Directional traffic service guarantee method of PCC self-intelligence network based on AI intelligent agent

Abstract

The invention belongs to the technical field of communication, and particularly relates to a directional traffic service guarantee method of a PCC self-intelligence network based on an AI agent. Comprising the following steps: an AI identifies a live broadcast application and issues a policy control and charging PCC rule; service experience detection: the NWDAF receives user flow data, and the NWDAF continuously monitors the experience quality of the live broadcast service; establishing a service awareness baseline: establishing the service awareness baseline for different live broadcast applications through AI learning: based on a user level signing protocol, analyzing experience class data of a service flow in combination with a network element; according to the experience class data analysis result of the service flow, when the service experience index is lower than the service awareness baseline, the definite positioning analysis is carried out, then a dynamic guarantee mechanism is triggered according to needs, and oriented dynamic QoS optimization is carried out on a user network. According to the invention, the problems of network congestion and network resource waste of the PCC self-intelligence network are solved.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a directional traffic service assurance method for a PCC self-intelligent network based on an AI agent. Background Art

[0002] With the rapid development of technologies such as 5G, cloud computing, the Internet of Things, and artificial intelligence, networks are facing increasingly complex challenges. How to improve the intelligence level of the network and realize network automation, self-adaptation, self-optimization, and self-healing, self-intelligent networks have emerged. Self-intelligent networks are a new type of network architecture based on artificial intelligence technology. Self-intelligent networks can realize autonomous perception, autonomous decision-making, and autonomous execution of the network.

[0003] PCC (Policy Control and Charging) self-intelligent network is an automated and intelligent network architecture for policy control and charging under the framework of self-intelligent network. The existing protection strategy based on PCC directional traffic service is generated on PCF (Policy and Charging Enforcement Function), which has the following shortcomings:

[0004] 1. The standard process, PCC parameters and judgment logic are relatively rigid and not very scalable. 2. As long as the customer uses the service, the guarantee is turned on, which consumes a lot of network resources; especially during peak service periods, this strategy may aggravate network congestion and affect user experience. 3. The existing user's QOS (Quality of Service) guarantee strategy cannot automatically provide QOS guarantees to users according to network conditions, resulting in high-value users' service experience being affected when the network is congested, while low-value users occupy a large amount of network resources.

[0005] However, users have a strong demand for high quality of mobile networks, especially for e-commerce services such as live streaming and walking broadcasts. Customers are concerned about issues such as buffering delays and freezes in video services and the inability to open mini-programs. If the experience is not good for a long time, complaints will arise.

[0006] AI agents can proactively understand the goals set by users, and then carefully plan a series of specific steps required to achieve the goals. This shift from passive acceptance to proactive planning, execution, and adjustment is the significant advantage and difference of AI agents compared to traditional artificial intelligence systems. On the other hand, the AI ​​unit in the 5G core network, the network data analysis function (NWDAF), can output accurate prediction, reasoning, and decision-making results, providing strong support for network operation and maintenance.

[0007] The self-intelligent network works in coordination with the AI ​​agent and the NWDAF, providing an effective means for processing complex scene analysis and executing tasks. To this end, the present invention provides a directional traffic service assurance method for a PCC self-intelligent network based on an AI agent, so as to solve the problems of network congestion and network resource waste in the PCC self-intelligent network. Summary of the invention

[0008] In order to solve the above technical problems, namely the network congestion and network resource waste existing in the PCC self-intelligent network.

[0009] The present invention provides the following technical solutions:

[0010] The method for ensuring directional traffic services of a PCC self-intelligent network based on an AI agent includes the following steps:

[0011] AI identifies live broadcast applications and issues policy control and billing PCC rules: AI technology is used to identify directional traffic services including live broadcast application traffic in the network. The network will dynamically issue policy control and billing PCC rules to users based on their contract information and current network conditions.

