LTE and 5G network service load prediction and network planning method and system
By using quantum-inspired machine learning models and network digital twin technology in LTE and 5G networks, combining feature engineering and cross-validation to dynamically adjust the network slice configuration, the shortcomings of business load prediction and network planning in the existing technology are solved, and efficient , accurate network resource management and optimization.
Patent Information
- Application Number
- CN202410940422.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art fails to fully utilize the advantages of quantum computing in LTE and 5G network service load prediction, resulting in insufficient improvement in model performance and lack of deep fusion and dynamic feedback regulation, resulting in insufficient real-time and flexibility of network planning.
Collect multi-source data through network monitoring systems and sensors, clean and normalize, use feature engineering to extract features, build a quantum-inspired machine model, combine cross-verification technology to optimize model parameters, create highly simulated network digital twins, and input quantum The machine learning model predicts business load data, simulates network performance, and dynamically adjusts the network slice configuration based on the evaluation results.
It realizes the optimization of network performance and efficient allocation of resources, improves prediction accuracy and real-time network planning, ensures high-quality service levels in high loads or emergencies, and integrates the latest data science, quantum computing and network management. Progress supports efficient operation and future expansion of 5G networks.
Smart Images

Figure CN119095061A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of communication network management and optimization, and in particular to a method and system for LTE and 5G network service load prediction and network planning. Background Art
[0002] In the rapid development of contemporary communication technology, with the maturity and popularization of the fourth-generation mobile communication system (LTE) and the rapid development of the fifth-generation (5G) network, the diversification of network services and the explosive growth of data traffic have put forward higher requirements for network management and optimization. Traditional network planning methods mainly rely on historical data statistical analysis and manual experience.
[0003] However, existing technologies have exposed some shortcomings in practice. Although the application of machine learning technology has improved prediction efficiency, most models are still based on classical algorithms and ignore the potential advantages brought by quantum computing, such as high-speed parallel processing capabilities, and fail to achieve a breakthrough improvement in model performance. Furthermore, although the application of network digital twin technology can simulate network behavior, it lacks deep integration with prediction models and a dynamic feedback adjustment mechanism for prediction results, resulting in insufficient real-time and flexibility in network planning. Summary of the invention
[0004] In view of the above existing problems, the present invention is proposed.
[0005] Therefore, the present invention provides a method and system for LTE and 5G network service load prediction and network planning to solve the problem of how to efficiently and accurately predict LTE and 5G network service loads and dynamically optimize network slice configuration accordingly to improve resource allocation efficiency and network performance.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0007] In the first aspect, an embodiment of the present invention provides a method for LTE and 5G network service load forecasting and network planning, which includes collecting multi-source data through a network monitoring system and sensors, cleaning and normalizing the data, using feature engineering technology to extract relevant features for network load forecasting, building a quantum-inspired machine model, inputting the processed feature data into the model for learning and optimizing the model parameters through cross-validation technology, creating a highly simulated network digital twin, inputting the service load data predicted by the quantum machine learning model, simulating the network performance after implementation, evaluating the network performance and resource allocation effect, dynamically adjusting the network slice configuration according to the evaluation results, and implementing network planning.
[0008] As a preferred solution of the LTE and 5G network service load prediction and network planning method described in the present invention, the network monitoring system and sensors collect multi-source data by installing traffic analyzers, sniffers and agent programs on key network nodes and devices to track and record data passing through these nodes, configuring router switches to support the Simple Network Management Protocol, setting the frequency of data collection, and collecting data in real time.
