WO2021012697A1 - 应用mos模型的训练方法、设备及系统 - Google Patents

应用mos模型的训练方法、设备及系统 Download PDF

Info

Publication number
WO2021012697A1
WO2021012697A1 PCT/CN2020/080958 CN2020080958W WO2021012697A1 WO 2021012697 A1 WO2021012697 A1 WO 2021012697A1 CN 2020080958 W CN2020080958 W CN 2020080958W WO 2021012697 A1 WO2021012697 A1 WO 2021012697A1
Authority
WO
WIPO (PCT)
Prior art keywords
uplink
downlink
rate
entity
network
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2020/080958
Other languages
English (en)
French (fr)
Inventor
蒯少锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP20843289.8A priority Critical patent/EP3993320B1/en
Publication of WO2021012697A1 publication Critical patent/WO2021012697A1/zh
Priority to US17/581,935 priority patent/US12113679B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5003—Managing SLA; Interaction between SLA and QoS
    • H04L41/5009—Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/08—Testing, supervising or monitoring using real traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/147—Network analysis or design for predicting network behaviour
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters

Definitions

  • This application relates to the field of communication technology, and in particular to a training method, device and system for applying a mean opinion score (MOS) model.
  • MOS mean opinion score
  • NWDAF network data analysis function
  • 3GPP 3rd generation partnership project 23.791
  • NWDAF can learn from the fifth generation core network (5rd generation core, 5GC) network functions (NFs), application functions (AFs), and operator network operation and maintenance systems (operation, administration, and maintenance).
  • NFs fifth generation core network
  • AFs application functions
  • OAM operator network operation and maintenance systems
  • the observed service experience related network data analytics in 3GPP 23.288 6.4 defines how NWDAF subscribes to data from NF and AF and realizes the establishment of a MOS model for the specified service process.
  • the NWDAF entity subscribes the service quality MOS level of the specified service from the AF entity, and the NWDAF entity subscribes the network performance indicators of the network transmission carrying this service from the 5GC NFs entity, and then the NWDAF entity subscribes to the service quality MOS level, And the network performance index of the network transmission subscribed from the 5GC NFs entity, establish the MOS model of the designated service (that is, the mathematical relationship between the service quality MOS level and the network performance index of the network transmission).
  • the network performance index and service quality MOS level of network transmission need to be subscribed from the 5GC NFs entity and the AF entity respectively. Therefore, to achieve measurement synchronization, it is necessary to rely on the 5GC NFs entity and the AF entity. Ability and cooperation, which is difficult to implement in the current network.
  • the embodiments of the present application provide training methods, equipment and systems for applying the MOS model, which are used to solve the problem that the existing service MOS model training method needs to rely on the capabilities of the AF entity and the cooperation of the AF entity and is difficult to implement in the existing network.
  • a training method for applying an average opinion score MOS model includes: a central network data analysis function (center NWDAF, C-NWDAF) entity to an edge network data analysis function (edge NWDAF, E-NWDAF) entity Sending a first subscription request, where the first subscription request is used to request to subscribe to the service quality MOS level of the target service and the corresponding first network performance index, where the first network performance index is the network performance index of the network transmission that carries the target service;
  • the C-NWDAF entity receives the service quality MOS level and the first network performance index from the E-NWDAF entity; the C-NWDAF entity establishes the MOS model of the target service according to the service quality MOS level and the first network performance index.
  • the service quality MOS level is used to measure the key indicators of business quality in the business process by means of instruments, tools, etc., and then the measurement results of the key indicators are given this time through the experience model.
  • the comprehensive score of the business can also be called the business experience quality score.
  • a 5-point value can be used to give a service quality evaluation score of 1, 2, 3, 4, or 5.
  • establishing the MOS model of the target service refers to establishing the mathematical relationship between the service quality MOS level of the target service and the network performance index, which is explained here in a unified manner, and will not be repeated in the following.
  • the operator deploys distributed NWDAF entities, including C-NWDAF entities and E-NWDAF entities, and C-NWDAF entities can be from E-NWDAF entities
  • the service quality MOS level of the subscribed target service and the corresponding first network performance index are simultaneously measured by the E-NWDAF entity to obtain the service quality MOS level of the target service and the corresponding first network performance index. That is to say, the service quality MOS level of the target service and the corresponding first network performance index are obtained within the operator, so it is easier to implement and the measurement result is more accurate.
  • the service quality MOS level is determined based on the service experience data of the target service.
  • the service experience data includes one or more of the following parameters: initial buffering delay of the video service, playback buffering duration, bit rate, service rate, frame rate, Fluency, clarity; and, the resolution of the terminal device.
  • the bit rate of the video service in the embodiment of the present application refers to the number of data bits transmitted per unit time during data transmission, and the general unit is kilobits per second (kbps).
  • the frame rate of video services is a measurement used to measure the number of display frames, and the unit of measurement is the number of display frames per second ((frames per second, FPS) or Hertz (Hz).
  • the resolution of terminal equipment includes display resolution and image resolution
  • the display resolution can also be called the screen resolution, which refers to the number of pixels that the display can display; the image resolution is the number of pixels contained in a unit inch. This is a unified description and will not be repeated here.
  • the first network performance index includes one or more of the following parameters: the number of uplink packet loss on the path between the terminal device and the access network device, the uplink packet loss rate, and the downlink packet loss Number, downlink packet loss rate, round-trip time (RTT), uplink error packet number, uplink error rate, downlink error packet number, downlink error rate, uplink out-of-sequence packet number, uplink out-of-sequence Packet rate, downlink out-of-sequence packet number, downlink out-of-sequence packet rate, uplink retransmission packet number, uplink retransmission rate, downlink retransmission packet number, downlink retransmission rate, uplink average packet interval, uplink average packet jitter, downlink average packet Interval, downlink average packet jitter, uplink rate, and downlink rate; and, the number of uplink packet loss, uplink packet loss rate, downlink packet loss, and downlink packet loss rate on the path between the access network device and the user
  • RTT round
  • the error code refers to that after receiving, judging, and regenerating, some bits in the digital code stream have errors, which damages the quality of the transmitted information. It can be understood as an error occurred during the transmission process.
  • data pack Out of order means that when a data packet is too large, a data packet will be split into multiple data packets that meet the transmission requirements. Each data packet will have a corresponding sequence number for the opposite end to reorganize, but due to different intermediate routes or network Quality problems may cause the data packets sent later to arrive at the opposite end first, resulting in disorder. Among them, a high proportion of packets disorder will also lead to poor communication quality. Jitter refers to the degree of change in the delay of packet data packets.
  • the user plane entity in the embodiment of the present application is mainly used to perform the forwarding of user data packets.
  • the user plane entity can be a user plane function (UPF) network element.
  • UPF user plane function
  • the user plane entity can still be a UPF network element, or Other names are not limited in the embodiment of this application.
  • the application function entity in the embodiment of the present application is mainly used to provide application layer service functions to terminal devices.
  • the application function entity may be an AF network element.
  • future communications, such as the sixth-generation 6G communication the application function entity may still be an AF network element or have other names. Not limited.
  • the method further includes: the C-NWDAF entity obtains a second network performance index corresponding to the MOS level, where the second network performance index is the network performance index of the wireless carrying the target service; C-NWDAF The entity establishes the MOS model of the target service according to the service quality MOS level and the first network performance index, including: C-NWDAF entity according to the second network performance index, as well as the service quality MOS level and the first network performance index , Eliminate poor service quality MOS samples caused by abnormal terminal equipment and application function entities, and establish a MOS model of the target service.
  • the samples with poor service quality MOS refer to samples with low MOS scores due to the abnormality of the terminal equipment and application functional entities, which are explained here in a unified manner, and will not be repeated here.
  • the MOS model of the target service established based on the training method of applying the MOS model provided by the embodiment of the present application not only considers the first network performance index, but also considers the network performance index of the wireless carrying the target service, as well as the terminal equipment and application.
  • a sample of poor service quality MOS caused by abnormal functional entities, so the MOS model of the target service is more accurate and comprehensive.
  • the method further includes: the C-NWDAF entity obtains the first load of the first network function entity corresponding to the MOS level, the first network function entity is a network data provider function entity; C-NWDAF The entity establishes the MOS model of the target service according to the service quality MOS level and the first network performance index, including: the C-NWDAF entity according to the first load of the first network function entity, and the service quality MOS level and the first network performance index A network performance index, excluding samples of poor service quality MOS caused by abnormal terminal equipment and application function entities, and establishing a MOS model of the target service.
  • the MOS model of the target service established based on the training method of applying the MOS model provided by the embodiment of the present application not only considers the first network performance index, but also considers the first load of the first network functional entity corresponding to the MOS level and the factors A sample of poor service quality MOS caused by the abnormality of terminal equipment and application functional entities, so the MOS model of the target service is more accurate and comprehensive.
  • the method further includes: the C-NWDAF entity obtains the second network performance index corresponding to the MOS level and the first load of the first network function entity corresponding to the MOS level, and the second network performance index
  • the first network function entity is a network data provider function entity
  • the C-NWDAF entity establishes the MOS of the target service according to the service quality MOS level and the first network performance index
  • the model includes: the C-NWDAF entity excludes terminal equipment and application function entities based on the second network performance index and the first load of the first network function entity, as well as the service quality MOS level and the first network performance index A sample of poor service quality MOS caused by the abnormality, and a MOS model of the target service is established.
  • the MOS model of the target service established based on the training method of applying the MOS model provided by the embodiments of the present application takes into account not only the first network performance index, but also the network performance index of the wireless carrying the target service and the first corresponding MOS level.
  • the C-NWDAF entity acquiring the first load of the first network function entity corresponding to the MOS level includes: the C-NWDAF entity sends a second subscription request to the network storage function entity, the second subscription request Used to request to subscribe to the first load of the first network function entity; the C-NWDAF entity receives the first load of the first network function entity from the network storage function entity.
  • the network storage function entity in the embodiment of the present application is used to store the network function entity (such as the first network function entity) and the description information of the services it provides, and to support service discovery, network element or entity discovery, etc.
  • the network storage function entity can be a network storage function (network repository function, NRF) entity.
  • NRF network repository function
  • the network storage function entity can still be an NRF entity, or Other names are not limited in the embodiment of this application.
  • the C-NWDAF entity can obtain the first load of the first network function entity corresponding to the MOS level.
  • the C-NWDAF entity obtains the second network performance index corresponding to the MOS level, including: the C-NWDAF entity sends a third subscription request to the operator's network operation and maintenance OAM entity, the third subscription The request is used to request to subscribe to the second network performance indicator; the C-NWDAF entity receives the second network performance indicator from the OAM entity. Based on this solution, the C-NWDAF entity obtains the second network performance index corresponding to the MOS level.
  • the first load of the first network function entity includes one or more of the following parameters: the number of sessions of the first network function entity, the number of users of the first network function entity, and the The resource utilization rate of the first network function entity. That is, the MOS model of the target service in the embodiment of the present application also includes one of the number of sessions with the first network function entity, the number of users of the first network function entity, and the resource utilization rate of the first network function entity. Or multiple related, so the MOS model of the target business is more complete.
  • the second network performance index includes one or more of the following parameters: the number of sessions of the access network device, the number of radio resource control (radio resource control, RRC) connected users, congestion status, and resources Occupancy rate, wireless measurement indicators of the terminal device, and location information of the terminal device.
  • RRC radio resource control
  • the MOS model of the target service in the embodiment of this application also includes the number of sessions with the access network device, the number of RRC connected users, the congestion state and resource occupancy rate, the wireless measurement index of the terminal device, and the location of the terminal device.
  • One or more of the information is related, so the MOS model of the target business is more complete.
  • the method before the C-NWDAF entity sends the first subscription request to the E-NWDAF entity, the method further includes: the C-NWDAF entity receives a fourth subscription request from the second network function entity, and the fourth The subscription request is used to request to subscribe to the MOS model of the target service.
  • the second network function entity is the consumer function entity; after the C-NWDAF entity establishes the MOS model of the target service, the method further includes: 2.
  • the network function entity sends the MOS model of the target service. That is, in the training method for applying the MOS model provided by the embodiment of the present application, the C-NWDAF entity may send the first subscription request message to the E-NWDAF entity based on the trigger of the consumer functional entity.
  • the consumer function entity in the embodiment of the present application refers to the requester that initiates a data analysis request or subscription to the C-NWDAF entity or the NWDAF entity, which is explained here in a unified manner and will not be repeated in the following.
  • the method further includes: the C-NWDAF entity receives a fifth subscription request from the policy control entity, and the fifth subscription request is used to request service experience guarantee for the target service of the target terminal device.
  • the guaranteed service quality MOS level requirement is the first service quality MOS level;
  • the C-NWDAF entity sends a sixth subscription request to the E-NWDAF entity, and the sixth subscription request is used to request to subscribe to the target terminal device using the target service trigger event ,
  • the second service quality MOS level of the target service and the corresponding first network performance index after the trigger event is triggered, the C-NWDAF entity receives the second service quality MOS level and the corresponding first network performance index from the E-NWDAF entity A network performance index; in the case where the second service quality MOS level is different from the first service quality MOS level, the C-NWDAF entity corresponds to the first network performance index corresponding to the second service quality MOS level and the target service
  • the first network performance index corresponding to the first service quality MOS level in the MOS model
  • the method further includes: the C-NWDAF entity receives a fifth subscription request from the policy control entity, and the fifth subscription request is used to request service experience guarantee for the target service of the target terminal device.
  • the service quality MOS level requirement for experience assurance is the first service quality MOS level; the C-NWDAF entity determines that the target terminal device will move from the service range of the first access network device to the second access device based on the movement trajectory of the target terminal device.
  • a training method for applying an average opinion score MOS model includes: an edge network data analysis function E-NWDAF entity receives a first subscription request from a central network data analysis function C-NWDAF entity, and the first subscription The request is used to request subscription to the service quality MOS level of the target service and the corresponding first network performance index.
  • the first network performance index is the network performance index of the network transmission that carries the target service; the E-NWDAF entity obtains the service quality MOS level And the first network performance indicator; the E-NWDAF entity sends the service quality MOS level and the first network performance indicator to the C-NWDAF entity.
  • the operator deploys distributed NWDAF entities, including C-NWDAF entities and E-NWDAF entities, and C-NWDAF entities can be from E-NWDAF entities
  • the service quality MOS level of the subscribed target service and the corresponding first network performance index are simultaneously measured by the E-NWDAF entity to obtain the service quality MOS level of the target service and the corresponding first network performance index. That is to say, the service quality MOS level of the target service and the corresponding first network performance index are obtained within the operator, so it is easier to implement and the measurement result is more accurate.
  • the E-NWDAF entity obtains the service quality MOS level, including: the E-NWDAF entity obtains the service experience data of the target service; the E-NWDAF entity determines the service quality according to the service experience data of the target service MOS level. Based on this solution, the E-NWDAF entity can obtain the service quality MOS level.
  • the service experience data includes one or more of the following parameters: the initial buffering delay, playback buffering duration, bit rate, service rate, frame rate of the video service , Fluency, clarity; and, the resolution of the terminal device.
  • the first network performance index includes one or more of the following parameters: the number of uplink packet loss on the path between the terminal device and the access network device, the uplink packet loss rate, and the downlink packet loss Number, Downlink Packet Loss Rate, Round Trip Delay RTT, Uplink Error Packets, Uplink Error Rate, Downlink Error Packets, Downlink Error Rate, Uplink Out of Order Packets, Uplink Out of Order Packet Rate, Downlink Out of Order Packets Number, downlink out-of-sequence packet rate, uplink retransmission packet number, uplink retransmission rate, downlink retransmission packet number, downlink retransmission rate, uplink average packet interval, uplink average packet jitter, downlink average packet interval, downlink average packet jitter, Uplink rate and downlink rate; and, the number of uplink packet loss, uplink packet loss rate, downlink packet loss, downlink packet loss rate, RTT, and uplink error packets
  • a communication device for implementing the above-mentioned various methods.
  • the communication device may be the C-NWDAF entity in the foregoing first aspect, or a device including the foregoing C-NWDAF entity; or, the communication device may be the E-NWDAF entity in the foregoing first aspect, or may include the foregoing E-NWDAF The physical device.
  • the communication device includes a module, unit, or means corresponding to the foregoing method, and the module, unit, or means can be implemented by hardware, software, or hardware executing corresponding software.
  • the hardware or software includes one or more modules or units corresponding to the above-mentioned functions.
  • a communication device including: a processor and a memory; the memory is used to store computer instructions, and when the processor executes the instructions, the communication device executes the method described in any of the above aspects.
  • the communication device may be the C-NWDAF entity in the foregoing first aspect, or a device including the foregoing C-NWDAF entity; or, the communication device may be the E-NWDAF entity in the foregoing first aspect, or may include the foregoing E-NWDAF The physical device.
  • a communication device including: a processor; the processor is configured to couple with a memory, and after reading an instruction in the memory, execute the method according to any of the foregoing aspects according to the instruction.
  • the communication device may be the C-NWDAF entity in the foregoing first aspect, or a device including the foregoing C-NWDAF entity; or, the communication device may be the E-NWDAF entity in the foregoing first aspect, or may include the foregoing E-NWDAF The physical device.
  • a computer-readable storage medium stores instructions that, when run on a computer, enable the computer to execute the method described in any of the above aspects.
  • a computer program product containing instructions which when running on a computer, enables the computer to execute the method described in any of the above aspects.
  • a communication device for example, the communication device may be a chip or a chip system
  • the communication device includes a processor for implementing the functions involved in any of the foregoing aspects.
  • the communication device further includes a memory for storing necessary program instructions and data.
  • the communication device is a chip system, it may be composed of chips, or may include chips and other discrete devices.
  • a communication system in a ninth aspect, includes a central network data analysis function C-NWDAF entity and an edge network data analysis function E-NWDAF entity; the C-NWDAF entity is used to send the first data to the E-NWDAF entity.
  • a subscription request where the first subscription request is used to request to subscribe to the service quality average opinion score MOS level of the target service and the corresponding first network performance index, where the first network performance index is the network performance index of the network transmission that carries the target service;
  • the E-NWDAF entity is used to receive the first subscription request from the C-NWDAF entity, and after obtaining the service quality MOS level and the first network performance index, send the service quality MOS level and the first network performance index to the C-NWDAF entity A network performance index;
  • the C-NWDAF entity is used to receive the service quality MOS level and the first network performance index from the E-NWDAF entity, and establish the target service according to the service quality MOS level and the first network performance index MOS model.
  • the technical effect of the ninth aspect can be referred to the above-mentioned first aspect or the second aspect, which will not be repeated here.
  • the C-NWDAF entity is also used to execute the training method for applying the MOS model described in any one of the possible designs of the first aspect.
  • the E-NWDAF entity is also used to execute the training method of applying the MOS model described in any one of the possible designs of the second aspect.
  • Figure 1 is a schematic diagram of the location of NWDAF in the existing 3GPP23.791;
  • Figure 2 is a schematic diagram of existing resource customization on demand
  • FIG. 3 is a schematic diagram of the architecture of a communication system provided by an embodiment of the application.
  • FIG. 4 is a schematic structural diagram of a communication device provided by an embodiment of the application.
  • FIG. 5 is a schematic flowchart of a training method using a MOS model provided by an embodiment of the application
  • FIG. 6 is a schematic diagram of a first network performance index provided by an embodiment of this application.
  • FIG. 7 is a schematic diagram 1 of the flow of a service experience guarantee method provided by an embodiment of the application.
  • FIG. 8 is a schematic diagram of the architecture corresponding to the service experience guarantee method shown in FIG. 7;
  • FIG. 9 is a schematic diagram of the second flow of the service experience guarantee method provided by an embodiment of the application.
  • FIG. 10 is a schematic diagram of the architecture corresponding to the service experience guarantee method shown in FIG. 9;
  • FIG. 11 is a schematic structural diagram of a C-NWDAF entity provided by an embodiment of this application.