[0012] Service experience detection: Based on AI recognition of live broadcast applications, NWDAF receives user traffic data and continuously monitors the experience quality of live broadcast services;

[0013] Establishing a service perception baseline: Establishing a service perception baseline for different live broadcast applications through AI learning, which serves as an important basis for evaluating whether the live broadcast service experience meets the standards;

[0014] Experience data analysis of user service flows: Based on user-level contract agreements, experience data analysis of service flows combined with network elements;

[0015] Implement delimited positioning analysis: Based on the experience data analysis results of the service flow, when the service experience index is lower than the service perception baseline, implement delimited positioning analysis, and then trigger the dynamic guarantee mechanism as needed to perform targeted dynamic service quality QoS optimization on the user network.

[0016] Furthermore, the policy control and charging PCC rules include trigger conditions at the cell granularity. When the live broadcast user is located in a specific cell, a specific network policy or action will be triggered.

[0017] Furthermore, PCC rules are issued when a user session is established, and during the session, PCC rules are dynamically adjusted as needed and the adjusted PCC rules are issued. The adjusted PCC rules define how the network charges, controls and manages the user's data traffic.

[0018] Furthermore, through the NWDAF visual page, various service experience indicators of the experience quality of live user services are intuitively displayed, including but not limited to service experience indicators such as the uplink rate, latency, and jitter of the live broadcast picture.

[0019] Furthermore, the user-level contract agreement is signed using the accounting system, down to the granularity of single-user services.

[0020] Furthermore, delimited positioning analysis is implemented, including when the live broadcast service experience of VIP users declines, NWDAF reports a poor quality event, and the reported information includes the user's real-time location information, and obtains the capacity and resource usage of each cell; based on the real-time location information of the delimited positioning analysis and the capacity and resource usage of the cell, the dynamic guarantee mechanism is then triggered as needed.

[0021] Furthermore, the dynamic guarantee mechanism includes:

[0022] Control plane instructions to establish a dedicated bearer: When necessary, the network will use the control plane instructions to establish a dedicated bearer to provide services to users;

[0023] Instruct the user plane to establish a dedicated bearer: After the C plane instructs to establish a dedicated bearer, the user plane is responsible for actually establishing the dedicated bearer;

[0024] Air interface resource reservation and 5QI scheduling: After the dedicated bearer is established, the network will reserve the necessary air interface resources for the service. At the same time, the network will schedule according to parameters including 5QI to ensure that the service can obtain sufficient network resources.

[0025] After the dedicated bearer is established, the service is carried on the dedicated bearer: Once the dedicated bearer is successfully established and configured, the network will instruct the user to establish a dedicated bearer through the control plane to provide services to the user when needed.

[0026] Furthermore, targeted dynamic service quality QoS optimization of user networks includes: in network congestion scenarios, for users who have signed up for guaranteed bit rate (GBR) services, when a decline in service experience is detected, the network will trigger the GBR guarantee mechanism, dynamically adjust the service quality QoS retransmission strategy based on the user's service needs and contract information, and allocate more resources to the user.

[0027] Furthermore, the quality of service (Qos) reprint strategy includes adjusting parameters including 5QI, guaranteed bit rate and MBR maximum bit rate (GBR).

[0028] Furthermore, the NWDAF topology network is constructed as follows:

[0029] UPF enables the real-time user flow data recording and reporting function, and sends user flow data to NWDAF every second;

[0030] NWDAF configures the uplink bandwidth baselines of VIP users' eight live broadcast applications, and displays the difference between the VIP users' actual experience indicators and the target baselines in real time. If the target is not met for seven consecutive cycles, the guarantee process is automatically triggered.

[0031] As an AF application function, NWDAF automatically generates N5 interface messages, which contain key information such as user number and target uplink bandwidth;

[0032] The PCF policy control function generates a target 5QI based on the information provided by the NWDAF, and the PCF sends the target 5QI and guaranteed bandwidth information to relevant network elements;

[0033] SMF dedicated bearer establishment process: The SMF session management function receives the policy information sent by the PCF and initiates a dedicated bearer establishment request to the AMF and RAN;

[0034] Based on the request of SMF, UPF establishes a dedicated bearer process for VIP users to ensure the QoS of data transmission;

[0035] The wireless network side performs differentiated scheduling of VIP users’ traffic based on 5QI information to ensure service priority and bandwidth guarantee;

[0036] After the VIP user's live broadcast ends, NWDAF sends a notification SMS to the SMS agent via the HTTPS protocol to inform the user that the guarantee has ended;

