Artificial intelligence inferred quality of experience, quality of service, and minimization of drive test data

Abstract

Systems, methods, and apparatus for wireless communication are described. A method of wireless communication includes transmitting, by a network node, a request for input data. The method also includes receiving, by the network node and in response to the request, input data. The described techniques may be employed by a network device or by a wireless device.

Description

Technical Field

[0001] This patent document is generally directed to wireless communications. Background Art

[0002] Mobile communication technology is driving the world towards an increasingly connected and networked society. Compared with existing wireless networks, the next generation of systems and wireless communication technologies will need to support a wider range of use case characteristics and provide a more complex and sophisticated range of access requirements and flexibility.

[0003] Long Term Evolution (LTE) is a wireless communication standard for mobile devices and data terminals developed by the Third Generation Partnership Project (3GPP). LTE Evolution (LTE-A) is a wireless communication standard that is an enhanced version of the LTE standard. The fifth generation wireless system (i.e. 5G) advances the LTE and LTE-A wireless standards and is committed to supporting higher data rates, a large number of connections, ultra-low latency, high reliability, and other emerging business requirements. Summary of the invention

[0004] Disclosed are techniques for training artificial intelligence (AI) functions used in wireless communications. This patent document describes in detail data requests and input data for training AI functions. Further disclosed are techniques for inferring data using AI functions. This patent document describes in detail the inferred data. This patent document also describes in detail feedback requests and feedback responses.

[0005] A first example wireless communication method includes transmitting, by a network node, a request for input data. The method also includes receiving, by the network node and in response to the request, the input data.

[0006] A second example wireless communication method includes providing, by a network node, training of an artificial intelligence (AI) function using input data. The method also includes transmitting, by the network node and based on the training, an updated AI function.

[0007] A third example wireless communication method includes determining, by a network node and based on input data, using an artificial intelligence (AI) function, inferred data.

[0008] A fourth example wireless communication method includes transmitting, by a network node, inferred data. The method also includes transmitting, by the network node, a request for feedback associated with the inferred data. The method also includes receiving, by the network node and in response to the request, a feedback response.

[0009] A fifth example wireless communication method includes receiving, by a network node, a request for input data. The method also includes transmitting, by the network node and in response to the request, the input data.

[0010] In yet another exemplary embodiment, a device configured or operable to perform the above method is disclosed. The device may include a processor configured to implement the above method.

[0011] In another exemplary embodiment, the above method is embodied in the form of processor executable code and stored in a non-transitory computer readable storage medium. When executed by a processor, the code included in the computer readable storage medium causes the processor to implement the method described in this patent document.

[0012] The above and other aspects and embodiments thereof will be described in more detail in the drawings, the description and the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 An artificial intelligence (AI) framework is shown.

[0014] Figure 2 AI training performed by the core network (CN) nodes and AI inference performed by the gNodeB (gNB) are shown.

[0015] Figure 3 Shows AI training and inference performed by gNB.

[0016] Figure 4 AI training performed by a central unit (CU) and AI inference performed by distributed units (DU) are shown.

[0017] Figure 5 AI training and inference performed by the CU are shown.

[0018] Figure 6 The feedback procedure is shown.

[0019] Figure 7 is an exemplary flow chart for requesting input data.

[0020] Figure 8 is an exemplary flow chart for training AI functions.

[0021] Fig. 9 is an exemplary flow chart for determining inferred data.

[0022] Fig.10 is an exemplary flow chart for requesting feedback.

[0023] Fig.11 is an exemplary flow chart for transmitting input data.

[0024] Fig.12 An exemplary block diagram of a hardware platform that may be part of a network device or a communication device is shown.

[0025] Fig.13 An exemplary wireless communication including a base station (BS) and a user equipment (UE) based on some embodiments of the disclosed technology is shown. DETAILED DESCRIPTION

[0026] The example titles of the following sections are used to facilitate understanding of the disclosed subject matter and do not limit the scope of the claimed subject matter in any way. Therefore, one or more features of one example section can be combined with one or more features of another example section. In addition, for clarity of explanation, 5G terminology is used, but the technology disclosed in this document is not limited to 5G technology and can be used in wireless systems implementing other protocols.

[0027] I. Introduction

[0028] By using artificial intelligence (AI) / machine learning (ML) techniques, a radio access network (RAN) node / core network (CN) / distributed unit (DU) or other entity can predict quality of experience (QoE) measurement data, quality of service (QoS) configuration and / or minimization of drive test (MDT) measurement data by using historical QoE, QoS and MDT data and / or inference data provided from other entities. By using the predicted data, the network can allocate / reallocate more appropriate third generation partnership project (3GPP) / non-3GPP resources and modify the mobility policy of a specific user equipment (UE). The newly generated / measured data will then be used as feedback data to evaluate the modification. Based on the evaluation, further received historical data (e.g., QoE measurement data, QoS configuration, MDT measurement data, etc.) and AI / ML techniques, the network entity will update the AI ​​model and perform AI inference.

[0029] AI / ML technology has been widely used in various fields, especially in the industrial field. With the assistance of AI / ML, production efficiency has been greatly improved compared with before. In the field of telecommunications, AI / ML can also be used to improve the system. For example, AI technology can be applied to beam management in air communication interfaces. In some embodiments, beam management typically relies on an exhaustive search for beam sweeps. For another example, AI technology can be applied to channel state information (CSI) feedback. These AI / ML models used in the system can be trained and managed on each network node and may need to be transported or transmitted to other network nodes. Figure 1 The functional framework of RAN intelligence defined in TR 37.817 is shown in FIG.

[0030] II. Embodiment 1

[0031] like Figure 2As shown, Example 1 shows AI training performed by operations, administration and maintenance (OAM) / CN nodes and AI inference performed by gNodeB (gNB) nodes.

[0032] In this embodiment, CN or OAM is responsible for AI model training, and the next generation (NG) RAN (NG-RAN) node is responsible for model inference. If CN is involved in this embodiment, the procedure between CN and NG-RAN node is NG Application Protocol (NGAP) procedure, which will be introduced in 3GPP protocol. If OAM is involved in this embodiment, the procedure between OAM and NG-RAN node is based on the implementation method, and only the stage 2 description can be introduced in 3GPP protocol.

[0033] Before this procedure takes place, all involved network entities may already be deployed with AI models, so that NG-RAN Node 1 and NG-RAN Node 2 can provide inferred data when they receive a data request in step 3 or 8.

[0034] Step 1: The UE has configured QoE measurement and is performing QoE data collection.

[0035] Step 2: The UE periodically transmits measurement reports to NG-RAN node 1.

[0036] Step 3: OAM / CN sends a message to NG-RAN node 1 and requests data for AI functions (e.g., model training). This message may include at least one of the following information.

[0037] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 1 only needs to provide data for one or more specific UEs.

[0038] Processing indicator: This indication is used to inform the NG-RAN node 1 that the required information can be used for training and / or inference. The NG-RAN node 1 can provide different amounts of data based on this indication.

[0039] Historical QoS configuration request message: This message is used to request historical QoS configuration information stored in / available in the NG-RAN node 1.

[0040] This information may contain an indication of historical QoS configuration requirements. In this case, the NG-RAN node 1 may provide all historical QoS configurations that meet the requirements in this message.

[0041] This information may contain detailed QoS configuration parameters. In this case, the NG-RAN node 1 may provide all historical QoS configurations with specific parameters that meet the requirements in this message. At least one of the following parameters may be included in this message.

[0042] Historical QoS Configuration Requirement Indication, 5G QoS Identifier (5QI) Reference, QoS Class Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, and Average Guaranteed Flow Bit Rate.

[0043] Historical QoE (RVQoE) Required Message: This message is used to request the historical QoE information available at the NG-RAN node 1.

[0044] This information may contain an indication of historical QoE requirements. In this case, the NG-RAN node 1 may provide all historical QoE data that meets the requirements in this message.

[0045] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 1 may provide all historical QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0046] Historical QoE requirement indication, QoE level identifier, service type, area scope, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric (wherein the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playout delay of video, video resolution, buffer occupancy level, etc.

[0047] Historical MDT data request message: This message is used to request the historical MDT information available at the NG-RAN node 1.

[0048] This information may contain an indication for historical MDT data requirements. In this case, the NG-RAN node 1 may provide all historical MDT data that meets the requirements in this message.

[0049] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 1 may provide all historical MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0050] M1: Downlink (DL) signal volume measurement results of the serving cell and intra-frequency / inter-frequency / inter-radio access technology (RAT) neighboring cells, including cell / beam level measurements of new radio (NR) cells only.

[0051] M2: Power headroom measurement performed by the UE.

[0052] M3: Received interference power measurement.

[0053] M4: Data volume measurement for DL ​​and uplink (UL) separately (per data radio bearer (DRB) per UE).

[0054] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0055] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0056] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0057] M8: Received Signal Strength Indicator (RSSI) measurement by UE (for Wireless Local Area Network (WLAN) / Bluetooth measurements).

[0058] M9: Round Trip Time (RTT) measurement by UE (for WLAN measurements).

[0059] Inferred QoS Data Request: This message is used to request the inferred QoS information available at the NG-RAN Node 1.

[0060] This information may contain an indication of the requirements for inferred QoS configurations. In this case, the NG-RAN node 1 may provide all inferred QoS configurations that meet the requirements in this message.

[0061] This information may contain detailed QoS configuration parameters. In this case, the NG-RAN node 1 may provide all inferred QoS configurations regarding specific parameters that meet the requirements in this message. At least one of the following parameters may be included in this message.

[0062] Inferred QoS Configuration Required Indication: This is used to indicate that the NG-RAN node 1 needs to provide inferred QoS data.

[0063] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0064] Inferred QoE Data Request: This information is used to request the inferred QoE information available at the NG-RAN Node 1.

[0065] This information may contain an indication for inferred QoE requirements. In this case, the NG-RAN node 1 may provide all inferred QoE data that meets the requirements in this message.

[0066] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 1 may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0067] Inferred QoE Required Indication: This is used to indicate that the NG-RAN node 1 needs to provide inferred QoE data.

[0068] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0069] Inferred MDT Data Request: This information is used to request the inferred MDT information available at the NG-RAN Node 1.

[0070] This message may contain an indication of the requirements for inferred MDT information. In this case, the NG-RAN node 1 may provide all inferred MDT data that meets the requirements in this message.

[0071] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 1 may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0072] Inferred MDT Required Indication: This is used to indicate that the NG-RAN node 1 needs to provide inferred MDT data.

[0073] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0074] M2: Power headroom measurement performed by the UE.

[0075] M3: Received interference power measurement.

[0076] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0077] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0078] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0079] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0080] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0081] M9: RTT measurement performed by UE (for WLAN measurement).

[0082] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The NG-RAN node 1 can only provide the inferred data if the confidence of the inferred data is not less than this threshold. Each inferred data type (QoE, QoS, MDT) may have an associated confidence requirement. In another alternative, one confidence information may be used to qualify all inferred data in this message.

[0083] Time related information.

