Enhanced Wireless Device (WD) Latency Performance Analysis to Support Application Layer Artificial Intelligence / Machine Learning (AI / ML) Operations

Abstract

Disclosed are methods, network data analysis functions (NWDAF), and network functions (NF) for enhanced wireless device (WD) latency performance analysis to support application layer artificial intelligence / machine learning (AI / ML) operations. In one embodiment, the method in the NWDAF includes collecting WD latency performance inputs from a network function (NF), wherein the latency performance inputs include multiple transmit data volume values ​​and transmit time values. The method includes determining a set of latency performance analyses to support a federated learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmissions between the NWDAF and the NF, at least in part based on the latency performance inputs. The method also includes transmitting the set of latency performance analyses to the NF.

Description

Technical Field

[0002] ,

[0005] , , , ,

[0004] , , , ,

[0003] ,

[0001] This disclosure relates to wireless communication, and more particularly to enhanced wireless device (WD) latency performance analysis to support application layer artificial intelligence (AI) / machine learning (ML) operations.

Background Art

[0002] ​​​​​​​​​​​​​​​​​Note 1: Specific input and output parameters for latency performance analysis within a group of WDs to support application AI / ML FL operation will be determined during the normative phase.

[0006] Analysis ID: WD Latency Performance in CR#0616 (to be added to 3GPP TS23.288) general Section 2.1.2.1 of 3GPP Technical Standard (TS) 23.288 describes how NWDAF may provide service consumers with WD latency performance analysis in the form of statistics, forecasts, or both. NWDAF collects WD latency performance relational input data from 5G core network functions (NF), operational administration and maintenance (OAM), and application functions (AF). Consumers may either subscribe to analysis notifications (i.e., a subscribe-notification model) or request a single notification (i.e., a request-response model).

[0007] WD latency performance refers to the time delay required to complete the transmission of a specific data volume from the WD to the AF or from the AF to the WD. Given the expected number of data transmissions to be repeated, or the expected time interval between data transmissions, WD latency performance can be provided as the average value of the data packet transmission latency performance within the analysis target period. WD latency performance analysis can be used, for example, to assist an AF hosting an AI / ML-based service for member selection in federated learning.

[0008] WD latency performance analysis may be provided individually for each WD or for a list of WDs, as specified in section 2.1.2.3 of 3GPP 23.288.

[0009] A service consumer can be an NF (for example, an AF, or a Network Exposure Function (NEF)).

[0010] Consumers of these analyses may specify the following in their requests: - Analysis ID = "WD Latency Performance", - Target of analysis report: Single WD (Subscription Persistence Identifier (SUPI) / Public Subscription Identifier (GPSI)) or group of WD (list of SUPI / GPSIs) - Optionally include the following analysis filter information: - Data Network Name (DNN), - S-Network Slice Selection Assistance Information (NSSAI), - Application ID, - Area of ​​Interest (AOI): Limits the scope of the WD latency performance analysis to the provided area. - An optional list of required analysis subsets (see section 2.1.2.3 of 3GPP TS23.288), - Data Volume Uplink / Downlink (UL / DL): Specifies a specific data volume that has been sent once from WD to AF or from AF to WD. - Quality of Service (QoS) requirements (e.g., 5G QoS identifier (5QI), QoS characteristics), - Optionally, the expected number of times data transmission will be repeated within the analysis target period. - Optionally, the expected time interval between data transmissions. - Optionally, a request regarding the geographical distribution of WDs (i.e., areas of interest (AoI)), - The analysis target period indicates the time period for which statistics or forecasts are required. - During the subscription, the notification correlation ID and notification target address are included. - At your discretion, the desired level of accuracy in the analysis, - Any preferred order of results for the list of WD latency performance, - Ranking criteria: "WD Latency Performance", - Order: Ascending or descending order, - Optional, reporting thresholds. This indicates the conditions regarding the level that should be reached for each subset of analysis in order to be notified by NWDAF (see Section 2.1.2.3 of 3GPP TS23.288), for example, NWDAF may provide the percentage of WDs that have reached a certain reporting threshold (one or more). - The maximum number of WDs, at your discretion.

[0011] Input data NWDAF, which supports analysis of WD latency performance, should be able to collect WD performance information from AF, Operational Administration and Maintenance (OAM), and 5GC NF.

[0012] More detailed information collected by the NWDAF from the OAM is specified in Table 2.1.2.2-1, and more detailed information collected by the NWDAF from the relevant 5GC NFs (i.e., User Plane Function (UPF), Session Management Function (SMF), and Access and Mobility Management Function (AMF)) is specified in Table 2.1.2.2-2. TIFF2026508989000002.tif249170

[0013] NWDAF subscribes to network data from OAM in Table 2.1.2.2-1 by using the services provided by OAM, as described in Section 6.2.3 of 3GPP TS23.288.

[0014] Note 1: Whether a WD (one or more) supports a slice can be checked by retrieving details of the registered Access and Mobility Management Function (AMF) from the Unified Data Management (UDM), or by querying the AMF regarding which slices are used by the WD (one or more) in the current registration. (Alternatively, if a Network Slice Admission Control Function (NSACF) is deployed, the NSACF may provide a report on which slices are used by the WD (one or more).)

[0015] Note 2: User commitment checks from the UDM may be applied for these analyses. TIFF2026508989000003.tif88170

[0016] Output analysis The NWDAF that supports the WD latency performance analysis provides the analysis results to the consumer NF, e.g., AF or NEF. The analysis results provided by the NWDAF can be the WD latency performance statistics defined in Table 2.1.2.3-1 or the predictions defined in Table 2.1.2.3-2. TIFF2026508989000004.tif255167TIFF2026508989000005.tif43170TIFF2026508989000006.tif255167TIFF2026508989000007.tif78170

[0017] An exemplary procedure for the WD latency performance analysis is shown in Figure 1.

[0018] Procedure The NWDAF may provide the WD latency performance analysis to a 5GC NF (e.g., AF or NEF). 1. A consumer NF, e.g., AF or NEF, requests an analysis for the WD latency performance from the NWDAF (in some cases, if the consumer NF is AF, via the NEF) or subscribes to an analysis for the WD latency performance from the NWDAF and provides the input information specified in 2.1.2.1 to the 5GC. 2a - b. The NWDAF subscribes to the service data from the AMF in Table 2.1.2.2-2 using the Namf_EventExposure_Subscribe service to collect the (one or more) WD locations for the WD or group of WDs. Note: If the NWDAF requires WD location information with a finer granularity than TA / cell, the NWDAF collects location data from the GMLC instead of the AMF. 2c. The NWDAF subscribes to service data from the SMF in Table 2.1.2.2-2 by invoking Nsmf_EventExposure_Subscribe (event ID, (one or more) SUPIs or application IDs). To provide the required analysis, the NWDAF may subscribe to WD information and to N4 session relationship input data from the SMF as defined in Table 2.1.2.2-2. 2d~e. The N4 relationship input data is provided to the SMF by the UPF. 2f. The SMF provides the required input data to the NWDAF. 2g~h. The NWDAF may subscribe to input data from the OAM in Table 2.1.2.2-1 according to the data collection principle from the OAM described in section 6.2.3 of 3GPP TS23.288. 3. The NWDAF derives the required analysis in the form of latency performance statistics or latency performance prediction or both. 4. Depending on the service used in step 1, the NWDAF uses either the Nnwdaf_AnalyticsInfo_Request response or the Nnwdaf_AnalyticsSubscription_Notify to provide the required latency performance analysis to the NF, and 5~7. If the NF subscribed to WD latency performance analysis in step 1, when the NWDAF generates a new analysis, the NWDAF notifies the consumer of the newly generated analysis.

[0019] However, in the current version of the new analysis ID (WD latency performance), specific properties of the AI / ML traffic (for federated learning) are not considered.

[0020] A new analysis ID for WD latency performance was proposed at the SA2#154 ad hoc electronic meeting to support application layer federative learning (FL) behavior, as described in the approved CR#0616 (which will be added to 3GPP TS23.288).

