TECHNICAL FIELD

[001] Disclosed are embodiments related to systems and method for discovering a network node's capabilities.

BACKGROUND

[002] 1 . Overall Architecture of NG-RAN

[003] FIG. 1 illustrates the current 5G RAN (a.k.a., the Next Generation RAN (NG-RAN)) architecture.

The NG-RAN architecture is described in 3GPP Technical Specification (TS) 38.401 v17.0.0 (“TS 38.401”). The NG- RAN consists of a set of base stations (denoted "gNBs”) connected to a 5G core network (5GC) through an NG interface. The gNBs can be interconnected through an Xn interface.

[004] As specified in 3GPP TS 38.300 v17.1 .0 ("TS 38.300”), a gNB may consist of a gNB central unit (gNB-CU) and one or more gNB distributed units (gNB-DU(s)). A gNB-CU and a gNB-DU are connected via an F1 interface. Typically, one gNB-DU is connected to only one gNB-CU. The NG, Xn and F1 interfaces are logical interfaces. A gNB-CU may comprise a gNB-CU control plane (CP) function (gNB-CU-CP) and a gNB-CU user plane (UP) function (gNB-CU-UP).

[005] As also specified in TS 38.300, NG-RAN could also consist of a set of ng-eNBs, an ng-eNB may consist of an ng-eNB-CU and one or more ng-eNB-DU(s). An ng-eNB-CU and an ng-eNB-DU is connected via W1 interface. The general principle described in this section also applies to ng-eNB and W1 interface, if not explicitly specified otherwise.

[006] A gNB can support FDD mode, TDD mode, or dual mode operation.

[007] In case of network sharing with multiple cell ID broadcast, each Cell Identity associated with a subset of PLMNs corresponds to a gNB-DU and the gNB-CU it is connected to, i.e. the corresponding gNB-DUs share the same physical layer cell resources. For resiliency, a gNB-DU may be connected to multiple gNB-CUs by appropriate implementation.

[008] For NG-RAN, the NG and Xn-C interfaces for a gNB consisting of a gNB-CU and gNB-DUs, terminate in the gNB-CU. For EN-DC, the S1-U and X2-C interfaces for a gNB consisting of a gNB-CU and gNB- DUs, terminate in the gNB-CU. The gNB-CU and connected gNB-DUs are only visible to other gNBs and the 5GC as a gNB. A possible deployment scenario is described in Annex A of TS 38.300. [009] The node hosting user plane part of NR PDCP (e.g. gNB-CU, gNB-CU-UP, and for EN-DC, MeNB or SgNB depending on the bearer split) shall perform user inactivity monitoring and further informs its inactivity or (re)activation to the node having C-plane connection towards the core network (e.g. over E1, X2). The node hosting NR RLC (e.g. gNB-DU) may perform user inactivity monitoring and further inform its inactivity or (re)activation to the node hosting control plane, e.g. gNB-CU or gNB-CU-CP.

[0010] UL PDCP configuration (i.e. how the UE uses the UL at the assisting node) is indicated via X2-C (for EN-DC), Xn-C (for NG-RAN) and F1-C. Radio Link Outage/Resume for DL and/or UL is indicated via X2-U (for EN-DC), Xn-U (for NG-RAN) and F1-U.

[0011] The NG-RAN is layered into a Radio Network Layer (RNL) and a Transport Network Layer (TNL). The NG-RAN architecture, i.e., the NG-RAN logical nodes and interfaces between them, is defined as part of the RNL. For each NG-RAN interface (NG, Xn, F1) the related TNL protocol and the functionality are specified. The TNL provides services for user plane transport, signalling transport.

[0012] In NG-Flex configuration, each NG-RAN node is connected to all AMFs of AMF Sets within an AMF Region supporting at least one slice also supported by the NG-RAN node. The AMF Set and the AMF Region are defined in 3GPP TS 23.501, v17.6.0.

[0013] If security protection for control plane and user plane data on TNL of NG-RAN interfaces has to be supported, NDS/IP 3GPP TS 33.501 shall be applied. The overall architecture for separation of gNB-CU-CP and gNB-CU-UP is depicted in FIG. 2 and specified in TS 37.483, v17.2.0.

[0014] A gNB may consist of a gNB-CU-CP, multiple gNB-CU-UPs and multiple gNB-DUs. The gNB-CU- CP is connected to the gNB-DU through the F1-C interface. The gNB-CU-UP is connected to the gNB-DU through the F1-U interface. The gNB-CU-UP is connected to the gNB-CU-CP through the E1 interface. For resiliency, a gNB- DU and/or a gNB-CU-UP may be connected to multiple gNB-CU-CPs by appropriate implementation One gNB-DU can be connected to multiple gNB-CU-UPs under the control of the same gNB-CU-CP. One gNB-CU-UP can be connected to multiple DUs under the control of the same gNB-CU-CP. The connectivity between a gNB-CU-UP and a gNB-DU is established by the gNB-CU-CP using Bearer Context Management functions. The gNB-CU-CP selects the appropriate gNB-CU-UP(s) for the requested services for the UE. In case of multiple CU-UPs they belong to same security domain as defined in TS 33.210, v17.1.0. Data forwarding between gNB-CU-UPs during Intra-gNB- CU-CP handover within a gNB may be supported by Xn-U.

[0015] 2. Resource Status Reporting Initiation [0016] The resource status reporting procedure, which is illustrated in FIG. 3A, is used by an NG-RAN node to request the reporting of load measurements to another NG-RAN node. The procedure uses non UE- associated signalling.

[0017] As shown in FIG. 3A, NG-RAN nodei initiates the procedure by sending the RESOURCE STATUS REQUEST message to NG-RAN node2 to start a measurement, stop a measurement or add cells to report for a measurement. Upon receipt, NG-RAN node2 shall: I) initiate the requested measurement according to the parameters given in the request in case the Registration Request \E set to "start"; or ii) stop all cells measurements and terminate the reporting in case the Registration Request \E is set to "stop"; or ill) add cells indicated in the Cell To Report List IE to the measurements initiated before for the given measurement IDs, in case the Registration Requests is set to "add". If measurements are already initiated for a cell indicated in the Cell To Report List IE, this information shall be ignored.

[0018] If the Registration Request \E is set to "start" in the RESOURCE STATUS REQUEST message and the Report Characteristics IE indicates cell specific measurements, the Cell To Report List \E shall be included.

[0019] If Registration Request IE is set to "add" in the RESOURCE STATUS REQUEST message, the Cell To Report List \E shall be included.

[0020] If NG-RAN node2 is capable to provide all requested resource status information, it shall initiate the measurement as requested by NG-RAN nodei and respond with the RESOURCE STATUS RESPONSE message.

[0021] 2.1 Interaction with other procedures

[0022] When starting a measurement, the Report Characteristics IE in the RESOURCE STATUS REQUEST indicates the type of objects NG-RAN node2 shall perform measurements on. For each cell, NG-RAN node2 shall include in the RESOURCE STATUS UPDATE message the following information Elements (lEs):

TABLE 1 - lEs included in RESOURCE STATUS UPDATE MESSAGE

[0023] If the Reporting Periodicity IE in the RESOURCE STATUS REQUEST is present, this indicates the periodicity for the reporting of periodic measurements, the NG-RAN node2 shall report only once, unless otherwise requested within the Reporting Periodicity IE.

[0024] If any of the requested measurements cannot be initiated, NG-RAN node2 shall, instead of sending the RESOURCE STATUS RESPONSE message, send the RESOURCE STATUS FAILURE message with an appropriate cause value (see FIG. 3B). [0025] 2.2 Abnormal Conditions

[0026] For the same Measurement ID, if the initiating NG-RAN nodel does not receive either the RESOURCE STATUS RESPONSE message or the RESOURCE STATUS FAILURE message, the NG-RAN nodel may reinitiate the Resource Status Reporting Initiation procedure towards the same NG-RAN node, provided that the content of the new RESOURCE STATUS REQUEST message is identical to the content of the previously unacknowledged RESOURCE STATUS REQUEST message.

[0027] If the NG-RAN node2 receives a RESOURCE STATUS REQUEST message which includes the Registration Request IE set to "add" or "stop" and if the NG-RAN node2 Measurement ID value received in the RESOURCE STATUS REQUEST message is not used, the NG-RAN node2 shall initiate RESOURCE STATUS FAILURE message with an appropriate cause value.

[0028] If the Report Characteristics IE bitmap is set to "0" (all bits are set to "0") in the RESOURCE STATUS REQUEST message then NG-RAN node2 shall initiate a RESOURCE STATUS FAILURE message with an appropriate cause value.