[0084] Time information of the required data: The CN / OAM can request the NG-RAN node 1 to provide a data block for one or more specific UEs within a certain duration (or start and end time).

[0085] Reporting frequency: The NG-RAN node 1 may periodically transmit data required by one or more specific UEs to the CN / OAM.

[0086] Time-related information can coexist in the message. If so, the NG-RAN node shall send the data block that meets the duration limit in the "Time information of required data" in the first response message. The NG-RAN node 1 can then send the remaining stored data or newly generated data to the OAM / CN by using the configured "Reporting frequency".

[0087] Step 4: Based on the requirements received in step 3, NG-RAN node 1 may transmit the data required for the AI ​​function to OAM / CN. The message may contain at least one of the following information.

[0088] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 1 only needs to provide historical data for one or more specific UEs.

[0089] Null Indicator: This optional indicator shall be sent to the CN / OAM if no information is available on the NG-RAN Node 1 side.

[0090] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 3.

[0091] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 3.

[0092] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 1. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0093] Time information of data: Time information of the provided data. It can be the start and end time, or duration.

[0094] Step 4a: If the data is too large or needs to be transmitted regularly, multiple messages will be used for data transmission. The message may contain at least one of the following information.

[0095] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 1 only needs to provide historical data for one or more specific UEs.

[0096] Null Indicator: This optional indicator shall be sent to the CN / OAM if no information is available on the NG-RAN Node 1 side.

[0097] If available, historical QoE data and / or historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 3.

[0098] If available, the inferred QoE data and / or inferred QoS data as well as inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 3.

[0099] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 1. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0100] Time information of data: Time information of the provided data. It can be the start and end time, or duration.

[0101] Step 5: After collecting enough data, CN / OAM can perform the AI ​​model training process with AI / ML capabilities enabled.

[0102] This step may be reflected in the 3GPP specification as a stage 2 description.

[0103] Step 6: After the AI ​​model is generated by CN / OAM, the new model will be deployed / updated to NG-RAN Node 1.

[0104] Step 7: The UE continues to send measurement data (if any) to NG-RAN node 1.

[0105] Step 8: NG-RAN node 1 sends a message to NG-RAN node 2 and requests data for AI functions (e.g., model inference). This message may include at least one of the following information.

[0106] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 2 only needs to provide data for one or more specific UEs.

[0107] Processing indicator: This indication is used to inform the NG-RAN node 2 that the required information can be used for online training and / or inference. The NG-RAN node 2 can provide different amounts of data based on this indication.

[0108] Historical QoS configuration request information: This message is used to request historical QoS configuration information stored in / available in the NG-RAN node 2.

[0109] This information may contain an indication of historical QoS configuration requirements. In this case, the NG-RAN node 2 may provide all historical QoS configurations that meet the requirements in this message.

[0110] This information may include detailed QoS configuration parameters. In this case, the NG-RAN node 2 may provide all historical QoS configurations that meet the specific parameters required in this message. The message may include at least one of the following parameters.

[0111] Historical QoS Configuration Requirement Indication, 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0112] Historical QoE (RVQoE) Required Information: This information is used to request the historical QoE information available at the NG-RAN node 2.

[0113] This information may contain an indication of historical QoE requirements. In this case, the NG-RAN node 2 may provide all historical QoE data that meets the requirements in this message.

[0114] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 2 may provide all historical QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0115] Historical QoE requirement indication, QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric. The QoE score can be a numeric range, such as 0 to 10. Where 10 represents good quality and 0 represents poor quality, or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); The QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of the video, video resolution, buffer occupancy level, etc.

[0116] Historical MDT Information Request message: This message is used to request the historical MDT information available at the NG-RAN node 2.

[0117] This message may contain an indication for historical MDT information requirements. In this case, the NG-RAN node 2 may provide all historical MDT data that meets the requirements in this message.

[0118] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 2 may provide all historical MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0119] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0120] M2: Power headroom measurement performed by the UE.

[0121] M3: Received interference power measurement.

[0122] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0123] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0124] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0125] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0126] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0127] M9: RTT measurement performed by UE (for WLAN measurement).

[0128] Inferred QoS Data Request: This information is used to request the inferred QoS information available at the NG-RAN Node 2.

[0129] This information may contain an indication of the requirements for inferred QoS configurations. In this case, the NG-RAN node 2 may provide all inferred QoS configurations that meet the requirements in this message.

[0130] This information may contain detailed QoS configuration parameters. In this case, the NG-RAN node 2 may provide all inferred QoS configurations regarding specific parameters that meet the requirements in this message. At least one of the following parameters may be included in this message.

[0131] Inferred QoS Configuration Required Indication: This is used to indicate that the NG-RAN Node 2 needs to provide inferred QoS data.

[0132] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0133] Inferred QoE Data Request: This information is used to request the inferred QoE information available at the NG-RAN Node 2.

[0134] This information may contain an indication for inferred QoE requirements. In this case, the NG-RAN node 2 may provide all inferred QoE data that meets the requirements in this message.

[0135] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 2 may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0136] Inferred QoE Required Indication: This is used to indicate that the NG-RAN node 2 needs to provide inferred QoE data.

[0137] The start and end time of this inferred data or the duration of this inferred data.

[0138] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0139] Inferred MDT Data Request: This information is used to request the inferred MDT information available on the NG-RAN node 2.

[0140] This message may contain an indication of the requirements for inferred MDT information. In this case, the NG-RAN node 2 may provide all inferred MDT data that meets the requirements in this message.

[0141] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 2 may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0142] Inferred MDT Required Indication: This is used to indicate that the NG-RAN node 2 needs to provide inferred MDT data.

[0143] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0144] M2: Power headroom measurement performed by the UE.

[0145] M3: Received interference power measurement.

[0146] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0147] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0148] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0149] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0150] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0151] M9: RTT measurement performed by UE (for WLAN measurement).

[0152] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The NG-RAN node 2 can only provide the inferred data if the confidence of the inferred data is not lower than this threshold. Each inferred data type (QoE, QoS, MDT) can have an associated confidence requirement. In another alternative, one confidence information can be used to qualify all inferred data in this message.

[0153] Time related information.

[0154] Time information of the required data: NG-RAN Node 1 can request NG-RAN Node 2 to provide data blocks for one or more specific UEs within a certain duration (or start and end time).

[0155] Reporting frequency: NG-RAN node 2 may periodically transmit data required by one or more specific UEs to NG-RAN node 1.

[0156] Time-related information can coexist in the message. If so, the NG-RAN node 2 shall send the data block that meets the duration limit in the "Time information of required data" in the first response message. The NG-RAN node 2 can then send the remaining stored data or newly generated data to the OAM / CN by using the configured "Reporting frequency".

[0157] Feedback request indication: This indication is used by the NG-RAN node 1 to request feedback on the inferences. With this indication, after the NG-RAN node 2 receives the inference data from the NG-RAN node 1, it shall send feedback information to the NG-RAN node 1 (if available).

[0158] Step 9: Based on the requirements received in step 8, NG-RAN node 2 may transmit the data required for the AI ​​function to NG-RAN node 1. The message may contain at least one of the following information.

[0159] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 2 only needs to provide historical data for one or more specific UEs.

[0160] Null Indicator: This optional indicator shall be sent to the NG-RAN node 1 if no information is available on the NG-RAN node 2 side.

[0161] If available, historical QoE data, historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0162] Inferred QoE data, inferred QoS data and inferred MDT data will be added to the message if available. Detailed parameters / metrics of this data can be found in step 8.

[0163] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 2. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0164] Time information of historical data: time information of the provided data.

[0165] Step 9a: If the historical data is too large or needs to be transmitted regularly, more than one message will be used for historical data transmission. The message may contain at least one of the following information.

[0166] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 2 only needs to provide historical data for one or more specific UEs.

[0167] Null Indicator: This optional indicator shall be sent to the NG-RAN node 1 if no information is available on the NG-RAN node 2 side.

[0168] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0169] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 8.

[0170] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 2. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0171] Time information of historical data: time information of the provided data.

[0172] Step 10: After collecting sufficient data, the NG-RAN node 1 can perform the AI ​​model inference process with the AI / ML function enabled. The NG-RAN node 1 will generate predicted QoE data and / or predicted QoS configuration and / or predicted MDT configuration / data for one or more specific UEs for the next one or more subsequent timings.

[0173] This step may be reflected in the 3GPP specification as a stage 2 description.

[0174] Step 11: After the model inference, the NG-RAN node 1 may trigger the inference action procedure. With the assistance of the inferred / predicted QoE, QoS data, the NG-RAN node 1 may perform some procedures including but not limited to the following procedures.

[0175] NG-RAN node 1 can reallocate 3GPP / non-3GPP resources for each UE in the NG-RAN node.

[0176] The NG-RAN node 1 may update the mobility policy of one or more specific UEs.

[0177] The NG-RAN node 1 may send the inferred / predicted data to other NG-RAN nodes for use in the inference procedure.

[0178] The NG-RAN node 1 may send the inferred / predicted data to the CN / OAM for further QoS / QoE configuration etc.

[0179] This step may be reflected in the 3GPP specification as a stage 2 description.

[0180] Step 12: A detailed description of the feedback procedure can be found in Example 5.

[0181] III. Embodiment 2

[0182] like Figure 3 As shown, Example 2 illustrates AI training and inference of gNB.

[0183] The procedures involved between NG-RAN nodes are Xn Application Protocol (XnAP) enhancements.

[0184] Before this procedure takes place, all involved network entities may already be deployed with AI models, so that NG-RAN Node 1 and NG-RAN Node 2 can provide inferred data when they receive the data request in step 3 / 8.

[0185] Step 1: The UE has configured QoE measurement and is performing QoE data collection.

[0186] Step 2: The UE periodically transmits measurement reports to NG-RAN node 1.

[0187] Step 3: NG-RAN node 1 sends a message to NG-RAN node 2 and requests data for AI functions (e.g., model training). This message may include at least one of the following information.

[0188] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 2 only needs to provide historical data for one or more specific UEs.

[0189] Processing indicator: This indication is used to inform the NG-RAN node 2 that the required information can be used for online training and / or inference. The NG-RAN node 2 can provide different amounts of data based on this indication.

[0190] Historical QoS configuration request information: This message is used to request historical QoS configuration information stored in / available in the NG-RAN node 2.

[0191] The information may contain an indication of historical QoS configuration requirements. In this case, the NG-RAN node 2 may provide all historical QoS configurations that meet the requirements of one or more specific UEs in the message.

[0192] The information may include detailed QoS configuration parameters. In this case, the NG-RAN node 2 may provide all historical QoS configurations for specific parameters that meet the requirements of one or more specific UEs in the message. The message may include at least one of the following parameters.

[0193] Historical QoS Configuration Requirement Indication, 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0194] Historical QoE (RVQoE) Required Information: This information is used to request the historical QoE information available at the NG-RAN node 2.

[0195] This information may contain an indication of historical QoE requirements. In this case, the NG-RAN node 2 may provide all historical QoE data that meets the requirements of one or more specific UEs in the message.

