A 5G-A Network Slicing Dynamic Adjustment and Resource Allocation System for Low-Altitude Economy

Abstract

This invention relates to the field of network communication technology, specifically a 5G-A network slice dynamic adjustment and resource allocation system for the low-altitude economy. The system includes a mobility management feature integration module, a network slice service level protocol evaluation module, a mobility management demand prediction benchmark construction module, a multi-dimensional trend prediction module, a joint optimization decision generation module, and a strategy verification and execution module. The system acquires a mobility management feature dataset of the wireless network; evaluates the network slice service level protocol to obtain network slice status evaluation results; constructs a mobility management demand prediction benchmark; performs multi-dimensional trend analysis on the network slice status evaluation results to obtain prediction results; performs joint optimization decision-making on standard network slice configuration strategies to obtain network slice adjustment strategies; and executes the verified slice adjustment strategies to achieve dynamic adjustment and resource allocation of network slices. This invention improves the efficiency of dynamic adjustment and resource allocation of network slices.

Description

Technical Field

[0001] This invention relates to the field of network communication technology, and in particular to a 5G-A network slicing dynamic adjustment and resource allocation system for the low-altitude economy. Background Technology

[0002] In low-altitude economic scenarios, wireless networks need to support diverse and dynamically changing service demands. However, existing network slicing technologies struggle to effectively integrate multi-dimensional characteristics related to mobility management, failing to accurately capture the correlation between real-time mobility management types, priorities, latency sensitivity, and bandwidth requirements. This results in a lack of comprehensive data support for network slicing service level agreement (SSR) evaluations, making it difficult to accurately reflect the impact of real-time network load and wireless channel quality on sliced ​​services. Consequently, resource allocation deviates from actual service needs, failing to guarantee differentiated service quality for different services.

[0003] Existing technologies lack scientific benchmarks for predicting mobility management needs and multi-dimensional trend analysis mechanisms, making it difficult to accurately predict network slice growth by combining historical slice status information with expected business growth trends. This results in a lack of foresight in network slice adjustment strategies. Furthermore, the strategy optimization process fails to adequately consider the gap between resource demand and configuration supply, and the feasibility verification process is inadequate, easily leading to policy conflicts or resource compatibility issues. This not only wastes network resources but may also cause slice service interruptions or performance degradation, failing to meet the network's needs for dynamic adjustment and efficient resource allocation in low-altitude economic scenarios. Therefore, improving the efficiency of dynamic network slice adjustment and resource allocation has become an urgent problem to be solved. Summary of the Invention

[0004] To achieve the above objectives, this invention provides a 5G-A network slice dynamic adjustment and resource allocation system for the low-altitude economy. The system comprises a mobility management feature integration module, a network slice service level agreement evaluation module, a mobility management demand prediction benchmark construction module, a multi-dimensional trend prediction module, a joint optimization decision generation module, and a strategy verification and execution module, wherein:

[0005] The mobility management feature integration module is used to integrate the real-time mobility management type, mobility management priority, latency sensitivity and bandwidth requirements in the wireless network into the mobility management feature dataset of the wireless network.

[0006] The network slice service level protocol evaluation module is used to evaluate the real-time load status and wireless channel quality parameters of the wireless network based on the mobility management feature dataset, and obtain the network slice status evaluation result of the wireless network.

[0007] The mobility management demand prediction benchmark construction module is used to construct the mobility management demand prediction benchmark of the wireless network based on the mobility management feature dataset and the expected demand and historical slice status information database of the wireless network.

[0008] The multi-dimensional trend prediction module is used to perform multi-dimensional trend analysis on the network slice status assessment results based on the mobility management demand prediction benchmark, and obtain the prediction results of the wireless network.

[0009] The joint optimization decision generation module is used to perform joint optimization decision on the standard network slice configuration strategy of the wireless network based on the prediction results, so as to obtain the network slice adjustment strategy of the wireless network.

[0010] The policy verification and execution module is used to verify the feasibility of the network slice adjustment policy and execute the verified slice adjustment policy to realize the dynamic adjustment and resource allocation of the network slices of the wireless network.

[0011] In a preferred embodiment, when the mobility management feature integration module integrates the real-time mobility management types, mobility management priorities, latency sensitivity, and bandwidth requirements in the wireless network into a mobility management feature dataset for the wireless network, it is specifically used for:

[0012] Extract the real-time mobility management type parameters and corresponding mobility management priority policy parameters from the wireless network;

[0013] The real-time mobility management type parameter and the mobility management priority policy parameter are subjected to network protocol quantization to obtain the mobility management policy vector of the wireless network;

[0014] The network slice selection auxiliary information and service subscription data of the wireless network are used to perform performance quality screening to obtain the latency sensitivity level parameters and bandwidth requirement parameters of the wireless network.

[0015] The mobility management policy vector, the latency sensitivity level parameter, and the bandwidth requirement parameter are correlated in multiple dimensions to obtain the mobility management feature dataset of the wireless network.

[0016] In a preferred embodiment, the network slicing service level protocol evaluation module performs a network slicing service level protocol evaluation on the real-time load status and wireless channel quality parameters of the wireless network based on the mobility management feature dataset, and obtains the network slicing status evaluation result of the wireless network, specifically for:

[0017] Obtain the real-time load status of the wireless network;

[0018] The channel spectrum parameters captured in the wireless network are used as the wireless channel quality parameters of the wireless network.

[0019] The real-time load status and the wireless channel quality parameters are time-fused and aligned to obtain a snapshot of the network operation status of the wireless network.

[0020] Based on the latency, bandwidth, and reliability indicators defined in the network slicing service level protocol of the wireless network, a matching degree analysis is performed on the mobility management feature dataset and the network operation status snapshot to obtain the sub-item evaluation results of the mobility management feature dataset.

[0021] Based on preset evaluation weight rules, the sub-evaluation results are weighted and aggregated to obtain the network slice status evaluation result of the wireless network.

[0022] In a preferred embodiment, the calculation formula for the network slice state evaluation result is as follows:

[0023] ;

[0024] In the formula, The network slice state evaluation result, For the mobility management feature dataset, the mobility management type is... For a specific mobility management type among the mobility management types, The type weight coefficient is the pre-defined evaluation weight rule. This is the balance coefficient for the service demand matching degree in the evaluation weighting rules. For the specific mobility management type The service quality requirement vector, This is the balance coefficient for network state stability in the evaluation weighting rule. For the specific mobility management type The service capability vector that actually guarantees business operations. This is the network operating state vector in the network operating state snapshot. For the type in the evaluation weighting rule The ideal state threshold vector.

[0025] In a preferred embodiment, when the mobility management demand prediction benchmark construction module constructs the mobility management demand prediction benchmark for the wireless network based on the mobility management feature dataset and the expected demand and historical slice status information database of the wireless network, it is specifically used for:

[0026] The expected business growth information and service quality targets for low-altitude economic scenarios in the wireless network are integrated into the network planning strategy data of the wireless network.

[0027] Based on the historical slice status information database in the wireless network, the mobility management feature dataset is traversed and analyzed to obtain the joint feature sequence of the wireless network;

[0028] Based on the expected business growth information, the joint feature sequence is extrapolated to obtain the initial prediction of the mobility management feature dataset;

[0029] Using the network planning strategy data as constraints, the initial prediction is corrected for compliance, resulting in the standard prediction of the mobility management feature dataset.

[0030] The standard predicted quantity is correlated and mapped with the key performance indicator benchmark values ​​in the historical slice status information database to obtain the mobility management demand prediction benchmark of the wireless network.

[0031] In a preferred embodiment, when the mobility management demand forecasting benchmark construction module performs trend extrapolation on the joint feature sequence based on the expected business growth information to obtain the initial forecast amount of the mobility management feature dataset, it is specifically used for:

[0032] The expected business growth information is separated into business data to obtain the business volume growth data and the expected introduction characteristics of new businesses.

[0033] The joint feature sequence is periodically decomposed to obtain the long-term trend component and periodic fluctuation component of the joint feature sequence.

[0034] Based on the long-term trend components and the business volume growth data, the joint feature sequence is extrapolated over time to obtain the basic feature prediction values ​​of the mobility management feature dataset.

[0035] By combining the periodic fluctuation component with the expected introduction features of new services, the predicted values ​​of the basic features are superimposed and corrected to obtain the initial predicted values ​​of the mobility management feature dataset.

[0036] In a preferred embodiment, when the mobility management demand prediction baseline construction module performs an association mapping between the standard prediction quantity and the key performance indicator baseline values ​​in the historical slice status information database to obtain the mobility management demand prediction baseline for the wireless network, it is specifically used for:

[0037] Based on the historical slice status information database, the historical key performance index data associated with the features in the standard prediction quantity are matched and filtered to obtain the historical key index benchmark data of the standard prediction quantity.

[0038] The feature items related to mobility management type, priority and service quality requirements in the standard predicted quantity are matched and mapped with the historical key indicator benchmark data to establish a feature benchmark association table of the standard predicted quantity.

[0039] Based on the feature benchmark association table, the service quality requirement features in the standard prediction quantity are quantified for network slicing requirements to obtain the target values ​​of the key performance indicators of the wireless network.

[0040] The target values ​​of the key performance indicators are integrated with the mobility management type and priority characteristics in the standard forecasts to form the mobility management demand forecasting benchmark for the wireless network.

