Method and system for adjusting network quality and computer equipment

Abstract

The invention relates to the technical field of network resource management, and discloses a network quality adjusting method and system and computer equipment. The method comprises the following steps: when the network quality of user equipment is poor, obtaining network state data and resource state data of a service area where the user equipment is located; extracting a first key field of the network state data and a second key field of the resource state data; based on verification results of the first key field and the second key field, determining a current congestion state of a base station corresponding to the service area and a first target adjustment strategy matched with the current congestion state; and adjusting the first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain the first network quality. By implementing the technical scheme of the invention, the network performance can be effectively improved, the data transmission efficiency is optimized, and the user experience is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network resource management, and in particular to a method, system and computer device for adjusting network quality. Background Art

[0002] Modern mobile communication networks, with their high bandwidth, low latency and massive connection capabilities, have become the core communication infrastructure in scenarios such as smart cities and industrial automation. Their network architecture separates the control plane from the user plane and combines technologies such as network slicing to achieve flexible decoupling of service logic and transmission resources.

[0003] In terms of network resource management, related technologies mainly adopt an event-triggered congestion detection mechanism, relying on load indicators periodically reported by the base station side, and performing resource reallocation through manual policy configuration. For example, when a cell is detected to be overloaded, maintenance personnel need to manually adjust antenna parameters or switch users to adjacent base stations, and this lagged intervention is likely to cause service continuity interruption. Summary of the Invention

[0004] In view of this, the present invention provides a method, system and computer device for adjusting network quality, aiming to solve the problem of service continuity interruption caused by relying on manual lagged intervention in related technologies.

[0005] In a first aspect, the present invention provides a method for adjusting network quality, including: when the network quality of a user equipment is poor, obtaining network state data and resource state data of the service area where the user equipment is located; extracting a first key field of the network state data and a second key field of the resource state data; determining the current congestion state of the base station corresponding to the service area and a first target adjustment strategy matching the current congestion state based on the verification results of the first key field and the second key field; and adjusting first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain a first network quality.

[0006] The method for adjusting network quality provided by the embodiments of the present invention can quickly identify the congestion state of the base station by obtaining network state data and resource state data in real time and accurately extracting key fields for verification, thereby dynamically matching the optimal adjustment strategy, realizing automatic detection and intelligent decision-making of network quality problems. By specifically adjusting the network parameters of the user equipment and the base station, the response speed and accuracy of congestion control are significantly improved, while reducing the dependence on manual intervention, ensuring the efficient utilization of network resources and the continuous optimization of user experience.

[0007] In an alternative embodiment, based on the verification results of the first keyword field and the second keyword field, determining the current congestion status of the base station corresponding to the service area and the first target adjustment policy matching the current congestion status includes: obtaining a preset user subscription configuration; verifying the first keyword field and the second keyword field according to the user subscription configuration to obtain a first data verification result corresponding to the first keyword field and a second data verification result corresponding to the second keyword field; when the first data verification result and / or the second data verification result indicates that the user equipment passes the verification, analyzing the first data verification result and / or the second data verification result to determine the current congestion status of the base station; matching the current congestion status with a preset adjustment policy, and determining the first target adjustment policy corresponding to the current congestion status based on the matching result.

[0008] The network quality adjustment method provided by the embodiments of the present invention can quickly determine the compliance of user equipment by automatically verifying keyword fields through a preset user subscription configuration, and dynamically analyze the current congestion status of the base station in combination with real-time verification results, significantly improving the accuracy and efficiency of congestion identification. By intelligently matching the current congestion status with a preset adjustment policy, it can accurately generate a target policy adapted to the current network environment, realize the dynamic optimal allocation of network resources, reduce the complexity of manual configuration, enhance the real-time nature of congestion control and the flexibility of policy adaptation, thereby effectively improving the network response speed and service quality.

[0009] In an alternative embodiment, the user subscription configuration includes a quality of service profile and a maximum physical resource block utilization threshold; the first keyword field includes a session identifier, and the second keyword field includes a physical resource block utilization rate; verifying the first keyword field and the second keyword field according to the user subscription configuration to obtain a first data verification result corresponding to the first keyword field and a second data verification result corresponding to the second keyword field includes: determining a quantization index value matching the session identifier; detecting whether the quantization index value is within a preset range of the quality of service profile to obtain the first data verification result; detecting whether the physical resource block utilization rate is lower than the maximum physical resource block utilization threshold to obtain the second data verification result.

[0010] The network quality adjustment method provided by the embodiments of the present invention can accurately match the user subscription policy with the actual network status through a predefined quality of service profile and a maximum physical resource block utilization threshold, combined with a dual verification mechanism of session identifier and real-time resource utilization rate. By detecting whether the quantization index value meets the service quality range and whether the physical resource block utilization rate is lower than the preset upper limit, it not only ensures that the service quality of user services strictly follows the subscription agreement, but also effectively avoids network resource overload, thereby improving the compliance of resource allocation and network stability, while reducing manual intervention and enhancing the adaptability to dynamic network changes.

[0011] In an alternative embodiment, the current congestion state of the base station is determined by analyzing the first data verification result and / or the second data verification result, including: obtaining the coverage ranges of the respective base stations corresponding to the service area; determining the number of target user equipment corresponding to each base station based on the verification result, where the target user equipment is the user equipment that passes the verification; determining the user equipment density corresponding to each base station based on the coverage range and the number of target user equipment; determining the congestion index of each base station based on the user equipment density and the physical resource block utilization rate in the resource status data; the congestion index is used to characterize the current congestion state.

[0012] The network quality adjustment method provided by the embodiments of the present invention can accurately quantify the congestion index of each base station by combining the coverage range of the base station with the number of target user equipment that passes the verification, dynamically calculate the user equipment density, and further correlate with the physical resource block utilization rate, realizing an objective assessment of the network congestion state. This not only improves the accuracy of congestion identification but also provides a scientific basis for dynamic resource allocation, thereby effectively preventing network overload and improving resource utilization efficiency and service quality.

[0013] In an alternative embodiment, the current congestion state is matched with a preset adjustment strategy, and a first target adjustment strategy corresponding to the current congestion state is determined based on the matching result, including: obtaining a first preset policy template library, where the first preset policy template library includes a first mapping relationship between the congestion state and the preset adjustment policy template; extracting the first target adjustment strategy corresponding to the current congestion state from the first preset policy template library based on the current congestion state.

[0014] The network quality adjustment method provided by the embodiments of the present invention can establish a mapping relationship between the congestion state and the adjustment strategy through a predefined policy template library, quickly match the current network state and extract the corresponding strategy, significantly improving the response efficiency of congestion control, reducing the delay and error of manual decision-making, and ensuring the consistency and reliability of the strategy at the same time.

[0015] In an alternative embodiment, signaling characteristic parameters of the user equipment are obtained; the signaling characteristic parameters are associated with the device identifier of the user equipment to generate an association result; based on the association result, a signaling behavior pattern library is constructed by classifying according to the device identifier, and the signaling behavior pattern library is used to identify the signaling behavior characteristics corresponding to different device identifiers.

[0016] The network quality adjustment method provided by the embodiments of the present invention can accurately identify the typical signaling behavior characteristics of different devices by dynamically associating the signaling characteristic parameters of the user equipment with the device identifier and classifying and constructing the signaling behavior pattern library, effectively solving the network compatibility problem caused by the differences in signaling implementation of devices from different manufacturers, and providing a reliable basis for abnormal signaling detection and targeted optimization at the same time.

[0017] In an alternative embodiment, obtain the current signaling characteristics of the user equipment; compare the current signaling characteristics with the historical signaling behavior characteristics of the corresponding device identifier in the signaling behavior pattern library, and determine whether the user equipment is in an abnormal signaling state based on the comparison result; when the user equipment is in an abnormal signaling state, extract the target device identifier of the user equipment; based on the target device identifier, extract the corresponding initial policy template from the second preset policy template library, and adjust the initial policy template based on the abnormal signaling characteristics corresponding to the abnormal signaling state to generate a second target adjustment policy; adjust the second network parameters of the user equipment and the base station according to the second target adjustment policy to obtain the second network quality.

[0018] The network quality adjustment method provided by the embodiments of the present invention can accurately identify the abnormal signaling state by comparing the current signaling characteristics of the user equipment with the historical signaling behavior pattern library in real time, and quickly match the preset policy template based on the device identifier. By dynamically adjusting the policy template in combination with the abnormal characteristics to generate a customized adjustment policy, it realizes the differential optimization of network parameters for different device manufacturer characteristics and abnormal scenarios, which not only improves the accuracy and response efficiency of abnormal detection, but also avoids the device compatibility problems caused by the "one-size-fits-all" policy. Finally, the network quality is significantly improved through dynamic parameter adjustment, while reducing the manual intervention cost.

[0019] In an alternative embodiment, obtain the user layer resource utilization rate after network parameter adjustment according to the first target adjustment policy and / or the second target adjustment policy; determine the current deviation degree based on the difference between the user layer resource utilization rate and the preset resource utilization rate; determine the target alarm level corresponding to the current deviation degree based on the second mapping relationship between the deviation degree and the alarm level; execute the inspection policy corresponding to the target alarm level to generate the corresponding inspection result.

[0020] The network quality adjustment method provided by the embodiments of the present invention can accurately quantify the network performance deviation by obtaining the user layer resource utilization rate after network parameter adjustment in real time and calculating its deviation from the preset value. Based on the preset alarm level mapping relationship, it realizes hierarchical alarm response, and can dynamically trigger differential inspection policies according to the deviation degree, which not only avoids overprocessing of minor deviations, but also can take timely intervention for high-risk deviations, thereby improving the accuracy and response efficiency of network anomaly detection. At the same time, by automatically executing the inspection policy matching the alarm level, the manual operation and maintenance cost is significantly reduced, and the controllability and traceability of the network optimization process are ensured.

[0021] Second aspect, the present invention provides a network quality adjustment system, including: a user equipment, configured to monitor the signal quality of the service area where it is located, and when the signal quality is lower than a preset threshold, obtain the network status data of the service area; a base station, communicatively connected to the user equipment, configured to receive the network status data sent by the user equipment, and obtain the resource status data of the service area; a server cluster, communicatively connected to the user equipment and the base station respectively, configured to receive the network status data and the resource status data sent by the base station, extract the first key field of the network status data, and the second key field of the resource status data, determine the current congestion status of the base station corresponding to the service area and the first target adjustment strategy matching the current congestion status based on the verification results of the first key field and the second key field, and adjust the first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain the first network quality.