[0021] The latency performance of a WD group is output information from a WD latency performance analysis used to evaluate the performance of a group of WDs to support associative learning. The motivation for introducing WD latency performance analysis is to provide latency relationship performance of WD groups to support member selection in associative learning for AI / ML services, in particular in this release. The performance of associative learning is based on the quality of service of the WD group. Therefore, output based on the performance of the WD group is essential for NEF and (in this release) AF to select WD members and evaluate WD performance during model training. Statistics and predictions of the WD group performance should be performed by NWDAF and provided to the consumer.

[0022] The nature of AI / ML traffic (for example, for federative learning) is extremely different from other normal traffic, such as retransmitting the same AI / ML data or sending different AI / ML data at multiple time intervals with and without ultra-low latency requirements, different data volumes of various AI / ML data to be sent, and different transmission conditions for sending AI / ML data between different network entities. However, the new analysis ID does not take into account the distinction between the specific nature of AI / ML traffic and other normal traffic, which can lead to analysis results being unusable to effectively support application layer AI / ML operations (e.g., FL). [Overview of the Initiative]

[0023] Some embodiments advantageously provide methods and network nodes for enhanced wireless device (WD) latency performance analysis to support application layer artificial intelligence (AI) / machine learning (ML) operations.

[0024] In some embodiments, specific properties of the AI / ML traffic for federative learning are considered along with latency relationship analysis of the AI / ML traffic. New inputs and outputs are added to the analyzed ID WD latency performance.

[0025] In some embodiments, the nature of the AI / ML traffic for federative learning is considered. These may include one or more of the following: • Existing inputs are extended, for example, transmitted data volume and transmission time. • New inputs are added to the analysis ID WD latency performance, including round-trip related time, paired timestamps (start and end times) (list), IP filter information, application location, application server instance address, and / or • New outputs will be added, including the following: - Estimated number of transmissions (statistics / forecast), - Statistics / forecasts regarding UL latency for target transmissions. - Statistics / predictions regarding DL latency for target transmission. - Statistics / predictions regarding round-trip latency for target transmission, and / or, - Statistics / predictions regarding maximum latency for target transmission.

[0026] Some embodiments, for example, take into account the specific nature of AI / ML traffic for federative learning (FL) and focus on latency relationship analysis with respect to AI / ML traffic. Existing inputs are extended and new inputs and outputs are added to the analysis ID WD latency performance to enable these analyses to support application layer AI / ML operations (e.g., federative learning).

[0027] Analysis IDs for WD latency performance were considered at the SA2#154 ad-hoc electronic meeting to support application layer federated learning operations. However, the new analysis IDs do not take into account the distinction between certain characteristics of AI / ML traffic (e.g., for FL) and other normal traffic. This may result in the analysis results not being used to effectively support application layer AI / ML operations (e.g., FL).

[0028] According to one embodiment, a method is provided in a core network node configured to include a Network Data Analysis Function (NWDAF). The method includes collecting a Radio Device (WD) latency performance input from a Network Function (NF), wherein the latency performance input includes a plurality of transmit data volume values ​​and transmit time values. The method includes determining a set of latency performance analyses to support a federated learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmit latency between the WD and the Application Function (AF), at least in part based on the latency performance input. The method also includes transmitting the set of latency performance analyses to the NF.

[0029] In some embodiments, the latency prediction and latency statistics include a time delay to complete the transmission of a target volume of data from the WD to the NF or from the NF to the WD. In some embodiments, the latency prediction and latency statistics include an estimated number of retransmissions to complete the transmission of a target volume of data. In some embodiments, the latency prediction and latency statistics include at least one of uplink packet delay, downlink packet delay, and round-trip packet delay. In some embodiments, the latency prediction and latency statistics include a value averaged over the analysis target period. In some embodiments, the latency prediction and latency statistics include a valid period for latency performance analysis. In some embodiments, the latency prediction and latency statistics include the maximum packet delay observed for communicating with the NF. In some embodiments, the latency prediction and latency statistics include a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied. In some embodiments, at least one of the latency prediction and latency statistics is grouped into latency classes according to latency performance ranges. In some embodiments, at least one of the latency prediction and latency statistics includes an indication of the percentage of WD in each latency class. In some embodiments, NF is implemented as an application function.

[0030] In another embodiment, a core network node is provided that is configured to include a Network Data Analysis Function (NWDAF). The core network node is configured to collect radio device (WD) latency performance input from a network function (NF), wherein the latency performance input includes multiple transmit data volume values ​​and transmit time values. The core network node is configured to determine a set of latency performance analyses to support a federated learning process, wherein the set of latency performance analyses includes, at least in part, a latency prediction and a latency statistic relating to uplink and downlink transmit latency between the WD and the application function (AF). The core network node is also configured to transmit the set of latency performance analyses to the NF.

[0031] In some embodiments, the latency prediction and latency statistics include a time delay to complete the transmission of a target volume of data from the WD to the NF or from the NF to the WD. In some embodiments, the latency prediction and latency statistics include an estimated number of retransmissions to complete the transmission of a target volume of data. In some embodiments, the latency prediction and latency statistics include at least one of uplink packet delay, downlink packet delay, and round-trip packet delay. In some embodiments, the latency prediction and latency statistics include a value averaged over the analysis target period. In some embodiments, the latency prediction and latency statistics include a valid period for latency performance analysis. In some embodiments, the latency prediction and latency statistics include the maximum packet delay observed for communicating with the NF. In some embodiments, the latency prediction and latency statistics include a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied. In some embodiments, at least one of the latency prediction and latency statistics is grouped into latency classes according to latency performance ranges. In some embodiments, at least one of the latency prediction and latency statistics includes an indication of the percentage of WD in each latency class. In some embodiments, NF is implemented as an application function.

[0032] In another embodiment, a method is provided for a network node configured to include a network function (NF) and to communicate with a network data analysis function (NWDAF). The method includes transmitting a latency performance input to the NWDAF, wherein the latency performance input includes a plurality of transmit data volume values ​​and transmit time values. The method also includes receiving a set of latency performance analyses to support a federate learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmit latency between the NWDAF and the NF, at least in part based on the latency performance input. The method also includes hosting an artificial intelligence / machine learning (AI / ML) based service at least in part based on a federate learning process.

[0033] In some embodiments, the latency prediction and latency statistics include a time delay to complete the transmission of a target volume of data from the WD to the NF or from the NF to the WD. In some embodiments, the latency prediction and latency statistics include an estimated number of retransmissions to complete the transmission of a target volume of data. In some embodiments, the latency prediction and latency statistics include at least one of uplink packet delay, downlink packet delay, and round-trip packet delay. In some embodiments, the latency prediction and latency statistics include a value averaged over the analysis target period. In some embodiments, the latency prediction and latency statistics include a valid period for latency performance analysis. In some embodiments, the latency prediction and latency statistics include the maximum packet delay observed for communicating with the NF. In some embodiments, the latency prediction and latency statistics include a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied. In some embodiments, at least one of the latency prediction and latency statistics is grouped into latency classes according to latency performance ranges. In some embodiments, at least one of the latency prediction and latency statistics includes an indication of the percentage of WD in each latency class. In some embodiments, NF is implemented as an application function.

[0034] In another embodiment, a network node is provided that is configured to include a network function (NF) and to communicate with a network data analysis function (NWDAF). The network node is configured to transmit latency performance inputs to the NWDAF, wherein the latency performance inputs include multiple transmit data volume values ​​and transmit time values. The network node is also configured to receive a set of latency performance analyses to support a federative learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmit latency between the NWDAF and the NF, at least in part based on the latency performance inputs. The network node is further configured to host artificial intelligence / machine learning (AI / ML) based services, at least in part based on the federative learning process.

