Information processing apparatus, information processing method, and communication system

Abstract

To enable suitable network control based on prediction of quality of service experience (QoE).SOLUTION: Acquiring at least one of first information indicating an observation result of QoE relating to target communication and second information indicating an observation result of communication performance relating to the communication from at least one network node with respect to the target communication, and generating a learning model for predicting a change of the QoE when assuming that a specific network control parameter is applied, by using at least one of the first information and the second information.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, and a communication system. [Background technology]

[0002] There is a wireless communication system that reduces congestion in a cellular network by allocating the connection destinations of a group of terminals from a cellular base station to a nearby wireless LAN access point (see, for example, Patent Document 1). There is also a wireless communication device that transmits channel selection order information stored in a storage device as information indicating the priority according to the communication quality for each of a plurality of channels used for wireless communication to one or more partner devices (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2014-551800 A [Patent Document 2] Japanese Patent Application Publication No. 2018-201238 Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present disclosure is to provide an information processing device and an information processing method that enable suitable network control based on prediction of quality of experience (QoE). [Means for solving the problem]

[0005] One aspect of the present disclosure is an information processing device that includes a control unit that performs the following operations: acquiring, from at least one network node, at least one of first information indicating observed results of quality of service experience (QoE) related to the target communication and second information indicating observed results of communication performance related to the communication; and using at least one of the first information and the second information, generating a learning model that predicts changes in QoE when a specific network control parameter is applied.

[0006] One aspect of the present disclosure is an information processing device that includes a control unit that transmits one or more network control parameter sets related to a target communication to a network node, receives a response to the one or more network control parameter sets from the network node, and selects a network control parameter set to be used for controlling the communication from the one or more network control parameter sets based on the response.

[0007] Other aspects include a communication system, an information processing method corresponding to the above-mentioned information processing device, a program for causing a computer to execute the information processing method, or a computer-readable storage medium non-temporarily storing the program. [Effects of the Invention]

[0008] According to the present disclosure, it is possible to perform suitable network control based on prediction of service quality of experience (QoE). [Brief explanation of the drawings]

[0009] [Figure 1] Figures 1(A) and 1(B) are explanatory diagrams of a fifth generation mobile communication system (5G) network. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of an information processing device. [Figure 3] FIG. 3 is a sequence diagram illustrating a first operation example of the communication system. [Figure 4]FIG. 4 is a sequence diagram illustrating a second operation example of the communication system. [Figure 5] FIG. 5 is a sequence diagram illustrating a third operation example of the communication system. [Figure 6] FIG. 6 is a sequence diagram showing a fourth operation example of the communication system. DETAILED DESCRIPTION OF THE INVENTION

[0010] A communication system and a communication control method according to an embodiment will be described below with reference to the drawings. The configurations of the embodiments are examples, and the present disclosure is not limited to the configurations of the embodiments.

[0011] <Communication system configuration> FIG. 1(A) shows components (entities) constituting a fifth-generation mobile communication system (5G network). In FIG. 1(A), UE (User Equipment) 2 is a terminal of a user (subscriber). RAN (Radio Access Network) 3 is an access network to a 5G core network (5GC). RAN 3 is composed of a base station (gNB) 3A. The 5G network consists of 5GC and RAN 3, and UE 2, DN (Data Network) 5, and AF (Application Function) 12 are connected to the 5G network. DN 5 is a network outside 5GC, such as the Internet. AF 12 is a general-purpose network function that provides auxiliary services outside 5GC. For example, AF 12 is an application executed on an external server. Each of NFs 11a to 11k is a function realized, for example, by one or more computers (information processing devices) executing a program.

[0012] 5GC is composed of a set of components (called network nodes) with specific functions called NFs (Network Functions). Figure 1(A) shows the following NFs 11 that make up 5GC: UPF (User Plane Function) 11a AMF (Access and Mobility Management Function)11b SMF (Session Management Function)11c PCF(Policy Control Function)11d NEF(Network Exposure Function)11e NRF(Network Repository Function)11g NSSF(Network Slice Selection Function)11h AUSF(Authentication Server Function)11i UDM(Unified Data Management)11j NWDAF(Network Data Analytics Function)11k

[0013] The UPF 11a performs routing and forwarding of user packets (user plane packets sent and received by the UE 2), packet inspection, and QoS processing. The AMF 11b is a device that accommodates the UE 2 in the 5GC area. The AMF 11b accommodates the RAN 3 and performs subscriber authentication control and UE 2 location (mobility) management. The UDM 11j provides subscriber information, or acquires, registers, deletes, and changes the status of the UE 2.