  • Fig. 12 is a schematic structural diagram of an E-NWDAF entity provided by an embodiment of the application.
  • enhanced mobile broadband requires large-capacity, high-speed, and dynamic bandwidth allocation.
  • Ultra-reliable and low latency communications (ultra-reliable and low latency communications) require high reliability, high availability, and low latency, which can support the reliable operation of mission-critical services such as automated factories and remote surgery, and can meet autonomous driving.
  • Remote control drones and other delay-critical services require low latency;
  • massive machine type communications (mMTC) requires large-capacity, high-speed, and dynamic bandwidth allocation, which can be used for smart cities and other Internet of Things ( The internet of things (IoT) service provides connections for billions of devices, with a density of one million devices per square kilometer.
  • the purpose of QoS is to "customize on demand" under the condition of limited resources to provide network services with differentiated service quality for the business.
  • the resources in the network such as the air interface bearer resources between the terminal equipment and the access network equipment or the next generation user plane (NG-U) between the access network equipment and the UPF entity)
  • Tunnel tunnel
  • different QoS rules can be allocated as needed to provide differentiated service quality at the lowest cost.
  • QoS usually has two meanings. One is how the quality of service is, that is, the specific indicators (parameters) that characterize QoS; the other is how to ensure these indicators, that is, the mechanism to achieve QoS.
  • 5G QoS identifier 5G QoS identifier, 5QI
  • resource type resource type
  • default priority Level level default priority level
  • packet delay estimation packet delay budget, PDB
  • packet error rate packetet error rate
  • MDBV default (default) maximum data burst volume
  • MDBV maximum data burst volume
  • Table 1 The average window (default averaging window, DAV) and exemplary services are shown in Table 1:
  • the 5QI in Table 1 is defined as the end-to-end network service quality level.
  • Each transmission node in the network (such as access network equipment or core network equipment, etc.) guarantees the transmission data according to the QoS profile, but whether it can be guaranteed
  • the end-to-end service experience quality is unknown.
  • the NWDAF entity is introduced in 3GPP 23.791 and 23.228, and the MOS of the specified service is obtained by inference by subscribing to the network performance indicators (packet loss rate, RTT and bit error rate) of the network transmission from the 5GC NFs entity and subscribing to the service quality MOS level from the AF entity
  • the model is used to support operators to provide differentiated service quality assurance configuration QoS files for end users, third-party application (over the top, OTT) vendors, and industries carried on the operator's network.
  • the communication system 30 includes a C-NWDAF entity 301 and an E-NWDAF entity 302.
  • the C-NWDAF entity 301 and the E-NWDAF entity 302 are distributed structures of NWDAF entities.
  • the E-NWDAF entity 302 may be deployed on the 5GC NFs side, on the side of the access network device, or inside the terminal device.
  • the E-NWDAF entity 302 can be embedded in the terminal device in the form of a software development kit (SDK), or deployed together with the UPF entity in the form of network function virtualization (NFV) through a software plug-in , Or deploy it in the periphery of the access network equipment or UPF entity by means of general server software.
  • SDK software development kit
  • NFV network function virtualization
  • the E-NWDAF entity 302 is mainly used to collect the service quality MOS level of the target service and the corresponding first network performance index, and provide the service quality MOS level of the target service and the corresponding first network performance index to
  • the first network performance indicator is a network performance indicator of a network transmission that carries the target service.
  • the C-NWDAF entity 301 is the central node, which is mainly used to obtain the service quality MOS level of the target service and the corresponding first network performance index from the E-NWDAF entity 302, and according to the service quality MOS level of the target service and the corresponding first network Performance indicators, establish the MOS model of the target business.
  • the communication system may further include one or more of a network storage function entity 303, an OAM 304, or a terminal device 305.
  • the C-NWDAF entity 301 can also interact with one or more of the network storage function entity 303, terminal device 305, or OAM304 to collect data, and cooperate with the service quality MOS level of the target service and the corresponding first network performance index to complete the final MOS reasoning and analysis.
  • the terminal device in the embodiment of the present application may be a device used to implement a wireless communication function, such as a terminal or a chip that can be used in a terminal.
  • the terminal may be a user equipment (UE), an access terminal, a terminal unit, a terminal station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, and wireless communication in a 5G network or a future evolved PLMN.
  • the access terminal can be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), with wireless communication Functional handheld devices, computing devices or other processing devices connected to wireless modems, vehicle-mounted devices or wearable devices, virtual reality (VR) terminal devices, augmented reality (AR) terminal devices, industrial control (industrial) Wireless terminal in control), wireless terminal in self-driving, wireless terminal in remote medical, wireless terminal in smart grid, wireless terminal in transportation safety (transportation safety) Terminal, wireless terminal in smart city, wireless terminal in smart home, etc.
  • the terminal can be mobile or fixed.
  • the access network equipment in the embodiments of this application refers to equipment that accesses the core network, such as base stations, broadband network gateways (BNG), aggregation switches, non-3GPP access equipment, etc. .
  • the base station may include various forms of base stations, such as: macro base stations, micro base stations (also called small stations), relay stations, access points, and so on.
  • the related functions of the C-NWDAF entity or the E-NWDAF entity in the embodiments of the present application can be implemented by one device, or by multiple devices, or by one or more functions in one device.
  • Module implementation this embodiment of the application does not specifically limit this. It is understandable that the above functions can be network elements in hardware devices, software functions running on dedicated hardware, or a combination of hardware and software, or instantiated on a platform (for example, a cloud platform) Virtualization function.
  • FIG. 4 is a schematic structural diagram of a communication device 400 provided by an embodiment of this application.
  • the communication device 400 includes one or more processors 401, a communication line 402, and at least one communication interface (in FIG. 4, the communication interface 404 and one processor 401 are taken as an example for illustration), optional
  • the memory 403 may also be included.
  • the processor 401 may be a general-purpose central processing unit (central processing unit, CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more programs for controlling the execution of the program of this application. integrated circuit.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • the communication line 402 may include a path for connecting different components.
  • the communication interface 404 may be a transceiver module for communicating with other devices or communication networks, such as Ethernet, RAN, wireless local area networks (WLAN), etc.
  • the transceiver module may be a device such as a transceiver or a transceiver.
  • the communication interface 404 may also be a transceiver circuit located in the processor 401 to implement signal input and signal output of the processor.
  • the memory 403 may be a device having a storage function. For example, it can be read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM), or other types that can store information and instructions Dynamic storage devices can also be electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), or other optical disk storage, optical disc storage ( Including compact discs, laser discs, optical discs, digital universal discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store desired program codes in the form of instructions or data structures and can be stored by a computer Any other media taken, but not limited to this.
  • the memory may exist independently and is connected to the processor through the communication line 402. The memory can also be integrated with the processor.
  • the memory 403 is used to store computer-executed instructions for executing the solution of the present application, and the processor 401 controls the execution.
  • the processor 401 is configured to execute computer-executable instructions stored in the memory 403, so as to implement the training method for applying the MOS model provided in the embodiment of the present application.
  • the processor 401 may also execute the processing-related functions in the training method for applying the MOS model provided in the following embodiments of the present application, and the communication interface 404 is responsible for communicating with other devices or communication networks. Communication, this embodiment of the application does not specifically limit this.
  • the computer-executable instructions in the embodiments of the present application may also be referred to as application program code, which is not specifically limited in the embodiments of the present application.
  • the processor 401 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 4.
  • the communication device 400 may include multiple processors, such as the processor 401 and the processor 408 in FIG. 4. Each of these processors can be a single-CPU (single-CPU) processor or a multi-core (multi-CPU) processor.
  • the processor here may refer to one or more devices, circuits, and/or processing cores for processing data (for example, computer program instructions).
  • the communication device 400 may further include an output device 405 and an input device 406.
  • the output device 405 communicates with the processor 401 and can display information in a variety of ways.
  • the output device 405 may be a liquid crystal display (LCD), a light emitting diode (LED) display device, a cathode ray tube (CRT) display device, or a projector (projector) Wait.
  • the input device 406 communicates with the processor 401 and can receive user input in a variety of ways.
  • the input device 406 may be a mouse, a keyboard, a touch screen device, or a sensor device.
  • the aforementioned communication device 400 may sometimes be called a communication device, and it may be a general-purpose device or a special-purpose device.
  • the communication device 400 may be an AP such as a server, a router, a switch, or a bridge, or the communication device 400 may be an STA such as a mobile phone, a tablet computer, a computer notebook, a smart watch, or a smart TV.
  • the embodiment of the present application does not limit the communication device 400 type.
  • a training method for applying a MOS model includes the following steps:
  • the consumer NF entity sends a subscription request 1 to the C-NWDAF entity.
  • the C-NWDAF entity receives the subscription request 1 from the consumer NF entity, and the subscription request 1 is used to request subscription to the MOS model of the specified service.
  • step S501 is an optional step, that is, the training method of applying the MOS model provided in the embodiment of the present application may not perform step S501, which is explained here in a unified manner, and will not be repeated in the following.
  • the C-NWDAF entity sends a subscription request 2 to the E-NWDAF entity.
  • the E-NWDAF entity receives the subscription request 2 from the C-NWDAF entity.
  • the subscription request 2 is used to request the service quality MOS level of the target service and the corresponding first network performance index, and the first network performance index is the bearer The network performance index of the network transmission of the target service.
  • the first network performance index includes one or more of the following parameters:
  • uplink packet loss Number of uplink packet loss, uplink packet loss rate, downlink packet loss, downlink packet loss rate, RTT, uplink error packets, uplink error rate, and downlink error packets on the path between the terminal device and the access network device Number, downlink bit error rate, uplink out-of-sequence packet number, uplink out-of-sequence packet rate, downlink out-of-sequence packet number, downlink out-of-sequence packet rate, uplink retransmission packet number, uplink retransmission rate, downlink retransmission packet number, downlink retransmission Transmission rate, uplink average packet interval, uplink average packet jitter, downlink average packet interval, downlink average packet jitter, uplink rate and downlink rate;
  • the RTT in the embodiment of the present application includes uplink RTT and downlink RTT, where the uplink RTT can be understood as the time interval for the terminal device to send uplink data packets and receive the response frame of the uplink data packet, and the downlink RTT can be understood as AF The time interval between the entity sending the downlink data packet and receiving the response frame of the downlink data packet is described here in a unified manner, and will not be repeated here.
  • the provided schematic diagram of the first network performance index includes one or more of the network performance index between the terminal device and the gNB, the network performance index between the gNB and the UPF entity, or the network performance index between the UPF entity and the AF entity One.
  • the E-NWDAF entity obtains the service quality MOS level of the target service and the corresponding first network performance index.
  • the E-NWDAF entity may collect the user's original service data by means of traffic mirroring or implanting NF proxy, and complete the following processing:
  • the service experience data includes one or more of the following parameters: initial buffering delay of the video service, playback buffering time, bit rate, service rate, frame rate, fluency , Resolution; and, the resolution of the terminal device.
  • the E-NWDAF entity sends a notification message 2 to the C-NWDAF entity.
  • the C-NWDAF entity receives a notification message 2 from the E-NWDAF entity.
  • the notification message 2 includes the service quality MOS level of the target service and the corresponding first network performance index.
  • the training method for applying the MOS model may further include the following steps
  • the C-NWDAF entity sends a subscription request 3 to the NRF entity/5GC NFs entity.
  • the NRF entity/5GC NFs entity receives the subscription request 3 from the C-NWDAF entity, and the subscription request 3 is used to request to subscribe to the first load of the 5GC NFs entity.
  • the first load of the 5GC NFs entity includes one or more of the following parameters:
  • the number of sessions of the 5GC NFs entity The number of sessions of the 5GC NFs entity, the number of users of the 5GC NFs entity, and the resource utilization of the 5GC NFs entity.
  • the resource utilization of the 5GC NFs entity may include, for example, the utilization of CPU, memory or network IO.
  • the NRF entity/5GC NFs entity obtains the first load of the 5GC NFs entity.
  • the NRF entity/5GC NFs entity sends a notification message 3 to the C-NWDAF entity.
  • the C-NWDAF entity receives the notification message 3 from the NRF entity/5GC NFs entity, and the notification message 3 includes the first load of the 5GC NFs entity.
  • the training method for applying the MOS model provided in the embodiment of the present application may further include the following steps S508-S510:
  • the C-NWDAF entity sends a subscription request 4 to the OAM.
  • the OAM receives a subscription request 4 from the C-NWDAF entity, and the subscription request 4 is used to request to subscribe to a second network performance index.
  • the second network performance index is a network performance index of the wireless carrying the target service.
  • the second network performance index includes one or more of the following parameters:
  • the number of sessions of the access network equipment the number of RRC connected users, the congestion state and resource occupancy rate, the wireless measurement indicators of the terminal equipment, and the location information of the terminal equipment.
  • the number of RRC connected users may include, for example, the number of users in an RRC_inactive state and the number of users in an RRC_active state, respectively.
  • the resource utilization of the access network device may include, for example, uplink or downlink physical resource block (physical resource bank, PRB) utilization, CPU or memory utilization.
  • uplink or downlink physical resource block physical resource bank, PRB
  • CPU or memory utilization may include, for example, uplink or downlink physical resource block (physical resource bank, PRB) utilization, CPU or memory utilization.
  • PRB physical resource bank
  • the wireless measurement index of the terminal device may include, for example, reference signal receiving power (RSRP), reference signal receiving quality (RSRQ), or signal-to-interference plus noise ratio, SINR).
  • RSRP reference signal receiving power
  • RSRQ reference signal receiving quality
  • SINR signal-to-interference plus noise ratio
  • the OAM obtains the second network performance index.
  • the OAM sends a notification message 4 to the C-NWDAF entity.
  • the C-NWDAF entity receives a notification message 4 from the OAM, and the notification message 4 includes the second network performance indicator.
  • the training method for applying the MOS model provided in the embodiment of the present application may further include the following step S511:
  • the terminal device sends a measurement report to the C-NWDAF entity.
  • the C-NWDAF entity receives the measurement report from the terminal device.
  • the measurement report includes the information of the terminal device, the service quality MOS level of the target service and the number of uplink packet loss, uplink packet loss rate, downlink packet loss, and downlink packet loss rate on the path between the corresponding terminal device and the access network device , RTT, uplink average packet interval, uplink average packet jitter, downlink average packet interval, downlink average packet jitter, uplink rate and downlink rate, or one or more of the wireless measurement indicators of the terminal device.
  • the information of the terminal device may include, for example, the model, CPU, or memory capacity of the terminal device.
  • the training method for applying the MOS model provided in the embodiment of the present application may further include the following step S512:
  • the C-NWDAF entity establishes the MOS model of the target service according to the service quality MOS level and the first network performance index.
  • the MOS model of the target service includes the model relationship between the service quality MOS level and the first network performance index.
  • the C-NWDAF entity establishes the MOS model of the target service according to the service quality MOS level and the first network performance index, including: C-
  • the NWDAF entity excludes the business caused by the abnormality of the terminal equipment and the AF entity based on the terminal equipment measurement report, the second network performance index, and/or the first load of the 5GC NFs entity, the service quality MOS level and the first network performance index Samples of poor quality MOS, and establish MOS model of the target business.
  • the MOS model of the target service includes the service quality MOS level and the second network performance index, and/or the first load of the 5GC NFs entity, and the model relationship between the service quality MOS level and the first network performance index.
  • the training method for applying the MOS model provided in the embodiment of the present application may further include the following step S513:
  • the C-NWDAF entity sends a notification message 1 to the consumer NF entity.
  • the consumer NF entity receives a notification message 1 from the C-NWDAF entity, and the notification message 1 includes the MOS model of the target service.
  • the consumer NF entity in the area under the jurisdiction of the E-NWDAF entity may also directly subscribe to the E-NWDAF entity the MOS model of the specified service, and then the E-NWDAF entity obtains the service quality of the target service After the MOS level and the corresponding first network performance index, the MOS model of the target service is established according to the service quality MOS level and the first network performance index, and the target service is sent to the consumer NF entity in the area under the jurisdiction of the E-NWDAF entity
  • the MOS model of this application does not specifically limit this.
  • the terminal device can also send the above measurement report to the E-NWDAF entity, and the E-NWDAF entity can also subscribe to the service quality MOS level and the second network performance index, and/or the first load of the 5GC NFs entity, and then the E-NWDAF entity It is an entity that can rule out the abnormality of the terminal device and AF entity based on the measurement report of the terminal device, the second network performance index, and/or the first load of the 5GC NFs entity, as well as the service quality MOS level and the first network performance index
  • the MOS model of the target service is established based on the sample of poor service quality MOS, which is not specifically limited in the embodiment of this application.
  • the C-NWDAF entity or the E-NWDAF entity after the C-NWDAF entity or the E-NWDAF entity obtains the MOS model of the target service, it can generate a suggested QoS file according to the MOS model of the target service to support the operator as an end user, OTT vendors and industries provide differentiated service quality. If the consumer NF entity receives the recommended QoS file sent by the C-NWDAF entity or the E-NWDAF entity, it can choose to use the corresponding QoS file according to the pre-configured priority rules or the corresponding judgment algorithm. This is a unified explanation. Repeat.
  • C-NWDAF entities can be from E-NWDAF entities Subscribe to the service quality MOS level of the target service and the corresponding first network performance index, and the E-NWDAF entity simultaneously measures the service quality MOS level of the target service and the corresponding first network performance index, that is to say, obtains it inside the operator
  • the service quality MOS level of the target service and the corresponding first network performance index are therefore easier to implement and the measurement results are more accurate.
  • the MOS model of the target service established based on the training method using the MOS model provided by the embodiment of the present application not only considers the first network performance index, but also takes into account the network performance index of the wireless carrying the target service, and/ Or, the first load of the 5GC NFs entity corresponding to the MOS level, and the poor quality MOS sample of the service quality caused by the abnormality of the terminal equipment and the AF entity, so the MOS model of the target service is more accurate and comprehensive.
  • the actions of the C-NWDAF entity or the E-NWDAF entity in the foregoing steps S501 to S513 may be executed by the processor 401 in the communication device 400 shown in FIG. 4 calling the application program code stored in the memory 403. There are no restrictions.
  • the following gives a relevant example of providing users with real-time service experience guarantee based on the MOS model of the target service provided by the embodiment of the present application in combination with the current service experience quality of the terminal device.
  • Example 1 Taking service experience guarantee in a 5G network as an example, as shown in FIG. 7, a service experience guarantee method provided by an embodiment of the present application includes the following steps:
  • the AF entity applies to the policy control function (PCF) entity through the network exposure function (NEF) entity to provide service experience guarantee for the target service of the target terminal device.
  • PCF policy control function
  • NEF network exposure function
  • the AF entity may also directly apply to the PCF entity for service experience guarantee for the target service, which is not specifically limited in the embodiment of this application.
  • the service experience guarantee in the embodiment of the present application can also be replaced with QoS optimization, which is described in a unified manner here, and will not be repeated in the following.
  • the PCF entity sends a subscription request 5 to the C-NWDAF entity.
  • the C-NWDAF entity receives a subscription request 5 from the PCF entity.
  • the subscription request 5 is used to request service experience guarantee for the target service of the target terminal device.
  • the service quality MOS level requirement of the service experience guarantee is the first service quality MOS. grade.
  • the C-NWDAF entity sends a subscription request 6 to the E-NWDAF entity.
  • the E-NWDAF entity receives a subscription request 6 from the C-NWDAF entity, and the subscription request 6 is used to request to subscribe to the trigger event of the target terminal device using the target service, the second service quality MOS level of the target service, and the corresponding first network Performance.
  • the E-NWDAF entity obtains the second service quality MOS level of the target service and the corresponding first network performance index, and sends a notification message 6 to the C-NWDAF entity, where the notification message 6 includes the current target service The second service quality MOS level and the corresponding first network performance index.
  • the C-NWDAF entity sends a subscription request 7 to the OAM.
  • the OAM receives the subscription request 7 from the C-NWDAF entity, and the subscription request 7 is used to request to subscribe to the second network performance indicator.
  • the OAM entity After obtaining the second network performance indicator, the OAM entity sends a notification message 7 to the C-NWDAF entity.
  • the C-NWDAF entity receives the notification message 7 from the OAM, and the notification message 7 includes the current second network performance index.
  • the C-NWDAF entity compares the first network performance index corresponding to the second service quality MOS level with the first service quality in the MOS model of the target service The first network performance index corresponding to the MOS level is matched, and the C-NWDAF entity matches the second network performance index corresponding to the second service quality MOS level with the second network corresponding to the first service quality MOS level in the MOS model of the target service Performance indicators are matched to obtain network analysis results.