[0037] For live broadcast services, UPF generates user-level aggregate data records (ADRs) for subsequent business analysis and optimization.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention provides a comprehensive and intelligent guarantee strategy for directional traffic services such as live broadcasting through steps such as AI recognition of live broadcasting applications, perception of service experience, establishment of service perception baseline, and on-demand triggering of dynamic guarantee mechanisms. It realizes the intention-driven dynamic experience guarantee of live broadcasting services and solves the problems of network congestion and network resource waste in PCC self-intelligent networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a target framework diagram of the self-intelligent network of the present invention;

[0041] Figure 2 It is a structural diagram of the network data analysis function (NWDAF) of the present invention;

[0042] Figure 3 This is a schematic diagram of a 5G core network based on NWDAF of the present invention;

[0043] Figure 4It is a schematic diagram of the NWDAF networking architecture and interfaces of the present invention;

[0044] Figure 5 The figure is a flow chart of the directional flow service assurance method of the present invention. DETAILED DESCRIPTION

[0045] The technical solution of the present invention will be clearly described below in conjunction with the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention, and all other embodiments obtained by ordinary technicians in the field without making creative work are within the protection scope of the present invention.

[0046] The self-intelligent network is a new network architecture based on artificial intelligence technology. Its ultimate goal is to improve the overall quality and operational efficiency of the communication network. The self-intelligent network can realize the network's autonomous perception, autonomous decision-making, and autonomous execution. Its core essence lies in the deep integration of automation and intelligence. Figure 1 The following is the target framework diagram of the self-intelligent network.

[0047] AI agents are designed to be highly autonomous intelligent entities that can not only perceive their surroundings acutely, but also make decisions independently and perform corresponding actions. Unlike the passive response mode based on user prompt input commonly used in current large models, AI agents can actively understand the goals set by users, and then carefully plan a series of specific steps required to achieve the goals. By continuously receiving external feedback and combining its own autonomous reasoning ability, AI agents can creatively construct effective prompts to achieve predetermined goals in an autonomously driven manner. This transformation from passive acceptance to active planning, execution and adjustment is the significant advantage and difference of AI agents compared to traditional artificial intelligence systems.

[0048] like Figure 2 As shown in the figure, the AI ​​unit in the 5G core network, the Network Data Analysis Function (NWDAF), has built a closed-loop control system that integrates data collection, model training, and reasoning analysis, and supports a network big data analysis architecture that can respond to diverse needs. As the intelligent hub and big data power source of the 5G network, NWDAF is responsible for comprehensively collecting massive data from network management, business operations, network infrastructure, and terminal devices. Through an efficient data model training process, NWDAF can output accurate prediction, reasoning, and decision-making results, providing strong support for network operations and maintenance.

[0049] The collaboration between the self-intelligent network, AI agent, and network data analysis function (NWDAF) provides an effective means to ensure the directional traffic services of the PCC self-intelligent network.

[0050] The present invention mainly adopts 5G network networking, such as Figure 3 The figure shows a schematic diagram of the 5G core network based on NWDAF. The Network Data Analytics Function (NWDAF) integrates AI technology. NWDAF is a new function added to the 5G core network intelligent architecture. It collects data from network functions, OAM and application layers and uses artificial intelligence technology for analysis. For example, operation data, performance data, load data, etc. are collected from access and AMF (mobile management function), SMF (session management function), etc., and AI agents are used for analysis and prediction, and then the algorithm model is applied to the 5G network. Figure 3 The network elements are as follows:

[0051] Core Access and Mobility Management Function (AMF);

[0052] Session Management Function (SMF);

[0053] Unified Data Management (UDM);

[0054] User plane function (UPF);

[0055] Policy Control function (PCF);

[0056] Network Data Analytics Function (NWDAF);

[0057] Network Function Repository (NF Repository Function, NRF);

[0058] Network Exposure Function (NEF);

[0059] Unified Data Repository (UDR);

[0060] Network Slice Selection Function (NSSF);

[0061] Authentication Server Function (AUSF);

[0062] Application Function (AF);

[0063] Data Network (DN).

