Handling of applicability of ai / ML information for inactive state

Abstract

According to an example, a method performed by a User Equipment (UE) for storing or deleting information is provided The UE has a functionality that uses an artificial intelligence or machine learning (AI / ML) model. The method comprises transitioning to an inactive state, and storing or deleting at least part of information associated with the functionality.

Description

[0001] APPLICABILITY HANDLING FOR INACTIVE STATE

[0002] Background

[0003] Artificial Intelligence (Al) and Machine Learning (ML) have been investigated, both in academia and industry, as promising tools to optimize the design of the air-interface in wireless communication networks. Example use cases include using autoencoders for Channel State Information (CSI) compression to reduce the feedback overhead and improve channel prediction accuracy; using deep neural networks for classifying Line-of-Sight (LOS) and Non-LOS (NLOS) conditions to enhance the positioning accuracy; and using reinforcement learning for beam selection at the network side and / or the User Equipment (UE) side to reduce the signaling overhead and beam alignment latency; using deep reinforcement learning to learn an optimal precoding policy for complex Multiple Input Multiple Output (MIMO) precoding problems.

[0004] In 3rd Generation Partnership Project (3GPP) New Radio (NR) standardization work, a new Release 18 (Rel.18) study item on AI / ML for the New Radio (NR) air interface started in May 2022. This study item explored the benefits of augmenting the air-interface with features enabling improved support of AI / ML based algorithms for enhanced performance and / or reduced complexity / overhead. Through studying a few selected use cases (CSI feedback, beam management, and positioning), this study item aims at laying the foundation for future air-interface use cases leveraging AI / ML techniques. The analysis carried out during Rel.18 is now considered in the context of Rel.19 work item, see for example RP-234039, New WID on Artificial Intelligence (Al) / Machine Learning (ML) for NR Air Interface, Source: Qualcomm (Moderator), 3GPP TSG RAN Meeting #102, Edinburgh, Scotland, December 11-15, 2023. Additionally, during the rel. 19, a new study item addressing AIML for mobility has been approved. In the context of this new study item, 3GPP will investigate methods for cell-level measurement predictions, and mobility event predictions (e.g. RLF, handover failure, mobility-related events predictions such as A3 / A5), see for example RP-234055, Study on Artificial Intelligence (Al) / Machine Learning (ML) for mobility in NR, 3GPP TSG RAN Meeting #102, Edinburgh, GB, December 11-15, 2023.

[0005] Applicability reporting has been discussed during the Rel.18 study item, to allow a User Equipment (UE) to inform the gNodeB (gNB) about the applicability of an AIML model / functionality while the UE is connected to this gNB. An AIML model / functionality may be applicable or not depending on a number of factors, so called applicability conditions, that are only partly under the control of the gNB. For example, whether the UE has an AI / ML model that is applicable given the current location of the UE, or given the current speed of the UE, is not something that the network can control or it can know, because typically it is assumed that the UE-side model is not trained and generated by the gNB. Rather, it is typically assumed that the UE-side model is trained and generated by a node outside the RAN, such as an OTT server or core network (CN) function controlled by the UE-vendor or by the Mobile Network Operator (MNO).

[0006] Two types of applicability reporting were identified during the Rel.18 study item and are currently being discussed in RAN2 for the normative phase, the so-called reactive approach and the proactive approach. In the proactive approach, the network enquires the UE capabilities and configures the UE to report the applicability of an AI / ML functionality and, based on the reported information the network configures the UE with an inference configuration, as shown in Figure 1 , which illustrates an example of proactive reporting of applicability, which includes the following steps:

[0007] Step 1 : Network sends UECapabilityEnqiry message to initiate the procedure to a UE reporting its AI / ML supported functionalities

[0008] Step 2: UE sends UECapablitylnformation message to network, containing supported functionalities at the UE side

[0009] Step 3: Network configures UE that it is allowed to provide its applicable functionalities

[0010] Step 4: UE sends applicable functionalities to network upon change of applicable functionality / condition

[0011] Step 5: Network sends inference configuration for the applicable functionalities to the UE

[0012] Step 6: Start inference / monitoring based on network / UE activation / deactivation

[0013] Figure 2 illustrates an example of reactive reporting of applicability. In this example reactive approach, the network enquires the UE capabilities and configures the UE with the AI / ML functionality (possibly including the inference configuration) in response to which the UE is able to determine the applicability of the A / ML functionality and, in case the configured AI / ML functionality is applicable, the functionality could be up and running as soon as possible, without the need to an additional reconfiguration, as follows:

[0014] Step 1 : Network sends UECapabilityEnquiry message to initiate the procedure to a UE reporting its AI / ML supported functionalities. Step 2: UE sends UECapablitylnformation message to network, containing supported functionalities at the UE side.

[0015] Step 3: Network provides network configurations and initiates UE to report its applicable functionalities.

[0016] Step 4: UE sends applicable functionalities to network.

[0017] Step 5: Network sends updated inference configuration for applicable functionalities reported in Step 4 to the UE.

[0018] Step 6: Start inference / monitoring based on network / UE activation / deactivation.

[0019] In New Radio (NR), a UE may be in RRC_CONNECTED, RRCJDLE or RRCJNACTIVE. When the UE is in RRCJNACTIVE the following is supported:

[0020] • PLMN selection;

[0021] • Broadcast of system information;

[0022] • Cell re-selection mobility;

[0023] • Paging is initiated by NG-RAN (RAN paging);

[0024] • RAN-based notification area (RNA) is managed by NG- RAN;

[0025] • DRX for RAN paging configured by NG-RAN;

[0026] • 5GC - NG-RAN connection (both C / U-planes) is established for UE;

[0027] • The UE AS context is stored in NG-RAN and the UE;

[0028] • NG-RAN knows the RNA which the UE belongs to.

[0029] When the UE is in RRCJNACTIVE the UE remains in CM-CONNECTED and can move within an area configured by NG-RAN (the RNA) without notifying NG-RAN. In RRCJNACTIVE, the last serving gNB node keeps the UE context and the UE-associated NG connection with the serving AMF and UPF. The UE context (also called UE Inactive AS context) includes a subset of the configuration(s) the UE has stored when the UE is in RRC_CONNECTED.

[0030] There currently exist certain challenges. For example, the framework for the applicability report has been described for a UE connected to the current Primary Cell (PCell) i.e. in case the AI / ML functionality is associated to that PCell. However, when the UE is transitioned to an Inactive state the UE may either resume in the same cell and / or resume in another cell, for which a previously activated AI / ML functionality may still be considered activated, and / or for which a previously configured AI / ML functionality may still be considered valid, and / or for which a previously applicable AI / ML functionality may still be considered applicable. For example, at the time of resuming the RRC connection, the applicability conditions of the network and / or of the UE may have changed, and the network may not if an AIML model / functionality previously configured and / or activated is still applicable or not.

[0031] In other words, without addressing this scenario, it is not clear how the UE should behave when the UE transitions to Inactive state (e.g. RRCJNACTIVE)

[0032] This problem is related to the Work Item on AI / ML for PHY part of Rel-19 and 5G evolution. However, the same problem is existing in the AI / ML for Mobility feature which may be specified in Rel-20 (ongoing Study Item), and possible specifications for 6G, for Rel-10 and Rel-21 related to AI / ML.

[0033] Summary

[0034] Certain embodiments may provide one or more of the following technical advantage(s). For example, according to examples of this disclosure, the UE stores (or selectively discards) one or more information associated to an AI / ML functionality when it transitions to Inactive state so that when the UE is resuming the connection the UE does not need to receive further AI / ML configurations to have an AI / ML functionality up and running, and / or the UE does not need to further report an applicability information of an AI / ML functionality, and / or the UE does not need to further receive a command to activate (or deactivate) an AI / ML functionality. The teachings of certain embodiments may improve network performance.

[0035] One aspect of the present disclosure provides a method performed by a User Equipment, UE, for storing or deleting information. The UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model. The method comprises transitioning to an inactive state, and storing or deleting at least part of information associated with the functionality.

[0036] Another aspect of the present disclosure provides a method performed by a User Equipment, UE, for restoring information. The UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model. The method comprises transitioning to a connected state, and restoring at least part of information associated with the functionality.

[0037] A further aspect of the present disclosure provides a method performed by a first Radio Access Network, RAN, node for storing or deleting information. The method comprises determining that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, and storing or deleting at least part of information associated with the functionality.

[0038] A still further aspect of the present disclosure provides a method performed by a first Radio Access Network, RAN, node for determining information. The method comprises determining that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, and determining at least part of information associated with the functionality.

[0039] Another aspect of the present disclosure provides a tangible, non-transient computer- readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations a User Equipment, UE, for storing or deleting information. The UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model. The operations comprise transitioning to an inactive state, and storing or deleting at least part of information associated with the functionality.

[0040] An additional aspect of the present disclosure provides a tangible, non-transient computer- readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations a User Equipment, UE, for restoring information. The UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model. The operations comprise transitioning to a connected state, and restoring at least part of information associated with the functionality.

[0041] Another aspect of the present disclosure provides a tangible, non-transient computer- readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations in a first Radio Access Network, RAN, node for storing or deleting information. The operations comprise determining that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, and storing or deleting at least part of information associated with the functionality.

[0042] A further aspect of the present disclosure provides a tangible, non-transient computer- readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations in a first Radio Access Network, RAN, node for determining information. The operations comprise determining that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, and determining at least part of information associated with the functionality. A further aspect of the present disclosure provides apparatus in a User Equipment, UE, for storing or deleting information. The UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model. The apparatus comprises processing circuitry and a memory. The apparatus is configured to transition to an inactive state, and store or delete at least part of information associated with the functionality.

[0043] Another aspect of the present disclosure provides apparatus in a User Equipment, UE, for restoring information. The UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model. The apparatus comprises processing circuitry and a memory. The apparatus is configured to transition to a connected state, and restore at least part of information associated with the functionality.

[0044] A further aspect of the present disclosure provides apparatus in a first Radio Access Network, RAN, node for storing or deleting information. The apparatus comprises processing circuitry and a memory. The apparatus is configured to determine that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, and store or delete at least part of information associated with the functionality.

[0045] An additional aspect of the present disclosure provides apparatus in a first Radio Access Network, RAN, node for determining information. The apparatus comprises processing circuitry and a memory. The apparatus is configured to determine that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and determine at least part of information associated with the functionality.

[0046] Brief Description of the Drawings

[0047] For a better understanding of the embodiments of the present disclosure, and to show how it may be put into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0048] Figure 1 illustrates an example of proactive reporting of applicability;

[0049] Figure 2 illustrates an example of reactive reporting of applicability;

[0050] Figure 3 depicts an example of a method performed by a UE for storing or deleting information;

[0051] Figure 4 depicts an example of a method performed by a UE for restoring information;

[0052] Figure 5 depicts an example of a method performed by a first RAN node for storing or deleting information; Figure 6 depicts an example of a method performed by a first RAN node for determining information;

[0053] Figure 7 illustrates an example of communications in a network in a suspend / resume procedure;

[0054] Figure 8 illustrates another example of communications in a network in a suspend / resume procedure;

[0055] Figure 9 illustrates another example of communications in a network according to examples of this disclosure;

[0056] Figure 10 illustrates another example of communications in a network according to examples of this disclosure;

[0057] Figure 11 shows an example of a communication system in accordance with some embodiments;

[0058] Figure 12 shows a UE in accordance with some embodiments;

[0059] Figure 13 shows a network node in accordance with some embodiments; and

[0060] Figure 14 is a block diagram illustrating a virtualization environment in which functions implemented by some embodiments may be virtualized.

[0061] Detailed Description

[0062] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0063] Figure 3 depicts a method 300 in accordance with particular embodiments, such as for example method performed by a User Equipment (UE) for storing or deleting information, wherein the UE has a functionality that uses an artificial intelligence or machine learning (AI / ML) model. The method 300 may be performed by a UE or wireless device (e.g. the UE QQ112 or UE QQ200 as described later with reference to Figures 11 and 12 respectively). The method begins at step 302 with transitioning to an inactive state, and step 304 with storing or deleting at least part of information associated with the functionality. The functionality that uses the AI / ML model may comprise for example a UE positioning functionality

[0064] In some examples, the method 300 may comprise receiving a message on a first cell and / or from a first Radio Access Network, RAN, node, wherein transitioning to the inactive state is performed after or in response to the message. The message may comprise for example a a Radio Resource Control, RRC, message or a RRC release message. The message may in some examples include a suspend configuration, an instruction for the UE to perform a suspend procedure, and / or an instruction for the UE to transition to the inactive state. The message may in some examples include an indication of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality. The message may for example identify the at least part of the information associated with the functionality.

[0065] In some examples, the information associated with the functionality comprises at least one of the following non-limiting examples:

[0066] • a configuration of the functionality;

[0067] • applicability of the functionality;

[0068] • a state of the functionality.

[0069] Storing or deleting at least part of the information associated with the functionality may comprise one or more of the following non-limiting examples:

[0070] • storing or deleting the configuration of the functionality;

[0071] • storing or deleting the applicability of the functionality;

[0072] • storing or deleting the state of the functionality.

[0073] The configuration of the functionality comprises one or more of a training configuration, a monitoring configuration, an applicability reporting configuration and / or an inference configuration. Additionally or alternatively, the applicability of the functionality may comprise for example an indication of whether the functionality is applicable or not applicable. Additionally or alternatively, applicability of the functionality may comprise for example a configuration of the UE with which the functionality is applicable. Additionally or alternatively, the status of the functionality may comprise for example an indication of whether the functionality is active or not active. The message may indicate one or more of the following non-limiting examples:

[0074] • whether to store or delete the configuration of the functionality;

[0075] • whether to store or delete the applicability of the functionality;

[0076] • whether to store or delete the state of the functionality.