[0041] In a preferred embodiment, when the multi-dimensional trend prediction module performs multi-dimensional trend analysis on the network slice status assessment results based on the mobility management demand prediction benchmark to obtain the prediction results of the wireless network, it is specifically used for:

[0042] The deviation of the network slice status assessment results is compared with the mobility management demand prediction benchmark to obtain the difference feature set of the wireless network;

[0043] By performing a trend extrapolation of the service quality of the network slices in the differential feature set in chronological order, the time series analysis results of the wireless network are obtained.

[0044] Based on the time series analysis results and the network slice type of the wireless network, key factors are identified in the differential feature set to determine the dominant factors for slice allocation of the wireless network.

[0045] Based on the mobility management demand prediction benchmark, the time series analysis results and the dominant factors of slice allocation are heterogeneously fused to obtain the prediction results of the wireless network.

[0046] In a preferred embodiment, when the joint optimization decision generation module performs joint optimization decision-making on the standard network slice configuration strategy of the wireless network based on the prediction results to obtain the network slice adjustment strategy of the wireless network, it is specifically used for:

[0047] The prediction results are analyzed for potential quality risks to obtain the predicted values ​​of service quality risk and network slice performance requirements of the wireless network.

[0048] The resource configuration templates and scheduling rules of the slice instances in the wireless network are integrated into the standard network slice configuration strategy of the wireless network;

[0049] Based on the predicted performance requirements and the standard network slicing configuration strategy, a gap analysis is performed on the resource requirements and configuration supply of the wireless network to obtain the adjustment requirements information of the wireless network.

[0050] Based on the adjustment requirement information, the standard network slice configuration strategy is modified with constraints to obtain the network slice adjustment strategy for the wireless network.

[0051] In a preferred embodiment, when the policy verification and execution module performs feasibility verification on the network slice adjustment policy and executes the verified slice adjustment policy to achieve dynamic adjustment and resource allocation of the network slices of the wireless network, it is specifically used for:

[0052] The network slice adjustment strategy is subjected to policy conflict detection to confirm the slice adjustment strategy of the network slice adjustment strategy;

[0053] The slice adjustment strategy is simulated in the isolated test environment of the wireless network, and network performance indicators and alarm information are collected during the simulation process to obtain a verification feedback report of the wireless network.

[0054] Based on the verification feedback report, the execution risk and expected benefits of the slice adjustment strategy are evaluated, and the strategy is optimized and adjusted to obtain the target slice adjustment strategy for the wireless network.

[0055] The target slice adjustment strategy is encoded and encapsulated to obtain the slice reconfiguration instruction sequence of the target slice adjustment strategy;

[0056] The network function entities of the wireless network are issued execution instructions according to the slice reconfiguration instruction sequence, and the network status is monitored in real time to confirm the completion of the dynamic adjustment and resource allocation of the network slice.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. This invention achieves deep integration and precise quantification of multi-dimensional features of mobility management through multi-module collaborative operation, constructs a comprehensive feature dataset that fits business needs, accurately judges the network slice status by combining a scientific evaluation model, and establishes a reliable prediction benchmark based on historical data and expected needs. Through multi-dimensional trend analysis, it obtains accurate prediction results, significantly improving the accuracy of network slice status assessment and the foresight of business demand prediction. It can provide precise service support for differentiated services and ensure the stability and adaptability of service quality.

[0059] 2. This invention achieves targeted optimization of network slice configuration strategies through a joint optimization decision-making mechanism. Combined with rigorous strategy conflict detection, resource compatibility checks, and simulation verification processes, it ensures the feasibility and effectiveness of adjustment strategies. Furthermore, through precise command issuance and real-time network status monitoring, it completes dynamic adjustments and resource allocation, significantly improving the timeliness of network slice adjustments and the rationality of resource allocation. This effectively improves network resource utilization, ensures the high efficiency and stability of network operation, and provides a solid network guarantee for the continuous development of diversified services in low-altitude economic scenarios. Attached Figure Description

[0060] Figure 1 A system architecture diagram of a 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy provided in an embodiment of the present invention;

[0061] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0064] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0065] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0066] In practice, the server-side equipment deployed in a 5G-A network slice dynamic adjustment and resource allocation system for the low-altitude economy may consist of one or more devices. This system can be implemented as a mobility management instance, a virtual machine, or hardware devices. For example, it can be implemented as a mobility management instance deployed on one or more devices in a cloud node. Simply put, this system can be understood as software deployed on a cloud node to provide a 5G-A network slice dynamic adjustment and resource allocation system for the low-altitude economy to various user terminals. Alternatively, it can also be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage various user terminals. Alternatively, this 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy can also be implemented as a server consisting of many identical or different types of hardware devices, with one or more hardware devices set up to provide each user terminal with a 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy.

[0067] In terms of implementation, a 5G-A network slicing dynamic adjustment and resource allocation system for the low-altitude economy and the user terminal are mutually compatible. Specifically, if the 5G-A network slicing dynamic adjustment and resource allocation system for the low-altitude economy is implemented as an application installed on a cloud service platform, then the user terminal acts as a client establishing a communication connection with that application; or if the 5G-A network slicing dynamic adjustment and resource allocation system for the low-altitude economy is implemented as a website, then the user terminal acts as a webpage; or if the 5G-A network slicing dynamic adjustment and resource allocation system for the low-altitude economy is implemented as a cloud service platform, then the user terminal acts as a mini-program within an instant messaging application.

[0068] like Figure 1 The diagram shown is a system architecture diagram of a 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy provided by an embodiment of the present invention.

[0069] The 5G-A network slicing dynamic adjustment and resource allocation system 100 for low-altitude economy described in this invention can be set up in a cloud server. In terms of implementation, it can be implemented as one or more service devices, or as an application installed in the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed as a website. Depending on the implemented functions, the 5G-A network slicing dynamic adjustment and resource allocation system 100 for low-altitude economy may include a mobility management feature integration module 101, a network slice service level agreement evaluation module 102, a mobility management demand prediction benchmark construction module 103, a multi-dimensional trend prediction module 104, a joint optimization decision generation module 105, and a strategy verification and execution module 106. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0070] In this embodiment of the invention, in a 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy, each of the above modules can be implemented independently and can call other modules. This calling can be understood as a module connecting to multiple modules of another type and providing corresponding services to those connected modules. In this embodiment of the invention, the 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy can adjust its applicability without modifying the program code by adding modules and directly calling them, achieving cluster-based horizontal expansion. This allows for quick and flexible expansion of the 5G-A network slice dynamic adjustment and resource allocation system for low-altitude economy. In practical applications, the above modules can be set in the same device or different devices, or in virtual devices, such as service instances in a cloud server.

[0071] The following describes, with reference to specific embodiments, each component and specific workflow of a 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy:

[0072] The mobility management feature integration module 101 is used to integrate the real-time mobility management type, mobility management priority, latency sensitivity and bandwidth requirements in the wireless network into the mobility management feature dataset of the wireless network.

[0073] In this embodiment of the invention, when the mobility management feature integration module integrates the real-time mobility management type, mobility management priority, latency sensitivity, and bandwidth requirements in the wireless network into a mobility management feature dataset for the wireless network, it is specifically used for:

[0074] Extract the real-time mobility management type parameters and corresponding mobility management priority policy parameters from the wireless network;

[0075] The real-time mobility management type parameter and the mobility management priority policy parameter are subjected to network protocol quantization to obtain the mobility management policy vector of the wireless network;

[0076] The network slice selection auxiliary information and service subscription data of the wireless network are used to perform performance quality screening to obtain the latency sensitivity level parameters and bandwidth requirement parameters of the wireless network.

[0077] The mobility management policy vector, the latency sensitivity level parameter, and the bandwidth requirement parameter are correlated in multiple dimensions to obtain the mobility management feature dataset of the wireless network.