  • the network analysis results in the embodiments of the present application are used to characterize what causes the difference between the second service quality MOS level and the first service quality MOS level, such as a difference between the terminal device and the current access network device
  • the network performance index is poor, or a certain network performance index between the current access network device and the UPF entity is poor, or a certain network performance index between the UPF entity and the AF entity is poor, etc., etc. Make specific restrictions.
  • steps S705-S706 are optional steps. If steps S705-S706 are not performed, the C-NWDAF entity in step S707 does not need to associate the second network performance index corresponding to the second service quality MOS level with the target In the MOS model of the service, the second network performance index corresponding to the first service quality MOS level is matched, which is explained here in a unified manner, and will not be repeated in the following.
  • the C-NWDAF entity sends a notification message 5 to the PCF entity.
  • the PCF entity receives a notification message 5 from the C-NWDAF entity, and the notification message 5 includes the network analysis result.
  • the PCF entity may notify the session management function (SMF) entity or the UPF entity to update the QoS policy and network parameter tuning of the terminal device according to the network analysis result, such as wireless scheduling priority, The transmission layer differentiated services code point (DSCP) level, etc., which are not specifically limited in the embodiment of the present application.
  • SMS session management function
  • the schematic diagram of the architecture corresponding to the service experience guarantee method shown in FIG. 7 may be as shown in FIG. 8. Based on the service experience guarantee method provided by the embodiments of the present application, real-time service experience guarantee can be provided for users.
  • Example 2 Taking the service experience guarantee in a 5G network as an example, as shown in FIG. 9, another service experience guarantee method provided by an embodiment of the present application includes the following steps:
  • the AF entity applies to the PCF entity through the NEF entity for service experience guarantee for the target service of the target terminal device.
  • the AF entity may also directly apply to the PCF entity for service experience guarantee for the target service, which is not specifically limited in the embodiment of this application.
  • the service experience guarantee in the embodiment of the present application can also be replaced with QoS optimization, which is described in a unified manner here, and will not be repeated in the following.
  • the PCF entity sends a subscription request 5 to the C-NWDAF entity.
  • the C-NWDAF entity receives a subscription request 5 from the PCF entity.
  • the subscription request 5 is used to request service experience guarantee for the target service of the target terminal device.
  • the service quality MOS level requirement of the service experience guarantee is the first service quality MOS. grade.
  • the C-NWDAF entity determines that the target terminal device will move from the service range of the first access network device to the service range of the second access network device according to the movement track of the target terminal device.
  • the movement trajectory of the target terminal device may be obtained by the E-NWDAF entity and then sent to the C-NWDAF entity, or it may be obtained by the C-NWDAF entity in other ways. This is not specifically limited.
  • the C-NWDAF entity sends a subscription request 8 to the E-NWDAF entity.
  • the E-NWDAF entity receives a subscription request 8 from the C-NWDAF entity, and the subscription request 8 is used to request to subscribe to the trigger event of the target terminal device using the target service and the first network performance index associated with the second access network device.
  • the first network performance index associated with the second access network device may include one or more of the following parameters, for example:
  • the number of uplink packet loss, uplink packet loss rate, downlink packet loss, downlink packet loss rate, RTT, uplink error packet number, uplink error rate, and downlink error on the path between the terminal device and the second access network device Number of code packets, downlink bit error rate, uplink out-of-sequence packet number, uplink out-of-sequence packet rate, downlink out-of-sequence packet number, downlink out-of-sequence packet rate, uplink retransmission packet number, uplink retransmission rate, downlink retransmission packet number, Downlink retransmission rate, Downlink retransmission rate, uplink average packet interval, uplink average packet jitter, downlink average packet interval, downlink average packet jitter, uplink rate and downlink rate;
  • the number of uplink packet loss, the uplink packet loss rate, the number of downlink packet loss, the downlink packet loss rate, RTT, the number of uplink error packets, and the uplink error rate on the path between the second access network device and the user plane entity Downlink error packet number, downlink error rate, uplink out-of-sequence packet number, uplink out-of-sequence packet rate, downlink out-of-sequence packet rate, downlink out-of-sequence packet rate, uplink retransmission packet number, uplink retransmission rate, downlink retransmission Number of packets, downlink retransmission rate, uplink average packet interval, uplink average packet jitter, downlink average packet interval, downlink average packet jitter, uplink rate, and downlink rate;
  • the E-NWDAF entity obtains the first network performance indicator associated with the second access network device, and sends a notification message 8 to the C-NWDAF entity, the notification message 8 includes the second access network device association The first network performance index.
  • the C-NWDAF entity sends a subscription request 9 to the OAM.
  • the OAM receives a subscription request 9 from the C-NWDAF entity, and the subscription request 9 is used to request a subscription to the second network performance indicator associated with the second access network device.
  • the second network performance index associated with the second access network device includes one or more of the following parameters:
  • the number of sessions of the second access network device The number of sessions of the second access network device, the number of RRC connected users, the congestion state and resource occupancy rate, the wireless measurement index of the terminal device, and the location information of the terminal device.
  • the number of RRC connected users, resource utilization, and wireless measurement indicators of terminal equipment reference may be made to the description of the embodiment shown in FIG. 5, which is not repeated here.
  • the OAM entity After obtaining the second network performance indicator associated with the second access network device, the OAM entity sends a notification message 9 to the C-NWDAF entity.
  • the C-NWDAF entity receives a notification message 9 from the OAM, and the notification message 9 includes the second network performance index associated with the second access network device.
  • the C-NWDAF entity matches the first network performance index and the second network performance index associated with the second access network device with the MOS model of the target service, and determines the first network performance index and the second network performance index associated with the second access network device.
  • the third service quality MOS level of the target service corresponding to the second network performance index.
  • the C-NWDAF entity will compare the first network performance index corresponding to the third service quality MOS level with the first service in the MOS model of the target service.
  • the first network performance index corresponding to the quality MOS level is matched, and the C-NWDAF entity matches the second network performance index corresponding to the third service quality MOS level with the second network performance index corresponding to the first service quality MOS level in the MOS model of the target service. 2.
  • the network performance index is matched to obtain the network prediction result.
  • the network prediction result in the embodiment of the present application is used to characterize what causes the difference between the third service quality MOS level and the first service quality MOS level, such as a difference between the terminal device and the current access network device
  • the network performance index is poor, or a certain network performance index between the current access network device and the UPF entity is poor, or a certain network performance index between the UPF entity and the AF entity is poor, etc., etc. Make specific restrictions.
  • steps S906-S907 are optional steps. If steps S906-S907 are not performed, the C-NWDAF entity in step S908 does not need to associate the second network performance index associated with the second access network device with the target To match the MOS model of the service, it is only necessary to determine the third service quality MOS level of the target service corresponding to the first network performance index associated with the second access network device. At the same time, if steps S906-S907 are not performed, the C-NWDAF entity in step S909 does not need to compare the second network performance index corresponding to the third service quality MOS level with the second network performance index corresponding to the first service quality MOS level in the MOS model of the target service. Second, the network performance indicators are matched, which are described here in a unified manner, and will not be repeated here.
  • the C-NWDAF entity sends a notification message 5 to the PCF entity.
  • the PCF entity receives a notification message 5 from the C-NWDAF entity, and the notification message 5 includes the network prediction result.
  • the PCF entity may suggest that the UPF entity update the user's QoS policy after entering the service range of the second access network device according to the network prediction result, which is not specifically limited in the embodiment of the present application.
  • the schematic diagram of the architecture corresponding to the service experience guarantee method shown in FIG. 9 may be as shown in FIG. 10. Based on the service experience guarantee method provided by the embodiments of the present application, real-time service experience guarantee can be provided for users.
  • the methods and/or steps implemented by the C-NWDAF entity can also be implemented by components that can be used for the C-NWDAF entity; the methods and/or steps implemented by the E-NWDAF entity, It can also be implemented by components that can be used for E-NWDAF entities.
  • an embodiment of the present application also provides a communication device.
  • the communication device may be the C-NWDAF entity in the foregoing method embodiment, or a device including the foregoing C-NWDAF entity, or a component that can be used for the C-NWDAF entity;
  • the communication device may be the E-NWDAF entity in the foregoing method embodiment, or a device containing the foregoing E-NWDAF entity, or a component that can be used for the E-NWDAF entity.
  • the communication device includes hardware structures and/or software modules corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • FIG. 11 shows a schematic structural diagram of a C-NWDAF entity 110.
  • the C-NWDAF entity 110 includes a transceiver module 1101 and a processing module 1102.
  • the transceiver module 1101 may also be referred to as a transceiver unit to implement a transceiver function, and may be, for example, a transceiver circuit, transceiver, transceiver or communication interface.
  • the transceiver module 1101 is configured to send a first subscription request to the E-NWDAF entity.
  • the first subscription request is used to request the service quality MOS level of the target service and the corresponding first network performance index.
  • the first network performance index is the bearer The network performance index of the network transmission of the target service;
  • the transceiver module 1101 is also used to receive the service quality MOS level and the first network performance index from the E-NWDAF entity;
  • the processing module 1102 is used to according to the service quality MOS level and the first network performance Indicators, establish a MOS model of the target business.
  • the processing module 1102 is further configured to obtain a second network performance index corresponding to the MOS level, and/or the first load of the first network functional entity corresponding to the MOS level, and the second network performance index is the value of the target service.
  • the first network function entity is the network data provider function entity; the processing module 1102 is used to establish the MOS model of the target service according to the service quality MOS level and the first network performance index, including: processing module 1102, According to the second network performance index, and/or the first load of the first network function entity, the service quality MOS level and the first network performance index, exclude the service quality MOS caused by the abnormality of the terminal equipment and the application function entity For poor quality samples, establish a MOS model of the target business.
  • the processing module 1102 is configured to obtain the first load of the first network functional entity corresponding to the MOS level, and includes: the processing module 1102 is configured to send a second subscription request to the network storage functional entity through the transceiver module 1101, and the second The subscription request is used to request to subscribe to the first load of the first network function entity; the first load of the first network function entity is received from the network storage function entity through the transceiver module 1101.
  • the processing module 1102 is configured to obtain a second network performance index corresponding to the MOS level, and includes: a processing module 1102, configured to send a third subscription request to the operation and maintenance OAM entity of the operator network through the transceiver module 1101, The third subscription request is used to request to subscribe to the second network performance index; the second network performance index is received from the OAM entity through the transceiver module 1101.
  • the transceiver module 1101 is further configured to receive a fourth subscription request from a second network function entity, the fourth subscription request is used to request a subscription to the MOS model of the target service, and the second network function entity is a consumer function entity; 1101. It is also used to send the MOS model of the target service to the second network function entity.
  • the transceiver module 1101 is further configured to receive a fifth subscription request from the policy control entity.
  • the fifth subscription request is used to request service experience guarantee for the target service of the target terminal device.
  • the service quality MOS level requirement for service experience guarantee is The first service quality MOS level;
  • the transceiver module 1101 is also used to send a sixth subscription request to the E-NWDAF entity, the sixth subscription request is used to request to subscribe to the trigger event of the target terminal device using the target service, and the second service quality of the target service
  • the transceiver module 1101 is also used to receive the second service quality MOS level and the corresponding first network performance index from the E-NWDAF entity after the trigger event is triggered;
  • the processing module 1102 It is also used to correspond the first network performance index corresponding to the second service quality MOS level to the first service quality MOS level in the MOS model of the target service when the second service quality MOS level is different from the first service quality MOS level
  • the transceiver module 1101 is further configured to receive a fifth subscription request from the policy control entity.
  • the fifth subscription request is used to request service experience guarantee for the target service of the target terminal device.
  • the service quality MOS level requirement for service experience guarantee is The first service quality MOS level;
  • the processing module 1102 is also used to determine that the target terminal device will move from the service range of the first access network device to the movement range of the second access network device according to the movement trajectory of the target terminal device;
  • the module 1101 is also used to send a sixth subscription request to the E-NWDAF entity.
  • the sixth subscription request is used to request a subscription to the trigger event of the first terminal device using the target service and the first network performance index associated with the second access network device
  • the transceiver module 1101 is also used to receive the first network performance index associated with the second access network device from the E-NWDAF entity after the trigger event is triggered; the processing module 1102 is also used to associate the second access network device
  • the first network performance index of the second access network device is matched with the MOS model of the target service to determine the third service quality MOS level of the target service corresponding to the first network performance index associated with the second access network device; the processing module 1102 is also used for 3.
  • the first network performance index corresponding to the third service quality MOS level is changed to the first network corresponding to the first service quality MOS level in the MOS model of the target service
  • the performance indicators are matched to obtain the network prediction result; the transceiver module 1101 is also used to send the network prediction result to the policy control entity, and the network prediction result is used for network optimization.
  • the C-NWDAF entity 110 is presented in the form of dividing various functional modules in an integrated manner.
  • the "module” here can refer to a specific ASIC, circuit, processor and memory that executes one or more software or firmware programs, integrated logic circuit, and/or other devices that can provide the above-mentioned functions.
  • the C-NWDAF entity 110 may take the form of the communication device 400 shown in FIG. 4.
  • the processor 401 in the communication device 400 shown in FIG. 4 may invoke the computer execution instructions stored in the memory 403 to make the communication device 400 execute the training method of applying the MOS model in the foregoing method embodiment.
  • the functions/implementation process of the transceiver module 1101 and the processing module 1102 in FIG. 11 can be implemented by the processor 401 in the communication device 400 shown in FIG. 4 calling the computer execution instructions stored in the memory 403.
  • the function/implementation process of the processing module 1102 in FIG. 11 can be implemented by the processor 401 in the communication device 400 shown in FIG. 4 calling a computer execution instruction stored in the memory 403, and the function of the transceiver module 1101 in FIG. 11
  • the implementation process can be implemented through the communication interface 404 in the communication device 400 shown in FIG. 4.
  • the C-NWDAF entity 110 provided in this embodiment can perform the above-mentioned training method using the MOS model, the technical effects that can be obtained can refer to the above-mentioned method embodiment, which will not be repeated here.
  • FIG. 12 shows a schematic structural diagram of an E-NWDAF entity 120.
  • the E-NWDAF entity 120 includes a transceiver module 1201 and a processing module 1202.
  • the transceiver module 1201 may also be referred to as a transceiver unit to implement a transceiver function, and may be, for example, a transceiver circuit, transceiver, transceiver, or communication interface.
  • the transceiver module 1201 is configured to receive a first subscription request from the C-NWDAF entity, and the first subscription request is used to request to subscribe to the service quality average opinion score MOS level of the target service and the corresponding first network performance index, and the first network performance
  • the index is the network performance index of the network transmission carrying the target service
  • the processing module 1202 is used to obtain the service quality MOS level and the first network performance index
  • the transceiver module 1201 is also used to send the service quality MOS level and the first network performance index to the C-NWDAF entity 1.
  • Network performance index is the network performance index of the network transmission carrying the target service.
  • the processing module 1202 is used to obtain the service quality MOS level, including: the processing module 1202 is used to obtain service experience data of the target service; and determine the service quality MOS level according to the service experience data of the target service.
  • the E-NWDAF entity 120 is presented in the form of dividing various functional modules in an integrated manner.
  • the "module” here can refer to a specific ASIC, circuit, processor and memory that executes one or more software or firmware programs, integrated logic circuit, and/or other devices that can provide the above-mentioned functions.
  • the E-NWDAF entity 120 may take the form of the communication device 400 shown in FIG. 4.
  • the processor 401 in the communication device 400 shown in FIG. 4 may invoke the computer execution instructions stored in the memory 403 to make the communication device 400 execute the training method of applying the MOS model in the foregoing method embodiment.
  • the functions/implementation process of the transceiver module 1201 and the processing module 1202 in FIG. 12 may be implemented by the processor 401 in the communication device 400 shown in FIG. 4 calling the computer execution instructions stored in the memory 403.
  • the function/implementation process of the processing module 1202 in FIG. 12 can be implemented by the processor 401 in the communication device 400 shown in FIG. 4 calling a computer execution instruction stored in the memory 403, and the function of the transceiver module 1201 in FIG.
  • the implementation process can be implemented through the communication interface 404 in the communication device 400 shown in FIG. 4.
  • the E-NWDAF entity 120 provided in this embodiment can perform the above-mentioned training method using the MOS model, the technical effects that can be obtained can refer to the above-mentioned method embodiment, which will not be repeated here.
  • one or more of the above modules or units can be implemented by software, hardware or a combination of both.
  • the software exists in the form of computer program instructions and is stored in the memory, and the processor can be used to execute the program instructions and implement the above method flow.
  • the processor can be built in SoC (system on chip) or ASIC, or it can be an independent semiconductor chip.
  • SoC system on chip
  • ASIC application specific integrated circuit
  • the processor's internal processing is used to execute software instructions for calculations or processing, and may further include necessary hardware accelerators, such as field programmable gate array (FPGA), PLD (programmable logic device) , Or a logic circuit that implements dedicated logic operations.
  • FPGA field programmable gate array
  • PLD programmable logic device
  • the hardware can be a CPU, a microprocessor, a digital signal processing (digital signal processing, DSP) chip, a microcontroller unit (MCU), an artificial intelligence processor, an ASIC, Any one or any combination of SoC, FPGA, PLD, dedicated digital circuit, hardware accelerator, or non-integrated discrete device can run necessary software or do not rely on software to perform the above method flow.
  • DSP digital signal processing
  • MCU microcontroller unit
  • an artificial intelligence processor an ASIC
  • Any one or any combination of SoC, FPGA, PLD, dedicated digital circuit, hardware accelerator, or non-integrated discrete device can run necessary software or do not rely on software to perform the above method flow.
  • an embodiment of the present application further provides a communication device (for example, the communication device may be a chip or a chip system), and the communication device includes a processor for implementing the method in any of the foregoing method embodiments.
  • the communication device further includes a memory.
  • the memory is used to store necessary program instructions and data, and the processor can call the program code stored in the memory to instruct the communication device to execute the method in any of the foregoing method embodiments.
  • the memory may not be in the communication device.
  • the communication device is a chip system, it may be composed of a chip, or may include a chip and other discrete devices, which is not specifically limited in the embodiment of the present application.
  • A/B can mean A or B; in this application, “and/or "It's just an association relationship that describes the associated objects. It means that there can be three kinds of relationships.
  • a and/or B can mean: A alone exists, A and B exist at the same time, and B exists alone. , B can be singular or plural.
  • plural means two or more than two. "The following at least one item (a)” or similar expressions refers to any combination of these items, including any combination of a single item (a) or plural items (a).
  • At least one item (a) of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple .
  • words such as “first” and “second” are used to distinguish the same items or similar items with substantially the same function and effect. Those skilled in the art can understand that words such as “first” and “second” do not limit the quantity and order of execution, and words such as “first” and “second” do not limit the difference.
  • words such as “exemplary” or “for example” are used as examples, illustrations, or illustrations. Any embodiment or design solution described as “exemplary” or “for example” in the embodiments of the present application should not be construed as being more preferable or advantageous than other embodiments or design solutions. To be precise, words such as “exemplary” or “for example” are used to present related concepts in a specific manner to facilitate understanding.
  • the computer may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
  • a software program it may be implemented in the form of a computer program product in whole or in part.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the computer instructions may be transmitted from a website, computer, server, or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, or may include one or more data storage devices such as servers and data centers that can be integrated with the medium.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Environmental & Geological Engineering (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本申请实施例提供应用MOS模型的训练方法、设备及系统，用于解决现有的业务的MOS模型的训练方法需要依赖AF实体的能力和AF实体的配合，在现网中难以实施的问题。方法包括：中心网络数据分析功能C-NWDAF实体向边缘网络数据分析功能E-NWDAF实体发送第一订阅请求，所述第一订阅请求用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，所述第一网络性能指标为承载所述目标业务的网络传输的网络性能指标；C-NWDAF实体从E-NWDAF实体接收所述业务质量MOS等级和所述第一网络性能指标，并根据所述业务质量MOS等级和所述第一网络性能指标，建立所述目标业务的MOS模型。