[0077] System information (SI) may indicate one or more of the following non-limiting examples:

[0078] • whether to store or delete the configuration of the functionality;

[0079] • whether to store or delete the applicability of the functionality;

[0080] • whether to store or delete the state of the functionality. In some examples, system information includes an indication of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality. The system information may also identify the at least part of the information associated with the functionality in some examples.

[0081] The method 300 may in some examples comprise storing the at least part of the information associated with the functionality in an inactive access stratum, AS, context for the UE.

[0082] Transitioning to the inactive state may be performed in some examples as part of a suspend procedure. Additionally or alternatively, transitioning to the inactive state may be performed in some examples from a connected state. The connected state may comprise a Radio Resource Control, RRC, Connected state. The inactive state may comprise a Radio Resource Control, RRC, Inactive state.

[0083] In some examples, transitioning to the inactive state in step 302 may be performed when the UE is connected to a first cell or a first Radio Access Network, RAN, node. The emthdo 300 may also in some examples comprise performing cell selection to the first cell or the first RAN node, and may also comprise performing cell reselection to a second cell or a second RAN node.

[0084] The method 300 may in some examples comprise transitioning to a connected state. In such examples, the method 300 may also comprise receiving a resume message, wherein transitioning to the connected state is performed after or in response to the resume message. The resume message may be received for example on the first cell or from the first RAN node, or on the second cell or from the second RAN node. After transitioning to the connected state, the method 300 may in some examples comprise at least one of the following:

[0085] • determining that the functionality is applicable if the stored at least part of the information indicates that the functionality is applicable and the UE is connected to the first cell or the first RAN node;

[0086] • determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is applicable and the UE is connected to the second cell or the second RAN node;

[0087] • determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable;

[0088] • determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable; • determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is not applicable.

[0089] After transitioning to the connected state, the method 300 may in some examples comprise determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE.

[0090] In some examples storing or deleting at least part of the information associated with the functionality in step 304 may comprises one or more of the following:

[0091] • storing the at least part of the information;

[0092] • starting a timer; and

[0093] • deleting the at least part of the information on expiry of the timer.

[0094] The method 300 may in some examples comprise deleting the stored at least part of the information after one or more of the following non-limiting examples:

[0095] • transitioning to an idle state;

[0096] • performing a setup procedure;

[0097] • failure of a resume procedure.

[0098] Figure 4 depicts a method 400 in accordance with particular embodiments, such as for example a method performed by a User Equipment (UE) for restoring information, wherein the UE has a functionality that uses an artificial intelligence or machine learning (AI / ML) model. The method 400 may be performed by a UE or wireless device (e.g. the UE QQ112 or UE QQ200 as described later with reference to Figures 11 and 12 respectively). The method begins at step 402 with transitioning to a connected state, and step 404 with restoring at least part of information associated with the functionality, e.g. from local or on- device storage. The functionality that uses the AI / ML model may comprises for example a UE positioning functionality.

[0099] In some examples the information associated with the functionality comprises at least one of the following non-limiting examples:

[0100] • a configuration of the functionality;

[0101] • applicability of the functionality;

[0102] • a state of the functionality.

[0103] Restoring at least part of the information associated with the functionality may comprises one or more of the following non-limiting examples: • restoring the configuration of the functionality;

[0104] • restoring the applicability of the functionality;

[0105] • restoring the state of the functionality.

[0106] The configuration of the functionality may comprise for example one or more of a training configuration, a monitoring configuration, an applicability reporting configuration and / or an inference configuration. The applicability of the functionality may comprise for example an indication of whether the functionality is applicable or not applicable. The status of the functionality may comprise for example an indication of whether the functionality is active or not active.

[0107] In some examples, the method 400 may in some examples comprise restoring the at least part of the information associated with the functionality from an inactive access stratum, AS, context for the UE. Transitioning to the connected state in step 402 may in some examples be performed as part of a suspend procedure, and / or may be performed from an inactive state. The inactive state may comprise for example a Radio Resource Control, RRC, Inactive state. The connected state may comprise for example a Radio Resource Control, RRC, Connected state.

[0108] Transitioning to the connected state in step 402 of the method 400 may in some examples be performed to connect to a first cell or a first Radio Access Network, RAN, node. The method 400 may also in some examples comprise performing cell selection to the first cell or the first RAN node.

[0109] In some examples, transitioning to the connected state in step 402 of the method 400 may be performed to connect to a second cell or a second Radio Access Network, RAN, node. The method 400 may also in some examples comprise performing cell reselection to the second cell or the second RAN node.

[0110] In some examples, the method 400 comprises, before transitioning to the connected state, receiving a resume message, wherein transitioning to the connected state is performed after or in response to the resume message. The resume message may be received for example, on the first cell or from the first RAN node, or on the second cell or from the second RAN node. The method 400 may also in some examples comprise, after transitioning to the connected state, at least one of the following examples:

[0111] • determining that the functionality is applicable if the restored at least part of the information indicates that the functionality is applicable and the UE is connected to the first cell or the first RAN node; • determining that the functionality is not applicable if the restored at least part of the information indicates that the functionality is applicable and the UE is connected to the second cell or the second RAN node;

[0112] • determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable;

[0113] • determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable;

[0114] • determining that the functionality is not applicable if the restored at least part of the information indicates that the functionality is not applicable.

[0115] In some examples, the method 400 may comprise, after transitioning to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE. The method 400 may also comprise sending an indication of the applicability of the functionality. The method 400 may also comprise sending the indication of the applicability of the functionality on the first cell or to the first network node, or on the second cell or to the second network node.

[0116] Figure 5 depicts a method 500 in accordance with particular embodiments, such as for example a method performed by a first Radio Access Network (RAN) node for storing or deleting information. The method 500 may be performed by a network node (e.g. the network node QQ110 or network node QQ300 as described later with reference to Figures 11 and 13 respectively). The method begins at step 502 with determining that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning (AI / ML) model, and step 504 with storing or deleting at least part of information associated with the functionality. The functionality that uses the AI / ML model may comprise for example a UE positioning functionality.

[0117] In some examples, the method 500 comprises determining that the UE transitions to the inactive state comprises sending a message to the UE, wherein the UE transitions to the inactive state after or in response to the message. The message may comprise for example a Radio Resource Control, RRC, message or a RRC release message.

[0118] In some examples, the message includes a suspend configuration, an instruction for the UE to perform a suspend procedure, and / or an instruction for the UE to transition to the inactive state. Additionally or alternatively, in some examples, the message includes an indication to the UE of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality. The message may for example identify the at least part of the information associated with the functionality.

[0119] The information associated with the functionality may comprise at least one of the following non-limiting examples:

[0120] • a configuration of the functionality;

[0121] • applicability of the functionality;

[0122] • a state of the functionality.

[0123] Storing or deleting at least part of the information associated with the functionality may comprise one or more of the following non-limiting examples:

[0124] • storing or deleting the configuration of the functionality;

[0125] • storing or deleting the applicability of the functionality;

[0126] • storing or deleting the state of the functionality.

[0127] The configuration of the functionality may comprise for example one or more of a training configuration, a monitoring configuration, an applicability reporting configuration and / or an inference configuration. Additionally or alternatively, the applicability of the functionality comprises an indication of whether the functionality is applicable or not applicable, and / or comprises a configuration of the UE with which the functionality is applicable. The status of the functionality may for example comprise an indication of whether the functionality is active or not active.

[0128] The message may in some examples indicate to the UE one or more of the following nonlimiting examples:

[0129] • whether to store or delete the configuration of the functionality;

[0130] • whether to store or delete the applicability of the functionality;

[0131] • whether to store or delete the state of the functionality.

[0132] The method 400 may in some examples broadcasting system information that indicates one or more of the following non-limiting examples:

[0133] • whether to store or delete the configuration of the functionality;

[0134] • whether to store or delete the applicability of the functionality;

[0135] • whether to store or delete the state of the functionality. In some examples, the method 500 comprises broadcasting system information that includes an indication of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality. The system information may for example identify the at least part of the information associated with the functionality. In some examples, the method 500 comprises storing the at least part of the information associated with the functionality in an inactive access stratum, AS, context for the UE.

[0136] The UE transitioning to the inactive state may in some examples be performed as part of a suspend procedure, and / or may be performed from a connected state. The connected state comprises for example a Radio Resource Control, RRC, Connected state, and / or the inactive state comprises a Radio Resource Control, RRC, Inactive state.

[0137] In some examples, the method 500 may comprise determining that the UE transitions to a connected state. The method 500 may also comprise receiving a resume message from the UE or sending a resume message to the UE, wherein transitioning to the connected state is performed after or in response to the resume message. The method 500 may also comprise, after the UE transitions to the connected state, at least one of the following:

[0138] • determining that the functionality is applicable if the stored at least part of the information indicates that the functionality is applicable;

[0139] • determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable;

[0140] • determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable;

[0141] • determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is not applicable.

[0142] The method 500 may also in some examples comprise, after the UE transitions to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE.

[0143] In some examples, the method 500 may comprise deleting the stored at least part of the information after one or more of the following non-limiting examples:

[0144] • the UE transitions to an idle state;

[0145] • performing a setup procedure;

[0146] • failure of a resume procedure. Figure 6 depicts a method 600 in accordance with particular embodiments, for example a method performed by a first Radio Access Network, RAN, node for determining information. The method 600 may be performed by a network node (e.g. the network node QQ110 or network node QQ300 as described later with reference to Figures 11 and 13 respectively). The method begins at step 602 with determining that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning (AI / ML) model, and step 604 with determining at least part of information associated with the functionality.

[0147] The information associated with the functionality may comprise at least one of the following non-limiting examples:

[0148] • a configuration of the functionality;

[0149] • applicability of the functionality;

[0150] • a state of the functionality.

[0151] Determining the at least part of the information associated with the functionality in step 604 of the method 600 may comprise one or more of the following non-limiting examples:

[0152] • determining the configuration of the functionality;

[0153] • determining the applicability of the functionality;

[0154] • determining the state of the functionality.

[0155] The configuration of the functionality may for example comprise one or more of a training configuration, a monitoring configuration, an applicability reporting configuration and / or an inference configuration. The applicability of the functionality may comprise for example an indication of whether the functionality is applicable or not applicable. The applicability of the functionality may comprise for example a configuration of the UE with which the functionality is applicable. The status of the functionality may comprise for example an indication of whether the functionality is active or not active.

[0156] In some examples, determining the at least part of the information associated with the functionality in step 604 of the method 600 comprises determining the at least part of the information associated with the functionality from an inactive access stratum, AS, context for the UE. The inactive AS context for the UE may in some examples be stored at the first RAN node. The method 600 may also comprise receiving the inactive AS context for the UE from a second RAN node or another network node. The second RAN node or the another network node may be associated with a last serving cell for the UE. The method 600 may also in some examples comprise sending, to the second RAN node or the another network node, a context request message for the inactive AS context for the UE before receiving the inactive AS context for the UE from the second RAN node or the another network node.

[0157] In some examples, determining that the UE transitions to the connected state in step 602 of the method 600 comprises receiving a resume message from the UE or sending a resume message to the UE.

[0158] The method 600 may in some examples comprise, after the UE transitions to the connected state, at least one of the following examples:

[0159] • determining that the functionality is applicable if the stored at least part of the information indicates that the functionality is applicable;

[0160] • determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable;

[0161] • determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable;

[0162] • determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is not applicable.

[0163] In some examples, the method 600 may comprise, after the UE transitions to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE. In some examples, the method 600 may comprise determining that the functionality is not applicable if the first RAN node is not associated with a last serving cell of the UE.

[0164] Further examples of this disclosure are described below.

[0165] According to examples of this disclosure, a UE in a connected state (e.g.

[0166] RRC_CONNECTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration), from a last serving network node, and in response to the message the UE transitions from the connected state (e.g.

[0167] RRC_CONNECTED) to an Inactive state (e.g. RRCJNACTIVE), and in response to the message and to transitioning to the Inactive state the UE stores or deletes (at least partially) one or more information associated to an AI / ML functionality the UE was operating in of the first cell e.g. stored within the UE Inactive Access Stratum (AS) context. Examples of the one or more information associated to the AI / ML functionality are:

[0168] • An AI / ML functionality configuration e.g. inference configuration;

[0169] • An applicability indication for an AI / ML functionality; A ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0170] According to examples of this disclosure, the UE in Inactive state, which had stored the one or more information associated to the AI / ML functionality initiates a resume procedure in a second cell (e.g. different from the first cell) and restored at least one of the one or more information associated to the AI / ML functionality.

[0171] In the context of this disclosure, the first cell may be called the last serving cell i.e. that may be the Primary Cell (PCell) the UE was connected to when the UE receives the message for transitioning to Inactive state. That last serving cell is associated to a radio access network (RAN) node, which may be called a last serving RAN node e.g. last serving gNodeB (gNB), or last serving 6G RAN node. Or, in more general terms, what is called ‘the first cell’ is the ‘last network entity visible to the UE’ (e.g. a cell in the case of NR) which the UE is connected to and from where the UE receives the message for transitioning to Inactive state.

[0172] In the context of this disclosure, the message in response to which the UE transitions to Inactive state may (e.g. RRCJNACTIVE) correspond to an RRC message, such as an RRC Release including a suspend configuration (e.g. IE SuspendConfig).