[0196] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 2 may provide all historical QoE data about specific parameters / metrics that meet the requirements of one or more specific UEs in the message. At least one of the following parameters / metrics may be included in the message.

[0197] Historical QoE requirement indication, QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0198] Historical MDT Information Request message: This message is used to request the historical MDT information available at the NG-RAN node 2.

[0199] The message may contain an indication of a requirement for historical MDT information. In this case, the NG-RAN node 2 may provide all historical MDT data that meets the requirements of one or more specific UEs in the message.

[0200] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 1 may provide all historical MDT data about specific parameters / metrics that meet the requirements of one or more specific UEs in the message. At least one of the following parameters / metrics may be included in the message.

[0201] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0202] M2: Power headroom measurement performed by the UE.

[0203] M3: Received interference power measurement.

[0204] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0205] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0206] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0207] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0208] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0209] M9: RTT measurement performed by UE (for WLAN measurement).

[0210] Inferred QoS Data Request: This information is used to request inferred QoS information available at the NG-RAN node 2 for one or more specific UEs.

[0211] This information may contain an indication of the requirements for inferred QoS configurations. In this case, the NG-RAN node 2 may provide all inferred QoS configurations that meet the requirements in this message.

[0212] The information may include detailed QoS configuration parameters. In this case, the NG-RAN node 2 may provide all inferred QoS configurations regarding specific parameters that meet the requirements of one or more specific UEs in the message. The message may include at least one of the following parameters.

[0213] Inferred QoS Configuration Required Indication: This is used to indicate that the NG-RAN Node 2 needs to provide inferred QoS data.

[0214] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0215] Inferred QoE data request: This information is used to request inferred QoE information available at the NG-RAN node 2 for one or more specific UEs.

[0216] This information may contain an indication for inferred QoE requirements. In this case, the NG-RAN node 2 may provide all inferred QoE data that meets the requirements in this message.

[0217] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 2 may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0218] Inferred QoE Required Indication: This is used to indicate that the NG-RAN node 2 needs to provide inferred QoE data.

[0219] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0220] Inferred MDT Data Request: This information is used to request the inferred MDT information available at the NG-RAN node 2 for one or more specific UEs.

[0221] This message may contain an indication of the requirements for inferred MDT information. In this case, the NG-RAN node 2 may provide all inferred MDT data that meets the requirements in this message.

[0222] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 2 may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0223] Inferred MDT Required Indication: This is used to indicate that the NG-RAN node 2 needs to provide inferred MDT data.

[0224] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0225] M2: Power headroom measurement performed by the UE.

[0226] M3: Received interference power measurement.

[0227] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0228] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0229] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0230] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0231] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0232] M9: RTT measurement performed by UE (for WLAN measurement).

[0233] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The NG-RAN node 2 can only provide the inferred data if the confidence of the inferred data is not lower than this threshold. Each inferred data type (QoE, QoS, MDT) can have an associated confidence requirement. In another alternative, one confidence information can be used to qualify all inferred data in this message.

[0234] Time related information.

[0235] Time information of the required data: NG-RAN Node 1 can request NG-RAN Node 2 to provide data blocks for one or more specific UEs within a certain duration (or start and end time).

[0236] Reporting frequency: NG-RAN node 2 may periodically transmit data required by one or more specific UEs to NG-RAN node 1.

[0237] Time-related information can coexist in the message. If so, the NG-RAN node 2 shall send the data block that meets the duration limit in the "Time information of required data" in the first response message. The NG-RAN node 2 can then send the remaining stored data or newly generated data to the NG-RAM node 1 by using the configured "Reporting Frequency".

[0238] Step 4: Based on the requirements received in step 3, NG-RAN node 2 may transmit the data required for the AI ​​function to NG-RAN node 1. The message may contain at least one of the following information.

[0239] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 1 only needs to provide the required data for one or more specific UEs.

[0240] Null Indicator: This optional indicator shall be sent to the NG-RAN node 1 if no information is available on the NG-RAN node 2 side.

[0241] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 3.

[0242] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 3.

[0243] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 1. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0244] Time information of data: Time information of the provided data. It can be the start and end time, or duration.

[0245] Step 4a: If the data is too large or needs to be transmitted regularly, multiple messages will be used for data transmission. The message may contain at least one of the following information.

[0246] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 1 only needs to provide the required data for one or more specific UEs.

[0247] Null Indicator: This optional indicator shall be sent to the NG-RAN node 1 if no information is available on the NG-RAN node 2 side.

[0248] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 3.

[0249] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 3.

[0250] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 1. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0251] Time information of data: Time information of the provided data. It can be the start and end time, or duration.

[0252] Step 5: After collecting enough data, NG-RAN Node 1 can perform the AI ​​model training process with AI / ML functionality enabled.

[0253] This step may be reflected in the 3GPP specification as a stage 2 description.

[0254] Step 6: After the AI ​​model is generated in NG-RAN Node 1, the new model will be deployed / updated to NG-RAN Node 1.

[0255] Step 7: The UE continues to send measurement data (if any) to NG-RAN node 1.

[0256] Step 8: NG-RAN node 1 sends a message to NG-RAN node 2 and requests data for AI functions (e.g., model inference). This message may include at least one of the following information.

[0257] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 1 only needs to provide the required data for one or more specific UEs.

[0258] Processing indicator: This indication is used to inform the NG-RAN node 1 that the required information can be used for online training and / or inference. The NG-RAN node 2 can provide different amounts of data based on this indication.

[0259] Historical QoS configuration request information: This message is used to request historical QoS configuration information stored in / available in the NG-RAN node 2.

[0260] This information may contain an indication of historical QoS configuration requirements. In this case, the NG-RAN node 2 may provide all historical QoS configurations that meet the requirements in this message.

[0261] This information may include detailed QoS configuration parameters. In this case, the NG-RAN node 2 may provide all historical QoS configurations that meet the specific parameters required in this message. The message may include at least one of the following parameters.

[0262] Historical QoS Configuration Requirement Indication, 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0263] Historical QoE (RVQoE) Required Information: This information is used to request the historical QoE information available at the NG-RAN node 2.

[0264] This information may contain an indication of historical QoE requirements. In this case, the NG-RAN node 2 may provide all historical QoE data that meets the requirements in this message.

[0265] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 2 may provide all historical QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0266] Historical QoE requirement indication, QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0267] Historical MDT Information Request message: This message is used to request the historical MDT information available at the NG-RAN node 2.

[0268] This message may contain an indication for historical MDT information requirements. In this case, the NG-RAN node 2 may provide all historical MDT data that meets the requirements in this message.

[0269] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 2 may provide all historical MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0270] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0271] M2: Power headroom measurement performed by the UE.

[0272] M3: Received interference power measurement.

[0273] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0274] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0275] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0276] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0277] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0278] M9: RTT measurement performed by UE (for WLAN measurement).

[0279] Inferred QoS Data Request: This information is used to request the inferred QoS information available at the NG-RAN Node 2.

[0280] This information may contain an indication of the requirements for inferred QoS configurations. In this case, the NG-RAN node 2 may provide all inferred QoS configurations that meet the requirements in this message.

[0281] This information may contain detailed QoS configuration parameters. In this case, the NG-RAN node 2 may provide all inferred QoS configurations regarding specific parameters that meet the requirements in this message. At least one of the following parameters may be included in this message.

[0282] Inferred QoS Configuration Required Indication: This is used to indicate that the NG-RAN Node 2 needs to provide inferred QoS data.

[0283] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0284] Inferred QoE Data Request: This information is used to request the inferred QoE information available at the NG-RAN Node 2.

[0285] This information may contain an indication for inferred QoE requirements. In this case, the NG-RAN node 2 may provide all inferred QoE data that meets the requirements in this message.

[0286] This information may contain detailed QoE parameters / metrics. In this case, the NG-RAN node 2 may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0287] Inferred QoE Required Indication: This is used to indicate that the NG-RAN node 2 needs to provide inferred QoE data.

[0288] The start and end time of this inferred data or the duration of this inferred data.

[0289] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0290] Inferred MDT Data Request: This information is used to request the inferred MDT information available on the NG-RAN node 2.

[0291] This message may contain an indication of the requirements for inferred MDT information. In this case, the NG-RAN node 2 may provide all inferred MDT data that meets the requirements in this message.

[0292] This information may contain detailed MDT parameters / metrics. In this case, the NG-RAN node 2 may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0293] Inferred MDT Required Indication: This is used to indicate that the NG-RAN node 2 needs to provide inferred MDT data.

[0294] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0295] M2: Power headroom measurement performed by the UE.

[0296] M3: Received interference power measurement.

[0297] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0298] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0299] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0300] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0301] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0302] M9: RTT measurement performed by UE (for WLAN measurement).

[0303] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The NG-RAN node 2 can only provide the inferred data if the confidence of the inferred data is not lower than this threshold. Each inferred data type (QoE, QoS, MDT) may have an associated confidence requirement. In another alternative, one confidence information may be used to qualify all inferred data in this message.

[0304] Time related information.

[0305] Time information of the required data: NG-RAN Node 1 can request NG-RAN Node 2 to provide data blocks for one or more specific UEs within a certain duration (or start and end time).

[0306] Reporting frequency: NG-RAN node 2 may periodically transmit data required by one or more specific UEs to NG-RAN node 1.

[0307] Time-related information can coexist in the message. If so, the NG-RAN node 2 shall send the data block that meets the duration limit in the "Time information of required data" in the first response message. The NG-RAN node 2 can then send the remaining stored data or newly generated data to the NG-RAM node 2 by using the configured "Reporting frequency".

[0308] Step 9: Based on the request received in step 8, NG-RAN node 2 may transmit the historical data of the AI ​​function to NG-RAN node 1. The message may contain at least one of the following information.

[0309] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 2 only needs to provide historical data for one or more specific UEs.

[0310] Null Indicator: This optional indicator shall be sent to the NG-RAN Node 1 if no information is available on the NG-RAN Node 1 side.

[0311] If available, historical QoE data, historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0312] Inferred QoE data, inferred QoS data and inferred MDT data will be added to the message if available. Detailed parameters / metrics of this data can be found in step 8.

[0313] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 2. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0314] Time information of historical data: time information of the provided data.

[0315] Step 9a: If the historical data is too large or needs to be transmitted regularly, more than one message will be used for historical data transmission. The message may contain at least one of the following information.

[0316] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The NG-RAN node 2 only needs to provide historical data for one or more specific UEs.

[0317] Null Indicator: This optional indicator shall be sent to the NG-RAN Node 1 if no information is available on the NG-RAN Node 1 side.

[0318] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0319] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 8.

[0320] Confidence Information for Inferred Data: This is an information element with a value from 0 to 100 (or can be a ratio) generated by the NG-RAN Node 2. It is used to mark the level of confidence / accuracy that the inferred data should have. Each type of inferred data (QoE, QoS, MDT) can have one confidence information associated with it. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0321] Time information of historical data: time information of the provided data.

[0322] Step 10: After collecting sufficient data, the NG-RAN node 1 can perform the AI ​​model inference process with the AI / ML function enabled. The NG-RAN node 1 will generate predicted QoE data and / or predicted QoS configuration and / or MDT data for one or more specific UEs for the next one or more subsequent timings.

