dynamic adjustment module;

[0060] 1. The network resource demand prediction module is used to analyze the demand change trend of spectrum resource and base station resource usage in time series, establish a network resource demand prediction model based on the analysis results, and use the network resource demand prediction model to obtain the prediction results of network resource demand, specifically:

[0061] I. Collect spectrum usage and base station load in different time periods and organize them into data set P1, and extract resource demand impact features in traffic data and organize them into data set P2;

[0062] The data of spectrum usage and base station load are organized into a data set P1, which is expressed as: P1 = {(a1, b1), (a2, b2), (a3, b3), …, (a n ,b n )},a1~a n are the spectrum usage in different time periods, b1~b n are the base station loads in different time periods, and n is the number of acquisitions;

[0063] The factors affecting resource demand in the traffic data are extracted, including traffic patterns, user distribution, special events (such as holidays, large-scale events), etc., and the resource demand influencing features are formed after normalization. After serialization, they are organized into a data set P2, which is expressed as: P2 = {R1, R2, R3, ..., R n}, R1~R n are the resource demand impact feature sets in different time periods, n is the number of collections, R i={r1,r2,r3,…,r m}, r1~r m are different resource impact characteristics respectively, and m is the number of resource demand impact characteristic items.

[0064] II. Take dataset P2 as input set and dataset P1 as output set to train and construct network resource demand prediction model;

[0065] The network resource demand prediction model includes two parts: spectrum demand prediction model and base station load prediction model, which can be expressed as:

[0066]

[0067] ,in is the spectrum demand prediction model, and the output result is recorded as y1. is the base station load prediction model, and the output result is recorded as y2, ε1, ε2,…,ε z is the coefficient of the autoregressive term, is the autoregressive term, z is the order of the autoregressive term, d is the number of differences, x j is the resource demand impact feature of the jth input, j ranges from 1 to m, m is the number of resource demand impact feature items, σ1,σ2,…,σ q is the coefficient of the moving average term, γ j The error term of the j-th resource demand influencing term, q is the order of the moving average term, γ1,γ2,…,γ q is the error term of the moving average term.

[0068] III. Optimize the parameters of the network resource demand prediction model to ensure the accuracy of the model;

[0069] The training data set is divided into k subsets of equal size. In each iteration, one subset is selected as the validation set and the rest as the training set. This process is repeated k times, and a different subset is selected as the validation set each time. The optimal model parameters are determined based on the loss function.

[0070] The loss function is expressed as: where y1 k is the spectrum demand predicted for the kth time, Δy1 k is the actual spectrum usage for the kth time, y2 k is the k-th predicted base station load, Δy2 k is the actual base station load for the kth time, k ranges from 1 to L, and L is the total number of training times.

[0071] IV. Obtain the traffic data generated by users within the operation scope in real time through the monitoring layer;

[0072] V. Extract resource demand characteristics from traffic data;

[0073] The flow data is transmitted to the feature extraction step using stream processing. Because the amount of flow data is huge, extracting it after it is transmitted will reduce the computing efficiency. Therefore, the stream processing method is adopted. The control layer extracts the resource demand characteristics in the data while receiving the data.

[0074] VI. Input resource demand characteristics into the network resource demand prediction model and output the prediction results of network resource demand.

[0075] 2. A network resource dynamic adjustment module, which is used to generate and execute a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction results of network resource demand;

[0076] Generate a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction result of network resource demand, which is specifically divided into the following sub-steps:

[0077] I. Spectrum and base stations are used as resource adjustment items to construct a network resource dynamic adjustment function;

[0078] The network resource dynamic adjustment function includes two functions: spectrum frequency adjustment function and base station load adjustment function. The spectrum adjustment function is expressed as L1(y1). The calculation process is: If it is true, it means that the current spectrum frequency is too large, then execute τ φ =max(τ1,τ2,…,τ Φ )-t to reduce the spectrum frequency, if If it is true, it means that the current spectrum frequency is too small, then execute τ φ =min(τ1,τ2,…,τ Φ )+t to increase the spectrum frequency, if If it is true, it means that the current spectrum frequency just meets the predicted demand, then the frequency adjustment of each segment of the spectrum τ1, τ2, ..., τ Φ , where y1 is the predicted spectrum demand, c φ is the current frequency of the φth segment of the spectrum, τ φ is the adjustment amount of the φth segment spectrum frequency, l1 φ is the working efficiency of the φth spectrum segment, φ ranges from 1 to Φ, Φ is the number of spectrum segments, τ1, τ2, …, τ Φ is the adjustment amount of each spectrum, and t is the adjustment step.

[0079] The base station load adjustment function is expressed as L2(y2), and the calculation process is: This indicates that the current base station load is too high, so execute o =max(υ1,υ1,…,υ Ψ )-t to reduce the base station load, if This indicates that the current base station load is too low, so execute o =min(υ1,υ1,…,υ Ψ )+t to increase the base station load. This means that the current base station load just meets the predicted demand of the base station, and the load adjustment of each base station is returned υ1,υ1,…,υ Ψ , where y2 is the predicted base station load, h o is the current load of the oth base station, υ o is the adjustment amount of o base stations, l2 o is the base station transmission distance, o takes values ​​from 1 to Ψ, Ψ is the total number of base stations, and t is the adjustment step size.

[0080] II. Pass the prediction result of network resource demand as input parameter to the network resource dynamic adjustment function, and return the network resource dynamic adjustment strategy;

[0081] (3) The forwarding layer forwards data based on the network resources adjusted by the control layer;

[0082] After the control layer executes the dynamic adjustment strategy of network resources, the forwarding layer will forward data again, and the data monitoring point of each base station is responsible for real-time monitoring of the efficiency of data forwarding. If it is lower than a certain threshold, it will initiate dynamic adjustment to the control layer again to improve the utilization efficiency of network resources.