[0022] Third aspect, the present invention provides a network quality adjustment device, including: a first acquisition module, configured to obtain the network status data and the resource status data of the service area where the user equipment is located when the network quality of the user equipment is poor; a first extraction module, configured to extract the first key field of the network status data, and the second key field of the resource status data; a first determination module, configured to determine the current congestion status of the base station corresponding to the service area and the first target adjustment strategy matching the current congestion status based on the verification results of the first key field and the second key field; a first adjustment module, configured to adjust the first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain the first network quality.

[0023] Fourth aspect, the present invention provides a computer device, including: a memory and a processor, communicatively connected to each other between the memory and the processor, the memory stores computer instructions, and the processor executes the computer instructions to execute the network quality adjustment method according to the first aspect or any corresponding embodiment thereof.

[0024] Fifth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the network quality adjustment method according to the first aspect or any corresponding embodiment thereof.

[0025] Sixth aspect, the present invention provides a computer program product, including computer instructions, and the computer instructions are used to cause a computer to execute the network quality adjustment method according to the first aspect or any corresponding embodiment thereof. Description of the Drawings

[0026] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0027] Figure 1 is a schematic diagram of a network quality adjustment system according to an embodiment of the present invention;

[0028] Figure 2 is a flowchart of a network quality adjustment method according to an embodiment of the present invention;

[0029] Figure 3 is a flowchart of another network quality adjustment method according to an embodiment of the present invention;

[0030] Figure 4 is a flowchart of yet another network quality adjustment method according to an embodiment of the present invention;

[0031] Figure 5 is a flowchart of still another network quality adjustment method according to an embodiment of the present invention;

[0032] Figure 6 is a structural block diagram of a network quality adjustment device according to an embodiment of the present invention;

[0033] Figure 7 is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Specific Embodiments

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0035] As the fifth-generation mobile communication technology, 5G network has advantages such as high speed, low latency, and large connection, and is widely used in multiple fields such as industrial Internet, intelligent transportation, and smart agriculture. Its main design goal is to meet diverse service requirements, covering enhanced mobile broadband (eMBB), ultra-reliable low-latency communication (URLLC), and massive machine-type communication (mMTC). The 5G network architecture includes the 5G core network (5GC) and the 5G radio access network (NG-RAN), which are connected through the NG interface, adopt a service-based architecture (SBA), support network function virtualization (NFV) and software-defined network (SDN). At the same time, the 5G network realizes the separation of the control plane and the user plane (CUPS), and introduces network slicing technology to meet the needs of different service scenarios.

[0036] However, there are still some significant deficiencies in the current 5G technology solutions in practical applications, especially in network congestion control, data management, user experience, and privacy protection. For example, although the current network congestion control methods (such as backoff timers and cause values) can alleviate network load, they often require manual operation by users to take effect, with low efficiency and affecting user experience. In addition, differences in the implementation of signaling behavior by different vendors may also exacerbate network congestion. In terms of data management, with the rapid increase in the amount of data in the 5G network, the existing mechanisms lack intelligence and dynamic optimization, resulting in complex data collection, storage, and sharing processes and difficulty in real-time adjustment. At the same time, the issue of data privacy protection has become increasingly prominent. The high-frequency signals of the 5G network have poor penetration and are easily interfered by obstacles, resulting in limited signal coverage, while high energy consumption and dense base station deployment further increase the operating cost.

[0037] In view of this, the technical solution of the present invention provides a more intelligent and dynamic 5G network data management and congestion control mechanism. By introducing a unified intelligent data management system, the present invention can dynamically collect, deduplicate, and integrate data from user equipment (UE), radio access network (RAN), and core network, and at the same time support data life cycle management (including creation, storage, sharing, and destruction). This mechanism not only improves the efficiency of data management but also ensures data security and privacy protection. In addition, the present invention also adopts an intelligent optimization algorithm to dynamically adjust data exposure and resource allocation according to the real-time network state to improve data utilization efficiency and reduce network load. Through these improvements, the present invention aims to solve the deficiencies of the current 5G network in congestion control, data management, and privacy protection, thereby enhancing user experience and network performance.

[0038] In this embodiment, a system for adjusting network quality is provided, as Figure 1 shown, the system includes:

[0039] User equipment 1 is used to monitor the signal quality of the service area where it is located. When the signal quality is lower than the preset threshold, it obtains the network status data of the service area.

[0040] Base station 2 is communicatively connected to user equipment 1 and is used to receive the network status data sent by user equipment 1 and obtain the resource status data of the service area.

[0041] The server cluster 3 is communicatively connected to user equipment 1 and base station 2 respectively. It is used to receive the network status data and resource status data sent by base station 2, extract the first key field of the network status data and the second key field of the resource status data, determine the current congestion status of base station 2 corresponding to the service area and the first target adjustment strategy matching the current congestion status based on the verification results of the first key field and the second key field, and adjust the first network parameters of user equipment 1 and base station 2 according to the first target adjustment strategy to obtain the first network quality.

[0042] User equipment 1 (UE) refers to the terminal equipment (such as mobile phones, Internet of Things devices) accessing the 5th Generation Mobile Communication Technology (5G) network. It is responsible for monitoring the signal quality of the service area and sending network status data to base station 2 (gNB) through Radio Resource Control (RRC) connection.

[0043] The service area refers to the geographical range covered by the 5G cell to which user equipment 1 is currently connected, which is determined by the wireless signal coverage range of base station 2.

[0044] The signal quality refers to the wireless link quality parameters of the service area monitored by user equipment 1, including Reference Signal Receiving Power (RSRP) and Reference Signal Receiving Quality (RSRQ), which are used to evaluate the connection stability between user equipment 1 and base station 2.

[0045] The preset threshold is the signal quality threshold defined by the network side. For example, when RSRP or RSRQ is lower than this threshold, user equipment 1 triggers the reporting process of network status data.

[0046] Network status data is dynamic information related to the service area reported by user equipment 1. For example, it may include: radio link quality (RSRP / RSRQ), service feature identifier, and implicit location identifier (location information that protects user privacy) based on the user's permanent identifier (Subscription Permanent Identifier, SUPI), etc. Among them, the service feature identifier may include, for example, PDU session ID, service type associated with the 5G Quality of Service Identifier (5G QoS Identifier, 5QI), etc.

[0047] Specifically, user equipment 1 uses the built-in wireless communication module to continuously monitor the signal quality parameters of the serving cell (i.e., the service area), including reference signal received power and reference signal received quality. These parameters reflect the radio link quality between user equipment 1 and base station 2. When it is detected that the RSRP or RSRQ is lower than the signal quality threshold preset by the network side (for example, RSRP < -110 dBm or RSRQ < -12 dB), user equipment 1 triggers the generation process of the enhanced measurement report. This report contains the following core data:

[0048] Radio link quality: the current value of RSRP / RSRQ;

[0049] Service feature identifier: the PDU session ID associated with 5QI (used to identify the service type, such as voice, video, or data);

[0050] Implicit location identifier: the obfuscated location information based on SUPI (to avoid directly exposing the user's coordinates).

[0051] Subsequently, user equipment 1 sends the enhanced measurement report to base station 2 through the RRC connection, completing the active reporting of network status data.

[0052] Base station 2 refers to the base station 2 equipment in the 5G Radio Access Network (RAN), which is responsible for receiving the network status data reported by user equipment 1, and after attaching the resource status data, forwarding it to the Access and Mobility Management Function (AMF) in server cluster 3 through the N2 interface.

[0053] Resource status data refers to the data on the usage of radio resources provided by base station 2. For example, it may include: Physical Resource Block (PRB) utilization rate, Physical Downlink Control Channel (PDCCH) Control Channel Element (CCE) utilization rate, Random Access Channel (RACH) load level, and time synchronization identifier (gNB local timestamp, measurement report sequence number), etc.

[0054] Specifically, after receiving the enhanced measurement report sent by user equipment 1 through the N2 interface, base station 2 adds the local radio resource status of base station 2, including:

[0055] PRB utilization rate: reflecting the occupancy rate of radio channel resources;

[0056] PDCCH CCE utilization rate: measuring the control channel load;

[0057] RACH load level (level 0 - 15): indicating the density of random access requests.

[0058] And append the time synchronization identifier, including the gNB local timestamp (for cross - node time alignment) and the measurement report sequence number (to prevent replay attacks). Forward the integrated data to the AMF in the core network through the N2 interface to trigger the subsequent access permission verification process.

[0059] Server cluster 3 refers to the Network Data Analytics Function (NWDAF) and related core network functions in the 5G core network, which are responsible for integrating data, analyzing congestion status, generating optimization strategies, and coordinating policy execution. Among them, related core network functions may include, for example, Policy Control Function (PCF), Session Management Function (SMF), AMF, etc. Multiple servers can form server cluster 3, and server cluster 3 has a main server with central control.

[0060] The first key field refers to the core field extracted from the network status data. The second key field refers to the core field extracted from the resource status data. The verification result refers to the result of the legality and compliance check of the first key field and the second key field by the NWDAF in server cluster 3.

[0061] The current congestion state refers to the degree of network congestion of base station 2, and the first target adjustment strategy refers to the optimization strategy generated based on the congestion state. The first network parameter refers to the key configuration parameters that need to be adjusted for user equipment 1 and base station 2, and the first network quality refers to the improvement result of the network performance index after adjustment.

[0062] Specifically, the NWDAF in the server cluster 3 extracts key fields from the received network status data and resource status data: the first key field (such as service feature identifier) in the network status data and the second key field (such as PRB utilization rate) in the resource status data. On this basis, the NWDAF verifies these key fields to check their legality and compliance.