[0029] If the NG-RAN node2 receives a RESOURCE STATUS REQUEST message which includes the Registration Request IE set to "start" and the NG-RAN nodel Measurement ID IE corresponding to an existing ongoing load measurement reporting, then NG-RAN node2 shall initiate a RESOURCE STATUS FAILURE message with an appropriate cause value.

[0030] 3. Resource Status Reporting

[0031] The Resource Status Reporting procedure, which is illustrated in FIG. 3C, is initiated by an NG- RAN node to report the result of measurements admitted by the NG-RAN node following a successful Resource Status Reporting Initiation procedure. The procedure uses non UE-associated signalling. NG-RAN node2 shall report the results of the admitted measurements in RESOURCE STATUS UPDATE message. The admitted measurements are the measurements that were successfully initiated during the preceding Resource Status Reporting Initiation procedure.

[0032] 4. Ongoing 3GPP discussion

[0033] The 3GPP Technical Report (TR) 37.817 V17.0.0 (“TR 37.817”) has been produced as outcome of the Study Item (SI) "Enhancement for Data Collection for NR and EN-DC” defined in 3GPP Technical Document (Tdoc) No. RP-201620. The normative work based on the conclusion of Rel-17 SI is currently undertaken in 3GPP Rel-18, the related Work Item (Wl) is described in RP-213602.

[0034] During a previous meeting (i.e. , meeting RAN3#117-e), the following agreements were made: define a new procedure over Xn which can be used for AI/ML related information, e.g., predicted information. The new procedure for reporting of AI/ML related information, e.g., predicted information, should be based in a requested way, like resource status report procedure.

[0035] Subsequently, the following agreements were made: Introduce a new Class 1 procedure for initiating the reporting of AI/ML Related Information and a Class 2 procedure for Data Reporting of AI/ML Related Information. The response message of the new procedure for AI/ML Related Information indicates if the requested information can be provided. AI/ML capability exchange in NG-RAN can be achieved by means of procedures for AI/ML information request, AI/ML information response, and AI/ML Information Request Failure.

[0036] Accordingly, RAN3 has agreed on introducing a new class 1 procedure for subscribing to AI/ML - related information similar to the Resource Status Reporting Initiation procedure previously described. Tentative names for the messages included in this procedure are: AI/ML Information Request, similar to RESOURCE STATUS REQUEST; AI/ML Information Response, similar to RESOURCE STATUS RESPONSE; and AI/ML Information Request Failure, similar to RESOURCE STATUS FAILURE.

[0037] When reporting predictions (a type of AI/ML-related information), one problem recognized by many companies is that of the validity time of predictions, i.e. , the time interval where the prediction is valid. Legacy measurements like the ones found in Resource Status Reporting, e.g., number of active UEs, can be performed by the same algorithm independently of how often these values are reported. In the case of predictions, an algorithm (or model) that can predict the average number of active UEs in the following 2 minutes may not be able to predict the average number of active UEs in the following 5 minutes.

[0038] Therefore, indicating the validity time of predictions has been recognized as a topic for further discussion: How to indicate validity time (e.g., implicitly with a new prediction when the previous prediction becomes invalid, explicitly with every prediction in the AI/ML output or by the request to the prediction) shall be discussed on a case by case basis.

[0039] As recognized in the agreement, one possible way of indicating the validity time is implicitly with the arrival of a new prediction. In the case of periodic reporting, the reporting periodicity (e.g., Reporting Periodicity IE as in the RESOURCE STATUS REQUEST message) determines how often a node reports the subscribed information; thus, the reporting periodicity determines the validity time of the predictions.

SUMMARY

[0040] Certain challenges presently exist. For instance, following the agreements made in 3GPP, an NG- RAN node will learn about the AI/ML supported information of other NG-RAN nodes by means of trial-and-error processes. When the NG-RAN node 1 sends a request to an NG-RAN node 2 to receive data, node 1 learns, based on the response, if NG-RAN node 2 supports AI/ML for a specific use case. The NG-RAN node 1 will gather the information about whether the other NG-RAN node 2 supports AI/ML for a specific use case, depending on whether the request is accepted or rejected.

[0041] If one follows the current mechanisms to successfully complete procedures for information reporting or to fail them then the criteria to follow in NG/RAN are: (1) when the reporting node can provide all the information asked for by the requesting node, the reporting node responds with the AI/ML Information Response message accepting to provide the information; and (2) when the reporting node cannot provide any of the information asked for by the requesting node, the reporting node responds with the AI/ML Information Request Failure message rejecting the AI/ML information request with an appropriate cause value.

[0042] Procedures specified so far for measurement reporting do not, however, state whether from a failed measurement request the requesting node can deduce capabilities at the reporting node to be able to report the measurements. Hence a first problem is how can failed or unsuccessful procedures indicate the actual capabilities of the reporting node.

[0043] A further problem arises in the case that the reporting node can provide some of the information asked for by the requesting node. This case is not straightforward and the legacy procedure that is used as guideline, i.e. , the NG-RAN Resource Status Reporting (Initiation) procedure, offers no solution in that case. Furthermore, TS 38.423, v17.2.0 ("TS 38.423”), states the following: If any of the requested measurements cannot be initiated, NG- RAN node2 shall send the RESOURCE STATUS FAILURE message with an appropriate cause value.

[0044] This makes the problem of how to deduce AI/ML capabilities more complex, because a failure may imply that none of the measurements are supported, or available at a given point in time, while some of those measurements may actually be supported/avai lable. Also, some of the measurements requested may not be supported, i.e., the reporting node has no capabilities to provide them, while some other measurements may simply be unavailable, i.e., the reporting node has the capabilities to provide them, but the measurement cannot be provided at that precise point in time.

[0045] With the expectation that AI/ML capabilities can differ significantly in different nodes due to varying hardware and software capabilities due to different types of nodes, software versions, etc., the above-mentioned problem can be very limiting when realizing information exchange required for AI/ML functionalities.

[0046] As previously noted, one issue related to predictions is their validity time; this issue is not present in legacy measurements like the ones found in Resource Status Reporting. Given that the new procedure for reporting predictions agreed in 3GPP is based on the legacy Resource Status Reporting (Initiation) procedure, there is a need to address this difference. For example, if NG-RAN node 1 requests a certain prediction with a validity time of 1 minute and NG-RAN node 2 can only provide the prediction with a validity time of 2 minutes, the legacy procedure would most likely fail. However, in the case of predictions, it is not clear what the outcome should be. In other words, the validity time of predictions gives another dimension to this problem that was not considered in legacy procedures.

[0047] Accordingly, in one aspect there is provided a method performed by a first network node. The method includes the first network node transmitting to a second network node a first request message for causing the second network node to initiate reporting of at least a first information item. The method also includes the first network node receiving from the second network node a first response message responsive to the first request message. The first response message comprises: I) information indicating that the reporting of the first information item cannot currently be initiated and ii) information indicating an expected time at which reporting of the first information item can be initiated.

[0048] In another aspect there is a first network node configured to perform such method.

[0049] In another aspect there is provided a method performed by a second network node. The method includes the second network node receiving, from a first network node, a first request message for causing the second network node to initiate reporting of at least a first information item. The method also includes the second network node transmitting to the first network node a first response message responsive to the first request message. The first response message comprises information indicating that the reporting of the first information item cannot currently be initiated and ii) information indicating an expected time at which reporting of the first information item can be initiated.

[0050] In another aspect there is a second network node configured to perform such method.

[0051] In another aspect there is provided a computer program comprising instructions which when executed by processing circuitry of a network node causes the network node to perform any of the methods disclosed herein. In some embodiments, there is provided a carrier containing the computer program wherein the carrier is one of an electronic signal, an optical signal, a radio signal, and a computer readable storage medium. In another aspect there is provided a network node that is configured to perform the methods disclosed herein. The network node may include memory and processing circuitry coupled to the memory.

[0052] An advantage of the embodiments disclosed herein is that they avoid the need to reject requests for AI/ML assistance data when only some of the requested information is not available or supported at the receiving node. This has the advantage of reducing the convergence time of the learning at the requesting node of what are the capabilities at the reporting node because the requesting node learns which of the overall set of requested information is supported and which is not. In the case of predictions being available with different validity times than the one requested, the embodiments provide a simple way of signaling this issue. Therefore, the embodiments enable the requesting node to differentiate between information not being available because of validity time issues and information not reported because they are not supported, i.e. because of capability limitations at the reporting node.

BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate various embodiments.

[0054] FIG. 1 illustrates the current 5G RAN (a.k.a., the Next Generation RAN (NG-RAN)) architecture.

[0055] FIG. 2 depicts the overall architecture for separation of gNB-CU-CP and gNB-CU-UP.

[0056] FIG. 3A is a message flow diagram.

[0057] FIG. 3B is a message flow diagram.

[0058] FIG. 3C is a message flow diagram.

[0059] FIG. 4 is a message flow diagram according to an embodiment.

[0060] FIG. 5A is a flowchart illustrating a process according to an embodiment.

[0061] FIG. 5B is a flowchart illustrating a process according to an embodiment.

[0062] FIG. 6 is a block diagram of a network node according to an embodiment.

[0063] FIG. 7A is a flowchart illustrating a process according to an embodiment.

[0064] FIG. 7B is a flowchart illustrating a process according to an embodiment.

DETAILED DESCRIPTION

[0065] Terminology

[0066] As used herein a "network node” can be a node in a radio access network (RAN), an 0AM, a core network (ON) node (e.g., a ON function), an 0AM, an SMO, a Network Management System (NMS), a Non-Real Time RAN Intelligent Controller (Non-RT RIO), a Real-Time RAN Intelligent Controller (RT-RIC), a gNB, eNB, en- gNB, ng-eNB, gNB-CU, gNB-CU-CP (gNB-CU control plane (CP)), gNB-CU-UP (gNB-CU user plane (UP)), eNB-CU, eNB-CU-CP, eNB-CU-UP, lAB-node, lAB-donor DU, lAB-donor-CU, IAB-DU, IAB-MT, O-CU, O-CU-CP, O-CU-UP, O-DU, O-RU, O-eNB, a user equipment (UE). References to network nodes herein should be understood such that a network node may be a physical node or a function or logical entity of any kind, e.g. a software entity implemented in a data center or a cloud, e.g. using one or more virtual machines, and two network nodes may well be implemented as logical software entities in the same data center or cloud. [0067] As used herein a "RAN node” is a node or entity in a radio access network (RAN), such as a gNB, eNB, en-gNB, ng-eNB, gNB-CU, gNB-CU-CP, gNB-CU-UP, eNB-CU, eNB-CU-CP, eNB-CU-UP, lAB-node, IAB- donor DU, lAB-donor-CU, IAB-DU, IAB-MT, O-CU, O-CU-CP, O-CU-UP, O-DU, O-RU, or O-eNB.

[0068] A function, or entity of a RAN node is to be intended as one of the functional entities comprised in a RAN node. A RAN node in split deployment comprises different functions or entities. For example, a gNB comprising a gNB-CU, one or more gNB-DU, one or more gNB-CU-CP.

[0069] The terms model training, model optimizing, model optimization, model updating are herein used interchangeably with the same meaning unless explicitly specified otherwise.

[0070] The terms model changing, modify or similar are herein used interchangeably with the same meaning unless explicitly specified otherwise. In particular, they refer to the fact that the type, structure, parameters, connectivity of a model may have changed compared to a previous format/configuration of the model.

[0071] The terms artificial intelligence (Al) and machine learning (ML) are used interchangeably. Hence, the abbreviations Al, ML, and AI/ML are used interchangeably.

[0072] The terms Al model, ML model, AI/ML model, and AIML model are herein used interchangeably with the term model. A model is as an application or an algorithm or a function including processes based on ML. For example, it may refer to the ML model itself as well as the support software packages needed for it to run properly, e.g., this may include the software necessary for data preparation.

[0073] Data collection refers to a process of collecting data for the purpose of model training, data analytics, and/or inference.

[0074] The embodiments disclosed herein are independent with respect to specific AI/ML model types or learning problems/setting (e.g. supervised learning, unsupervised learning, reinforcement learning, hybrid learning, centralized learning, federated learning, distributed learning, ...). Non limiting examples of AI/ML algorithms may include supervised learning algorithms, deep learning algorithms, reinforcement learning type of algorithms (such as DQN, A2C, A3C, etc.), contextual multi-armed bandit algorithms, autoregression algorithms, etc., or combinations thereof. Such algorithms may exploit functional approximation models, hereafter referred to as AI/ML models, such as neural networks (e.g. feedforward neural networks, deep neural networks, recurrent neural networks, convolutional neural networks, etc.). Examples of reinforcement learning algorithms may include deep reinforcement learning (such as deep Q-network (DQN), proximal policy optimization (PPO), double Q- learning), actor-critic algorithms (such as Advantage actor-critic algorithms, e.g. A2C or A3C, actor-critic with experience replay, etc), policy gradient algorithms, off-policy learning algorithms, etc. [0075] The network nodes described herein can be directly connected (i.e., with a signaling connection between them) or indirectly connected (i.e., with a signaling connection traversing intermediate network nodes between them, relaying the signaling information exchanged between them). Additionally, as used herein transmitting a message to an intended recipient encompasses transmitting the message directly to the intended recipient or transmitting the message indirectly to the intended recipient (i.e., one or more other nodes are used to relay the message from the source node to the intended recipient). Likewise, as used herein receiving a message from a sender encompasses receiving the message directly from the sender or indirectly from the sender (i.e., one or more nodes are used to relay the message from the sender to the receiving node).

[0076] The term "CU” is used herein as short for "gNB-CU” (and may also refer to an eNB-CU or an IAB- donor-CU”). The term "CU-CP” is used herein as short for "gNB-CU-CP” (and may also refer to an eNB-CU-CP or an lAB-donor-CU-CP). The term "CU-UP” is used herein as short for "gNB-CU-UP” (and may also refer to an eNB-CU- UP or an lAB-donor-CU-UP). The term "DU” is used herein as short for "gNB-DU” (and may also refer to an eNB-DU or an IAB-donor-DU)).

[0077] The terms "AI/ML capabilities”, "Al capabilities”, "ML capabilitiies”, "AI/ML related capabilities” are used interchangeably. ML capabilities refer to the support that a network node can offer when ML is used. For example, the ability to support certain use cases, to produce certain output (and at which pace), which AI/ML model types are supported, which AI/ML algorithms are supported.

[0078] The terms "AI/ML related requirements”, "AI/ML requirements” are used interchangeably and indicate requirements for executing an AI/ML model by a network node. Requirements refer to attributes and aspects needed for the AI/ML model to function according to design. They include aspects such as "processing power”, "memory consumption”, "minimum hardware”, "recommended hardware”, "frequency of input”, "security requirements”, "user consent requirements.”

[0079] In this disclosure it is often taken as an example that of RAN nodes. However, the person skilled in the art can easily deduce that the methods apply to all the nodes involved in ML based processes where exchange of ML capabilities is needed. Such nodes may be CN nodes, management systems, UE, external systems.

[0080] Introduction

[0081] In embodiments disclosed herein a network node (a.k.a., the "requesting node” or "node-1”) sends to another network node (a.k.a., the "reporting node” or "node-2”) a request for reporting several information items (e.g., several AI/ML-related information items), but the reporting node can only report some of the requested information. In order to enable the requesting node to understand the capabilities of the reporting node, the reporting node responds with a list of the information that can be provided and also with a list of the information that cannot be provided with an appropriate cause value. In this way, the requesting node will be able to understand the capabilities of the reporting node and be able to adapt its requests in the future.

[0082] The embodiments herein rely on an indication of why a requested piece of information (e.g., a measurement object or other information item) was not provided. Such indication may consist of a cause value associated to each failed to be reported item of information. The requesting node will therefore become aware, via the cause values for the information items that cannot be reported, about the capability related reasons for the unavailability of the information and decide accordingly. For example, it might be conveyed that an information item is not available for the time being, but it may be available at a future time, hence the requesting node may decide to wait and later make another request for the information item.

[0083] The reporting node can also give an indication of when the unavailable information will be available. It can also indicate alternative set of parameters that it can support.

[0084] The requesting node can provide a preferred set of parameters, as well as indication to the reporting node to report the information when available.

[0085] In some embodiments a flag is used by the requesting node to indicate to the reporting node that the requesting node accepts partial success outcomes on a particular request. This is justified to enable the requesting node to have a faster understanding of the reporting node's capabilities. Namely, procedures stating partial success for the requested measurements enable the requesting node to learn which measurements were available/not-available/supported/not-supported. Without the use of partial success, the reporting node would have to try requesting many measurements combinations in order to determine which measurements are not supported or unavailable and therefore cause a procedure failure.