[0014] The SMF 11c manages PDU (Protocol Data Unit) sessions and controls the UPF 11a to implement QoS control and policy control. The PDU session is a virtual communication path for transmitting and receiving data between the UE 2 and a DN (Data Network) 5. The DN 5 is a data network (such as the Internet) external to the 5GC.

[0015] The PCF 11d performs QoS control, policy control, and billing control under the control of the SMF 11c. QoS control involves controlling the quality of communication, such as priority packet forwarding. Policy control involves controlling communication, such as QoS based on network or subscriber information, whether packet forwarding is possible, and billing. The NEF 11e controls functions such as the AF (Application Function) 12. It mediates communication between external nodes and nodes within the control plane. For example, AF12 For example, it is an information processing device (server, terminal, etc.) that implements an application outside of 5GC.

[0016] The NRF 11g stores and manages information about NFs (e.g., AMF, SMF, UPF, etc.) within the 5GC. In response to an inquiry about an NF desired to be used, the NRF 11g can return multiple NF candidates to the inquiry source.

[0017] The NSSF11h has the function of selecting the network slice to be used by the subscriber from among the network slices generated by network slicing. A network slice is a virtual network with specifications according to the application.

[0018] The AUSF 11i is a subscriber authentication server that performs subscriber authentication under the control of the AMF 11b. The UDM 11j is a database that stores subscriber-related information, etc. The NWDAF 11k is an NF that has the function of collecting and analyzing data from each NF 11, the OAM terminal 8 (Figure 1(B)), the AF 12, etc., and provides network analysis information.

[0019] Each NF forming 5GC is composed of one or more information processing devices (general-purpose devices or appliances (dedicated devices)). Information processing devices are installed in special buildings called data centers. Data centers are also called station buildings. As shown in Figure 1(B), one or more data centers 6 are placed within the communication area of ​​5GC (Figure 1(B) shows an example of 3), and data centers 6 are connected by communication lines 7. Each data center 6 is provided with an OAM (Operations, Administration, and Maintenance) terminal 8. The OAM terminal 8 has the function of operating, managing, and maintaining the network (5GC).

[0020] In 5GC, multiple NFs of the same type may be prepared. For example, an NF 11 may be prepared for each data center 6. Also, one NF 11 may be shared between data centers 6. Also, multiple NFs 11 of the same type may be configured in one data center 6. The number of data centers 6, the number of NFs 11, and the correspondence between the NFs 11 and the data centers 6 may be set as appropriate.

[0021] <Configuration of information processing device> Fig. 2 is a diagram showing an example of the configuration of an information processing device that can operate as each of the UE 2, NFs 11a to 11k, OAM terminal 8, and AF 12. In Fig. 3, the information processing device 20 can be configured using a dedicated or general-purpose information processing device (computer) such as a personal computer (PC), a workstation (WS), or a server machine. However, the information processing device 20 may also be a collection (cloud) of one or more computers.

[0022] The information processing device 20 includes a processor 21 as a processing unit or control unit (controller), a storage device 22, a communication interface 23 (communication IF23), an input device 24, and a display 25, which are interconnected via a bus 26.

[0023] The storage device 22 includes a main storage device and an auxiliary storage device. The main storage device is used as at least one of a storage area for programs and data, a program development area, a program work area, and a buffer area for communication data. The main storage device is configured with RAM (Random Access Memory) or a combination of RAM and ROM (Read Only Memory). The auxiliary storage device is used as a storage area for data and programs. A non-volatile storage medium is used as the auxiliary storage device. Examples of non-volatile storage media include a hard disk, a solid state drive (SSD), a flash memory, and an EEPROM (Electrically Erasable Programmable Read-Only Memory). The storage device 22 may also include a drive device for a disk recording medium.