[0063] According to the verification result, the NWDAF can evaluate the current network congestion state of the base station 2 in the service area. For example, if the resource status of base station 2 indicates a high load, it is determined that the base station 2 is currently in a congested state. Next, the server cluster 3 formulates a first target adjustment strategy that matches this state. This adjustment strategy can, for example, involve changing some key network parameters, such as adjusting the resource allocation of base station 2, optimizing the access strategy of user equipment 1, adjusting radio resource control, etc., to improve the overall network performance. After adjustment, the first network parameters (such as radio channel configuration, access strategy, etc.) of user equipment 1 and base station 2 will be optimized accordingly, thereby achieving an improvement in network quality and finally obtaining the improvement result of the first network quality.

[0064] The network quality adjustment system provided by the embodiments of the present invention realizes the accurate identification and adaptive optimization of network congestion status through the collaborative reporting of real-time monitoring by user equipment and base station resource status data, combined with the dynamic verification and intelligent analysis of multi-dimensional key fields by the server cluster. Utilizing the data integration ability of the 5G core network NWDAF, it deeply integrates service characteristics, radio resource load, and privacy-protected location information to generate a real-time adjustment strategy that matches the congestion state. While protecting user privacy, it significantly improves network quality through dynamic parameter optimization, forms a closed-loop optimization mechanism of "terminal-network-cloud" linkage, effectively solves the problem of passive response delay in traditional networks, and has advantages such as high timeliness, strategy accuracy, and optimized resource utilization.

[0065] According to the embodiments of the present invention, there is provided an embodiment of a method for adjusting network quality. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0066] In this embodiment, a method for adjusting network quality is provided, which can be used in a computer device. The computer device can be any server in the above-mentioned server cluster. Figure 2 It is a flowchart of the method for adjusting network quality according to an embodiment of the present invention, as Figure 2 shown. The process includes the following steps:

[0067] Step S201, when the network quality of the user equipment is poor, obtain the network status data and resource status data of the service area where the user equipment is located.

[0068] When the user equipment detects that the signal quality (such as RSRP or RSRQ) in the service area is lower than a preset threshold (for example, RSRP < -110 dBm or RSRQ < -12 dB), it triggers the reporting process of the network status data. The user equipment sends an enhanced measurement report containing the radio link quality (RSRP / RSRQ), service characteristic identifier (such as PDU session ID and service type associated with 5QI), and implicit location identifier based on SUPI to the base station through the RRC connection. After receiving the network status data reported by the user equipment, the base station attaches its own resource status data, including PRB utilization rate, PDCCH CCE utilization rate, RACH load level, and time synchronization identifier, and sends the integrated data to the Access and Mobility Management Function (AMF) server in the server cluster through the N2 interface. The Network Data Analysis Function (NWDAF) server in the server cluster receives and stores these data through the core network interface (such as N2 / Namf), and completes the acquisition of the network status data and resource status data.

[0069] Step S202, extract the first key field of the network status data and the second key field of the resource status data.

[0070] The Network Data Analysis Function (NWDAF) server extracts the first key field from the received network status data and the second key field from the resource status data. The extraction process is completed through data parsing and filtering. For example, specific fields are located from JSON or binary protocols, and redundant information is removed to ensure that subsequent analysis focuses on the core metrics affecting network congestion.

[0071] Step S203, based on the verification results of the first key field and the second key field, determine the current congestion status of the base station corresponding to the service area and the first target adjustment strategy matching the current congestion status.

[0072] The NWDAF performs legality and compliance verification on the first key field and the second key field. After the verification passes, it comprehensively analyzes the current congestion status in combination with multi-dimensional data. According to the current congestion status, the NWDAF selects a matching first target adjustment policy from a predefined policy library, such as dynamically adjusting resource allocation (reserving PRBs for high-priority services), optimizing PDCCH configuration (increasing CCE aggregation level), reducing RACH competition intensity (expanding the preamble range), or triggering the user equipment to switch to a neighboring cell (selecting a cell with lower load based on implicit location identification). Policy generation depends on preset rules or machine learning models to ensure precise matching with the current network state.

[0073] Step S204: Adjust the first network parameters of the user equipment and the base station according to the first target adjustment policy to obtain the first network quality.

[0074] The server cluster coordinates policy execution through each core network function server (such as PCF, SMF, AMF). For example, the NWDAF sends an adjustment instruction to the Policy Control Function (PCF) server. The PCF sends a policy update request to the Session Management Function (SMF) server through the Npcf interface. The SMF then modifies the QoS rules of the User Plane Function (UPF) through the N4 interface. At the same time, the AMF sends a Radio Resource Control (RRC) reconfiguration instruction to the base station through the N2 interface. After the adjustment is completed, the user equipment and the base station report new measurement reports. The NWDAF evaluates indicators such as the improvement amplitude of RSRP / RSRQ and the decrease ratio of PRB utilization rate. If the expected threshold is reached (such as an RSRP increase of 10 dB and the PRB utilization rate drops below 70%), it is determined that the improvement of the first network quality is successful.

[0075] The network quality adjustment method provided by the embodiments of the present invention can quickly identify the congestion status of the base station by obtaining network status data and resource status data in real time and accurately extracting key fields for verification, so as to dynamically match the optimal adjustment policy, realizing the automatic detection and intelligent decision-making of network quality problems. By specifically adjusting the network parameters of the user equipment and the base station, it significantly improves the response speed and accuracy of congestion control, while reducing the dependence on manual intervention, ensuring the efficient utilization of network resources and the continuous optimization of the user experience.

[0076] In this embodiment, a network quality adjustment method is provided, which can be used in a computer device, such as any server in the above-mentioned server cluster. Figure 3 is a flowchart of the network quality adjustment method according to the embodiments of the present invention, as Figure 3 shown, the process includes the following steps:

[0077] Step S301: When the network quality of the user equipment is poor, obtain the network status data and resource status data of the service area where the user equipment is located. For details, please refer to Figure 2 Step S201 of the embodiment shown, which will not be elaborated here.

[0078] Step S302: Extract the first key field of the network status data and the second key field of the resource status data. For details, please refer to Figure 2 Step S202 of the embodiment shown, which will not be elaborated here.

[0079] Step S303: Based on the verification results of the first key field and the second key field, determine the current congestion status of the base station corresponding to the service area and the first target adjustment strategy matching the current congestion status.

[0080] Specifically, the above Step S303 includes:

[0081] Step S3031: Obtain the preset user subscription configuration.

[0082] The user subscription configuration refers to the set of service parameters subscribed by the user from the operator and is stored in the Unified Data Management (UDM) server. For example, it may include: user identifier (SUPI / SUCI), authorized network slice type (S-NSSAI list), subscribed QoS profile (such as 5QI value, guaranteed bit rate, packet loss rate, etc., service quality requirements), and privacy protection constraints (such as data anonymization level, geographical restrictions), etc. These configurations are used to verify whether the user equipment has the right to access the network and determine whether its service level meets the subscription conditions. Specifically, the preset user subscription configuration is stored in the UDM and is obtained by the AMF by invoking the Nudm service interface to send a query request to the UDM. When the AMF receives the enhanced measurement report of the user equipment forwarded by the base station, it sends a service query to the UDM according to the user identifier (such as SUPI / SUCI) in the report. The user subscription data returned by the UDM includes: user authorized network slice type (S-NSSAI list), subscribed QoS profile (such as 5QI value, guaranteed bit rate, packet loss rate), privacy protection constraints (such as data anonymization level, geographical restrictions), and service level agreement (SLA) parameters (such as maximum threshold of PRB utilization rate). These data are encapsulated in JSON or XML format and transmitted through the HTTP / 2 protocol to ensure data integrity and real-time performance. The UDM also records the policy version (ETag) and the effective timestamp to support dynamic updates and conflict resolution.

[0083] Step S3032: Verify the first key field and the second key field according to the user subscription configuration, to obtain a first data verification result corresponding to the first key field and a second data verification result corresponding to the second key field.

[0084] The first data verification result refers to the compliance check result of the network status data. For example, whether the 5QI value corresponding to the service type requested by the user (such as voice, video) belongs to the range of the subscribed Quality of Service (QoS) profile. The second data verification result refers to the compliance check result of the resource status data. For example, whether the radio resource utilization rate of the base station (such as PRB utilization rate, PDCCH CCE utilization rate) is lower than the maximum threshold of the subscribed Service Level Agreement (SLA). Specifically, according to the user request, verify the first key field in the network status data. For example, the service type requested by the user (such as voice, video, etc.) is compared with the subscribed QoS profile to check whether the service type requested by the user meets the service quality requirements agreed at the time of subscription (such as whether the corresponding 5QI value is within the range of the subscribed QoS profile). If the network status data meets the requirements of the subscribed QoS profile, generate a first data verification result indicating that the network status data is compliant. Verify the second key field in the resource status data. For example, check whether the radio resource utilization rate of the base station is lower than the maximum threshold value of the service level agreement agreed at the time of user subscription (for example, whether the PRB utilization rate is lower than the subscribed maximum threshold). If the resource utilization situation of the base station meets the requirements of the subscription agreement, generate a second data verification result indicating that the resource status data is compliant.

[0085] In some optional embodiments, the user subscription configuration includes a quality of service profile and a maximum physical resource block utilization threshold; the first key field includes a session identifier, and the second key field includes a physical resource block utilization rate; the above step S3032 includes:

[0086] Step a1: Determine the quantization metric value that matches the session identifier.

[0087] The session identifier refers to the PDU session ID (Protocol Data Unit Session Identifier), which is used to uniquely identify the service session of the user equipment in the 5G network, distinguish different service types (such as voice, video, data download), and associate with the QoS flow parameters. The quantization metric value refers to the specific value in the QoS flow parameters, such as the 5QI value. Specifically, according to the PDU session ID (session identifier), the QoS flow parameters related to this session are found, especially the 5QI value. The 5QI value is a quality of service metric related to a specific service session (such as voice, video, etc.), which can reflect the network performance requirements of this session. Through the 5QI value associated with the session identifier, the specific quantization metric of this session is determined.

[0088] Step a2, detect whether the quantization metric value is within the preset range of the quality of service profile to obtain the first data verification result.

[0089] The quality of service profile refers to a configuration file related to the network quality of service (QoS) of the user equipment, which contains a set of predefined parameters and rules for managing and guaranteeing the quality of network services. Specifically, the QoS flow parameters monitored in real time are compared with the preset range defined in the subscribed quality of service profile. For example, detect whether the 5QI value belongs to the subscribed quality of service profile. If it belongs to the subscribed quality of service profile, the first data verification result is generated as "passed"; otherwise, it is "not passed".