[0173] In the context of this disclosure, the term “AI / ML functionality” may be called a “supported functionality” the UE can indicate by using UE capability signaling. A supported functionality is one or more functionalities for and / or associated to beam management and / or CSI reporting, or mobility operations, such as the reporting of time domain and / or spatial domain or frequency domain predictions (inference). It could be said as the ability the UE has to produce an output of an inference function. For example, reporting of time-domain prediction(s) of SSB and / or CSI-RS measurement information (e.g. predicted RSRP) may be considered as an AI / ML functionality which is a “supported functionality” by the UE when the UE reports a capability associated to it (via RRC or LPP signaling).

[0174] For example, “spatial domain prediction for beam management or mobility procedure e.g., handover or reconfiguration with sync, or Primary cell (PCell) change, or Primary Secondary Cell Group cell (PSCell) change” or a related functionality (e.g. reporting and inference of spatial domain info) may be a supported functionality in which the UE may report that is capable of performing and reporting inference / prediction of a set A of beams or cells (e.g. predicted L1 RSRP values of one or more beams or one or more SSB indexes of a cell or predicted L1 or L3 RSRP values of one or more cells) based on measurements performed on a set B of beams (e.g. measured L1 / L3 RSRP values of one or more beams or one or more SSB indexes of a cell), in the case of spatial domain predictions.

[0175] For example, “frequency domain prediction for beam management or mobility procedure e.g., handover” or a related functionality (e.g. reporting and inference of frequency domain info) may be a supported functionality in which the UE may indicate that is capable of performing and reporting inference (e.g., prediction of the radio link quality of a set A of beams or cells (e.g. predicted L1 RSRP values of one or more beams or one or more SSB indexes of a cell or predicted L1 or L3 RSRP values of one or more cells) based on measurements performed on a set B of beams (e.g. measured L1 / L3 RSRP values of one or more beams or one or more SSB indexes of a cell or one or more cells), in the case of frequency domain predictions. For example, “time domain prediction for beam management or a mobility procedure e.g., handover or reconfiguration with sync, or Primary cell (PCell) change, or Primary Secondary Cell Group cell (PSCell) change” or a related functionality (e.g. reporting and inference of time domain info) may be a supported functionality in which the UE may report that is capable of performing and reporting inference of a set of A of beams (e.g. predicted L1 / L3 RSRP values of one or more beams or one or more SSB indexes of a cell in future time instances or the L1 / L3 RSRP value of one or more cells in the future time instances) based on measurements performed on a set B of beams or cells (e.g. measured L1 / L3 RSRP values of one or more beams or one or more SSB indexes of a cell and / or L1 / L3 RSRP value of one or more cells), in the case of time domain predictions.

[0176] For example, beam management - DL Tx beam prediction for both UE-sided model and NW-sided model, including: o Spatial-domain DL T ransmitted (Tx) beam prediction for Set A of beams based on measurement results of Set B of beams (“BM-Case1”) o Temporal DL Tx beam prediction for Set A of beams based on the historic measurement results of Set B of beams (“BM-Case2”)

[0177] For example, positioning accuracy enhancements, including: o Direct AI / ML positioning, such as:

[0178] ■ UE-based positioning with UE-side model, direct AI / ML positioning

[0179] ■ UE-assisted / LMF-based positioning with LMF-side model, direct AI / ML positioning

[0180] ■ NG-RAN node assisted positioning with LMF-side model, direct AI / ML positioning o AI / ML assisted positioning, such as: ■ UE-assisted / LMF-based positioning with UE-side model, AI / ML assisted positioning

[0181] ■ NG-RAN node assisted positioning with gNB-side model, AI / ML assisted positioning

[0182] For example, CSI compression e.g., considering extending the spatial / frequency compression to spatial / temporal / frequency compression, cell / site specific models, CSI compression plus prediction (compared to Rel-18 non-AI / ML based approach) For example, “Radio Link Failure prediction of serving and / or neighbour cells” or a related functionality (e.g. reporting and inference of RLF prediction of serving and / or neighbour cell(s)) may be a supported functionality in which the UE may report that is capable of performing and reporting inference of an RLF in future time instances.

[0183] For example, “Handover Failure (HOF) prediction of a cell” or a related functionality (e.g. reporting and inference of HOF prediction of a cell) may be a supported functionality in which the UE may report that is capable of performing and reporting inference of an HOF in future time instances.

[0184] In the context of this disclosure the one or more information associated to an AI / ML functionality (of the first cell the UE transitions to Inactive state or of a cell the UE may later resume) may correspond to one or more of the following:

[0185] - a) An AI / ML functionality configuration (e.g. inference configuration); o the UE has received before the message transitioning the UE to Inactive state (e.g. RRC Release with suspend configuration).

[0186] - b) An applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’

[0187] - c) A ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. a) AI / ML functionality configuration

[0188] In the context of this disclosure, an AI / ML functionality configuration (received in the first cell e.g. last serving cell) may in one option include one or more parameters, I E(s), fields and / or configuration(s) necessary and / or sufficient for the UE to operate the AI / ML functionality e.g. in the first cell, or in another cell in which the UE may later resume. That may include an inference configuration or an inference related configuration (which may also be considered a full and / or complete inference configuration, sufficient for the operation of the AI / ML functionality e.g. while the UE is at least in the first cell). In other words, when the UE receives in connected sate the inference configuration or an inference related configuration for an AI / ML functionality, while it is in the first cell and before it transitions to Inactive state, the UE can generate inference information (e.g. as output of an AI / ML model associated with the AI / ML functionality) and possibly report to the network, while connected to the first cell.

[0189] In the context of this disclosure, an inference configuration or an inference related configuration may correspond to a Channel State information (CSI) measurement configuration (e.g. in an IE CSI-MeasConfig, CSI-ReportConfig, CSI-ResourceConfig) associated to a set A and or set B of beams for a beam management AI / ML functionality. The inference configuration may further include one or more of:

[0190] - Synchronization Signal Block (SSB) identifiers associated to a serving cell and / or a neighbour cell;

[0191] - CSI-RS resource identifiers associated to a serving cell and / or a neighbour cell; Beam identifiers associated to a serving cell and / or a neighbour cell;

[0192] Mobility Reference Signal(s) identifiers associated to a serving cell and / or a neighbour cell;

[0193] - Candidate inference configuration(s) Set A and / or B (1); Set A and / or B (2); Set A and / or B (3), etc.

[0194] In the context of this disclosure, an inference configuration or an inference related configuration may correspond to a mobility prediction configuration or Radio resource management (RRM) measurement configuration associated to a mobility procedure to run the inference and report the predictions. That may include one or more of:

[0195] Measurement configurations (e.g., measConfig) including one or more of o a list of one or more measurement objects to add / modify / remove by the UE which may further includes

[0196] ■ SSB frequency

[0197] ■ CSI-RS frequency

[0198] ■ SSB subcarrier spacing

[0199] ■ measurement timing configuration

[0200] ■ SSB and or CSI-RS beam consolidation configuration / threshold

[0201] ■ Number of SSB or CSI-RS measurements to average o a list of one or more report configuration to add / modify / remove by the UE o SpCell RSRP measurement controlling when the UE is required to perform measurements on non-serving cells o Measurement gap configuration e.g., a measurement gap ID, identifiying the measurement gap ID per FR. o Measurement quantity configuration In the context of this disclosure, an inference configuration or an inference related configuration may include a first set (set A) of measurement resources (e.g. beams, SSB indexes and / or CSI-RS resource identifiers, Mobility Refence Signal identifiers) in which the UE performs radio measurement predictions (inferences, such as predicted RSRP values), and a second set (set B) of radio measurement resources (e.g. beams, SSB indexes and / or CSI-RS resource identifiers, Mobility Refence Signal identifiers) in which the UE can perform radio measurement in order to determine the radio measurement predictions on the first set. That may also include one or more configuration(s) associated to network side (NW-side) additional conditions reflecting the NW operational properties, such as:

[0202] - Set A and / or Set B of resources, represented e.g. by an ID associated to the set of resources or to the resources within the set.

[0203] Mapping relationship of Set A and Set B, including ordering to (a set of ID, or resource)

[0204] - Consistency of downlink spatial domain transmission filters corresponding to the beams in Set A and Set B.

[0205] - QCL assumption

[0206] - The order of model input and model output between RS and Tx beams can be predefined.

[0207] - gNB transmission power UE distribution

[0208] - gNB antenna height and / or other antenna properties Network deployment scenarios (e.g., ISD, Umi / Uma)

[0209] NW-side resource configuration(s) which may be considered as NW implementationbased configurations which may possibly impact the inference performance for a UE sided model. For instance, beam and Tx port mapping relationship in the gNodeB for a given cell, NW antenna shape, Antenna dip angle, height of the tower / gNB, etc. NW load in terms of connected users, radio resource utilizations (PDCCH / RACH / PUCCH / PUSCH / PDSCH resource load, number of configured bearers, etc)

[0210] The inference configuration or an inference related configuration which the UE has received in the first cell may include a list of IDs referring to the set A and set B (or to the resources within the set A / B), and referring to one or more NW-side additional conditions.

[0211] In the context of this disclosure, the AI / ML functionality configuration may include one or more of the following: - An inference configuration for a beam management functionality (e.g. time-domain prediction of beam information) o In one example, the inference configuration includes the reporting configuration including parameters indicating how the UE is to report timedomain predictions of beam information (e.g. beam indexes and / or SSB indexes and / or time-domain prediction of beam measurements) and / or spatial-domain predictions of beam information. o In one example, the inference configuration includes the resource configuration for resources (e.g. SSB indexes and / or CSI-RS resources) which the UE measures and provides as input to an AI / ML model (or inference function) to produce inference outputs e.g. the actual time-domain predictions of beam information (e.g. beam indexes and / or SSB indexes and / or time-domain prediction of beam measurements) and / or spatial-domain predictions of beam information to be included in a report.

[0212] - An inference configuration for a L3 Mobility functionality o In one example, the inference configuration includes the reporting configuration including parameters indicating how the UE is to report timedomain predictions for neighbour cell(s) which are candidates for a connected mode inter-cell mobility procedure, or for serving cells. The time-domain predictions may be cell identifiers, time-domain predictions of measurements, such as predicted RSRP, predicted RSRQ, as predicted SI NR and / or spatial- domain predictions of cell(s). o In one example, the inference configuration includes the configuration for the UE to predict the occurrence of a Radio Link failure (RLF) in the second cell. o In one example, the inference configuration includes the configuration for the UE to predict the future occurrence of a Handover failure (HOF) when the UE is in second cell and later would move to yet another cell. o In one example, the inference configuration includes the configuration for the UE to predict the future occurrence of the fulfillment of a measurement reporting event such as an event A1, A2, A3, A4, A5, A6, B1, B2, etc.

[0213] - An inference configuration for positioning functionality

[0214] - An inference configuration for CSI reporting functionality

[0215] - A configurating enabling the UE to determine whether the AI / ML functionality is applicable or not. o In one option the UE has reported whether the AI / ML functionality of the second cell (which is the target cell) was applicable or not. However, as that may have changed from the time the UE ahs transmitted the report (until the time in which the UE is to access the second cell in the handover), the UE...

[0216] Network conditions such as Set A / set B configuration(s).

[0217] - A state indication for the AI / ML functionality, e.g., ‘activated’, ‘inactivated’, ‘deactivated’.

[0218] - An indication on whether the UE is allowed to consider the AI / ML functionality as ‘activated’ when the functionality is determined by the UE to be applicable.

[0219] The AI / ML functionality configuration may include also an identifier of the AIML functionality to which the configuration (e.g. inference configuration) is referred to, wherein the AIML functionality could be for example, beam management functionality, spatial beam management functionality, temporal beam management functionality, L3 mobility functionality, positioning functionality, CSI compression functionality, CSI prediction functionality, etc.

[0220] In the context of this disclosure an AI / ML functionality configuration, which the UE in connected state receives while it is connected to the first cell may in another option include one or more parameters, I E(s), fields and / or configuration(s) necessary and / or sufficient for the UE to report the applicability of the AI / ML functionality (e.g. in the first cell, before the UE transitions to Inactive state or in another cell in which the UE may later resume), such as an applicability reporting configuration. In other words, when the UE receives the applicability reporting configuration for an AI / ML functionality the UE can determine whether the AI / ML functionality, supported by the UE, and / or associated configuration(s) of that AI / ML functionality, is applicable or not applicable. The applicability reporting configuration may include one or more of the following:

[0221] - An indication that the UE is allowed to do UE assistance information reporting e.g. by configuring it in the IE OtherConfig in the RRCReconfiguration message.

[0222] - An indication of the AI / ML functionality for which the UE should transmit the applicability reporting e.g. indications of the applicability associated the indicated AIML functionality.

[0223] One or more NW-side additional condition(s) (included e.g. in the IE OtherConfig in the RRCReconfiguration message, e.g. for the UE to determine whether the AI / ML model I functionality has been trained under similar conditions, such as one or more of the following: o Configuration(s) related to the Mapping relationship of Set A and Set B, including ordering to (a set of IDs, or resources) Configuration(s) related to the consistency of downlink spatial domain transmission filters corresponding to the beams in Set A and Set B. In that context, consistency may correspond to one or more of:

[0224] ■ Set size consistency for Set B, Set A: consistency in number of beams and / or associated resources for Set B and Set A, across training and inference

[0225] ■ periodicity consistency for Set B, Set A: consistency in periodicity of beams and / or associated resources for Set B and Set A, across training and inference

[0226] ■ relationship of Set A / Set B (Set B is a subset of Set A or not): consistency in relationship of beams and / or associated resources for Set B and Set A, i.e, whether Set B is a subset of Set A, across training and inference Configuration(s) related to the Quasi-Co-Location (QCL) assumption(s) Beam configuration(s) of the network such as:

[0227] ■ Beam characteristics, e.g., beam boresight direction (azimuth and elevation), 3dB beamwidth. In one sub-option the beam characteristics may be associated to an identifier indicated to the UE during training and AI / ML configuration, for checking of the consistency between training and inference.