[0064] The network data analysis function (NWDAF) has unified data aggregation, model training and network strategy generation, and central AI reasoning training to form a large closed loop of network-level analysis and decision-making. By analyzing the UE experience status, network element service status, and applying intelligent models and algorithms, it generates UE-level, network element-level and network-level control strategies, including UE-level experience guarantee and congestion collaborative control, to improve the quality and efficiency of the entire network operation. Figure 4 The NWDAF topology network construction of the present invention is as follows:

[0065] 1) UPF enables real-time UFDR reporting (every second): UPF enables real-time UFDR (User Flow Data Record) reporting function and sends user flow data to NWDAF every second.

[0066] 2) NWDAF configuration and application baseline establishment: NWDAF configures the uplink bandwidth baseline of VIP users' 8 live broadcast applications, and displays the difference between the target baseline of the actual experience index of VIP users in real time. If the target is not met for 7 consecutive cycles (adjustable), the guarantee process will be automatically triggered.

[0067] 3) NWDAF generates N5 interface message: As AF (Application Function), NWDAF automatically generates N5 interface message, which contains key information such as user number and target uplink bandwidth.

[0068] 4) PCF policy generation and delivery: PCF (Policy Control Function) generates the target 5QI (5G QoS Identifier) ​​based on the information provided by NWDAF. PCF delivers the target 5QI and guaranteed bandwidth information to relevant network elements.

[0069] 5) SMF dedicated bearer establishment process: SMF (Session Management Function) receives the policy information sent by PCF and initiates a dedicated bearer establishment request to AMFRAN (referring to the joint function of AMF and RAN).

[0070] 6) UPF establishes a dedicated bearer process: Based on the request of SMF, UPF establishes a dedicated bearer process for VIP users to ensure the QoS of data transmission.

[0071] 7) Wireless differentiated scheduling: The wireless network side performs differentiated scheduling of VIP users’ traffic based on 5QI information to ensure service priority and bandwidth guarantee.

[0072] 8) End of user protection: After the VIP user’s live broadcast ends, NWDAF sends a notification SMS to the SMS proxy through the HTTPS protocol to inform the user that the protection has ended.

[0073] 9) User-level aggregated ADR generation: For live broadcast services, UPF generates user-level aggregated ADR (Aggregated Data Record) for subsequent business analysis and optimization.

[0074] Example

[0075] like Figure 5 As shown, the present invention provides a directional traffic service assurance method for a PCC self-intelligent network based on an AI agent, wherein the directional traffic service assurance method uses the analysis and processing of network elements such as the core network UPF and NWDAF to identify dynamically triggered 5G QoS assurance capabilities. Through AI learning, an application assurance baseline is dynamically established. When the quality of VIP users is poor, a dedicated GBR is dynamically triggered, an experience report is produced, and a VIP assurance effect report is pushed. Specifically, the following steps are included:

[0076] 1. AI identification of live broadcast applications: AI technologies including deep learning and machine learning are used to identify directional traffic services including live broadcast application traffic in the network. Based on the identification of live broadcast application traffic, the network management system can manage it in a targeted manner to ensure the priority and stability of the live broadcast service.

[0077] 2. Send PCC rules: When a live broadcast user tries to establish a session, the network will dynamically send policy control and charging (PCC) rules to the user based on the user's contract information and current network conditions. These rules contain trigger conditions (Trigger) at the cell granularity, which means that when the user is in a specific cell, a specific network policy or action will be triggered. In addition to sending them when the session is established, PCC rules are dynamically adjusted as needed during the session and the adjusted PCC rules are sent. These rules define how the network charges, controls, and manages the user's data traffic.

[0078] 3. Send user flow data to NWDAF: UPF enables the real-time UFDR (User Flow Data Record) reporting function and sends user flow data to NWDAF every second.

[0079] 4. Service experience detection: Based on AI recognition of live broadcast applications, NWDAF receives user traffic data and continuously monitors the experience quality of live broadcast services. NWDAF perceives the experience quality of live broadcast services, including but not limited to service experience indicators such as uplink rate, latency, and jitter of live broadcast images. Through the NWDAF visual page, various service experience indicators of live broadcast user experience are intuitively displayed. NWDAF is used to build an intelligent experience network that is user-perceivable, service-identifiable, and optimizes self-enclosed loops.