[0323] This step may be reflected in the 3GPP specification as a stage 2 description.

[0324] Step 11: After the model inference, the NG-RAN node 1 may trigger the inference action procedure. With the assistance of the inferred / predicted QoE, QoS data, the NG-RAN node 1 may perform some procedures including but not limited to the following procedures.

[0325] NG-RAN node 1 can reallocate 3GPP / non-3GPP resources for each UE in the NG-RAN node.

[0326] The NG-RAN node 1 may update the mobility policy of one or more specific UEs.

[0327] The NG-RAN node 1 may send the inferred / predicted data to other NG-RAN nodes for use in the inference procedure.

[0328] NG-RAN node 1 may send the inferred / predicted data to NG-RAN node 2 for further QoS / QoE configuration etc.

[0329] This step may be reflected in the 3GPP specification as a stage 2 description.

[0330] Step 12: A detailed description of the feedback procedure can be found in Example 5.

[0331] IV. Example 3

[0332] like Figure 4 As shown, Example 3 shows AI training performed by a central unit (CU) and AI inference performed by a distributed unit (DU).

[0333] The procedure involved between DU and CU is F1 Application Protocol (F1AP) enhancement.

[0334] Before this procedure occurs, all involved network entities may already be deployed with AI models, so that the CU and DU can provide inferred data when they receive the data request in step 3 / 8.

[0335] Step 1: The UE has configured QoE measurement and is performing QoE data collection.

[0336] Step 2: The UE periodically transmits measurement reports to the DU.

[0337] Step 3: CU sends a message to DU and requests data for AI functions (e.g., model training). This message may include at least one of the following information.

[0338] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. A DU only needs to provide historical data for one or more specific UEs.

[0339] Processing indicator: This indicator is used to inform the DU that the required information can be used for online training and / or inference. The DU can provide different amounts of data based on this indicator.

[0340] Historical QoS configuration request message: This message is used to request historical QoS configuration information stored in / available in the DU.

[0341] This information may contain an indication of historical QoS configuration requirements. In this case, the DU may provide all historical QoS configurations that meet the requirements in this message.

[0342] This information may contain detailed QoS configuration parameters. In this case, the DU may provide all historical QoS configurations for specific parameters that meet the requirements in this message. The message may include at least one of the following parameters.

[0343] Historical QoS Configuration Requirement Indication, 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0344] Historical QoE (RVQoE) Required Information: This information is used to request historical QoE information available at the DU.

[0345] This information may contain an indication of historical QoE requirements. In this case, the DU may provide all historical QoE data that meets the requirements in this message.

[0346] This information may contain detailed QoE parameters / metrics. In this case, the DU may provide all historical QoE data on specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0347] Historical QoE requirement indication, QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0348] Historical MDT information request message: This message is used to request historical MDT information available at the DU.

[0349] This message may contain an indication for historical MDT information requirements. In this case, the DU may provide all historical MDT data that meets the requirements in this message.

[0350] This information may contain detailed MDT parameters / metrics. In this case, the DU may provide all historical MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0351] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0352] M2: Power headroom measurement performed by the UE.

[0353] M3: Received interference power measurement.

[0354] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0355] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0356] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0357] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0358] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0359] M9: RTT measurement performed by UE (for WLAN measurement).

[0360] Inferred QoS Data Request: This information is used to request the inferred QoS information available at the DU.

[0361] This message may contain an indication of inferred QoS configuration requirements. In this case, the DU may provide all inferred QoS configurations that meet the requirements in this message.

[0362] This information may contain detailed QoS configuration parameters. In this case, the DU may provide all inferred QoS configurations that meet the specific parameters required in this message. This message may include at least one of the following parameters.

[0363] Inferred QoS Configuration Required Indication: This is used to indicate that the DU needs to provide inferred QoS data.

[0364] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0365] Inferred QoE data request: This information is used to request the inferred QoE information available at the DU.

[0366] This information may contain an indication for inferred QoE requirements. In this case, the DU may provide all inferred QoE data that meets the requirements in this message.

[0367] This information may contain detailed QoE parameters / metrics. In this case, the DU may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0368] Inferred QoE requirement indication: This is used to indicate that the DU needs to provide inferred QoE data.

[0369] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0370] Inferred MDT Data Request: This information is used to request the inferred MDT information available at the DU.

[0371] This message may contain an indication of the inferred MDT information requirements. In this case, the DU may provide all inferred MDT data that meets the requirements in this message.

[0372] This information may contain detailed MDT parameters / metrics. In this case, the DU may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0373] Inferred MDT Required Indication: This is used to indicate that the DU needs to provide inferred MDT data.

[0374] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0375] M2: Power headroom measurement performed by the UE.

[0376] M3: Received interference power measurement.

[0377] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0378] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0379] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0380] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0381] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0382] M9: RTT measurement performed by UE (for WLAN measurement).

[0383] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The DU can only provide inferred data if the confidence of the inferred data is not lower than this threshold. Each inferred data type (QoE, QoS, MDT) can have an associated confidence requirement. In another alternative, a confidence information can be used to limit all inferred data in this message.

[0384] Time related information.

[0385] Time information of the required data: The CU can request the DU to provide a data block for one or more specific UEs within a certain duration (or start and end time).

[0386] Reporting frequency: The DU can periodically transmit data required by one or more specific UEs to the CU.

[0387] Time-related information can coexist in messages. If so, the NG-RAN node shall send the data block that meets the duration limit in the "Time Information of Required Data" in the first response message. The DU can then send the remaining stored data or newly generated data to the CU by using the configured "Reporting Frequency".

[0388] Step 4: Based on the request received in step 3, the DU may transmit historical data for AI functions to the CU. The message may include at least one of the following information.

[0389] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. A DU only needs to provide historical data for one or more specific UEs.

[0390] Null Indicator: If there is no information available on the DU side, this optional indicator will be sent to the CU.

[0391] If available, historical QoE data, historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 3.

[0392] Inferred QoE data, inferred QoS data and inferred MDT data will be added to the message if available. Detailed parameters / metrics of this data can be found in step 3.

[0393] Confidence Information for Inferred Data: This is an information element generated by the DU with a value from 0 to 100 (or can be a ratio). It is used to mark the level of trust / accuracy that the inferred data should reach. Each inferred data type (QoE, QoS, MDT) can have an associated confidence information. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0394] Time information of data: Time information of the provided data. It can be the start and end time, or duration.

[0395] Step 4a: If the data is too large or needs to be transmitted regularly, multiple messages will be used for data transmission. The message may contain at least one of the following information.

[0396] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. A DU only needs to provide historical data for one or more specific UEs.

[0397] Null Indicator: If there is no information available on the DU side, this optional indicator will be sent to the CU.

[0398] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 3.

[0399] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 3.

[0400] Confidence Information for Inferred Data: This is an information element generated by the DU with a value from 0 to 100 (or can be a ratio). It is used to mark the level of trust / accuracy that the inferred data should reach. Each inferred data type (QoE, QoS, MDT) can have an associated confidence information. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0401] Time information of data: Time information of the provided data. It can be the start and end time, or duration.

[0402] Step 5: After collecting enough data, the CU can perform the AI ​​model training process with AI / ML capabilities enabled.

[0403] This step may be reflected in the 3GPP specification as a stage 2 description.

[0404] Step 6: After the CU generates the AI ​​model, the new model will be deployed / updated to the DU.

[0405] Step 7: The UE continues to send measurement data to the DU (if available).

[0406] Step 8: DU sends a message to CU and requests data for AI functions (e.g., model inference). This message may include at least one of the following information.

[0407] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The CU only needs to provide historical data for one or more specific UEs.

[0408] Processing indicator: This indicator is used to inform the CU that the required information can be used for online training and / or inference. The CU can provide different amounts of data based on this indicator.

[0409] Historical QoS configuration request information: This message is used to request historical QoS configuration information stored in the CU / available in the DU.

[0410] This information may contain an indication of historical QoS configuration requirements. In this case, the DU may provide all historical QoS configurations that meet the requirements in this message.

[0411] This information may contain detailed QoS configuration parameters. In this case, the CU may provide all historical QoS configurations for specific parameters that meet the requirements in this message. The message may include at least one of the following parameters.

[0412] Historical QoS Configuration Requirement Indication, 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0413] Historical QoE (RVQoE) Required Information: This information is used to request historical QoE information available at the CU.

[0414] This information may contain an indication of historical QoE requirements. In this case, the CU may provide all historical QoE data that meets the requirements in this message.

[0415] This information may contain detailed QoE parameters / metrics. In this case, the CU may provide all historical QoE data for specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0416] Historical QoE requirement indication, QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0417] Historical MDT information request message: This message is used to request historical MDT information available at the CU.

[0418] This message may contain an indication for historical MDT information requirements. In this case, the CU may provide all historical MDT data that meets the requirements in this message.

[0419] This information may contain detailed MDT parameters / metrics. In this case, the CU may provide all historical MDT data for specific parameters / metrics that meet the requirements in this message. The message may include at least one of the following parameters / metrics.

[0420] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0421] M2: Power headroom measurement performed by the UE.

[0422] M3: Received interference power measurement.

[0423] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0424] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0425] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0426] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0427] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0428] M9: RTT measurement performed by UE (for WLAN measurement).

[0429] Inferred QoS Data Request: This information is used to request the inferred QoS information available at the CU.

[0430] This message may contain instructions for inferring QoS configuration requirements. In this case, the CU may provide all inferred QoS configurations that meet the requirements in this message.

[0431] This message may contain detailed QoS configuration parameters. In this case, the CU may provide all inferred QoS configurations that meet the specific parameters required in this message. This message may include at least one of the following parameters.

[0432] Inferred QoS configuration required indication: This is used to indicate that the CU needs to provide inferred QoS data.

[0433] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0434] Inferred QoE data request: This information is used to request inferred QoE information available at the CU.

[0435] This information may contain an indication for inferring QoE requirements. In this case, the CU may provide all inferred QoE data that meets the requirements in this message.

[0436] This information may contain detailed QoE parameters / metrics. In this case, the CU may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0437] Inferred QoE requirement indication: This is used to indicate that the DU needs to provide inferred QoE data.

[0438] The start and end time of this inferred data or the duration of this inferred data.

[0439] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0440] Inferred MDT Data Request: This information is used to request the inferred MDT information available at the CU.

[0441] This message may contain an indication of the inferred MDT information requirements. In this case, the CU may provide all inferred MDT data that meets the requirements in this message.

[0442] This information may contain detailed MDT parameters / metrics. In this case, the CU may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0443] Inferred MDT Required Indication: This is used to indicate that the DU needs to provide inferred MDT data.

[0444] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0445] M2: Power headroom measurement performed by the UE.

[0446] M3: Received interference power measurement.

[0447] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0448] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0449] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0450] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0451] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0452] M9: RTT measurement performed by UE (for WLAN measurement).

