Group management methods, systems, and storage media for reducing UE latency differences

Abstract

This invention discloses a group management method, system, and storage medium to reduce UE latency differences. The invention involves the AF (Active Provider) requesting the UDM (User Device Manager) to construct UE groups for application services. After the UE establishes a session with the AF and engages in traffic exchange, the UE initiates a new service request to the AF. The AF sends a new Quality of Service (QoS) request to the PCF (Process Control Fund), prompting the PCF to request the NWDAF (Network Window Data Center) to perform UE group analysis. The NWDAF analyzes the UEs initiating new service requests using an integrated strong classifier to classify them into corresponding groups. Based on the analysis results, the PCF issues QoS policies and routing policies to the SMF (Service Management Fund). The SMF generates QoS rules based on the QoS policies and routing rules based on the routing policies. UE traffic is then routed via ULCL (Universal Flow Control Class) and PSA (Power Segment Allocation Class) to the EAS (Electronic Access Provider) serving the same UE group. This invention enables dynamic UE group management and reduces latency differences within UE groups.

Description

Technical Field

[0001] This invention relates to the field of 5G communication technology, and in particular to a group management method, system and storage medium for reducing UE latency differences. Background Technology

[0002] In related technologies, many applications, such as XR applications, consist of multiple application components running on different UEs for purposes such as identifying, modeling, and rendering a set of objects. Processing of application components for multiple UEs should run on the same EAS instance to reduce latency differences between applications on different UEs. Existing technologies primarily manage a group of UEs through pre-configured UE groups. However, for many applications, the UEs may change dynamically, making pre-configured UE groups inflexible, resulting in inconvenient and slow applications, and failing to effectively guarantee latency differences between UEs, thus impacting the user's network experience. Summary of the Invention

[0003] This invention aims to address at least one of the technical problems existing in the prior art. To this end, this invention proposes a group management method, system, and storage medium for reducing UE latency differences, which can effectively reduce UE latency differences and enhance network flexibility.

[0004] On one hand, embodiments of the present invention provide a group management method for reducing UE latency differences, comprising the following steps:

[0005] AF requests UDM to build a UE group for application services;

[0006] After the UE establishes a session with the AF and performs traffic interaction, the UE initiates a new service request to the AF.

[0007] AF sends a new quality of service request to PCF based on the new service request;

[0008] PCF requests NWDAF to perform UE group analysis based on the new Quality of Service Request information;

[0009] NWDAF analyzes UEs that initiate new service requests by integrating a strong classifier, so as to classify the UEs that initiate new service requests into the corresponding groups;

[0010] Based on the analysis results, PCF issues service quality policies and routing policies to SMF;

[0011] SMF generates quality of service rules based on the quality of service policy and routing rules based on the routing policy, and distributes the quality of service rules and routing rules to ULCL and PSA;

[0012] Traffic from a UE initiating a new service request is routed through ULCL and PSA to the EAS serving the same UE group.

[0013] In some embodiments, the AF request UDM to build a UE group for application services includes:

[0014] The AF sends a UE group creation request to the NEF. The creation request includes the UE's GPSI list, EAS deployment information, UE group type, and UE group description information.

[0015] NEF authorizes the AF to create the request and forwards the creation request to the UDM;

[0016] The UDM assigns a group identifier to the UE group corresponding to the creation request and requests that the UE group information be stored in the UDR;

[0017] Once the UE group information is successfully stored in the UDR, the UDM will respond with a NEF.

[0018] The AF initiates an event notification request to the NEF. The event notification request includes an event identifier, which includes UE group information notification and EAS deployment information.

[0019] NEF authorizes the event notification request and forwards the event notification request to UDR;

[0020] After UDR responds to NEF, NEF responds to AF.

[0021] In some embodiments, when the PCF requests the NWDAF to perform UE group analysis according to a new Quality of Service Request instruction, if the NWDAF is performing UE group analysis for the first time, the method further includes the following steps:

[0022] NWDAF requests a UE group dataset from UDR to train a preset classifier using the UE group dataset, which includes a UE group identifier and historical service instruction indicator monitoring samples for each UE in the UE group.

[0023] In some embodiments, training a preset classifier using the UE group dataset includes:

[0024] Initialize the weights of each UE sample within the UE group dataset;

[0025] The weak learning classifier is iteratively trained using a UE group dataset with preset weights;

[0026] By constructing a linear combination of iteratively trained weak learning classifiers, an ensemble strong classifier is obtained.

[0027] In some embodiments, the iterative training of the weak learning classifier using a UE group dataset with preset weights includes:

[0028] The UE group dataset with preset weights is input into a weak learning classifier, which includes either a linear regression model or a decision tree.