[0035] In some embodiments, the latency prediction and latency statistics include a time delay to complete the transmission of a target volume of data from the WD to the NF or from the NF to the WD. In some embodiments, the latency prediction and latency statistics include an estimated number of retransmissions to complete the transmission of a target volume of data. In some embodiments, the latency prediction and latency statistics include at least one of uplink packet delay, downlink packet delay, and round-trip packet delay. In some embodiments, the latency prediction and latency statistics include a value averaged over the analysis target period. In some embodiments, the latency prediction and latency statistics include a valid period for latency performance analysis. In some embodiments, the latency prediction and latency statistics include the maximum packet delay observed for communicating with the NF. In some embodiments, the latency prediction and latency statistics include a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied. In some embodiments, at least one of the latency prediction and latency statistics is grouped into latency classes according to latency performance ranges. In some embodiments, at least one of the latency prediction and latency statistics includes an indication of the percentage of WD in each latency class. In some embodiments, NF is implemented as an application function.

[0036] When considered in conjunction with the attached drawings, a more complete understanding of these embodiments, as well as their associated advantages and features, will be more readily apparent by referring to the following detailed description. [Brief explanation of the drawing]

[0037] [Figure 1] This is a diagram showing the latency performance analysis of the procedure WD. [Figure 2]This is a schematic diagram of an exemplary network architecture illustrating a communication system connected to a host computer via an intermediate network, based on the principles described herein. [Figure 3] This is a block diagram of a host computer communicating with a wireless device via a network node, at least partially over a wireless connection, according to some embodiments of the present disclosure. [Figure 4] This flowchart illustrates an exemplary method implemented in a communication system including a host computer, a network node, and a wireless device for running a client application on a wireless device, according to some embodiments of the present disclosure. [Figure 5] This flowchart illustrates an exemplary method implemented in a communication system including a host computer, a network node, and a wireless device for receiving user data in a wireless device, according to some embodiments of the present disclosure. [Figure 6] This flowchart illustrates an exemplary method implemented in a communication system including a host computer, a network node, and a wireless device for receiving user data from a wireless device on a host computer, according to some embodiments of the present disclosure. [Figure 7] This flowchart illustrates an exemplary method implemented in a communication system including a host computer, a network node, and a wireless device for receiving user data on a host computer, according to some embodiments of the present disclosure. [Figure 8] This is an illustrative process flowchart for network node latency performance analysis of Enhanced Wireless Devices (WD) to support application layer artificial intelligence (AI) / machine learning (ML) operations. [Figure 9] This is an exemplary process flowchart for a network node configured to include a network data analysis function (NWDAF) according to the principles disclosed herein. [Figure 10]This is a flowchart illustrating an exemplary process in a network node configured to include a network function (NF) and to communicate with a network data analysis function (NWDAF), according to the principles disclosed herein. [Modes for carrying out the invention]

[0038] Before describing exemplary embodiments in detail, it should be noted that embodiments primarily exist as combinations of apparatus components and processing steps relating to enhanced wireless device (WD) latency performance analysis for supporting application layer artificial intelligence (AI) / machine learning (ML) operations. Accordingly, components are represented by conventional symbols in the drawings where appropriate, and only their specific details relevant to understanding the embodiments are shown, so as not to obscure this disclosure with details that would be readily apparent to those skilled in the art who are interested in the description herein. Similar numbers refer to similar elements throughout the description.

[0039] As used herein, relational terms such as “first” and “second,” “upper” and “lower” may be used simply to distinguish one entity or element from another, without necessarily requiring or implying any physical or logical relationship or order between such entities or elements. The terminology used herein is for the purpose of describing specific embodiments and does not limit the concepts described herein. As used herein, the singular forms “a,” “an,” and “the” also include the plural form unless the context otherwise explicitly indicates. Furthermore, as used herein, the terms “comprises,” “comprising,” “includes,” and / or “including” specify the presence of the described feature, complete, step, action, element, and / or component, but do not exclude the presence or addition of one or more other features, complete, step, action, element, component, and / or groups thereof.

[0040] In the embodiments described herein, joining terms such as “in communication with” may be used to indicate electrical or data communication, which can be achieved, for example, by physical contact, induction, electromagnetic radiation, radio signaling, infrared signaling, or optical signaling. Those skilled in the art will understand that multiple components can interact with each other, and that modifications and variations are possible for achieving electrical and data communication.

[0041] In some embodiments described herein, terms such as “coupled” and “connected” may be used herein to indicate a connection, though not necessarily directly, and may include wired and / or wireless connections.

[0042] As used herein, the term “network node” can refer to any type of network node present in a radio network, which may further comprise any of the following: base stations (BS), radio base stations, base transceiver stations (BTS), base station controllers (BSC), radio network controllers (RNC), g-node B (gNB), evolved node B (eNB or e-node B), node B, MSR radio nodes such as multi-standard radio (MSR) BS, multi-cell / multicast cooperative entities (MCE), radio access backhaul integrated transmission (IAB) nodes, relay nodes, donor node control relays, radio access points (AP), transmit points, transmit nodes, remote radio units (RRU), remote radio heads (RRH), core network nodes (e.g., mobile management entities (MME), self-organizing network (SON) nodes, cooperative nodes, positioning nodes, MDT nodes, etc.), external nodes (e.g., third-party nodes, nodes outside the current network), nodes in distributed antenna systems (DAS), spectrum access system (SAS) nodes, element management systems (EMS), etc. Network nodes may also include test equipment. The term “wireless node” as used herein may also be used to refer to wireless devices (WDs) or wireless network nodes, etc.

[0043] In some embodiments, the non-limiting terms "wireless device (WD)" and "user equipment (UE)" are used interchangeably. A WD as used herein can be any type of wireless device capable of communicating with a network node or another WD via wireless signals, such as a wireless device (WD). A WD can also be a wireless communication device, a target device, a D2D (device to device) WD, a machine-type WD or a WD capable of machine-to-machine communication (M2M), a low-cost and / or low-complexity WD, a sensor equipped with a WD, a tablet, a mobile terminal, a smartphone, a laptop embedded equipment (LEE), a laptop mounted equipment (LME), a USB dongle, customer premises equipment (CPE), an Internet of Things (IoT) device, or a narrowband IoT (NB-IoT) device.

[0044] In some embodiments, the general term “wireless network node” is used. A wireless network node can be any type of wireless network node, which may comprise any of the following: base stations, wireless base stations, base station transceiver stations, base station controllers, network controllers, RNCs, evolved node B (eNB), node B, gNB, multicell / multicast cooperative entity (MCE), IAB node, relay node, access point, wireless access point, remote radio unit (RRU), or remote radio head (RRH).

[0045] This disclosure may use terminology from a specific radio system, such as 3GPP LTE and / or New Radio (NR), but it should be noted that this should not be considered to limit the scope of this disclosure to the aforementioned systems only. However, other radio systems, including Wideband Code Division Multiple Access (WCDMA), Global Interoperability for Microwave Access (WiMAX), Ultra Mobile Broadband (UMB), and GSM (Global System for Mobile Communications), may also benefit from leveraging the ideas covered within this disclosure.

[0046] It should be further noted that the functions described herein as being performed by wireless devices or network nodes may be distributed across multiple wireless devices and / or network nodes. In other words, the functions of network nodes and wireless devices described herein are not limited to being performed by a single physical device, but can actually be distributed across several physical devices.

[0047] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as they would ordinarily be understood by those skilled in the art to which this disclosure belongs. Terms used herein should be interpreted as having the meanings of those terms in the context of this specification and the related art, and not in an ideal or overly formal sense unless explicitly specified herein.

[0048] Some embodiments provide enhanced wireless device (WD) latency performance analysis to support application layer artificial intelligence (AI) / machine learning (ML) operations.