[0009] As a preferred solution of the LTE and 5G network service load prediction and network planning method of the present invention, wherein: the multi-source data is collected through the network monitoring system and sensors, the data is cleaned and normalized, and the relevant features of the network load prediction are extracted using feature engineering technology, which is specifically as follows:
[0010] The data is cleaned and normalized by using the Z-Score method to remove outliers, then using the median filling method to complete, and finally using Min-Max to normalize all numerical features. The normalized expression is:
[0011]
[0012] Among them, D′ is the data set after the previous processing, Ω(D′) represents the data set after max-min normalization processing, ω d is the new value of any element d in the data set D′ after normalization, max(D′) and min(D′) represent the maximum and minimum values in the data set D′, respectively;
[0013] Use the Pearson correlation coefficient formula to extract the minimum, maximum and average values of time series features;
[0014] Use K-means clustering algorithm to extract user distribution features and find the cluster center that minimizes the intra-class sum of squares through iteration;
[0015] Extract network resource usage, the average load rate expression is:
[0016]
[0017] Among them, L avg is the average load rate of the network equipment, L j is the load rate of the jth device, and m is the total number of devices;
[0018] The idle resource ratio expression is:
[0019]
[0020] Among them, P idle is the idle resource ratio, C jis the total resource capacity of the jth device, U j is the amount of resources currently used by the device.
[0021] As a preferred solution of the LTE and 5G network service load prediction and network planning method of the present invention, wherein: the construction of the quantum-inspired machine model, inputting the processed feature data into the model for learning and optimizing the model parameters through cross-validation technology are specifically as follows:
[0022] The classical features are usually mapped to quantum features to generate quantum states, and the expression is:
[0023]
[0024] Among them, x i is the i-th normalized eigenvalue in the eigenvector, i is the ground state consisting of n qubits, and |ψ(x)> is the quantum state;
[0025] The similarity is calculated by calculating the inner product between quantum states to obtain the quantum kernel, which is expressed as:
[0026] K Q (x i ,x j )=|<ψ(x i )|ψ(x j )>| 2 ;
[0027] Among them, K Q (x i ,x j ), is the quantum kernel function;
[0028] Using the quantum kernel K Q The model is constructed in the classic machine learning framework, and the cross-validation technique is used to optimize the model parameters. The expression is:
[0029]
[0030] Among them, θ is the learnable parameter of the model, D train represents the training data set, is the optimal model parameter value, O opt is the optimization goal;
[0031] The predicted business load data is output by the quantum machine learning model, and the expression is:
[0032] y predicted =sign(∑ i α i K Q (x i ,x j )+b);
[0033] Among them, y predicted is the output result of the business load forecast, sign is a symbolic function, b is the bias term of the quantum machine learning model, and α i is the support vector.
[0034] As a preferred solution of the LTE and 5G network service load prediction and network planning method of the present invention, wherein: the highly simulated network digital twin is created and input into the quantum machine learning model, and the predicted service load data is expressed as:
[0035] F total (D raw θ c ,φ,η,λ)=O opt (g sim (T twin ,y predicted ;φ);η);λ);
[0036] Among them, Q pred is a quantum machine learning prediction model, T twin is a network digital twin, g sim is the simulation function, F total This is the final overall evaluation result.
[0037] As a preferred solution of the LTE and 5G network service load prediction and network planning method of the present invention, the network slice configuration is dynamically adjusted according to the evaluation results to implement network planning, which is specifically as follows:
[0038] When the evaluation result F total If it is lower than f, adjust the network configuration or the source allocation strategy;
[0039] When the evaluation result F total If it is higher than f, it is normal and network planning is required.
[0040] As a preferred solution of the LTE and 5G network service load prediction and network planning method of the present invention, the network slice configuration is dynamically adjusted according to the evaluation results to implement network planning, which is specifically as follows:
[0041] Analyze the evaluation results to ensure the latest network performance indicators and resource usage, updated in real time from the network monitoring system and digital twin, and total Identify bottleneck areas of network performance;
[0042] Based on F total and predictedThe analysis results are used to sort the adjustment requirements of different slices, and the resource allocation of each slice is automatically adjusted through SDN / NFV technology. For the identified bottlenecks, the boundaries of the slices are dynamically adjusted, and slices are merged and split.
[0043] In the second aspect, the present invention provides a LTE and 5G network business load prediction and network planning system, including a data preprocessing module, a quantum-inspired machine learning model construction module, a network digital twin simulation module and a network evaluation and planning module. The data preprocessing module is used to collect multi-source data through a network monitoring system and sensors, clean and normalize the data, and use feature engineering technology to extract relevant features for network load prediction. The quantum-inspired machine learning model construction module is used to build a quantum-inspired machine model, input the processed feature data into the model for learning and optimize the model parameters through cross-validation technology. The network digital twin simulation module is used to build a highly simulated network digital twin, input the business load data predicted by the quantum machine learning model, simulate the network performance after implementation, and evaluate the network performance and resource allocation effect. The network evaluation and planning module is used to dynamically adjust the network slice configuration according to the evaluation results, and perform real-time network planning.