Description

应用MOS模型的训练方法、设备及系统
本申请要求于2019年07月23日提交国家知识产权局、申请号为201910668701.5、申请名称为“应用MOS模型的训练方法、设备及系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信技术领域，尤其涉及应用平均意见得分(mean opinion score，MOS)模型的训练方法、设备及系统。
背景技术
第三代合作伙伴计划(3rd generation partnership project，3GPP)23.791中对网络数据分析功能(network data analytics function，NWDAF)的定位如图1所示。具体的，NWDAF可以从第五代核心网(5rd generation core，5GC)网络功能(network functions，NFs)、应用功能(application functions，AFs)以及运营商网络的操作维护系统(operation，administration，and maintenance，OAM)订阅基础数据，然后对订阅的基础数据按照场景进行分析之后，将分析结果反馈给NF和AF执行后续的处理。
针对业务体验保障场景，在3GPP 23.288 6.4被观测到的业务体验相关的网络数据分析(observed service experience related network data analytics)中定义了NWDAF如何从NF和AF订阅数据并实现指定业务的MOS模型的建立过程。具体的，NWDAF实体从AF实体订阅指定业务的业务质量MOS等级，NWDAF实体从5GC NFs实体订阅承载本次业务的网络传输的网络性能指标，进而NWDAF实体根据从AF实体订阅的业务质量MOS等级、以及从5GC NFs实体订阅的网络传输的网络性能指标，建立指定业务的MOS模型(即业务质量MOS等级与网络传输的网络性能指标的数学关系)。
然而，由于上述业务的MOS模型的建立方法中，网络传输的网络性能指标和业务质量MOS等级需要分别从5GC NFs实体和AF实体订阅，因此要实现测量同步，需要依赖5GC NFs实体和AF实体的能力和配合，这在现网中难以实施。
发明内容
本申请实施例提供应用MOS模型的训练方法、设备及系统，用于解决现有的业务的MOS模型的训练方法需要依赖AF实体的能力和AF实体的配合，在现网中难以实施的问题。
为达到上述目的，本申请的实施例采用如下技术方案：
第一方面，提供一种应用平均意见得分MOS模型的训练方法，该方法包括：中心网络数据分析功能(center NWDAF，C-NWDAF)实体向边缘网络数据分析功能(edge NWDAF，E-NWDAF)实体发送第一订阅请求，该第一订阅请求用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，该第一网络性能指标为承载该目标业务的网络传输的网络性能指标；C-NWDAF实体从该E-NWDAF实体接收该业务质量MOS等级和该第一网络性能指标；C-NWDAF实体根据该业务质量MOS等级和该第一网络性能指标，建立该目标业务的MOS模型。其中，本申请实施例中，业务质量MOS等级是在一次业务过程中，通过仪器、工具等手段测量业务过程中的业务质量关键指标，然后对关键指标的测 量结果通过体验模型给出的本次业务的综合得分，也可以称之为业务体验质量评分。例如，可以采用5分值，给出业务质量评价分数为1分、2分、3分、4分或5分。此外，本申请实施例中，建立该目标业务的MOS模型是指建立目标业务的业务质量MOS等级与网络性能指标的数学关系，在此统一说明，以下不再赘述。基于本申请实施例提供的应用MOS模型的训练方法，由于本申请实施例中，运营商部署分布式NWDAF实体，包括C-NWDAF实体和E-NWDAF实体，C-NWDAF实体可以从E-NWDAF实体订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，由E-NWDAF实体同步测量得到目标业务的业务质量MOS等级和对应的第一网络性能指标。也就是说，在运营商内部获取目标业务的业务质量MOS等级和对应的第一网络性能指标，因此更容易实施，测量结果也更为准确。
在一种可能的设计中，该业务质量MOS等级是根据该目标业务的业务体验数据确定的。示例性的，在该目标业务为视频业务的情况下，该业务体验数据包括以下参数中的一个或多个：视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。其中，本申请实施例中的视频业务的码率是指数据传输时，单位时间传送的数据位数，一般单位是千位每秒(kbps)。视频业务的帧率是用于测量显示帧数的量度，测量单位为每秒显示帧数((frames per second，FPS)或赫兹(Hz)。终端设备的分辨率包括显示分辨率和图像分辨率，显示分辨率也可以称之为屏幕分辨率，是指显示器所能显示的像素有多少；图像分辨率则是单位英寸中所包含的像素点数。在此统一说明，以下不再赘述。
在一种可能的设计中，该第一网络性能指标包括以下参数中的一个或多个：终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、往返时延(round-trip time，RTT)、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；以及，该接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；以及，该用户面实体与应用功能实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。其中，本申请实施例中，误码是指经接收、判决、再生后，数字码流中的某些比特发生了差错，使传输的信息质量产生损伤，可以理解为传输过程中，发生错误的数据包。乱序是指当数据包过大时，会将一个数据包拆分为多个满足传输要求的数据包，每个数据包都会有对应的序号以供对端重组，但由于中间路由不同或者网络质量问题，可能会导致后发送的数据包先到达对端，从而产生乱序，其中，包乱序比例高，也会导致通信质量差。抖动是指分组数据包延迟的变化程度，如果网络发生拥塞，排队延迟将影响端到端的延迟，并导致通过同一连接传输的分组延迟各不相同，而抖动，就是用来 描述这样一延迟变化的程度。此外，本申请实施例中的用户面实体主要用于执行用户数据包的转发。在第五代5G通信系统中，用户面实体可以是用户面功能(user plane function，UPF)网元，在未来通信如第六代6G通信中，用户面实体仍可以是UPF网元，或者有其它名称，本申请实施例对此不作限定。本申请实施例中的应用功能实体主要用于面向终端设备提供应用层业务功能。在第五代5G通信系统中，应用功能实体可以是AF网元，在未来通信如第六代6G通信中，应用功能实体仍可以是AF网元，或者有其它名称，本申请实施例对此不作限定。
在一种可能的设计中，该方法还包括：C-NWDAF实体获取该MOS等级对应的第二网络性能指标，该第二网络性能指标为承载该目标业务的无线的网络性能指标；C-NWDAF实体根据该业务质量MOS等级和该第一网络性能指标，建立该目标业务的MOS模型，包括：C-NWDAF实体根据该第二网络性能指标，以及该业务质量MOS等级和该第一网络性能指标，排除因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，建立该目标业务的MOS模型。其中，本申请实施例中，业务质量MOS质差样本是指因终端设备和应用功能实体的异常导致MOS评分较低的样本，在此统一说明，以下不再赘述。基于本申请实施例提供的应用MOS模型的训练方法建立的目标业务的MOS模型，由于不仅考虑到第一网络性能指标，还考虑到承载目标业务的无线的网络性能指标，以及因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，因此目标业务的MOS模型更为准确和全面。
在一种可能的设计中，该方法还包括：C-NWDAF实体获取该MOS等级对应的第一网络功能实体的第一负荷，该第一网络功能实体为网络数据提供者功能实体；C-NWDAF实体根据该业务质量MOS等级和该第一网络性能指标，建立该目标业务的MOS模型，包括：C-NWDAF实体根据该第一网络功能实体的第一负荷，以及该业务质量MOS等级和该第一网络性能指标，排除因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，建立该目标业务的MOS模型。基于本申请实施例提供的应用MOS模型的训练方法建立的目标业务的MOS模型，由于不仅考虑到第一网络性能指标，还考虑到MOS等级对应的第一网络功能实体的第一负荷，以及因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，因此目标业务的MOS模型更为准确和全面。
在一种可能的设计中，该方法还包括：C-NWDAF实体获取该MOS等级对应的第二网络性能指标和该MOS等级对应的第一网络功能实体的第一负荷，该第二网络性能指标为承载该目标业务的无线的网络性能指标，该第一网络功能实体为网络数据提供者功能实体；C-NWDAF实体根据该业务质量MOS等级和该第一网络性能指标，建立该目标业务的MOS模型，包括：C-NWDAF实体根据该第二网络性能指标和该第一网络功能实体的第一负荷，以及该业务质量MOS等级和该第一网络性能指标，排除因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，建立该目标业务的MOS模型。基于本申请实施例提供的应用MOS模型的训练方法建立的目标业务的MOS模型，由于不仅考虑到第一网络性能指标，还考虑到承载目标业务的无线的网络性能指标和MOS等级对应的第一网络功能实体的第一负荷，以及因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，因此目标业务的MOS模型更为准确和全面。
在一种可能的设计中，C-NWDAF实体获取该MOS等级对应的第一网络功能实体的 第一负荷，包括：C-NWDAF实体向网络存储功能实体发送第二订阅请求，该第二订阅请求用于请求订阅该第一网络功能实体的第一负荷；C-NWDAF实体从该网络存储功能实体接收该第一网络功能实体的第一负荷。其中，本申请实施例中的网络存储功能实体用于保存网络功能实体(如第一网络功能实体)以及其提供服务的描述信息，以及支持服务发现，网元或实体发现等。在第五代5G通信系统中，网络存储功能实体可以是网络存储功能(network repository function，NRF)实体，在未来通信如第六代6G通信中，网络存储功能实体仍可以是NRF实体，或者有其它名称，本申请实施例对此不作限定。
基于该方案，C-NWDAF实体可以获取该MOS等级对应的第一网络功能实体的第一负荷。
在一种可能的设计中，C-NWDAF实体获取该MOS等级对应的第二网络性能指标，包括：C-NWDAF实体向运营商网络的操作和维护OAM实体发送第三订阅请求，该第三订阅请求用于请求订阅该第二网络性能指标；C-NWDAF实体从该OAM实体接收该第二网络性能指标。基于该方案，C-NWDAF实体获取该MOS等级对应的第二网络性能指标。
在一种可能的设计中，该第一网络功能实体的第一负荷包括以下参数中的一个或多个：该第一网络功能实体的会话数、该第一网络功能实体的用户数、以及该第一网络功能实体的资源利用率。也就是说，本申请实施例中的目标业务的MOS模型还与第一网络功能实体的会话数、该第一网络功能实体的用户数、以及该第一网络功能实体的资源利用率中的一个或多个相关，因此目标业务的MOS模型更完整。
在一种可能的设计中，该第二网络性能指标包括以下参数中的一个或多个：接入网设备的会话数、无线资源控制(radio resource control，RRC)连接用户数、拥塞状态和资源占用率，终端设备的无线测量指标，以及该终端设备的位置信息。也就是说，本申请实施例中的目标业务的MOS模型还与接入网设备的会话数、RRC连接用户数、拥塞状态和资源占用率，终端设备的无线测量指标，以及该终端设备的位置信息中的一个或多个相关，因此目标业务的MOS模型更完整。
在一种可能的设计中，在C-NWDAF实体向该E-NWDAF实体发送第一订阅请求之前，该方法还包括：C-NWDAF实体从第二网络功能实体接收第四订阅请求，该第四订阅请求用于请求订阅该目标业务的MOS模型，第二网络功能实体为消费者功能实体；在C-NWDAF实体建立该目标业务的MOS模型之后，该方法还包括：C-NWDAF实体向该第二网络功能实体发送该目标业务的MOS模型。也就是说，本申请实施例提供的应用MOS模型的训练方法中，C-NWDAF实体可以基于消费者功能实体的触发向E-NWDAF实体发送第一订阅请求消息。其中，本申请实施例中的消费者功能实体是指向C-NWDAF实体或者NWDAF实体发起数据分析请求或订阅的请求方，在此统一说明，以下不再赘述。
在一种可能的设计中，该方法还包括：C-NWDAF实体从策略控制实体接收第五订阅请求，该第五订阅请求用于请求对目标终端设备的目标业务进行业务体验保障，该业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；C-NWDAF实体向E-NWDAF实体发送第六订阅请求，该第六订阅请求用于请求订阅该目标终端设备使用该目标业务的触发事件、该目标业务的第二业务质量MOS等级和对应的第一网络性能指标；在该触发事件被触发之后，C-NWDAF实体从该E-NWDAF实体接收该第二业务质量MOS等级和对应的第一网络性能指标；在该第二业务质量MOS等级和该第一业务质量MOS等级不同 的情况下，该C-NWDAF实体将该第二业务质量MOS等级对应的第一网络性能指标与该目标业务的MOS模型中该第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络分析结果；C-NWDAF实体向该策略控制实体发送该网络分析结果，该网络分析结果用于进行网络优化。也就是说，基于本申请实施例提供的目标业务的MOS模型，结合终端设备当前的业务体验质量可以为用户提供实时业务体验保障，进而可以提升用户的业务体验感知。
在一种可能的设计中，该方法还包括：C-NWDAF实体从策略控制实体接收第五订阅请求，该第五订阅请求用于请求对目标终端设备的该目标业务进行业务体验保障，该业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；C-NWDAF实体根据该目标终端设备的移动轨迹，确定该目标终端设备将从第一接入网设备的服务范围移动到第二接入网设备的移动范围；C-NWDAF实体向该E-NWDAF实体发送第六订阅请求，该第六订阅请求用于请求订阅第一终端设备使用该目标业务的触发事件、以及该第二接入网设备关联的第一网络性能指标；在该触发事件被触发之后，该C-NWDAF实体从该E-NWDAF实体接收该第二接入网设备关联的第一网络性能指标；C-NWDAF实体将该第二接入网设备关联的第一网络性能指标与该目标业务的MOS模型进行匹配，确定该第二接入网设备关联的第一网络性能指标对应的该目标业务的第三业务质量MOS等级；在该第三业务质量MOS等级和该第一业务质量MOS等级不同的情况下，C-NWDAF实体将与该第三业务质量MOS等级对应的第一网络性能指标与该目标业务的MOS模型中该第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络预测结果；C-NWDAF实体向该策略控制实体发送该网络预测结果，该网络预测结果用于进行网络优化。也就是说，基于本申请实施例提供的目标业务的MOS模型，结合终端设备当前的业务体验质量可以为用户提供实时业务体验保障，进而可以提升用户的业务体验感知。
第二方面，提供一种应用平均意见得分MOS模型的训练方法，该方法包括：边缘网络数据分析功能E-NWDAF实体从中心网络数据分析功能C-NWDAF实体接收第一订阅请求，该第一订阅请求用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，该第一网络性能指标为承载该目标业务的网络传输的网络性能指标；E-NWDAF实体获取该业务质量MOS等级和该第一网络性能指标；E-NWDAF实体向该C-NWDAF实体发送该业务质量MOS等级和该第一网络性能指标。基于本申请实施例提供的应用MOS模型的训练方法，由于本申请实施例中，运营商部署分布式NWDAF实体，包括C-NWDAF实体和E-NWDAF实体，C-NWDAF实体可以从E-NWDAF实体订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，由E-NWDAF实体同步测量得到目标业务的业务质量MOS等级和对应的第一网络性能指标。也就是说，在运营商内部获取目标业务的业务质量MOS等级和对应的第一网络性能指标，因此更容易实施，测量结果也更为准确。
在一种可能的设计中，E-NWDAF实体获取该业务质量MOS等级，包括：E-NWDAF实体获取该目标业务的业务体验数据；E-NWDAF实体根据该目标业务的业务体验数据确定该业务质量MOS等级。基于该方案，E-NWDAF实体可以获取业务质量MOS等级。
示例性的，在该目标业务为视频业务的情况下，该业务体验数据包括以下参数中的一个或多个：该视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。
在一种可能的设计中，该第一网络性能指标包括以下参数中的一个或多个：终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、往返时延RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；以及，该接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；以及，该用户面实体与应用功能实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
第三方面，提供了一种通信装置用于实现上述各种方法。该通信装置可以为上述第一方面中的C-NWDAF实体，或者包含上述C-NWDAF实体的装置；或者，该通信装置可以为上述第一方面中的E-NWDAF实体，或者包含上述E-NWDAF实体的装置。该通信装置包括实现上述方法相应的模块、单元、或手段(means)，该模块、单元、或means可以通过硬件实现，软件实现，或者通过硬件执行相应的软件实现。该硬件或软件包括一个或多个与上述功能相对应的模块或单元。
第四方面，提供了一种通信装置，包括：处理器和存储器；该存储器用于存储计算机指令，当该处理器执行该指令时，以使该通信装置执行上述任一方面所述的方法。该通信装置可以为上述第一方面中的C-NWDAF实体，或者包含上述C-NWDAF实体的装置；或者，该通信装置可以为上述第一方面中的E-NWDAF实体，或者包含上述E-NWDAF实体的装置。
第五方面，提供了一种通信装置，包括：处理器；该处理器用于与存储器耦合，并读取存储器中的指令之后，根据该指令执行如上述任一方面所述的方法。该通信装置可以为上述第一方面中的C-NWDAF实体，或者包含上述C-NWDAF实体的装置；或者，该通信装置可以为上述第一方面中的E-NWDAF实体，或者包含上述E-NWDAF实体的装置。
第六方面，提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机可以执行上述任一方面所述的方法。
第七方面，提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机可以执行上述任一方面所述的方法。
第八方面，提供了一种通信装置(例如，该通信装置可以是芯片或芯片系统)，该通信装置包括处理器，用于实现上述任一方面中所涉及的功能。在一种可能的设计中，该通信装置还包括存储器，该存储器，用于保存必要的程序指令和数据。该通信装置是芯片系统时，可以由芯片构成，也可以包含芯片和其他分立器件。
其中，第三方面至第八方面中任一种设计方式所带来的技术效果可参见上述第一方面 或第二方面中不同设计方式所带来的技术效果，此处不再赘述。
第九方面，提供了一种通信系统，该通信系统包括中心网络数据分析功能C-NWDAF实体和边缘网络数据分析功能E-NWDAF实体；C-NWDAF实体，用于向E-NWDAF实体发送第一订阅请求，该第一订阅请求用于请求订阅目标业务的业务质量平均意见得分MOS等级和对应的第一网络性能指标，该第一网络性能指标为承载该目标业务的网络传输的网络性能指标；E-NWDAF实体，用于从C-NWDAF实体接收该第一订阅请求，并获取该业务质量MOS等级和该第一网络性能指标之后，向该C-NWDAF实体发送该业务质量MOS等级和该第一网络性能指标；C-NWDAF实体，用于从E-NWDAF实体接收该业务质量MOS等级和该第一网络性能指标，并根据该业务质量MOS等级和该第一网络性能指标，建立该目标业务的MOS模型。其中，第九方面的技术效果可参考上述第一方面或第二方面，在此不再赘述。
在一种可能的设计中，该C-NWDAF实体还用于执行上述第一方面任一种可能的设计中所述的应用MOS模型的训练方法。
在一种可能的设计中，该E-NWDAF实体还用于执行上述第二方面任一种可能的设计中所述的应用MOS模型的训练方法。
附图说明
图1为现有的3GPP23.791中对NWDAF的定位示意图；
图2为现有的资源按需定制示意图；
图3为本申请实施例提供的通信系统的架构示意图；
图4为本申请实施例提供的通信设备的结构示意图；
图5为本申请实施例提供的应用MOS模型的训练方法流程示意图；
图6为本申请实施例提供的第一网络性能指标的示意图；
图7为本申请实施例提供的业务体验保障方法流程示意图一；
图8为图7所示的业务体验保障方法对应的架构示意图；
图9为本申请实施例提供的业务体验保障方法流程示意图二；
图10为图9所示的业务体验保障方法对应的架构示意图；
图11为本申请实施例提供的C-NWDAF实体的结构示意图；
图12为本申请实施例提供的E-NWDAF实体的结构示意图。
具体实施方式
为了方便理解本申请实施例的技术方案，首先给出本申请相关技术的简要介绍如下。