[0049] Referring again to the drawings, similar elements are referenced by similar reference numbers, and Figure 2 shows a schematic diagram of a communication system 10, such as a 3GPP type cellular network capable of supporting standards such as LTE and / or NR (5G), comprising an access network 12, such as a wireless access network, and a core network 14, according to one embodiment. The access network 12 comprises several network nodes 16a, 16b, 16c (collectively referred to as network nodes 16), such as NBs, eNBs, gNBs, or other types of wireless access points, each defining a corresponding coverage area 18a, 18b, 18c (collectively referred to as a coverage area 18). Each network node 16a, 16b, 16c can connect to the core network 14 via a wired or wireless connection 20. A first wireless device (WD) 22a located in a coverage area 18a is configured to wirelessly connect to a corresponding network node 16a or to be paged by a corresponding network node 16a. A second WD22b in coverage area 18b can wirelessly connect to the corresponding network node 16b. Although multiple WDs 22a, 22b (collectively referred to as wireless device 22) are shown in this example, the disclosed embodiments are equally applicable to situations where only one WD is in the coverage area, or where only one WD is connected to the corresponding network node 16. For convenience, only two WDs 22 and three network nodes 16 are shown, but it should be noted that the communication system may include more WDs 22 and network nodes 16.

[0050] Furthermore, it is conceivable that WD22 may be configured to communicate simultaneously with two or more network nodes 16 and two or more types of network nodes 16, as well as / or to communicate with them separately. For example, WD22 may have dual connectivity with a network node 16 that supports LTE and the same or different network nodes 16 that support NR. As an example, WD22 may communicate with an eNB for LTE / E-UTRAN and a gNB for NR / NG-RAN.

[0051] The communication system 10 may itself be connected to a host computer 24, which may be embodied in the hardware and / or software of a standalone server, a cloud implementation server, a distributed server, or as a processing resource in a server farm. The host computer 24 may be owned or controlled by a service provider, or may be operated by or on behalf of a service provider. Connections 26, 28 between the communication system 10 and the host computer 24 may extend directly from the core network 14 to the host computer 24, or may extend via an optional intermediate network 30. The intermediate network 30 may be one of a public network, a private network, or a hosted network, or a combination of two or more of these. The intermediate network 30 may be a backbone network or the internet, if any. In some embodiments, the intermediate network 30 may comprise two or more subnets (not shown).

[0052] The communication system in Figure 2, as a whole, enables connectivity between one of the connected WD22a, 22b and the host computer 24. The connectivity can be described as an over-the-top (OTT) connection. The host computer 24 and the connected WD22a, 22b are configured to communicate data and / or signaling over the OTT connection, using the access network 12, the core network 14, an optional intermediate network 30 and possible further infrastructure (not shown) as intermediaries. The OTT connection can be transparent in the sense that at least some of the participating communication devices through which the OTT connection passes are unaware of the routing of uplink and downlink communications. For example, network node 16 may not be aware of, or does not need to be aware of, the past routing of incoming downlink communications with data originating from the host computer 24 that should be forwarded (e.g., handed over) to the connected WD22a. Similarly, network node 16 does not need to be aware of the future routing of outgoing uplink communications originating from WD22a and destined for host computer 24.

[0053] Network node 16 may be configured to include an AI / ML unit which can be configured to implement a federative learning process including multiple inputs of artificial intelligence / machine learning (AI / ML), where the multiple inputs include transmitted data information. Network node 16 may be configured to include NF32 which can be configured to host an AI / ML-based service based at least partially on the federative learning process. Core network node 36 in core network 14 may include NWDAF34, and network node 16 may include NF32 which receives latency performance analysis from core network node 36. NWDAF34 may be configured to determine a set of latency performance analyses to support the federative learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between WD22 and application functions (AF), based at least partially on latency performance inputs.

[0054] Next, an exemplary implementation of the WD22, network node 16, host computer 24, and core network node 36 described in the previous paragraph, according to one embodiment, will be described with reference to Figure 3. In the communication system 10, the host computer 24 comprises hardware (HW) 38, including a communication interface 40 configured to set up and maintain wired or wireless connections to the interfaces of different communication devices of the communication system 10. The host computer 24 further comprises a processing circuit 42 which may have memory and / or processing capabilities. The processing circuit 42 may include a processor 44 and memory 46. In particular, in addition to or instead of a processor and memory such as a central processing unit, the processing circuit 42 may comprise integrated circuits for processing and / or control, such as one or more processors and / or processor cores and / or FPGAs (field-programmable gate arrays) and / or ASICs (application-specific integrated circuits) adapted to execute instructions. The processor 44 may be configured to access memory 46 (for example, to write to memory 46 and / or read from memory 46), and memory 46 may include any kind of volatile and / or non-volatile memory, such as cache and / or buffer memory and / or RAM (random access memory) and / or ROM (read-only memory) and / or optical memory and / or EPROM (erasable programmable read-only memory).

[0055] The processing circuit 42 may be configured to control any of the methods and / or processes described herein, and / or to cause such methods and / or processes to be carried out, for example, by the host computer 24. The processor 44 corresponds to one or more processors 44 for carrying out the host computer 24 functions described herein. The host computer 24 includes memory 46 configured to store data, programmatic software code, and / or other information described herein. In some embodiments, the software 48 and / or host application 50 may include instructions that, when executed by the processor 44 and / or processing circuit 42, cause the processor 44 and / or processing circuit 42 to carry out the processes described herein with respect to the host computer 24. The instructions may be software related to the host computer 24.

[0056] Software 48 may be executable by processing circuit 42. Software 48 includes a host application 50. The host application 50 may be able to operate to provide services to remote users, such as a WD22 connected via an OTT connection 52 that terminates at the host computer 24. When providing services to remote users, the host application 50 may provide user data transmitted using the OTT connection 52. "User data" may be data and information as described herein as implementing the functions described. In one embodiment, the host computer 24 may be configured to provide control and functionality to a service provider and may be operated by or on behalf of the service provider. The processing circuit 42 of the host computer 24 may enable the host computer 24 to observe, monitor, and control the network node 16 and / or wireless device 22, transmit to the network node 16 and / or wireless device 22, and / or receive from the network node 16 and / or wireless device 22.

[0057] The communication system 10 further includes a network node 16 provided within the communication system 10, the network node 16 including hardware 58 that enables the network node 16 to communicate with the host computer 24 and WD22. The hardware 58 may include a communication interface 60 for setting up and maintaining wired or wireless connections with the interfaces of different communication devices of the communication system 10, and a wireless interface 62 for setting up and maintaining at least a wireless connection 64 with WD22 located in the coverage area 18 served by the network node 16. The wireless interface 62 may be formed as, for example, one or more RF transmitters, one or more RF receivers, and / or one or more RF transceivers, or may include them. The communication interface 60 may be configured to facilitate a connection 66 to the host computer 24. The connection 66 may be direct, or the connection 66 may pass through the core network 14 of the communication system 10 and / or one or more intermediate networks 30 outside the communication system 10.

[0058] In the embodiments shown, the hardware 58 of the network node 16 further includes a processing circuit 68. The processing circuit 68 may include a processor 70 and a memory 72. More specifically, in addition to, or instead of, a processor and memory such as a central processing unit, the processing circuit 68 may include an integrated circuit for processing and / or control, for example, one or more processors and / or processor cores and / or FPGAs (field-programmable gate arrays) and / or ASICs (application-specific integrated circuits) adapted to execute instructions. The processor 70 may be configured to access the memory 72 (e.g., write to and / or read from the memory 72), and the memory 72 may include any kind of volatile and / or non-volatile memory, for example, cache and / or buffer memory and / or RAM (random access memory) and / or ROM (read-only memory) and / or optical memory and / or EPROM (erasable programmable read-only memory).

[0059] Therefore, the network node 16 further has software 74 stored either internally in memory 72 or in external memory (e.g., a database, storage array, network storage device, etc.) accessible by the network node 16 via an external connection. The software 74 may be executable by processing circuit 68. Processing circuit 68 may be configured to control any of the methods and / or processes described herein, and / or to cause such methods and / or processes to be carried out by the network node 16, for example. Processor 70 corresponds to one or more processors 70 for carrying out the network node 16 functions described herein. Memory 72 is configured to store data, programmatic software code, and / or other information described herein. In some embodiments, the software 74 may include instructions that, when executed by the processor 70 and / or processing circuit 68, cause the processor 70 and / or processing circuit 68 to carry out the processes described herein with respect to the network node 16. For example, the processing circuit 68 of the network node 16 may include an AI / ML unit that can be configured to implement a federated learning process involving multiple inputs of artificial intelligence / machine learning (AI / ML), where the multiple inputs include transmit data information. NF32 may be configured to host an AI / ML-based service based at least partially on the federated learning process. In some embodiments, NWDAF34 may also be configured in the network node 16.