[0044] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the LTE and 5G network service load prediction and network planning method as described in the first aspect of the present invention is implemented.
[0045] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the LTE and 5G network service load prediction and network planning method as described in the first aspect of the present invention.
[0046] The beneficial effects of the present invention are as follows: network performance optimization and efficient resource allocation are achieved through multi-stage technical means. First, multi-source data collection is carried out through monitoring systems and sensors deployed at key nodes of the network, and detailed data cleaning, outlier processing and normalization are performed. Key features are extracted by combining feature engineering technology, which lays a solid data foundation for the prediction model. Then, the innovative quantum-inspired machine learning model is used to not only accelerate the learning process, but also significantly improve the prediction accuracy through advanced feature mapping and quantum kernel techniques. The constructed network digital twin is used as a virtual test bench. After inputting the predicted load data, it can accurately simulate the network behavior, help in-depth evaluation and optimization of network planning, and finally, according to the simulation results of the digital twin, the network slice configuration is dynamically adjusted to realize the intelligent and dynamic management of network resources, ensuring that high-quality service levels can be maintained even under high load or emergency conditions. The entire solution integrates the latest progress in data science, quantum computing and network management, and provides strong technical support for the efficient operation and future expansion of 5G networks. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0048] Figure 1 This is a flowchart of the LTE and 5G network service load prediction and network planning method in Example 1. DETAILED DESCRIPTION
[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.
[0050] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0051] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.
[0052] Example 1, reference Figure 1, which is the first embodiment of the present invention, and provides a method for predicting LTE and 5G network service load and network planning, comprising the following steps:
[0053] S1. Collect multi-source data through network monitoring systems and sensors, clean and normalize the data, and use feature engineering technology to extract relevant features for network load prediction.
[0054] Furthermore, network monitoring systems and sensors collect multi-source data by installing traffic analyzers, sniffers, and agents on key network nodes and devices to track and record data passing through these nodes (including but not limited to data frame profiles, Ethernet frame header information, IP packet header information), and configuring network devices (such as routers and switches) to support the Simple Network Management Protocol (SNMP) for monitoring device status and performance parameters, setting the frequency of data collection, and collecting data in real time.
[0055] The data is cleaned and normalized by using the Z-Score method to remove outliers and then using the median filling method to complete the data. Finally, all numerical features are normalized using Min-Max to ensure that all feature values are mapped to the [0,1] interval, ensuring the consistency and comparability of the data. The normalized expression is:
[0056]
[0057] Among them, D′ is the data set after the preliminary processing, Ω(D′) represents the data set after max-min normalization processing, ω d : represents the new value of any element d in the data set D′ after normalization, max(D′) and min(D′) represent the maximum and minimum values in the data set D′, respectively;
[0058] Using feature engineering technology, the relevant features of network load prediction are extracted as follows:
[0059] Use the Pearson correlation coefficient formula to identify network time series features that are closely related to network load, including but not limited to minimum, maximum, and average values;
[0060] The K-means clustering algorithm is used to extract user distribution features. The cluster center that minimizes the intra-class sum of squares is found through iteration. At the same time, the average load rate and idle resource ratio of network devices are calculated to further refine the usage of network resources. These carefully extracted features then become the cornerstone of the prediction model.
[0061] The average load factor expression is:
[0062]
[0063] Among them, L avg is the average load rate of the network equipment, L j is the load rate of the jth device, and m is the total number of devices;
[0064] The idle resource ratio expression is:
[0065]
[0066] Among them, P idle is the idle resource ratio, C j is the total resource capacity of the jth device, U j is the amount of resources currently used by the device.
[0067] S2. Build a quantum-inspired machine model, input the processed feature data into the model for learning, and optimize the model parameters through cross-validation technology.