随着5G网络的到来，移动网络的架构及承载的业务都发生了很大的变化，产生了车联网、云增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)、高清直播、工业控制等新的应用场景，对5G网络提出了更高的端到端服务质量(quality of service，QoS)要求。比如，增强移动宽带业务(enhanced mobile broadband，eMBB)要求大容量、高速、动态带宽分配，可以高速上传下载GB量级的视频内容，可以为超高清视频、VR/AR等业务动态分配带宽；超高可靠低时延业务(ultra-reliable and low latency communications)要求高可靠、高可用、低时延，可以支持自动化工厂、远程手术等关键任务(mission-critical)业务可靠运行，可以满足自动驾驶，远程控制无人机等时延关键(delay-critical)业务的低时延需求；大物联业务(massive machine type communications，mMTC)要求大容量、高 速、动态带宽分配，可以为智慧城市等物联网(internet of things，IoT)业务提供十亿级设备的连接，密度可以达到百万设备/平方千米。
QoS的目的是在资源有限的情况下，“按需定制”，为业务提供差异化服务质量的网络服务。如图2所示，网络中的资源(如终端设备与接入网设备之间的空口承载资源或者接入网设备与UPF实体之间的下一代网络用户面(next generation user plane，NG-U)隧道(tunnel))对于所有用户、业务共享，但针对不同的业务(如摄像、微信、优酷或者淘宝)，可以按需分配不同的QoS规则，以最低成本提供差异化的服务质量。
QoS通常有两种含义，一是服务质量怎么样，即表征QoS的具体指标(参数)；二是如何保证这些指标，即实现QoS的机制。3GPP 23.501的5.7.4定义了标准的5G QoS的规格(即表征QoS的具体指标(参数))，包括5G QoS标识(5G QoS identifier，5QI)值以及对应的资源类型(resource type)、默认优先级水平(default priority level)、包延时估算(packet delay budget，PDB)、包错误率(packet error rate，PER)、默认(default)最大数据突发流量(maximum data burst volume，MDBV)、默认平均窗口(default averaging window，DAV)和示例性的服务，如表一所示：
表一
Figure PCTCN2020080958-appb-000001
表一中的5QI定义为端到端的网络服务质量等级，网络中的各传输节点(如接入网设备或者核心网设备等)根据QoS文件(QoS profile)对传输数据进行保障，但是否可以保障端到端的业务体验质量未知。因此在3GPP 23.791、23.228中引入NWDAF实体，通过从5GC NFs实体订阅网络传输的网络性能指标(丢包率、RTT和误码率)和从AF实体订阅业务质量MOS等级，推理获取指定业务的MOS模型，用以支撑运营商为终端用户、运营商网络上承载的第三方应用(over the top，OTT)厂商、行业提供差异化服务质量保障配置QoS文件。
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述。需要说明的是，本申请实施例描述的网络架构以及业务场景是为了更加清楚的说明本申请实施例 的技术方案，并不构成对于本申请实施例提供的技术方案的限定，本领域普通技术人员可知，随着网络架构的演变和新业务场景的出现，本申请实施例提供的技术方案对于类似的技术问题，同样适用。
如图3所示，为本申请实施例提供的一种通信系统30，该通信系统30包括C-NWDAF实体301和E-NWDAF实体302。C-NWDAF实体301和E-NWDAF实体302为NWDAF实体的分布式结构。
其中，E-NWDAF实体302可以部署在5GC NFs侧、接入网设备侧或者终端设备内部。比如，E-NWDAF实体302可以以软件开发工具包(software development kit，SDK)方式内嵌在终端设备中，或者通过软件插件以网络功能虚拟化(network function virtualization，NFV)方式和UPF实体共同部署，或者采用通用服务器软件方式部署在接入网设备或者UPF实体的周边等。本申请实施例中，E-NWDAF实体302主要用于采集目标业务的业务质量MOS等级和对应的第一网络性能指标，并将目标业务的业务质量MOS等级和对应的第一网络性能指标提供给C-NWDAF实体301，该第一网络性能指标为承载目标业务的网络传输的网络性能指标。
C-NWDAF实体301为中心节点，主要用于从E-NWDAF实体302获取目标业务的业务质量MOS等级和对应的第一网络性能指标，并根据目标业务的业务质量MOS等级和对应的第一网络性能指标，建立目标业务的MOS模型。可选的，如图3所示，该通信系统还可以包括网络存储功能实体303、OAM304或者终端设备305中的一个或多个。
C-NWDAF实体301还可以与网络存储功能实体303、终端设备305、或者OAM304中的一个或多个交互以收集数据，配合目标业务的业务质量MOS等级和对应的第一网络性能指标，完成最终MOS推理和分析。
需要说明的是，本申请实施例中的实体也可以称之为网元，在此统一说明，以下不再赘述。
基于该通信系统进行应用MOS模型训练的方法可参考后续方法实施例，在此不再赘述。
可选的，本申请实施例中的终端设备，可以是用于实现无线通信功能的设备，例如终端或者可用于终端中的芯片等。其中，终端可以是5G网络或者未来演进的PLMN中的用户设备(user equipment，UE)、接入终端、终端单元、终端站、移动站、移动台、远方站、远程终端、移动设备、无线通信设备、终端代理或终端装置等。接入终端可以是蜂窝电话、无绳电话、会话启动协议(session initiation protocol，SIP)电话、无线本地环路(wireless local loop，WLL)站、个人数字处理(personal digital assistant，PDA)、具有无线通信功能的手持设备、计算设备或连接到无线调制解调器的其它处理设备、车载设备或可穿戴设备，虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、工业控制(industrial control)中的无线终端、无人驾驶(self driving)中的无线终端、远程医疗(remote medical)中的无线终端、智能电网(smart grid)中的无线终端、运输安全(transportation safety)中的无线终端、智慧城市(smart city)中的无线终端、智慧家庭(smart home)中的无线终端等。终端可以是移动的，也可以是固定的。
可选的，本申请实施例中的接入网设备指的是接入核心网的设备，例如可以是基站，宽带网络业务网关(broadband network gateway，BNG)，汇聚交换机，非3GPP接入设备 等。基站可以包括各种形式的基站，例如：宏基站，微基站(也称为小站)，中继站，接入点等。
可选的，本申请实施例中的C-NWDAF实体或者E-NWDAF实体的相关功能可以由一个设备实现，也可以由多个设备共同实现，还可以是由一个设备内的一个或多个功能模块实现，本申请实施例对此不作具体限定。可以理解的是，上述功能既可以是硬件设备中的网络元件，也可以是在专用硬件上运行的软件功能，或者是硬件与软件的结合，或者是平台(例如，云平台)上实例化的虚拟化功能。
例如，本申请实施例中的C-NWDAF实体或者E-NWDAF实体的相关功能可以通过图4中的通信设备400来实现。图4所示为本申请实施例提供的通信设备400的结构示意图。该通信设备400包括一个或多个处理器401，通信线路402，以及至少一个通信接口(图4中仅是示例性的以包括通信接口404，以及一个处理器401为例进行说明)，可选的还可以包括存储器403。
处理器401可以是一个通用中央处理器(central processing unit，CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制本申请方案程序执行的集成电路。
通信线路402可包括一通路，用于连接不同组件之间。
通信接口404，可以是收发模块用于与其他设备或通信网络通信，如以太网，RAN，无线局域网(wireless local area networks，WLAN)等。例如，所述收发模块可以是收发器、收发机一类的装置。可选的，所述通信接口404也可以是位于处理器401内的收发电路，用以实现处理器的信号输入和信号输出。
存储器403可以是具有存储功能的装置。例如可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过通信线路402与处理器相连接。存储器也可以和处理器集成在一起。
其中，存储器403用于存储执行本申请方案的计算机执行指令，并由处理器401来控制执行。处理器401用于执行存储器403中存储的计算机执行指令，从而实现本申请实施例中提供的应用MOS模型的训练方法。
或者，可选的，本申请实施例中，也可以是处理器401执行本申请下述实施例提供的应用MOS模型的训练方法中的处理相关的功能，通信接口404负责与其他设备或通信网络通信，本申请实施例对此不作具体限定。
可选的，本申请实施例中的计算机执行指令也可以称之为应用程序代码，本申请实施例对此不作具体限定。
在具体实现中，作为一种实施例，处理器401可以包括一个或多个CPU，例如图4中的CPU0和CPU1。
在具体实现中，作为一种实施例，通信设备400可以包括多个处理器，例如图4中的处理器401和处理器408。这些处理器中的每一个可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，通信设备400还可以包括输出设备405和输入设备406。输出设备405和处理器401通信，可以以多种方式来显示信息。例如，输出设备405可以是液晶显示器(liquid crystal display，LCD)，发光二级管(light emitting diode，LED)显示设备，阴极射线管(cathode ray tube，CRT)显示设备，或投影仪(projector)等。输入设备406和处理器401通信，可以以多种方式接收用户的输入。例如，输入设备406可以是鼠标、键盘、触摸屏设备或传感设备等。
上述的通信设备400有时也可以称为通信装置，其可以是一个通用设备或者是一个专用设备。例如该通信设备400可以为服务器、路由器、交换机或者网桥等AP，或者，该通信设备400可以为手机、平板电脑，电脑笔记本，智能手表，智能电视等STA，本申请实施例不限定通信设备400的类型。
下面将结合图2至图4对本申请实施例提供的应用MOS模型的训练方法进行具体阐述。
需要说明的是，本申请下述实施例中各个实体之间的消息名字或消息中各参数的名字等只是一个示例，具体实现中也可以是其他的名字，本申请实施例对此不作具体限定。
结合图3所示的通信系统，以网络存储功能实体为5G网络中的NRF实体为例，如图5所示，为本申请实施例提供的一种应用MOS模型的训练方法，包括如下步骤：
S501、可选的，消费者NF实体向C-NWDAF实体发送订阅请求1。相应的，C-NWDAF实体接收来自消费者NF实体的订阅请求1，该订阅请求1用于请求订阅指定业务的MOS模型。
需要说明的是，步骤S501为可选步骤，即本申请实施例提供的应用MOS模型的训练方法也可以不执行步骤S501，在此统一说明，以下不再赘述。
S502、C-NWDAF实体向E-NWDAF实体发送订阅请求2。相应的，E-NWDAF实体接收来自C-NWDAF实体的订阅请求2，该订阅请求2用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，该第一网络性能指标为承载目标业务的网络传输的网络性能指标。
可选的，本申请实施例中，第一网络性能指标包括以下参数中的一个或多个：
终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
以及，接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行 平均包抖动、上行速率以及下行速率；
以及，用户面实体与AF实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
需要说明的是，本申请实施例中的RTT包括上行RTT和下行RTT，其中，上行RTT可以理解为终端设备发送上行数据包以及接收上行数据包的响应帧的时间间隔，下行RTT可以理解为AF实体发送下行数据包以及接收下行数据包的响应帧的时间间隔，在此统一说明，以下不再赘述。
示例性的，以接入网设备为5G网络中的下一代节点B(next generation nodeB，gNB)，用户面实体为5G网络中的UPF实体为例，如图6所示，为本申请实施例提供的第一网络性能指标的示意图，包括终端设备和gNB之间的网络性能指标、gNB与UPF实体之间的网络性能指标、或者UPF实体与AF实体之间的网络性能指标中的一个或多个。
S503、E-NWDAF实体获取目标业务的业务质量MOS等级和对应的第一网络性能指标。
可选的，本申请实施例中，E-NWDAF实体可以通过流量镜像或植入NF代理等方式采集用户的原始业务数据，并完成以下处理：
第一，识别用户的业务类型，如视频业务或者VR业务。
第二，根据用户的业务体验测量业务体验数据，并根据业务体验数据确定业务质量MOS等级。
示例性的，在目标业务为视频业务的情况下，业务体验数据包括以下参数中的一个或多个：视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。
第三，测量承载目标业务的网络传输的网络性能指标，即测量上述第一网络性能指标。
S504、E-NWDAF实体向C-NWDAF实体发送通知消息2。C-NWDAF实体接收来自E-NWDAF实体的通知消息2，该通知消息2包括目标业务的业务质量MOS等级和对应的第一网络性能指标。
可选的，本申请实施例提供的应用MOS模型的训练方法还可以包括如下步骤
S505-S507：
S505、C-NWDAF实体向NRF实体/5GC NFs实体发送订阅请求3。相应的，NRF实体/5GC NFs实体接收来自C-NWDAF实体的订阅请求3，该订阅请求3用于请求订阅5GC NFs实体的第一负荷。
可选的，本申请实施例中，5GC NFs实体的第一负荷包括以下参数中的一个或多个：
5GC NFs实体的会话数、5GC NFs实体的用户数、以及5GC NFs实体的资源利用率。
示例性的，5GC NFs实体的资源利用率例如可以包括CPU、内存或者网络IO的利用率。
S506、NRF实体/5GC NFs实体获取5GC NFs实体的第一负荷。
S507、NRF实体/5GC NFs实体向C-NWDAF实体发送通知消息3。相应的，C-NWDAF 实体接收来自NRF实体/5GC NFs实体的通知消息3，该通知消息3包括5GC NFs实体的第一负荷。
可选的，本申请实施例提供的应用MOS模型的训练方法还可以包括如下步骤S508-S510：
S508、C-NWDAF实体向OAM发送订阅请求4。相应的，OAM接收来自C-NWDAF实体的订阅请求4，该订阅请求4用于请求订阅第二网络性能指标，第二网络性能指标为承载目标业务的无线的网络性能指标。
可选的，本申请实施例中，第二网络性能指标包括以下参数中的一个或多个：
接入网设备的会话数、RRC连接用户数、拥塞状态和资源占用率，终端设备的无线测量指标，以及终端设备的位置信息。
示例性的，RRC连接用户数例如可以包括分别处于RRC_非激活(inactive)状态的用户数和RRC_激活(active)状态下的用户数。
示例性的，接入网设备的资源利用率例如可以包括上行或下行物理资源块(physical resource bank，PRB)利用率、者CPU或者内存的利用率。
示例性的，终端设备的无线测量指标例如可以包括参考信号接收功率(reference signal receiving power，RSRP)、参考信号接收质量(reference signal receiving quality，RSRQ)或者信噪比(signal to interference plus noise ratio，SINR)。
S509、OAM获取第二网络性能指标。
S510、OAM向C-NWDAF实体发送通知消息4。相应的，C-NWDAF实体接收来自OAM的通知消息4，该通知消息4包括第二网络性能指标。
可选的，本申请实施例提供的应用MOS模型的训练方法还可以包括如下步骤S511：
S511、终端设备向C-NWDAF实体发送测量报告。相应的，C-NWDAF实体接收来自终端设备的测量报告。测量报告包括终端设备的信息，目标业务的业务质量MOS等级和对应的终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率，或者终端设备的无线测量指标中的一个或多个。
示例性的，终端设备的信息例如可以包括终端设备的型号、CPU或者内存容量。
在C-NWDAF实体获得所需信息之后，本申请实施例提供的应用MOS模型的训练方法还可以包括如下步骤S512：
S512、C-NWDAF实体根据业务质量MOS等级和第一网络性能指标，建立目标业务的MOS模型。该目标业务的MOS模型包括业务质量MOS等级与第一网络性能指标的模型关系。
可选的，在执行步骤S505-S507或者执行步骤S508-S510或者步骤S511的情况下，C-NWDAF实体根据业务质量MOS等级和第一网络性能指标，建立目标业务的MOS模型，包括：C-NWDAF实体根据终端设备测量报告、第二网络性能指标，和/或，5GC NFs实体的第一负荷，以及业务质量MOS等级和第一网络性能指标，排除因终端设备和AF实体的异常导致的业务质量MOS质差样本，建立目标业务的MOS模型。此时，该目标业务的MOS模型包括业务质量MOS等级与第二网络性能指标，和/或，5GC NFs实体的第一负荷，以及业务质量MOS等级和第一网络性能指标的模型关系。
可选的，本申请实施例中，若执行上述步骤S501，则本申请实施例提供的应用MOS模型的训练方法还可以包括如下步骤S513：
S513、C-NWDAF实体向消费者NF实体发送通知消息1。相应的，消费者NF实体接收来自C-NWDAF实体的通知消息1，该通知消息1包括目标业务的MOS模型。
可选的，本申请实施例中，E-NWDAF实体所辖区域内的消费者NF实体也可以直接向E-NWDAF实体订阅指定业务的MOS模型，进而，E-NWDAF实体获取目标业务的业务质量MOS等级和对应的第一网络性能指标之后，根据业务质量MOS等级和第一网络性能指标，建立目标业务的MOS模型，并向E-NWDAF实体所辖区域内的消费者NF实体发送该目标业务的MOS模型，本申请实施例对此不作具体限定。当然，终端设备也可以向E-NWDAF实体发送上述测量报告，E-NWDAF实体也可以订阅业务质量MOS等级与第二网络性能指标，和/或，5GC NFs实体的第一负荷，进而E-NWDAF是实体可以根据终端设备的测量报告、第二网络性能指标，和/或，5GC NFs实体的第一负荷，以及业务质量MOS等级和第一网络性能指标，排除因终端设备和AF实体的异常导致的业务质量MOS质差样本，建立目标业务的MOS模型，本申请实施例对此不作具体限定。
可选的，本申请实施例中，C-NWDAF实体或者E-NWDAF实体获得目标业务的MOS模型之后，可以根据该目标业务的MOS模型生成建议的QoS文件，用以支撑运营商为终端用户、OTT厂商、行业提供差异化服务质量。若消费者NF实体接收到C-NWDAF实体或者E-NWDAF实体发送的建议的QoS文件，可以根据预先配置的优先级规则或对应的判断算法选择使用相应的QoS文件，在此统一说明，以下不赘述。
基于本申请实施例提供的应用MOS模型的训练方法，由于本申请实施例中，运营商部署分布式NWDAF实体，包括C-NWDAF实体和E-NWDAF实体，C-NWDAF实体可以从E-NWDAF实体订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，由E-NWDAF实体同步测量得到目标业务的业务质量MOS等级和对应的第一网络性能指标，也就是说，在运营商内部获取目标业务的业务质量MOS等级和对应的第一网络性能指标，因此更容易实施，测量结果也更为准确。另一方面，基于本申请实施例提供的应用MOS模型的训练方法建立的目标业务的MOS模型，由于不仅考虑到第一网络性能指标，还考虑到承载目标业务的无线的网络性能指标，和/或，MOS等级对应的5GC NFs实体的第一负荷，以及因终端设备和AF实体的异常导致的业务质量MOS质差样本，因此目标业务的MOS模型更为准确和全面。
上述步骤S501至S513中的C-NWDAF实体或者E-NWDAF实体的动作可以由图4所示的通信设备400中的处理器401调用存储器403中存储的应用程序代码来执行，本申请实施例对此不作任何限制。
下面给出基于本申请实施例提供的目标业务的MOS模型，结合终端设备当前的业务体验质量为用户提供实时业务体验保障的相关示例。
示例1、以5G网络中的业务体验保障为例，如图7所示，本申请实施例提供的一种业务体验保障方法包括如下步骤：
S701、AF实体通过网络开放功能(network exposure function，NEF)实体向策略控制功能(policy control function，PCF)实体申请对目标终端设备的目标业务进行业务体验保障。
可选的，本申请实施例中，对于可信应用，AF实体也可以直接向PCF实体申请对目标业务的业务体验保障，本申请实施例对此不作具体限定。
可选的，本申请实施例中的业务体验保障也可以替换为QoS优化，在此统一说明，以下不再赘述。
S702、PCF实体向C-NWDAF实体发送订阅请求5。相应的，C-NWDAF实体接收来自PCF实体的订阅请求5，该订阅请求5用于请求对目标终端设备的目标业务进行业务体验保障，业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级。