[0228] ■ Set A / Set B related info, e.g., the beam index of set B.

[0229] ■ Information about the beam codebook and / or indexing / mapping of Set A and Set B i.e. info on whether the AI / ML Model I functionality is trained with a data set with a certain beam codebook and index / mapping of Set A / Set B, inference works for the same beam codebook and index / mapping of Set A / Set B. Configuration(s) related to the order of model input and model output between RS and Tx beams can be pre-defined. Configuration(s) related to the transmission power and / or power levels the gNodeB and / or the serving cells are operating Configuration(s) related to the UE distribution Configuration(s) related to Antenna height Configuration(s) related to the deployment scenarios (e.g., ISD, Umi / Uma / rural / indoor / indoor office / indoor factory, specific area(s)) Configuration(s) related to UE speed - An indication of an identifier (associated ID) associated to one or more network conditions, so that the UE assumes that NW-side additional conditions with the same associated ID are consistent at least within a cell.

[0230] In one option, NW-side additional condition may be associated to an inference configuration (e.g. resource set A, to be inferred and / or estimated and / or predicted, and / or resource set B, in which the UE should perform the measurement to infer / estimate / predict the radio measurement assocatiated to the set A resources) and / or one training configuration (e.g. resource of CSI resources configured by the gNB at the time of the UE performing UE-side model training) identified by the same associated ID, for the second cell (which is a neighbour cell which may become the target cell in a handover). The UE may performing training of one or AI / ML functionalities I models with different sets of collected data via training configuration identified by its associated ID (i.e. one associated ID->one training configuration- >one Al model).

[0231] - An inference configuration. b) Applicability indication for an AI / ML functionality

[0232] In the context of this disclosure, the applicability indication of an AI / ML functionality associated to a cell (e.g. the first cell, or another cell the UE may later resume to cell fulfilling an event) may comprises one or more of the following: o An indication indicating that the AI / ML functionality is ‘applicable’

[0233] ■ In one option this indication corresponds to a field and / or IE and / or parameter; o An indication indicating that the AI / ML functionality is ‘not applicable’ (or ‘non- applicable’)

[0234] ■ In one option this indication corresponds to a field and / or IE and / or parameter;

[0235] ■ In one option this indication corresponds to the absence of a field and / or the absence of an IE and / or the absence of a parameter; o An indication of an applicability status, which may take one or more values, such as ‘applicable’ or ‘not applicable’; o A recommended (or preferred) AI / ML functionality configuration for which the AI / ML functionality becomes applicable i.e. the indication indicates that the AI / ML functionality may not be applicable for a configuration(x) but it may be applicable for a configuration(y), wherein the applicability indication corresponds to an indication of configuration(y) o An identifier associated to a one or more NW-side additional conditions associated to the AI / ML model e.g. the identifier of the NW configuration(s) (settings) in which the AI / ML model was trained.

[0236] In the context of this disclosure, the UE has determined the applicability indication for the AI / ML functionality of the second cell based on one or more UE-conditions and / or one or more network conditions and / or based on an inference configuration. Hence, prior to transmit the applicability indication, the UE evaluates the applicability of the AIML model / functionality and takes into account the NW-side additional conditions, and UE-side additional conditions, wherein the latter could be represented by:

[0237] • UE speed

[0238] • UE battery status

[0239] • UE antenna properties (layout, Ml MO configuration, device orientation etc.)

[0240] • UE radio configuration

[0241] • UE traffic type c) ‘State’ of the AI / ML functionality

[0242] In the context of this disclosure the state’ of the AI / ML functionality is considered to be ‘activated’ (or active) when the UE has an AI / ML configuration (e.g. ) for an AI / ML functionality which is ‘applicable’ and the UE is configured by the network and / or indicated by the network while the UE is the first cell to produce inference outputs (e.g. predictions) which may be reported and / or based on which the UE needs to take further decisions.

[0243] In the context of this disclosure the state’ of the AI / ML functionality is considered to be ‘deactivated’ (or no activated) when the UE has an AI / ML configuration (e.g. ) for an AI / ML functionality which is ‘non applicable’ or the UE is configured by the network and / or indicated by the network while the UE is the first cell to NOT produce inference outputs (e.g. predictions).

[0244] In the context of this disclosure the UE determines the state’ of the AI / ML functionality based on the applicability of the AI / ML functionality (e.g. state set to ‘activated’ when the AI / ML functionality is applicable; state set to ‘deactivated’ when the AI / ML functionality is not applicable), and / or based on a network command (e.g. a MAC Control Element) received before the message indicating the transition to Inactive state , and / or based on an RRC parameter. When the UE transitions to Inactive state and the UE stores the state, the UE stores the latest state of the AI / ML functionality, determined before the UE transitions to Inactive state.

[0245] In the following we show embodiments for a suspend procedure and for a resume procedure. However, combinations of suspend and resume procedures are also considered e.g. UE transitions to Inactive state and stores information of an AI / ML functionality and, when it initiates a resume procedure it restores the information of the AI / ML functionality, before it receives the resume message. Non-limiting examples of these suspend / resume combinations of sections 2.7.1.2 and 2.7.1.3 are shown in some figures in section 2.7.1.4.

[0246] Suspend procedure

[0247] In one embodiment, a UE in a connected state (e.g. RRC_CON NESTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration), from a last serving network node, and in response to the message the UE transitions from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRCJNACTIVE), and in response to the message and to transitioning to the Inactive state the UE stores one or more information associated to an AI / ML functionality of the first cell e.g. stored within the UE Inactive Access Stratum (AS) context.

[0248] In one example, the UE stores an AI / ML functionality configuration e.g. inference configuration the UE has received before the message transitioning the UE to Inactive state (e.g. RRC Release with suspend configuration). For example, the UE may have received the AI / ML functionality configuration e.g. inference configuration when it was in connected state, in an RRC Reconfiguration message. This means that the UE keeps / maintains the AI / ML functionality configuration received while in connected mode prior to receive the RRC release message.

[0249] In one example, the UE stores an applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’. In other words, the UE remembers whether the AI / ML functionality was ‘applicable’ or ‘not applicable’ when the UE transitioned to the Inactive state. The applicability of an AI / ML functionality may have been determined by the UE when the UE was in connected state in the first cell.

[0250] In one example, the UE stores a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. In other words, the UE remembers what state the AI / ML functionality had when the UE was in connected state and transitioned to Inactive state (e.g. RRCJNACTIVE). The last serving network node, associated with the first cell (called last serving cell), which transmits the message indicating the UE to transition from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRCJNACTIVE), also stores one or more information associated to an AI / ML functionality of the first cell e.g. stored within the UE Inactive Access Stratum (AS) context. In one option the last serving network node stores the one or more information associated to an AI / ML when it determines that the UE has transitions to the Inactive state e.g. upon reception of an acknowledgement message in response to the message transitioning the UE to Inactive state.

[0251] In one example, the last serving network node stores an AI / ML functionality configuration e.g. inference configuration which it has transmitted to the UE before the message transitioning the UE to Inactive state (e.g. RRC Release with suspend configuration).

[0252] In one example, the last serving network node stores an applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’.

[0253] In one example, the last serving network node stores a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0254] In one embodiment, a UE in a connected state (e.g. RRC_CONNECTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration) and in response to the message the UE transitions from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRC NACTIVE), and in response to the message the UE deletes one or more information associated to an AI / ML functionality of the first cell. In one option of that embodiment, the UE partially deletes the one or more information associated to an AI / ML functionality of the first cell.

[0255] In one example, the UE deletes (or releases) an AI / ML functionality configuration e.g. inference configuration the UE has received before the message transitioning the UE to Inactive state (e.g. RRC Release with suspend configuration). For example, the UE may have received the AI / ML functionality configuration e.g. inference configuration when it was in connected state, in an RRC Reconfiguration message.

[0256] In one example, the UE deletes (or releases) the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. In other words, the UE forgets what state the AI / ML functionality had when the UE was in connected state. o In one sub-option, the UE stores the AI / ML functionality configuration e.g. inference configuration the UE has received before the message (e.g. RRC Release with suspend configuration) but deletes (or releases) the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. o In one sub-option, the UE stores the AI / ML functionality configuration e.g. inference configuration the UE has received before the message (e.g. RRC Release with suspend configuration) and the applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’, but deletes (or releases) the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0257] In one example, the UE deletes (or releases) an applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’. In other words, the UE forgets whether the AI / ML functionality was applicable or not applicable. o In one sub-option, the UE stores the AI / ML functionality configuration e.g. inference configuration the UE has received before the message (e.g. RRC Release with suspend configuration) but deletes (or releases) the applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’.

[0258] The last serving network node, associated with the first cell (called last serving cell), which transmits the message (e.g. RRC Release including a suspend configuration) indicating the UE to transition from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRC NACTIVE), deletes one or more information associated to an AI / ML functionality of the first cell. In one option of that embodiment, the last serving network node partially deletes the one or more information associated to an AI / ML functionality of the first cell.

[0259] In one example, the last serving network node deletes (or releases) an AI / ML functionality configuration e.g. inference configuration which it has transmitted to UE before the message transitioning the UE to Inactive state (e.g. RRC Release with suspend configuration).

[0260] In one example, the last serving network node deletes (or releases) the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. In other words, the last serving network node forgets what state the AI / ML functionality had when the UE was in connected state. o In one sub-option, the last serving network node stores the AI / ML functionality configuration e.g. inference configuration but deletes (or releases) the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. o In one sub-option, the last serving network node stores the AI / ML functionality configuration e.g. inference configuration and the applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’, but deletes (or releases) the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0261] In one example, the last serving network node deletes (or releases) an applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’. In other words, the last serving network node forgets whether the AI / ML functionality was applicable or not applicable. o In one sub-option, the last serving network node stores the AI / ML functionality configuration e.g. inference configuration but deletes (or releases) the applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’.

[0262] In one embodiment, a UE in a connected state (e.g. RRC_CONNECTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration) and in response to the message the UE transitions from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRCJNACTIVE), and in response to a parameter within the message the UE determines to delete or to store one or more information associated to an AI / ML functionality of the first cell. In one option, in the case the one or more information associated to an AI / ML functionality of the first cell is stored within the UE Inactive Access Stratum AS context.

[0263] In one option, the UE stores the AI / ML functionality configuration e.g. inference configuration the UE has received before the message (e.g. RRC Release with suspend configuration) and stores or deletes the applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’, depending on the parameter within the message for transitioning the UE to Inactive state. o For example, the parameter may indicate ‘store AI / ML applicability indication’ based on which the UE stores the applicability indication for an AI / ML functionality of the first cell; or, the parameter may indicate ‘AI / ML applicability indication' based on which the UE stores the applicability indication for an AI / ML functionality of the first cell.

[0264] The last serving network node which transmits the message indicating the UE to transition from the connected state to an Inactive state includes in the message the parameter indicating the UE to determine to delete or to store one or more information associated to an AI / ML functionality of the first cell. And, depending on the parameter included the the last serving network node determines itself to delete or to store one or more information associated to an AI / ML functionality of the first cell. In one option, the last serving network node stores the AI / ML functionality configuration e.g. inference configuration and stores or deletes the applicability indication for an AI / ML functionality of the first cell e.g. ‘applicable’ or ‘not applicable’, depending on the parameter within the message for transitioning the UE to Inactive state. o For example, the parameter may indicate ‘store AI / ML applicability indication’ so that the last serving network node stores the applicability indication for an AI / ML functionality of the first cell; or, the parameter may indicate ‘AI / ML applicability indication' so that the last serving network node stores the applicability indication for an AI / ML functionality of the first cell.

[0265] In one embodiment, a UE in a connected state (e.g. RRC_CONNECTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration) and in response to the message the UE transitions from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRCJNACTIVE), and in response to the message the UE stores one or more information associated to an AI / ML functionality of the first cell e.g. within the UE Inactive Access Stratum (AS) context. Then, upon entering the Inactive state (e.g. RRCJNACTIVE) the UE performs cell selection to the first cell and camps in the first cell.

[0266] In one option, while the UE is camping in the Inactive state in the first cell the UE keeps the one or more information associated to the AI / ML functionality stored. o In one sub-option, when the UE moves to a second cell (e.g. cell re-selection from the first cell to the second cell) the UE keeps stored the one or more information associated to an AI / ML functionality of the first cell.

[0267] ■ In one example for this sub-option, the UE keeps stored the AI / ML functionality configuration e.g. inference configuration; this does not preclude the deletion of other information associated to an AI / ML functionality, such as the ‘state’ or the ‘applicability indication’.

[0268] ■ In one method, the UE keeps the one or more information associated to the Al ML functionality as long as the Al ML functionality is still applicable in the first cell. For example, if the UE-side additional conditions change but the AIML functionality is still trained to operate in the first cell under these new conditions, the UE keeps the associated information. In one option, while the UE is camping in the Inactive state in the first cell the UE releases(deletes) the one or more information associated to the AI / ML functionality stored. o This can happen if the UE-side additional conditions change and the AIML functionality is not trained to operate in the first cell under these new conditions

[0269] In one sub-option, when the UE moves to a second cell (e.g. cell re-selection from the first cell to the second cell) the UE releases (deletes) at least one of the one or more information associated to an AI / ML functionality of the first cell.

[0270] ■ In one example for this sub-option, the UE deletes the applicability indication for an AI / ML functionality of the first cell;

[0271] ■ In one example for this sub-option, the UE deletes the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0272] ■ This does not preclude the storing of other information associated to an AI / ML functionality. For example, the UE may delete the applicability indication for an AI / ML functionality of the first cell and / or the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’, but keep stored the AI / ML functionality configuration e.g. inference configuration.