[0068] Furthermore, we build a quantum-inspired machine model, transform traditional features into quantum states, use quantum mapping functions, calculate similarities through the inner product of quantum states, and form quantum cores. This process not only improves computational efficiency, but also increases the model's ability to process complex data. The processed feature data is input into the model for learning and the model parameters are optimized through cross-validation technology to ensure the accuracy of predictions. The details are as follows:
[0069] The classical features are usually mapped to quantum features to generate quantum states, and the expression is:
[0070]
[0071] Among them, x i is the i-th normalized eigenvalue in the eigenvector, i is the ground state consisting of n qubits, and |ψ(x)> is the quantum state;
[0072] The similarity is calculated by calculating the inner product between quantum states to obtain the quantum kernel, which is expressed as:
[0073] K Q (x i ,x j )=|<ψ(x i )|ψ(x j )>| 2 ;
[0074] Among them, K Q (x i ,x j ), is the quantum kernel function, used to measure the input data x i and x j Similarity after quantum feature mapping;
[0075] Using the quantum kernel K Q The model is constructed in the classic machine learning framework, and the cross-validation technique is used to optimize the model parameters. The expression is:
[0076]
[0077] Among them, θ is all the learnable parameters of the model, D train represents the training data set, is the optimal model parameter value, O opt is the optimization goal, which is to find the optimal set of model parameters by minimizing the loss function ("Loss");
[0078] The predicted business load data is output by the quantum machine learning model, and the expression is:
[0079] y predicted =sign(∑ i α i K Q (x i ,x j )+b);
[0080] Among them, y predicted is the output result of business load forecasting, sign is a symbolic function, and b is the bias term of the model; α i corresponds to the support vectors.
[0081] S3. Create a highly simulated network digital twin, input the business load data predicted by the quantum machine learning model, simulate the network performance after implementation, and evaluate the network performance and resource allocation effect.
[0082] Furthermore, a highly simulated network digital twin is created and the business load data predicted by the quantum machine learning model is input into the highly simulated network digital twin, which is like a virtual mirror of the real network and can receive the predicted data of the quantum machine learning model. It simulates the actual performance of the network under the predicted load and provides an intuitive simulation platform for network planning, helping to evaluate the network performance and effectiveness of resource allocation under different configurations. The expression is:
[0083] F total (D raw θ c ,φ,η,λ)=O opt (g sim (T twin ,y predicted ;φ);η);λ);
[0084] Among them, Q predIt is a quantum machine learning prediction model used to predict future business loads. twin It is a network digital twin that simulates the real network environment. sim is the simulation function, F total This is the final overall evaluation result.
[0085] S4. Dynamically adjust network slice configuration based on evaluation results and implement network planning.
[0086] Furthermore, the network slice configuration is dynamically adjusted according to the evaluation results. The implementation of network planning is to set a threshold f (f is the minimum adopted indicator of key performance) for key performance indicators (including but not limited to throughput, latency, packet loss rate) and resource usage (including but not limited to CPU, memory, bandwidth utilization). By comparing F total Compared with the set threshold f, it evaluates whether the network performance meets the standard and whether the resource allocation is reasonable. If the evaluation result F total If the evaluation result F is lower than f, the network configuration or resource allocation strategy is adjusted. total If it exceeds f, it is normal and the solution can be used for network planning.
[0087] Furthermore, the network slice configuration is dynamically adjusted according to the evaluation results to implement network planning, as follows:
[0088] Analyze the evaluation results and confirm that the latest network performance indicators (such as throughput, latency, packet loss rate) and resource usage (such as CPU, memory, bandwidth utilization) are updated in real time from the network monitoring system and digital twins. total Identify bottleneck areas of network performance, including but not limited to the highest latency areas within a specific time window, slices with resource utilization below a threshold, and services with frequent SLA violations;
[0089] Based on F total and predicted Based on the analysis results, for areas where the business volume is predicted to increase significantly, increase resource reservation and expand the slice size. According to the business importance, SLA requirements and current network conditions, sort the adjustment needs of different slices to ensure that key businesses are given priority. Through SDN / NFV technology, automatically adjust the resource allocation of each slice, including but not limited to bandwidth allocation, reconfiguration of computing resources, and optimization of storage resources. For identified bottlenecks, dynamically adjust the boundaries of slices, merge or split slices to optimize resource utilization and improve service quality.