[0060] The communication system 10 further includes the WD22 already mentioned. The WD22 may have hardware 80 which may include a radio interface 82 configured to set up and maintain a radio connection 64 with a network node 16 serving the coverage area 18 in which the WD22 is currently located. The radio interface 82 may be formed as, for example, one or more RF transmitters, one or more RF receivers, and / or one or more RF transceivers, or may include them.

[0061] The WD22 hardware 80 further includes a processing circuit 84. The processing circuit 84 may include a processor 86 and memory 88. More specifically, in addition to, or instead of, a processor and memory such as a central processing unit, the processing circuit 84 may include an integrated circuit for processing and / or control, such as one or more processors and / or processor cores and / or FPGAs (field-programmable gate arrays) and / or ASICs (application-specific integrated circuits) adapted to execute instructions. The processor 86 may be configured to access memory 88 (e.g., write to memory 88 and / or read from memory 88), and memory 88 may include any kind of volatile and / or non-volatile memory, such as cache and / or buffer memory and / or RAM (random access memory) and / or ROM (read-only memory) and / or optical memory and / or EPROM (erasable programmable read-only memory).

[0062] Therefore, the WD22 may further include software 90, which may be stored, for example, in memory 88 in the WD22 or in external memory accessible by the WD22 (e.g., a database, storage array, network storage device, etc.). The software 90 may be executable by processing circuit 84. The software 90 may include a client application 92. The client application 92 may operate to provide services to human or non-human users via the WD22, with the support of a host computer 24. On the host computer 24, a running host application 50 may communicate with a running client application 92 via an OTT connection 52 that terminates in the WD22 and the host computer 24. When providing services to a user, the client application 92 may receive request data from the host application 50 and provide user data in response to the request data. The OTT connection 52 may transfer both the request data and the user data. The client application 92 may interact with the user to generate the user data that the client application 92 provides.

[0063] Processing circuit 84 may be configured to control any of the methods and / or processes described herein, and / or to cause such methods and / or processes to be carried out, for example, by WD22. Processor 86 corresponds to one or more processors 86 for carrying out the WD22 functions described herein. WD22 includes memory 88 configured to store data, programmatic software code, and / or other information described herein. In some embodiments, software 90 and / or client application 92 may include instructions that, when executed by processor 86 and / or processing circuit 84, cause processor 86 and / or processing circuit 84 to carry out the processes described herein with respect to WD22.

[0064] The core network node 36 may include processing circuitry 94, which includes a processor and memory (not shown). The processor may include a central processing unit and / or processing circuitry, which includes one or more processors and / or processor cores and / or FPGAs (Field Programmable Gate Arrays) and / or ASICs (Application-Specific Integrated Circuits) adapted to perform processing and / or control. The processor may be configured to access memory (e.g., write to memory and / or read from memory), and the memory may include any kind of volatile and / or non-volatile memory, such as cache and / or buffer memory and / or RAM (Random Access Memory) and / or ROM (Read-Only Memory) and / or optical memory and / or EPROM (Erasable Programmable Read-Only Memory).

[0065] Therefore, the core network node 36 may further include software stored in memory on the core network node or in external memory accessible by the core network node (e.g., a database, storage array, network storage device, etc.). The software may be executable by the processing circuit 94 to perform the functions of the NWDAF as described herein. The processing circuit 94 may be configured to include an NWDAF 34 which can be configured to determine a set of latency performance analyses to support the federative learning process. In some embodiments, NF 32 may also be configured on the core network node 36.

[0066] The core network node 36 may also include a radio interface 96 configured to set up and maintain a radio connection with the network node 16 serving the coverage area 18 where the WD22 is currently located. The radio interface 96 may be formed as, for example, one or more RF transmitters, one or more RF receivers, and / or one or more RF transceivers, or may include them. In some embodiments, the internal workings of the network node 16, WD22, host computer 24, and core network node 36 may be as shown in Figure 3, and separately, the surrounding network topology may be as shown in Figure 2.

[0067] In Figure 3, the OTT connection 52 is depicted abstractly to illustrate communication between the host computer 24 and the wireless device 22 via the network node 16, without explicit reference to the intermediary devices and the precise routing of messages through these devices. The network infrastructure may determine the routing, and the network infrastructure may be configured to hide the routing from the WD22, the service provider operating the host computer 24, or both. While the OTT connection 52 is active, the network infrastructure may also make decisions to dynamically change the routing (for example, based on network load balancing considerations or reconfiguration).

[0068] The wireless connection 64 between WD22 and network node 16 follows the teachings of embodiments described throughout this disclosure. One or more of the various embodiments improve the performance of OTT services provided to WD22 using an OTT connection 52 in which the wireless connection 64 may form the final segment. More precisely, some teachings of these embodiments may improve data rate, latency, and / or power consumption, thereby providing benefits such as reduced user latency, relaxed file size limits, better responsiveness, and extended battery life.

[0069] In some embodiments, a measurement procedure may be provided for the purpose of monitoring data rate, latency, and other factors, which one or more embodiments improve. Further optional network functions may be provided for reconfiguring the OTT connection 52 between the host computer 24 and the WD22 in response to variations in the measurement results. The measurement procedure and / or network function for reconfiguring the OTT connection 52 may be implemented in the software 48 of the host computer 24, or in the software 90 of the WD22, or both. In embodiments, a sensor (not shown) may be deployed in or in relation to a communication device through which the OTT connection 52 passes, and the sensor may participate in the measurement procedure by supplying values ​​of the monitored quantities exemplified above, or values ​​of other physical quantities through which the software 48, 90 can calculate or estimate the monitored quantities. Reconfiguring the OTT connection 52 may include message formatting, retransmission settings, preferred routing, etc., and the reconfiguration may not need to affect the network node 16, and may be unknown to or imperceptible to the network node 16. Several such procedures and functions are known and practiced in the art. In some embodiments, the measurement may involve proprietary WD signaling that facilitates the measurement of the host computer 24, such as throughput, propagation time, and latency. In some embodiments, the measurement may be implemented such that software 48, 90 monitors propagation time, errors, etc., and software 48, 90 uses an OTT connection 52 to send messages, in particular empty or "dummy" messages.

[0070] Accordingly, in some embodiments, the host computer 24 includes a processing circuit 42 configured to provide user data and a communication interface 40 configured to forward the user data to the cellular network for transmission to the WD22. In some embodiments, the cellular network also includes a network node 16 having a radio interface 62. In some embodiments, the network node 16 is configured to perform the functions and / or methods described herein for preparing / starting / maintaining / supporting / terminating transmissions to the WD22 and / or preparing / terminating / maintaining / supporting / terminating transmissions from the WD22, and / or the processing circuit 68 of the network node 16 is configured to perform them.

[0071] In some embodiments, the host computer 24 includes a processing circuit 42 and a communication interface 40, the communication interface 40 being configured to receive user data originating from a transmission from the WD 22 to the network node 16. In some embodiments, the WD 22 is configured to perform the functions and / or methods described herein for preparing / starting / maintaining / supporting / terminating transmissions to the network node 16 and / or preparing / terminating / maintaining / supporting / terminating transmissions from the network node 16, and / or includes a radio interface 82 and / or processing circuit 84 configured to perform them.

[0072] Figures 2 and 3 show various "units," such as the AI / ML unit 32, within each processor. These units can be implemented such that a portion of the unit is stored in corresponding memory within the processing circuit. In other words, the units can be implemented in hardware or as a combination of hardware and software within the processing circuit.

[0073] Figure 4 is a flowchart illustrating an exemplary method implemented in a communication system, such as the communication system in Figures 2 and 3, according to one embodiment. The communication system may include a host computer 24, a network node 16, and a WD22, which may be described with reference to Figure 3. In a first step of the method, the host computer 24 provides user data (block S100). In an optional substep of the first step, the host computer 24 provides user data by running a host application, such as host application 50 (block S102). In a second step, the host computer 24 initiates a transmission to carry the user data to the WD22 (block S104). In an optional third step, the network node 16 transmits the user data carried in the transmission initiated by the host computer 24 to the WD22, in accordance with the teachings of the embodiments described throughout this disclosure (block S106). In an optional fourth step, WD22 executes a client application, such as a client application 92, which is related to the host application 50 executed by the host computer 24 (block S108).

[0074] Figure 5 is a flowchart illustrating an exemplary method implemented in a communication system, such as the communication system of Figure 2, according to one embodiment. The communication system may include a host computer 24, a network node 16, and a WD22, which may be described with reference to Figures 2 and 3. In a first step of the method, the host computer 24 provides user data (block S110). In an optional substep (not shown), the host computer 24 provides user data by running a host application, such as host application 50. In a second step, the host computer 24 initiates a transmission to carry the user data to the WD22 (block S112). The transmission may proceed via the network node 16, as taught in the embodiments described throughout this disclosure. In an optional third step, the WD22 receives the user data carried in the transmission (block S114).

[0075] Figure 6 is a flowchart illustrating an exemplary method implemented in a communication system, such as the communication system in Figure 2, according to one embodiment. The communication system may include a host computer 24, a network node 16, and a WD 22, which may be described with reference to Figures 2 and 3. In an optional first step of the method, the WD 22 receives input data provided by the host computer 24 (block S116). In an optional substep of the first step, the WD 22 runs a client application 92 that provides user data in response to the received input data provided by the host computer 24 (block S118). In an optional second step, either additionally or alternatively, the WD 22 provides user data (block S120). In an optional substep of the second step, the WD provides user data by running a client application, such as the client application 92 (block S122). When providing user data, the executed client application 92 may further consider user input received from the user. Regardless of the specific format in which the user data is provided, WD22 may initiate transmission of the user data to the host computer 24 in an optional third substep (block S124). In a fourth step of the method, the host computer 24 receives the user data transmitted from WD22 in accordance with the teachings of the embodiments described throughout this disclosure (block S126).

[0076] Figure 7 is a flowchart illustrating an exemplary method implemented in a communication system, such as the communication system of Figure 2, according to one embodiment. The communication system may include a host computer 24, a network node 16, and a WD22, which may be described with reference to Figures 2 and 3. In an optional first step of the method, the network node 16 receives user data from the WD22 (block S128), in accordance with the teachings of the embodiments described throughout this disclosure. In an optional second step, the network node 16 initiates a transmission of the received user data to the host computer 24 (block S130). In a third step, the host computer 24 receives the user data carried in the transmission initiated by the network node 16 (block S132).

[0077] Figure 8 is a flowchart of an exemplary process at network node 16 for enhanced wireless device (WD) latency performance analysis to support application layer artificial intelligence (AI) / machine learning (ML) operations. One or more blocks described herein may be implemented by one or more elements of network node 16, such as by one or more of the processing circuit 68 (including the AI / ML unit 32), processor 70, wireless interface 62, and / or communication interface 60. Network node 16 may be configured to implement a federated learning process involving multiple inputs of artificial intelligence / machine learning (AI / ML), such as via the processing circuit 68 and / or processor 70 and / or wireless interface 62 and / or communication interface 60, where the multiple inputs include transmitted data information (block S134). The federated learning process includes a WD latency performance analysis configured to produce multiple outputs, where the multiple outputs include latency statistics (block S136).

[0078] In some embodiments, the transmitted data information includes at least one of the following: round-trip transmitted data volume, start and end times for the data transmission, and filter information. In some embodiments, the latency statistics include uplink and downlink latency statistics for the target transmission. In some embodiments, the latency statistics include round-tip latency statistics for the target transmission. In some embodiments, the latency statistics include maximum latency for the target transmission.

[0079] Figure 9 is a flowchart of an exemplary process in a core network node 36 configured to include a network data analysis function (NWDAF34) configured according to the principles disclosed herein. One or more blocks described herein may be carried out by one or more elements of the core network node 36, such as by one or more of the processing circuitry 94 (including the NWDAF34) and the radio interface 96. The core network node 36, such as via the processing circuitry 94 and the radio interface 96, collects a radio device (WD) 22 latency performance input from a network function (NF) 32, and is configured to collect a radio device (WD) 22 latency performance input that includes a plurality of transmit data volume values ​​and transmit time values ​​(block S138). The method includes determining a set of latency performance analyses to support a federative learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between WD22 and application functions (AF), at least in part based on latency performance inputs (block S140). The method also includes transmitting the set of latency performance analyses to NF32 (block S142).

[0080] In some embodiments, at least one of the latency predictions and latency statistics includes the time delay required to complete the transmission of a target volume of data from WD22 to NF32 or from NF32 to WD22. In some embodiments, at least one of the latency predictions and latency statistics includes the estimated number of retransmissions required to complete the transmission of a target volume of data. In some embodiments, at least one of the latency predictions and latency statistics includes at least one of uplink packet delay, downlink packet delay, and round-trip packet delay. In some embodiments, at least one of the latency predictions and latency statistics includes a value averaged over the analysis target period. In some embodiments, at least one of the latency predictions and latency statistics includes the validity period for latency performance analysis. In some embodiments, at least one of the latency predictions and latency statistics includes the maximum packet delay observed for communicating with NF32. In some embodiments, at least one of the latency predictions and latency statistics includes a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied. In some embodiments, at least one of the latency prediction and latency statistics is grouped into latency classes according to the latency performance range. In some embodiments, at least one of the latency prediction and latency statistics includes an indication of the WD22 percentage in each latency class. In some embodiments, NF is implemented as an application function.

[0081] Figure 10 is a flowchart of an exemplary process in a network node 16 configured to include a network function (NF32) configured according to the principles disclosed herein. One or more blocks described herein may be implemented by one or more elements of the network node 16, such as by one or more of the processing circuit 68 (including NWDAF34), processor 70, radio interface 62 and / or communication interface 60. The network node 16, such as via the processing circuit 68 and / or processor 70 and / or radio interface 62 and / or communication interface 60, is configured to transmit a latency performance input to the NWDAF34, the latency performance input being configured to transmit a latency performance input including a plurality of transmit data volume values ​​and transmit time values ​​(block S144). The method includes receiving a set of latency performance analyses to support a federative learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between WD22 and NF32, at least in part based on a latency performance input (block S146). The method also includes hosting an artificial intelligence / machine learning (AI / ML) based service at least in part based on a federative learning process (block S148).

[0082] In some embodiments, the latency prediction and latency statistics include a time delay to complete the transmission of a target volume of data from WD22 to NF32 or from NF32 to WD22. In some embodiments, the latency prediction and latency statistics include an estimated number of retransmissions to complete the transmission of a target volume of data. In some embodiments, the latency prediction and latency statistics include at least one of uplink packet delay, downlink packet delay, and round-trip packet delay. In some embodiments, the latency prediction and latency statistics include a value averaged over the analysis target period. In some embodiments, the latency prediction and latency statistics include a valid period for latency performance analysis. In some embodiments, the latency prediction and latency statistics include the maximum packet delay observed for communicating with NF32. In some embodiments, the latency prediction and latency statistics include a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied. In some embodiments, at least one of the latency prediction and latency statistics is grouped into latency classes according to the latency performance range. In some embodiments, at least one of the latency prediction and latency statistics includes an indication of the WD22 percentage in each latency class. In some embodiments, NF is implemented as an application function.

[0083] Having described the general process flow of the configuration of this disclosure and provided examples of hardware and software configurations for implementing the processes and functions of this disclosure, the following sections provide configuration details and examples for enhanced wireless device (WD) latency performance analysis to support application layer artificial intelligence (AI) / machine learning (ML) operations.

[0084] Input for WD latency performance analysis for associative learning Referring to Table 1 below, expanding an existing input from a single value to a (possible) list of values ​​may include one or more of the following: • To represent the possibility of different data volumes for different AI / ML traffic in a single federative learning process, change the input "Transmit Data Volume" from a single value to a list of (possible) values, and further expand that input to "Round Trip Transmit Data Volume", and • Change the input "transmission time" from a single value to a list of (possible) values ​​to represent the possibility of different transmission times for different AI / ML traffic data volumes in a single federative learning process.

[0085] Some embodiments may include adding new inputs for WD latency performance analysis. This may include one or more of the following: • If round-trip transmission data volumes are provided, the "time associated with the round trip" (for example, the time interval for the total round-trip time for each data volume) may also be provided by the AF. • To obtain information about the start and end times of data transmissions with (one or more) transmission data volumes for AI / ML traffic in a federative learning process, a new input "timestamp" may be added to identify the start and end times of data transmissions with transmission data volumes. This new input may be used for analysis of specific time intervals, for example, tomorrow from 10:00 to 10:30. To obtain more detailed information about AI / ML traffic in order to generate statistics and predictions regarding latency performance over specific time intervals (for example, using the service flow for AI / ML traffic between WD22 and AF from 10:00 to 10:30), new inputs, "IP filter information," "application location," and "application server instance address" may be implemented. TIFF2026508989000008.tif210170

[0086] Output of WD latency performance analysis for associative learning Referring to Tables 2 and 3 below, to obtain the UL / DL / round-trip delay analysis results for WD22 communicating with the application for the complete transmission of the target data volume (for example, NWDAF34 may estimate the number of transmissions for the data volume), there may be one single transmission or multiple retransmissions for the data volume to achieve QoS requirements. New outputs may be added for WD latency performance analysis, and new outputs may include one or more of the following: • Estimated number of transmissions (statistics / forecast) • Statistics / forecasts regarding UL latency for target transmissions. • Statistics / forecasts regarding DL latency for target transmission. • Statistics / predictions regarding round-trip latency for target transmission, and / or • Statistics / predictions regarding maximum latency for target transmission. TIFF2026508989000009.tif255167TIFF2026508989000010.tif241170TIFF20265089890 00011.tif255167TIFF2026508989000012.tif255167TIFF2026508989000013.tif114170

[0087] Some embodiments may include one or more of the following: Embodiment A1. A network node configured to communicate with a wireless device (WD), wherein the network node is Implementing a federated learning process that includes artificial intelligence / machine learning (AI / ML) with multiple inputs, where the multiple inputs include transmitted data information. A wireless interface configured to perform and / or a processing circuit configured to perform, A federative learning process includes a WD latency performance analysis configured to produce multiple outputs, each containing latency statistics, on a network node. Embodiment A2. The network node according to Embodiment A1, wherein the transmitted data information includes at least one of the following: a round-trip transmitted data volume, start and end times for data transmission, and filter information. Embodiment A3. A network node according to Embodiment A1 or A2, wherein the latency statistics include uplink and downlink latency statistics for target transmission. Embodiment A4. A network node according to any one of Embodiments A1 to A3, wherein the latency statistics include round-tip latency statistics for target transmission. Embodiment A5. A network node according to any one of Embodiments A1 to A4, wherein the latency statistics include the maximum latency for target transmission. Embodiment B1. A method implemented in a network node configured to communicate with a wireless device (WD), wherein the method is: Implementing a federated learning process that includes artificial intelligence / machine learning (AI / ML) with multiple inputs, where the multiple inputs include transmitted data information. Includes, A method that includes a federative learning process, a WD latency performance analysis configured to produce multiple outputs, each output containing latency statistics. Embodiment B2. The method according to Embodiment B1, wherein the transmitted data information includes at least one of the following: a round-trip transmitted data volume, start and end times for data transmission, and filter information. Embodiment B3. The method according to Embodiment B1 or B2, wherein the latency statistics include uplink and downlink latency statistics for target transmission. Embodiment B4. The method according to any one of Embodiments B1 to B3, wherein the latency statistics include round-tip latency statistics for target transmission. Embodiment B5. The method according to any one of Embodiments B1 to B4, wherein the latency statistics include the maximum latency for target transmission.

[0088] As will be understood by those skilled in the art, the concepts described herein may be embodied as methods, data processing systems, computer program products, and / or computer storage media for storing executable computer programs. Accordingly, the concepts described herein may take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware embodiments, all of which may be generally referred to herein as “circuits” or “modules.” Any process, step, action, and / or function described herein may be carried out by and / or associated with a corresponding module, which may be implemented in software and / or firmware and / or hardware. Furthermore, this disclosure may take the form of a computer program product on a tangible computer-readable storage medium having computer program code embodied in a medium that can be executed by a computer. Any suitable tangible computer-readable medium may be used, including hard disks, CD-ROMs, electronic storage devices, optical storage devices, or magnetic storage devices.

[0089] Several embodiments have been described herein with reference to flowcharts and / or block diagrams illustrating methods, systems, and computer program products. It will be understood that each block in a flowchart and / or block diagram, as well as combinations of blocks in a flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a dedicated computer, or other programmable data processing device for creating a machine (thereby creating a dedicated computer), and so those instructions executed via the processor of the computer or other programmable data processing device create means for implementing a function / action specified in one or more blocks of a flowchart and / or block diagram.

[0090] These computer program instructions may also be stored in computer-readable memory or storage medium that can guide a computer or other programmable data processing device to function in a particular manner, and so the instructions stored in computer-readable memory may produce a product that includes instruction means for implementing a function / action specified in one or more blocks of a flowchart and / or block diagram.

[0091] Computer program instructions can also be loaded into a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device in order to create a computer implementation process; therefore, instructions executed on a computer or other programmable device provide steps for implementing a function / action specified in one or more blocks of a flowchart and / or block diagram.

[0092] It should be understood that the functions / actions mentioned within a block may occur in a different order than those shown in the illustrative diagram of the operation. For example, depending on the functions / actions involved, two blocks shown consecutively may, in effect, be executed substantially concurrently, or blocks may sometimes be executed in reverse order. Some of the diagrams include arrows on the communication path to indicate the primary direction of communication, but it should be understood that communication may occur in the opposite direction to the illustrated arrows.

[0093] Computer program code for performing the operations of the concepts described herein may be written in an object-oriented programming language such as Python, Java®, or C++. However, computer program code for performing the operations of the disclosure may also be written in a conventional procedural programming language such as the C programming language. The program code may run entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer. In the latter scenario, the remote computer may be connected to the user's computer through a local area network (LAN) or wide area network (WAN), or the connection may be made to an external computer (for example, via the Internet using an Internet service provider).

[0094] Many different embodiments have been disclosed herein in relation to the above description and drawings. It will be understood that a literal description and illustration of every combination and partial combination of these embodiments would be excessively repetitive and obscure. Therefore, all embodiments may be combined in some way and / or in combination, and this specification, including the drawings, should be construed as constituting a complete written description of all combinations and partial combinations of the embodiments described herein, and all combinations and partial combinations of the modes and processes of making and using them, and shall support any claims for any such combination or partial combination.

[0095] The abbreviations that may be used in the above explanation include the following: AF Application Function AMF access and mobility management functions AI artificial intelligence FL Associative Learning FQDN (Fully Qualified Domain Name) GMLC Gateway Mobile Location Center ML (Machine Learning) NEF Network Publishing Function NF Network Function NWDAF Network Data Analysis Function OAM Operations Administration and Maintenance SMF session management function UPF User Plane Functionality 5GC 5G Core Network

[0096] It will be understood by those skilled in the art that the embodiments described herein are not limited to those specifically shown and described herein. Furthermore, it should be noted that not all of the accompanying drawings are to a constant scale unless otherwise stated above. In light of the above teachings, various modifications and variations are possible without departing from the following claims.