[0080] 4. Establish a service perception baseline: Based on the service perception feature library, through AI learning, establish a service perception baseline for different live broadcast applications; this baseline is an important basis for evaluating whether the live broadcast service experience meets the standards. Ensure the standardization and measurability of the service experience.

[0081] 5. Accurate analysis of user experience data on service flows: Through the contract signing of the accounting system, it can be refined to the granularity of single-user services to achieve personalized experience guarantee. Through the analysis of user-level contract agreements by the AI ​​intelligent feature library, combined with the experience data analysis of service flows by network elements, it can be refined to the experience data analysis of single-user service flows to comprehensively judge the optimization needs.

[0082] 6. Implement delimited positioning analysis: Based on the experience data analysis results of the service flow, when the service experience index is lower than the service perception baseline, implement delimited analysis. For example, if the live broadcast service experience of VIP users declines (such as freezes, increased delays, etc.), the network will immediately report these poor quality events, and the reported information will include the user's real-time location information; the network management system will obtain the capacity and resource usage of each cell in real time. Then trigger the dynamic guarantee mechanism as needed;

[0083] 7. Closed-loop security management, differentiated optimization of user networks: The network automatically triggers dynamic QoS (quality of service) and other optimization actions to provide users with differentiated business experience services. In special scenarios such as network congestion, a special channel is opened for the user to ensure a smooth live broadcast experience in special scenarios such as network congestion, thus realizing closed-loop security management.

[0084] For users who have signed up for guaranteed bit rate (GBR) services, when the service experience is detected to be degraded, the network will trigger the GBR guarantee mechanism and dynamically adjust the quality of service (QoS) retransmission strategy based on the user's service needs and contract information, including adjusting parameters such as 5QI, GBR (guaranteed bit rate) and MBR (maximum bit rate). More resources are allocated to users to ensure that their service experience meets the predetermined standards.

[0085] Dynamic protection mechanisms include:

[0086] C-plane instructs the establishment of a dedicated bearer: When necessary, the network will instruct the establishment of a dedicated bearer (dedicated bearer) through the control plane (C-plane) to provide services to users;

[0087] Instruct the U-plane to establish a dedicated bearer: Once the C-plane instructs to establish a dedicated bearer, the user plane (U-plane) will be responsible for actually establishing this dedicated bearer;

[0088] Air interface resource reservation and 5QI scheduling: After the dedicated bearer is established, the network will reserve the necessary air interface resources for the service. At the same time, the network will schedule according to parameters such as 5QI to ensure that the service can obtain sufficient network resources;

[0089] After the dedicated bearer is established, the service is carried on the dedicated bearer: Once the dedicated bearer is successfully established and configured, when the user needs it, the network will instruct the establishment of a dedicated bearer (dedicated bearer) through the control plane (C plane) to provide services to the user.

[0090] 8. Experience report generation: After the guarantee service is completed, the VIP live broadcast user terminal can immediately receive an automatically generated experience report. The report includes the guarantee duration and various experience indicators, providing users with intuitive service experience feedback.

[0091] Through the analysis and processing of core network UPF (User Plane Function), NWDAF and other network elements, the 5G network QoS assurance capability is dynamically triggered by application identification. Through AI learning, the application assurance baseline is dynamically established. When the quality of VIP users is poor, the GBR dedicated carrier is dynamically triggered and the VIP assurance effect report is pushed.

[0092] In summary, the present invention provides a comprehensive and intelligent guarantee strategy for directional traffic services such as live broadcasting through steps such as AI recognition of live broadcasting applications, perception of service experience, establishment of service perception baseline, and on-demand triggering of dynamic guarantee mechanism, thus realizing the intention-driven dynamic experience guarantee of live broadcasting services.

[0093] The directional flow service assurance method according to the present invention is used to test the directional flow service:

[0094] Intelligent core network live broadcast acceleration traffic background: a computer is connected to an engineering mobile phone (5qi=5) for packet injection, a computer is connected to a CPE (5qi=5) for packet injection, and two commercial mobile phones are connected to a CPE (5qi=5) for Baidu live broadcast business.