[0078] By determining the coverage area of ​​the wireless network and all terminal devices connected to the network, the movement status of each terminal device in the network is monitored in real time to clarify the real-time mobility management type. At the same time, the service level agreement and network resource allocation rules pre-defined by the network operator are retrieved. Based on the service assurance standards corresponding to different mobility management types, the mobility management priority policy uniquely corresponding to each mobility management type is matched and extracted one by one. This ensures that the extracted real-time mobility management type parameters and mobility management priority policy parameters form an accurate one-to-one correspondence. All extracted data comes from the terminal status information transmitted in real time by the network and the preset service agreement files, with no omissions or incorrect matches. Referring to the parameter quantification standards explicitly defined in the communication protocols used by wireless networks, the extracted real-time mobility management type parameters are fixedly identified according to the type classification coding rules defined in the protocol. For example, the mobility management type for stationary states is identified as "QZ-Stationary", low-speed mobility states as "DS-Low-Speed", and high-speed mobility states as "GS-High-Speed". At the same time, the mobility management priority policy parameters are converted into corresponding fixed description intervals according to the priority level division standards set in the protocol. For example, the highest priority corresponds to the "optimal protection level" description interval, the second highest priority corresponds to the "second best protection level" description interval, and so on. Each priority policy corresponds to a unique description interval. Then, the identified mobility management type characters and the converted priority description intervals are combined in an orderly manner according to the order of "type identifier - priority description interval" to form an ordered data set with a clear correspondence. This set is the mobility management policy vector of the wireless network.Collect selection assistance information for all configured network slices in the wireless network, including the service scenario, resource configuration limits, and performance guarantee indicators for each slice. Simultaneously, retrieve the service subscription data for each access terminal device, covering information such as the type of subscribed service, service quality commitments, and the permitted network resource range. Based on preset performance quality screening criteria, first extract the response time limit indicator related to data transmission latency and the maximum transmission rate indicator related to data transmission capacity from the network slice selection assistance information. Then, combine this with the agreed service latency tolerance threshold and bandwidth usage requirements in the service subscription data to compare and analyze the network slice assistance information for each service. The system checks whether the performance indicators in the information meet the contract requirements, removes the slice auxiliary information and corresponding service contract data that do not meet the latency threshold or bandwidth range, and retains the information data that meets the requirements. From the retained information data, it clarifies the latency allowable range for each service and divides it into four specific latency sensitivity levels: extremely high sensitivity, high sensitivity, medium sensitivity, and low sensitivity. At the same time, it determines the basic bandwidth value and maximum carrying bandwidth value required for the normal operation of each service, integrates these two bandwidth values ​​to form the bandwidth requirement parameter, and ensures that the selected latency sensitivity level parameter and bandwidth requirement parameter can accurately reflect the actual operation requirements of the service and the performance support capability of the network slice. Using the unique identifier of the terminal device accessing the wireless network as the association benchmark, the characteristic dimensions of the mobility management policy vector, latency sensitivity level parameter, and bandwidth requirement parameter are first defined. The characteristic dimensions of the mobility management policy vector include mobility type identifier and priority description range; the characteristic dimension of the latency sensitivity level parameter is the specific level category; and the characteristic dimensions of the bandwidth requirement parameter are the base bandwidth value and the maximum bearer bandwidth value. Then, the characteristics of each feature in the mobility management policy vector corresponding to each terminal device, the level category of the latency sensitivity level parameter, and the base bandwidth value and maximum bearer bandwidth value of the bandwidth requirement parameter are associated according to the principle of "unique terminal". The mapping relationship between "identifier - mobility management policy vector - latency sensitivity level parameter - bandwidth requirement parameter" is bound to ensure that there are no cross-correlation errors among the data characteristics of the same terminal device. Finally, all the bound data are integrated and arranged in the order of the terminal device accessing the network to form a data set containing complete mobility management-related characteristics of each terminal device. This data set is the mobility management feature dataset of the wireless network. Each piece of data in the dataset contains four complete pieces of information: unique terminal identifier, mobility management policy vector, latency sensitivity level parameter, and bandwidth requirement parameter, and the relationship between each piece of information is clear and unambiguous.

[0079] The beneficial effects include accurately extracting the mobility management type and corresponding priority policy of wireless network terminals, forming a clearly defined mobility management policy vector through quantitative methods according to communication protocol specifications, and obtaining latency sensitivity level parameters and bandwidth requirement parameters that meet the actual needs of services by combining network slicing selection auxiliary information and service subscription data. Through precise correlation and integration of multi-dimensional features, a complete mobility management feature dataset is formed, ensuring that data extraction is complete and matching is error-free, and that the correlation between various parameters is clear and unambiguous. This dataset can accurately reflect the core needs of terminal mobility status and service operation, providing comprehensive and reliable data support for efficient mobility management of wireless networks.

[0080] The network slice service level protocol evaluation module 102 is used to evaluate the real-time load status and wireless channel quality parameters of the wireless network based on the mobility management feature dataset, and obtain the network slice status evaluation result of the wireless network.

[0081] In this embodiment of the invention, the network slicing service level protocol evaluation module performs a network slicing service level protocol evaluation on the real-time load status and wireless channel quality parameters of the wireless network based on the mobility management feature dataset, and obtains the network slicing status evaluation result of the wireless network, specifically for:

[0082] Obtain the real-time load status of the wireless network;

[0083] The channel spectrum parameters captured in the wireless network are used as the wireless channel quality parameters of the wireless network.

[0084] The real-time load status and the wireless channel quality parameters are time-fused and aligned to obtain a snapshot of the network operation status of the wireless network.

[0085] Based on the latency, bandwidth, and reliability indicators defined in the network slicing service level protocol of the wireless network, a matching degree analysis is performed on the mobility management feature dataset and the network operation status snapshot to obtain the sub-item evaluation results of the mobility management feature dataset.

[0086] Based on preset evaluation weight rules, the sub-evaluation results are weighted and aggregated to obtain the network slice status evaluation result of the wireless network.

[0087] The characteristic feature is that the calculation formula for the network slice state evaluation result is as follows:

[0088] ;

[0089] In the formula, The network slice state evaluation result, For the mobility management feature dataset, the mobility management type is... For a specific mobility management type among the mobility management types, The type weight coefficient is the pre-defined evaluation weight rule. This is the balance coefficient for the service demand matching degree in the evaluation weighting rules. For the specific mobility management type The service quality requirement vector, This is the balance coefficient for network state stability in the evaluation weighting rule. For the specific mobility management type The service capability vector that actually guarantees business operations. This is the network operating state vector in the network operating state snapshot. For the type in the evaluation weighting rule The ideal state threshold vector.

[0090] By monitoring the resource usage of network devices such as the core gateway and access points of the wireless network, including the proportion of processor usage, actual memory usage, data transmission queue length, and port data throughput, and collecting real-time status information reported by each network device, which covers the current working status, resource allocation, and data flow efficiency of the device, all collected device resource usage data and status information are summarized and integrated to comprehensively reflect the current resource carrying capacity of the wireless network, and finally form a real-time load status of the wireless network that can intuitively reflect the current status of network resource usage.

[0091] By deploying spectrum monitoring equipment within the coverage area of ​​the wireless network, all channels occupied by the wireless network are continuously scanned. Various information related to the channel spectrum, such as signal strength, signal attenuation, interference signal strength, and actual utilization of spectrum resources, are captured during the transmission process of each channel. These captured channel spectrum-related information are directly determined as wireless channel quality parameters that can characterize the transmission quality of the wireless network, ensuring that the parameters fully cover the core quality characteristics during the channel transmission process.

[0092] First, a unified time base is established, using the standard time commonly used by wireless networks as the time reference. The collection time points of real-time load status and wireless channel quality parameters are recorded separately. For data with completely consistent collection and capture time points, they are directly associated and bound. For data with time differences, the collection data closest to the capture time point of the wireless channel quality parameters in the real-time load status is matched according to the principle of the shortest time interval. This ensures that each set of wireless channel quality parameters corresponds to the most appropriate real-time load status data. Through this precise matching and integration in the time dimension, a complete data set that reflects the wireless network resource status and channel quality at a specific point in time is formed. This data set is the snapshot of the network operation status of the wireless network.

[0093] The network slicing service level agreement (SSLA) signed in the wireless network is retrieved to clarify the core indicators set for the network slice, such as latency cap standards, bandwidth guarantee standards, and data transmission reliability standards. Latency sensitivity level parameters and bandwidth requirement parameters are extracted from the mobility management feature dataset. Corresponding information such as actual data transmission latency, actual available bandwidth, and data transmission success probability are extracted from network operation status snapshots. The latency sensitivity level parameters in the mobility management feature dataset are compared with the latency cap standards stipulated in the agreement and the actual data transmission latency in the network operation status snapshots to determine whether the actual latency meets the agreement standards and its compatibility with latency sensitivity. Similarly, the actual available bandwidth is compared with the bandwidth guarantee standards of the agreement and the bandwidth requirement parameters in the mobility management feature dataset, and the data transmission success probability is compared with the reliability standards stipulated in the agreement. The comparative analysis results for each indicator dimension are recorded separately, forming independent evaluation conclusions for the mobility management feature dataset in the three dimensions of latency, bandwidth, and reliability. These independent evaluation conclusions together constitute the sub-evaluation results of the mobility management feature dataset.

[0094] The preset evaluation weighting rules are pre-defined by network operators based on the importance and service requirements of the services carried by the network slices. These rules determine the proportion of three sub-evaluation indicators—latency, bandwidth, and reliability—in the overall evaluation. According to these rules, the evaluation results for the latency dimension are multiplied by their corresponding latency weight, the evaluation results for the bandwidth dimension are multiplied by their corresponding bandwidth weight, and the evaluation results for the reliability dimension are multiplied by their corresponding reliability weight. The weighted results of the three dimensions are then summed. This weighted summation method aggregates and integrates all sub-evaluation results, ultimately yielding a network slice status evaluation result that comprehensively reflects the degree to which the network slice service level protocol is met.

[0095] The mobility management types in the mobility management feature dataset are determined by sorting out the service scenarios carried by network slices. During the sorting process, all service types accessing the network slices are collected, and the core management requirements of each service during the mobility process are analyzed, such as handover management, location update management, and session continuity management. Services with the same management requirements are grouped into one category, and the specific management behaviors and technical indicators corresponding to each type of management requirement are clarified. Finally, a set of mobility management types containing multiple independent types is formed, and each type corresponds to a specific mobility management scenario.

[0096] The type weight coefficients in the preset evaluation weight rules are determined based on the service importance calibration of various mobility management types. Historical data of network slicing operation are collected, and the service volume ratio, service interruption loss and user experience sensitivity corresponding to each mobility management type are statistically analyzed. These indicators are quantified into importance scores, and the importance scores of all types are normalized so that the sum of all weight coefficients is a fixed benchmark. The higher the importance score of a type, the larger the corresponding type weight coefficient, ensuring that the evaluation results of important types account for a higher proportion in the overall evaluation.