[0024] The communication IF 23 is a circuit that performs communication processing. For example, the communication IF 23 is a network interface card (NIC). The communication IF 23 may also be a wireless communication circuit that performs wireless communication (5G, wireless LAN (Wi-Fi), BLE, etc.). F23 may be a combination of a circuit that performs wired communication processing and a wireless communication circuit.

[0025] The input device 24 includes keys, buttons, a pointing device, a touch panel, etc., and is used to input information. The display 25 is, for example, a liquid crystal display, etc., and displays information and data.

[0026] The processor 21 performs various processes by executing various programs stored in the storage device 22. When the processor 21 executes the programs stored in the storage device 22, the information processing device 20 can operate as each of the UE 2, NFs 11a to 11k, OAM terminal 8, and AF 12 (external server).

[0027] The processor 21 is, for example, a Central Processing Unit (CPU). A CPU is also called a Microprocessor Unit (MPU). The processor 21 may have a single processor configuration or a multi-processor configuration. Furthermore, a single physical CPU connected via a single socket may have a multi-core configuration. The processor 21 may include an arithmetic unit with various circuit configurations, such as a Digital Signal Processor (DSP) or a Graphics Processing Unit (GPU). Furthermore, the processor 21 may have a configuration in which it cooperates with at least one of an integrated circuit (IC), other digital circuit, and analog circuit. The integrated circuit may be an LSI, an Application Specific Integrated Circuit (ASIC), or a programmable logic circuit (PLC). The PLD includes, for example, a CPLD and a Field-Programmable Gate Array (FPGA). The processor 21 is, for example, a microcontroller. This also includes what are called MCUs, SoCs (System-on-a-chips), system LSIs, chipsets, etc. The entities that execute the processes of the above-mentioned processor 21, ASICs, PLDs, MCUs, SoCs, chipsets, etc. are examples of "circuitry."

[0028] The communication system of this embodiment extends the communication quality prediction function provided by the 5GC NWDAF11k and provides a mechanism for simulating the impact of changes to control parameters by the NF11 on the quality of service experience (QoE) before the changes are applied.

[0029] When NF11 sets network control parameters, it is required to select from multiple setting candidates the option that offers the best balance between resource consumption and QoE. However, QoE is the product of the combined effects of the properties and behaviors of many network components, including AF12, and it is difficult for each NF11 to accurately predict changes in QoE caused by changes in control parameters.

[0030] Therefore, in the communication system according to this embodiment, a new interface is defined for notifying the NWDAF 11k of control parameter candidates that the NF 11, called "consumer NF (also written as NF (consumer)," can apply in the future. The NWDAF 11k predicts the conditional QoE assuming that each control parameter candidate is applied, using a prediction model (an example of a learning model) that is trained based on communication performance and QoE observation information collected from the NF 11 and the AF 12, called "producer NF (NF (producer)," respectively). The NWDAF 11k responds to the consumer NF with predicted performance values ​​for these multiple hypotheses. The consumer NF can refer to the predicted performance values, select the optimal setting candidate, and apply it to actual network control.

[0031] FIG. 3 is a sequence diagram showing a first operation example of the communication system shown in FIG. The first operation example is a process performed by a consumer NF, an NWDAF11k, a producer NF, an NEF11e, and The operation is performed by AF12 (one or more information processing devices operating as consumer NF, NWDAF11k, producer NF, NEF11e, and AF12, respectively). The number of consumer NFs and producer NFs is an appropriate number of 1 or 2 or more. The consumer NFs and producer NFs may be existing NFs or new NFs. Furthermore, the operation of NWDAF11k in the first operation example may be performed by an existing NF other than NWDAF11k or a new NF.

[0032] The operation performed in the first operation example consists of a learning phase and an inference phase. In this case, the NWDAF 11k (the information processing device 20 operating as the NWDAF 11k) performs the following operations.

[0033] 1. The NWDAF 11k collects QoE observation information and communication performance observation information. The QoE observation information can be collected from the AF 12, and the communication performance observation information can be collected from the AF 12 and / or one or more producer NFs.