S703、C-NWDAF实体向E-NWDAF实体发送订阅请求6。相应的，E-NWDAF实体接收来自C-NWDAF实体的订阅请求6，订阅请求6用于请求订阅目标终端设备使用目标业务的触发事件、目标业务的第二业务质量MOS等级和对应的第一网络性能指标。
其中，第一网络性能指标的相关描述可参考图5所示的实施例，在此不再赘述。
S704、在触发事件被触发之后，E-NWDAF实体获取目标业务的第二业务质量MOS等级和对应的第一网络性能指标，并向C-NWDAF实体发送通知消息6，通知消息6包括当前目标业务的第二业务质量MOS等级和对应的第一网络性能指标。
S705、C-NWDAF实体向OAM发送订阅请求7。相应的，OAM接收来自C-NWDAF实体的订阅请求7，订阅请求7用于请求订阅第二网络性能指标。
其中，第二网络性能指标的相关描述可参考图5所示的实施例，在此不再赘述。
S706、OAM实体获取第二网络性能指标之后，向C-NWDAF实体发送通知消息7。相应的，C-NWDAF实体接收来自OAM的通知消息7，通知消息7包括当前的第二网络性能指标。
S707、在第二业务质量MOS等级和第一业务质量MOS等级不同的情况下，C-NWDAF实体将第二业务质量MOS等级对应的第一网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第一网络性能指标进行匹配，以及，C-NWDAF实体将第二业务质量MOS等级对应的第二网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第二网络性能指标进行匹配，得到网络分析结果。
可选的，本申请实施例中的网络分析结果用于表征什么原因导致的第二业务质量MOS等级和第一业务质量MOS等级不同，如终端设备和当前的接入网设备之间的某个网络性能指标较差、或者当前的接入网设备与UPF实体之间的某个网络性能指标较差、或者UPF实体与AF实体之间的某个网络性能指标较差，等等，在此不做具体限定。
需要说明的是，上述步骤S705-S706为可选的步骤，若不执行步骤S705-S706，则步骤S707中C-NWDAF实体不需要将第二业务质量MOS等级对应的第二网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第二网络性能指标进行匹配，在此统一说明，以下不再赘述。
S708、C-NWDAF实体向PCF实体发送通知消息5。相应的，PCF实体接收来自C-NWDAF实体的通知消息5，通知消息5包括网络分析结果。
可选的，本申请实施例中，PCF实体可以根据网络分析结果通知会话管理功能(session management function，SMF)实体或者UPF实体更新终端设备的QoS策略和网络参数调优，如无线调度优先级、传输层差分服务代码点(differentiated services code point，DSCP)级别等，本申请实施例对此不作具体限定。
图7所示的业务体验保障方法对应的架构示意图可以如图8所示。基于本申请实施例提供的业务体验保障方法，可以为用户提供实时业务体验保障。
示例2、以5G网络中的业务体验保障为例，如图9所示，本申请实施例提供的另一种业务体验保障方法包括如下步骤：
S901、AF实体通过NEF实体向PCF实体申请对目标终端设备的目标业务进行业务体验保障。
可选的，本申请实施例中，对于可信应用，AF实体也可以直接向PCF实体申请对目标业务的业务体验保障，本申请实施例对此不作具体限定。
可选的，本申请实施例中的业务体验保障也可以替换为QoS优化，在此统一说明，以下不再赘述。
S902、PCF实体向C-NWDAF实体发送订阅请求5。相应的，C-NWDAF实体接收来自PCF实体的订阅请求5，该订阅请求5用于请求对目标终端设备的目标业务进行业务体验保障，业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级。
S903、C-NWDAF实体根据目标终端设备的移动轨迹，确定目标终端设备将从第一接入网设备的服务范围移动到第二接入网设备的服务范围。
可选的，本申请实施例中，目标终端设备的移动轨迹可以是E-NWDAF实体获取之后发送给C-NWDAF实体的，也可以是C-NWDAF实体通过其他方式获取的，本申请实施例对此不作具体限定。
S904、C-NWDAF实体向E-NWDAF实体发送订阅请求8。相应的，E-NWDAF实体接收来自C-NWDAF实体的订阅请求8，订阅请求8用于请求订阅目标终端设备使用目标业务的触发事件、以及第二接入网设备关联的第一网络性能指标。
示例性的，本申请实施例中，第二接入网设备关联的第一网络性能指标例如可以包括以下参数中的一个或多个：
终端设备和第二接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
以及，第二接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
以及，用户面实体与AF实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
S905、在触发事件被触发之后，E-NWDAF实体获取第二接入网设备关联的第一网络 性能指标，并向C-NWDAF实体发送通知消息8，通知消息8包括第二接入网设备关联的第一网络性能指标。
S906、C-NWDAF实体向OAM发送订阅请求9。相应的，OAM接收来自C-NWDAF实体的订阅请求9，订阅请求9用于请求订阅第二接入网设备关联的第二网络性能指标。
可选的，本申请实施例中，第二接入网设备关联的第二网络性能指标包括以下参数中的一个或多个：
第二接入网设备的会话数、RRC连接用户数、拥塞状态和资源占用率，终端设备的无线测量指标，以及终端设备的位置信息。其中，RRC连接用户数、资源利用率以及终端设备的无线测量指标的相关描述可参考图5所示的实施例的描述，在此不再赘述。
S907、OAM实体获取第二接入网设备关联的第二网络性能指标之后，向C-NWDAF实体发送通知消息9。相应的，C-NWDAF实体接收来自OAM的通知消息9，通知消息9包括第二接入网设备关联的第二网络性能指标。
S908、C-NWDAF实体将第二接入网设备关联的第一网络性能指标和第二网络性能指标与目标业务的MOS模型进行匹配，确定第二接入网设备关联的第一网络性能指标和第二网络性能指标对应的目标业务的第三业务质量MOS等级。
S909、在第三业务质量MOS等级和第一业务质量MOS等级不同的情况下，C-NWDAF实体将与第三业务质量MOS等级对应的第一网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第一网络性能指标进行匹配，以及，C-NWDAF实体将与第三业务质量MOS等级对应的第二网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第二网络性能指标进行匹配，得到网络预测结果。
可选的，本申请实施例中的网络预测结果用于表征什么原因导致的第三业务质量MOS等级和第一业务质量MOS等级不同，如终端设备和当前的接入网设备之间的某个网络性能指标较差、或者当前的接入网设备与UPF实体之间的某个网络性能指标较差、或者UPF实体与AF实体之间的某个网络性能指标较差，等等，在此不做具体限定。
需要说明的是，上述步骤S906-S907为可选的步骤，若不执行步骤S906-S907，则步骤S908中C-NWDAF实体不需要将第二接入网设备关联的第二网络性能指标与目标业务的MOS模型进行匹配，仅需要确定第二接入网设备关联的第一网络性能指标对应的目标业务的第三业务质量MOS等级即可。同时，若不执行步骤S906-S907，则步骤S909中C-NWDAF实体不需要将第三业务质量MOS等级对应的第二网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第二网络性能指标进行匹配，在此统一说明，以下不再赘述。
S910、C-NWDAF实体向PCF实体发送通知消息5。相应的，PCF实体接收来自C-NWDAF实体的通知消息5，通知消息5包括网络预测结果。
可选的，本申请实施例中，PCF实体可以根据网络预测结果建议UPF实体在进入第二接入网设备的服务范围之后更新用户的QoS策略，本申请实施例对此不作具体限定。
图9所示的业务体验保障方法对应的架构示意图可以如图10所示。基于本申请实施例提供的业务体验保障方法，可以为用户提供实时业务体验保障。
可以理解的是，以上各个实施例中，由C-NWDAF实体实现的方法和/或步骤，也可以由可用于C-NWDAF实体的部件实现；由E-NWDAF实体实现的方法和/或步骤，也可以 由可用于E-NWDAF实体的部件实现。
上述主要从各个实体之间交互的角度对本申请实施例提供的方案进行了介绍。相应的，本申请实施例还提供了通信装置，该通信装置可以为上述方法实施例中的C-NWDAF实体，或者包含上述C-NWDAF实体的装置，或者为可用于C-NWDAF实体的部件；或者，该通信装置可以为上述方法实施例中的E-NWDAF实体，或者包含上述E-NWDAF实体的装置，或者为可用于E-NWDAF实体的部件。可以理解的是，该通信装置为了实现上述功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
比如，以通信装置为上述方法实施例中的C-NWDAF实体为例，图11示出了一种C-NWDAF实体110的结构示意图。该C-NWDAF实体110包括收发模块1101和处理模块1102。所述收发模块1101，也可以称为收发单元用以实现收发功能，例如可以是收发电路，收发机，收发器或者通信接口。
其中，收发模块1101，用于向E-NWDAF实体发送第一订阅请求，第一订阅请求用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，第一网络性能指标为承载目标业务的网络传输的网络性能指标；收发模块1101，还用于从E-NWDAF实体接收业务质量MOS等级和第一网络性能指标；处理模块1102，用于根据业务质量MOS等级和第一网络性能指标，建立目标业务的MOS模型。
可选的，处理模块1102，还用于获取MOS等级对应的第二网络性能指标，和/或，MOS等级对应的第一网络功能实体的第一负荷，第二网络性能指标为承载目标业务的无线的网络性能指标，第一网络功能实体为网络数据提供者功能实体；处理模块1102，用于根据业务质量MOS等级和第一网络性能指标，建立目标业务的MOS模型，包括：处理模块1102，用于根据第二网络性能指标，和/或，第一网络功能实体的第一负荷，以及业务质量MOS等级和第一网络性能指标，排除因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，建立目标业务的MOS模型。
可选的，处理模块1102，用于获取MOS等级对应的第一网络功能实体的第一负荷，包括：处理模块1102，用于通过收发模块1101向网络存储功能实体发送第二订阅请求，第二订阅请求用于请求订阅第一网络功能实体的第一负荷；通过收发模块1101从网络存储功能实体接收第一网络功能实体的第一负荷。
可选的，处理模块1102，用于获取MOS等级对应的第二网络性能指标，包括：处理模块1102，用于通过收发模块1101向运营商网络的操作和维护OAM实体发送第三订阅请求，第三订阅请求用于请求订阅第二网络性能指标；通过收发模块1101从OAM实体接收第二网络性能指标。
可选的，收发模块1101，还用于从第二网络功能实体接收第四订阅请求，第四订阅请求用于请求订阅目标业务的MOS模型，第二网络功能实体为消费者功能实体；收发模块1101，还用于向第二网络功能实体发送目标业务的MOS模型。
可选的，收发模块1101，还用于从策略控制实体接收第五订阅请求，第五订阅请求用 于请求对目标终端设备的目标业务进行业务体验保障，业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；收发模块1101，还用于向E-NWDAF实体发送第六订阅请求，第六订阅请求用于请求订阅目标终端设备使用目标业务的触发事件、目标业务的第二业务质量MOS等级和对应的第一网络性能指标；收发模块1101，还用于在触发事件被触发之后，从E-NWDAF实体接收第二业务质量MOS等级和对应的第一网络性能指标；处理模块1102，还用于在第二业务质量MOS等级和第一业务质量MOS等级不同的情况下，将第二业务质量MOS等级对应的第一网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络分析结果；收发模块1101，还用于向策略控制实体发送网络分析结果，网络分析结果用于进行网络优化。
可选的，收发模块1101，还用于从策略控制实体接收第五订阅请求，第五订阅请求用于请求对目标终端设备的目标业务进行业务体验保障，业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；处理模块1102，还用于根据目标终端设备的移动轨迹，确定目标终端设备将从第一接入网设备的服务范围移动到第二接入网设备的移动范围；收发模块1101，还用于向E-NWDAF实体发送第六订阅请求，第六订阅请求用于请求订阅第一终端设备使用目标业务的触发事件、以及第二接入网设备关联的第一网络性能指标；收发模块1101，还用于在触发事件被触发之后，从E-NWDAF实体接收第二接入网设备关联的第一网络性能指标；处理模块1102，还用于将第二接入网设备关联的第一网络性能指标与目标业务的MOS模型进行匹配，确定第二接入网设备关联的第一网络性能指标对应的目标业务的第三业务质量MOS等级；处理模块1102，还用于在第三业务质量MOS等级和第一业务质量MOS等级不同的情况下，将与第三业务质量MOS等级对应的第一网络性能指标与目标业务的MOS模型中第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络预测结果；收发模块1101，还用于向策略控制实体发送网络预测结果，网络预测结果用于进行网络优化。
其中，上述方法实施例涉及的各步骤的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在本实施例中，该C-NWDAF实体110以采用集成的方式划分各个功能模块的形式来呈现。这里的“模块”可以指特定ASIC，电路，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。在一个简单的实施例中，本领域的技术人员可以想到该C-NWDAF实体110可以采用图4所示的通信设备400的形式。
比如，图4所示的通信设备400中的处理器401可以通过调用存储器403中存储的计算机执行指令，使得通信设备400执行上述方法实施例中的应用MOS模型的训练方法。
具体的，图11中的收发模块1101和处理模块1102的功能/实现过程可以通过图4所示的通信设备400中的处理器401调用存储器403中存储的计算机执行指令来实现。或者，图11中的处理模块1102的功能/实现过程可以通过图4所示的通信设备400中的处理器401调用存储器403中存储的计算机执行指令来实现，图11中的收发模块1101的功能/实现过程可以通过图4中所示的通信设备400中的通信接口404来实现。
由于本实施例提供的C-NWDAF实体110可执行上述的应用MOS模型的训练方法，因此其所能获得的技术效果可参考上述方法实施例，在此不再赘述。
比如，以通信装置为上述方法实施例中的E-NWDAF实体为例，图12示出了一种E-NWDAF实体120的结构示意图。该E-NWDAF实体120包括收发模块1201和处理模块1202。所述收发模块1201，也可以称为收发单元用以实现收发功能，例如可以是收发电路，收发机，收发器或者通信接口。
其中，收发模块1201，用于从C-NWDAF实体接收第一订阅请求，第一订阅请求用于请求订阅目标业务的业务质量平均意见得分MOS等级和对应的第一网络性能指标，第一网络性能指标为承载目标业务的网络传输的网络性能指标；处理模块1202，用于获取业务质量MOS等级和第一网络性能指标；收发模块1201，还用于向C-NWDAF实体发送业务质量MOS等级和第一网络性能指标。
可选的，处理模块1202，用于获取业务质量MOS等级，包括：处理模块1202，用于获取目标业务的业务体验数据；根据目标业务的业务体验数据确定业务质量MOS等级。
其中，上述方法实施例涉及的各步骤的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在本实施例中，该E-NWDAF实体120以采用集成的方式划分各个功能模块的形式来呈现。这里的“模块”可以指特定ASIC，电路，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。在一个简单的实施例中，本领域的技术人员可以想到该E-NWDAF实体120可以采用图4所示的通信设备400的形式。
比如，图4所示的通信设备400中的处理器401可以通过调用存储器403中存储的计算机执行指令，使得通信设备400执行上述方法实施例中的应用MOS模型的训练方法。
具体的，图12中的收发模块1201和处理模块1202的功能/实现过程可以通过图4所示的通信设备400中的处理器401调用存储器403中存储的计算机执行指令来实现。或者，图12中的处理模块1202的功能/实现过程可以通过图4所示的通信设备400中的处理器401调用存储器403中存储的计算机执行指令来实现，图12中的收发模块1201的功能/实现过程可以通过图4中所示的通信设备400中的通信接口404来实现。
由于本实施例提供的E-NWDAF实体120可执行上述的应用MOS模型的训练方法，因此其所能获得的技术效果可参考上述方法实施例，在此不再赘述。
需要说明的是，以上模块或单元的一个或多个可以软件、硬件或二者结合来实现。当以上任一模块或单元以软件实现的时候，所述软件以计算机程序指令的方式存在，并被存储在存储器中，处理器可以用于执行所述程序指令并实现以上方法流程。该处理器可以内置于SoC(片上系统)或ASIC，也可是一个独立的半导体芯片。该处理器内处理用于执行软件指令以进行运算或处理的核外，还可进一步包括必要的硬件加速器，如现场可编程门阵列(field programmable gate array，FPGA)、PLD(可编程逻辑器件)、或者实现专用逻辑运算的逻辑电路。
当以上模块或单元以硬件实现的时候，该硬件可以是CPU、微处理器、数字信号处理(digital signal processing，DSP)芯片、微控制单元(microcontroller unit，MCU)、人工智能处理器、ASIC、SoC、FPGA、PLD、专用数字电路、硬件加速器或非集成的分立器件中的任一个或任一组合，其可以运行必要的软件或不依赖于软件以执行以上方法流程。
可选的，本申请实施例还提供了一种通信装置(例如，该通信装置可以是芯片或芯片 系统)，该通信装置包括处理器，用于实现上述任一方法实施例中的方法。在一种可能的设计中，该通信装置还包括存储器。该存储器，用于保存必要的程序指令和数据，处理器可以调用存储器中存储的程序代码以指令该通信装置执行上述任一方法实施例中的方法。当然，存储器也可以不在该通信装置中。该通信装置是芯片系统时，可以由芯片构成，也可以包含芯片和其他分立器件，本申请实施例对此不作具体限定。
其中，在本申请的描述中，除非另有说明，“/”表示前后关联的对象是一种“或”的关系，例如，A/B可以表示A或B；本申请中的“和/或”仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况，其中A,B可以是单数或者复数。并且，在本申请的描述中，除非另有说明，“多个”是指两个或多于两个。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a，b，或c中的至少一项(个)，可以表示：a，b，c，a-b，a-c，b-c，或a-b-c，其中a，b，c可以是单个，也可以是多个。另外，为了便于清楚描述本申请实施例的技术方案，在本申请的实施例中，采用了“第一”、“第二”等字样对功能和作用基本相同的相同项或相似项进行区分。本领域技术人员可以理解“第一”、“第二”等字样并不对数量和执行次序进行限定，并且“第一”、“第二”等字样也并不限定一定不同。同时，在本申请实施例中，“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言，使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念，便于理解。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件程序实现时，可以全部或部分地以计算机程序产品的形式来实现。该计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或者数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line，DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可以用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)，光介质(例如，DVD)、或者半导体介质(例如固态硬盘(solid state disk，SSD))等。
尽管在此结合各实施例对本申请进行了描述，然而，在实施所要求保护的本申请过程中，本领域技术人员通过查看所述附图、公开内容、以及所附权利要求书，可理解并实现所述公开实施例的其他变化。在权利要求中，“包括”(comprising)一词不排除其他组成部分或步骤，“一”或“一个”不排除多个的情况。单个处理器或其他单元可以实现权利要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施，但这并不表示这些措施不能组合起来产生良好的效果。
尽管结合具体特征及其实施例对本申请进行了描述，显而易见的，在不脱离本申请的 精神和范围的情况下，可对其进行各种修改和组合。相应地，本说明书和附图仅仅是所附权利要求所界定的本申请的示例性说明，且视为已覆盖本申请范围内的任意和所有修改、变化、组合或等同物。显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (37)