[0086] Additional Details

[0087] Embodiments disclosed herein provide improvements to the framework currently discussed for AI/ML in RAN. An example of the AI/ML signaling procedure is shown in FIG. 4. The steps shown in FIG. 4 are described below.

[0088] 1 . A first network node 401 (referred to herein as "node-1”) sends an Xn setup request to a second network node 402 (referred to herein as "node-2”).

[0089] 2. Node-2 responds to the setup request by sending to node-1 an Xn setup response. In this manner, an Xn connection is established between node-1 and node 2.

[0090] 3. Node-1 performs measurement with respect to a least a first UE.

[0091] 4. Node-1 transmits a data request message to node-2. [0092] 5. Node-2 responds to the data request message by sending to node-1 a data response message.

[0093] 6. Node-2 also responds to the data request message by some time later sending to node-1 a data update message.

[0094] 7. Node-1 uses data receive from node-2 to produce a ML inference using a ML model.

[0095] 8. Node-2 transmits a data request message to node-1 .

[0096] 9. Node-1 responds to the data request message by sending to node-2 a data response message.

[0097] 10. Node-1 also responds to the data request message by some time later sending to node-1 a data update message.

[0098] 11 . Node-1 transmits a data request message to node-2.

[0099] 12. Node-2 responds to the data request message by sending to node-1 a data response message.

[00100] 13. Node-2 also responds to the data request message by some time later sending to node-1 a data update message.

[00101] In one embodiment a first network node (the requesting node) (e.g., 401 in FIG. 4) requests specific information (e.g., a set of one or more information items, such as a set of one or more measurement objects) from a second network node (the reporting node) (e.g., node-2 in FIG. 4). The requested information can be input, output or feedback. The second network node realizes that some of the requested information can be reported but some cannot. The second network node indicates to the first network node a list of the information that can be reported and it also indicates the information that cannot be reported.

[00102] In one embodiment, for each information item (e.g., a measurement, a prediction, or other information item) that cannot be reported a cause value is added indicating the cause of such failure to report the information item. As an example, some cause values may indicate that the reporting node is capable of reporting the information item but the information item cannot be reported at this time because the information is temporarily unavailable; some other cause value may indicate that the information cannot be reported because the reporting node does not support reporting the information item; some other cause value may indicate that the information cannot be reported because the reporting period cannot be respected by the reporting node.

[00103] In one embodiment a failure to report due to a reporting period that may not be respected by the reporting node may be associated to the validity time of the prediction corresponding to the information not being reported. Namely, in cases when the validity time for a prediction requested to be reported by the requesting node corresponds to the reporting periodicity of that prediction, a failure to report the prediction with an indication, e.g. a cause value, that the reporting is not possible due to a reporting period that cannot be respected may imply that the reporting node is not able to generate a prediction with the requested validity time. Namely, by means of such indication the requesting node also learns on the capability at the reporting node to predict a given metric with a specific validity time.

[00104] In another embodiment, for each information item (a.k.a., piece of information) that cannot be reported, the reporting node includes in the reporting message an IE corresponding to the information item that cannot be reported.

[00105] In both cases of expressing the reason for failure to report via cause value or via explicit IE, the information conveyed by the cause value, or the new IE may be one of the following (namely one of the following values may be associated to the failure to configure the requested information):

[00106] measurement not supported”,

[00107] measurement temporarily not available”,

[00108] measurement not supported with requested reporting periodicity”,

[00109] measurement not available with requested reporting periodicity”,

[00110] measurement not supported with current combination of requested information”,

[00111] measurement not available with current combination of requested information.”

[00112] To facilitate the description of the embodiments, the following text is based on new procedures specified for AI/ML which comprise the following messages: AI/ML ASSISTANCE DATA REQUEST, AI/ML ASSISTANCE DATA RESPONSE, AI/ML ASSISTANCE DATA FAILURE, and AI/ML ASSISTANCE DATA UPDATE.

[00113] This example and the name of the messages are only used to describe the embodiments and do not limit the scope of the embodiments. For instance, the tentative names (according to the RAN3 meeting agreements) for the first three messages are AI/ML Information Request, AI/ML Information Response, and AI/ML Information Request Failure, although the exact names of the procedure are still under discussion.

[00114] In the AI/ML ASSISTANCE DATA REQUEST message the first network node requests specific information and in the AI/ML ASSISTANCE DATA RESPONSE message, the second network node denotes the information that is available and also the information that is unavailable. For the unavailable information, cause values or a new indication IE is provided to clearly indicate the reason of the unavailability of the information and whether the information is expected to be available at a later point in time, in one example providing an estimated time when the information will be available. [00115] Following the example, after the estimated time passes, the requesting node can send a new AI/ML ASSISTANCE DATA REQUEST message and successfully establish the requested information reporting. In another example the reporting node will start reporting after the estimated time passes without the need for a new AI/ML ASSISTANCE DATA REQUEST message if this is agreed with the requesting node.

[00116] In one option the AI/ML Assistance Data Request message from the first network node to the second network node may include a flag to indicate that it accepts partial outcomes pertaining to the current request, i.e. , that the procedure should not fail if some of the requested information is not available, and the second network node should report the available information. By partially accepting the request of the requesting node the reporting node informs the requesting node of its capabilities for the subset of accepted and failed to report measurements

[00117] In a related option, the request sent by the requesting node may include, for each information item requested, a flag indicating that the requested information item is critical, i.e., even if partial outcomes are accepted, the lack of a critical information item should result in the procedure failing.

[00118] In a related option, a similar flag can indicate that the specific requested information is the most critical and as a result if they are available the procedure can be successful irrespective of the presence of other information not carrying an analogous flag.

[00119] In a related option, if the reporting node will make an effort to acquire the requested information, then it will know which information items to prioritize. By means of these enhancements, it is possible to deduce whether the reporting node is capable of reporting information items according to a specific combination of a subset of them. Namely, if the reporting node's capabilities are such that it is not possible to report all or some information together, by means of indicating a priority or weight for the requested measurements, the reporting node is able to check if it can report information with the highest priority (i.e. whether reporting of such information is available and/or supported) and the requesting node is able to learn the capability at the reporting node to report a combination of information.

[00120] If the flag indicating the most critical information is present in the request, then in another example, if these most critical information items are available, the requesting node might also indicate to the reporting node to report the rest of the information when they become available, or in a separate measurement report update procedure. The reporting node can then just send the other information when available via the AI/ML Assistance Data Update message without extra request.

[00121] In one option, the AI/ML Assistance Data Request message from the first network node to the second network node may include a flag to indicate that the first network node will accept the reporting of requested information items at a different periodicity and/or validity time than requested. In this embodiment, the reporting node may reply in a number of ways, including: [00122] (1) The reporting node may reply stating that the requested reporting period and/or validity time is not supported and that a different reporting period and/or validity time is supported;

[00123] (2) The reporting node may reply stating that the requested reporting period and/or validity time is not available and that a different reporting period and/or validity time is available.

[00124] The above helps the requesting node to learn the capabilities of the reporting node in terms of respecting certain reporting period/validity time for specific information items.

[00125] For information items that can be reported as per requests from the requesting node, the updates will be reported according to the specified periodicity and/or validity time in the AI/ML Assistance Data Request via the AI/ML Assistance Data Update messages. For information items that are not available, in one option, the second network node indicates in the response message (e.g., in the AI/ML ASSISTANCE DATA RESPONSE message), an (alternative) set of information items. For example, if the first network node requests to receive predictions with a reporting periodicity of X=500 ms, and the second network node can provide predictions at a reporting periodicity of Y=1000 ms, the second network node can indicate that the requested reporting period and/or validity time is either not supported or not available as well as the alternative (supported/avai lable) reporting periodicity and/or validity time in the response message.

[00126] The latter may lead to the first (requesting) network node sending another AI/ML ASSISTANCE DATA REQUEST message, either accepting the alternative set of parameters or cancelling the AI/ML assistance data reporting procedure.

[00127] In a related embodiment, the first network node may signal in the AI/ML Assistance Data Request message a list of preferred reporting periodicities, validity times, and/or other related parameters ordered by preference. In this way, if the option with the highest preference is not available or supported, but another is, the second network node can directly signal in the response message that this option is selected; thus, there is no need of a further message from the first network node as in the previous example. The second network (the reporting node) node also signals in the response message whether the highest priority set of information to be reported cannot be reported because such reporting is unavailable or not supported.