[0034] The QoE observation information is information indicating the QoE obtained by the AF 12 observing the target communication. The communication performance observation information is information indicating the communication performance (transfer speed, latency, etc.) obtained by the AF 12 and / or the producer NF observing the target communication. The QoE observation information is an example of first information indicating the observation result of the QoE related to the target communication, and the communication performance observation information is an example of second information indicating the observation result of the communication performance related to the communication.

[0035] For example, a message requesting QoE observation information or communication performance observation information is transmitted from the NWDAF 11k to the target NF, and a response message corresponding to the request is transmitted from the target NF to the NWDAF 11k.

[0036] When the collection target is QoE observation information, the target NF is the AF12, and when the collection target is communication performance observation information, the target NF is at least one of the AF12 and the producer NF. Note that the message requesting communication performance observation information may be obtained from both the AF12 and the producer NF, or from only one of them.

[0037] When the NWDAF 11k communicates with the AF 12, the request message is "Naf_EventExposure_Subscribe (EventID=Service Experience Information)" and the response message is "Naf_EventExposure_Notify". When the NWDAF 11k communicates with the producer NF, the request message is "Nnf_EventExposure_Subscribe (EventID = Service Experience Information)" and the response message is "Nnf_EventExposure_Notify."

[0038] 2. The NWDAF11k uses the history of QoE observation information and communication performance observation information as learning data to train a "conditional QoE prediction model." The conditional QoE prediction model has the function of predicting changes in QoE when specific network control parameters are applied in a given communication environment.

[0039] During the inference phase, the following actions occur: 1. A consumer NF (NF (consumer)) initiates a procedure to change network control parameters when the required communication quality (QoS) for the target communication is not met or is predicted to not be met in the future. 2. The consumer NF generates one or more candidates for new network control parameters (referred to as candidate parameters). 3. The consumer NF selects one or more candidate parameters (network control parameter sets). A message requesting a parameter update simulation test (conditional QoE prediction) is sent to NWDAF11k with the parameter as an argument (see Figure 3). <1> ). 4. Upon receiving the request, the NWDAF 11k collects QoE observation information from the AF 12 and collects communication performance observation information from at least one of the AF 12 and the producer NF (NF (producer)) (see FIG. 3). <2> ). That is, QoE observation information and communication performance observation information are collected through message exchange similar to that performed in 1. of the learning phase. 5. The NWDAF 11k receives as input the QoE observation information, communication performance observation information, and candidate parameters, and uses a conditional QoE prediction model to calculate the conditional QoE expected when the candidate parameters are applied. 6. The NWDAF 11k responds with the set of candidate parameters and the conditional predicted value of QoE to the consumer NF. 7. The consumer NF selects candidate parameters that offer a good balance between resource consumption and predicted QoE and applies them to actual network control.

[0040] The communication system according to the embodiment performs the operations shown in FIG. 3, and thereby the following advantageous effects can be obtained. That is, the consumer NF can change the setting parameters (for example, How does the increase or decrease in the amount of communication bandwidth allocated to a network slice affect QoE? This allows us to estimate the network environment and select and apply optimal or suitable network control parameter settings. This makes it possible to adapt quickly to changes in communication conditions and provide a stable QoE, compared to control methods that search for optimal network control parameters by trial and error, while repeatedly changing network control parameter settings and observing the actual QoE.

[0041] Those who utilize the above-mentioned communication system include automobile manufacturers who operate communication services for connected cars (e.g., those who want to monitor the behavior of communication networks via NEF11e). Examples include operators of communication networks, etc.

[0042] In the operation example shown in FIG. 3, the following two service operations are newly defined between the consumer NF and the NWDAF 11k. The service operations are message sending and receiving. It is done by faith. (1) Nnwdaf_Simulation_Request: This message requests NWDAF11k to predict the QoE assuming that the candidate parameters specified by the arguments are applied. <1> In this case, the signal is transmitted from the consumer NF to the NWDAF11k. (2) Nnwdaf_Simulation_Request Response: QoE prediction result for the requesting NF This is a message that responds to the message in Figure 3. <4> In this case, it is transmitted from NWDAF11k to consumer NF.