[0090] This embodiment also provides a LTE and 5G network service load prediction and network planning system, including:
[0091] Multi-source data acquisition module, data preprocessing module, feature engineering module, quantum-inspired machine learning model construction module, network digital twin simulation module, network evaluation and planning module; the multi-source data acquisition module is used to collect the operation data and environmental information of LTE and 5G networks in real time through the network monitoring system and sensors deployed at key nodes, providing a comprehensive data basis for predictive analysis; the data preprocessing module is used to clean the received multi-source data, remove outliers and redundant information, and implement normalization processing to ensure data consistency and availability; the feature engineering module is used to extract key features that affect network load, such as historical load trends, user behavior patterns, etc., based on the preprocessed data, to improve prediction Accuracy; the quantum-inspired machine learning model construction module is used to build a machine learning model based on quantum computing principles, using the extracted features as input, performing efficient calculations and learning through quantum state evolution, and optimizing model parameters through cross-validation to improve load forecasting accuracy; the network digital twin simulation module is used to build a high-fidelity network digital twin, taking the business load data predicted by the quantum machine learning model as input, simulating the future network status, and providing real-time feedback on network performance changes and resource allocation efficiency to assist in decision making; the network evaluation and planning module is used to dynamically adjust the network slicing configuration strategy based on the evaluation results of the digital twin simulation module, to achieve the optimal layout of network resources and improve the overall network performance and service quality.
[0092] This embodiment also provides a computer device, which is applicable to the case of LTE and 5G network service load prediction and network planning methods, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the LTE and 5G network service load prediction and network planning methods proposed in the above embodiments.
[0093] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
[0094] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for implementing LTE and 5G network service load prediction and network planning as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, SRAM for short), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM for short), erasable programmable read-only memory (Erasable Programmable Read Only Memory, EPROM for short), programmable read-only memory (Programmable Red-Only Memory, PROM for short), read-only memory (Read-Only Memory, ROM for short), magnetic storage, flash memory, disk or optical disk.
[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A method for predicting LTE and 5G network traffic load and network planning, Features: include, Collect multi-source data through network monitoring systems and sensors, clean and normalize the data, and use feature engineering technology to extract relevant features for network load prediction; Build a quantum-inspired machine model, input the processed feature data into the model for learning, and optimize the model parameters through cross-validation techniques; Create a highly simulated network digital twin, input the business load data predicted by the quantum machine learning model, simulate the network performance after implementation, and evaluate the network performance and resource allocation effect; Dynamically adjust network slice configuration based on the evaluation results and implement network planning.
2. The LTE and 5G network service load prediction and network planning method as claimed in claim 1, Features: Network monitoring systems and sensors collect multi-source data by installing traffic analyzers, sniffers, and agents on key network nodes and devices to track and record data passing through these nodes, configuring routers and switches to support simple network management protocols, setting the frequency of data collection, and collecting data in real time.
3. The LTE and 5G network traffic load prediction and network planning method as claimed in claim 2, Features: The multi-source data is collected through the network monitoring system and sensors, the data is cleaned and normalized, and the relevant features of network load prediction are extracted using feature engineering technology, as follows: The data is cleaned and normalized by using the Z-Score method to remove outliers, then using the median filling method to complete, and finally using Min-Max to normalize all numerical features. The normalized expression is: Among them, D′ is the data set after the previous processing, Ω(D′) represents the data set after max-min normalization processing, ω d is the new value of any element d in the data set D′ after normalization, max(D′) and min(D′) represent the maximum and minimum values in the data set D′, respectively; Use the Pearson correlation coefficient formula to extract the minimum, maximum and average values of time series features; Use K-means clustering algorithm to extract user distribution features and find the cluster center that minimizes the intra-class sum of squares through iteration; Extract network resource usage, the average load rate expression is: Among them, L avg is the average load rate of the network equipment, L j is the load rate of the jth device, and m is the total number of devices; The idle resource ratio expression is: Among them, P idle is the idle resource ratio, C j is the total resource capacity of the jth device, U j is the amount of resources currently used by the device.