[00128] In a related embodiment, the first network node may signal in the AI/ML Assistance Data Request message a list of reporting periodicities, validity times and/or other related parameters with a granularity per reported measurement characteristic and/or per reported type of measurement characteristic, and/or per reported measurement object and/or per reported type of measurement object.

[00129] In one example, related to a granularity per measurement characteristic, a different reporting periodicity and/or validity time is indicated for Predicted Radio Resources and for Predicted Energy Efficiency. [00130] In one example, related to a granularity per type of measurement characteristic, a different reporting periodicity is indicated for Predicted Radio Resources and for UE performance Feedback/lndicator.

[00131] In one example, related to a granularity per measurement object, a first reporting periodicity and/or validity time are preferred for the reporting associated to first tier neighbor cells, and a second reporting periodicity is preferred for reporting of second tier neighbor cells.

[00132] In one example, related to a granularity per type of measurement object, a first reporting periodicity and/or validity time are preferred for cells of a first Radio Access Technology (e.g. NR), and a second reporting periodicity is preferred for cells of a second Radio Access Technology (e.g. LTE)).

[00133] Following the request as described above, the first network node may receive in the response message, say for instance an AI/ML Assistance Data Response message, and with the granularities indicated in the request message, corresponding lists of reporting periodicity, validity times or other parameters that the second network node accepts, or rejects, or alternatives to the values for the same parameters provided in the request message. For each information item that failed to be reported by the reporting node, the reporting node may include a reason for such failure, e.g. it could state whether the reporting time and/or reporting period for each of the requested information item is un-available or not supported.

[00134] Following an initialization phase, during the reporting of the requested information, the responding node may inform the requesting node that for a given measurement characteristic and/or type of measurement characteristics and/or measurement object and/or type of measurement object the previously agreed parameters to be reported cannot be satisfied due to unavailability of them. The reporting node may include reasons for such unavailability. For example, the reporting node may include an indication that the reporting is unavailable due to shortage of hardware or memory resources. Such indication may be provided to the requesting node via the message used to update the information reported. With this, the reporting node allows the requesting node to learn that, in certain conditions, the requested measurements become unavailable, which is related to capability learning.

[00135] In another option, a second network node which had previously accepted reporting the requested metrics at a specified periodicity, may have to suspend reporting all or a part of the reported information due to transient events such as high CPU load, interface overload, etc. An analogous situation may arise if a previously accepted validity time is no longer available, e.g., if the AI/ML model providing the predictions was disabled for bad performance. In such situations, the reporting node in one example may indicate to the requesting node about the changes via the AI/ML ASSISTANCE DATA UPDATE message which gives the possibility to the reporting node to indicate potential issues and hinders for providing the reporting. This message may contain information about the newly supported periodicities, validity times, or measurements objects. On reception of this message, the requesting node may initiate an AI/ML Assistance Data Request message with the suggested parameters. [00136] In an extension of the previously specified option about weights corresponding to the requested information, a second network node that will have to suspend part of the reported data, may take into account the suggested weights/importance of the requested objects and suspend those that have the least weights/importance prior to suspending reporting of other objects that have a higher priority. With such suspension, the reporting node issues a signaling to the requesting node, where it indicates why the information reporting was suspended. As an example, such indication may indicate that the information items that failed to be reported are no longer available due to e.g. resource shortage at the reporting node.

[00137] In a related embodiment, and if the first network node provided a list of preferred reporting periodicities, validity times, or other related parameters ordered by preference, the second network node may choose another of the less preferred options if the accepted one is no longer available. The second network node may indicate the change in the accepted option via the next AI/ML ASSISTANCE DATA UPDATE message.

[00138] In another example if the reporting node can report some information with one periodicity or in general set of parameters and some other information (probably partially overlapping) with another frequency or in general set of parameters, the reporting node will take under consideration the flags about critical information (if present), the weights and priority of the information (if present) as indicated by the requesting node and decide which set of parameters to use.

[00139] To summarize, the request message (e.g, AI/ML Assistance Data Request) can indicate, apart from the requested information items and reporting periodicity (and/or validity time for predictions), any of the following: i) a list with preferred reporting periodicities and/or validity times for predictions ii) a weight indicating level of importance of the requested info iii) priority of each requested info iv) indication to report measurements when available v) a flag that it accepts partial success for the metrics specified in the request (optional flags indicating which information is critical and should not result in a partial success).

[00140] The response message, i.e., AI/ML Assistance Data Response can indicate apart from the measurements that can be reported, the measurements that cannot be reported together with: cause value; IE indicating the reason why the measurement cannot be reported; estimated time when the info can be available, in case the measurement is supported and not available; alternative set of parameters, such as supported reporting periodicities and/or validity times; Indication of whether a given combination of requested information is not supported or not available to be reported together; Indication of whether the reporting period and/or the validity time requested on a per one or more information requested or for all of the information requested can be supported or not supported or whether it is available or not available.

[00141] Additionally, as per the AI/ML ASSISTANCE INFORMATION UPDATE, the reporting node may indicate whether the ongoing reporting of certain information becomes unavailable due to specific reasons, e.g. lack of resources.

[00142] A possible realization of the messages involved is given below:

[00143] The AI-ML ASSISTANCE DATA REQUEST

[00144] This message is sent by NG-RAN nodei to NG-RAN node2 to initiate the requested assistance data reporting in support to AI/ML functions, according to the parameters given in the message. The table 2 below shows the lEs that may be included in the message:

TABLE 2: lEs for the AI-ML ASSISTANCE DATA REQUEST

[00145] AI-ML ASSISTANCE DATA RESPONSE

[00146] This message is sent by NG-RAN node2 to NG-RAN nodei to indicate that the reporting for the requested assistance data, for all or for a subset of the measurement objects included in the assistance data request, is successfully initiated. If some requested objects cannot be reported, a list of cause values is included. The table 3 below shows the lEs that may be included in the message:

TABLE 3: lEs for the AI-ML ASSISTANCE DATA RESPONSE

[00147] AI-ML ASSISTANCE DATA UPDATE

[00148] This message is sent by NG-RAN node2 to NG-RAN nodei to report the results of the requested assistance data. The table 4 below shows the lEs that may be included in the message:

TABLE 4: lEs for the AI-ML ASSISTANCE DATA UPDATE

[00149] In other embodiments, the suggested message structure which allows for exchange of capability/support to exchange a part of the requested metrics/status can be used in other procedures where each node's capabilities can widely vary, and the nodes do not have to exchange more than one request or response to understand a neighbor's capability.

[00150] FIG. 5A is a flow chart illustrating a process 500, according to an embodiment, performed by a first network node (e.g., node-1). Process 500 may begin in step s502. Step s502 comprises the first network node transmitting to a second network node (e.g., node-2) a first request message for causing the second network node to initiate reporting of at least a first information item. Step s504 comprises the first network node receiving from the second network node a first response message responsive to the first request message. The first response message comprises: I) information indicating that the reporting of the first information item cannot currently be initiated and ii) information indicating an expected time at which reporting of the first information item can be initiated.

[00151] In some embodiments, the first response message further comprises cause information indicating a failure cause for the first information element for which the reporting cannot be initiated.

[00152] In some embodiments, the cause information indicates that reporting of the first information item is temporarily unavailable.

[00153] In some embodiments, the at least the first information item is used by the first network node as Artificial Intelligence, Al, or Machine Learning, ML, assistance data.

[00154] In some embodiments, the first request message comprises a bit string, one bit in the bit string corresponds to the first information item, the bit in the bit string corresponding to the first information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the first information item, another bit in the bit string corresponds to a second information item, and the bit in the bit string corresponding to the second information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the second information item.

[00155] In some embodiments, the first request message further comprises an instruction to the second network node to report at least the first information item when the first information item becomes available.

[00156] In some embodiments, the first request message comprises information indicating a reporting periodicity.

[00157] In some embodiments, the first request message further comprises an information element indicating a requested prediction validity time for a prediction of at least the first information item.

[00158] In some embodiments, the first request message is also for causing the second network node to initiate reporting of a second information item, the method further comprises receiving an update message transmitted by the second network node, and the update message comprises a measurement or prediction related to the second information item. [00159] In some embodiments, the first request message is also for causing the second network node to initiate reporting of a second information item, and the first response message further comprises a measurement or prediction related to the second information item.

[00160] In some embodiments, the first request message is also for causing the second network node to initiate reporting of a second information item, and the first request message further comprises information indicating that the second information item is a critical information item.

[00161] In some embodiments, the first request message further comprises weight information associated with the first information item and indicating a level of importance of the first information item.