[0043] Examples of candidate parameters include, but are not limited to, the following: Communication bandwidth allocated to each UE2 or network slice UE2 or UE group whose registration with the network is denied UE or UE group that loses communication with the network Quality of Service (QoS) settings

[0044] For example, in 5GC, if the communication quality required now or in the future is not met, it is conceivable to reduce the number of UEs 2 connected to the 5G network by refusing to register one or more predetermined UEs 2 with the 5G network (location registration). Alternatively, it is conceivable to reduce the load on the 5G network by throttling (restricting) the communication band of one or more predetermined UEs 2. The consumer NF may give advance notice, via the NEF 11e and the AF 12, to UEs 2 for which registration with the 5G network is to be refused or for which the communication band is to be throttled. Upon receiving the notification, the AF 12 and the NEF 11e can take measures to prevent a decrease in QoE, such as enlarging the receiving buffer for video data provided to the UE 2 or pre-caching web content that the user is likely to view in the future.

[0045] Examples of information that NWDAF11k may collect from producers NF and AF include the following: [Example of QoE observation information] "Service data related to Observed Service Experience" specified in Table 6.4.2-1 of 3GPP (registered trademark) TS 23.288, and "performance information" specified in Table 6.4.2-1a (collected from AF12).

[0046] "Service Data Regarding the Observed Service Experience" may include, for example: Application ID: For example, collected from AF. It is an identifier to identify the service and support analysis for each service type (desired service level). IP filter information: collected from the AF, for example, and used to identify the UE's service flows for the application. Application location: For example, collected from AF / NEF. The location of the application represented by a list of DNAIs. If the DNAIs used by the application are statically defined, the NEF maps the AF-Service-Identifier information to a list of DNAIs. It is possible. Service Experience: Collected, for example, from the AF. It refers to the SLA and QoE per service flow established during the flight. For example, MOS or Video MOS as specified in ITU-T P.1203.3

[11] , or other non-video or voice related QoE. It can be any of the MOSs customized for any kind of service. UE ID: Collected from AF, for example. UE associated with the Service Experience value. A list of IDs. If the AF is not trusted, a GPSI is provided. If the AF is trusted, a SUPI is provided. Service Experience Contribution Weights: Collected, for example, from the AF. A list of service experience contribution weights associated with each provided UE ID. QoE metrics: collected from the UE, e.g., via the AF. QoE metrics observed at the UE, e.g., in accordance with TS 26.114

[27] , TS 26.247

[28] , TS 26.118

[29] , TS 26.346

[30] , TS 26.512

[31] , or ASP-specific QoE metrics in TS 26.512

[31] as agreed in the SLA with the MNO. The listed QoE metrics and measurements may be used. Timestamp: collected e.g. from the AF. Timestamp associated with the service experience provided by the AF. Required if the service experience is provided by an ASP. Application Server Instance: collected from, for example, AF. The IP address or FQDN of the application server with which E had a communication session.

[0047] Performance information may include, for example: UE identifier: For example, the IP address of the UE to be measured. UE location: For example, the location of the UE when the performance measurement is made. · Application ID: For example, service and support analysis (desired level of service) by service type. IP filter information: e.g., the UE service flow for an application is used for identification. Locations of Application: For example, applications indicated by a list of DNAIs Indicates the position of the button. Application Server Instance address: For example, the address of the application server instance to which the UE is connected when the measurement is made. The IP address / FQDN of the application server with the Performance Data: For example, performance related to the communication session between the UE and the application server: average packet delay, average loss rate, and throughput. . Timestamp: For example, the timestamp associated with performance data provided by AF. It's a tamp.