[0453] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The CU can only provide inferred data if the confidence of the inferred data is not lower than this threshold. Each inferred data type (QoE, QoS, MDT) can have an associated confidence requirement. In another alternative, one confidence information can be used to limit all inferred data in this message.

[0454] Time related information.

[0455] Time information of the required data: A DU can request a CU to provide a data block for one or more specific UEs within a certain duration (or start and end time).

[0456] Reporting frequency: The CU can periodically transmit data required by one or more specific UEs to the DU.

[0457] Time-related information can coexist in the message. If so, the CU shall send the data block that meets the duration limit in the "Time Information of Required Data" in the first response message. The CU can then send the remaining stored data or newly generated data to the DU by using the configured "Report Frequency".

[0458] Feedback request indication: This indication is used by the DU to request feedback on the inference. Through this indication, after the CU receives the inference data from the DU, it should send feedback information to the DU (if available).

[0459] Step 9: Based on the requirements received in step 8, the CU may transmit the data required for the AI ​​function to the DU. The message may include at least one of the following information.

[0460] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The CU only needs to provide historical data for one or more specific UEs.

[0461] Null Indicator: This optional indicator will be sent to the DU if no information is available on the CU side.

[0462] If available, historical QoE data, historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0463] Inferred QoE data, inferred QoS data and inferred MDT data will be added to the message if available. Detailed parameters / metrics of this data can be found in step 8.

[0464] Confidence information for inferred data: This is an information element generated by the CU with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. Each inferred data type (QoE, QoS, MDT) can have an associated confidence information. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0465] Time information of historical data: time information of the provided data.

[0466] Step 9a: If the historical data is too large or needs to be transmitted regularly, more than one message will be used for historical data transmission. The message may contain at least one of the following information.

[0467] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. The CU only needs to provide historical data for one or more specific UEs.

[0468] Null Indicator: This optional indicator will be sent to the DU if no information is available on the CU side.

[0469] If available, historical QoE data, historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0470] Inferred QoE data, inferred QoS data and inferred MDT data will be added to the message if available. Detailed parameters / metrics of this data can be found in step 8.

[0471] Confidence information for inferred data: This is an information element generated by the CU with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. Each inferred data type (QoE, QoS, MDT) can have an associated confidence information. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0472] Time information of historical data: time information of the provided data.

[0473] Step 10: After collecting enough data, the DU can perform the AI ​​model inference process with the AI / ML function enabled. The DU will generate predicted QoE data and / or predicted QoS configuration and / or predicted MDT data for one or more specific UEs for the next one or more subsequent timings.

[0474] This step may be reflected in the 3GPP specification as a stage 2 description.

[0475] Step 11: After model inference, the DU can trigger the inference action procedure. With the assistance of the inferred / predicted QoE, QoS data, the DU can perform some processes including but not limited to the following processes.

[0476] The DU may reschedule 3GPP / non-3GPP resources for the UE based on the received inference data.

[0477] The DU may send QoS / RB modification requirements to the CU based on an evaluation of the received inference data, etc.

[0478] This step may be reflected in the 3GPP specification as a stage 2 description.

[0479] Step 12: A detailed description of the feedback procedure can be found in Example 5.

[0480] V. Embodiment 4

[0481] like Figure 5 As shown, Example 4 shows AI training and inference performed by a central unit (CU).

[0482] The procedure involved between DU and CU is F1AP enhancement.

[0483] Before this procedure occurs, all involved network entities may already be deployed with AI models, so that the CU and DU can provide inferred data when they receive the data request in step 3 / 8.

[0484] Step 1: The UE has configured QoE measurement and is performing QoE data collection.

[0485] Step 2: The UE periodically transmits measurement reports to the DU.

[0486] Steps 3, 4, 4a and 5 are the same as those in Example 3.

[0487] Step 6: CU can send the updated AI model to DU.

[0488] In this embodiment, the CU also uses the updated AI model locally.

[0489] Step 7: The UE continues to send measurement data to the DU (if available).

[0490] Step 8: The CU sends a message to the DU and requests data for AI functions (e.g., model inference). This message may include at least one of the following information.

[0491] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. A DU only needs to provide historical data for one or more specific UEs.

[0492] Processing indicator: This indicator is used to inform the DU that the required information can be used for online training and / or inference. The DU can provide different amounts of data based on this indicator.

[0493] Historical QoS configuration request message: This message is used to request historical QoS configuration information stored in / available in the DU.

[0494] This information may contain an indication of historical QoS configuration requirements. In this case, the DU may provide all historical QoS configurations that meet the requirements in this message.

[0495] This information may contain detailed QoS configuration parameters. In this case, the DU may provide all historical QoS configurations for specific parameters that meet the requirements in this message. The message may include at least one of the following parameters.

[0496] Historical QoS Configuration Requirement Indication, 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0497] Historical QoE (RVQoE) Required Information: This information is used to request historical QoE information available at the DU.

[0498] This information may contain an indication of historical QoE requirements. In this case, the DU may provide all historical QoE data that meets the requirements in this message.

[0499] This information may contain detailed QoE parameters / metrics. In this case, the DU may provide all historical QoE data for specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0500] Historical QoE requirement indication, QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0501] Historical MDT information request message: This message is used to request historical MDT information available at the DU.

[0502] This message may contain an indication for historical MDT information requirements. In this case, the DU may provide all historical MDT data that meets the requirements in this message.

[0503] This information may contain detailed MDT parameters / metrics. In this case, the DU may provide all historical MDT data for specific parameters / metrics that meet the requirements in this message. The message may include at least one of the following parameters / metrics.

[0504] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0505] M2: Power headroom measurement performed by the UE.

[0506] M3: Received interference power measurement.

[0507] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0508] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0509] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0510] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0511] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0512] M9: RTT measurement performed by UE (for WLAN measurement).

[0513] Inferred QoS Data Request: This information is used to request the inferred QoS information available at the DU.

[0514] This message may contain an indication of inferred QoS configuration requirements. In this case, the DU may provide all inferred QoS configurations that meet the requirements in this message.

[0515] This information may contain detailed QoS configuration parameters. In this case, the DU may provide all inferred QoS configurations that meet the specific parameters required in this message. This message may include at least one of the following parameters.

[0516] Inferred QoS Configuration Required Indication: This is used to indicate that the DU needs to provide inferred QoS data.

[0517] 5QI Reference or QoS Level Identifier, Service Type, Packet Delay Range, Maximum Packet Delay, Minimum Packet Delay, Average Packet Delay, Packet Error Rate Range, Maximum Packet Error Rate, Minimum Packet Error Rate, Average Packet Error Rate, Packet Loss Rate Range, Maximum Packet Loss Rate, Minimum Packet Loss Rate, Average Packet Loss Rate, Guaranteed Flow Bit Rate Range, Maximum Guaranteed Flow Bit Rate, Minimum Guaranteed Flow Bit Rate, Average Guaranteed Flow Bit Rate.

[0518] Inferred QoE data request: This information is used to request the inferred QoE information available at the DU.

[0519] This information may contain an indication for inferred QoE requirements. In this case, the DU may provide all inferred QoE data that meets the requirements in this message.

[0520] This information may contain detailed QoE parameters / metrics. In this case, the DU may provide all inferred QoE data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0521] Inferred QoE requirement indication: This is used to indicate that the DU needs to provide inferred QoE data.

[0522] The start and end time of this inferred data or the duration of this inferred data.

[0523] QoE level identifier, service type, area range, QoE type (signaling-based QoE, management-based QoE), RAN-visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric, average QoE metric, where the QoE score can be a numeric range, such as 0 to 10, where 10 represents good quality and 0 represents poor quality; or the QoE score can be an enumeration type to indicate quality, such as (poor, medium, good); where the QoE metric is a subset of the QoE metric data guaranteed for the UE, such as round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution, buffer occupancy level, etc.

[0524] Inferred MDT Data Request: This information is used to request the inferred MDT information available at the DU.

[0525] This message may contain an indication of the inferred MDT information requirements. In this case, the DU may provide all inferred MDT data that meets the requirements in this message.

[0526] This information may contain detailed MDT parameters / metrics. In this case, the DU may provide all inferred MDT data about specific parameters / metrics that meet the requirements in this message. At least one of the following parameters / metrics may be included in this message.

[0527] Inferred MDT Required Indication: This is used to indicate that the DU needs to provide inferred MDT data.

[0528] M1: DL signal level measurement results of the serving cell and intra-frequency / inter-frequency / inter-RAT neighboring cells, including only the cell / beam level measurement of the NR cell.

[0529] M2: Power headroom measurement performed by the UE.

[0530] M3: Received interference power measurement.

[0531] M4: Data volume measurement for DL ​​and UL separately (per DRB per UE).

[0532] M5: Average UE throughput measurements performed by the gNB for DL ​​and UL separately (per DRB per UE and per UE for DL, per DRB per UE and per UE for UL).

[0533] M6: Packet delay measurement for DL ​​and UL separately (per DRB per UE).

[0534] M7: Packet loss rate measurement for DL ​​and UL (per DRB per UE) separately.

[0535] M8: RSSI measurement by UE (for WLAN / Bluetooth measurements).

[0536] M9: RTT measurement performed by UE (for WLAN measurement).

[0537] Confidence requirement for inferred data: This is an information element with a value from 0 to 100 (or can be a ratio). It is used to mark the confidence level / accuracy that the inferred data should reach. The DU can only provide inferred data if the confidence of the inferred data is not lower than this threshold. Each inferred data type (QoE, QoS, MDT) can have an associated confidence requirement. In another alternative, a confidence information can be used to limit all inferred data in this message.

[0538] Time related information.

[0539] Time information of the required data: The CU can request the DU to provide a data block for one or more specific UEs within a certain duration (or start and end time).

[0540] Reporting frequency: The DU can periodically transmit data required by one or more specific UEs to the CU.

[0541] Time-related information can coexist in the message. If so, the DU shall send the data block that meets the duration limit in the "Time Information of Required Data" in the first response message. The DU can then send the remaining stored data or newly generated data to the CU by using the configured "Report Frequency".

[0542] Feedback request indication: This indication is used by the CU to request feedback on the inference. Through this indication, after the DU receives the inference data from the CU, it should send feedback information to the CU (if available).

[0543] Step 9: Based on the requirements received in step 8, the DU may transmit the data required for the AI ​​function to the CU. The message may include at least one of the following information.

[0544] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. A DU only needs to provide historical data for one or more specific UEs.

[0545] Null Indicator: If there is no information available on the DU side, this optional indicator will be sent to the CU.

[0546] If available, historical QoE data, historical QoS data and historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0547] Inferred QoE data, inferred QoS data and inferred MDT data will be added to the message if available. Detailed parameters / metrics of this data can be found in step 8.

[0548] Confidence Information for Inferred Data: This is an information element generated by the DU with a value from 0 to 100 (or can be a ratio). It is used to mark the level of trust / accuracy that the inferred data should reach. Each inferred data type (QoE, QoS, MDT) can have an associated confidence information. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0549] Time information of historical data: time information of the provided data.