Claims

1. A method in a core network node (36) configured to include a network data analysis function (NWDAF) (34), wherein the method is: S138) Collecting a wireless device (WD) (22) latency performance input from a network function (NF) (32), wherein the latency performance input includes a plurality of transmit data volume values ​​and transmit time values. Determining a set of latency performance analyses to support a federative learning process (S140), wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between the WD (22) and the application function (AF), at least in part based on the latency performance input. The set of latency performance analysis is transmitted to the NF (32) (S142) Methods that include...

2. The method according to claim 1, wherein at least one of latency prediction and latency statistics includes a time delay to complete the transmission of a target volume of data from WD(22) to NF(32) or from NF(32) to WD(22).

3. The method according to claim 2, wherein at least one of latency prediction and latency statistics includes an estimated number of retransmissions to complete the transmission of the target volume of data.

4. The method according to claim 1 or 2, wherein the at least one of latency prediction and latency statistics includes at least one of uplink packet delay, downlink packet delay, and round-trip packet delay.

5. The method according to any one of claims 1 to 4, wherein the at least one of the latency prediction and latency statistics includes a value averaged over the analysis target period.

6. The method according to any one of claims 1 to 5, wherein the at least one of latency prediction and latency statistics includes an effective period for the latency performance analysis.

7. The method according to any one of claims 1 to 6, wherein the at least one of latency prediction and latency statistics includes the maximum packet delay observed for communicating with the NF(32).

8. The method according to any one of claims 1 to 7, wherein the at least one of the latency prediction and latency statistics includes a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied.

9. The method according to any one of claims 1 to 8, wherein at least one of latency prediction and latency statistics is grouped into latency classes according to latency performance ranges.

10. The method according to claim 9, wherein at least one of latency prediction and latency statistics includes an indication of the percentage of WD(22) in each latency class.

11. The method according to any one of claims 1 to 10, wherein the NF is implemented as the application function.

12. A core network node (36) configured to include a network data analysis function (NWDAF) (34), wherein the core network node (36) is Collecting a wireless device (WD) (22) latency performance input from a network function (NF) (32), wherein the latency performance input includes a plurality of transmit data volume values ​​and transmit time values. Determining a set of latency performance analyses to support a federative learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between the WD(22) and application functions (AF), at least in part based on the latency performance input. The set of latency performance analysis is transmitted to the NF(32) A core network node (36) configured to perform this task.

13. The core network node (36) according to claim 12, wherein at least one of latency prediction and latency statistics includes a time delay to complete the transmission of a target volume of data from WD(22) to NF(32) or from NF(32) to WD(22).

14. The core network node (36) according to claim 13, wherein at least one of latency prediction and latency statistics includes an estimated number of retransmissions to complete the transmission of the target volume of data.

15. The core network node (36) according to claim 12 or 13, wherein the at least one of latency prediction and latency statistics includes at least one of uplink packet delay, downlink packet delay and round-trip packet delay.

16. The core network node (36) according to any one of claims 12 to 15, wherein the at least one of latency prediction and latency statistics includes a value averaged over the analysis target period.

17. The core network node (36) according to any one of claims 12 to 16, wherein at least one of latency prediction and latency statistics includes a validity period for the latency performance analysis.

18. The core network node (36) according to any one of claims 12 to 17, wherein at least one of latency prediction and latency statistics includes the maximum packet delay observed for communicating with the NF (32).

19. The core network node (36) according to any one of claims 12 to 18, wherein at least one of latency prediction and latency statistics includes a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied.

20. The core network node (36) according to any one of claims 12 to 19, wherein at least one of latency predictions and latency statistics is grouped into latency classes according to latency performance ranges.

21. The core network node (36) according to claim 20, wherein at least one of latency prediction and latency statistics includes an indication of the percentage of WD(22) in each latency class.

22. The core network node according to any one of claims 12 to 21, wherein the NF is implemented as the application function.

23. A method in a network node configured to include a network function (NF) (32) and to communicate with a network data analysis function (NWDAF) (34), wherein the method is Transmitting a latency performance input to the NWDAF (34) (S144), wherein the latency performance input includes a plurality of transmit data volume values ​​and transmit time values ​​(S144), Receiving a set of latency performance analyses to support a federative learning process (S146), wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between a wireless device (WD) (22) and an application function (AF), at least in part based on the latency performance input. Hosting an artificial intelligence / machine learning (AI / ML) based service at least partially on the aforementioned federated learning process (S148) Methods that include...

24. The method according to claim 23, wherein at least one of latency prediction and latency statistics includes a time delay to complete the transmission of a target volume of data from WD(22) to NF(32) or from NF(32) to WD(22).

25. The method according to claim 24, wherein at least one of latency prediction and latency statistics includes an estimated number of retransmissions to complete the transmission of the target volume of data.

26. The method according to any one of claims 23 to 25, wherein the at least one of latency prediction and latency statistics includes at least one of uplink packet delay, downlink packet delay and round-trip packet delay.

27. The method according to any one of claims 23 to 26, wherein the at least one of the latency prediction and latency statistics includes a value averaged over a target analysis period.

28. The method according to any one of claims 23 to 27, wherein at least one of latency prediction and latency statistics includes an effective period for the latency performance analysis.

29. The method according to any one of claims 23 to 28, wherein at least one of latency prediction and latency statistics includes the maximum packet delay observed for communicating with the NF(32).

30. The method according to any one of claims 23 to 29, wherein the at least one of the latency prediction and latency statistics includes a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied.

31. The method according to any one of claims 23 to 30, wherein at least one of latency predictions and latency statistics is grouped into latency classes according to latency performance ranges.

32. The method according to claim 31, wherein at least one of latency prediction and latency statistics includes an indication of the percentage of WD(22) in each latency class.

33. The method according to any one of claims 23 to 32, wherein the NF is implemented as the application function.

34. A network node (16) configured to include a network function (NF) (32) and to communicate with a network data analysis function (NWDAF) (34), wherein the network node (16) Transmitting a latency performance input to the NWDAF (34), wherein the latency performance input includes a plurality of transmit data volume values ​​and transmit time values. Receiving a set of latency performance analyses to support a federative learning process, wherein the set of latency performance analyses includes at least one of latency predictions and latency statistics relating to uplink and downlink transmission latency between a wireless device (WD) (22) and an application function (AF), at least in part based on the latency performance input. Hosting an artificial intelligence / machine learning (AI / ML) based service at least partially on the aforementioned federated learning process and A network node (16) configured to perform this task.

35. The network node (16) according to claim 34, wherein at least one of latency prediction and latency statistics includes a time delay to complete the transmission of a target volume of data from WD(22) to NF(32) or from NF(32) to WD(22).

36. The network node (16) according to claim 35, wherein at least one of latency prediction and latency statistics includes an estimated number of retransmissions to complete the transmission of the target volume of data.

37. The network node (16) according to claim 35 or 36, wherein the at least one of latency prediction and latency statistics includes at least one of uplink packet delay, downlink packet delay and round-trip packet delay.

38. The network node (16) according to any one of claims 34 to 37, wherein the at least one of latency prediction and latency statistics includes a value averaged over the analysis target period.

39. The network node (16) according to any one of claims 34 to 38, wherein at least one of latency prediction and latency statistics includes a validity period for the latency performance analysis.

40. The network node (16) according to any one of claims 34 to 39, wherein the at least one of latency prediction and latency statistics includes the maximum packet delay observed for communicating with the NF (32).

41. The network node (16) according to any one of claims 34 to 40, wherein at least one of latency prediction and latency statistics includes a spatial effectiveness parameter indicating the area to which the latency performance analysis is applied.

42. The network node (16) according to any one of claims 34 to 41, wherein at least one of latency predictions and latency statistics is grouped into latency classes according to latency performance ranges.

43. The network node (16) according to claim 42, wherein at least one of latency prediction and latency statistics includes an indication of the percentage of WD(22) in each latency class.

44. The network node (16) according to any one of claims 34 to 43, wherein the NF is implemented as the application function.

Images