  1. 一种应用平均意见得分MOS模型的训练方法，其特征在于，所述方法包括：
    中心网络数据分析功能C-NWDAF实体向边缘网络数据分析功能E-NWDAF实体发送第一订阅请求，所述第一订阅请求用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，所述第一网络性能指标为承载所述目标业务的网络传输的网络性能指标；
    所述C-NWDAF实体从所述E-NWDAF实体接收所述业务质量MOS等级和所述第一网络性能指标；
    所述C-NWDAF实体根据所述业务质量MOS等级和所述第一网络性能指标，建立所述目标业务的MOS模型。
  2. 根据权利要求1所述的方法，其特征在于，所述业务质量MOS等级是根据所述目标业务的业务体验数据确定的。
  3. 根据权利要求2所述的方法，其特征在于，在所述目标业务为视频业务的情况下，所述业务体验数据包括以下参数中的一个或多个：
    所述视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。
  4. 根据权利要求1-3任一项所述的方法，其特征在于，所述第一网络性能指标包括以下参数中的一个或多个：
    终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、往返时延RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述用户面实体与应用功能实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
  5. 根据权利要求1-4任一项所述的方法，其特征在于，所述方法还包括：
    所述C-NWDAF实体获取所述MOS等级对应的第二网络性能指标，和/或，所述MOS等级对应的第一网络功能实体的第一负荷，所述第二网络性能指标为承载所述目标业务的无线的网络性能指标，所述第一网络功能实体为网络数据提供者功能实体；
    所述C-NWDAF实体根据所述业务质量MOS等级和所述第一网络性能指标，建立所述目标业务的MOS模型，包括：
    所述C-NWDAF实体根据所述第二网络性能指标，和/或，所述第一网络功能实体的第一负荷，以及所述业务质量MOS等级和所述第一网络性能指标，排除因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，建立所述目标业务的MOS模型。
  6. 根据权利要求5所述的方法，其特征在于，所述C-NWDAF实体获取所述MOS等级对应的第一网络功能实体的第一负荷，包括：
    所述C-NWDAF实体向网络存储功能实体发送第二订阅请求，所述第二订阅请求用于请求订阅所述第一网络功能实体的第一负荷；
    所述C-NWDAF实体从所述网络存储功能实体接收所述第一网络功能实体的第一负荷。
  7. 根据权利要求5或6所述的方法，其特征在于，所述C-NWDAF实体获取所述MOS等级对应的第二网络性能指标，包括：
    所述C-NWDAF实体向运营商网络的操作和维护OAM实体发送第三订阅请求，所述第三订阅请求用于请求订阅所述第二网络性能指标；
    所述C-NWDAF实体从所述OAM实体接收所述第二网络性能指标。
  8. 根据权利要求5-7任一项所述的方法，其特征在于，所述第一网络功能实体的第一负荷包括以下参数中的一个或多个：
    所述第一网络功能实体的会话数、所述第一网络功能实体的用户数、以及所述第一网络功能实体的资源利用率。
  9. 根据权利要求5-8任一项所述的方法，其特征在于，所述第二网络性能指标包括以下参数中的一个或多个：
    接入网设备的会话数、无线资源控制RRC连接用户数、拥塞状态和资源占用率，终端设备的无线测量指标，以及所述终端设备的位置信息。
  10. 根据权利要求1-9任一项所述的方法，其特征在于，在所述C-NWDAF实体向所述E-NWDAF实体发送第一订阅请求之前，所述方法还包括：
    所述C-NWDAF实体从第二网络功能实体接收第四订阅请求，所述第四订阅请求用于请求订阅所述目标业务的MOS模型，第二网络功能实体为消费者功能实体；
    在所述C-NWDAF实体建立所述目标业务的MOS模型之后，所述方法还包括：
    所述C-NWDAF实体向所述第二网络功能实体发送所述目标业务的MOS模型。
  11. 根据权利要求1-10任一项所述的方法，其特征在于，所述方法还包括：
    所述C-NWDAF实体从策略控制实体接收第五订阅请求，所述第五订阅请求用于请求对目标终端设备的所述目标业务进行业务体验保障，所述业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；
    所述C-NWDAF实体向所述E-NWDAF实体发送第六订阅请求，所述第六订阅请求用于请求订阅所述目标终端设备使用所述目标业务的触发事件、所述目标业务的第二业务质量MOS等级和对应的第一网络性能指标；
    在所述触发事件被触发之后，所述C-NWDAF实体从所述E-NWDAF实体接收所述第二业务质量MOS等级和对应的第一网络性能指标；
    在所述第二业务质量MOS等级和所述第一业务质量MOS等级不同的情况下，所 述C-NWDAF实体将所述第二业务质量MOS等级对应的第一网络性能指标与所述目标业务的MOS模型中所述第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络分析结果；
    所述C-NWDAF实体向所述策略控制实体发送所述网络分析结果，所述网络分析结果用于进行网络优化。
  12. 根据权利要求1-10任一项所述的方法，其特征在于，所述方法还包括：
    所述C-NWDAF实体从策略控制实体接收第五订阅请求，所述第五订阅请求用于请求对目标终端设备的所述目标业务进行业务体验保障，所述业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；
    所述C-NWDAF实体根据所述目标终端设备的移动轨迹，确定所述目标终端设备将从第一接入网设备的服务范围移动到第二接入网设备的移动范围；
    所述C-NWDAF实体向所述E-NWDAF实体发送第六订阅请求，所述第六订阅请求用于请求订阅第一终端设备使用所述目标业务的触发事件、以及所述第二接入网设备关联的第一网络性能指标；
    在所述触发事件被触发之后，所述C-NWDAF实体从所述E-NWDAF实体接收所述第二接入网设备关联的第一网络性能指标；
    所述C-NWDAF实体将所述第二接入网设备关联的第一网络性能指标与所述目标业务的MOS模型进行匹配，确定所述第二接入网设备关联的第一网络性能指标对应的所述目标业务的第三业务质量MOS等级；
    在所述第三业务质量MOS等级和所述第一业务质量MOS等级不同的情况下，所述C-NWDAF实体将与所述第三业务质量MOS等级对应的第一网络性能指标与所述目标业务的MOS模型中所述第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络预测结果；
    所述C-NWDAF实体向所述策略控制实体发送所述网络预测结果，所述网络预测结果用于进行网络优化。
  13. 一种应用平均意见得分MOS模型的训练方法，其特征在于，所述方法包括：
    边缘网络数据分析功能E-NWDAF实体从中心网络数据分析功能C-NWDAF实体接收第一订阅请求，所述第一订阅请求用于请求订阅目标业务的业务质量MOS等级和对应的第一网络性能指标，所述第一网络性能指标为承载所述目标业务的网络传输的网络性能指标；
    所述E-NWDAF实体获取所述业务质量MOS等级和所述第一网络性能指标；
    所述E-NWDAF实体向所述C-NWDAF实体发送所述业务质量MOS等级和所述第一网络性能指标。
  14. 根据权利要求13所述的方法，其特征在于，所述E-NWDAF实体获取所述业务质量MOS等级，包括：
    所述E-NWDAF实体获取所述目标业务的业务体验数据；
    所述E-NWDAF实体根据所述目标业务的业务体验数据确定所述业务质量MOS等级。
  15. 根据权利要求14所述的方法，其特征在于，在所述目标业务为视频业务的情 况下，所述业务体验数据包括以下参数中的一个或多个：
    所述视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。
  16. 根据权利要求13-15任一项所述的方法，其特征在于，所述第一网络性能指标包括以下参数中的一个或多个：
    终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、往返时延RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述用户面实体与应用功能实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
  17. 一种中心网络数据分析功能C-NWDAF实体，其特征在于，所述C-NWDAF实体包括：处理模块和收发模块；
    所述收发模块，用于向边缘网络数据分析功能E-NWDAF实体发送第一订阅请求，所述第一订阅请求用于请求订阅目标业务的业务质量平均意见得分MOS等级和对应的第一网络性能指标，所述第一网络性能指标为承载所述目标业务的网络传输的网络性能指标；
    所述收发模块，还用于从所述E-NWDAF实体接收所述业务质量MOS等级和所述第一网络性能指标；
    所述处理模块，用于根据所述业务质量MOS等级和所述第一网络性能指标，建立所述目标业务的MOS模型。
  18. 根据权利要求17所述的C-NWDAF实体，其特征在于，所述业务质量MOS等级是根据所述目标业务的业务体验数据确定的。
  19. 根据权利要求18所述的C-NWDAF实体，其特征在于，在所述目标业务为视频业务的情况下，所述业务体验数据包括以下参数中的一个或多个：
    所述视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。
  20. 根据权利要求17-19任一项所述的C-NWDAF实体，其特征在于，所述第一网络性能指标包括以下参数中的一个或多个：
    终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、 下行丢包率、往返时延RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述用户面实体与应用功能实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
  21. 根据权利要求17-20任一项所述的C-NWDAF实体，其特征在于，
    所述处理模块，还用于获取所述MOS等级对应的第二网络性能指标，和/或，所述MOS等级对应的第一网络功能实体的第一负荷，所述第二网络性能指标为承载所述目标业务的无线的网络性能指标，所述第一网络功能实体为网络数据提供者功能实体；
    所述处理模块，用于根据所述业务质量MOS等级和所述第一网络性能指标，建立所述目标业务的MOS模型，包括：
    所述处理模块，用于根据所述第二网络性能指标，和/或，所述第一网络功能实体的第一负荷，以及所述业务质量MOS等级和所述第一网络性能指标，排除因终端设备和应用功能实体的异常导致的业务质量MOS质差样本，建立所述目标业务的MOS模型。
  22. 根据权利要求21所述的C-NWDAF实体，其特征在于，所述处理模块，用于获取所述MOS等级对应的第一网络功能实体的第一负荷，包括：
    所述处理模块，用于通过所述收发模块向网络存储功能实体发送第二订阅请求，所述第二订阅请求用于请求订阅所述第一网络功能实体的第一负荷；通过所述收发模块从所述网络存储功能实体接收所述第一网络功能实体的第一负荷。
  23. 根据权利要求21或22所述的C-NWDAF实体，其特征在于，所述处理模块，用于获取所述MOS等级对应的第二网络性能指标，包括：
    所述处理模块，用于通过所述收发模块向运营商网络的操作和维护OAM实体发送第三订阅请求，所述第三订阅请求用于请求订阅所述第二网络性能指标；通过所述收发模块从所述OAM实体接收所述第二网络性能指标。
  24. 根据权利要求21-23任一项所述的C-NWDAF实体，其特征在于，所述第一网络功能实体的第一负荷包括以下参数中的一个或多个：
    所述第一网络功能实体的会话数、所述第一网络功能实体的用户数、以及所述第一网络功能实体的资源利用率。
  25. 根据权利要求21-24任一项所述的C-NWDAF实体，其特征在于，所述第二网络性能指标包括以下参数中的一个或多个：
    接入网设备的会话数、无线资源控制RRC连接用户数、拥塞状态和资源占用率，终端设备的无线测量指标，以及所述终端设备的位置信息。
  26. 根据权利要求17-25任一项所述的C-NWDAF实体，其特征在于，
    所述收发模块，还用于从第二网络功能实体接收第四订阅请求，所述第四订阅请求用于请求订阅所述目标业务的MOS模型，第二网络功能实体为消费者功能实体；
    所述收发模块，还用于向所述第二网络功能实体发送所述目标业务的MOS模型。
  27. 根据权利要求17-26任一项所述的C-NWDAF实体，其特征在于，
    所述收发模块，还用于从策略控制实体接收第五订阅请求，所述第五订阅请求用于请求对目标终端设备的所述目标业务进行业务体验保障，所述业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；
    所述收发模块，还用于向所述E-NWDAF实体发送第六订阅请求，所述第六订阅请求用于请求订阅所述目标终端设备使用所述目标业务的触发事件、所述目标业务的第二业务质量MOS等级和对应的第一网络性能指标；
    所述收发模块，还用于在所述触发事件被触发之后，从所述E-NWDAF实体接收所述第二业务质量MOS等级和对应的第一网络性能指标；
    所述处理模块，还用于在所述第二业务质量MOS等级和所述第一业务质量MOS等级不同的情况下，将所述第二业务质量MOS等级对应的第一网络性能指标与所述目标业务的MOS模型中所述第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络分析结果；
    所述收发模块，还用于向所述策略控制实体发送所述网络分析结果，所述网络分析结果用于进行网络优化。
  28. 根据权利要求17-26任一项所述的C-NWDAF实体，其特征在于，
    所述收发模块，还用于从策略控制实体接收第五订阅请求，所述第五订阅请求用于请求对目标终端设备的所述目标业务进行业务体验保障，所述业务体验保障的业务质量MOS等级要求为第一业务质量MOS等级；
    所述处理模块，还用于根据所述目标终端设备的移动轨迹，确定所述目标终端设备将从第一接入网设备的服务范围移动到第二接入网设备的移动范围；
    所述收发模块，还用于向所述E-NWDAF实体发送第六订阅请求，所述第六订阅请求用于请求订阅第一终端设备使用所述目标业务的触发事件、以及所述第二接入网设备关联的第一网络性能指标；
    所述收发模块，还用于在所述触发事件被触发之后，从所述E-NWDAF实体接收所述第二接入网设备关联的第一网络性能指标；
    所述处理模块，还用于将所述第二接入网设备关联的第一网络性能指标与所述目标业务的MOS模型进行匹配，确定所述第二接入网设备关联的第一网络性能指标对应的所述目标业务的第三业务质量MOS等级；
    所述处理模块，还用于在所述第三业务质量MOS等级和所述第一业务质量MOS等级不同的情况下，将与所述第三业务质量MOS等级对应的第一网络性能指标与所 述目标业务的MOS模型中所述第一业务质量MOS等级对应的第一网络性能指标进行匹配，得到网络预测结果；
    所述收发模块，还用于向所述策略控制实体发送所述网络预测结果，所述网络预测结果用于进行网络优化。
  29. 一种边缘网络数据分析功能E-NWDAF实体，其特征在于，所述E-NWDAF实体包括：处理模块和收发模块；
    所述收发模块，用于从中心网络数据分析功能C-NWDAF实体接收第一订阅请求，所述第一订阅请求用于请求订阅目标业务的业务质量平均意见得分MOS等级和对应的第一网络性能指标，所述第一网络性能指标为承载所述目标业务的网络传输的网络性能指标；
    所述处理模块，用于获取所述业务质量MOS等级和所述第一网络性能指标；
    所述收发模块，还用于向所述C-NWDAF实体发送所述业务质量MOS等级和所述第一网络性能指标。
  30. 根据权利要求29所述的E-NWDAF实体，其特征在于，所述处理模块，用于获取所述业务质量MOS等级，包括：
    所述处理模块，用于获取所述目标业务的业务体验数据；根据所述目标业务的业务体验数据确定所述业务质量MOS等级。
  31. 根据权利要求30所述的E-NWDAF实体，其特征在于，在所述目标业务为视频业务的情况下，所述业务体验数据包括以下参数中的一个或多个：
    所述视频业务的初始缓冲时延、播放缓冲时长、码率、业务速率、帧率、流畅度、清晰度；以及，终端设备的分辨率。
  32. 根据权利要求29-31任一项所述的E-NWDAF实体，其特征在于，所述第一网络性能指标包括以下参数中的一个或多个：
    终端设备和接入网设备之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、往返时延RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述接入网设备和用户面实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率；
    以及，所述用户面实体与应用功能实体之间的路径上的上行丢包数、上行丢包率、下行丢包数、下行丢包率、RTT、上行误码包数、上行误码率、下行误码包数、下行误码率、上行乱序包数、上行乱序包率、下行乱序包数、下行乱序包率、上行重传包数、上行重传率、下行重传包数、下行重传率，上行平均包间隔、上行平均包抖动、下行平均包间隔、下行平均包抖动、上行速率以及下行速率。
  33. 一种通信装置，其特征在于，所述通信装置包括处理器和存储器；
    所述存储器用于存储计算机指令；
    所述处理器用于执行所述计算机指令，以使得所述通信装置执行如权利要求1-16任一项所述的方法。
  34. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有指令；当其在计算机上运行时，使得所述计算机执行如权利要求1-16任一项所述的方法。
  35. 一种通信系统，所述通信系统包括中心网络数据分析功能C-NWDAF实体和边缘网络数据分析功能E-NWDAF实体；
    所述C-NWDAF实体，用于向所述E-NWDAF实体发送第一订阅请求，所述第一订阅请求用于请求订阅目标业务的业务质量平均意见得分MOS等级和对应的第一网络性能指标，所述第一网络性能指标为承载所述目标业务的网络传输的网络性能指标；
    所述E-NWDAF实体，用于从所述C-NWDAF实体接收所述第一订阅请求，并获取所述业务质量MOS等级和所述第一网络性能指标之后，向所述C-NWDAF实体发送所述业务质量MOS等级和所述第一网络性能指标；
    所述C-NWDAF实体，用于从所述E-NWDAF实体接收所述业务质量MOS等级和所述第一网络性能指标，并根据所述业务质量MOS等级和所述第一网络性能指标，建立所述目标业务的MOS模型。
  36. 一种计算机程序产品，其特征在于，包括指令，当所述指令在计算机上运行时，使得所述计算机执行如权利要求1-16任一项所述的方法。
  37. 一种芯片，其特征在于，包括：处理器和接口，所述处理器通过所述接口与存储器耦合，当所述处理器执行所述存储器中的计算机程序或指令时，使得权利要求1-16任一项所述的方法被执行。
PCT/CN2020/080958 2019-07-23 2020-03-24 应用mos模型的训练方法、设备及系统 Ceased WO2021012697A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP20843289.8A EP3993320B1 (en) 2019-07-23 2020-03-24 Training method, device, and system using mos model
US17/581,935 US12113679B2 (en) 2019-07-23 2022-01-23 Training method for application MOS model, device, and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910668701.5 2019-07-23
CN201910668701.5A CN112311564B (zh) 2019-07-23 2019-07-23 应用mos模型的训练方法、设备及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/581,935 Continuation US12113679B2 (en) 2019-07-23 2022-01-23 Training method for application MOS model, device, and system