[0090] Step a3, detect whether the physical resource block utilization rate is lower than the maximum physical resource block utilization rate threshold to obtain the second data verification result.

[0091] The physical resource block utilization rate refers to the occupancy ratio of the physical resource blocks (Physical Resource Block) on the radio side, which reflects the load situation of the base station radio resources. The maximum physical resource block utilization rate threshold refers to the upper limit of the PRB utilization rate specified in the service level agreement (SLA) subscribed by the user. If the actual PRB utilization rate exceeds this threshold, the network will reject the access request of the user equipment (such as returning the cause value #68 "network congestion"). Specifically, the detection of the physical resource block utilization rate is completed by real-time collecting the radio resource status data of the base station (such as the available PRB ratio). The maximum PRB utilization rate threshold is specified by the service level agreement subscribed by the user (for example, 70%). Compare the current physical resource block utilization rate (such as 65%) with the threshold: if the actual utilization rate is lower than the threshold, the second data verification result is "passed"; if it exceeds (such as 75%), it is marked as "not passed".

[0092] In the above embodiments, through the predefined quality of service profile and the maximum physical resource block utilization threshold, combined with the dual verification mechanism of the session identifier and the real-time resource utilization rate, the user subscription policy can be accurately matched with the actual network status. By detecting whether the quantified index value meets the quality of service range and whether the physical resource block utilization rate is lower than the preset upper limit, not only is the quality of service of the user service strictly in accordance with the subscription agreement ensured, but also network resource overload is effectively avoided, thereby improving the compliance of resource allocation and network stability. At the same time, manual intervention is reduced, and the adaptability to dynamic network changes is enhanced.

[0093] Step S3033, when the first data verification result and / or the second data verification result indicates that the user equipment passes the verification, analyze the first data verification result and / or the second data verification result to determine the current congestion status of the base station.

[0094] When the user equipment passes the verification, the NWDAF performs spatio-temporal correlation analysis, correlates the tracking area identity (TAI) history record of the user equipment with the NR cell identity (NR Cell ID) trajectory, so as to achieve spatio-temporal positioning with high precision. Obtain the geographical coordinates of the base station through the Operation Administration and Maintenance (OAM) server to provide geographical reference for the base station load data. The PRB utilization rate of the base station is aligned according to a 15-minute time window. Subsequently, based on the first data verification result and / or the second data verification result, determine the current congestion status of the base station.

[0095] In some alternative embodiments, the above step S3033 includes:

[0096] Step b1, when the first data verification result and / or the second data verification result indicates that the user equipment passes the verification, obtain the coverage ranges of the respective base stations corresponding to the service area.

[0097] Coverage refers to the wireless signal coverage area of a base station, which is defined by the geographical coordinates of the base station (such as longitude and latitude) and a wireless signal propagation model (such as TAI tracking area identifier). Specifically, when a user equipment passes the verification, it obtains the coverage of each base station in the service area for congestion analysis. The coverage is defined by the geographical coordinates of the base station (such as longitude and latitude) and the tracking area identifier (TAI), and the data is sourced from OAM or the network configuration database. For example, through the spatio-temporal correlation analysis module of NWDAF, the TAI historical record of the UE is combined with the NR cell identifier (NR Cell ID) trajectory, and the geographical coordinate information of the base station provided by OAM is invoked to determine the signal coverage area of each base station (such as a circular area with a radius of 500 meters). In addition, the coverage can also be dynamically corrected by combining the implicit location identifier reported by the user equipment (fuzzy location based on SUPI) and the radio link quality (such as RSRP, RSRQ) through a signal propagation model.

[0098] Step b2, based on the verification result, determine the number of target user equipments corresponding to each base station, where the target user equipment is the user equipment that has passed the verification.

[0099] The number of target user equipments refers to the number of user equipments that have passed the access permission verification. The verification conditions can include, for example: user subscription data (such as SUPI / SUCI, authorized S-NSSAI list), 5QI value matching the QoS profile, and the current PRB utilization not exceeding the limit, etc. Specifically, query the UDM to obtain user subscription data (such as SUPI / SUCI, authorized S-NSSAI list, QoS profile), and perform dynamic access control in combination with the real-time network status (such as the PRB utilization of the serving cell). For example, when the 5QI value of the user equipment matches the subscribed QoS profile and the current PRB utilization is lower than the maximum threshold specified by the SLA, the user equipment is allowed to access the network. By correlating TAI and NR Cell ID, count the number of user equipments that meet the above conditions within the coverage of each base station. For example, if 100 user equipments initiate access requests within the coverage area of a certain base station, and 80 of them pass the verification (5QI matches and PRB is not exceeded), then the number of target user equipments is 80.

[0100] Step b3, based on the coverage and the number of target user equipments, determine the user equipment density corresponding to each base station.

[0101] User equipment density refers to the number of user equipments that have passed the verification per unit coverage area. Specifically, the density formula is: For example, if the coverage area of a certain base station is 1.5 square kilometers and the number of target user equipments is 120, then the user equipment density is 80 UE / km 2 。

[0102] Step b4: Determine the congestion index of each base station based on the user equipment density and the physical resource block utilization rate in the resource status data. The congestion index is used to characterize the current congestion status.

[0103] The congestion index is a dynamic indicator calculated by integrating the PRB utilization rate and the user equipment density, and is used to determine the network congestion status (for example, when the congestion index value exceeds 0.75, it is marked as a congestion hotspot). Specifically, the congestion index (DCI) is obtained by weighted calculation of the PRB utilization rate and the user equipment density, and the formula is:

[0104] DCI = α × physical resource block utilization rate + β × user equipment density.

[0105] Among them, the weights α and β can be dynamically adjusted (for example, α = 0.6, β = 0.4). For example, the physical resource block utilization rate of a certain base station is 75% (0.75), and the user equipment density is 100 UE / km 2 (normalized to 0.5), then DCI is: 0.6 × 0.75 + 0.4 × 0.5 = 0.65.

[0106] When the DCI exceeds the preset congestion threshold (such as 0.75) and lasts for a certain time window (such as 3 15-minute cycles), this area is marked as a congestion hotspot.

[0107] In the above embodiment, by combining the coverage range of the base station with the number of target user equipment that has passed verification, the user equipment density is dynamically calculated, and further associated with the physical resource block utilization rate, the congestion index of each base station can be accurately quantified, realizing an objective evaluation of the network congestion status. This not only improves the accuracy of congestion identification, but also provides a scientific basis for dynamic resource allocation, thereby effectively preventing network overload and improving resource utilization efficiency and service quality.

[0108] Step S3034: Match the current congestion status with the preset adjustment strategy, and determine the first target adjustment strategy corresponding to the current congestion status based on the matching result.

[0109] The preset adjustment strategy refers to a predefined network optimization strategy template, which is triggered according to different congestion statuses. For example, it can include:

[0110] QoS adjustment strategy: Dynamically adjust the bandwidth allocation ratio of the UPF to give priority to high-priority services (such as voice);

[0111] Signaling optimization strategy: Set the RRC Inactivity Timer range and dynamically calculate the backoff time parameter;

[0112] Resource allocation strategy: Sort the policy actions according to the 5QI priority (voice > video > data);

[0113] Security Policy: When there are conflicts, follow the hierarchical priority (emergency services > regular services) and the principle of the latest timestamp.

[0114] These policies are stored in the Unified Data Repository (UDR) server and are matched and distributed by the PCF for execution. Specifically, the preset adjustment policies are stored in the UDR, classified by scenario and priority. The matching process is as follows:

[0115] Policy Template Matching: The PCF extracts the matching template from the UDR based on the current congestion status and service type (such as voice service). For example, for severe congestion of voice service, it matches the QoS adjustment template of "Guarantee 5QI=1 traffic first".

[0116] Variable Binding and Priority Sorting: Bind dynamic parameters (such as the current PRB utilization rate, DCI value) to the variables of the policy template and sort the actions according to the 5QI priority (voice > video > data). For example, preferentially reduce the bandwidth allocation ratio of data services;

[0117] Conflict Resolution: If multiple policies conflict (such as the need for capacity expansion and traffic limiting at the same time), the PCF arbitrates according to the hierarchical rules; emergency service policies (such as URLLC slices) take precedence over regular services; for policies at the same level, the principle of the latest timestamp is adopted to ensure that the latest policy takes effect.

[0118] The finally generated first target adjustment policy includes specific actions (such as adjusting the UPF shunt weight, dynamically setting the RRC Inactivity Timer to 60 seconds) and is distributed to the SMF / AMF for execution through the Npcf interface.

[0119] In some alternative embodiments, the above step S3034 includes:

[0120] Step c1, obtain the first preset policy template library, which includes the first mapping relationship between the congestion status and the preset adjustment policy template.

[0121] The first preset policy template library refers to the database storing the preset congestion control policies, including the policy templates corresponding to different congestion statuses. The policy templates are sorted according to the 5QI priority (voice > video > data). Specifically, the first mapping relationship establishes the association between the congestion status and the policy through the metadata of the policy template (such as tags, priorities, applicable scenarios).

[0122] Step c2, based on the current congestion status, extract the first target adjustment policy corresponding to the current congestion status from the first preset policy template library.

[0123] When the current congestion state is detected, the PCF retrieves the policy that matches the DCI value and service priority from the policy template library, and issues the policy (such as adjusting QoS flow parameters or RACH resource allocation) to the SMF or AMF through the Npcf interface, and ensures the consistency of the policy version through the ETag mechanism. For example, the PCF generates a policy triggered by DCI overrun, extracts the UDR policy template and sorts it according to the 5QI priority, and issues a QoS adjustment policy to the SMF and an RRC parameter to the AMF.

[0124] In the above embodiment, by establishing the mapping relationship between the congestion state and the adjustment policy through the predefined policy template library, the current network state can be quickly matched and the corresponding policy can be extracted, which significantly improves the response efficiency of congestion control, reduces the delay and error of manual decision-making, and ensures the consistency and reliability of the policy at the same time.

[0125] Step S304, adjust the first network parameters of the user equipment and the base station according to the first target adjustment policy to obtain the first network quality. For details, please refer to Figure 2 Step S204 of the embodiment shown, which will not be elaborated here.