[0550] Step 9a: If the historical data is too large or needs to be transmitted regularly, more than one message will be used for historical data transmission. The message may contain at least one of the following information.

[0551] One or more UE IDs: One or more UE IDs used to uniquely identify a UE. A DU only needs to provide historical data for one or more specific UEs.

[0552] Null Indicator: If there is no information available on the DU side, this optional indicator will be sent to the CU.

[0553] If available, historical QoE data and / or historical QoS data and / or historical MDT data will be added to this message. Detailed parameters / metrics of this data can be found in step 8.

[0554] If available, the inferred QoE data and / or inferred QoS data and / or inferred MDT data will be added in the message. Detailed parameters / metrics of this data can be found in step 8.

[0555] Confidence Information for Inferred Data: This is an information element generated by the DU with a value from 0 to 100 (or can be a ratio). It is used to mark the level of trust / accuracy that the inferred data should reach. Each inferred data type (QoE, QoS, MDT) can have an associated confidence information. In another alternative, one confidence information can be used to mark all inferred data in this message.

[0556] Time information of historical data: time information of the provided data.

[0557] Step 10: After collecting enough data, the CU can perform the AI ​​model inference process with the AI / ML function enabled. The CU will generate predicted QoE data and / or predicted QoS configuration and / or predicted MDT data for one or more specific UEs for the next one or more subsequent timings.

[0558] This step may be reflected in the 3GPP specification as a stage 2 description.

[0559] Step 11: After the model is inferred, the CU can trigger the inference action procedure. With the assistance of the inferred / predicted QoE, QoS data, the CU can perform some processes including but not limited to the following processes.

[0560] The CU can reallocate 3GPP / non-3GPP resources for each UE in the NG-RAN node.

[0561] The CU may update the mobility policy of one or more specific UEs.

[0562] The CU may send the inferred / predicted data to other network entities (e.g., DU, other NG-RAN nodes) for inference procedures.

[0563] The CU may send the inferred / predicted data to (e.g., DU, other NG-RAN nodes) for further QoS / QoE configuration, etc.

[0564] This step may be reflected in the 3GPP specification as a stage 2 description.

[0565] Step 12: A detailed description of the feedback procedure can be found in Example 5.

[0566] VI. Example 5

[0567] like Figure 6 As shown, Example 5 shows a feedback procedure.

[0568] Step 1: Entity 1 makes an inference and sends the inference data to entity 2.

[0569] [Optional] Step 2: Entity 1 may send a feedback request message to entity 2. The message may include at least one of the following information.

[0570] Feedback request indication: This indicates that entity 2 needs to provide feedback data for the received inference data.

[0571] UE ID: One or more UEs used to uniquely identify one or more UEs.

[0572] Timing Information: This indicates that the feedback is reflected in the timing information used to infer the data.

[0573] Feedback data type request: This indication is used by entity 1 to request specific feedback of one / multiple data types (eg, QoS, QoE, MDT, etc.).

[0574] Feedback request: This indication is used for entity 1 to request entity 2 to provide feedback information for inference data. At least one of the following information shall be included in this aspect.

[0575] The level of relationship between two of the following information: QoE data, QoS configuration, MDT data. The detailed data types can be found in the previous embodiments. How the level of relationship between two data is represented by a value within a value range (0,100), (range 0 to 10), a ratio or "low, medium, high", etc.

[0576] The level of precision of the inferred data. This can be expressed as a value in the range (0,100), (range 0,10), a ratio, or "low, medium, high", "low, high", etc.

[0577] Reflection on QoE data. Is the QoE data better than the previous data before AI / ML optimization?

[0578] This step may be defined by 3GPP. However, if there is no such message, entity 2 may also automatically send a feedback message to entity 1.

[0579] Step 3: Entity 2 sends a response message to entity 1. At least one of the following information should be included in the message.

[0580] UE ID: One or more UEs used to uniquely identify one or more UEs.

[0581] Timing Information: This indicates that the feedback is reflected in the timing information used to infer the data.

[0582] UE QoS Configuration List: This contains the actual QoS configuration of the UE.

[0583] UE QoE Information List: This contains the actual QoE measurement data of the UE.

[0584] UE MDT Information List: This contains the actual MDT measurement data of the UE.

[0585] Feedback Information: Detailed data about the feedback depends on the requirements. At least one of the following information will be included in this aspect.

[0586] The level of relationship between two of the following information: QoE data, QoS configuration, MDT data. The detailed data types can be found in the previous embodiments. How the level of relationship between two data is represented by a value within a value range (0,100), (range 0 to 10), a ratio or "low, medium, high", etc.

[0587] The level of precision of the inferred data. This can be expressed as a value in the range (0,100), (range 0,10), a ratio, or "low, medium, high", "low, high", etc.

[0588] Reflection on QoE data. Is the QoE data better than the previous data before AI / ML optimization?

[0589] Based on the procedure shown in this embodiment, two feedback mechanisms can be used.

[0590] Working mode 1.

[0591] In this alternative, entity 2 may send feedback data (eg, UEQoS, QoE, MDT, confidence, etc.) for a timing to entity 1.

[0592] Working mode 2.

[0593] In this alternative, entity 2 may send a list containing one or more feedback data for different timings (e.g., UE QoS, QoE, MDT, timing, etc.) to entity 1. Based on the implementation or 3GPP definition, confidence may be used to reflect the inferred data in this list. Alternatively, the inferred data for each timing may be linked to confidence information.

[0594] This embodiment may require different enhancements.

[0595] If entities 1 and 2 are CN and RAN nodes, NGAP enhancement is required.

[0596] If entities 1 and 2 are RAN nodes, XnAP enhancement is required.

[0597] If entities 1 and 2 are CU and / or DU, F1AP enhancement is required.

[0598] The input data consists of historical data (QoE, QoS, MDT) and / or inferred data (QoE, QoS, MDT). The input data can be used by an entity responsible for the AI / ML model function and as shown in the above embodiments for AI model training and / or model inference. When the AI / ML entity receives the input data, it can send the input data to the AI / ML model function. With the assistance of the AI / ML model function, the entity can generate a new or updated AI / ML model function. In addition, with the assistance of the AI / ML model function, inferred data in different fields (e.g., QoE, QoS, MDT) can be generated based on the received input data.

[0599] By using the inferred data (e.g., QoE, QoS, MDT), an entity (e.g., a different entity in different embodiments) can reschedule 3GPP / non-3GPP resources for a specific UE by considering the current and future conditions of the specific UE. In addition, the entity can update the mobility policy of the UE by considering the inferred data of the UE and other inferred data from neighboring nodes of the UE's serving node.

[0600] Furthermore, by receiving inferred data from multiple UEs from the same / different entities, network entities (e.g., CU, gNB) can optimize their local resources for better energy saving and efficiency by considering the predicted situations.

[0601] Figure 7 is an exemplary flow chart for requesting input data. Operation 702 includes transmitting a request for input data by a network node. Operation 704 includes receiving input data by the network node in response to the request. In some embodiments, the method may be implemented according to embodiments 1 to 4. In some embodiments, further steps of performing the method may be based on better system performance than conventional protocols.

[0602] Figure 8 is an exemplary flow chart for training an AI function. Operation 802 includes providing, by a network node, training for an artificial intelligence (AI) function using input data. Operation 804 includes transmitting, by a network node, an updated AI function based on the training. In some embodiments, the method may be implemented according to embodiments 1 to 4. In some embodiments, further steps of performing the method may be based on better system performance than conventional protocols.

[0603] Fig. 9is an exemplary flow chart for determining inferred data. Operation 902 includes determining, by a network node, inferred data using an artificial intelligence (AI) function based on input data. In some embodiments, the method may be implemented according to embodiments 1 to 4. In some embodiments, further steps of performing the method may be based on better system performance than conventional protocols.

[0604] In some embodiments, the network node is an operations, administration, and maintenance (OAM) node or a core network (CN) node, wherein transmitting the request comprises transmitting the request to a radio access network (RAN) node, and wherein receiving the input data comprises receiving the input data from the RAN node. In some embodiments, the network node is a first radio access network (RAN) node, wherein transmitting the request comprises transmitting the request to a second RAN node, and wherein receiving the input data comprises receiving the input data from the second RAN node. In some embodiments, the network node is a central unit (CU), wherein transmitting the request comprises transmitting the request to a distributed unit (DU), and wherein receiving the input data comprises receiving the input data from the DU. In some embodiments, the network node is a distributed unit (DU), wherein transmitting the request comprises transmitting the request to a central unit (CU), and wherein receiving the input data comprises receiving the input data from the CU.

[0605] In some embodiments, the request includes at least one of: a processing indicator, historical quality of service (QoS) configuration requirement information, historical quality of experience (QoE) data requirement information, historical minimization of drive tests (MDT) data requirement information, inferred QoS configuration requirement information, inferred QoE data requirement information, and inferred MDT data requirement information.

[0606] In some embodiments, the historical QoS configuration requirement information includes at least one of the following: a historical QoS configuration requirement indication, a 5G QoS identifier (5QI) reference, a QoS level identifier, a service type, a packet delay range, a maximum packet delay, a minimum packet delay, an average packet delay, a packet error rate range, a maximum packet error rate, a minimum packet error rate, an average packet error rate, a packet loss rate range, a maximum packet loss rate, a minimum packet loss rate, an average packet loss rate, a guaranteed flow bit rate range, a maximum guaranteed flow bit rate, a minimum guaranteed flow bit rate, and an average guaranteed flow bit rate. In some embodiments, the historical QoS configuration requirement indication includes a historical QoS configuration requirement, wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU), or a distributed unit (DU) provides all historical QoS configurations that meet the historical QoS configuration requirements.

[0607] In some embodiments, the historical QoE data requirement information includes at least one of the following: a historical QoE data requirement indication, a QoE level identifier, a service type, an area range, a QoE type, a radio access network (RAN) visible QoE data, a QoE report container, a codec rate, a QoE score, a QoE metric range, a maximum QoE metric, a minimum QoE metric, and an average QoE metric, wherein the QoE metric includes at least one of the following: round trip time, jitter duration, impairment duration, throughput, initial playback delay of a video, video resolution, and a buffer occupancy level. In some embodiments, the historical QoE data requirement indication includes a historical QoE data requirement, wherein the first radio access network (RAN) node, the second RAN node, a central unit (CU), or a distributed unit (DU) provides all historical QoE data that meets the historical QoE requirement.

[0608] In some embodiments, the historical MDT data requirement information includes at least one of the following: a historical MDT data requirement indication, a downlink (DL) signal volume measurement result, a power headroom measurement performed by a user equipment (UE), a received interference power measurement, a data volume measurement, an average UE throughput measurement, a packet delay measurement, a packet loss rate measurement, a received signal strength indicator (RSSI) measurement, and a round trip time (RTT) measurement. In some embodiments, the historical MDT data requirement indication includes a historical MDT data requirement, wherein the first radio access network (RAN) node, the second RAN node, the central unit (CU), or the distributed unit (DU) provides all historical MDT data that meets the historical MDT data requirement.