[0097] The balance coefficient for service demand matching in the evaluation weighting rules is obtained by calibrating with historical evaluation data. A large number of network slice status evaluation cases are collected, and the deviation between the evaluation results and the actual network operation effect under different balance coefficient values ​​is recorded. The average evaluation error corresponding to each coefficient is calculated, and the coefficient with the smallest average evaluation error is selected as the balance coefficient for service demand matching. This coefficient is used to adjust the influence of the service demand matching part in the overall evaluation result.

[0098] The service quality requirement vector for a specific mobility management type is extracted and quantified from the service level agreement. During extraction, the standard requirements for the corresponding service in terms of latency, bandwidth, packet loss rate, reliability, etc. are clearly defined. The requirements of each dimension are converted into specific values ​​and arranged in a preset fixed order to form an ordered set of values. This set is the service quality requirement vector for a specific mobility management type. Each component of the vector corresponds to a standard value of a service quality dimension.

[0099] The balance coefficient for network state stability in the evaluation weighting rules is determined based on historical network operation data. Evaluation samples under different network load scenarios are selected to test the impact of different balance coefficient values ​​on the evaluation results. The degree of fit between the evaluation results and the actual network stability is compared, and the fit index is calculated. The coefficient with the highest fit is selected as the balance coefficient for network state stability. This coefficient is used to balance the evaluation weights of the service demand matching part and the network state stability part.

[0100] The service capability vector for the actual service assurance of a specific mobility management type is obtained by real-time monitoring of network performance. During monitoring, dedicated performance monitoring nodes are deployed for each dimension of the service quality requirement vector corresponding to that type. The actual performance data of the network in that dimension is continuously collected. After removing outliers, the average value is taken as the actual capability value of that dimension. These actual capability values ​​are arranged in the order of the components of the service quality requirement vector to form the service capability vector for the actual service assurance. The vector components correspond one-to-one with the components of the service quality requirement vector.

[0101] The network operation status vector in the network operation status snapshot is formed by extracting and quantifying snapshot data. The network operation status snapshot is a collection of network operation data collected at a fixed time point. Core indicators related to specific mobility management types, such as current network load, resource utilization, and link status, are extracted from the snapshot. These indicators are quantified into specific values ​​and arranged in a preset order to form a network operation status vector. This vector intuitively reflects the actual operation level of the network at the snapshot time.

[0102] The ideal state threshold vector for each type in the evaluation weighting rules is determined based on industry standards and business requirements. Referring to the standard specifications for network slicing operation in the communications industry and combining the optimal operating scenario for a specific mobility management type, the ideal values ​​for each dimension of the network operating state vector are determined. These ideal values ​​are arranged in the order of the components of the network operating state vector to form the ideal state threshold vector for each type. This vector serves as a benchmark for measuring whether the network state is stable.

[0103] The network slice status assessment result is a quantitative indicator that comprehensively measures the service capabilities and operational status of network slices under various mobility management types. Its value directly reflects the overall operational quality of the network slice.

[0104] For each type of mobility management, the service quality requirement vector is multiplied by the corresponding components of the service capability vector that the business actually guarantees, and then summed to obtain the service demand matching degree for that type. The larger the value, the higher the fit between the service capability and the business needs. This value is then multiplied by the balance coefficient of the service demand matching degree to determine the specific contribution of this part in the evaluation.

[0105] The deviation between the network state and the ideal state threshold vector is obtained by multiplying the corresponding components of the network state vector and summing them. The deviation is then subtracted from the fixed benchmark value to obtain the network state stability index. The larger the value, the more stable the network operation. This index is multiplied by the balance coefficient of network state stability to form the evaluation contribution of the stability component.

[0106] The contribution of service demand matching is added to the contribution of network state stability to obtain the evaluation sub-result for a single mobility management type. This sub-result is then multiplied by the weight coefficient of that type to highlight the evaluation impact of important types. Finally, the weighted evaluation sub-results of all types are summed to obtain the network slice state evaluation result.

[0107] The higher the evaluation result value, the better the overall performance of network slices in terms of service matching and operational stability under various mobility management scenarios, and the more reliable the support for services. The lower the value, the more likely the network slice has insufficient service capabilities or unstable operation. This indicator provides a precise quantitative basis for the resource adjustment, optimization and upgrading of network slices.

[0108] The beneficial effects include comprehensively acquiring resource occupancy and status information of core wireless network devices, accurately determining wireless channel quality parameters that characterize channel transmission quality, achieving precise time fusion and alignment of real-time load status and wireless channel quality parameters through a unified time base, forming a complete snapshot of network operation status reflecting the network operation status at a specific point in time, and conducting targeted matching degree analysis on mobility management feature datasets and network operation status snapshots based on core indicators agreed upon in the network slicing service level agreement. This yields independent sub-item evaluation results for each dimension, which are then weighted and aggregated according to preset reasonable evaluation weight rules. By comprehensively considering the importance of each sub-item indicator, a comprehensive and accurate evaluation result reflecting the degree of compliance with the network slicing service level agreement is obtained. This provides a reliable basis for subsequent network slicing adjustments and optimizations, and dynamic resource allocation for the wireless network, ensuring accurate adaptation of network slicing service quality to business operation requirements.

[0109] The mobility management demand prediction benchmark construction module 103 is used to construct the mobility management demand prediction benchmark of the wireless network based on the mobility management feature dataset and the expected demand and historical slice status information database of the wireless network.

[0110] In this embodiment of the invention, when the mobility management demand prediction benchmark construction module constructs the mobility management demand prediction benchmark for the wireless network based on the mobility management feature dataset and the expected demand and historical slice status information database of the wireless network, it is specifically used for:

[0111] The expected business growth information and service quality targets for low-altitude economic scenarios in the wireless network are integrated into the network planning strategy data of the wireless network.

[0112] Based on the historical slice status information database in the wireless network, the mobility management feature dataset is traversed and analyzed to obtain the joint feature sequence of the wireless network;

[0113] Based on the expected business growth information, the joint feature sequence is extrapolated to obtain the initial prediction of the mobility management feature dataset;

[0114] Using the network planning strategy data as constraints, the initial prediction is corrected for compliance, resulting in the standard prediction of the mobility management feature dataset.

[0115] The standard predicted quantity is correlated and mapped with the key performance indicator benchmark values ​​in the historical slice status information database to obtain the mobility management demand prediction benchmark of the wireless network.

[0116] When the mobility management demand forecasting benchmark construction module performs trend extrapolation on the joint feature sequence based on the expected business growth information to obtain the initial forecast amount of the mobility management feature dataset, it is specifically used for:

[0117] The expected business growth information is separated into business data to obtain the business volume growth data and the expected introduction characteristics of new businesses.

[0118] The joint feature sequence is periodically decomposed to obtain the long-term trend component and periodic fluctuation component of the joint feature sequence.

[0119] Based on the long-term trend components and the business volume growth data, the joint feature sequence is extrapolated over time to obtain the basic feature prediction values ​​of the mobility management feature dataset.

[0120] By combining the periodic fluctuation component with the expected introduction features of new services, the predicted values ​​of the basic features are superimposed and corrected to obtain the initial predicted values ​​of the mobility management feature dataset.

[0121] When the mobility management demand forecasting baseline construction module performs the association mapping between the standard forecast quantity and the key performance indicator baseline values ​​in the historical slice status information database to obtain the mobility management demand forecasting baseline of the wireless network, it is specifically used for:

[0122] Based on the historical slice status information database, the historical key performance index data associated with the features in the standard prediction quantity are matched and filtered to obtain the historical key index benchmark data of the standard prediction quantity.

[0123] The feature items related to mobility management type, priority and service quality requirements in the standard predicted quantity are matched and mapped with the historical key indicator benchmark data to establish a feature benchmark association table of the standard predicted quantity.

[0124] Based on the feature benchmark association table, the service quality requirement features in the standard prediction quantity are quantified for network slicing requirements to obtain the target values ​​of the key performance indicators of the wireless network.

[0125] The target values ​​of the key performance indicators are integrated with the mobility management type and priority characteristics in the standard forecasts to form the mobility management demand forecasting benchmark for the wireless network.

[0126] This process involves collecting information on various services offered by wireless networks for the low-altitude economic scenario, including the types of services planned for access, the expected low-altitude coverage area for each service, the planned scale of terminal devices to be accessed, and the timeframe for service implementation. Using market demand survey data, industry trend analysis reports, and network operators' business expansion plans, the growth scale and rate of growth for each service type in the future are determined, forming anticipated business growth information. Simultaneously, the process retrieves the service quality requirements preset by network operators for the low-altitude economic scenario, covering latency tolerance standards, bandwidth guarantee standards, data transmission reliability standards, and continuous operation stability requirements for business data transmission. The growth characteristics of each service in the anticipated business growth information are then mapped one-to-one with the corresponding service quality targets by service type, clarifying the service quality requirements that services of different growth scales need to meet. Finally, this data is integrated to form structured network planning strategy data for the wireless network.

[0127] The historical slice status information database of the wireless network is reviewed. This database contains operational status data, mobility management feature records, service quality performance data, and network resource allocation information for network slices in different time periods. Historical mobility management feature data with the same structure as the current mobility management feature dataset is extracted from the historical slice status information database. Each feature in the current mobility management feature dataset is compared one by one according to the category of mobility management feature. All relevant historical data in the historical slice status information database are traversed to find historical feature records that are related to the current features. The current features and corresponding historical features are combined in an orderly manner according to time sequence and service type to form a continuous data sequence containing the current features and related historical features. This data sequence is the joint feature sequence of the wireless network.