[0126] The network quality adjustment method provided by the embodiment of the present invention can automatically verify the key fields through the preset user subscription configuration, quickly judge the compliance of the user equipment, and dynamically analyze the current congestion state of the base station in combination with the real-time verification results, which significantly improves the accuracy and efficiency of congestion identification. By intelligently matching the current congestion state with the preset adjustment policy, the target policy adapted to the current network environment can be accurately generated, realizing the dynamic optimal allocation of network resources, reducing the complexity of manual configuration, enhancing the real-time performance of congestion control and the flexibility of policy adaptation, thereby effectively improving the network response speed and service quality.

[0127] In this embodiment, a network quality adjustment method is provided, which can be used in a computer device, such as any server in the above server cluster, Figure 4 is a flowchart of the network quality adjustment method according to the embodiment of the present invention, as Figure 4 shown, and the process includes the following steps:

[0128] Step S401, when the network quality of the user equipment is poor, obtain the network state data and resource state data of the service area where the user equipment is located. For details, please refer to Figure 3 Step S301 of the embodiment shown, which will not be elaborated here.

[0129] Step S402, extract the first key field of the network state data and the second key field of the resource state data. For details, please refer to Figure 3 Step S302 of the embodiment shown, which will not be elaborated here.

[0130] Step S403: Based on the verification results of the first keyword field and the second keyword field, determine the current congestion status of the base station corresponding to the service area and the first target adjustment strategy that matches the current congestion status. For details, please refer to Figure 3 Step S303 of the embodiment shown, which will not be elaborated here.

[0131] Step S404: Adjust the first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain the first network quality. For details, please refer to Figure 3 Step S304 of the embodiment shown, which will not be elaborated here.

[0132] Step S405: Obtain the signaling feature parameters of the user equipment; associate the signaling feature parameters with the device identifier of the user equipment to generate an association result; based on the association result, construct a signaling behavior pattern library according to the device identifier classification, and the signaling behavior pattern library is used to identify the signaling behavior characteristics corresponding to different device identifiers.

[0133] The signaling feature parameters refer to the parameters related to the signaling behavior generated by the user equipment during the communication process. For example, they may include: RRC reconfiguration delay distribution (such as the time consumption for connection establishment and handover), NAS signaling retransmission rate (the repeated transmission ratio of non-access stratum signaling), TAC code associated device model (identifying the device manufacturer and model through the Type Allocation Code), and signaling frequency (such as the number of RRC connection requests per unit time), etc. These parameters are used to quantify the signaling behavior characteristics of the device. Specifically, the signaling feature parameters are obtained by parsing the interaction signaling logs between the user equipment (UE) and the base station (gNB) and the core network functions (such as AMF, SMF). Specifically, it includes:

[0134] RRC layer parameters: Extract the delay distribution (such as average delay, maximum delay) and signaling interaction frequency (such as the number of RRC requests per second) from the processes of RRC connection establishment, reconfiguration, handover, etc.;

[0135] NAS layer parameters: Statistically calculate the retransmission rate and failure reason values (such as #68 network congestion) from the non-access stratum signaling (such as attachment request, session management);

[0136] Device feature parameters: Associate the device manufacturer and model through the TAC code (Type Allocation Code), and obtain the device hardware configuration information in combination with the OAM (Operation, Administration and Maintenance system);

[0137] Service feature parameters: Based on the PDU session ID and 5QI value, extract the service type (voice, video, data) and the corresponding QoS requirements (such as guaranteed bit rate, packet loss rate).

[0138] These parameters are collected in real time by the NWDAF from the signaling logs provided by the AMF and the session data reported by the SMF, and a structured data set is formed through data preprocessing (such as deduplication and standardization).

[0139] Device identifier refers to the credential that uniquely identifies a user device. For example, it can be the TAC code, which is used to associate the manufacturer and model. The association result refers to the output after matching the signaling feature parameters with the device identifier. For example, parameters such as the RRC reconfiguration delay distribution and the NAS signaling retransmission rate corresponding to a certain SUPI are associated and stored. Specifically, a device identifier field (such as the TAC code for the device model) is embedded in the signaling log to ensure that each signaling record contains the unique device identifier; the NWDAF uses a global transaction ID (such as generated based on the timestamp and base station ID) to align the data from different network functions (AMF, SMF, PCF) in terms of time and space, forming an association table of "device identifier - signaling feature - time window"; when the device undergoes handover or reattachment, the UDM updates the device identifier in the user subscription data in real time and synchronizes it to the NWDAF through the Nudm interface to ensure the timeliness of the association result. The finally generated association result is stored in the form of key-value pairs. For example:

[0140] Key: TAC code (manufacturer A - model X);

[0141] Value: Mean RRC reconfiguration delay = 50ms, NAS retransmission rate = 2%, historical signaling frequency baseline = 2 times / second.

[0142] The signaling behavior pattern library (which can also be called the signaling fingerprint library) stores the signaling behavior data of different user devices. For example, it can include: manufacturer device characteristics (such as the average RRC reconfiguration delay of a certain manufacturer's device is 50ms), normal behavior baselines (such as the mean and standard deviation of historical signaling frequencies), and abnormal pattern markers (such as records of specific devices frequently triggering signaling storms), etc. Specifically, historical signaling features are grouped according to the device identifier (such as the TAC code) to form a device-level data set; statistical features (such as mean, standard deviation, percentile) and temporal features (such as trend changes within a sliding window) are extracted from each group of data; a reference model of normal signaling behavior of the device is established through machine learning (such as clustering algorithms) or statistical analysis (such as Gaussian distribution fitting); the signaling feature combinations of historical abnormal events (such as signaling storms, frequent handover failures) are recorded and the root causes are marked (such as device compatibility issues); the pattern data is classified and stored in the UDR according to the device identifier to support fast retrieval and update. For example, the entries in the signaling behavior pattern library of a certain manufacturer's device (TAC code 12345) include: Normal range: RRC reconfiguration delay ≤ 80ms, NAS retransmission rate ≤ 5%; Abnormal marker: If the delay > 120ms and the retransmission rate > 10%, it is determined as "device firmware defect".

[0143] The network quality adjustment method provided by the embodiments of the present invention can accurately identify the typical signaling behavior characteristics of different devices by dynamically associating the signaling characteristic parameters of user equipment with device identifiers and classifying and constructing a signaling behavior pattern library, effectively solving the network compatibility problems caused by differences in signaling implementation of devices from different manufacturers, and at the same time providing a reliable basis for abnormal signaling detection and targeted optimization.

[0144] Step S406: Obtain the current signaling characteristics of the user equipment; compare the current signaling characteristics with the historical signaling behavior characteristics corresponding to the device identifier in the signaling behavior pattern library, and determine whether the user equipment is in an abnormal signaling state based on the comparison result.

[0145] The current signaling characteristics refer to the signaling parameters generated by the user equipment during real-time communication. Specifically, the AMF obtains the RRC connection status (such as the Inactivity Timer value) and the RACH access delay from the gNB; the SMF extracts the QoS flow violation rate (such as the actual rate is lower than 90% of the subscribed value) from the UPF; the NWDAF analyzes the NAS signaling records provided by the AMF (such as the attachment request timestamp and the number of retransmissions). The NWDAF performs real-time processing on the original data, including: statistically analyzing the signaling frequency and delay distribution at a granularity of 5 minutes; associating the service type (such as eMBB, URLLC) and QoS parameters based on the PDU session ID; binding the real-time characteristics with the device identifier to form the current signaling feature vector (such as [delay = 70ms, retransmission rate = 8%, frequency = 4 times / second]), and reporting it to the PCF through the Nnwdaf interface for policy generation.

[0146] An abnormal signaling state refers to a significant deviation of the real-time signaling behavior of a user equipment from the normal baseline in the historical pattern library. For example: abnormal frequency (the number of signaling interactions exceeds two standard deviations of the historical mean), abnormal latency (the RRC reconfiguration latency is higher than 30% of the characteristic value of the manufacturer's equipment), and abnormal retransmission (the NAS signaling retransmission rate exceeds the maximum threshold allowed by the subscribed SLA), etc. Specifically, abnormal detection is achieved by comparing the current signaling characteristics with the historical baseline in the signaling behavior pattern library. NWDAF extracts the corresponding normal behavior model from the signaling behavior pattern library according to the device identifier (such as the TAC code), including parameters such as the mean and standard deviation of the latency. Then, deviation calculation is performed, and statistical test methods are used to determine whether the current value exceeds the historical distribution, such as whether the latency exceeds the mean plus three standard deviations. In addition, a rule engine is also used to make a judgment according to the predefined threshold rules. For example, an alarm is triggered when the NAS retransmission rate exceeds 10%. To avoid misjudgment, multi-dimensional correlation analysis can also be performed by combining spatio-temporal characteristics (such as base station load and UE density) and service context (such as high-priority voice services). Finally, the abnormal signaling state is determined based on the detection result. For example, for a device (TAC code is 12345), the current RRC latency is 130 ms (while the historical mean is 80 ms and the standard deviation is 10 ms), and the NAS retransmission rate is 12% at the same time. At this time, NWDAF will determine that the device is in an abnormal state and associate it with the device compatibility problem.

[0147] Step S407, when the user equipment is in an abnormal signaling state, extract the target device identifier of the user equipment; based on the target device identifier, extract the corresponding initial policy template from the second preset policy template library, and adjust the initial policy template based on the abnormal signaling characteristics corresponding to the abnormal signaling state to generate a second target adjustment policy; adjust the second network parameters of the user equipment and the base station according to the second target adjustment policy to obtain the second network quality.

[0148] The target device identifier refers to the device identifier (such as the TAC code) that triggers the abnormal signaling state and is used to locate the device that needs special processing. Specifically, when the user equipment is determined to be in an abnormal signaling state, by analyzing its signaling characteristic parameters (such as RRC reconfiguration latency, NAS signaling retransmission rate) and comparing them with the historical behavior patterns in the signaling fingerprint library, abnormal characteristics that significantly deviate from the normal range are identified (such as the signaling frequency exceeding two standard deviations of the historical mean). Subsequently, the unique identifier of the device, such as the TAC code, is extracted, and the device manufacturer and model are associated (such as the TAC code of XX device manufacturer and model is "861234").