[0048] [Example of communication performance observation information] Other data includes the history of communication traffic generated by UE2, future communication demand, and QoE indicators (collected from AF12). The "QoS flow level network" defined in Table 6.4.2-2 of 3GPP (registered trademark) TS 23.288 "QoS flow-level network data" may include: Timestamp: Collected from the 5GC NF. A timestamp associated with the collected information. Location information: For example, collected from AMF. UE location information (such as cell ID or TAI). Finer granularity location (1~max): e.g., collected from GMLC Indicates the location of the UE. ·>UE location: GAD shape or location coordinates (see TS 23.032

[34] ). ·>Timestamp: The timestamp when the position was measured. ·>LCS: Accuracy of QoS measurements. UE ID: Collected e.g. from AMF. List of SUPIs. If no UE ID is provided as target for Slice Service Experience Analytics report, AMF will return UE IDs matching the AMF event filter. DNN: DNN of the PDU session including the QoS flow, e.g. collected from the SMF. S-NSSAI: Collected from e.g. SMF. S-NSSAI of the PDU session containing the QoS flow. · Application ID: Used by the NWDAF collected from the SMF to identify the application service provider and application of the QoS flow. DNAI: Collected, for example, from SMF. Identifies the access DN to which the PDU session connects. PDU Session type: Type of PDU session, e.g. collected from SMF. SSC Mode: SSC mode selected for the PDU session, e.g. collected from SMF. Do. Access Types: A list of access types used in the PDU session, e.g. collected from SMF. IP filter information: collected, for example, from the SMF. Used by the NWDAF to identify service data flows for QoS flow policy control and / or differentiated charging, and provided by the SMF. QFI: QoS flow identifier, e.g. collected from SMF. QoS Flow Bit Rate: Collected, for example, from the UPF. The observed bit rate in the uplink (UL) direction and the observed bit rate in the downlink (DL) direction. QoS Flow Packet Delay: Collected e.g. from UPF. Observed packet delay in UL direction and observed packet delay for DL ​​direction. Packets sent: Collected from UPF, for example. Indicates the number of packets sent. Packet retransmissions: Collected from, for example, UPF or AF. Indicates the number of observed packet retransmissions.

[0049] "UE level network data" (collection source is NF) specified in Tables 6.4.2-3 and 6.4.2-4 of 3GPP (registered trademark) TS 23.288. "UE level network data" may include the following. The description of the collection source is an example. Timestamp: A timestamp associated with the collected information, e.g. collected from the OAM. Reference Signal Received Power: e.g. collected from OAM. Per-UE measurement of received power level in the network cell. Includes SS-RSRP, CSI-RSRP as specified in clause 5.5 of TS 38.331

[14] , and E-UTRA RSRP as specified in clause 5.5.5 of TS 36.331

[15] . Reference signal reception quality: collected, for example, from OAM. Includes SS-RSRQ, CSI-RSRQ as specified in TS 38.331

[14] clause 5.5, and E-UTRA RSRQ as specified in TS 36.331

[15] clause 5.5.5. A per-UE measurement of reception quality in a network cell. Signal-to-noise and interference ratio: e.g. collected from OAM. Section 5.1 of TS 38.215

[12] Per-UE measurements of received signal-to-noise-and-interference ratio in network cells, including SS-SINR, CSI-SINR, and E-UTRA RS-SINR, as specified. DL and UL RAN throughput: collected e.g. from OAM. DL and UL throughput measurements per UE as specified in TS 37.320

[20] clauses 5.2.1.1 and 5.4.1.1. DL and UL RAN packet delay: Collected, for example, from OAM. DL and UL packet delay, including packet delay per QCI per UE as specified in clause 5.2.1.1 of TS 37.320

[20] and packet delay per DRB per UE as specified in clause 5.4.1.1 of TS 37.320

[20] . Measured per UE packet delay. DL and UL RAN packet loss rates: Collected, for example, from OAM. DL and UL packet loss rates include the packet loss rate per QCI per UE as specified in clause 5.2.1.1 of TS 37.320

[20] and the packet loss rate per DRB per UE as specified in clause 5.4.1.1 of TS 37.320

[20] . Measured packet loss rate per UE for both UL and UL. Cell ID and frequency mapping information: For example, collected from OAM. Cell ID and frequency mapping information. Cell energy saving status: collected for example from OAM. List of cells in the target area that are in energy saving status as specified in clauses 3.1 and 6.2 of TS 28.310

[24] .