[0128] Based on the clearly defined business growth scale and growth rate in the expected business growth information, we analyze the changing patterns of various features in the joint feature sequence as the business grows. For example, when the business scale expands, we observe the growth trend of bandwidth requirements and the changing tendency of latency sensitivity level parameters. Combining the expected business growth time cycle, we extend the joint feature sequence to the future time dimension according to the identified changing patterns, and infer the possible values ​​or states of various features at different future time points. This ensures that the extended feature data matches the rhythm and scale of the expected business growth. We then integrate these inferred future feature data to form the initial prediction of the mobility management feature dataset.

[0129] Clearly define the various quality of service constraints included in the network planning strategy data, including core requirements such as latency caps, minimum guaranteed bandwidth, and minimum reliability standards. Compare each characteristic data in the initial forecast with its corresponding constraint, checking whether the latency-related data in the initial forecast exceeds the latency cap in the constraint, whether the bandwidth-related data is lower than the minimum guaranteed bandwidth in the constraint, and whether the reliability-related data does not meet the minimum reliability standard in the constraint. For initial forecast data that exceeds the constraint range, adjust it according to the requirements of the constraint. For example, correct forecasts that exceed the latency cap to the latency cap value in the constraint, and increase forecasts that are lower than the minimum guaranteed bandwidth value to the minimum guaranteed bandwidth value. Ensure that the adjusted forecasts fully comply with the constraint requirements of the network planning strategy data, and finally form the standard forecasts for the mobility management feature dataset.

[0130] The key performance indicator benchmarks for network slices during normal and stable operation are extracted from the historical slice status information database of the wireless network. These benchmarks include benchmark latency, benchmark bandwidth, and benchmark reliability. These benchmarks are derived from statistical data of the time period during which the network slices met the quality of service requirements and operated at their optimal state. Each characteristic data in the standard prediction is matched with the corresponding key performance indicator benchmark to determine the correspondence and deviation range between each standard prediction and the benchmark. For example, the benchmark bandwidth range corresponding to the bandwidth prediction and the difference range between the latency prediction and the benchmark latency. The correlation results of all standard predictions and their corresponding benchmarks are integrated to form a reference standard that can clearly define the correspondence between future mobility management requirements and historical best benchmarks. This reference standard is the mobility management requirement prediction benchmark for the wireless network.

[0131] We analyze all data content included in the expected business growth information, identifying various types of information related to the expansion of existing business scale, including the growth in the number of existing business users, the increase in total data transmission volume, and the increasing trend of business usage frequency. This data, which directly reflects changes in the volume of existing business, is categorized separately to form the business volume growth data for the expected business growth information. Simultaneously, we filter out business-related content not currently deployed in the wireless network from the expected business growth information. This includes information that reflects the uniqueness of the new business, such as its type attributes, service scenario characteristics, data transmission mode, and special network resource requirements. We integrate this information to form the expected introduction characteristics of the new business, ensuring that the two types of data are completely separate and accurately correspond to their respective definitions during the business data separation process.

[0132] By deeply analyzing the patterns of change of each feature in the joint feature sequence over time, and comparing the changes in feature data in different historical time periods, the recurring temporal patterns in feature changes are identified, such as regular fluctuations at fixed time intervals or cycles. The parts of the joint feature sequence that conform to these recurring patterns are extracted to form the periodic fluctuation component of the joint feature sequence. At the same time, after removing the periodic fluctuation component, the remaining feature data that shows an overall trend of continuous increase, continuous decrease, or stabilization is extracted. This part of the data can reflect the core trend of the feature in the long-term time dimension, which is the long-term trend component of the joint feature sequence. The decomposition process strictly follows the temporal correlation of the feature data to ensure that both types of components completely cover all the information of the original joint feature sequence.

[0133] Using the time axis as a benchmark, the characteristic change pattern reflected by the long-term trend component is combined with the business expansion speed reflected by the business volume growth data to clarify the correlation logic between the two. For example, when the business volume growth data shows that the business scale continues to expand, the bandwidth requirement-related features in the long-term trend component show a synchronous upward trend. According to this correlation logic, following the time order of the original joint feature sequence, the long-term trend component is extended to the future time dimension. During the extension process, the growth rate of the business volume growth data is referenced to ensure that the extended trend matches the business volume growth rate, while maintaining the continuity and rationality of the feature data. Through this time-series extension inference, the preliminary prediction results of various features at different future time points are obtained. This result is the basic feature prediction value of the mobility management feature dataset.

[0134] First, the regular variation patterns reflected by the periodic fluctuation components are superimposed onto the predicted values ​​of basic features according to their corresponding time periods. For example, if the periodic fluctuation components show that latency sensitivity will increase in stages during specific periods of the day, the increase during those periods is superimposed onto the corresponding time nodes of the predicted values ​​of basic features, so that the predicted values ​​reflect periodic fluctuation characteristics. Next, the impact of the expected introduction of new services on the predicted values ​​of various basic features is analyzed. For example, if the expected introduction of new services includes higher bandwidth requirements, the predicted values ​​of the corresponding bandwidth-related basic features are adjusted according to the degree of these requirements. If the new services have higher requirements for latency sensitivity, the predicted values ​​of the latency sensitivity-related basic features are increased accordingly. Through the superposition of periodic fluctuation patterns and the adjustment of new service introduction requirements, the predicted values ​​of basic features are comprehensively corrected, and finally, the initial predicted values ​​of the mobility management feature dataset that integrates long-term trends, periodic fluctuations, and new service requirements are obtained.

[0135] This study delves into all feature types included in the standard predictive quantity, clarifying the specific attributes and representations of core features such as mobility management policy vectors, latency sensitivity level parameters, and bandwidth requirement parameters. It then traverses the historical slice state information database of the wireless network, which stores key performance indicator records under different mobility management scenarios during past network operations. This includes latency performance data, bandwidth support data, and reliability compliance data corresponding to various features. Based on the feature attributes of the standard predictive quantity, historical key performance indicator data that are completely consistent with or highly compatible with these features in terms of scenario type, service attributes, and network environment conditions are precisely matched and filtered from the historical slice state information database. Simultaneously, historical data unrelated to the standard predictive quantity features or mismatched with scenarios are removed, ensuring that the filtered historical data accurately reflects the optimal past operating state of the corresponding features. This process ultimately forms the historical key indicator benchmark data for the standard predictive quantity.

[0136] The standard forecast quantities extract three core feature items: mobility management type, mobility management priority, and service quality requirements. Mobility management type covers specific types such as stationary, low-speed, and high-speed mobility. Mobility management priority includes different levels such as highest, second highest, medium, and low. Service quality requirements involve related requirements such as latency, bandwidth, and reliability. At the same time, the historical feature information corresponding to each data item in the historical key indicator benchmark data is sorted out to establish a one-to-one correspondence between feature items and historical key indicator benchmark data. For example, the mobility management type of "high-speed mobility" in the standard forecast quantity is bound to the key performance indicator data of high-speed mobility scenarios in the historical key indicator benchmark data. The "highest" priority is bound to the benchmark data corresponding to the highest priority in history. Specific latency requirements are bound to the benchmark data that meets the latency requirements in history. All binding relationships are organized in a unified format to form a structured table containing feature item names, historical key indicator benchmark data content, and correspondence descriptions. This table is the feature benchmark association table of the standard forecast quantity.

[0137] Based on the feature benchmark association table, we first clarify the specific range and performance level of the historical key indicator benchmark data corresponding to the service quality requirement features in the standard prediction quantity. For example, the feature benchmark association table shows that the historical latency benchmark data corresponding to a certain service quality requirement is stable at a low level, and the bandwidth benchmark data continuously meets the high transmission demand. Combining the current network planning strategy data of the wireless network for low-altitude economic scenarios, and referring to the expected business growth information on the service quality improvement demand, we specifically quantify and define the service quality requirement features. The latency requirement is quantified as a clear transmission time upper limit, the bandwidth requirement is quantified as the minimum guaranteed bandwidth and peak carrying bandwidth, and the reliability requirement is quantified as the minimum proportion of successful data transmission. This ensures that the quantified indicators are based on the stable performance of historical benchmark data and fully adapt to the current and future business operation needs. These quantified specific indicators are the key performance indicator target values ​​of the wireless network.

[0138] The key performance indicator (KPI) target values ​​are systematically integrated with the mobility management type and priority characteristics in the standard forecast. First, mobility management type is used as the core classification dimension, and mobility management priority characteristics and KPI target values ​​corresponding to the same mobility management type are grouped into one category. Then, within each category, they are sorted in descending order of mobility management priority, clarifying the differences in KPI target values ​​corresponding to different priorities. This ensures that the integrated content is logically clear and hierarchically distinct, containing both the core types and priority information of mobility management and clarifying the corresponding quantitative service quality targets. Ultimately, this forms a reference standard that can comprehensively guide the forecasting of mobility management needs in wireless networks. This reference standard serves as the benchmark for forecasting mobility management needs in wireless networks.

[0139] The beneficial effects include accurately integrating expected business growth information and service quality targets in low-altitude economic scenarios to form network planning strategy data that meets the needs; combining historical slice status information database to conduct traversal analysis of mobility management feature datasets to obtain joint feature sequences containing current and related historical features; extrapolating initial forecasts based on business growth trends; and correcting them with network planning strategy data as constraints to form compliant standard forecasts. By associating historical key performance indicator benchmarks, a mobility management demand forecasting benchmark is constructed to ensure that the forecasting benchmark not only conforms to business growth patterns but also meets service quality requirements. This provides an accurate and reliable reference for wireless network mobility management demand forecasting, ensuring the scientific and practical nature of subsequent forecast results.