[0149] The second preset policy template library stores predefined policy templates for different devices or abnormal scenarios. For example, for a certain manufacturer's device, the RRC Inactivity Timer is fixed at 60 seconds; when the signaling frequency limit is exceeded, API rate limiting or forced backoff timer is triggered, etc. The initial policy template refers to the basic policy framework pre-stored in the second preset policy template library, which contains default adjustment rules for specific device manufacturers or service types. Abnormal signaling features refer to characteristic parameters that deviate from the historical signaling behavior pattern of the device. For example, the RRC reconfiguration delay exceeds 2 times the standard deviation of the mean of similar devices; the NAS signaling retransmission rate is significantly higher than the benchmark value in the manufacturer's device fingerprint library; the signaling frequency of devices associated with a specific TAC code fluctuates abnormally (such as a sudden increase of 500%), etc. The second target adjustment policy refers to the optimized policy dynamically generated for the abnormal signaling state. Specifically, the initial policy template is matched from the second preset policy template library according to the target device identifier (such as the TAC code). For example, the historical adaptability problem of a certain manufacturer's device (TAC code is XYZ) may cause the dynamic Inactivity Timer parameter to fail, and the initial template may contain the default rule of fixing this parameter at 60 seconds. Subsequently, combined with the abnormal signaling features (such as the NAS signaling retransmission rate as high as 30%), the template is dynamically adjusted: the variables in the template (such as the Inactivity Timer value) are associated with real-time abnormal data, for example, the fixed value is adjusted to a dynamic range (50s - 70s); if the abnormality involves high-priority services (such as voice), then the QoS flow priority is reduced to release resources; according to the hierarchical rules (emergency services > regular services) and the principle of the latest timestamp, multi-policy conflicts are resolved. The finally generated second target adjustment policy (such as "fix the Inactivity Timer of device with TAC code XYZ at 60 seconds + limit API access to 20 QPS") is sent to the AMF / SMF for execution by the PCF.

[0150] The second network parameters refer to the adjustable parameters related to signaling optimization and resource allocation, including: RRC / NAS parameters (such as Inactivity Timer, RACH resource allocation ratio); the upper limit of the QoS flow guaranteed bit rate (GBR); the UPF shunt weight or the transmission network DSCP marking priority, etc. The second network quality refers to the network performance indicators after applying the second target adjustment policy. Specifically, the adjusted RRC parameters (such as the Inactivity Timer fixed at 60 seconds) are sent to the UE through the N1 NAS signaling to limit its signaling interaction frequency; the RACH resource allocation scheme is configured for the gNB through the N2 interface to reduce the access conflict of abnormal devices; the SMF adjusts the shunt weight of the UPF (such as the N3 / N6 interface bandwidth ratio changes from 7:3 to 6:4) to optimize the service traffic distribution.

[0151] The network quality adjustment method provided by the embodiments of the present invention can accurately identify abnormal signaling states by comparing the current signaling characteristics of user equipment with the historical signaling behavior pattern library in real time, and quickly match the preset policy templates based on the device identifier. By dynamically adjusting the policy templates in combination with abnormal features to generate customized adjustment policies, it realizes the differentiated optimization of network parameters for different device manufacturer characteristics and abnormal scenarios, improves both the accuracy and response efficiency of abnormal detection, and avoids device compatibility problems caused by the "one-size-fits-all" policy. Finally, it significantly improves the network quality through dynamic parameter adjustment while reducing the manual intervention cost.

[0152] Step S408: Obtain the user layer resource utilization rate after network parameter adjustment according to the first target adjustment policy and / or the second target adjustment policy; determine the current deviation degree based on the difference between the user layer resource utilization rate and the preset resource utilization rate; determine the target alarm level corresponding to the current deviation degree based on the second mapping relationship between the deviation degree and the alarm level; execute the inspection policy corresponding to the target alarm level to generate the corresponding inspection result.

[0153] The user layer resource utilization rate refers to the resource occupancy ratio of the user plane function (UPF). The preset resource utilization rate is a predefined resource usage target value. For example, the reduction rate of the user layer resource utilization rate should reach or exceed 20%. Specifically, the user layer resource utilization rate is obtained by SMF and AMF collecting the status data of the user plane and the control plane in real time. SMF extracts the utilization rate distribution of each PRB (statistical by service type) and the QoS flow violation rate (the actual rate is lower than 90% of the subscribed GBR) from the UPF through the N4 interface. AMF obtains the percentile of the RACH access delay (such as P50 / P95) and the distribution of signaling rejection reason values (such as the proportion of #68 network congestion) from the gNB through the N2 interface. These data are preprocessed (such as removing transient fluctuation outliers, aggregating by 5QI classification) and time-aligned (using TAI and second-level timestamps), and finally reported to NWDAF. The policy version ETag and the execution area TAI list are carried in the data label to ensure that NWDAF can accurately associate the policy execution effect with the change of resource utilization rate.

[0154] The deviation degree calculation formula is: For example, if the preset PRB utilization rate is 70% and the actual is 80%, the deviation degree is: This indicator quantifies the gap between the actual resource usage and the expected target, providing a basis for subsequent alarm classification. The calculation is completed by NWDAF, which dynamically compares the KPI target (such as the expected PRB utilization rate reduction rate ≥ 20%) defined in the policy template with the actual value.

[0155] The target alarm level refers to the alarm level divided according to the deviation degree. Specifically, the second mapping relationship defines the corresponding rules between the deviation degree and the alarm level:

[0156] When the deviation is less than 10%, it is regarded as normal, only the log is recorded and no intervention is required;

[0157] When the deviation is between 10% and 30%, it is regarded as a low risk, and the OAM system is notified to generate an operation and maintenance work order;

[0158] When the deviation reaches or exceeds 30%, it is regarded as a high risk, triggering an AUSF security review and suspending policy execution to prevent potential security issues or performance degradation.

[0159] Based on the calculated deviation value, NWDAF matches the mapping relationship through the rule engine and automatically triggers the alarm actions corresponding to the corresponding level.

[0160] Different alarm levels correspond to differentiated inspection strategies. Specifically, when the alarm level is normal, NWDAF only records the deviation data in the log file for subsequent trend analysis; when the alarm level is low risk, OAM generates an operation and maintenance task according to the work order template (such as optimizing the UPF shunt weight or adjusting the RACH resource allocation), and assigns it to the network administrator for processing; when the alarm level is high risk, AUSF starts the security handling process, verifies the policy execution signature to prevent malicious injection, UDM freezes the current policy version and generates a security event report, detailing the abnormal timeline, the scope of influence (TAI list) and the associated device list.

[0161] Finally, NWDAF and OAM cooperate to complete the root cause analysis (such as configuration errors or device compatibility issues), and execute the automated recovery mechanism according to the analysis results (such as configuration rollback or issuing device-specific policies), while generating an inspection result report, including handling measures, effect verification and subsequent optimization suggestions.

[0162] The network quality adjustment method provided by the embodiments of the present invention can accurately quantify the network performance deviation by obtaining the user layer resource utilization rate after adjusting the network parameters in real time and calculating its deviation from the preset value. Based on the preset alarm level mapping relationship, hierarchical alarm response is realized, and differentiated inspection strategies can be dynamically triggered according to the deviation degree, which not only avoids overprocessing of minor deviations, but also can take timely intervention for high-risk deviations, thus improving the accuracy and response efficiency of network anomaly detection. At the same time, by automatically executing the inspection strategy matching the alarm level, the manual operation and maintenance cost is significantly reduced, and the controllability and traceability of the network optimization process are ensured.

[0163] Below this embodiment, a complete intelligent data management and congestion control optimization system process based on the 5G network will be used to exemplarily illustrate the above network quality adjustment method and system, as Figure 5 shown.

[0164] When the signal of the serving cell of the UE is lower than the threshold, the UE establishes an RRC connection. It sends an enhanced measurement report to the gNB through the RRC connection, which includes RSRP and RSRQ and is used to reflect the quality of the radio link between the current UE and the base station; in addition, it also includes a service characteristic identifier, that is, the PDU session ID associated with the 5QI, which is used to describe the service currently being performed by the UE (voice, video, data download); and it also includes an implicit location identifier based on the SUPI to avoid directly transmitting coordinate information and protect user privacy. Through these data, the network side can understand the network environment and requirements of the UE in real time. After receiving the enhanced measurement report of the UE, the gNB forwards these reports to the AMF through the N2 interface. During the forwarding process, the gNB attaches the radio resource status and the time synchronization identifier. The radio resource status includes the available PRB ratio, the PDCCH CCE utilization rate, and the RACH load level (0-15 levels); the time synchronization identifier includes the gNB local timestamp, which is used to cooperate with the TAI to achieve cross-node time synchronization, and the measurement report sequence number, which is used to prevent replay attacks.

[0165] After the AMF receives the measurement report forwarded by the gNB, it performs UE access permission verification. It queries the user subscription data of the UDM through the Nudm service-based interface to confirm whether the UE has access permission. The user subscription data includes the SUPI / SUCI, the list of authorized S-NSSAIs, and the subscribed QoS profile. Based on this information, the AMF performs dynamic access control. If the 5QI value in the measurement report belongs to the subscribed QoS profile and the PRB utilization rate of the serving cell is lower than the maximum threshold of the subscribed SLA, the UE is allowed to access the network and the subsequent analysis process is triggered; otherwise, access is denied and the cause value #68 (network congestion) is returned. This process ensures that only UEs that meet the subscription conditions and whose network resources permit can access the network, and at the same time provides a trigger condition for subsequent intelligent data analysis. The SMF provides session-level data to the NWDAF through the Nsmf interface, and these data reflect the network resource usage and service performance. It includes the PDU session ID, the specific QoS flow parameters (including the 5QI value, the downlink and uplink guaranteed bitrates, and the packet loss rate), and the UPF load statistics (including the N3 interface throughput, the N6 interface latency, and the buffer occupancy rate). The data is reported every 5 minutes, and when the load change rate exceeds 10% / min, the report is immediately triggered. The PCF injects policy control parameters into the NWDAF through the Npcf interface, which are used to guide the NWDAF in data analysis and policy generation. It mainly includes network congestion control rules and data collection compliance constraints. The network congestion control rules define the 5QI ranges corresponding to services with different priorities and the corresponding actions. The data collection compliance constraints specify the anonymization level of the data, the data retention period, and the geographical restrictions. The policy version management adopts the ETag mechanism to achieve atomic updates of the policies, and when conflicts occur, the policy with the latest timestamp is preferentially selected. This ensures the timely update and consistency of the policies, while meeting compliance requirements such as GDPR and protecting user privacy. After receiving the data provided by the AMF, SMF, and PCF, the NWDAF establishes an analysis context. It correlates these data sources and generates a global transaction ID to ensure the end-to-end traceability of the entire process. At the same time, the NWDAF starts a data timeliness monitoring mechanism to ensure the real-time and accurate nature of the data. If the data is not updated after timing out, the NWDAF will trigger an alarm to remind the network administrator to intervene. This mechanism provides a reliable data basis for subsequent intelligent data analysis and ensures the efficiency and reliability of the entire process.