[0050] In addition, "UE-level network data" may include the following: The description of the collection source is an example. Timestamp: Collected from 5GC NF. The timestamp associated with the collected information. It's a tamp. Location information: For example, collected from AMF. UE location information (such as cell ID or TAI). Finer granularity location: collected from GMLC. Indicates the location. UE ID: Collected, for example, from AMF. A list of SUPIs. RAT Type: Collected e.g. from SMF. RAT type the UE is camped on.

[0051] Examples of predicted QoE information that the NWDAF 11k returns to the consumer NF include the following: - Predicted QoE value specified in Table 6.4.3-2 of 3GPP (registered trademark) TS 23.288 (responds with a "conditional" predicted value assuming the candidate parameters are applied) The predicted QoE value may be expressed as a number between 0 and a predetermined maximum value max, or other expressions (e.g., a probability distribution of the predicted value may be used to express the uncertainty of the predicted result). etc.) may also be adopted.

[0052] Predictions of QoE may include: Slice instance service experience (0~max): Each network slice instance A list of observed service experience information for the tansu. ·> S-NSSAI: Identifying network slices. NSI ID: Identifies a network slice instance within a network slice. . Network Slice Instance Service Experience: The service experience (average, variance) of the entire application of the network slice instance during the analysis period. ·>SUPI list (0~SUPImax): A list of SUPIs to which the service experience of the slice instance applies. Ratio: Estimated proportion of UEs with similar service experience (within a group or across all) between the UE). ·> Spatial effectiveness: the area where the service experience analysis of network slices is applied. Validity period: The validity period of the network slice service experience analysis defined in Section 6.1.3. Between. Confidence: How confident is this prediction? Application Service Experience (0~Max): Estimated service experience for each application A list of information. S-NSSAI: Identifies the network slice used to access the application do. ·> Application ID: The identifier of the application. ·> Service Experience Type: Type of service experience analysis such as voice, video, other. UE location: UE location information (TAI link) when the UE service is delivered The information includes the location of the gNB, its name, address, gNB ID, location coordinates, etc. ·> UPF information: Indicates the UPF serving the UE. ·> DNAI: Indicates which DNAI the UE service uses / camps on. ·> DNN: DNN of PDU sessions including QoS flows. Application Server Instance Address: Application Server Instance Indicates the instance (IP address of the application server) or the FQDN of the application server. Service Experience: Service experience (average, variance) for the analytical period. ·> SUPI list (0~SUPImax): A list of SUPIs with the same application service experience. ·> Proportion: Estimated proportion of UEs (within a group or across all UEs) that have similar service experience. ·> Spatial validity: the area to which the analysis of the application usage experience applies. ·> Validity Period: The validity period of the application service experience analysis as defined in Section 6.1.3. ·>Confidence: How confident are you in this prediction? ·> RAT type: Indicates the list of RAT types to which the application service experience analysis applies. Frequency: Indicates a list of carrier frequency values ​​of the serving cell of the UE to which the application service experience analysis is applied. ·> SSC Mode: The SSC mode selected for the PDU session used to associate with the application. ·> PDU Session Type: The type of PDU session used to associate with the application. Access Type: The access type used for the application's PDU session. A list of pools.

[0053] FIG. 4 is a sequence diagram showing a second operation example of the communication system, in which the PCF 11d triggers the process as a consumer NF. In FIG. 4, the PCF 11d sends a message to the NWDAF 11k requesting a parameter update simulation test (conditional QoE prediction) with one or more candidate parameters as arguments (see FIG. 4). <1> ).

[0054] Figure 4 <2> ~ <4> This is the same operation as the first operation example. However, in FIG. <4> The response message from the NWDAF 11k is sent to the PCF 11d. <5> In this case, the PCF 11d selects candidate parameters that provide the best balance between the conditional QoE and resource consumption. <6> Then, the PCF 11d applies the selected candidate parameters to the producer NF.