[0273] ■ In one method, the UE releases the one or more information associated to the AIML functionality upon determining that the AIML functionality is not applicable in the second cell. For example, if the AIML functionality is not trained to operate in the second cell, the UE deletes the associated information.

[0274] In one sub-option, when the UE moves to a second cell (e.g. cell re-selection from the first cell to the second cell) the UE keeps at least one of the one or more information associated to an AI / ML functionality of the first cell.

[0275] ■ In one method, the UE keeps the one or more information associated to the AIML functionality as long as the AIML functionality is still applicable in the second cell. For example, if the the AIML functionality is still trained to operate in the second cell, the UE keeps the associated information of the first cell.

[0276] In one embodiment, a UE in a connected state (e.g. RRC_CONNECTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration) and in response to the message the UE transitions from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRC NACTIVE), and in response to the message the UE stores one or more information associated to an AI / ML functionality of the first cell e.g. within the UE Inactive Access Stratum (AS) context. Then, upon entering the Inactive state (e.g. RRCJNACTIVE) the UE performs cell selection to a second cell and camps in the second cell, different from the first cell, in response to which the UE deletes at least parts of the one or more information associated to an AI / ML functionality of the first cell.

[0277] In one example, in response to selecting the second cell the UE deletes the applicability indication for an AI / ML functionality of the first cell;

[0278] In one example in response to selecting the second cell the UE deletes the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0279] - This does not preclude the storing of other information associated to an AI / ML functionality. For example, the UE may delete the applicability indication for an AI / ML functionality of the first cell and / or the ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’, but keep stored the AI / ML functionality configuration e.g. inference configuration.

[0280] In one embodiment, a UE in a connected state (e.g. RRC_CONNECTED) in a first cell (called last serving cell) receives a message (e.g. RRC Release including a suspend configuration), from a last serving network node, and in response to the message the UE transitions from the connected state (e.g. RRC_CONNECTED) to an Inactive state (e.g. RRCJNACTIVE), and in response to the message and to transitioning to the Inactive state the UE stores one or more information associated to an AI / ML functionality of the first cell e.g. stored within the UE Inactive Access Stratum (AS) context and starts a timer. While the timer is running the UE keeps stored the one or more information associated to an AI / ML functionality of the first cell; when the timer expires the UE deletes the one or more information associated to an AI / ML functionality of the first cell.

[0281] In one option, when a resume procedure is initiated the timer is stopped (when that is running).

[0282] In one option, when a resume procedure is initiated and the UE receives a resume message, the timer is stopped.

[0283] In one option, a timer value for the timer is received by the UE in the message from a last serving network node in response to which the UE transitions from the connected state to the Inactive state.

[0284] Resume procedure In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE restores one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context. In one example, the UE restores an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. Then, upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, the UE receives a resume message (e.g. RRC Resume message) from the second cell (i.e. current network node).

[0285] In one method, prior to transmit the resume request the UE evaluates the applicability of the AIML model / functionality considering the stored information associated to the AIML functionality. For example, the UE evaluates the applicability of the AIML model / functionality considering the AIML functionality configuration stored and received before entering RRCJNACTIVE mode, and the current UE-side additional conditions. This is because the UE-side additional conditions at the time of transmitting the RRC resume request may have changed compared with the UE-side additional conditions at the time of the UE receiving the RRC release message. The UE may then indicate in the RRC resume request an indication indicating whether the AIML model / functionality is applicable or not applicable. The RRC resume request may further indicate the one or more NW-side additional conditions (e.g. indicated via an associated ID) for which the AIML model / functionality was applicable prior to receive the RRC release message.

[0286] In one option, the resume message (e.g. RRC Resume) from the current network node does not contain an AI / ML functionality configuration e.g. inference configuration, in response to which the UE considers the restored AI / ML functionality configuration (e.g. inference configuration) the AI / ML functionality configuration (e.g. the inference configuration) to be utilized by the UE in the second cell the UE is resuming. o In one example, the network may not transmit any AIML functionality configuration in response of receiving a resume request indicating that the AIML model / functionality operated prior to receive the RRC release message is applicable. o In one example, the network may not transmit any AIML functionality configuration in response of receiving a resume request indicating the NW- side additional conditions for which the AIML model / functionality was applicable prior to receive the RRC release message, and determining that such NW-side additional conditions are the same as the current NW-side additional conditions at the point in time of transmitting the RRC Resume message to the UE.

[0287] In one option, the resume message (e.g. RRC Resume) contains an AI / ML functionality configuration, in response to which the UE applies the received AI / ML functionality configuration on top of the restored AI / ML functionality configuration. o In one sub-option, after the UE determines the AI / ML functionality to be used in the second cell, the UE determines an applicability indication for an AI / ML functionality e.g. ‘applicable’, or ‘not applicable’, based on the resulting configuration after the UE applied the received AI / ML functionality configuration on top of the restored AI / ML functionality configuration.

[0288] ■ In one sub-option, when the UE determines that the resulting configuration to be applicable, the UE activates the resulting AI / ML configuration.

[0289] ■ In one sub-option, when the UE determines that the resulting configuration to be not applicable, the UE considers the resulting AI / ML configuration to be deactivated.

[0290] ■ In one sub-option, when the UE determines the applicability indication for an AI / ML functionality e.g. ‘applicable’, or ‘not applicable’, for the resulting configuration, and the UE transmits to the current network node (of the second cell the UE is resuming) the applicability indication for an AI / ML functionality e.g. in a resume complete message (e.g. RRC Resume Complete message) or in a UE Assistance Information message.

[0291] • In one sub-option, when the UE determines the applicability indication for an AI / ML functionality e.g. ‘applicable’, or ‘not applicable’, for the resulting configuration, based one or more parameters in the resume message e.g. an applicability reporting configuration in the resume message. o In one example, the network transmits an AIML functionality configuration in response of receiving a resume request indicating that the AIML model / functionality operated prior to receive the RRC release message is not applicable. In such case, the network node may transmit a different AIML functionality configuration compared with the one the UE was operating at the time of receiving the RRC release message. o In one example, the network transmit an AIML functionality configuration in response of receiving a resume request indicating the NW-side additional conditions for which the AIML model / functionality was applicable prior to receive the RRC release message, and determining that such NW-side additional conditions are not the same as the current NW-side additional conditions at the point in time of transmitting the RRC Resume message to the UE. The network node may then transmit a different inference configuration including the current NW-side additional conditions.

[0292] In one option, the UE considers the AI / ML functionality to be ‘applicable’ in the second cell when the restored applicability indication for an AI / ML functionality indicates that the AI / ML functionality is ‘applicable’. o In one option, one additional condition the UE checks is whether the second cell is the same as the first cell i.e. the last serving.

[0293] ■ When the second cell (i.e. the cell the UE is resuming) is the same as the first cell (i.e. the last serving cell), the UE considers the AI / ML functionality to be ‘applicable’ in the second cell when the restored applicability indication for an AI / ML functionality indicates that the AI / ML functionality is ‘applicable’.

[0294] ■ When the second cell (i.e. the cell the UE is resuming) is NOT the same as the first cell (i.e. the last serving cell), the UE considers the AI / ML functionality to be ‘not applicable’ in the second cell regardless of what the restored applicability indication for an AI / ML functionality indicates.

[0295] In one option, the UE considers the AI / ML functionality to be ‘not applicable’ in the second cell when the restored applicability indication for an AI / ML functionality indicates that the AI / ML functionality is ‘ not applicable’.

[0296] In one option, the UE considers the AI / ML functionality in the second cell to be the restored applicability indication for an AI / ML functionality (e.g. ‘applicable’ or ‘not applicable’) based on a presence or absence of a parameter in the resume message. In one example, the UE considers the AI / ML functionality to be ‘activated’ state when the UE restored ‘state’ of the AI / ML functionality is ‘activated’.

[0297] In one example, the UE considers the AI / ML functionality to be ‘deactivated’ state when the UE restored ‘state’ of the AI / ML functionality is ‘deactivated’.

[0298] In one example, the UE considers the AI / ML functionality to be the restored ‘state’ of the AI / ML functionality based on a presence or absence of a parameter in the resume message. o In one option, the presence of a parameter indicating the state of the Al / ML functionality (e.g. ‘activated’) indicates that the UE shall follow what the parameter indicates and that the UE shall ignore the restored parameter. o In one option, the absence of a parameter indicating the state of the Al / ML functionality (e.g. ‘activated’) indicates that the UE shall follow what the restored parameter indicates.

[0299] The current network node associated to the second cell, receives the resume request (e.g. an RRC Resume Request message) from the UE, and in response transmits to the last serving network node (e.g. indicated in a resume identifier included in the resume request) a context request message (e.g. UE Context Request message over XnAP), in response to which it receives the UE AS Inactive context (stored in the ) which includes the previously stored one or more information associated to an AI / ML functionality. Based on that the current network node transmits to the UE a resume message (e.g. RRC Resume), according to one or more options disclosed above e.g. the resume message contain or not an AI / ML functionality configuration.

[0300] In one sub-option, the current network node receives the applicability indication for an AI / ML functionality e.g. in a resume complete message (e.g. RRC Resume Complete message) or in a UE Assistance Information message. That is done when the UE determines the applicability indication for an AI / ML functionality e.g. ‘applicable’, or ‘not applicable’, for the resulting configuration.

[0301] In one sub-option, the current network node transmits one or more parameters in the resume message (e.g. an applicability reporting configuration in the resume message) based on which the UE determines the applicability indication for an AI / ML functionality e.g. ‘applicable’, or ‘not applicable’, for the resulting configuration.

[0302] In one example, the current network node determines to include or not (presence or absence) a parameter in the resume message, based on which the UE considers the AI / ML functionality to be the restored ‘state’ of the AI / ML functionality .

[0303] In one option, the UE restores the one or more information associated to an AI / ML functionality when the second cell indicates in its system information (e.g. a System Information Block message the UE receives) that the UE shall restore the one or more information associated to an AI / ML functionality.

[0304] In one option, the current network node indicates in its system information (e.g. a System Information Block message) that the UE shall restore the one or more information associated to an AI / ML functionality e.g. when the current network node is capable of the AI / ML functionality.

[0305] In one option the indication in the system information to restore the one or more information associated to an AI / ML functionality may be defined per AI / ML functionality.

[0306] In one option, the UE deletes (or releases) the one or more information associated to an AI / ML functionality when the second cell indicates in its system information that the UE shall delete (or release) the one or more information associated to an AI / ML functionality.

[0307] In one option, the current network node indicates in its system information (e.g. a System Information Block message) that the UE shall delete the one or more information associated to an AI / ML functionality e.g. when the current network node is not capable of the AI / ML functionality.

[0308] In one option the indication in the system information to delete the one or more information associated to an AI / ML functionality may be defined per AI / ML functionality.

[0309] In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE restores one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context. In one example, the UE restores an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. Then, upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, the UE receives a message (e.g. RRC Release message including a suspend configuration) from the second cell (i.e. current network node) indicating the UE to remain in the Inactive state. In response to that message the UE stores once more the one or more information associated to an AI / ML functionality.

[0310] In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE restores one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context. In one example, the UE restores an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. Then, upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, the UE receives a message (e.g. RRC Release message NOT including a suspend configuration) from the second cell (i.e. current network node) indicating the UE to transition to an Idle state (e.g. RRCJDLE). In response to that message the UE deletes the one or more information associated to an AI / ML functionality.

[0311] In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE restores one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context. In one example, the UE restores an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. Then, upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, the UE receives a message (e.g. RRC Setup) from the second cell (i.e. current network node) indicating the UE to transmit a setup request i.e. to transition to Idle state and to attempt the transition to connected state. In response to that message the UE deletes the one or more information associated to an AI / ML functionality.

[0312] In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE restores one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context. In one example, the UE restores an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. Then, upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, the UE receives a reject message (e.g. RRC Reject) from the second cell (i.e. current network node) indicating the UE to wait for an amount of time (according to a wait timer value). In response to that message the UE stores once more the one or more information associated to an AI / ML functionality. In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE restores one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context. In one example, the UE restores an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’. Then, upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, the UE starts a timer (e.g. timer T319) and, when the timer expires before the UE receives a response message, the UE declares a failure in the resume procedure, transitions to an Idle state (e.g. RRCJDLE) and indicates the failure to its higher layers (e.g. Non-access stratum layer). In response to timer expiry (e.g. T319 expiry) the UE deletes the one or more information associated to an AI / ML functionality.

[0313] In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE transmits the resume request the UE receives a resume message (e.g. RRC Resume message) from the second cell (i.e. current network node) which includes an AI / ML configuration associated to an AI / ML functionality for the second cell (e.g. inference configuration); in response to the AI / ML configuration associated to an AI / ML functionality for the second cell (e.g. inference configuration), the UE determines an applicability indication of the AI / ML functionality for the second cell (e.g. ‘applicable’ or ‘not applicable’) and transmits the applicability indication of the AI / ML functionality for the second cell to the current network node e.g. in a resume complete message or in a UE Assistance Information message after or multiplexed with the resume complete message.

[0314] In one option, in response to the transmission of the applicability indication of the AI / ML functionality for the second cell (e.g. ‘applicable’) to the current network node, the UE receives a command (e.g. a MAC CE) indicating the UE to activate the AI / ML functionality.

[0315] In one embodiment, a UE in an Inactive state initiates a resume procedure in a second cell and transmits a resume request (e.g. an RRC Resume Request message) to a current network node (associated with the second cell). Upon transmitting the resume request to the second cell, the UE transmits the resume request the UE receives a resume message (e.g. RRC Resume message) from the second cell (i.e. current network node) which includes an AI / ML configuration associated to an AI / ML functionality for the second cell (e.g. inference configuration); in response to the AI / ML configuration associated to an AI / ML functionality for the second cell (e.g. inference configuration), the UE determines an applicability indication of the AI / ML functionality for the second cell (e.g. ‘applicable’ or ‘not applicable’) and, based on that, determines the state of the AI / ML functionality.