[0095] Before packet injection, ordinary users (5qi=9) and VIP users (5qi=9) were doing Baidu live broadcasting. The UDN real-time report showed that the uplink bandwidth was about 2Mb / s, and the customers could watch the live broadcast on their mobile phones without any lag.

[0096] After packet filling, if the VIP user's bandwidth is lower than the threshold (2Mb / s) for 7 consecutive seconds, the VIP user (5qi=4) triggers the GBR guarantee. The UDN real-time report shows that the VIP user's uplink bandwidth is about 2Mb / s, and the ordinary user's bandwidth is lower than 500Kb / s. The customer's mobile phone has no lag when watching the VIP user's live broadcast, but the ordinary user's live broadcast has serious lag. After the VIP user's live broadcast ends, a user experience report is sent to the customer's mobile phone, and each parameter is consistent with the actual situation.

[0097] In the NWDAF live broadcast report, it can be observed that before congestion, the uplink rate of ordinary users and VIP users doing live broadcast services was about 2.3 Mb / s; after congestion, the uplink rate of VIP users still met the guarantee strategy, achieving accurate guarantee of key networks and key user experience.

[0098] It can be seen that the present invention can improve user perception and reduce user complaints by finely identifying and guaranteeing services. User behavior diagnosis and analysis can be performed on services in the network from different dimensions (global level, server level, and cell level) to guide network optimization, so that operators can select different dimensions to count corresponding indicators according to their own network operation status, give corresponding guarantee strategies, and guarantee the user experience of key networks and important customer groups.

[0099] Test the maximum network speed for VIP users relative to ordinary users:

[0100] In a cell with an average wireless load of 60%, speed test software was used at the same location and time to test the speed of VIP users and ordinary users. Through multiple groups of service tests, it can be found that the uplink and downlink speeds of users in the GBR guarantee scenario are significantly better than those in the non-GBR guarantee scenario. The test results are shown in Table 1.

[0101] Table 1

[0102]

[0103] By comparing the test results of pilot verification of various scenario combinations, it can be found that NWDAF network intelligence can effectively improve the network experience and network quality of live broadcast services based on user perception, and can enable users to perceive the guaranteed situation in real time and improve the service experience. The existing service guarantee strategies of operators mainly include static contract QoS and capability opening. The unconditional guarantee method of static contract has low wireless resource efficiency and may affect the general user experience when the wireless load is heavy; the capability opening method is realized by opening network capabilities to OTT manufacturers. It is difficult for the operator network side to control business needs and intelligent allocation and scheduling of wireless resources. There are many OTT manufacturers and it is difficult to promote them one by one. After the introduction of NWDAF, the network is dynamically monitored through real-time data collection and analysis, and the guarantee strategy is activated on demand based on predetermined rules, such as poor quality, to improve resource utilization efficiency and service experience, which can effectively make up for the shortcomings of the existing service experience guarantee solution.

[0104] The above technical features constitute the best embodiment of the present invention, which has strong adaptability and best implementation effect. Non-essential technical features can be added or reduced according to actual needs to meet the needs of different situations.

[0105] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A directional traffic service assurance method for a PCC self-intelligent network based on an AI agent, characterized in that: The steps include: AI identifies live broadcast applications and issues policy control and billing PCC rules: AI technology is used to identify directional traffic services including live broadcast application traffic in the network. The network will dynamically issue policy control and billing PCC rules to users based on their contract information and current network conditions. Service experience detection: Based on AI recognition of live broadcast applications, NWDAF receives user traffic data and continuously monitors the experience quality of live broadcast services; Establishing a service perception baseline: Establishing a service perception baseline for different live broadcast applications through AI learning, which serves as an important basis for evaluating whether the live broadcast service experience meets the standards; Experience data analysis of user service flows: Based on user-level contract agreements, experience data analysis of service flows combined with network elements; Implement delimited positioning analysis: Based on the experience data analysis results of the service flow, when the service experience index is lower than the service perception baseline, implement delimited positioning analysis, and then trigger the dynamic guarantee mechanism as needed to perform targeted dynamic service quality QoS optimization on the user network.