[0055] FIG. 5 is a sequence diagram showing a third operation example of the communication system. The third operation example is a case where the NWDAF 11k triggers the processing. <0> In Fig. 5, the PCF 11 sets the trigger conditions for conditional QoE prediction. <1> In this case, when a predetermined trigger condition is met (for example, when a certain time has passed since the last QoE prediction), the QoE or When the conditional QoE prediction is initiated, the third In the example operation, the candidate parameters are generated by the NWDAF 11k. <2> from <6> The operation is the same as the second operation example (FIG. 4), so a description thereof will be omitted.

[0056] FIG. 6 is a sequence diagram showing a fourth operation example of the communication system. In the fourth operation example, the AF 12 triggers the processing. <0> In this case, the AF 12 detects the degradation of QoE and notifies the PCF 11d of the degradation of QoE via the NEF 11e. <1> ~ <6> The operation is the same as the second operation example, so the explanation will be omitted.

[0057] In the first to fourth operation examples described above, the information processing device 20 operating as NWDAF11k is an example of a first information processing device, and the consumer NF in the first operation example and the PCF11d in the second to fourth operation examples are examples of a second information processing device.

[0058] The above-described embodiment and modifications are merely examples, and the present disclosure may be modified as appropriate within the scope of the present disclosure. Furthermore, the processes and means described in the present disclosure may be freely combined and implemented as long as no technical contradiction occurs.

[0059] Furthermore, a process described as being performed by one device may be shared and executed by multiple devices. Alternatively, a process described as being performed by different devices may be executed by one device. In a computer system, the hardware configuration (server configuration) by which each function is realized can be flexibly changed.

[0060] The present disclosure can also be realized by providing a computer program implementing the functions described in the above embodiments to a computer, and having one or more processors in the computer read and execute the program. Such a computer program may be provided to the computer via a non-transitory computer-readable storage medium connectable to the computer's system bus or via a network. Non-transitory computer-readable storage media include any type of medium suitable for storing electronic instructions, such as any type of disk, including magnetic disks (e.g., floppy disks, hard disk drives (HDDs), etc.), optical disks (e.g., CD-ROMs, DVDs, Blu-ray disks), read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic cards, flash memory, or optical cards. [Explanation of symbols]

[0061] 11···NF, 11d···PCF, 11e···NEF, 11k···NWDAF, 20···Information processing device

Claims

1. acquiring, for a target communication, from at least one network node, at least one of first information indicative of observed results of a quality of experience (QoE) for the target communication and second information indicative of observed results of communication performance for the communication; generating a learning model using at least one of the first information and the second information to predict a change in QoE assuming that a particular network control parameter is applied; and An information processing device including a control unit that executes the above.

2. The information processing device acquiring, for a target communication, from at least one network node, at least one of first information indicative of observed results of a quality of experience (QoE) for the target communication and second information indicative of observed results of communication performance for the communication; generating a learning model using at least one of the first information and the second information to predict a change in QoE assuming that a particular network control parameter is applied; and An information processing method that performs the above.

3. transmitting to a network node one or more sets of network control parameters relating to the communication of interest; receiving a response to the one or more network control parameter sets from the network node; selecting, based on the response, from the one or more network control parameter sets, a network control parameter set to use for controlling the communication; An information processing device including a control unit that executes the above.

4. the response includes a predicted QoE for the communication calculated using the one or more network control parameter sets; The control unit performs the selection using the predicted value of QoE. The information processing device according to claim 3 .

5. The control unit makes the selection taking into consideration a balance between resource consumption related to the communication and a predicted value of QoE. The information processing device according to claim 4 .

6. The information processing device transmitting to a network node one or more sets of network control parameters relating to the communication of interest; receiving a response to the one or more network control parameter sets from the network node; selecting, based on the response, from the one or more network control parameter sets, a network control parameter set to use for controlling the communication; An information processing method that performs the above.

7. the response includes a predicted QoE for the communication calculated using the one or more network control parameter sets; The information processing device makes the selection using the predicted value of QoE. The information processing method according to claim 6.

8. The information processing device makes the selection taking into consideration a balance between resource consumption related to the communication and a predicted value of QoE. The information processing method according to claim 7.

9. A communication system comprising: a first information processing device that is the information processing device according to claim 1; and a second information processing device that is the information processing device according to claim 3, wherein the network node according to claim 3 is the first information processing device.

Images