[0316] In one option, when the UE determines the AI / ML functionality for the second cell to be ‘applicable’, the UE activates the AI / ML functionality.

[0317] In one option, when the UE determines the AI / ML functionality for the second cell to be ‘not applicable’, the UE deactivates the AI / ML functionality.

[0318] Examples including both suspend and resume related embodiments

[0319] Figure 7 illustrates an example of communications in a network in a suspend / resume procedure in which the UE stores one or more information of an AI / ML functionality when the UE transitions to Inactive state and restores one or more information of an AI / ML functionality when the UE initiates a resume procedure.

[0320] Figure 8 illustrates an example of communications in a network in a suspend / resume procedure in which the UE stores one or more information of an AI / ML functionality when the UE transitions to Inactive state and restores one or more information of an AI / ML functionality when the UE initiates a resume procedure.

[0321] Figures 9 and 10 each illustrate another example of communications in a network according to examples of this disclosure.

[0322] Further example embodiments

[0323] In one embodiment, before the UE stores and / or deletes the applicability indication of an AI / ML functionality associated to a cell (e.g. the first cell, or last serving cell, or another cell the UE is resuming), the UE determines whether the AI / ML functionality associated to the cell cell is ‘applicable’ or ‘not applicable’.

[0324] In one option, the UE is configured to include the applicability indication of an AI / ML functionality associated to a second cell in a measurement report (e.g. an RRC Measurement Report) or UE Assistance Information, and when the UE is configured, the UE determines whether the AI / ML functionality associated to a second cell is ‘applicable’ or ‘not applicable’. o In one example, the UE first waits until a neighbour cell is a triggered cell (i.e. fulfills the entering condition of the event) and only then determines whether the AI / ML functionality associated to a second cell is applicable or not applicable. The advantage is that determining applicability may require UE processing power and / or battery consumption, so here the UE only performs the actions when it knows it needs to report. o In one example, the UE does not wait for a neighbour cell to be a triggered cell and determines whether the AI / ML functionality associated to a second cell is applicable or not applicable when it receives the configuration for reporting. The advantage is that determining applicability may take some time, so such an approach would not delay the transmission of a measurement report because the UE needs to determines the applicability of the AI / ML functionality of the second cell. o In one example, the UE periodically determines whether the AI / ML functionality associated to a second cell is applicable or not applicable after it receives the configuration for reporting. The advantage is that the applicability is always checked, so when it is time to report no further delay is added to it due to the applicability check.

[0325] In one embodiment, before the UE stores and / or deletes the applicability indication of an AI / ML functionality associated to a cell (e.g. the first cell, or last serving cell, or another cell the UE is resuming), the UE receives from the last serving cell (associated to a last serving RAN node) a configuration with one or more parameters controlling how to include the applicability indication of an AI / ML functionality.

[0326] In one option, these one or more parameters are known to the source network node, when the second cell is a cell of the source network node.

[0327] In one option, these one or more parameters are received from the target network node (of the second cell) in an earlier handover (e.g. in a first HO request ack message), so that these one or more parameters are stored in the source network node and known to the source network node when a handover needs to be triggered. In one option, these one or more parameters are received from the target network node of the second cell to the source network node during the setup of an XnAp interface e.g. together with neighbour relations. These one or more parameters may be considered as a property of a neighbour cell in a neighbour relation stored in the source network node. In one option, these one or more parameters are received from the target network node in response to a first request from the source network node, sort of a pre-HO Request message. These one or more parameters could be limited to be needed only for the UE to report the applicability indication of an AI / ML functionality associated to a second cell, but the overall inference configuration to operate in the second cell (i.e. that would be required to be included in the HO command).

[0328] Below are provided example enumerated embodiments according to examples of this disclosure.

[0329] UE embodiments starting from a Suspend procedure

[0330] A1. A method at a UE in a connected state in a first cell comprising:

[0331] Receiving a message from a last serving network node and in response transitioning to an Inactive state, and

[0332] In response to the message and to transitioning to the Inactive state, storing one or more information associated to an AI / ML functionality.

[0333] A1*. A method at a UE in a connected state in a first cell comprising:

[0334] Receiving a message from a last serving network node and in response transitioning to an Inactive state and

[0335] In response to the message and to transitioning to the Inactive state, deleting one or more information associated to an AI / ML functionality.

[0336] A2. A method of A1 and A1*, wherein the one or more information associated to an AI / ML functionality comprises one or more of:

[0337] An AI / ML functionality configuration;

[0338] An applicability indication for an AI / ML functionality;

[0339] A ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0340] A3. A method of A1 , A1* and A2, wherein in response to a parameter within the message, deleting or to storing one or more information associated to an AI / ML functionality of the first cell.

[0341] A4. A method of A1 , A1*, A2, and A3, further comprising after transitioning to Inactive state performing cell selection to the first cell, and keeping the one or more information associated to the AI / ML functionality stored while camping in the first cell. A5. A method of A4, wherein further comprising releasing the one or more information associated to the AI / ML functionality stored when re-selecting from the first cell to a second cell, different form the first cell.

[0342] A6. A method of A1 , A1*, A2, and A3, further comprising after transitioning to Inactive state performing cell selection to a second cell (different from the first cell), and releasing the one or more information associated to the AI / ML functionality.

[0343] A7. A method of A1 , A1*, A2, and A3, further comprising after transitioning to Inactive state starting a timer and while the timer is running keeping stored the one or more information associated to the AI / ML functionality; and, when the timer expires, deleting the one or more information associated to the AI / ML functionality.

[0344] UE embodiments starting from a Resume procedure

[0345] B1. A method at a UE in an Inactive state for resuming a procedure, the method comprising:

[0346] Initiating a resume procedure in a second cell, transmitting a resume request to a current network node (associated with the second cell) and

[0347] Upon transmitting the resume request to the second cell, restoring one or more information associated to an AI / ML functionality when the one or more information associated to an AI / ML functionality were stored within the UE Inactive AS context.

[0348] B2. A method of B1 , wherein the restored one or more information associated to an AI / ML functionality comprises one or more of: an AI / ML functionality configuration e.g. inference configuration and / or an applicability indication for an AI / ML functionality_e.g. ‘applicable’ or ‘not applicable’ and / or a ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0349] B2* A method of B1 , wherein on the basis of the restored one or more information associated to an AI / ML functionality, the UE evaluates the applicability of the AI / ML functionality, and indicates an applicability indication in the resume request.

[0350] B3. A method of B1 and B2, B2*, further comprising: upon transmitting the resume request and restoring the one or more information associated to an AI / ML functionality, receiving a resume message.

[0351] B4. A method of B1 , B2 , B2*and B3, wherein: when the resume message does not contain an AI / ML functionality configuration e.g. inference configuration, considering the restored AI / ML functionality configuration (e.g. inference configuration) the AI / ML functionality configuration (e.g. the inference configuration) to be utilized by the UE in the second cell the UE is resuming; or when the resume message contains an AI / ML functionality configuration, applying the received AI / ML functionality configuration on top of the restored AI / ML functionality configuration.

[0352] B5. A method of B1 , B2 , B2*and B3, wherein:

[0353] After determining the AI / ML functionality to be used in the second cell, determining an applicability indication for an AI / ML functionality based on the resulting configuration after the UE applied the received AI / ML functionality configuration on top of the restored AI / ML functionality configuration.

[0354] Last serving network node embodiments starting from a Suspend procedure

[0355] C1. A method at a last serving network node, associated with a first cell, comprising:

[0356] Transmitting a message indicating the UE to transition from a connected state to an Inactive state and

[0357] Storing one or more information associated to an AI / ML functionality.

[0358] C1*. A method at a last serving network node, associated with a first cell, comprising:

[0359] Transmitting a message indicating the UE to transition from a connected state to an Inactive state and

[0360] Deleting one or more information associated to an AI / ML functionality.

[0361] C2. A method of C1 , wherein the one or more information associated to an AI / ML functionality comprises one or more of:

[0362] An AI / ML functionality configuration;

[0363] An applicability indication for an AI / ML functionality;

[0364] A ‘state’ of the AI / ML functionality e.g. ‘activated’ or ‘deactivated’.

[0365] C3. A method of C1 and C2, wherein the one or more information associated to an AI / ML functionality is stored in a UE Inactive AS Context. C4. A method of C1 and C2, further comprising:

[0366] Receiving a message from a current network node request the UE Inactive AS Context, wherein the UE context includes the one or more information associated to an AI / ML functionality;

[0367] And, in response to the message, transmitting to the current network node a response message including the UE context.

[0368] Current network node embodiments starting from a Resume procedure

[0369] D1. A method at a current network, associated to a second cell a UE is trying to rtesume to, the method comprising:

[0370] Receiving a resume request from a UE in Inactive state

[0371] In response transmitting to a last serving network node indicated in a resume identifier included in the resume request a context request message

[0372] In response to the context request message, receiving a UE AS Inactive context including the previously stored one or more information associated to an AI / ML functionality

[0373] And transmitting to the UE a resume message.

[0374] Figure 11 shows an example of a communication system QQ100 in accordance with some embodiments.

[0375] In the example, the communication system QQ100 includes a telecommunication network QQ102 that includes an access network QQ104, such as a radio access network (RAN), and a core network QQ106, which includes one or more core network nodes QQ108. The access network QQ104 includes one or more access network nodes, such as network nodes QQ110a and QQ110b (one or more of which may be generally referred to as network nodes QQ110), or any other similar 3rd Generation Partnership Project (3GPP) access nodes or non-3GPP access points. Moreover, as will be appreciated by those of skill in the art, a network node is not necessarily limited to an implementation in which a radio portion and a baseband portion are supplied and integrated by a single vendor. Thus, it will be understood that network nodes include disaggregated implementations or portions thereof. For example, in some embodiments, the telecommunication network QQ102 includes one or more Open- RAN (ORAN) network nodes. An ORAN network node is a node in the telecommunication network QQ102 that supports an ORAN specification (e.g., a specification published by the O-RAN Alliance, or any similar organization) and may operate alone or together with other nodes to implement one or more functionalities of any node in the telecommunication network QQ102, including one or more network nodes QQ110 and / or core network nodes QQ108.

[0376] Examples of an ORAN network node include an open radio unit (0-Rll), an open distributed unit (O-DU), an open central unit (O-CU), including an O-CU control plane (O-CU-CP) or an O-CU user plane (O-CU-UP), a RAN intelligent controller (near-real time or non-real time) hosting software or software plug-ins, such as a near-real time control application (e.g., xApp) or a non-real time control application (e.g., rApp), or any combination thereof (the adjective “open” designating support of an ORAN specification). The network node may support a specification by, for example, supporting an interface defined by the ORAN specification, such as an A1, F1, W1, E1 , E2, X2, Xn interface, an open fronthaul user plane interface, or an open fronthaul management plane interface. Moreover, an ORAN access node may be a logical node in a physical node. Furthermore, an ORAN network node may be implemented in a virtualization environment (described further below) in which one or more network functions are virtualized. For example, the virtualization environment may include an O-Cloud computing platform orchestrated by a Service Management and Orchestration Framework via an 0-2 interface defined by the O-RAN Alliance or comparable technologies. The network nodes QQ110 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs QQ112a, QQ112b, QQ112c, and QQ112d (one or more of which may be generally referred to as UEs QQ112) to the core network QQ106 over one or more wireless connections.

[0377] Example wireless communications over a wireless connection include transmitting and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors. Moreover, in different embodiments, the communication system QQ100 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of data and / or signals whether via wired or wireless connections. The communication system QQ100 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.

[0378] The UEs QQ112 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and / or operable to communicate wirelessly with the network nodes QQ110 and other communication devices. Similarly, the network nodes QQ110 are arranged, capable, configured, and / or operable to communicate directly or indirectly with the UEs QQ112 and / or with other network nodes or equipment in the telecommunication network QQ102 to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in the telecommunication network QQ102.

[0379] In the depicted example, the core network QQ106 connects the network nodes QQ110 to one or more host computing systems, such as host QQ116. These connections may be direct or indirect via one or more intermediary networks or devices. In other examples, network nodes may be directly coupled to hosts. The core network QQ106 includes one more core network nodes (e.g., core network node QQ108) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and / or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node QQ108. Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (ALISF), Subscription Identifier Deconcealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and / or a User Plane Function (UPF).

[0380] The host QQ116 may be under the ownership or control of a service provider other than an operator or provider of the access network QQ104 and / or the telecommunication network QQ102. The host QQ116 may host a variety of applications to provide one or more service. Examples of such applications include live and pre-recorded audio / video content, data collection services such as retrieving and compiling data on various ambient conditions detected by a plurality of UEs, analytics functionality, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and surveillance center, or any other such function performed by a server.

[0381] As a whole, the communication system QQ100 of Figure 11 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and / or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and / or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox.

[0382] In some examples, the telecommunication network QQ102 is a cellular network that implements 3GPP standardized features. Accordingly, the telecommunications network QQ102 may support network slicing to provide different logical networks to different devices that are connected to the telecommunication network QQ102. For example, the telecommunications network QQ102 may provide Ultra Reliable Low Latency Communication (URLLC) services to some UEs, while providing Enhanced Mobile Broadband (eMBB) services to other UEs, and / or Massive Machine Type Communication (mMTC) / Massive loT services to yet further UEs.