[0140] By accurately separating expected business growth information, the core data of business volume growth and the unique characteristics of new businesses are clearly identified, providing a precise basis for subsequent forecasts. The joint feature sequence is effectively decomposed to clarify long-term trends and cyclical fluctuations, ensuring that the logic of feature changes is clear. By combining long-term trends and business volume growth data for time-series extrapolation, the predicted values ​​of basic features closely match the actual business growth. The basic predicted values ​​are corrected by incorporating cyclical fluctuations and new business needs, so that the initial predicted values ​​not only reflect historical changes but also adapt to the needs of new business introduction. Finally, a comprehensive and accurate initial predicted value of the mobility management feature dataset is obtained, providing reliable data support for the construction of subsequent forecast benchmarks.

[0141] Based on a historical slice status information database, historical key indicator benchmark data that highly matches the characteristics of standard prediction quantities is precisely matched and filtered. Irrelevant and mismatched data are eliminated to ensure the relevance and reliability of the benchmark data. By establishing a clear correlation mapping relationship between mobility management type, priority, and service quality requirement features and historical key indicator benchmark data, a structured feature benchmark association table is formed, providing a clear basis for subsequent quantification work. Supported by the feature benchmark association table, the service quality requirement features are accurately quantified in combination with network planning strategies and business growth needs, resulting in key performance indicator target values ​​that fit actual needs. Finally, the key performance indicator target values ​​are systematically integrated with mobility management type and priority features to form a logically clear, hierarchical mobility management demand prediction benchmark that covers core management elements and quantified quality objectives. This provides a comprehensive and reliable reference for the accurate prediction of wireless network mobility management needs, ensuring the scientific nature and accuracy of subsequent prediction work.

[0142] The multi-dimensional trend prediction module 104 is used to perform multi-dimensional trend analysis on the network slice status assessment results based on the mobility management demand prediction benchmark, and obtain the prediction results of the wireless network.

[0143] In this embodiment of the invention, when the multi-dimensional trend prediction module performs multi-dimensional trend analysis on the network slice status assessment results based on the mobility management demand prediction benchmark to obtain the prediction results of the wireless network, it is specifically used for:

[0144] The deviation of the network slice status assessment results is compared with the mobility management demand prediction benchmark to obtain the difference feature set of the wireless network;

[0145] By performing a trend extrapolation of the service quality of the network slices in the differential feature set in chronological order, the time series analysis results of the wireless network are obtained.

[0146] Based on the time series analysis results and the network slice type of the wireless network, key factors are identified in the differential feature set to determine the dominant factors for slice allocation of the wireless network.

[0147] Based on the mobility management demand prediction benchmark, the time series analysis results and the dominant factors of slice allocation are heterogeneously fused to obtain the prediction results of the wireless network.

[0148] This study deconstructs the core components of network slicing status assessment results and mobility management demand prediction benchmarks. It clarifies the actual operational characteristics included in the assessment results, such as real-time latency, actual available bandwidth, and data transmission reliability, as well as the benchmark characteristics in the prediction benchmarks, such as key performance indicator target values, mobility management type adaptation standards, and priority guarantee requirements. Following the one-to-one correspondence logic of "actual characteristics - benchmark characteristics," the study compares and analyzes the two types of characteristics one by one to determine whether the actual operational characteristics meet the requirements of the benchmark characteristics. It accurately defines the specific manifestations of differences, including actual values ​​exceeding benchmark values, actual values ​​failing to reach benchmark values, and fluctuations between actual and benchmark values. At the same time, it records the scenario conditions, time points, and associated service types where differences occur. All the confirmed differences are classified and organized according to feature categories and difference types to form a complete and detailed set of differences in wireless network features.

[0149] This study analyzes all feature data related to network slice service quality (SQW) in a feature set, covering differences in core SQW dimensions such as latency, bandwidth, and reliability. It extracts the collection time point for each data point and arranges these SQW differences in chronological order to form a continuous time series. The study analyzes the patterns of SQW differences over time, such as a continuous decrease in latency, periodic fluctuations in bandwidth, and a gradual increase in reliability. By combining this with the wireless network's service operation cycle and network resource scheduling patterns, the study predicts future SQW trends, identifying the direction of trend continuation, magnitude of change, and potential turning points. These trend analysis conclusions are then integrated with the corresponding time series data to form a time series analysis result of the wireless network that reflects the temporal changes in SQW differences.

[0150] Clearly define the type attributes of various network slices in the wireless network, including core information such as the corresponding business scenarios, service objects, resource configuration modes, and performance assurance priorities. Based on the time series analysis results, observe the degree of impact of various differences in the feature set at different time stages on the network slice operation status. For example, some differences may lead to a continuous decline in slice service quality, while others may have a smaller impact on overall operation. Combined with the performance requirements of network slice types, analyze the degree of compatibility conflict between each difference item in the feature set and the slice type. Identify the difference factors that play a decisive role in slice service quality compliance, mobility management strategy implementation, and priority assurance. By judging whether the factor directly causes a core deviation between the actual operation status and the predicted baseline, and whether it affects the implementation of core slice functions, the dominant factors of wireless network slice allocation that play a key role in the effectiveness of network slice allocation are finally determined.

[0151] Using the mobility management demand forecasting benchmark as the core reference framework, this paper clarifies the constraint range of key performance indicator target values ​​in the forecasting benchmark, the core requirements of mobility management type and priority, and adapts the service quality difference change trend reflected in the time series analysis results to this framework to determine whether the trend conforms to the expected direction set by the benchmark. At the same time, the dominant factors of slice allocation are taken into consideration, and the influence of the dominant factors on the benchmark requirements is analyzed. For example, when the dominant factor is insufficient bandwidth, the forecasting logic for future bandwidth demand needs needs to be adjusted in combination with the changes in bandwidth difference in the time series trend. Following the integration logic of "benchmark constraint - trend adaptation - dominant factor correction", the trend information in the time series analysis results, the impact information of the dominant factors of slice allocation, and the core requirements of the forecasting benchmark are systematically integrated to ensure that the integrated information not only meets the benchmark constraint standards, but also reflects trend changes and core influencing factors. Finally, the paper forms a forecasting result for the wireless network that can accurately predict the future network slice operation status and the degree of satisfaction of mobility management requirements.

[0152] The beneficial effects include: accurately deconstructing the core features of network slice status assessment results and mobility management demand prediction benchmarks; clarifying the differences, scenario conditions, and related service types between the two through one-to-one comparison, forming a detailed set of difference features to provide accurate basis for subsequent analysis; sorting service quality-related difference data in chronological order, and deducing the changing trends by combining business operation cycles and resource scheduling rules to obtain time-series analysis results reflecting the time-series change patterns, clearly grasping the evolution direction of service quality differences; combining network slice type attributes with time-series analysis results to identify the core difference factors that play a decisive role in slice service quality and the implementation of mobility management strategies, clarifying the dominant factors in slice allocation; and using the mobility management demand prediction benchmark as a constraint framework to systematically integrate time-series trends, dominant factors in slice allocation, and benchmark requirements, ensuring compliance with benchmark constraints while fully reflecting trend changes and core impacts, ultimately obtaining accurate prediction results of future network slice operation status and the degree of mobility management demand satisfaction, providing reliable decision support for dynamic adjustment and resource optimization of wireless network slices.

[0153] The joint optimization decision generation module 105 is used to perform joint optimization decision on the standard network slice configuration strategy of the wireless network based on the prediction results, so as to obtain the network slice adjustment strategy of the wireless network.

[0154] In this embodiment of the invention, when the joint optimization decision generation module performs joint optimization decision-making on the standard network slice configuration strategy of the wireless network based on the prediction results to obtain the network slice adjustment strategy of the wireless network, it is specifically used for:

[0155] The prediction results are analyzed for potential quality risks to obtain the predicted values ​​of service quality risk and network slice performance requirements of the wireless network.

[0156] The resource configuration templates and scheduling rules of the slice instances in the wireless network are integrated into the standard network slice configuration strategy of the wireless network;

[0157] Based on the predicted performance requirements and the standard network slicing configuration strategy, a gap analysis is performed on the resource requirements and configuration supply of the wireless network to obtain the adjustment requirements information of the wireless network.

[0158] Based on the adjustment requirement information, the standard network slice configuration strategy is modified with constraints to obtain the network slice adjustment strategy for the wireless network.

[0159] This study delves into the core information contained in the forecast results, such as the predicted network slice operation status and the degree of fulfillment of mobility management requirements. Focusing on key service quality indicators (SMIs) such as latency, bandwidth, and reliability, it compares the actual predicted values ​​in the forecast results with the target values ​​of key performance indicators in the mobility management requirement forecast benchmark one by one. This identifies potential risk types in the forecast results that may lead to substandard service quality, service interruption, or degraded user experience. For example, latency exceeding the benchmark limit, bandwidth failing to meet expected business growth demands, and reliability failing to meet the corresponding priority guarantee standards. At the same time, based on the business growth trend, expected user scale expansion, and changes in mobility characteristics in the forecast results, it clarifies the specific requirements for core performance indicators such as latency, bandwidth, and reliability that network slices need to achieve in future operation. Ultimately, this results in a service quality risk profile for wireless networks that clearly defines risk types, risk scenarios, and the scope of impact, as well as predicted values ​​for network slice performance requirements that clearly define future performance standards.