2. The directional traffic service assurance method according to claim 1, characterized in that: The policy control and charging PCC rules contain trigger conditions at the cell granularity. When the live broadcast user is located in a specific cell, a specific network policy or action will be triggered.

3. The directional traffic service assurance method according to claim 2, characterized in that: PCC rules are issued when a user session is established. During the session, PCC rules are dynamically adjusted as needed and the adjusted PCC rules are issued. The adjusted PCC rules define how the network charges, controls, and manages the user's data traffic.

4. The method for ensuring directional traffic services according to claim 1, characterized in that: Through the NWDAF visual page, various service experience indicators of the experience quality of live user services are intuitively displayed, including but not limited to service experience indicators such as the uplink rate, latency, and jitter of the live broadcast picture.

5. The directional traffic service assurance method according to claim 1, characterized in that: The user-level contract agreement is signed using the accounting system, down to the granularity of single-user business.

6. The method for ensuring directional traffic services according to claim 5, characterized in that: Implement delimited positioning analysis, including when the live broadcast service experience of VIP users deteriorates, NWDAF reports a poor quality event, the reported information includes the user's real-time location information, and obtains the capacity and resource usage of each cell; based on the real-time location information of the delimited positioning analysis and the capacity and resource usage of the cell, the dynamic guarantee mechanism is then triggered as needed.

7. The directional traffic service assurance method according to claim 6, characterized in that: Dynamic protection mechanisms include: Control plane instructions to establish a dedicated bearer: When necessary, the network will use the control plane instructions to establish a dedicated bearer to provide services to users; Instruct the user plane to establish a dedicated bearer: After the C plane instructs to establish a dedicated bearer, the user plane is responsible for actually establishing the dedicated bearer; Air interface resource reservation and 5QI scheduling: After the dedicated bearer is established, the network will reserve the necessary air interface resources for the service. At the same time, the network will schedule based on parameters including 5QI. After the dedicated bearer is established, the service is carried on the dedicated bearer: Once the dedicated bearer is successfully established and configured, the network will instruct the user to establish a dedicated bearer through the control plane to provide services to the user when needed.

8. The method for ensuring directional traffic services according to claim 1, characterized in that: Targeted dynamic service quality QoS optimization of user networks includes: in network congestion scenarios, for users who have signed up for guaranteed bit rate (GBR) services, when the service experience is detected to have declined, the network will trigger the GBR guarantee mechanism, and dynamically adjust the service quality QoS retransmission strategy based on the user's business needs and contract information to allocate more resources to the user.

9. The method for ensuring directional traffic services according to claim 8, characterized in that: The quality of service (Qos) reprint strategy includes adjusting parameters such as 5QI, guaranteed bit rate and MBR maximum bit rate (GBR).

10. The method for ensuring directional traffic services according to claim 1, characterized in that: NWDAF topology network construction is as follows: UPF enables the real-time user flow data recording and reporting function, and sends user flow data to NWDAF every second; NWDAF configures the uplink bandwidth baselines of VIP users' eight live broadcast applications, and displays the difference between the VIP users' actual experience indicators and the target baselines in real time. If the target is not met for seven consecutive cycles, the guarantee process is automatically triggered. As an AF application function, NWDAF automatically generates N5 interface messages, which contain key information such as user number and target uplink bandwidth; The PCF policy control function generates a target 5QI based on the information provided by the NWDAF, and the PCF sends the target 5QI and guaranteed bandwidth information to relevant network elements; SMF dedicated bearer establishment process: The SMF session management function receives the policy information sent by the PCF and initiates a dedicated bearer establishment request to the AMF and RAN; Based on the request of SMF, UPF establishes a dedicated bearer process for VIP users to ensure the QoS of data transmission; The wireless network side performs differentiated scheduling of VIP users’ traffic based on 5QI information to ensure service priority and bandwidth guarantee; After the VIP user's live broadcast ends, NWDAF sends a notification SMS to the SMS agent via the HTTPS protocol to inform the user that the guarantee has ended; For live broadcast services, UPF generates user-level aggregate data records (ADRs) for subsequent business analysis and optimization.

Images