[0166] The NWDAF first performs spatio-temporal correlation analysis, correlating the TAI historical records of the UE with the NR Cell ID (5G cell identifier) trajectory to achieve spatio-temporal positioning with high precision. The geographical coordinates of the base station are obtained through OAM to provide a geographical reference for the base station load data. The PRB utilization rate of the base station is aligned according to a 15-minute time window. Then the dynamic congestion index (DCI) is calculated, and the formula is:

[0167] When the DCI exceeds the threshold (0.75) and lasts for 3 time windows, this area is marked as a congestion hotspot. Then, fine-grained signaling feature extraction is performed. The NWDAF analyzes the signaling logs provided by the AMF, and at the same time, extracts vendor feature parameters, including the device model associated with the TAC code, the RRC reconfiguration delay distribution, and the NAS signaling retransmission rate statistics. Through these signaling features, the NWDAF constructs a signaling fingerprint database for vendor devices. Then, interpretable load prediction is carried out. The NWDAF performs engineering processing on the input features of the model. The time series features include the load data of the past 2 hours, with a granularity of 5 minutes; the spatial features include the correlation coefficient of the loads of the adjacent 3 base stations; and the service features include the proportion of GBR services. Based on these features, the NWDAF generates a load heat map, and the load levels are divided into green (0.0 - 0.3), yellow (0.3 - 0.6), and red (0.6 - 1.0). At the same time, the prediction confidence is displayed through the standard deviation interval output by the model, providing an interpretable basis for network optimization. Then, the NWDAF selects a desensitization scheme according to the GDPR policy issued by the PCF. The desensitization levels are divided into Level 1 (TAI-level location, 500-meter grid) and Level 2 (urban area-level, 3-kilometer grid). The service types are mapped to standardized service categories, including eMBB, URLLC, and mMTC. In addition, different API access rate limits (QPS) are assigned to third-party applications to ensure the security and compliance of data opening. The NWDAF also outputs signaling optimization strategy proposals to the NEF, including recommended values for backoff parameters. These proposals are based on the analysis of UE behavior and network load, aiming to optimize the signaling behavior of UEs and reduce unnecessary signaling interactions. For desensitization level Level 1, the NWDAF recommends an API speed limit of 100 QPS; while for services with desensitization level Level 2, the API speed limit is 50 QPS. Through these optimization strategies, the NWDAF can improve the utilization efficiency of network resources, reduce the occurrence of signaling storms, and thus improve the overall performance and user experience of the network.

[0168] After that, the NWDAF collaborates with the PCF to initiate the policy generation process according to dynamic trigger conditions. The trigger conditions include congestion situations and signaling anomalies. When the DCI continuously exceeds the limit, the NWDAF will trigger the generation of congestion policies. In addition, if the signaling frequency exceeds two standard deviations of the historical mean, indicating a signaling anomaly, the policy generation will also be initiated. The NWDAF submits the analysis results, including DCI and signaling fingerprints, to the PCF through the Nnwdaf interface. The PCF extracts the matching policy templates from the UDR. Then, the policy variables are bound, and the policy actions are sorted according to the 5QI priority (sorting rule: voice service > video service > data service). For the congestion control policy, the PCF issues a QoS adjustment policy to the SMF through the Npcf interface to adjust the bandwidth allocation ratio of the N3 / N6 interfaces of the UPF to alleviate network congestion. In terms of the signaling optimization policy, the PCF issues RRC / NAS parameters to the AMF, dynamically setting the range of the RRC Inactivity Timer from 10s to 300s to optimize the signaling interaction efficiency. The AMF then executes these control instructions, issuing a RACH resource allocation plan to the gNB through the N2 interface. At the same time, backoff parameters are issued to the UE through the N1 NAS signaling. The backoff time for voice services is dynamically calculated based on the current PRB utilization rate, and the backoff time for data services is adjusted according to the DCI value, thus achieving precise control of UE behavior. At the same time, the PCF can also be responsible for resolving potential conflicts between policies. The PCF follows the hierarchical priority rule, with emergency service policies taking precedence over regular service policies. For conflicts between policies of the same level, arbitration is carried out according to the principle that the policy with the latest timestamp takes precedence. At the same time, the UDM is responsible for recording the policy version history, including the Etag and the effective timestamp, to ensure the traceability and consistency of the policies.

[0169] During policy execution, the network needs to track execution status in real time to ensure policy effectiveness and enable timely adjustments. The SMF extracts user plane status data from the UPF via the N4 interface, including per-PRB utilization distribution (by service type) and QoS flow violation rate (actual rate less than 90% of the contracted GBR). This data is pre-processed to remove transient outliers and aggregated by 5QI classification to more clearly present the execution status of different service types. The AMF obtains control plane status data from the gNB, including percentiles of RACH access latency (e.g., P50 / P95) and distribution of signaling rejection reason values (e.g., the proportion of #68 network congestion). This data undergoes normalization, including time alignment (using TAI and second-level timestamps) and device vendor tagging (based on TAC codes) to ensure data consistency and comparability. Finally, the SMF and AMF report the collected data to the NWDAF. The data tag includes the policy version ETag and the list of execution region TAIs, allowing the NWDAF to accurately identify the data source and associated policy version. NWDAF is responsible for identifying abnormal patterns in policy execution. First, it defines a baseline for expected performance, based on the KPI targets in the policy template. For example, the expected PRB utilization rate should decrease by 20% or more. Then, it calculates the actual deviation.

[0170] Depending on the degree of deviation, NWDAF triggers different levels of alarms:

[0171] When the deviation is less than 10%, it is considered normal and only needs to be recorded in the log without intervention;

[0172] When the deviation is between 10% and 30%, it is considered low risk and the OAM system is notified to generate an operation and maintenance work order;

[0173] When the deviation reaches or exceeds 30%, it is considered high risk, triggering an AUSF security review and suspending policy execution to prevent potential security issues or performance degradation.

[0174] The NWDAF collaborates with the OAM system to conduct root cause analysis of anomalies. Through the correlation analysis engine, it correlates policy execution data with network configuration change logs to check whether there are network element upgrades or parameter misconfigurations during the policy effective period. At the same time, it conducts joint analysis of signaling plane and user plane data to identify whether incorrect control plane parameters lead to deterioration of user plane performance. In addition, the NWDAF also conducts special reviews of vendor equipment, counts the abnormal policy execution rate according to the TAC code, and matches the signaling fingerprint library to identify device compatibility issues. For example, UEs of certain vendors may have poor adaptability to the dynamic Inactivity Timer, which may affect the execution effect of policies. When the NWDAF detects a high-risk deviation, the AUSF will initiate a security handling process. First, the AUSF verifies the policy execution signature to prevent malicious policy injection. The UDM then freezes the current policy version and prohibits new policies from overwriting to ensure network security. At the same time, a security event report is generated, detailing the abnormal timeline, the affected scope (TAI list), and the associated device list. According to the results of the root cause analysis, different automated recovery mechanisms are adopted:

[0175] If the root cause is a network configuration error, the OAM system automatically rolls back the most recent configuration change and performs recovery based on the configuration version library;

[0176] If the root cause is a device compatibility issue, the PCF issues a device-specific policy, such as setting the fixed Inactivity Timer to 60 seconds, to address the adaptation problem of specific devices.

[0177] By analyzing historical data over the past 30 days, the NWDAF identifies periodic congestion patterns, such as network congestion during the daily evening peak. Based on these analysis results, the NWDAF can generate predictive capacity expansion plans. These plans can automatically increase the capacity of the target cell 1 hour before the expected congestion occurs, thus effectively alleviating potential network congestion problems. This predictive capacity expansion not only improves the network's response speed but also optimizes resource allocation, ensuring that the network can still provide high-quality services during peak hours. The NWDAF dynamically adjusts the shunt weight of the UPF according to the PRB utilization rate on the radio side. This adjustment can ensure the reasonable allocation of network resources among different service types and optimize the overall network performance. At the same time, the core network and the transport network work together to optimize QoS mapping, converting 5QI into the DSCP marking of the transport network. This linkage mechanism can ensure that traffic of different service types obtains corresponding service quality guarantees during transmission, thereby enhancing the end-to-end performance of the network. Through lightweight cross-domain coordination, the network can achieve efficient resource coordination and optimization between different domains, ensuring that the network always maintains high performance and high reliability in a complex and changing environment.

[0178] In this embodiment, a network quality adjustment device is further provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0179] This embodiment provides a network quality adjustment device, as Figure 6 shown, including:

[0180] A first acquisition module 601, configured to acquire network status data and resource status data of the service area where the user equipment is located when the network quality of the user equipment is poor;

[0181] A first extraction module 602, configured to extract a first key field of the network status data and a second key field of the resource status data;

[0182] A first determination module 603, configured to determine the current congestion status of the base station corresponding to the service area and a first target adjustment policy matching the current congestion status based on the verification results of the first key field and the second key field;

[0183] A first adjustment module 604, configured to adjust the first network parameters of the user equipment and the base station according to the first target adjustment policy to obtain a first network quality.