[0160] This process involves collecting resource configuration templates for all deployed network slice instances in the wireless network. These templates include core information such as the computing resource allocation ratio, storage resource quota, network resource occupancy limit, and resource priority configuration for different service types for each slice instance. Simultaneously, it reviews the existing network slice scheduling rules in the wireless network, covering key rules such as slice selection mechanisms, dynamic resource adjustment trigger conditions, load balancing strategies, slice switching procedures, and fault recovery scheduling schemes. Resource configuration templates and scheduling rules are categorized and matched according to the type and attributes of the network slices. Resource configuration templates and scheduling rules corresponding to the same type of slice are bound and integrated, clarifying unified standards for resource allocation and scheduling execution for different slice types. This results in a structured and standardized standard network slice configuration strategy for the wireless network, ensuring that this strategy includes the resource configuration basis and scheduling execution criteria required for the operation of all slice instances.

[0161] The core performance indicators such as latency, bandwidth, and reliability specified in the network slice performance requirement forecast are systematically compared and analyzed with the corresponding resource configuration standards and scheduling guarantee capabilities in the standard network slice configuration strategy. The analysis verifies whether the computing and network resources configured in the standard strategy can support the bandwidth requirements in the performance requirement forecast, whether the latency control mechanism in the scheduling rules can meet the latency standard in the forecast, and whether the resource fault tolerance configuration can match the reliability requirements in the forecast. This process accurately identifies specific gaps, such as resource gaps where the performance requirement forecast exceeds the standard strategy's configuration, and mismatches between the performance requirement type and the standard strategy's scheduling direction. The dimensions, degree, and corresponding slice types of these gaps are clarified. This gap information is then organized according to the logic of "performance dimension - gap manifestation - corresponding slice," forming adjustment requirement information for the wireless network with a clear direction and scope for adjustment.

[0162] Based on the adjustment demand information and considering constraints such as the total resource capacity limit of the wireless network and the normal operation requirements of other slice instances, the standard network slice configuration strategy is specifically modified. If the adjustment demand information indicates a bandwidth resource gap, the resource configuration template for the corresponding slice type is modified to increase its network bandwidth allocation ratio. At the same time, the bandwidth allocation mechanism in the scheduling rules is optimized to ensure that bandwidth resources are tilted towards high-demand slices. If the adjustment demand indicates insufficient latency guarantee, the slice switching process and data transmission path selection mechanism in the scheduling rules are adjusted to reduce latency consumption in the transmission process. At the same time, the configuration related to computing resources in the resource configuration template is corrected to improve data processing efficiency. During the modification process, the principle of "meeting performance requirements, not exceeding resource constraints, and not affecting the operation of other slices" is strictly followed. Each parameter of the resource configuration template and each process of the scheduling rules is reviewed and adjusted one by one to ensure that the modified strategy can accurately respond to all requirements in the adjustment demand information, and finally form a network slice adjustment strategy for the wireless network that adapts to the future network slice performance requirements.

[0163] The beneficial effects include: accurately identifying potential service quality risks by breaking down prediction results; clarifying future performance requirements for network slices and providing targeted support for optimization decisions; integrating slice instance resource configuration templates and scheduling rules to form a unified and standardized network slice configuration strategy, ensuring that configuration and scheduling are systematic; conducting gap analysis by combining performance requirement predictions with the standard strategy to clearly define resource supply and demand gaps and adjustment directions; and making restrictive modifications to the standard strategy with adjustment requirements as the core, while taking into account resource constraints and other slice operation requirements, ultimately obtaining a network slice adjustment strategy that adapts to future performance requirements. This ensures that service quality meets standards while achieving efficient resource utilization, providing reliable support for stable optimization of wireless networks.

[0164] The policy verification and execution module 106 is used to verify the feasibility of the network slice adjustment policy and execute the verified slice adjustment policy to realize the dynamic adjustment and resource allocation of the network slices of the wireless network.

[0165] In this embodiment of the invention, when the policy verification and execution module performs feasibility verification on the network slice adjustment policy and executes the verified slice adjustment policy to achieve dynamic adjustment and resource allocation of the network slices of the wireless network, it is specifically used for:

[0166] The network slice adjustment strategy is subjected to policy conflict detection to confirm the slice adjustment strategy of the network slice adjustment strategy;

[0167] The slice adjustment strategy is simulated in the isolated test environment of the wireless network, and network performance indicators and alarm information are collected during the simulation process to obtain a verification feedback report of the wireless network.

[0168] Based on the verification feedback report, the execution risk and expected benefits of the slice adjustment strategy are evaluated, and the strategy is optimized and adjusted to obtain the target slice adjustment strategy for the wireless network.

[0169] The target slice adjustment strategy is encoded and encapsulated to obtain the slice reconfiguration instruction sequence of the target slice adjustment strategy;

[0170] The network function entities of the wireless network are issued execution instructions according to the slice reconfiguration instruction sequence, and the network status is monitored in real time to confirm the completion of the dynamic adjustment and resource allocation of the network slice.

[0171] This process involves analyzing the core elements of the network slice adjustment strategy, including resource configuration modifications, scheduling rule adjustments, slice instance addition / reduction plans, and performance parameter settings. Simultaneously, it retrieves the current operating configuration of the wireless network, other effective network slice strategies, global resource allocation cap rules, and service quality assurance protocols for each service. Each element of the network slice adjustment strategy is compared and verified against the aforementioned existing rules and configurations. The focus is on checking whether resource allocation requests exceed the total network resource capacity limit, whether scheduling rule adjustments contradict existing strategies, and whether slice instance adjustments affect the resource consumption and service quality of other normally operating slices. Potential resource conflicts, rule conflicts, and service impact conflicts within the strategy are clearly identified. Conflict points are recorded, and the policy's executable scope is confirmed. Finally, the conflict detection of the network slice adjustment strategy is completed, confirming the core adjustment content and conflict-free execution basis of the strategy.

[0172] An isolated test environment was built that is completely identical to the actual wireless network architecture, resource configuration, and service deployment. This environment operates independently of the live network and will not affect the actual operation of services. The network slice adjustment strategy, after conflict detection, was fully imported into the control center of the test environment. According to the adjustment sequence and execution logic set by the strategy, the service load and user access patterns under the live network operation scenario were simulated. Operations such as slice resource allocation adjustment, scheduling rule update, slice instance creation or deletion were executed step by step. During the simulation, key network performance indicators were continuously collected, including data transmission latency, actual available bandwidth, data transmission success rate, and resource utilization. At the same time, various alarm information generated in the test environment was captured in real time, such as resource allocation failure alarms, slice connection interruption alarms, and performance failure alarms. The collected performance indicator data was compared with preset standards, and the types, trigger times, and associated adjustment operations of alarm information were sorted out to form a verification feedback report of the wireless network that includes performance data, alarm details, and policy execution status.

[0173] A comprehensive analysis of the verification feedback report is conducted. From an execution risk perspective, the severity and probability of potential service interruptions, performance fluctuations, resource waste, and impacts on other slices after policy implementation are assessed. For example, alarm information is used to determine the risk of core business unavailability, and performance metrics are used to determine the risk of long-term performance failure. From an expected benefit perspective, the performance improvement of network slices, the improvement in resource utilization efficiency, and the optimization effect on service quality after policy implementation are analyzed. For example, bandwidth utilization and latency compliance rates are compared before and after simulated execution to clarify the actual value brought by the policy. Based on the risk and benefit assessment results, the network slice adjustment policy is optimized in a targeted manner. If high-risk issues exist, resource allocation ratios are adjusted, scheduling trigger conditions are optimized, or adjustment steps are split to reduce risk. If the benefits do not meet expectations, configuration parameters are further optimized to improve performance, ensuring that the adjusted policy achieves the expected benefits under controllable risks, ultimately yielding the target slice adjustment policy for the wireless network.

[0174] The execution logic, sequence, and corresponding network functional entities of each adjustment operation in the target slice adjustment strategy are clearly defined. The resource configuration requirements, scheduling rule update instructions, and slice instance management requirements in the strategy are transformed into standardized instruction formats that network devices and functional entities can recognize and execute. For example, resource allocation adjustment is transformed into specific resource quota allocation instructions, scheduling rule updates are transformed into rule parameter configuration instructions, and slice instance addition or deletion is transformed into instance creation or deletion instructions. All standardized instructions are arranged in an orderly manner according to the order of strategy execution to ensure that the instruction sequence can accurately map all the adjustment intentions of the target strategy. At the same time, the instruction format is validated to ensure that there are no syntax errors and no logical confusion, and finally a slice reconfiguration instruction sequence of the target slice adjustment strategy is formed.

[0175] Following the sequence of slice reconfiguration instructions, execution instructions are sent sequentially to the corresponding network functional entities through the wireless network control interface. These entities include core network gateways, access network base stations, server clusters, and resource scheduling controllers. During instruction issuance, the instruction reception status and execution progress of each network functional entity are monitored in real time. Simultaneously, network operation status data is continuously collected, including performance indicators, resource usage, and service operation status of each slice. The system verifies whether each instruction achieves the expected adjustment effect. If an instruction fails to execute or the adjustment effect is not met, subsequent instruction issuance is paused immediately to investigate the cause. Execution continues until all instructions are completed. Monitoring data confirms that the network slice configuration adjustment is complete, resource allocation reaches the target state, network operation is stable, and all performance indicators meet preset standards. Ultimately, dynamic adjustment and resource allocation of the wireless network slices are achieved.