[0383] In some examples, the UEs QQ112 are configured to transmit and / or receive information without direct human interaction. For instance, a UE may be designed to transmit information to the access network QQ104 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network QQ104. Additionally, a UE may be configured for operating in single- or multi-RAT or multi-standard mode. For example, a UE may operate with any one or combination of Wi-Fi, NR (New Radio) and LTE, i.e. being configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved- UMTS Terrestrial Radio Access Network) New Radio - Dual Connectivity (EN-DC).

[0384] In the example, the hub QQ114 communicates with the access network QQ104 to facilitate indirect communication between one or more UEs (e.g., UE QQ112c and / or QQ112d) and network nodes (e.g., network node QQ110b). In some examples, the hub QQ114 may be a controller, router, content source and analytics, or any of the other communication devices described herein regarding UEs. For example, the hub QQ114 may be a broadband router enabling access to the core network QQ106 for the UEs. As another example, the hub QQ114 may be a controller that sends commands or instructions to one or more actuators in the UEs. Commands or instructions may be received from the UEs, network nodes QQ110, or by executable code, script, process, or other instructions in the hub QQ114. As another example, the hub QQ114 may be a data collector that acts as temporary storage for UE data and, in some embodiments, may perform analysis or other processing of the data. As another example, the hub QQ114 may be a content source. For example, for a UE that is a VR device, display, loudspeaker, or other media delivery device, the hub QQ114 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub QQ114 then provides to the UE either directly, after performing local processing, and / or after adding additional local content. In still another example, the hub QQ114 acts as a proxy server or orchestrator for the UEs, in particular if one or more of the UEs are low energy loT devices.

[0385] The hub QQ114 may have a constant / persistent or intermittent connection to the network node QQ110b. The hub QQ114 may also allow for a different communication scheme and / or schedule between the hub QQ114 and UEs (e.g., UE QQ112c and / or QQ112d) , and between the hub QQ114 and the core network QQ106. In other examples, the hub QQ114 is connected to the core network QQ106 and / or one or more UEs via a wired connection. Moreover, the hub QQ114 may be configured to connect to an M2M service provider over the access network QQ104 and / or to another UE over a direct connection. In some scenarios, UEs may establish a wireless connection with the network nodes QQ110 while still connected via the hub QQ114 via a wired or wireless connection. In some embodiments, the hub QQ114 may be a dedicated hub - that is, a hub whose primary function is to route communications to / from the UEs from / to the network node QQ110b. In other embodiments, the hub QQ114 may be a non-dedicated hub - that is, a device which is capable of operating to route communications between the UEs and network node QQ110b, but which is additionally capable of operating as a communication start and / or end point for certain data channels.

[0386] Figure 12 shows a UE QQ200 in accordance with some embodiments. The UE QQ200 presents additional details of some embodiments of the UE QQ112 of Figure 11. As used herein, a UE refers to a device capable, configured, arranged and / or operable to communicate wirelessly with network nodes and / or other UEs. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage / playback device, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), an Augmented Reality (AR) or Virtual Reality (VR) device, wireless customer-premise equipment (CPE), vehicle, vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB- loT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0387] A UE may support device-to-device (D2D) communication, for example by implementing a 3GPP standard for sidelink communication, Dedicated Short-Range Communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X). In other examples, a UE may not necessarily have a user in the sense of a human user who owns and / or operates the relevant device. Instead, a UE may represent a device that is intended for sale to, or operation by, a human user but which may not, or which may not initially, be associated with a specific human user (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended for sale to, or operation by, an end user but which may be associated with or operated for the benefit of a user (e.g., a smart power meter).

[0388] The UE QQ200 includes processing circuitry QQ202 that is operatively coupled via a bus QQ204 to an input / output interface QQ206, a power source QQ208, a memory QQ210, a communication interface QQ212, and / or any other component, or any combination thereof. Certain UEs may utilize all or a subset of the components shown in Figure 12. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0389] The processing circuitry QQ202 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory QQ210. The processing circuitry QQ202 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above. For example, the processing circuitry QQ202 may include multiple central processing units (CPUs). The processing circuitry QQ202 may be configured to cause the UE QQ202 to perform the methods as described with reference to Figure 3 and / or 4.

[0390] In the example, the input / output interface QQ206 may be configured to provide an interface or interfaces to an input device, output device, or one or more input and / or output devices. Examples of an output device include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, an emitter, a smartcard, another output device, or any combination thereof. An input device may allow a user to capture information into the UE QQ200. Examples of an input device include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a web camera, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smartcard, and the like. The presence-sensitive display may include a capacitive or resistive touch sensor to sense input from a user. A sensor may be, for instance, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof. An output device may use the same type of interface port as an input device. For example, a Universal Serial Bus (USB) port may be used to provide an input device and an output device.

[0391] In some embodiments, the power source QQ208 is structured as a battery or battery pack. Other types of power sources, such as an external power source (e.g., an electricity outlet), photovoltaic device, or power cell, may be used. The power source QQ208 may further include power circuitry for delivering power from the power source QQ208 itself, and / or an external power source, to the various parts of the UE QQ200 via input circuitry or an interface such as an electrical power cable. Delivering power may be, for example, for charging of the power source QQ208. Power circuitry may perform any formatting, converting, or other modification to the power from the power source QQ208 to make the power suitable for the respective components of the UE QQ200 to which power is supplied.

[0392] The memory QQ210 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth. In one example, the memory QQ210 includes one or more application programs QQ214, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data QQ216. The memory QQ210 may store, for use by the UE QQ200, any of a variety of various operating systems or combinations of operating systems.

[0393] The memory QQ210 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD-DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and / or ISIM, other memory, or any combination thereof. The UICC may for example be an embedded UICC (eUlCC), integrated UICC (iUICC) or a removable UICC commonly known as ‘SIM card.’ The memory QQ210 may allow the UE QQ200 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to off-load data, or to upload data. An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory QQ210, which may be or comprise a device-readable storage medium.

[0394] The processing circuitry QQ202 may be configured to communicate with an access network or other network using the communication interface QQ212. The communication interface QQ212 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna QQ222. The communication interface QQ212 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network). Each transceiver may include a transmitter QQ218 and / or a receiver QQ220 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth). Moreover, the transmitter QQ218 and receiver QQ220 may be coupled to one or more antennas (e.g., antenna QQ222) and may share circuit components, software or firmware, or alternatively be implemented separately.

[0395] In the illustrated embodiment, communication functions of the communication interface QQ212 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communications such as Bluetooth, near-field communication, location-based communication such as the use of the global positioning system (GPS) to determine a location, another like communication function, or any combination thereof. Communications may be implemented in according to one or more communication protocols and / or standards, such as IEEE 802.11, Code Division Multiplexing Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, transmission control protocol / internet protocol (TCP / IP), synchronous optical networking (SONET), Asynchronous Transfer Mode (ATM), QUIC, Hypertext Transfer Protocol (HTTP), and so forth.

[0396] Regardless of the type of sensor, a UE may provide an output of data captured by its sensors, through its communication interface QQ212, via a wireless connection to a network node. Data captured by sensors of a UE can be communicated through a wireless connection to a network node via another UE. The output may be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to even out the load from reporting from several sensors), in response to a triggering event (e.g., when moisture is detected an alert is sent), in response to a request (e.g., a user initiated request), or a continuous stream (e.g., a live video feed of a patient). As another example, a UE comprises an actuator, a motor, or a switch, related to a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input the states of the actuator, the motor, or the switch may change. For example, the UE may comprise a motor that adjusts the control surfaces or rotors of a drone in flight according to the received input or to a robotic arm performing a medical procedure according to the received input.

[0397] A UE, when in the form of an Internet of Things (loT) device, may be a device for use in one or more application domains, these domains comprising, but not limited to, city wearable technology, extended industrial application and healthcare. Non-limiting examples of such an loT device are a device which is or which is embedded in: a connected refrigerator or freezer, a TV, a connected lighting device, an electricity meter, a robot vacuum cleaner, a voice controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door / window sensor, a flood / moisture sensor, an electrical door lock, a connected doorbell, an air conditioning system like a heat pump, an autonomous vehicle, a surveillance system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smart watch, a fitness tracker, a wearable for tactile augmentation or sensory enhancement, a water sprinkler, an animal- or item-tracking device, a sensor for monitoring a plant or animal, an industrial robot, an Unmanned Aerial Vehicle (UAV), and any kind of medical device, like a heart rate monitor or a remote controlled surgical robot. A UE in the form of an loT device comprises circuitry and / or software in dependence of the intended application of the loT device in addition to other components as described in relation to the UE QQ200 shown in Figure 12.

[0398] As yet another specific example, in an loT scenario, a UE may represent a machine or other device that performs monitoring and / or measurements, and transmits the results of such monitoring and / or measurements to another UE and / or a network node. The UE may in this case be an M2M device, which may in a 3GPP context be referred to as an MTC device. As one particular example, the UE may implement the 3GPP NB-loT standard. In other scenarios, a UE may represent a vehicle, such as a car, a bus, a truck, a ship and an airplane, or other equipment that is capable of monitoring and / or reporting on its operational status or other functions associated with its operation.

[0399] In practice, any number of UEs may be used together with respect to a single use case. For example, a first UE might be or be integrated in a drone and provide the drone’s speed information (obtained through a speed sensor) to a second UE that is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE may adjust the throttle on the drone (e.g. by controlling an actuator) to increase or decrease the drone’s speed. The first and / or the second UE can also include more than one of the functionalities described above. For example, a UE might comprise the sensor and the actuator, and handle communication of data for both the speed sensor and the actuators.

[0400] Figure 13 shows a network node QQ300 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), O-RAN nodes or components of an O-RAN node (e.g., O-RU, O-DU, O-CU).

[0401] Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on the provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations. A base station may be a relay node or a relay donor node controlling a relay. A network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units, distributed units (e.g., in an O- RAN access node) and / or remote radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio. Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).

[0402] Other examples of network nodes include multiple transmission point (multi-TRP) 5G access nodes, multi-standard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi-cel l / multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and / or Minimization of Drive Tests (MDTs).

[0403] The network node QQ300 includes a processing circuitry QQ302, a memory QQ304, a communication interface QQ306, and a power source QQ308. The network node QQ300 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components. In certain scenarios in which the network node QQ300 comprises multiple separate components (e.g., BTS and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs. In such a scenario, each unique NodeB and RNC pair, may in some instances be considered a single separate network node. In some embodiments, the network node QQ300 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory QQ304 for different RATs) and some components may be reused (e.g., a same antenna QQ310 may be shared by different RATs). The network node QQ300 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node QQ300, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node QQ300.

[0404] The processing circuitry QQ302 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic operable to provide, either alone or in conjunction with other network node QQ300 components, such as the memory QQ304, to provide network node QQ300 functionality. For example, the processing circuitry QQ302 may be configured to cause the network node to perform the methods as described with reference to Figure 5 and / or 6.

[0405] In some embodiments, the processing circuitry QQ302 includes a system on a chip (SOC). In some embodiments, the processing circuitry QQ302 includes one or more of radio frequency (RF) transceiver circuitry QQ312 and baseband processing circuitry QQ314. In some embodiments, the radio frequency (RF) transceiver circuitry QQ312 and the baseband processing circuitry QQ314 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry QQ312 and baseband processing circuitry QQ314 may be on the same chip or set of chips, boards, or units.

[0406] The memory QQ304 may comprise any form of volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and / or any other volatile or non-volatile, non-transitory device-readable and / or computer-executable memory devices that store information, data, and / or instructions that may be used by the processing circuitry QQ302. The memory QQ304 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and / or other instructions capable of being executed by the processing circuitry QQ302 and utilized by the network node QQ300. The memory QQ304 may be used to store any calculations made by the processing circuitry QQ302 and / or any data received via the communication interface QQ306. In some embodiments, the processing circuitry QQ302 and memory QQ304 is integrated.

[0407] The communication interface QQ306 is used in wired or wireless communication of signaling and / or data between a network node, access network, and / or UE. As illustrated, the communication interface QQ306 comprises port(s) / terminal(s) QQ316 to send and receive data, for example to and from a network over a wired connection. The communication interface QQ306 also includes radio front-end circuitry QQ318 that may be coupled to, or in certain embodiments a part of, the antenna QQ310. Radio front-end circuitry QQ318 comprises filters QQ320 and amplifiers QQ322. The radio front-end circuitry QQ318 may be connected to an antenna QQ310 and processing circuitry QQ302. The radio front-end circuitry may be configured to condition signals communicated between antenna QQ310 and processing circuitry QQ302. The radio front-end circuitry QQ318 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front-end circuitry QQ318 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters QQ320 and / or amplifiers QQ322. The radio signal may then be transmitted via the antenna QQ310.

[0408] Similarly, when receiving data, the antenna QQ310 may collect radio signals which are then converted into digital data by the radio front-end circuitry QQ318. The digital data may be passed to the processing circuitry QQ302. In other embodiments, the communication interface may comprise different components and / or different combinations of components.

[0409] In certain alternative embodiments, the network node QQ300 does not include separate radio front-end circuitry QQ318, instead, the processing circuitry QQ302 includes radio frontend circuitry and is connected to the antenna QQ310. Similarly, in some embodiments, all or some of the RF transceiver circuitry QQ312 is part of the communication interface QQ306. In still other embodiments, the communication interface QQ306 includes one or more ports or terminals QQ316, the radio front-end circuitry QQ318, and the RF transceiver circuitry QQ312, as part of a radio unit (not shown), and the communication interface QQ306 communicates with the baseband processing circuitry QQ314, which is part of a digital unit (not shown).

[0410] The antenna QQ310 may include one or more antennas, or antenna arrays, configured to send and / or receive wireless signals. The antenna QQ310 may be coupled to the radio front- end circuitry QQ318 and may be any type of antenna capable of transmitting and receiving data and / or signals wirelessly. In certain embodiments, the antenna QQ310 is separate from the network node QQ300 and connectable to the network node QQ300 through an interface or port.