[0029] Calculate the error of the weakly learned classifier between each category;

[0030] The weight of the weak learning classifier for each category in the final classifier is calculated based on the error.

[0031] The weights are updated based on the stated proportions.

[0032] In some embodiments, the PCF issues quality of service policies and routing policies to the SMF based on the analysis results, including:

[0033] Based on the analysis results, the PCF requests the UDR to update the UE group information;

[0034] The UDR adds the UE that initiates a new service request to the corresponding group and responds with the corresponding UE group information to the PCF.

[0035] The UDR notifies the AF of the UE group update request so that the AF can respond to the UDR;

[0036] PCF sends quality of service policies and routing policies to SMF.

[0037] In some embodiments, the quality of service request information includes quality of service parameters and an application identifier, wherein the quality of service parameters include latency data, bandwidth data, packet loss rate data, and reliability data.

[0038] In some embodiments, the deployment information of the EAS includes the data network name, data network access identifier, single network slice selection assistance information, application identifier, fully qualified domain name, and IP address of the edge application server.

[0039] On the other hand, embodiments of the present invention provide a group management system for reducing UE latency differences, including:

[0040] At least one memory for storing programs;

[0041] At least one processor is configured to load the program to execute the group management method for reducing UE latency differences.

[0042] On the other hand, embodiments of the present invention provide a storage medium storing a computer-executable program, which, when executed by a processor, is used to implement the group management method for reducing UE latency differences.

[0043] This invention provides a group management method for reducing UE latency differences, which has the following beneficial effects:

[0044] In this embodiment, the AF requests the UDM to construct a UE group for application services. After the UE establishes a session with the AF and exchanges traffic, the UE initiates a new service request to the AF. Then, the AF sends a new Quality of Service (QoS) request to the PCF based on the new service request, so that the PCF requests the NWDAF to perform UE group analysis based on the new QoS request. Then, the NWDAF analyzes the UE that initiated the new service request through an integrated strong classifier to classify the UE that initiated the new service request into the corresponding group. After the PCF issues QoS policies and routing policies to the SMF based on the analysis results, the SMF generates QoS rules based on the QoS policies and routing rules based on the routing policies, and issues the QoS rules and routing rules to the ULCL and PSA. The traffic of the UE that initiated the new service request is routed to the EAS serving the same UE group through the ULCL and PSA. This embodiment uses the intelligent analysis capabilities of 5GS to classify UEs and realizes dynamic UE group management according to the needs of applications, thereby reducing the latency difference of UE groups, enhancing network flexibility, improving the network intelligence level, and reducing the manpower consumption of network management.

[0045] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0046] The present invention will be further described below with reference to the accompanying drawings and embodiments, wherein:

[0047] Figure 1 This is a flowchart of a group management method for reducing UE latency differences according to an embodiment of the present invention;

[0048] Figure 2 This is a flowchart illustrating an AF request UDM to construct a UE group for application services according to an embodiment of the present invention;

[0049] Figure 3 This is a flowchart illustrating how traffic from a group of UEs is routed to the same EAS according to an embodiment of the present invention. Detailed Implementation

[0050] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0051] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0052] In the description of this invention, "several" means one or more, "multiple" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0053] In the description of this invention, unless otherwise explicitly defined, terms such as "setting," "installing," and "connecting" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0054] In the description of this invention, the terms "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0055] Before describing specific embodiments, the terms used in the embodiments of this application are explained as follows:

[0056] 5GS: The full English name is 5G System.

[0057] 5GC: 5G Core Network, is the core of the 5G mobile network. It establishes reliable and secure network connections for end users and provides access to their services. The core domain handles various essential functions in the mobile network, such as connectivity and mobility management, authentication and authorization, user data management, and policy management. 5G core network functions are entirely software-based and designed to be cloud-native, meaning they are independent of the underlying cloud infrastructure, enabling greater deployment agility and flexibility.

[0058] mMTC stands for Massive Machine Type Communication.

[0059] URLLC stands for Ultra Reliable Low Latency Communication.

[0060] XR: The full English name is Extended Reality.

[0061] SUPI: The full English name is Subscription Permanent Identifier, which means user permanent identifier.

[0062] GPSI stands for Generic Public Subscription Identifier.

[0063] UE stands for User Equipment. User equipment can be a mobile phone, tablet, laptop, or other device.

[0064] SMF stands for Session Management Function. Its functions include session management, such as session establishment, modification, and release, as well as channel maintenance between UPF and AN nodes.