4. The LTE and 5G network service load prediction and network planning method as claimed in claim 3, Features: The construction of the quantum-inspired machine model, inputting the processed feature data into the model for learning and optimizing the model parameters through cross-validation technology are as follows: The classical features are usually mapped to quantum features to generate quantum states, and the expression is: Among them, x i is the i-th normalized eigenvalue in the eigenvector, i is the ground state consisting of n qubits, and |ψ(x)> is the quantum state; The similarity is calculated by calculating the inner product between quantum states to obtain the quantum kernel, which is expressed as: K Q (x i ,x j )=|<ψ(x i )|ψ(x j ))| 2 ; Among them, K Q (x i , x j ), is the quantum kernel function; Using the quantum kernel K Q The model is constructed in the classic machine learning framework, and the cross-validation technique is used to optimize the model parameters. The expression is: Among them, θ is the learnable parameter of the model, D train represents the training data set, is the optimal model parameter value, O opt is the optimization goal; The predicted business load data is output by the quantum machine learning model, and the expression is: y predicted =sign(∑ i α i K Q (x i ,x j )+b); Among them, y predicted is the output result of the business load forecast, sign is a symbolic function, b is the bias term of the quantum machine learning model, and α i is the support vector.
5. The LTE and 5G network service load prediction and network planning method according to claim 4, Features: The highly simulated network digital twin is created and input into the quantum machine learning model to predict the business load data, and the expression is: F total (D raw ;θ c ,φ,η,λ)=O opt (g sim (T twin ,y predicted ;f); η); λ); Among them, Q pred is a quantum machine learning prediction model, T twin is a network digital twin, g sim is the simulation function, F total This is the final overall evaluation result.
6. The LTE and 5G network traffic load prediction and network planning method as claimed in claim 5, Features: The network slice configuration is dynamically adjusted according to the evaluation results to implement network planning, as follows: When the evaluation result F total If it is lower than f, adjust the network configuration and resource allocation strategy; When the evaluation result F total If it is higher than f, it is normal and network planning is required.
7. The LTE and 5G network traffic load prediction and network planning method as claimed in claim 6, Features: The network slice configuration is dynamically adjusted according to the evaluation results to implement network planning, as follows: Analyze the evaluation results to ensure the latest network performance indicators and resource usage, updated in real time from the network monitoring system and digital twin, and total Identify bottleneck areas of network performance; Based on F total and predicted The analysis results are used to sort the adjustment requirements of different slices, and the resource allocation of each slice is automatically adjusted through SDN / NFV technology. For the identified bottlenecks, the boundaries of the slices are dynamically adjusted, and slices are merged and split.
8. A LTE and 5G network service load prediction and network planning system, based on the LTE and 5G network service load prediction and network planning method according to any one of claims 1 to 7, Features: Including, data preprocessing module, quantum-inspired machine learning model building module, network digital twin simulation module and network evaluation and planning module; The data preprocessing module is used to collect multi-source data through the network monitoring system and sensors, clean and normalize the data, and use feature engineering technology to extract relevant features for network load prediction; The quantum-inspired machine learning model building module is used to build a quantum-inspired machine model, input the processed feature data into the model for learning, and optimize the model parameters through cross-validation technology; The network digital twin simulation module is used to build a highly simulated network digital twin, input the business load data predicted by the quantum machine learning model, simulate the network performance after implementation, and evaluate the network performance and resource allocation effect; The network evaluation and planning module is used to dynamically adjust the network slice configuration according to the evaluation results and perform real-time network planning.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. Features: When the processor executes the computer program, the steps of the LTE and 5G network service load prediction and network planning method described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, Features: When the computer program is executed by a processor, the steps of the LTE and 5G network service load prediction and network planning method described in any one of claims 1 to 7 are implemented.
Citation Information
Cited By
Network resource allocation method and network resource allocation system
CN120499022A