[0411] The antenna QQ310, communication interface QQ306, and / or the processing circuitry QQ302 may be configured to perform any receiving operations and / or certain obtaining operations described herein as being performed by the network node. Any information, data and / or signals may be received from a UE, another network node and / or any other network equipment. Similarly, the antenna QQ310, the communication interface QQ306, and / or the processing circuitry QQ302 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and / or signals may be transmitted to a UE, another network node and / or any other network equipment.

[0412] The power source QQ308 provides power to the various components of network node QQ300 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component). The power source QQ308 may further comprise, or be coupled to, power management circuitry to supply the components of the network node QQ300 with power for performing the functionality described herein. For example, the network node QQ300 may be connectable to an external power source (e.g., the power grid, an electricity outlet) via an input circuitry or interface such as an electrical cable, whereby the external power source supplies power to power circuitry of the power source QQ308. As a further example, the power source QQ308 may comprise a source of power in the form of a battery or battery pack which is connected to, or integrated in, power circuitry. The battery may provide backup power should the external power source fail.

[0413] Embodiments of the network node QQ300 may include additional components beyond those shown in Figure 13 for providing certain aspects of the network node’s functionality, including any of the functionality described herein and / or any functionality necessary to support the subject matter described herein. For example, the network node QQ300 may include user interface equipment to allow input of information into the network node QQ300 and to allow output of information from the network node QQ300. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node QQ300. In some embodiments providing a core network node, such as core network node 108 of FIGURE 11, some components, such as the radio front-end circuitry QQ318 and the RF transceiver circuitry QQ312 may be omitted. Figure 14 is a block diagram illustrating a virtualization environment QQ400 in which functions implemented by some embodiments may be virtualized. In the present context, virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments QQ400 hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized. In some embodiments, the virtualization environment QQ400 includes components defined by the O-RAN Alliance, such as an O-Cloud environment orchestrated by a Service Management and Orchestration Framework via an 0-2 interface. Virtualization may facilitate distributed implementations of a network node, UE, core network node, or host.

[0414] Applications QQ402 (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment Q400 to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.

[0415] Hardware QQ404 includes processing circuitry, memory that stores software and / or instructions executable by hardware processing circuitry, and / or other hardware devices as described herein, such as a network interface, input / output interface, and so forth. Software may be executed by the processing circuitry to instantiate one or more virtualization layers QQ406 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs QQ408a and QQ408b (one or more of which may be generally referred to as VMs QQ408), and / or perform any of the functions, features and / or benefits described in relation with some embodiments described herein. The virtualization layer QQ406 may present a virtual operating platform that appears like networking hardware to the VMs QQ408.

[0416] The VMs QQ408 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer QQ406. Different embodiments of the instance of a virtual appliance QQ402 may be implemented on one or more of VMs QQ408, and the implementations may be made in different ways. Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV). NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.

[0417] In the context of NFV, a VM QQ408 may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine. Each of the VMs QQ408, and that part of hardware QQ404 that executes that VM, be it hardware dedicated to that VM and / or hardware shared by that VM with others of the VMs, forms separate virtual network elements. Still in the context of NFV, a virtual network function is responsible for handling specific network functions that run in one or more VMs QQ408 on top of the hardware QQ404 and corresponds to the application QQ402.

[0418] Hardware QQ404 may be implemented in a standalone network node with generic or specific components. Hardware QQ404 may implement some functions via virtualization. Alternatively, hardware QQ404 may be part of a larger cluster of hardware (e.g. such as in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration QQ410, which, among others, oversees lifecycle management of applications QQ402. In some embodiments, hardware QQ404 is coupled to one or more radio units that each include one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units may communicate directly with other hardware nodes via one or more appropriate network interfaces and may be used in combination with the virtual components to provide a virtual node with radio capabilities, such as a radio access node or a base station. In some embodiments, some signaling can be provided with the use of a control system QQ412 which may alternatively be used for communication between hardware nodes and radio units.

[0419] Although the computing devices described herein (e.g., UEs, network nodes) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non- computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0420] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer- readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer- readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

Claims

1. Claims1. A method (300) performed by a User Equipment, UE, for storing or deleting information, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, the method comprising: transitioning (302) to an inactive state; and storing or deleting (304) at least part of information associated with the functionality.

2. The method of claim 1 , comprising receiving a message on a first cell and / or from a first Radio Access Network, RAN, node, wherein transitioning to the inactive state is performed after or in response to the message.

3. The method of claim 2, wherein the message includes an indication of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality.

4. The method of any of claims 1 to 3, wherein system information includes an indication of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality.

5. The method of claim 1 or 2, comprising storing the at least part of the information associated with the functionality in an inactive access stratum, AS, context for the UE.

6. The method of any of claims 1 to 5, wherein transitioning (302) to the inactive state is performed as part of a suspend procedure.

7. The method of any of claims 1 to 6, wherein transitioning (302) to the inactive state is performed when the UE is connected to a first cell or a first Radio Access Network, RAN, node.

8. The method of claim 7, comprising performing cell selection to the first cell or the first RAN node, or to a second cell or a second RAN node.

9. The method of any of claims 1 to 8, comprising receiving a resume message, and transitioning to the connected state after or in response to the resume message.

10. The method of claim 9 when dependent on claim 2 or 8, wherein the resume message is received on the first cell or from the first RAN node, or is received on the second cell orfrom the second RAN node, and wherein the method comprises, after transitioning to the connected state, at least one of: determining that the functionality is applicable if the stored at least part of the information indicates that the functionality is applicable and the UE is connected to the first cell or the first RAN node; determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is applicable and the UE is connected to the second cell or the second RAN node; determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable; determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable; determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is not applicable.

11. The method of claim 9 or 10, comprising, after transitioning to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE.

12. The method of any of claims 1 to 11, wherein storing or deleting (304) at least part of the information associated with the functionality comprises: storing the at least part of the information; starting a timer; and deleting the at least part of the information on expiry of the timer.

13. The method of any of claims 1 to 12, comprising deleting the stored at least part of the information after one or more of: transitioning to an idle state; performing a setup procedure; failure of a resume procedure.

14. A method (400) performed by a User Equipment, UE, for restoring information, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, the method comprising: transitioning (402) to a connected state; and restoring (404) at least part of information associated with the functionality.

15. The method of claim 14, wherein transitioning (402) to the connected state is performed to connect to a first cell or a first Radio Access Network, RAN, node, and the method comprises performing cell selection to the first cell or the first RAN node.

16. The method of claim 14, wherein transitioning (402) to the connected state is performed to connect to a second cell or a second Radio Access Network, RAN, node, and the method comprisesperforming cell reselection to the second cell or the second RAN node.

17. The method of any of claims 14 to 16, comprising, before transitioning (402) to the connected state, receiving a resume message, wherein transitioning to the connected state is performed after or in response to the resume message.

18. The method of claim 17 when dependent on claim 15, wherein the resume message is received on the first cell or from the first RAN node, or when dependent on claim 16, wherein the resume message is received on the second cell or from the second RAN node, and wherein the method comprises, after transitioning (402) to the connected state, at least one of: determining that the functionality is applicable if the restored at least part of the information indicates that the functionality is applicable and the UE is connected to the first cell or the first RAN node; determining that the functionality is not applicable if the restored at least part of the information indicates that the functionality is applicable and the UE is connected to the second cell or the second RAN node; determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable; determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable; determining that the functionality is not applicable if the restored at least part of the information indicates that the functionality is not applicable.

19. The method of any of claims 14 to 18, comprising, after transitioning (402) to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE.

20. The method of any of claims 15 to 19, comprising sending an indication of the applicability of the functionality on the first cell or to the first network node, or on the second cell or to the second network node.

21. A method (500) performed by a first Radio Access Network, RAN, node for storing or deleting information, the method comprising: determining (502) that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and storing or deleting (504) at least part of information associated with the functionality.

22. The method of claim 21 , wherein determining (502) that the UE transitions to the inactive state comprises sending a message to the UE, wherein the UE transitions to the inactive state after or in response to the message.

23. The method of claim 21 or 22, comprising broadcasting system information that includes an indication of whether to store the at least part of information associated with the functionality or to delete the least part of information associated with the functionality.

24. The method of any of claims 21 to 23, comprising storing the at least part of the information associated with the functionality in an inactive access stratum, AS, context for the UE.

25. The method of any of claims 21 to 24, comprising determining that the UE transitions to a connected state, and receiving a resume message from the UE or sending a resume message to the UE, wherein transitioning to the connected state is performed after or in response to the resume message.

26. The method of claim 25, comprising, after the UE transitions to the connected state, at least one of: determining that the functionality is applicable if the stored at least part of the information indicates that the functionality is applicable; determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable; determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable; determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is not applicable.

27. The method of claim 25 or 26, comprising, after the UE transitions to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE.

28. The method of any of claims 21 to 27, comprising deleting the stored at least part of the information after one or more of: the UE transitions to an idle state; performing a setup procedure; failure of a resume procedure.

29. A method (600) performed by a first Radio Access Network, RAN, node for determining information, the method comprising: determining (602) that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and determining (604) at least part of information associated with the functionality.

30. The method of claim 29, wherein determining (604) the at least part of the information associated with the functionality comprises determining the at least part of the information associated with the functionality from an inactive access stratum, AS, context for the UE.

31. The method of claim 30, wherein the inactive AS context for the UE is stored at the first RAN node, and the method comprises receiving the inactive AS context for the UE from a second RAN node or another network node.

32. The method of claim 31 , comprising sending, to the second RAN node or the another network node, a context request message for the inactive AS context for the UE before receiving the inactive AS context for the UE from the second RAN node or the another network node.

33. The method of any of claims 29 to 32, wherein determining (602) that the UE transitions to the connected state comprises receiving a resume message from the UE or sending a resume message to the UE.

34. The method of any of claims 29 to 33, comprising, after the UE transitions to the connected state, at least one of: determining that the functionality is applicable if the stored at least part of the information indicates that the functionality is applicable; determining that the functionality is applicable if an indication in the resume message indicates that the functionality is applicable; determining that the functionality is not applicable if an indication in the resume message indicates that the functionality is not applicable;determining that the functionality is not applicable if the stored at least part of the information indicates that the functionality is not applicable.

35. The method of any of claims 29 to 34, comprising, after the UE transitions to the connected state, determining applicability of the functionality based on the stored at least part of the information and / or current conditions of the UE.

36. The method of any of claims 29 to 35, comprising determining that the functionality is not applicable if the first RAN node is not associated with a last serving cell of the UE.

37. A tangible, non-transient computer-readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations a User Equipment, UE, for storing or deleting information, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, the operations comprising: transitioning (302) to an inactive state; and storing or deleting (304) at least part of information associated with the functionality.

38. The computer-readable medium of claim 37, comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform the method (300) of any of claims 2 to 13.

39. A tangible, non-transient computer-readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations a User Equipment, UE, for restoring information, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, the operations comprising: transitioning (402) to a connected state; and restoring (404) at least part of information associated with the functionality.

40. The computer-readable medium of claim 39, comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform the method (400) of any of claims 15 to 20.

41. A tangible, non-transient computer-readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations in a first Radio Access Network, RAN, node for storing or deleting information, the operations comprising:determining (502) that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and storing or deleting (504) at least part of information associated with the functionality.

42. The computer-readable medium of claim 41 , comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform the method (500) of any of claims 22 to 28.

43. A tangible, non-transient computer-readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform operations in a first Radio Access Network, RAN, node for determining information, the operations comprising: determining (602) that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and determining (602) at least part of information associated with the functionality.

44. The computer-readable medium of claim 43, comprising instructions that, when executed by processing circuitry, cause the processing circuitry to perform the method (600) of any of claims 30 to 36.

45. A computer-readable medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to carry out the method (300, 400, 500, 600) according to any of claims 1 to 36.

46. A computer program, comprising instructions that, when executed by processing circuitry, cause the processing circuitry to carry out the method (300, 400, 500, 600) according to any of claims 1 to 36.

47. A carrier containing the computer program of claim 46, wherein the carrier is one of an electronic signal, optical signal, radio signal, or computer-readable medium.

48. Apparatus in a User Equipment, UE, for storing or deleting information, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, the apparatus comprising processing circuitry and a memory, the apparatus configured to: transition (302) to an inactive state; and store or delete (304) at least part of information associated with the functionality.

49. The apparatus of claim 48, wherein the apparatus is configured to perform the method (300) of any of claims 2 to 13.

50. Apparatus in a User Equipment, UE, for restoring information, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model, the apparatus comprising processing circuitry and a memory, the apparatus configured to: transition (402) to a connected state; and restore (404) at least part of information associated with the functionality.

51. The apparatus of claim 50, wherein the apparatus is configured to perform the method (400) of any of claims 15 to 20.

52. Apparatus in a first Radio Access Network, RAN, node for storing or deleting information, the apparatus comprising processing circuitry and a memory, the apparatus configured to: determine (502) that a UE transitions to an inactive state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and store or delete (504) at least part of information associated with the functionality.

53. The apparatus of claim 52, wherein the apparatus is configured to perform the method (500) of any of claims 22 to 28.

54. Apparatus in a first Radio Access Network, RAN, node for determining information, the apparatus comprising processing circuitry and a memory, the apparatus configured to: determine (602) that a UE transitions to a connected state, wherein the UE has a functionality that uses an artificial intelligence or machine learning, AI / ML, model; and determine (604) at least part of information associated with the functionality.

55. The apparatus of claim 54, wherein the apparatus is configured to perform the method (600) of any of claims 30 to 36.

Images