[0176] The beneficial effects are as follows: comprehensive conflict detection eliminates contradictions between network slicing adjustment strategies and existing configurations and rules, solidifying the foundation for strategy execution; simulating strategy execution in an isolated test environment accurately collects performance data and alarm information, obtaining reliable verification feedback without affecting the operation of the live network; combining verification feedback to evaluate the risks and benefits of the strategy, optimizing it into a target strategy with controllable risks and satisfactory benefits; transforming the target strategy into a standardized instruction sequence to ensure accurate transmission of adjustment intentions; issuing instructions sequentially and monitoring the execution process and network status in real time, promptly handling problems to ensure full implementation of instructions, ultimately achieving dynamic adjustment of network slices and optimized resource allocation, ensuring stable network operation and satisfactory performance.

[0177] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0178] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A 5G-A network slicing dynamic adjustment and resource allocation system for the low-altitude economy, characterized in that, The system includes a mobility management feature integration module, a network slicing service level agreement evaluation module, a mobility management demand prediction benchmark construction module, a multi-dimensional trend prediction module, a joint optimization decision generation module, and a strategy verification and execution module, wherein: The mobility management feature integration module is used to integrate the real-time mobility management type, mobility management priority, latency sensitivity and bandwidth requirements in the wireless network into the mobility management feature dataset of the wireless network. The network slicing service level protocol evaluation module is used to evaluate the real-time load status and wireless channel quality parameters of the wireless network based on the mobility management feature dataset, and obtain the network slicing status evaluation result of the wireless network. Specifically, it is used for: Obtain the real-time load status of the wireless network; The channel spectrum parameters captured in the wireless network are used as the wireless channel quality parameters of the wireless network. The real-time load status and the wireless channel quality parameters are time-fused and aligned to obtain a snapshot of the network operation status of the wireless network. Based on the latency, bandwidth, and reliability indicators defined in the network slicing service level protocol of the wireless network, a matching degree analysis is performed on the mobility management feature dataset and the network operation status snapshot to obtain the sub-item evaluation results of the mobility management feature dataset. Based on preset evaluation weight rules, the sub-evaluation results are weighted and aggregated to obtain the network slice status evaluation result of the wireless network. The calculation formula for the network slice state evaluation result is as follows: ； In the formula, The network slice state evaluation result, For the mobility management feature dataset, the mobility management type is... For a specific mobility management type among the mobility management types, The type weight coefficient is the pre-defined evaluation weight rule. This is the balance coefficient for the service demand matching degree in the evaluation weighting rules. For the specific mobility management type The service quality requirement vector, This is the balance coefficient for network state stability in the evaluation weighting rule. For the specific mobility management type The service capability vector that actually guarantees business operations. This is the network operating state vector in the network operating state snapshot. For the type in the evaluation weighting rule The ideal state threshold vector; The mobility management demand prediction benchmark construction module is used to construct the mobility management demand prediction benchmark of the wireless network based on the mobility management feature dataset and the expected demand and historical slice status information database of the wireless network, specifically for: The expected business growth information and service quality targets for low-altitude economic scenarios in the wireless network are integrated into the network planning strategy data of the wireless network. Based on the historical slice status information database in the wireless network, the mobility management feature dataset is traversed and analyzed to obtain the joint feature sequence of the wireless network; Based on the expected business growth information, the joint feature sequence is extrapolated to obtain the initial prediction of the mobility management feature dataset; Using the network planning strategy data as constraints, the initial prediction is corrected for compliance, resulting in the standard prediction of the mobility management feature dataset. The standard predicted quantity is correlated and mapped with the key performance indicator benchmark values ​​in the historical slice status information database to obtain the mobility management demand prediction benchmark of the wireless network. The multi-dimensional trend prediction module is used to perform multi-dimensional trend analysis on the network slice status assessment results based on the mobility management demand prediction benchmark, and obtain the prediction results of the wireless network. The joint optimization decision generation module is used to perform joint optimization decision on the standard network slice configuration strategy of the wireless network based on the prediction results, so as to obtain the network slice adjustment strategy of the wireless network. The policy verification and execution module is used to verify the feasibility of the network slice adjustment policy and execute the verified slice adjustment policy to realize the dynamic adjustment and resource allocation of the network slices of the wireless network.

2. The 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy as described in claim 1, characterized in that, When the mobility management feature integration module integrates the real-time mobility management type, mobility management priority, latency sensitivity, and bandwidth requirements in the wireless network into a mobility management feature dataset for the wireless network, it is specifically used for: Extract the real-time mobility management type parameters and corresponding mobility management priority policy parameters from the wireless network; The real-time mobility management type parameter and the mobility management priority policy parameter are subjected to network protocol quantization to obtain the mobility management policy vector of the wireless network; The network slice selection auxiliary information and service subscription data of the wireless network are used to perform performance quality screening to obtain the latency sensitivity level parameters and bandwidth requirement parameters of the wireless network. The mobility management policy vector, the latency sensitivity level parameter, and the bandwidth requirement parameter are correlated in multiple dimensions to obtain the mobility management feature dataset of the wireless network.

3. The 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy as described in claim 1, characterized in that, When the mobility management demand forecasting benchmark construction module performs trend extrapolation on the joint feature sequence based on the expected business growth information to obtain the initial forecast amount of the mobility management feature dataset, it is specifically used for: The expected business growth information is separated into business data to obtain the business volume growth data and the expected introduction characteristics of new businesses. The joint feature sequence is periodically decomposed to obtain the long-term trend component and periodic fluctuation component of the joint feature sequence. Based on the long-term trend components and the business volume growth data, the joint feature sequence is extrapolated over time to obtain the basic feature prediction values ​​of the mobility management feature dataset. By combining the periodic fluctuation component with the expected introduction features of new services, the predicted values ​​of the basic features are superimposed and corrected to obtain the initial predicted values ​​of the mobility management feature dataset.

4. A 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy as described in claim 1, characterized in that, When the mobility management demand forecasting baseline construction module performs the association mapping between the standard forecast quantity and the key performance indicator baseline values ​​in the historical slice status information database to obtain the mobility management demand forecasting baseline of the wireless network, it is specifically used for: Based on the historical slice status information database, the historical key performance index data associated with the features in the standard prediction quantity are matched and filtered to obtain the historical key index benchmark data of the standard prediction quantity. The feature items related to mobility management type, priority and service quality requirements in the standard predicted quantity are matched and mapped with the historical key indicator benchmark data to establish a feature benchmark association table of the standard predicted quantity. Based on the feature benchmark association table, the service quality requirement features in the standard prediction quantity are quantified for network slicing requirements to obtain the target values ​​of the key performance indicators of the wireless network. The target values ​​of the key performance indicators are integrated with the mobility management type and priority characteristics in the standard forecasts to form the mobility management demand forecasting benchmark for the wireless network.

5. A 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy as described in claim 1, characterized in that, When the multi-dimensional trend prediction module performs multi-dimensional trend analysis on the network slice status assessment results based on the mobility management demand prediction benchmark to obtain the prediction results of the wireless network, it is specifically used for: The deviation of the network slice status assessment results is compared with the mobility management demand prediction benchmark to obtain the difference feature set of the wireless network; By extrapolating the service quality of the network slices in the difference feature set in chronological order, the time series analysis results of the wireless network are obtained. Based on the time series analysis results and the network slice type of the wireless network, key factors are identified in the differential feature set to determine the dominant factors for slice allocation of the wireless network. Based on the mobility management demand prediction benchmark, the time series analysis results and the dominant factors of slice allocation are heterogeneously fused to obtain the prediction results of the wireless network.

6. A 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy as described in claim 1, characterized in that, When the joint optimization decision generation module performs joint optimization decision-making on the standard network slice configuration strategy of the wireless network based on the prediction results to obtain the network slice adjustment strategy of the wireless network, it is specifically used for: The prediction results are analyzed for potential quality risks to obtain the predicted values ​​of service quality risk and network slice performance requirements of the wireless network. The resource configuration templates and scheduling rules of the slice instances in the wireless network are integrated into the standard network slice configuration strategy of the wireless network; Based on the predicted performance requirements and the standard network slicing configuration strategy, a gap analysis is performed on the resource requirements and configuration supply of the wireless network to obtain the adjustment requirements information of the wireless network. Based on the adjustment requirement information, the standard network slice configuration strategy is modified with constraints to obtain the network slice adjustment strategy for the wireless network.

7. A 5G-A network slicing dynamic adjustment and resource allocation system for low-altitude economy as described in claim 1, characterized in that, The policy verification and execution module, when performing feasibility verification on the network slice adjustment policy and executing the verified slice adjustment policy to achieve dynamic adjustment and resource allocation of the network slices of the wireless network, is specifically used for: The network slice adjustment strategy is subjected to policy conflict detection to confirm the slice adjustment strategy of the network slice adjustment strategy; The slice adjustment strategy is simulated in the isolated test environment of the wireless network, and network performance indicators and alarm information are collected during the simulation process to obtain a verification feedback report of the wireless network. Based on the verification feedback report, the execution risk and expected benefits of the slice adjustment strategy are evaluated, and the strategy is optimized and adjusted to obtain the target slice adjustment strategy for the wireless network. The target slice adjustment strategy is encoded and encapsulated to obtain the slice reconfiguration instruction sequence of the target slice adjustment strategy; The network function entities of the wireless network are issued execution instructions according to the slice reconfiguration instruction sequence, and the network status is monitored in real time to confirm the completion of the dynamic adjustment and resource allocation of the network slice.

Images