Publications (1)

Publication Number Publication Date
WO2021012697A1 true WO2021012697A1 (zh) 2021-01-28

Family

ID=74192788

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/080958 Ceased WO2021012697A1 (zh) 2019-07-23 2020-03-24 应用mos模型的训练方法、设备及系统

Country Status (4)

Country Link
US (1) US12113679B2 (zh)
EP (1) EP3993320B1 (zh)
CN (1) CN112311564B (zh)
WO (1) WO2021012697A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114745289A (zh) * 2022-04-19 2022-07-12 中国联合网络通信集团有限公司 网络性能数据的预测方法、装置、存储介质及设备
CN115052305A (zh) * 2021-03-09 2022-09-13 大唐移动通信设备有限公司 一种网络分析的转移方法、装置和可读存储介质

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11539606B2 (en) * 2020-02-10 2022-12-27 Celona, Inc. Method and apparatus for measuring packet loss rate via deep packet inspection at an intermediate node in a communication network
GB2593979B (en) 2020-03-20 2022-09-14 Samsung Electronics Co Ltd Improvements in and relating to data analytics in a telecommunication network
EP4209038A1 (en) * 2020-09-02 2023-07-12 Lenovo (Singapore) Pte. Ltd. Qos profile adaptation
WO2022174362A1 (en) * 2021-02-17 2022-08-25 Huawei Technologies Co., Ltd. Entities and methods for trained data model selection in 5g mobile networks
EP4340300A4 (en) * 2021-05-10 2025-02-26 Beijing Xiaomi Mobile Software Co., Ltd. MODEL INFERENCE METHOD, MODEL INFERENCE APPARATUS, AND STORAGE MEDIUM
CN115714990A (zh) * 2021-08-18 2023-02-24 中国移动通信有限公司研究院 无线业务传输控制方法、装置和功能实体
CN115767514B (zh) * 2021-09-03 2026-04-17 华为技术有限公司 一种通信方法、通信装置及通信系统
CN113568303B (zh) * 2021-09-26 2021-12-14 成都数默科技有限公司 一种基于pid控制算法的网络流量抓包限流丢包方法
CN114158076B (zh) * 2021-12-14 2023-07-14 中国联合网络通信集团有限公司 策略生成方法、设备及存储介质
WO2023109749A1 (en) * 2021-12-16 2023-06-22 Mediatek Singapore Pte. Ltd. Optimization of delivery of extended reality traffic over a wireless link
CN116866201A (zh) * 2022-03-28 2023-10-10 中国电信股份有限公司 人工智能训练方法和系统及相关设备
CN115243327B (zh) * 2022-07-15 2025-04-25 咪咕文化科技有限公司 一种实现网业协同的服务质量控制方法及系统
CN115942392B (zh) * 2022-11-10 2023-08-01 广州爱浦路网络技术有限公司 基于时间间隙的服务质量QoS动态配置方法及系统
TWI826194B (zh) * 2022-12-20 2023-12-11 明泰科技股份有限公司 相容於雲原生虛擬網路層的使用者層功能(upf)封包處理方法及計算裝置
EP4674162A1 (en) * 2023-02-27 2026-01-07 Telefonaktiebolaget LM Ericsson (publ) Uplink signal power in analytics
JP7803896B2 (ja) * 2023-03-22 2026-01-21 Kddi株式会社 複数のサイトそれぞれに配置された分析機能サーバの学習モデル訓練方法、分析機能サーバ及びプログラム
US20250081079A1 (en) * 2023-09-05 2025-03-06 Apple Inc. Adaptive Dynamic Link Selection

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104221411A (zh) * 2012-03-30 2014-12-17 高通股份有限公司 用于向网络以信号形式发送无线设备的能力的装置和方法
CN104782168A (zh) * 2012-11-08 2015-07-15 爱立信(中国)通信有限公司 Pcrf装置和用于pcrf的业务处理方法
US20180262924A1 (en) * 2017-03-10 2018-09-13 Huawei Technologies Co., Ltd. System and Method of Network Policy Optimization
US20190222489A1 (en) * 2018-04-09 2019-07-18 Intel Corporation NETWORK DATA ANALYTICS FUNCTION (NWDAF) INFLUENCING FIFTH GENERATION (5G) QUALITY OF SERVICE (QoS) CONFIGURATION AND ADJUSTMENT

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10484255B2 (en) * 2017-06-19 2019-11-19 Cisco Technology, Inc. Trustworthiness index computation in a network assurance system based on data source health monitoring
CN109600759B (zh) * 2017-09-30 2022-04-29 华为技术有限公司 一种策略确定的方法及通信装置
CN109561300B (zh) * 2018-12-27 2021-03-12 中国联合网络通信集团有限公司 业务质量检测方法及装置
US11038776B2 (en) * 2019-05-07 2021-06-15 Electronics And Telecommunications Research Institute Method and system for providing service experience analysis based on network data analysis
US12003384B2 (en) * 2019-07-05 2024-06-04 Nokia Technologies Oy Efficient handling of collected data or analytics data in network data analytics function scenarios
US11706110B2 (en) * 2019-07-12 2023-07-18 Verizon Patent And Licensing Inc. System and method of closed loop analytics for network automation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104221411A (zh) * 2012-03-30 2014-12-17 高通股份有限公司 用于向网络以信号形式发送无线设备的能力的装置和方法
CN104782168A (zh) * 2012-11-08 2015-07-15 爱立信(中国)通信有限公司 Pcrf装置和用于pcrf的业务处理方法
US20180262924A1 (en) * 2017-03-10 2018-09-13 Huawei Technologies Co., Ltd. System and Method of Network Policy Optimization
US20190222489A1 (en) * 2018-04-09 2019-07-18 Intel Corporation NETWORK DATA ANALYTICS FUNCTION (NWDAF) INFLUENCING FIFTH GENERATION (5G) QUALITY OF SERVICE (QoS) CONFIGURATION AND ADJUSTMENT

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3993320A4

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115052305A (zh) * 2021-03-09 2022-09-13 大唐移动通信设备有限公司 一种网络分析的转移方法、装置和可读存储介质
CN114745289A (zh) * 2022-04-19 2022-07-12 中国联合网络通信集团有限公司 网络性能数据的预测方法、装置、存储介质及设备

Also Published As

Publication number Publication date
US20220150130A1 (en) 2022-05-12
EP3993320A1 (en) 2022-05-04
CN112311564B (zh) 2022-04-22
US12113679B2 (en) 2024-10-08
EP3993320A4 (en) 2022-08-03
EP3993320B1 (en) 2025-05-07
CN112311564A (zh) 2021-02-02

Similar Documents

Publication Publication Date Title
EP3993320B1 (en) Training method, device, and system using mos model
CN110972193B (zh) 一种切片信息处理方法及装置
US11758416B2 (en) System and method of network policy optimization
JP7544782B2 (ja) サービス品質監視方法およびシステムならびにデバイス
US12231948B2 (en) Intelligent allocation of network resources
CN111225420B (zh) 一种用户接入控制方法、信息发送方法及装置
US10397758B2 (en) Function selection based on utilization level in 5G environments
WO2020057261A1 (zh) 通信方法和装置
CN110798860B (zh) 数据处理方法、设备及系统
WO2020063694A1 (zh) 一种切片信息处理方法及装置
CN111436030B (zh) 数据用量上报的方法、装置及系统
CN114930907B (zh) 一种数据传输方法、设备及系统
CN112019363B (zh) 确定业务传输需求的方法、设备及系统
WO2025131857A1 (en) Nwdaf xr analytics
WO2025139499A1 (zh) 基于非3GPP接入的QoS处理方法、装置、可读介质及设备
TWI821882B (zh) 丟包率的檢測方法、通信裝置及通信系統
US20250212112A1 (en) Service processing method and apparatus, device, and medium
WO2025107735A1 (zh) 一种语义评估方法及装置
WO2024169725A1 (zh) 一种通信方法及装置
CN120238972A (zh) 服务质量处理方法、装置、计算机可读介质及电子设备
WO2025113112A1 (zh) 通信方法、通信装置及通信系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20843289

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020843289

Country of ref document: EP

Effective date: 20220126

WWG Wipo information: grant in national office

Ref document number: 2020843289

Country of ref document: EP