[0184] In some optional implementation manners, the first determination module 603 includes:

[0185] An acquisition sub-module, configured to acquire a preset user subscription configuration;

[0186] A verification sub-module, configured to verify the first key field and the second key field according to the user subscription configuration to obtain a first data verification result corresponding to the first key field and a second data verification result corresponding to the second key field;

[0187] An analysis sub-module, configured to analyze the first data verification result and / or the second data verification result to determine the current congestion status of the base station when the first data verification result and / or the second data verification result indicates that the user equipment passes the verification;

[0188] A matching sub-module, configured to match the current congestion status with a preset adjustment policy and determine a first target adjustment policy corresponding to the current congestion status based on the matching result.

[0189] In some optional implementation manners, the verification sub-module includes:

[0190] A first determination unit, configured to determine a quantization metric value that matches the session identifier;

[0191] A first detection unit, configured to detect whether the quantization metric value is within a preset range of the quality of service profile, and obtain a first data verification result;

[0192] A second detection unit, configured to detect whether the physical resource block utilization rate is lower than a maximum physical resource block utilization rate threshold, and obtain a second data verification result.

[0193] In some alternative embodiments, the analysis sub-module includes:

[0194] A first acquisition unit, configured to acquire the coverage ranges of respective base stations corresponding to a service area;

[0195] A second determination unit, configured to determine, based on the verification result, the number of target user equipment corresponding to each base station, where the target user equipment is user equipment that passes the verification;

[0196] A third determination unit, configured to determine, based on the coverage range and the number of target user equipment, the user equipment density corresponding to each base station;

[0197] A fourth determination unit, configured to determine, based on the user equipment density and the physical resource block utilization rate in the resource status data, the congestion index corresponding to each base station; the congestion index is used to characterize the current congestion status.

[0198] In some alternative embodiments, the matching sub-module includes:

[0199] A second acquisition unit, configured to acquire a first preset policy template library, where the first preset policy template library includes a first mapping relationship between a congestion status and a preset adjustment policy template;

[0200] An extraction unit, configured to extract, based on the current congestion status, a first target adjustment policy corresponding to the current congestion status from the first preset policy template library.

[0201] In some alternative embodiments, the network quality adjustment device further includes:

[0202] A second acquisition module, configured to acquire signaling feature parameters of user equipment;

[0203] An association module, configured to associate the signaling feature parameters with the device identifier of the user equipment to generate an association result;

[0204] A construction module, configured to construct a signaling behavior pattern library classified by device identifier based on the association result, where the signaling behavior pattern library is used to identify signaling behavior characteristics corresponding to different device identifiers.

[0205] In some alternative embodiments, the network quality adjustment device further includes:

[0206] A third acquisition module, configured to acquire the current signaling feature of the user device;

[0207] A comparison module, configured to compare the current signaling feature with the historical signaling behavior features of the corresponding device identifier in the signaling behavior pattern library, and determine whether the user device is in an abnormal signaling state based on the comparison result;

[0208] A second extraction module, configured to extract the target device identifier of the user device when the user device is in an abnormal signaling state;

[0209] A third extraction module, configured to extract the corresponding initial policy template from the second preset policy template library based on the target device identifier, and adjust the initial policy template based on the abnormal signaling feature corresponding to the abnormal signaling state to generate a second target adjustment policy;

[0210] A second adjustment module, configured to adjust the second network parameters of the user device and the base station according to the second target adjustment policy to obtain a second network quality.

[0211] In some alternative embodiments, the network quality adjustment device further includes:

[0212] A fourth acquisition module, configured to acquire the user layer resource utilization rate after the network parameters are adjusted according to the first target adjustment policy and / or the second target adjustment policy;

[0213] A second determination module, configured to determine the current deviation degree based on the difference between the user layer resource utilization rate and the preset resource utilization rate;

[0214] A third determination module, configured to determine the target alarm level corresponding to the current deviation degree based on the second mapping relationship between the deviation degree and the alarm level;

[0215] A generation module, configured to execute the inspection policy corresponding to the target alarm level and generate a corresponding inspection result.

[0216] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding above-mentioned embodiments, and will not be repeated here.

[0217] The network quality adjustment device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0218] The network quality adjustment device provided by the embodiments of the present invention can quickly identify the congestion status of a base station by obtaining network status data and resource status data in real time and accurately extracting key fields for verification, so as to dynamically match the optimal adjustment strategy, realizing the automatic detection and intelligent decision-making of network quality problems. By specifically adjusting the network parameters of user equipment and base stations, the response speed and accuracy of congestion control are significantly improved, while reducing the dependence on manual intervention, ensuring the efficient utilization of network resources and the continuous optimization of user experience.

[0219] The embodiments of the present invention also provide a computer device having the above-mentioned Figure 6 network quality adjustment device.

[0220] Please refer to Figure 7 , Figure 7 which is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention. As Figure 7 shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (such as a server array, a set of blade servers, or a multi-processor system). Figure 7 In

[0221] <is taken as an example of one processor 10.

[0222] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiments.

[0223] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely disposed relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0224] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memory.

[0225] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0226] Embodiments of the present invention also provide a computer-readable storage medium. The methods according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and to be stored in a local storage medium, so that the methods described herein can be stored in such software processes on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium may further include a combination of the above types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.

[0227] A part of the present invention can be applied as a computer program product, for example, computer program instructions, which, when executed by a computer, can call or provide the methods and / or technical solutions according to the present invention through the operations of the computer. Those skilled in the art should understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0228] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for adjusting network quality, characterized in that, The method includes: When the network quality of the user equipment is poor, obtaining the network status data and resource status data of the service area where the user equipment is located; Extracting a first key field of the network status data and a second key field of the resource status data; Based on the verification results of the first key field and the second key field, determining the current congestion status of the base station corresponding to the service area and a first target adjustment strategy matching the current congestion status; Adjusting the first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain a first network quality.

2. The method according to claim 1, wherein The determining the current congestion status of the base station corresponding to the service area and a first target adjustment strategy matching the current congestion status based on the verification results of the first key field and the second key field includes: Obtaining a preset user subscription configuration; Verifying the first key field and the second key field according to the user subscription configuration to obtain a first data verification result corresponding to the first key field and a second data verification result corresponding to the second key field; When the first data verification result and / or the second data verification result indicates that the user equipment passes the verification, analyzing the first data verification result and / or the second data verification result to determine the current congestion status of the base station; Matching the current congestion status with a preset adjustment strategy, and determining the first target adjustment strategy corresponding to the current congestion status based on the matching result.

3. The method according to claim 2, wherein The user subscription configuration includes a quality of service profile and a maximum physical resource block utilization threshold; the first key field includes a session identifier, and the second key field includes a physical resource block utilization rate; The verifying the first key field and the second key field according to the user subscription configuration to obtain a first data verification result corresponding to the first key field and a second data verification result corresponding to the second key field includes: Determining a quantization index value matching the session identifier; Detecting whether the quantization index value is within a preset range of the quality of service profile to obtain the first data verification result; Detecting whether the physical resource block utilization rate is lower than the maximum physical resource block utilization threshold to obtain the second data verification result.

4. The method according to claim 2, wherein Analyzing the first data verification result and / or the second data verification result to determine the current congestion status of the base station includes: Obtaining the coverage ranges of the respective base stations corresponding to the service area; Determining the target number of user equipment corresponding to each base station based on the verification results, where the target user equipment is the user equipment that passes the verification; Based on the coverage range and the target number of user equipment, determining the user equipment density corresponding to each base station; Based on the user equipment density and the physical resource block utilization rate in the resource status data, determining the congestion index of each base station; the congestion index is used to characterize the current congestion status.

5. The method according to claim 2 or 4, characterized in that, Matching the current congestion state with a preset adjustment strategy, and determining the first target adjustment strategy corresponding to the current congestion state based on the matching result, including: Obtaining a first preset policy template library, where the first preset policy template library includes a first mapping relationship between a congestion state and a preset adjustment policy template; Based on the current congestion state, extracting the first target adjustment strategy corresponding to the current congestion state from the first preset policy template library.

6. The method according to claim 1, wherein It further includes: Obtaining the signaling feature parameters of the user equipment; Associating the signaling feature parameters with the device identifier of the user equipment to generate an association result; Based on the association result, constructing a signaling behavior pattern library classified by the device identifier, where the signaling behavior pattern library is used to identify the signaling behavior characteristics corresponding to different device identifiers.

7. The method according to claim 6, characterized in that It further includes: Obtaining the current signaling feature of the user equipment; Comparing the current signaling feature with the historical signaling behavior features of the corresponding device identifier in the signaling behavior pattern library, and determining whether the user equipment is in an abnormal signaling state based on the comparison result; When the user equipment is in an abnormal signaling state, extracting the target device identifier of the user equipment; Based on the target device identifier, extracting the corresponding initial policy template from a second preset policy template library, and adjusting the initial policy template based on the abnormal signaling features corresponding to the abnormal signaling state to generate a second target adjustment strategy; Adjusting the second network parameters of the user equipment and the base station according to the second target adjustment strategy to obtain a second network quality.

8. The method according to claim 7, wherein It further includes: Obtaining the user layer resource utilization rate after network parameter adjustment according to the first target adjustment strategy and / or the second target adjustment strategy; Based on the difference between the user layer resource utilization rate and a preset resource utilization rate, determining the current deviation degree; Based on a second mapping relationship between the deviation degree and an alarm level, determining the target alarm level corresponding to the current deviation degree; Executing an inspection strategy corresponding to the target alarm level to generate a corresponding inspection result.

9. A network quality adjustment system, characterized in that, The system includes: A user equipment, which is used to monitor the signal quality of the service area where it is located. When the signal quality is lower than a preset threshold, obtaining the network status data of the service area; A base station, which is communicatively connected to the user equipment, and is used to receive the network status data sent by the user equipment, and obtain the resource status data of the service area; A server cluster, which is communicatively connected to the user equipment and the base station respectively, and is used to receive the network status data and the resource status data sent by the base station, extract the first keyword field of the network status data and the second keyword field of the resource status data, and determine the current congestion state of the base station corresponding to the service area and the first target adjustment strategy matching the current congestion state based on the verification results of the first keyword field and the second keyword field, and adjusting the first network parameters of the user equipment and the base station according to the first target adjustment strategy to obtain a first network quality.

10. A computer device, characterized in that, It includes: A memory and a processor, which are communicatively connected to each other. Computer instructions are stored in the memory, and the processor executes the computer instructions to perform the method for adjusting network quality according to any one of claims 1 to 8.

Images