[0065] PCF stands for Policy Control function.

[0066] UDM stands for Unified Data Management, which provides operators with integrated data management for various networking scenarios including 2G, 3G, 4G, and 5G. It has efficient user data processing capabilities, simplifies network setup, is compatible with existing services, can expand 5G services, protects operators' investments, and provides users with the possibility of seamless network switching.

[0067] UDR stands for Unified Data Repository.

[0068] PSA stands for PDU Session Anchor.

[0069] UPF stands for User Plane Function. Its functions include session points for interconnecting external PDUs with data networks, packet routing, and forwarding. For example, it supports uplink classifiers to route traffic to the data network.

[0070] ULCL stands for Uplink Classifier.

[0071] EAS stands for Edge Application Server.

[0072] AF: The full name of the English word is Application Function. It refers to various services at the application layer. It can be an application within the operator (similar to the VoLTE AF in 4G) or a third-party AF (such as a video server or a service server). If it is an AF within the operator, it is in the same trusted domain as other NFs and can directly interact and access other NFs. However, a third-party AF is not in the trusted domain and must access other NFs through a NEF.

[0073] NEF stands for Network Exposure Function, which is located between the 5G core network and external third-party application functionalities. It manages all external applications that want to access network data exposed to the outside world. External applications wishing to access data within the 5G core network must go through the NEF. The NEF provides corresponding security guarantees to ensure the security of external applications accessing the 3GPP network, offering functions such as external application QoS customization capabilities, mobility state event subscription, and AF request distribution.

[0074] DNN: The full English name is Data Network Name.

[0075] DNAI stands for Data Network Access Identifier.

[0076] S-NSSAI stands for Single Network Slice Selection Assistance Information.

[0077] FQDN stands for Fully Qualified Domain Name.

[0078] In related technologies, with the development of technology, communication services have increasingly higher requirements for latency. URLLC and mMTC are application scenarios of 5G. In 5G networks, applications can be deployed on edge nodes to reduce the latency caused by terminal service traffic traversing too many network nodes. However, the widespread deployment of edge nodes and the increasing number of UEs also put a certain burden on UE management. Therefore, under the premise of meeting the latency requirements of UE services, enhancing UE group management is of great significance to the stability, reliability and controllability of the network.

[0079] Based on this, refer to Figure 1 This invention provides a group management method for reducing UE latency differences, including but not limited to the following steps:

[0080] Step S110: AF requests UDM to build a UE group for application services;

[0081] Step S120: After the UE establishes a session with the AF and performs traffic interaction, the UE initiates a new service request to the AF.

[0082] Step S130: AF sends a new quality of service request information to PCF based on the new service request;

[0083] Step S140: PCF requests NWDAF to perform UE group analysis based on the new Quality of Service Request information;

[0084] Step S150: NWDAF analyzes the UE that initiates a new service request by integrating a strong classifier, so as to classify the UE that initiates a new service request into the corresponding group;

[0085] Step S160: PCF issues service quality policies and routing policies to SMF based on the analysis results;

[0086] Step S170: SMF generates quality of service rules based on the quality of service policy and routing rules based on the routing policy, and distributes the quality of service rules and routing rules to ULCL and PSA;

[0087] Step S180: The traffic of the UE that initiates a new service request is routed through ULCL and PSA to the EAS serving the same UE group.

[0088] In this embodiment of the application, when the AF requests the UDM to build a UE group for the application service, such as Figure 2 As shown, including but not limited to the following steps:

[0089] The AF sends a UE group creation request to the NEF. The creation request includes the UE's GPSI list, EAS deployment information, UE group type, and UE group description information. The EAS deployment information includes the data network name (DNN), data network access identifier (DNAI), single network slice selection assistance information (S-NSSAI), application identifier (Application ID), fully qualified domain name (FQDN), and edge application server IP address (EAS IP address).

[0090] NEF authorizes the AF's creation request and forwards the creation request to the UDM;

[0091] The UDM assigns a group identifier to the UE group corresponding to the creation request and requests that the UE group information be stored in the UDR;

[0092] After the UDR responds with a successful UDM group information storage message, the UDM responds with a NEF.

[0093] After NEF responds to AF, AF initiates an event notification request to NEF. The event notification request includes an event identifier, which includes UE group information notification and EAS deployment information.

[0094] NEF authorizes the event notification request corresponding to AF and forwards the event notification request to UDR;

[0095] After UDR responds to NEF, NEF responds to AF.