[00162] In some embodiments, the first response message further comprises information indicating an alternative set of information items the second network node is capable of reporting.

[00163] In some embodiments, the first request message further comprises information indicating that the first network node accepts partial success outcomes.

[00164] FIG. 5B is a flow chart illustrating a process 550, according to an embodiment, performed by a second network node. Process 550 may begin in step s552. Step s552 comprises the second network node receiving, from a first network node, a first request message for causing the second network node to initiate reporting of at least a first information item. Step s554 comprises the second network node transmitting to the first network node a first response message responsive to the first request message, wherein the first response message comprises i) information indicating that the reporting of the first information item cannot currently be initiated and ii) information indicating an expected time at which reporting of the first information item can be initiated.

[00165] In some embodiments, the first response message further comprises cause information indicating a failure cause for the first information element for which the reporting cannot be initiated.

[00166] In some embodiments, the cause information indicates that reporting of the first information item is temporarily unavailable.

[00167] In some embodiments, the at least the first information item is used by the first network node as Artificial Intelligence, Al, or Machine Learning, ML, assistance data.

[00168] In some embodiments, the first request message comprises a bit string, one bit in the bit string corresponds to the first information item, the bit in the bit string corresponding to the first information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the first information item, another bit in the bit string corresponds to a second information item, and the bit in the bit string corresponding to the second information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the second information item. [00169] In some embodiments, the first request message further comprises an instruction to the second network node to report at least the first information item when the first information item becomes available.

[00170] In some embodiments, the first request message comprises information indicating a reporting periodicity.

[00171] In some embodiments, the first request message further comprises an information element indicating a requested prediction validity time for a prediction of at least the first information item.

[00172] In some embodiments, the first request message is also for causing the second network node to initiate reporting of a second information item, the method further comprises transmitting an update message to the first network node, and the update message comprises a measurement or prediction related to the second information item.

[00173] In some embodiments, the first request message is also for causing the second network node to initiate reporting of a second information item, and the first response message further comprises a measurement or prediction related to the second information item.

[00174] In some embodiments, the first request message is also for causing the second network node to initiate reporting of a second information item, and the first request message further comprises information indicating that the second information item is a critical information item.

[00175] In some embodiments, the first request message further comprises weight information associated with the first information item and indicating a level of importance of the first information item.

[00176] In some embodiments, the first response message further comprises information indicating an alternative set of information items the second network node is capable of reporting.

[00177] In some embodiments, the first request message further comprises information indicating that the first network node accepts partial success outcomes.

[00178] FIG. 6 is a block diagram of network node 600, according to some embodiments. As shown in FIG. 6, network node 600 may comprise: processing circuitry (PC) 602, which may include one or more processors (P) 655 (e.g., one or more general purpose microprocessors and/or one or more other processors, such as an application specific integrated circuit (ASIC), field-programmable gate arrays (FPGAs), and the like), which processors may be co-located in a single housing or in a single data center or may be geographically distributed (i.e., network node 600 may be a distributed computing apparatus); at least one network interface 648 (e.g., a physical interface or air interface) comprising a transmitter (Tx) 645 and a receiver (Rx) 647 for enabling network node 600 to transmit data to and receive data from other nodes connected to a network 110 (e.g., an Internet Protocol (IP) network) to which network interface 648 is connected (physically or wirelessly) (e.g., network interface 648 may be coupled to an antenna arrangement comprising one or more antennas for enabling network node 600 to wirelessly transmit/receive data); and a storage unit (a.k.a., "data storage system”) 608, which may include one or more nonvolatile storage devices and/or one or more volatile storage devices. In embodiments where PC 602 includes a programmable processor, a computer readable storage medium (CRSM) 642 may be provided. CRSM 642 may store a computer program (CP) 643 comprising computer readable instructions (CRI) 644. CRSM 642 may be a non- transitory computer readable medium, such as, magnetic media (e.g., a hard disk), optical media, memory devices (e.g., random access memory, flash memory), and the like. In some embodiments, the CRI 644 of computer program 643 is configured such that when executed by PC 602, the CRI causes network node 600 to perform steps described herein (e.g., steps described herein with reference to the flow charts). In other embodiments, network node 600 may be configured to perform steps described herein without the need for code. That is, for example, PC 602 may consist merely of one or more ASICs. Hence, the features of the embodiments described herein may be implemented in hardware and/or software.

[00179] For further clarification, advantages and technical effects of various embodiments of the present disclosure may be further clarified in view of the following, which is an excerpt of a change request for suggesting changes to 3GPP TS 38.423, titled "AI/ML capability discovery.”

[00180] Exchange of AI/ML supported use cases over Xn

[00181] In a previous meeting the following was agreed (see, e.g., reference [1]):

[00182] (1) Signalling describing the capability to support specific information predictions used for AI/ML is not pursued in this release;

[00183] (2) Signalling describing the capability to support specific AI/ML use cases is not pursued in this release; and

[00184] (3) AI/ML capability exchange in NG-RAN can be achieved by means of procedures for AI/ML information request, AI/ML information response and AI/ML Information Request Failure.

[00185] In addition, the following agreements ae captured in reference [2]:

[00186] (1) Introduce a new Class 1 procedure for initiating the reporting of AI/ML Related Information (e.g., predicted information) and a Class 2 procedure for Data Reporting of AI/ML Related Information (e.g., predicted information).

[00187] (2) The response message of the new procedure for AI/ML Related Information indicates if the requested measurement (e.g., prediction) can be provided. [00188] So, effectively, the NG-RAN node will learn about the AI/ML supported information of the other NG- RAN nodes by means of trial-and-error processes. When the NG-RAN node 1 sends a request to an NG-RAN node 2 to receive data, it learns if NG-RAN node 2 supports AI/ML for a specific use case based on the response. The NG- RAN node will gather the information about whether the other NG-RAN node supports AI/ML for a specific use case, depending on whether the request is accepted or rejected. This is in general terms the method already in use for other procedures.

[00189] This disclosure will now proceed to further analyse this process of AI/ML capability discovery. Three different cases can be identified.

[00190] First case is when the reporting node can provide all the information asked for by the requesting node. In that case the reporting node responds with AI/ML information response accepting to provide the information. Observation 1 : If the reporting node can provide all the information asked for by the requesting node, the reporting node responds with AI/ML information response accepting to provide the information.

[00191] Second case is when the reporting node cannot provide any of the information asked for by the requesting node. In that case the reporting node responds with AI/ML Information Request Failure rejecting the AI/ML information request with an appropriate cause value. Observation 2: If the reporting node cannot provide any of the information asked for by the requesting node, the reporting node responds with AI/ML Information Request Failure rejecting the AI/ML information request with an appropriate cause value.

[00192] The third case, which is less straightforward, is when the reporting node can provide some of the information asked for by the requesting node. In that case it is best that the reporting node responds with AI/ML information response indicating the information that can be provided, e.g., listing the information that cannot be provided with an appropriate cause value. In that way the requesting node will receive part of the requested information that could potentially be good enough and allow the requesting node to use it in at the Model Training and Model Inference functions. At the same time the requesting node will become aware, via the cause values for the information that cannot be provided, of the reasons for the unavailability of the information and decide analogously. For example, it might be conveyed that the measurements requested are not available, but that they may become available later. The requesting node may, in this case, decide to wait and make another request later. The reporting node may also indicate that the measurement is not supported, hence allowing the requesting node to understand that such measurement should not be requested again.

[00193] The approach described above was used in LTE. In fact, the Resource Status Reporting procedure over the X2AP allows for "partial success”, where the reporting node is able to list the Measurement Initiation Result IE, which indicates the measurements that were successfully configured and those that failed to be configured. [00194] If the same approach is followed for the XnAP, the AI/ML ASSISTANCE DATA REQUEST and AI/ML ASSISTANCE DATA RESPONSE messages can be encoded shown below.

[00195] AI-ML ASSISTANCE DATA REQUEST

[00196] This message is sent by NG-RAN nodei to NG-RAN node2 to initiate the requested assistance data reporting in support to AI/ML functions, according to the parameters given in the message.

[00197] Direction: NG-RAN nodei to NG-RAN node?:

[00198] AI-ML ASSISTANCE DATA RESPONSE [00199] This message is sent by NG-RAN node2 to NG-RAN nodei to indicate that the reporting for the requested assistance data, for all or for a subset of the measurement objects indicated in the bitmap included in the assistance data request, is successfully initiated.