[0083] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for dynamically adjusting 5G network resources, characterized in that: include: Step 1: Take the allocated spectrum resources and existing base stations as the main network resources, analyze the demand change trend of the main network resources in time series, and establish a network resource demand prediction model; Step 2: Separate the control layer from the forwarding layer. The control layer uses the network resource demand prediction model to obtain the prediction results of network resource demand. Step 3: The control layer generates and executes the network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction results of network resource demand. The specific steps are as follows: Use spectrum and base stations as resource adjustment items to build a dynamic adjustment function for network resources; The prediction result of network resource demand is passed as an input parameter to the network resource dynamic adjustment function, and the network resource dynamic adjustment strategy is returned; The network resource dynamic adjustment function includes two functions: spectrum frequency adjustment function and base station load adjustment function. The spectrum frequency adjustment function is expressed as , the calculation process is: If If it is true, it means that the current spectrum frequency is too large, then execute To reduce the frequency spectrum, if If it is true, it means that the current spectrum frequency is too small, then execute To increase the frequency of the spectrum, if If it is true, it means that the current spectrum frequency just meets the predicted demand, then the frequency adjustment of each segment of the spectrum is returned. , where y1 is the predicted spectrum demand, For the The current frequency of the spectrum segment, For the The adjustment amount of the frequency spectrum of the segment, For the The working efficiency of the spectrum segment, Value , is the number of spectrum segments, is the adjustment amount of each spectrum, and t is the adjustment step length; The base station load adjustment function is expressed as , the calculation process is: If , indicating that the current base station load is too high, then execute To reduce the base station load, if , indicating that the current base station load is too low, then execute To increase the base station load, if , indicating that the current base station load just meets the base station's predicted demand, then the load adjustment of each base station is returned , where y2 is the predicted base station load, is the current load of the oth base station, is the adjustment amount of o base stations, is the base station transmission distance, o value , is the total number of base stations, t is the adjustment step size; Step 4: The forwarding layer forwards data based on the adjusted network resources.

2. A 5G network resource dynamic adjustment method according to claim 1, characterized in that: Analyze the demand change trend of network resource entities in time series and establish a network resource demand prediction model, which specifically includes the following sub-steps: The spectrum usage and base station load conditions in different time periods are collected and organized into data set P1, and the resource demand impact characteristics in the traffic data are extracted and organized into data set P2; Take dataset P2 as input set and dataset P1 as output set to train and build network resource demand prediction model; Optimize the parameters of the network resource demand prediction model to ensure the accuracy of the model.

3. A 5G network resource dynamic adjustment method according to claim 1, characterized in that: The control layer uses the network resource demand prediction model to obtain the prediction results of network resource demand, which is divided into the following sub-steps: Set up monitoring points at the base station to obtain real-time traffic data generated by users within the operating range and transmit it to the control layer in the form of message queues; The control layer extracts resource demand characteristics from traffic data; Input resource demand characteristics into the network resource demand prediction model and output the prediction results of network resource demand.

4. A 5G network resource dynamic adjustment system, characterized in that: include: Monitoring layer, control layer and forwarding layer; The monitoring layer includes monitoring points deployed on all base stations within the operating range; The control layer includes a network resource demand prediction module and a network resource dynamic adjustment module, wherein the specific implementation process of the network resource dynamic adjustment module includes: Use spectrum and base stations as resource adjustment items to build a dynamic adjustment function for network resources; The prediction result of network resource demand is passed as an input parameter to the network resource dynamic adjustment function, and the network resource dynamic adjustment strategy is returned; The network resource dynamic adjustment function includes two functions: spectrum frequency adjustment function and base station load adjustment function. The spectrum adjustment function is expressed as , the calculation process is: If If it is true, it means that the current spectrum frequency is too large, then execute To reduce the frequency spectrum, if If it is true, it means that the current spectrum frequency is too small, then execute To increase the frequency of the spectrum, if If it is true, it means that the current spectrum frequency just meets the predicted demand, then the frequency adjustment of each segment of the spectrum is returned. , where y1 is the predicted spectrum demand, For the The current frequency of the spectrum segment, For the The adjustment amount of the frequency spectrum of the segment, For the The working efficiency of the spectrum segment, Value , is the number of spectrum segments, is the adjustment amount of each spectrum, and t is the adjustment step length; The base station load adjustment function is expressed as , the calculation process is: If , indicating that the current base station load is too high, then execute To reduce the base station load, if , indicating that the current base station load is too low, then execute To increase the base station load, if , indicating that the current base station load just meets the base station's predicted demand, then the load adjustment of each base station is returned , where y2 is the predicted base station load, is the current load of the oth base station, is the adjustment amount of o base stations, is the base station transmission distance, o value , is the total number of base stations, t is the adjustment step size; The forwarding layer forwards data based on the network resources adjusted by the control layer.

5. A 5G network resource dynamic adjustment system according to claim 4, characterized in that: The monitoring layer obtains the traffic data generated by users in real time, transmits it to the control layer in the form of message queues, and monitors the forwarding efficiency of network resources. If it is lower than the preset threshold, it initiates dynamic adjustments to the control layer.

6. A 5G network resource dynamic adjustment system according to claim 4, characterized in that: The network resource demand prediction module is used to analyze the demand change trend of spectrum resource and base station resource usage in time series, establish a network resource demand prediction model based on the analysis results, and use the network resource demand prediction model to obtain the prediction results of network resource demand.

7. A 5G network resource dynamic adjustment system according to claim 4, characterized in that: The network resource dynamic adjustment module is used to generate and execute a network resource dynamic adjustment strategy based on the network resource dynamic adjustment function and the prediction result of the network resource demand.