[0096] In this embodiment of the application, after the AF requests the UDM to build a UE group for the application service, the process of routing traffic of the UE group to the same EAS is executed. Wherein, as... Figure 3 As shown, the process includes, but is not limited to, the following steps:

[0097] Step 1: The UE establishes a session with the AF and performs traffic exchange.

[0098] Step 2: The UE initiates a new service request to the AF, where the new service includes XR service.

[0099] Step 3: The AF requests the PCF to send new Quality of Service (QoS) request information. The QoS request information includes UE SUPI, QoS parameters, and application identifier. The QoS parameters include latency data, bandwidth data, packet loss rate data, and reliability data.

[0100] Step 4: PCF requests NWDAF to perform UE packet analysis. The request message in this step includes an analysis ID, which includes information such as UE packet analysis, UE SUPI, requested QoS parameters (latency, bandwidth, packet loss rate, reliability, etc.), and application ID.

[0101] Step 5: NWDAF analyzes the UEs initiating new service requests using an ensemble strong classifier. In step 5, if this is the first time NWDAF is performing UE group analysis, NWDAF requests the UE group dataset from UDR to train a preset classifier using this dataset. UDR then returns the UE group dataset to NWDAF. The UE group dataset includes the UE group identifier and historical service instruction monitoring samples for each UE within the UE group. If NWDAF already has a trained ensemble strong classifier model, model training is not required.

[0102] Step 7: NWDAF performs group analysis on the UE that initiated the request and assigns the requesting UE to the appropriate group.

[0103] Step 8: NWDAF returns the analysis results to PCF.

[0104] Step 9: The PCF requests the UDR to update the UE group information. The UDR adds the requested UE to the corresponding group and responds to the PCF with the corresponding UE group information. The UE group information includes group ID, DNN / DNAI, S-NSSAI, ApplicationID, FQDN, EAS IP address, etc.

[0105] Step 10: The UDR notifies the AF of the UE group update status, and the AF responds to the UDR.

[0106] Step 11: PCF sends QoS policies and routing policies to SMF.

[0107] Step 12: SMF generates QoS rules and routing rules based on QoS policies and routing policies, and distributes the rules to ULCL and PSA.

[0108] Step 13: SMF responds to PCF.

[0109] Step fourteen: PCF response AF.

[0110] Step 15: AF responds to UE.

[0111] Step 16: UE traffic is routed through ULCL and PSA to the EAS serving the same UE group.

[0112] In this embodiment, when training a preset classifier using a UE group dataset, the weights of each UE sample in the UE group dataset are first initialized; the weak learning classifier is iteratively trained using the UE group dataset with preset weights; and a linear combination of the iteratively trained weak learning classifiers is constructed to obtain an ensemble strong classifier. Exemplarily, the training process can be described as follows:

[0113] Step 1: UE group dataset D = (X, Y), X ∈ R m×n ,Y∈R m×1 Where X represents the historical QoS metrics of the UE, such as latency, bandwidth, packet loss rate, and reliability; m represents the number of samples, i.e., the number of UE samples; n represents the number of features, i.e., the number of features such as latency, bandwidth, packet loss rate, and reliability; and Y represents the tag, i.e., the UE group ID.

[0114] Step 2: Initialize the weights W1 = (w... 11 ,w 12 ,…,w 1m ),in,

[0115] Step 3: Perform iterations 1, 2, ..., T:

[0116] Step 3.1: Assign weights W to the values ​​W. t The UE training dataset is input into a weak learning classifier, where the weak learning classifier includes one or more combinations of logistic regression or decision trees to obtain the base classifier: h t :X→{c1,c2,…,c K};

[0117] Step 3.2: Calculate the error of this classifier: Where I is an indicator function, which takes the value 1 when the condition in parentheses is met, and 0 otherwise;

[0118] Step 3.3: Calculate the weight of this classifier in the final classifier:

[0119] Step 3.4: Update the sample weights according to the proportions: i = 1, 2, ..., m; where Z t As the normalization factor,

[0120] Step 4: Construct a linear combination of weak classifiers to obtain an ensemble strong classifier:

[0121] In summary, the embodiments of this application can implement dynamic UE group management according to application needs; and utilize the analysis capabilities of NWDAF to classify UE groups, thereby improving the intelligence level of the network; simultaneously, based on 5GS dynamic group management, routing UEs within a group to a unified EAS service can reduce latency differences between UEs and enhance user experience. User experience.

[0122] This invention provides a group management system for reducing UE latency differences, comprising:

[0123] At least one memory for storing programs;

[0124] At least one processor is used to load the program for execution. Figure 1 The group management method shown reduces UE latency differences.