Direction: NG-RAN nodej NG-RAN nodei:

[00200] As it can be seen, the AI/ML ASSISTANCE DATA REQUEST includes the Partial Success Indicator IE, which indicates whether the requesting node accepts reporting of only some of the requested measurements or whether the requesting node considers lack of reporting for some of the measurements as a condition to fail the procedure. In some scenarios, it is important to enable the requesting node to indicate such information because there might be AI/ML processes which may only be able to work if the full list of measurements requested is provided.

[00201] Hence, if the Partial Success Indicator IE is set to "partial success allowed”, the reporting node may be able to report a subset of the requested measurements and fail some others. The failed measurements can be listed in the Measurement Initiation Result IE together with a cause value indicating the reason for failure. As an example, cause value "Measurement Temporarily not Available” may be used to indicate that the measurement is supported but it is temporarily not available, while cause value "Measurement not Supported For The Object” may be used to indicate that the measurement is not supported.

[00202] Proposal 1 : Adopt a solution based on "Partial Success” for the new AI/ML procedure for data reporting.

[00203] Proposal 2: If the reporting node can provide some of the information asked for by the requesting node and if the requesting node accepts partial reporting, the reporting node responds with AI/ML information response listing the information that can be provided and also listing the information that cannot be provided with an appropriate cause value.

[00204] In this disclosure the issues of exchanging the AI/ML support for the use cases defined in TR 37.817 over Xn are discussed, and the following observations and proposals are presented:

[00205] Observation 1 : If the reporting node can provide all the information asked for by the requesting node, the reporting node responds with AI/ML information response accepting to provide the information.

[00206] Observation 2: If the reporting node cannot provide any of the information asked for by the requesting node, the reporting node responds with AI/ML Information Request Failure rejecting the AI/ML information request with an appropriate cause value.

[00207] Proposal 1 : Adopt a solution based on "Partial Success” for the new AI/ML procedure for data reporting.

[00208] Proposal 2: If the reporting node can provide some of the information asked for by the requesting node and if the requesting node accepts partial reporting, the reporting node responds with AI/ML information response listing the information that can be provided and also listing the information that cannot be provided with an appropriate cause value. [00209] For clarification, below is provided a list of numbered exemplary embodiments of the present disclosure.

[00210] Embodiment 1. A method performed by a first network node, the method comprising: transmitting to a second network node a first request message for causing the second network node to initiate reporting of at least a first information item; and receiving from the second network node a first response message responsive to the first request message, wherein the first response message comprises: information indicating that the reporting of the first information item cannot currently be initiated; and information indicating an expected time at which reporting of the first information item can be initiated.

[00211] Embodiment 2. The method of embodiment 1, wherein the first response message further comprises cause information indicating a failure cause for the first information element for which the reporting cannot be initiated.

[00212] Embodiment 3. The method of embodiment 2, wherein the cause information indicates that reporting of the first information item is temporarily unavailable.

[00213] Embodiment 4. The method of any one of embodiments 1-3, wherein the at least the first information item is used by the first network node as Artificial Intelligence, Al, or Machine Learning, ML, assistance data.

[00214] Embodiment 5. The method of any one of embodiments 1-4, wherein the first request message comprises a bit string, one bit in the bit string corresponds to the first information item, the bit in the bit string corresponding to the first information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the first information item, another bit in the bit string corresponds to a second information item, and the bit in the bit string corresponding to the second information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the second information item.

[00215] Embodiment 6. The method of any one of embodiments 1-5, wherein the first request message further comprises an instruction to the second network node to report at least the first information item when the first information item becomes available.

[00216] Embodiment 7. The method of any one of embodiments 1-6, wherein the first request message comprises information indicating a reporting periodicity. [00217] Embodiment 8. The method of any one of embodiments 1-7, wherein the first request message further comprises an information element indicating a requested prediction validity time for a prediction of at least the first information item.

[00218] Embodiment 9. The method of any one of embodiments 1-8, wherein the first request message is also for causing the second network node to initiate reporting of a second information item, the method further comprises receiving an update message transmitted by the second network node, and the update message comprises a measurement or prediction related to the second information item.

[00219] Embodiment 10. The method of any one of embodiments 1-8, wherein the first request message is also for causing the second network node to initiate reporting of a second information item, and the first response message further comprises a measurement or prediction related to the second information item.

[00220] Embodiment 11. The method of any one of embodiments 1-10, wherein the first request message is also for causing the second network node to initiate reporting of a second information item, and the first request message further comprises information indicating that the second information item is a critical information item.

[00221] Embodiment 12. The method of any one of embodiments 1-11, wherein the first request message further comprises weight information associated with the first information item and indicating a level of importance of the first information item.

[00222] Embodiment 13. The method of any one of embodiments 1-12, wherein the first response message further comprises information indicating an alternative set of information items the second network node is capable of reporting.

[00223] Embodiment 14. The method of any one of embodiments 1-12, wherein the first request message further comprises information indicating that the first network node accepts partial success outcomes.

[00224] Embodiment 15. A method performed by a second network node, the method comprising: receiving, from a first network node, a first request message for causing the second network node to initiate reporting of at least a first information item; and transmitting to the first network node a first response message responsive to the first request message, wherein the first response message comprises: information indicating that the reporting of the first information item cannot currently be initiated; and information indicating an expected time at which reporting of the first information item can be initiated.

[00225] Embodiment 16. The method of embodiment 15, wherein the first response message further comprises cause information indicating a failure cause for the first information element for which the reporting cannot be initiated. [00226] Embodiment 17. The method of embodiment 16, wherein the cause information indicates that reporting of the first information item is temporarily unavailable.

[00227] Embodiment 18. The method of any one of embodiments 15-17, wherein the at least the first information item is used by the first network node as Artificial Intelligence, Al, or Machine Learning, ML, assistance data.

[00228] Embodiment 19. The method of any one of embodiments 15-18, wherein the first request message comprises a bit string, one bit in the bit string corresponds to the first information item, the bit in the bit string corresponding to the first information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the first information item, another bit in the bit string corresponds to a second information item, and the bit in the bit string corresponding to the second information item is set to a value of 1 to indicate that the first network node is requesting the second network node to initiate the reporting of the second information item.

[00229] Embodiment 20. The method of any one of embodiments 15-19, wherein the first request message further comprises an instruction to the second network node to report at least the first information item when the first information item becomes available.

[00230] Embodiment 21. The method of any one of embodiments 15-20, wherein the first request message comprises information indicating a reporting periodicity.

[00231] Embodiment 22. The method of any one of embodiments 15-21, wherein the first request message further comprises an information element indicating a requested prediction validity time for a prediction of at least the first information item.

[00232] Embodiment 23. The method of any one of embodiments 15-22, wherein the first request message is also for causing the second network node to initiate reporting of a second information item, the method further comprises transmitting an update message to the first network node, and the update message comprises a measurement or prediction related to the second information item.

[00233] Embodiment 24. The method of any one of embodiments 15-23, wherein the first request message is also for causing the second network node to initiate reporting of a second information item, and the first response message further comprises a measurement or prediction related to the second information item.

[00234] Embodiment 25. The method of any one of embodiments 15-24, wherein the first request message is also for causing the second network node to initiate reporting of a second information item, and the first request message further comprises information indicating that the second information item is a critical information item. [00235] Embodiment 26. The method of any one of embodiments 15-25, wherein the first request message further comprises weight information associated with the first information item and indicating a level of importance of the first information item.

[00236] Embodiment 27. The method of any one of embodiments 15-26, wherein the first response message further comprises information indicating an alternative set of information items the second network node is capable of reporting.

[00237] Embodiment 28. The method of any one of embodiments 15-26, wherein the first request message further comprises information indicating that the first network node accepts partial success outcomes.

[00238] Embodiment 29. A computer program comprising instructions which when executed by processing circuitry of a network node causes the network node to perform the method of any one of embodiments 1-28.

[00239] Embodiment 30. A carrier containing the computer program of embodiment 29, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, and a computer readable storage medium.

[00240] Embodiment 31 . A first network node configured to perform a method that comprises: transmitting to a second network node a first request message for causing the second network node to initiate reporting of at least a first information item; and receiving from the second network node a first response message responsive to the first request message, wherein the first response message comprises: information indicating that the reporting of the first information item cannot currently be initiated; and information indicating an expected time at which reporting of the first information item can be initiated.

[00241] Embodiment 32. The first network node of embodiment 31, wherein the first network node is further configured to perform the method of any one of embodiments 2-14.