[0609] In some embodiments, the inferred QoS configuration requirement information includes at least one of the following: an inferred QoS configuration indication, a 5G QoS identifier (5QI) reference, a QoS level identifier, a service type, a packet delay range, a maximum packet delay, a minimum packet delay, an average packet delay, a packet error rate range, a maximum packet error rate, a minimum packet error rate, an average packet error rate, a packet loss rate range, a maximum packet loss rate, a minimum packet loss rate, an average packet loss rate, a guaranteed flow bit rate range, a maximum guaranteed flow bit rate, a minimum guaranteed flow bit rate, and an average guaranteed flow bit rate. In some embodiments, the inferred QoS configuration requirement indication includes an inferred QoS configuration requirement, wherein the first radio access network (RAN) node, the second RAN node, the central unit (CU), or the distributed unit (DU) provides all inferred QoS configurations that meet the inferred QoS configuration requirement.

[0610] In some embodiments, the inferred QoE data requirement information includes at least one of the following: an inferred QoE requirement indication, a start time of the inferred QoE data, an end time of the inferred QoE data, a duration of the inferred QoE data, a QoE level identifier, a service type, an area range, a QoE type, a radio access network (RAN) visible QoE data, a QoE report container, a codec rate, a QoE score, a QoE metric range, a maximum QoE metric, a minimum QoE metric, and an average QoE metric, wherein the QoE metric includes at least one of the following: round trip time, jitter duration, impairment duration, throughput, initial playback delay of the video, video resolution, and buffer occupancy level. In some embodiments, the inferred QoE data requirement indication includes the inferred QoE data requirement, wherein the first radio access network (RAN) node, the second RAN node, the central unit (CU), or the distributed unit (DU) provides all inferred QoE data that meets the inferred QoE data requirement.

[0611] In some embodiments, the inferred MDT data requirement information includes at least one of the following: an inferred MDT data requirement indication, a downlink (DL) signal volume measurement result, a power headroom measurement performed by a user equipment (UE), a received interference power measurement, a data volume measurement, an average UE throughput measurement, a packet delay measurement, a packet loss rate measurement, a received signal strength indicator (RSSI) measurement, and a round trip time (RTT) measurement. In some embodiments, the inferred MDT data requirement indication includes an inferred MDT data requirement, wherein the first radio access network (RAN) node, the second RAN node, the central unit (CU), or the distributed unit (DU) provides all inferred MDT data that meets the inferred MDT data requirement.

[0612] In some embodiments, the input data includes at least one of: historical quality of service (QoS) data, historical quality of experience (QoE) data, historical minimization of drive tests (MDT) data, inferred QoS data, inferred QoE data, and inferred MDT data. In some embodiments, each inferred data of the inferred QoS data, inferred QoE data, and inferred MDT data is associated with a confidence requirement. In some embodiments, if the input data is too large or needs to be transmitted periodically, the input data is received in multiple transmissions.

[0613] Fig.10is an exemplary flow chart for requesting feedback. Operation 1002 includes transmitting inferred data by a network node. Operation 1004 includes transmitting a request for feedback associated with the inferred data by the network node. Operation 1006 includes receiving a feedback response by the network node in response to the request. In some embodiments, the method may be implemented according to Embodiment 5. In some embodiments, further steps of performing the method may be based on better system performance than conventional protocols.

[0614] In some embodiments, the inferred data is determined by using an artificial intelligence (AI) function. In some embodiments, the request includes a request for feedback associated with a data type of the inferred data, wherein the data type includes at least one of the following: a quality of service (QoS) type, a quality of experience (QoE) type, and a minimized drive test (MDT) type. In some embodiments, the request includes a request for feedback associated with a relationship level between at least two of the following: a quality of service (QoS) configuration, a quality of experience (QoE) data, and minimized drive test (MDT) data, wherein the relationship level includes at least one of the following: a numerical range, a ratio, and an enumerated quality list.

[0615] In some embodiments, the request includes a request for feedback associated with a level of precision of the inferred data, wherein the level of precision includes at least one of: a numerical range, a ratio, and an enumerated quality list. In some embodiments, the request includes a request for feedback associated with a comparison between prior data prior to using the AI ​​functionality and the inferred data, wherein the comparison includes at least one of: a numerical range, a ratio, and an enumerated quality list.

[0616] In some embodiments, the feedback response includes at least one of the following: a user equipment (UE) quality of service (QoS) configuration list, a UE quality of experience (QoE) information list, a UE minimization of drive tests (MDT) information list, and feedback information. In some embodiments, the feedback information includes a relationship level between at least two of the following: QoS configuration, QoE data, and MDT data, wherein the relationship level includes at least one of the following: a numerical range, a ratio, and an enumerated quality list.

[0617] In some embodiments, the feedback information includes a level of accuracy of the inferred data, wherein the level of accuracy includes at least one of: a numerical range, a ratio, and an enumerated quality list. In some embodiments, the feedback information includes a comparison between previous data before using the AI ​​function and the inferred data, wherein the comparison includes at least one of: a numerical range, a ratio, and an enumerated quality list. In some embodiments, the feedback response is received in multiple transmissions corresponding to different timings.

[0618] In some embodiments, the network node includes one of the following: a core network (CN) node, a radio access network (RAN) node, a central unit (CU) or a distributed unit (DU), wherein the transmission request and reception feedback response are enhanced by one of the following: next generation application protocol (NGAP) enhancement, Xn application protocol (XnAP) enhancement or F1 application protocol (F1AP) enhancement.

[0619] Fig.11 is an exemplary flow chart for transmitting input data. Operation 1102 includes receiving a request for input data by a network node. Operation 1104 includes transmitting the input data by the network node and responding to the request. In some embodiments, the method can be implemented according to embodiments 1 to 4. In some embodiments, further steps of performing the method can be based on better system performance than traditional protocols.

[0620] In some embodiments, the input data is used for an artificial intelligence (AI) function. In some embodiments, the network node is a radio access network (RAN) node, wherein receiving the request comprises receiving the request from an operations, administration, and maintenance (OAM) node or a core network (CN) node, and wherein transmitting the input data comprises transmitting the input data to the OAM node or the CN node.

[0621] In some embodiments, the network node is a first radio access network (RAN) node, wherein receiving the request comprises receiving the request from a second RAN node, and wherein transmitting the input data comprises transmitting the input data to the second RAN node.

[0622] In some embodiments, the network node is a distributed unit (DU), wherein receiving the request comprises receiving the request from a central unit (CU), and wherein transmitting the input data comprises transmitting the input data to the CU.

[0623] In some embodiments, the network node is a central unit (CU), wherein receiving the request comprises receiving the request from a distributed unit (DU), and wherein transmitting the input data comprises transmitting the input data to the DU.

[0624] Fig.12 An exemplary block diagram of a hardware platform 1200 is shown, which may be part of a network device (e.g., a base station, OAM, CN, RAN, CU, or DU) or a communication device (e.g., a user equipment (UE)). The hardware platform 1200 includes at least one processor 1210 and a memory 1205 having instructions stored thereon. The instructions executed by the processor 1210 configure the hardware platform 1200 to perform Figures 1 to 1112 and the operations in various embodiments described in this patent document. Transmitter 1215 transmits or sends information or data to another device. For example, a network device transmitter can send a message to a user equipment. Receiver 1220 receives information or data transmitted or sent by another device. For example, a user equipment can receive a message from a network device. For example, as described in this document, a UE or a network device can be implemented using hardware platform 1200.

[0625] The above-described embodiments are applicable to wireless communications. Fig.13 An example of a wireless communication system (e.g., a 5G or NR cellular network) is shown that includes a base station 1320 and one or more user equipment (UE) 1311, 1312, and 1313. In some embodiments, the UE uses a communication link to the network (sometimes referred to as an uplink direction, as depicted by dashed arrows 1331, 1332, 1333) to access the BS (e.g., the network), which then enables subsequent communication from the BS to the UE (e.g., shown in the direction from the network to the UE, sometimes referred to as a downlink direction, as shown by arrows 1341, 1342, 1343). In some embodiments, the BS sends information to the UE (sometimes referred to as a downlink direction, as depicted by arrows 1341, 1342, 1343), which then enables subsequent communication from the UE to the BS (e.g., shown in the direction from the UE to the BS, sometimes referred to as an uplink direction, as shown by dashed arrows 1331, 1332, 1333). A UE may be, for example, a smartphone, a tablet, a mobile computer, a machine-to-machine (M2M) device, an Internet of Things (IoT) device, etc. The UE described in this document may be communicatively coupled to Fig.13 1320 is depicted in FIG. The UE may also communicate with the BS for CSI communication.

[0626] Those skilled in the art will appreciate that this document discloses methods for requesting and receiving input data that can be used to train AI functions and determine QoS, QoE, and MDT data inferred by AI in wireless communications. This document also discloses methods for requesting and receiving feedback responses associated with AI-inferred data.

[0627] Some embodiments described herein are described in the general context of methods or processes, which can be implemented in one embodiment by a computer program product, which is embodied in a computer-readable medium, including computer-executable instructions, such as program codes, executed by a computer in a network environment. Computer-readable media may include removable and non-removable storage devices, including but not limited to read-only memory (ROM), random access memory (RAM), compact disk (CD), digital versatile disk (DVD), etc. Therefore, computer-readable media can include non-transient storage media. In general, program modules may include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. Computer or processor executable instructions, related data structures, and program modules represent examples of program codes for executing the steps of the methods disclosed herein. A specific sequence of such executable instructions or related data structures represents an example of corresponding behavior for implementing the functions described in these steps or processes.

[0628] In some disclosed embodiments, a device or module that can be implemented using hardware circuits, software, or a combination thereof. For example, a hardware circuit implementation can include discrete analog and / or digital components, such as being integrated as a part of a printed circuit board. Alternatively or additionally, the disclosed components or modules can be implemented as application specific integrated circuits (ASICs) and / or field programmable gate arrays (FPGAs) devices. Some embodiments additionally or alternatively include a digital signal processor (DSP), which is a dedicated microprocessor whose architecture is optimized for the operational requirements of digital signal processing related to the functions disclosed in the present application. Similarly, various components or subcomponents within each module can be implemented in software, hardware, or firmware. The connection between the modules and / or components within the modules can be provided using any of the connection methods and media known in the art, including but not limited to the communication performed by the Internet, wired, or wireless networks using appropriate protocols.

[0629] Although this document contains many details, these details should not be interpreted as limitations on the scope of the claimed invention or the content that can be claimed, but should be interpreted as descriptions of specific features of specific embodiments. Certain features described herein in the context of separate embodiments can also be implemented in combination in a single embodiment. On the contrary, the various features described in the context of a single embodiment can also be implemented individually or in any suitable sub-combination in multiple embodiments. In addition, although the above-mentioned features can be described as working in a specific combination, and even initially required to be so, in some cases one or more features from the required combination can be cut out from the required combination, and the required combination can be for a sub-combination or a variant of a sub-combination. Similarly, although the operations are depicted in a specific order in the accompanying drawings, this should not be understood as requiring the operations to be performed in the specific order or sequential order shown, or requiring the operations of all the diagrams to be performed to achieve the desired results.

[0630] Only some implementations and examples are described herein, and other implementations, improvements and variations can be made based on what is described and illustrated in this disclosure.

Claims

1. A wireless communication method, the method comprising: Requests for input data are transmitted by network nodes; as well as Input data is received by the network node and in response to the request.

2. The method according to claim 1, further comprising: providing, by the network nodes, training of an artificial intelligence (AI) function using the input data; as well as Updated AI functionality is transmitted by the network node based on the training. 3 . The method of claim 1 , further comprising determining, by the network node and based on the input data, inferred data using an artificial intelligence (AI) function.

4. The method according to any one of claims 1 or 2, wherein: The network node is an operations, administration and maintenance (OAM) node or a core network (CN) node, wherein transmitting the request comprises transmitting the request to a radio access network (RAN) node, and wherein receiving the input data comprises receiving the input data from a RAN node.

5. The method according to any one of claims 1 to 3, wherein: The network node is a first radio access network (RAN) node, wherein transmitting the request comprises transmitting the request to a second RAN node, and wherein receiving the input data comprises receiving the input data from the second RAN node.

6. The method according to any one of claims 1 to 3, wherein: The network node is a central unit (CU), wherein transmitting the request comprises transmitting the request to a distributed unit (DU), and wherein receiving the input data comprises receiving the input data from a DU.

7. The method according to claim 3, wherein: The network node is a distributed unit (DU), wherein transmitting the request comprises transmitting the request to a central unit (CU), and wherein receiving the input data comprises receiving the output data from the CU.

8. The method according to any one of claims 1 to 3, wherein: The request includes at least one of: a processing indicator, historical quality of service (QoS) configuration requirement information, historical quality of experience (QoE) data requirement information, historical minimization of drive tests (MDT) data requirement information, inferred QoS configuration requirement information, inferred QoE data requirement information, and inferred MDT data requirement information.

9. The method according to claim 8, wherein: The historical QoS configuration requirement information includes at least one of the following: a historical QoS configuration requirement indication, a 5G QoS identifier (5QI) reference, a QoS level identifier, a service type, a packet delay range, a maximum packet delay, a minimum packet delay, an average packet delay, a packet error rate range, a maximum packet error rate, a minimum packet error rate, an average packet error rate, a packet loss rate range, a maximum packet loss rate, a minimum packet loss rate, an average packet loss rate, a guaranteed flow bit rate range, a maximum guaranteed flow bit rate, a minimum guaranteed flow bit rate and an average guaranteed flow bit rate.

10. The method according to claim 9, wherein: The historical QoS configuration requirement indication includes a historical QoS configuration requirement, and wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU) or a distributed unit (DU) provides all historical QoS configurations that meet the historical QoS configuration requirement.

11. The method according to claim 8, wherein: The historical QoE data requirement information includes at least one of the following: historical QoS data requirement indication, QoE level identifier, service type, area range, QoE type, radio access network (RAN) visible QoE data, QoE report container, codec rate, QoE score, QoE metric range, maximum QoE metric, minimum QoE metric and average QoE metric, and wherein the QoE metric includes at least one of the following: round-trip time, jitter duration, impairment duration, throughput, initial playback delay of video, video resolution and buffer occupancy level.

12. The method according to claim 11, wherein: The historical QoE data requirement indication includes a historical QoE data requirement, and wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU), or a distributed unit (DU) provides all historical QoE data satisfying the historical QoE data requirement.

13. The method according to claim 8, wherein: The historical MDT data requirement information includes at least one of the following: a historical MDT data requirement indication, a downlink (DL) signal amount measurement result, a power headroom measurement performed by a user equipment (UE), a received interference power measurement, a data amount measurement, an average UE throughput measurement, a packet delay measurement, a packet loss rate measurement, a received signal strength indicator (RSSI) measurement, and a round trip time (RTT) measurement.

14. The method according to claim 13, wherein: The historical MDT data requirement indication includes a historical MDT data requirement, and wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU), or a distributed unit (DU) provides all historical MDT data satisfying the historical MDT data requirement.

15. The method according to claim 8, wherein: The inferred QoS configuration requirement information includes at least one of the following: an inferred QoS configuration requirement indication, a 5G QoS identifier (5QI) reference, a QoS level identifier, a service type, a packet delay range, a maximum packet delay, a minimum packet delay, an average packet delay, a packet error rate range, a maximum packet error rate, a minimum packet error rate, an average packet error rate, a packet loss rate range, a maximum packet loss rate, a minimum packet loss rate, an average packet loss rate, a guaranteed flow bit rate range, a maximum guaranteed flow bit rate, a minimum guaranteed flow bit rate and an average guaranteed flow bit rate.

16. The method according to claim 15, wherein: The inferred QoS configuration requirement indication includes an inferred QoS configuration requirement, and wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU) or a distributed unit (DU) provides all inferred QoS configurations that satisfy the inferred QoS configuration requirement.

17. The method according to claim 8, wherein: The inferred QoE data requirement information includes at least one of the following: an inferred QoE requirement indication, a start time of the inferred QoE data, an end time of the inferred QoE data, a duration of the inferred QoE data, a QoE level identifier, a service type, an area range, a QoE type, a radio access network (RAN) visible QoE data, a QoE report container, a codec rate, a QoE score, a QoE metric range, a maximum QoE metric, a minimum QoE metric, and an average QoE metric, and wherein the QoE metric includes at least one of the following: round-trip time, jitter duration, corruption duration, throughput, initial playback delay of the video, video resolution, and buffer occupancy level.

18. The method according to claim 17, wherein: The inferred QoE data requirement indication includes an inferred QoE data requirement, and wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU), or a distributed unit (DU) provides all inferred QoE data satisfying the inferred QoE data requirement.

19. The method according to claim 8, wherein: The inferred MDT data requirement information includes at least one of the following: an inferred MDT data requirement indication, a downlink (DL) signal amount measurement result, a power headroom measurement performed by a user equipment (UE), a received interference power measurement, a data amount measurement, an average UE throughput measurement, a packet delay measurement, a packet loss rate measurement, a received signal strength indicator (RSSI) measurement, and a round trip time (RTT) measurement.

20. The method according to claim 19, wherein: The inferred MDT data requirement indication includes an inferred MDT data requirement, and wherein a first radio access network (RAN) node, a second RAN node, a central unit (CU), or a distributed unit (DU) provides all inferred MDT data satisfying the inferred MDT data requirement.

21. The method according to any one of claims 1 to 3, wherein: The input data includes at least one of the following: historical quality of service (QoS) data, historical quality of experience (QoE) data, historical minimization of drive tests (MDT) data, inferred QoS data, inferred QoE data, and inferred MDT data.

22. The method according to claim 21, wherein: Each of the inferred QoS data, the inferred QoE data, and the inferred MDT data is associated with a confidence requirement.

23. The method according to any one of claims 1 to 3, wherein: If the input data is too large or needs to be transmitted periodically, the input data is received in multiple transmissions.

24. A wireless communication method, the method comprising: The inferred data is transmitted by the network nodes; transmitting, by a network node, a request for feedback associated with the inferred data; as well as A feedback response is received by the network node in response to the request.

25. The method according to claim 24, wherein: The inferred data is determined using artificial intelligence (AI) functionality.

26. The method according to any one of claims 24 or 25, wherein: The request includes a request for feedback associated with a data type of the inferred data, and wherein the data type includes at least one of: a quality of service (QoS) type, a quality of experience (QoE) type, and a minimization of drive tests (MDT) type.

27. The method according to any one of claims 24 or 25, wherein: The request comprises a request for feedback associated with a relationship level among at least two of: quality of service (QoS) configuration, quality of experience (QoE) data, and minimization of drive tests (MDT) data, and wherein the relationship level comprises at least one of: a numerical range, a ratio, and an enumerated quality list.

28. The method according to any one of claims 24 or 25, wherein: The request includes a request for feedback associated with a level of precision of the inferred data, and wherein the level of precision includes at least one of: a numerical range, a ratio, and an enumerated quality list.

29. The method according to claim 25, wherein: The request includes a request for feedback associated with a comparison between previous data prior to using the AI ​​functionality and the inferred data, and wherein the comparison includes at least one of: a numerical range, a ratio, and an enumerated quality list.

30. The method of claim 25, wherein: The feedback response includes at least one of the following: a user equipment (UE) quality of service (QoS) configuration list, a UE quality of experience (QoE) information list, a UE minimization of drive tests (MDT) information list, and feedback information.

31. The method according to claim 30, wherein: The feedback information includes a relationship level among at least two of the following: QoS configuration, QoE data, and MDT data, and wherein the relationship level includes at least one of the following: a numerical range, a ratio, and an enumerated quality list.

32. The method of claim 30, wherein: The feedback information includes a level of accuracy of the inferred data, and wherein the level of accuracy includes at least one of: a numerical range, a ratio, and an enumerated quality list.

33. The method of claim 30, wherein: The feedback information includes a comparison between previous data before using the AI ​​function and the inferred data, and wherein the comparison includes at least one of: a numerical range, a ratio, and an enumerated quality list.

34. The method according to any one of claims 24 or 25, wherein: The feedback response is received in multiple transmissions corresponding to different timings.

35. The method according to any one of claims 24 or 25, wherein: The network node includes one of the following: a core network (CN) node, a radio access network (RAN) node, a central unit (CU) or a distributed unit (DU), and wherein transmitting the request and receiving the feedback response are enhanced by one of the following: next generation application protocol (NGAP) enhancement, Xn application protocol (XnAP) enhancement or F1 application protocol (F1AP) enhancement.

36. A wireless communication method, the method comprising: Receiving, by a network node, a request for input data; as well as Input data is transmitted by the network node and in response to the request.

37. The method of claim 36, wherein: The input data is used for artificial intelligence (AI) functions.

38. The method according to any one of claims 36 or 37, wherein: The network node is a Radio Access Network (RAN) node, wherein receiving the request comprises receiving the request from an Operation, Administration, and Maintenance (OAM) node or a Core Network (CN) node, and wherein transmitting the input data comprises transmitting the input information to an OAM node or a CN node.

39. The method according to any one of claims 36 or 37, wherein: The network node is a first radio access network (RAN) node, wherein receiving the request comprises receiving the requirement from a second RAN node, and wherein transmitting the input data comprises transmitting the input data to the second RAN node.

40. The method according to any one of claims 36 or 37, wherein: The network node is a distributed unit (DU), wherein receiving the request comprises receiving the request from a central unit (CU), and wherein transmitting the input data comprises transmitting the input data to the CU.

41. The method according to any one of claims 36 or 37, wherein: The network node is a central unit (CU), wherein receiving the request comprises receiving the request from a distributed unit (DU), and wherein transmitting the input data comprises transmitting the input information to the DU.

42. An apparatus for wireless communication, the apparatus comprising a processor, wherein: The processor is configured to implement the method according to any one or more of claims 1 to 41 .

43. A computer readable program storage medium having stored thereon codes which, when executed by a processor, cause the processor to implement the method according to any one or more of claims 1 to 41.

Images