[0125] The content of the method embodiments of the present invention is applicable to the system embodiments. The specific functions implemented in the system embodiments are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above methods.

[0126] This invention provides a storage medium storing a computer-executable program, which, when executed by a processor, is used to implement... Figure 1 The group management method shown reduces UE latency differences.

[0127] The content of the method embodiments of the present invention is applicable to the storage medium embodiments. The specific functions implemented by the storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above methods.

[0128] Furthermore, embodiments of the present invention also provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform... Figure 1 The group management method shown reduces UE latency differences.

[0129] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments, and various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention. Furthermore, the embodiments of the present invention and the features thereof can be combined with each other unless otherwise specified.

Claims

1. A group management method for reducing UE latency differences, characterized in that, Includes the following steps: AF requests UDM to build a UE group for application services; After the UE establishes a session with the AF and performs traffic interaction, the UE initiates a new service request to the AF. AF sends a new quality of service request to PCF based on the new service request; PCF requests NWDAF to perform UE group analysis based on the new Quality of Service Request information; NWDAF analyzes UEs that initiate new service requests by integrating a strong classifier, so as to classify the UEs that initiate new service requests into the corresponding groups; Based on the analysis results, PCF issues service quality policies and routing policies to SMF; SMF generates quality of service rules based on the quality of service policy and routing rules based on the routing policy, and distributes the quality of service rules and routing rules to ULCL and PSA; Traffic from a UE initiating a new service request is routed through ULCL and PSA to the EAS serving the same UE group; The AF request UDM constructs a UE group for the application service, including: The AF sends a UE group creation request to the NEF. The creation request includes the UE's GPSI list, EAS deployment information, UE group type, and UE group description information. NEF authorizes the AF to create the request and forwards the creation request to the UDM; The UDM assigns a group identifier to the UE group corresponding to the creation request and requests that the UE group information be stored in the UDR; Once the UE group information is successfully stored in the UDR, the UDM will respond with a NEF. The AF initiates an event notification request to the NEF. The event notification request includes an event identifier, which includes UE group information notification and EAS deployment information. NEF authorizes the event notification request and forwards the event notification request to UDR; After UDR responds to NEF, NEF responds to AF; The PCF issues quality of service policies and routing policies to the SMF based on the analysis results, including: Based on the analysis results, the PCF requests the UDR to update the UE group information; The UDR adds the UE that initiates a new service request to the corresponding group and responds with the corresponding UE group information to the PCF. The UDR notifies the AF of the UE group update request so that the AF can respond to the UDR; PCF sends quality of service policies and routing policies to SMF.

2. The group management method for reducing UE latency differences according to claim 1, characterized in that, When the PCF requests the NWDAF to perform UE group analysis according to the new Quality of Service Request instruction, if the NWDAF is performing UE group analysis for the first time, the method further includes the following steps: NWDAF requests a UE group dataset from UDR to train a preset classifier using the UE group dataset, which includes a UE group identifier and historical service instruction indicator monitoring samples for each UE in the UE group.

3. The group management method for reducing UE latency differences according to claim 2, characterized in that, The step of training a preset classifier using the UE group dataset includes: Initialize the weights of each UE sample within the UE group dataset; The weak learning classifier is iteratively trained using a UE group dataset with preset weights; By constructing a linear combination of iteratively trained weak learning classifiers, an ensemble strong classifier is obtained.

4. The group management method for reducing UE latency differences according to claim 3, characterized in that, The iterative training of the weak learning classifier using a UE group dataset with preset weights includes: The UE group dataset with preset weights is input into a weak learning classifier, which includes either a linear regression model or a decision tree. Calculate the error of the weakly learned classifier between each category; The weight of the weak learning classifier for each category in the final classifier is calculated based on the error. The weights are updated based on the stated proportions.

5. The group management method for reducing UE latency differences according to claim 1, characterized in that, The quality of service request information includes quality of service parameters and application identifiers. The quality of service parameters include latency data, bandwidth data, packet loss rate data, and reliability data.

6. The group management method for reducing UE latency differences according to claim 1, characterized in that, The deployment information of the EAS includes the data network name, data network access identifier, single network slice selection assistance information, application identifier, fully qualified domain name, and IP address of the edge application server.

7. A group management system for reducing UE latency differences, characterized in that, include: At least one memory for storing programs; At least one processor is configured to load the program to execute the group management method for reducing UE latency differences as described in any one of claims 1-6.

8. A storage medium, characterized in that, It contains a computer-executable program, which, when executed by a processor, is used to implement the group management method for reducing UE latency differences as described in any one of claims 1-6.

Images