[00242] Embodiment 33. A second network node configured to perform a method that comprises: receiving, from a first network node, a first request message for causing the second network node to initiate reporting of at least a first information item; and transmitting to the first network node a first response message responsive to the first request message, wherein the first response message comprises: information indicating that the reporting of the first information item cannot currently be initiated; and information indicating an expected time at which reporting of the first information item can be initiated.

[00243] Embodiment 34. The second network node of embodiment 33, wherein the second network node further is configured to perform a method of any one of embodiments 16-28. [00244] For further clarification, below is provided a list of currently not claimed, numbered exemplary embodiments of the present disclosure.

[00245] Embodiment A1 . A method 700 (see FIG. 7A) performed by a first network node, the method comprising: transmitting (step s702) to second network node a first request message requesting at least a first information item (e.g., an energy efficiency value) and a second information item (e.g., a set of UE performance measurements); and receiving (step s704) from the second network node a first response message responsive to the first request message, wherein the first response message comprises the requested first information item but not the requested second information item, wherein the first response message further comprises information indicating a reason for why the first response message does not include the requested second information item.

[00246] Embodiment A2. The method of embodiment A1, wherein the information indicating a reason for why the first response message does not include the requested second information item comprises a cause value indicating the reason.

[00247] Embodiment A3. The method of embodiment A1 or A2, wherein the information indicating a reason for why the first response message does not include the requested second information item comprises information indicating that the second information item is currently not available, but may become available at a later point in time.

[00248] Embodiment A4. The method of embodiment A3, wherein the first response message further comprises information indicating a point in time at which the second network node expects the second information item to become available.

[00249] Embodiment A5. The method of embodiment A3 or A4, further comprising: as a result of receiving the information indicating that the second information item is currently not available, but may become available at a later point in time, the first network node transmitting to the second network node a second request message requesting at least the second information item; and receiving from the second network node a second response message responsive to the second request message, wherein the second response message comprises the requested second information item.

[00250] Embodiment A6. The method of any one of embodiments A1-A5, wherein the first request message further comprises an instruction to the second network node to report the first and/or second information item when it becomes available.

[00251] Embodiment A7. The method of any one of embodiments A1-A6, wherein the first request message further comprises an indication that the first network node accepts partial responses (i.e., the request indicates that the second network node should send a response if at least one requested information item is available).

[00252] Embodiment A8. The method of any one of embodiments A1-A7, wherein the first request message comprises a bit string, one bit in the bit string corresponds to the first information item, the bit in the bit string corresponding to the first information item is set to a value of 1 to indicate that the first network node is requesting the second network node to provide the first information item to the first network node, another bit in the bit string corresponds to second first information item, and the bit in the bit string corresponding to the second information item is set to a value of 1 to indicate that the first network node is requesting the second network node to provide the second information item to the first network node.

[00253] Embodiment A9. The method of any one of embodiments A1-A8, wherein the first request message comprises information indicating a reporting periodicity.

[00254] Embodiment A10. The method of embodiment A9, wherein the information indicating a reason for why the first response message does not include the requested second information item specifies that the second network node does not support reporting the second information item with the requested reporting periodicity.

[00255] Embodiment A11 . A method 750 (see FIG. 7B) performed by a second network node, the method comprising: receiving (step s752) from a first network node a first request message requesting at least an information item of a first type (e.g., a UE Performance Indicator) and a second information item of a second type (e.g., information specifying a current energy efficiency); obtaining a first information item of the first type; generating a first response message responsive to the first request message, wherein the first response message comprises i) the first information item and ii) information indicating a reason for why the first response message does not include an information item of the second type.

[00256] Embodiment A12. The method of embodiment A11, wherein the information indicating a reason for why the first response message does not include the requested second information item comprises a cause value indicating the reason.

[00257] Embodiment A13. The method of embodiment A11 or A12, wherein the information indicating a reason for why the first response message does not include the requested second information item comprises information indicating that an information items of the second type is currently not available, but may become available at a later point in time.

[00258] Embodiment A14. The method of embodiment A13, wherein the first response message further comprises information indicating a point in time at which the second network node expects the information item of the second type to become available. [00259] Embodiment A15. The method of embodiment A13 or A14, further comprising: receiving from the first network node a second request message requesting at least an information item of the second type; and transmitting to the first network node a second response message responsive to the second request message, wherein the second response message comprises the requested information item.

[00260] Embodiment A16. The method of any one of embodiments A11-A15, wherein the first request message further comprises an instruction to the second network node to report the first and/or second information item when it becomes available.

[00261] Embodiment A17. The method of any one of embodiments A11-A16, wherein the first request message further comprises an indication that the first network node accepts partial responses (i.e., the request indicates that the second network node should send a response if at least one requested information item is available).

[00262] Embodiment A18. The method of any one of embodiments A11-A17, wherein the first request message comprises a bit string, one bit in the bit string corresponds to the first information item, the bit in the bit string corresponding to the first information item is set to a value of 1 to indicate that the first network node is requesting the second network node to provide the first information item to the first network node, another bit in the bit string corresponds to second first information item, and the bit in the bit string corresponding to the second information item is set to a value of 1 to indicate that the first network node is requesting the second network node to provide the second information item to the first network node.

[00263] Embodiment A19. The method of any one of embodiments A11-A18, wherein the first request message comprises information indicating a reporting periodicity.

[00264] Embodiment A20. The method of embodiment A19, wherein the information indicating a reason for why the first response message does not include the requested second information item specifies that the second network node does not support reporting the second information item with the requested reporting periodicity.

[00265] Embodiment A21 . A computer program 643 comprising instructions 644 which when executed by processing circuitry 602) of a network node causes the network node to perform the method of any one of embodiments AA1-A20.

[00266] Embodiment A22. A carrier containing the computer program of embodiment A21, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, and a computer readable storage medium 642.

[00267] Embodiment A23. A first network node configured to perform a process that includes: transmitting (step s702) to second network node a first request message requesting at least a first information item (e.g., an energy efficiency value) and a second information item (e.g., a set of UE performance measurements); and receiving (step s704) from the second network node a first response message responsive to the first request message, wherein the first response message comprises the requested first information item but not the requested second information item, wherein the first response message further comprises information indicating a reason for why the first response message does not include the requested second information item.

[00268] Embodiment A24. A second network node configured to perform a process that includes: receiving (step s752) from a first network node a first request message requesting at least an information item of a first type (e.g., a UE Performance Indicator) and a second information item of a second type (e.g., information specifying a current energy efficiency); obtaining a first information item of the first type; generating a first response message responsive to the first request message, wherein the first response message comprises i) the first information item and ii) information indicating a reason for why the first response message does not include an information item of the second type.

[00269] Conclusion

[00270] The embodiments provide the possibility to provide requested information items (data, metrics, predictions, etc.) from one node to another in the context of AI/ML assistance data where there is not a complete support to exchange all of the requested information items.

[00271] Embodiments also support the case where there is full support for all metrics specified in the request and in the case of partial support, proposes an alternative to exchange partial capabilities in contrary to failure being the only option. This is done to enable a faster learning at the requesting node of the capabilities of the reporting node

[00272] Embodiments also provide the possibility of partial reporting of the requested AI/ML assistance data. Embodiments equip the NG-RAN nodes with the ability to efficiently exchange AI/ML assistance data supplying the flexibility lacking in the current procedure. Embodiments provide a method to the NG-RAN nodes to advantageously report the available AI/ML assistance data in the cases that would otherwise be terminated with a failure procedure.

[00273] While various embodiments are described herein, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of this disclosure should not be limited by any of the above-described exemplary embodiments. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the disclosure unless otherwise indicated herein or otherwise clearly contradicted by context.

[00274] Additionally, as used herein, transmitting a message to a device encompasses transmitting the message directly to the device or transmitting the message indirectly to the device (i.e., one or more nodes are used to relay the message from the source to the device). Likewise, as used herein, receiving a message from a device encompasses receiving the message directly from the device or indirectly from the device (i.e., one or more nodes are used to relay the message from the device to the receiving node).

[00275] Also, while the processes described above and illustrated in the drawings are shown as a sequence of steps, this was done solely for the sake of illustration. Accordingly, it is contemplated that some steps may be added, some steps may be omitted, the order of the steps may be re-arranged, and some steps may be performed in parallel.

[00276] References

[00277] [1] R3-225916, "SoD on AI AN1_General_Stage2”, 3GPP TSG-RAN WG3 Meeting #117bis-e;

[00278] [2] R3-225917, "Summary of email Discussion on CB: # Al RAN2_Xn Impact”, 3GPP TSG-RAN WG3 Meeting #117bis-e.

[00279] Abbreviations: