Method, apparatus and computer program

Abstract

An apparatus comprising means for receiving information defining a time period from a base station, a means for selecting one or more of a plurality of user equipment (UE) side models during the time

Description

A communication network can be seen as a facility that enables communications between two or more communication devices, or provides communication devices access to a data network. A mobile or wireless communication network is one example of a communication network. A communication device may be provided with a service by an application server. Such communication networks operate in according with standards such as those provided by 3GPP (Third Generation Partnership Project) or ETSI (European Telecommunications Standards Institute). Examples of standards are the so-called 5G (5th Generation) standards provided by 3GPP. SUMMARY Some example embodiments of this disclosure will be described with respect to certain aspects. These aspects are not intended to indicate key or essential features of the embodiments of this disclosure, nor are they intended to be used to limit the scope of thereof. Other features, aspects, and elements will be readily apparent to a person skilled in the art in view of this disclosure. According to an aspect, there is provided an apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to perform: receiving information defining a time period from a base station; selecting one or more of a plurality of user equipment side models during the time period; and sending information to the base station indicating the selected one or more of the plurality of user equipment side models. In examples, one or more network related conditions are stable during said time period. In examples, the selecting comprises: selecting the one or more of the plurality of user equipment side models based on the one or more network related conditions. In examples, the receiving information comprises: receiving information defining a time duration in which the one or more of the plurality of user equipment side models is to be selected. In examples, the receiving information comprises: receiving a first message from the base station which starts the time period and / or receiving a second message from the base station which ends the time period. In examples, the apparatus is caused to perform: sending to the base station information indicating a capability of the user equipment for supporting user equipment side model selection. In examples, the sending comprises: sending to the base station information indicating one or more capability conditions associated with the capability of the user equipment for supporting user equipment side model selection. In examples, the one or more capability conditions associated with the capability of the user equipment comprise one or more of: a minimum length of the time period; required assistance from the base station; and / or one or more machine learning enabled features. In examples, the receiving comprises: receiving from the base station network coverage area information associated with the time period. In examples, the network coverage area information comprises one or more of: one or more cells; one or more transmission reception points; one or more physical cell identifiers; and / or area information. In examples, the selecting of the one or more user equipment side models comprises: assessing a performance of one or more user equipment side models and selecting based on the performance of one or more of the one or more user equipment side models. In examples, the receiving comprises: receiving assistance information from the base station and using the assistance information when selecting one or more of the one or more user equipment side models. In examples, the assistance information comprises one or more of: one or more reference signals; and information about one or more of the network related conditions. In examples, the apparatus is caused to perform: determining one or more of the network related conditions. In examples, the selecting one or more of the plurality of user equipment side models during the time period comprises: comparing one or more network conditions associated with a respective user equipment side model and one or more current network conditions. In examples, the apparatus is caused to perform: assigning an identity to a respective one of the selected one or more user equipment side models and sending the assigned identity to the base station. In examples, the sending comprises: sending a first indicator to the base station indicating if there are one or more models appropriate for the one or more network conditions and / or sending a second indicator to the base station providing an identifier associated with a respective one of the one or more user equipment side models. In examples, the sending comprises: sending the associated time period and the network coverage area information with the second indicator to the base station. In examples, the receiving comprises: subsequently receiving from the base station one or more commands for the selected one or more of the plurality of user equipment side models. According to an aspect, there is provided an apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to perform: sending, to a user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected; and receiving, from the user equipment, information indicating the selected one or more of the plurality of user equipment side models. In examples, the apparatus is caused to perform: keeping one or more network related conditions stable during said time period. In examples, the sending comprises: sending to the user equipment network coverage area information associated with the time period. In examples, the sending comprises: sending assistance information to the user equipment to be used when selecting one or more of the one or more user equipment side models. In examples, the receiving comprises: receiving from the user equipment a first indicator indicating if there are one or more models appropriate for the one or more network conditions and / or for receiving a second indicator comprising an identifier associated with a respective one of the one or more user equipment side models. In examples, the receiving comprises: receiving the associated time period and the network coverage area information with the second indicator from the user equipment. According to an aspect, there is provided a method, comprising: receiving information from a base station defining a time period; selecting one or more of a plurality of user equipment side models during the time period; and sending information to the base station indicating the selected one or more of the plurality of user equipment side models. In examples, one or more network related conditions are stable during said time period. In examples, the selecting comprises: selecting the one or more of the plurality of user equipment side models based on the one or more network related conditions. In examples, the receiving information comprises: receiving information defining a time duration in which the one or more of the plurality of user equipment side models is to be selected. In examples, the receiving information comprises: receiving a first message from the base station which starts the time period and / or receiving a second message from the base station which ends the time period. In examples, the method comprises: sending to the base station information indicating a capability of the user equipment for supporting user equipment side model selection. In examples, the sending comprises: sending to the base station information indicating one or more capability conditions associated with the capability of the user equipment for supporting user equipment side model selection. In examples, the one or more capability conditions associated with the capability of the user equipment comprise one or more of: a minimum length of the time period; required assistance from the base station; and / or one or more machine learning enabled features. In examples, the receiving comprises: receiving from the base station network coverage area information associated with the time period. In examples, the network coverage area information comprises one or more of: one or more cells; one or more transmission reception points; one or more physical cell identifiers; and / or area information. In examples, the selecting of the one or more user equipment side models comprises: assessing a performance of one or more user equipment side models and selecting based on the performance of one or more of the one or more user equipment side models. In examples, the receiving comprises: receiving assistance information from the base station and using the assistance information when selecting one or more of the one or more user equipment side models. In examples, the assistance information comprises one or more of: one or more reference signals; and information about one or more of the network related conditions. In examples, the method comprises: determining one or more of the network related conditions. In examples, the selecting one or more of the plurality of user equipment side models during the time period comprises: comparing one or more network conditions associated with a respective user equipment side model and one or more current network conditions. In examples, the method comprises: assigning an identity to a respective one of the selected one or more user equipment side models and sending the assigned identity to the base station. In examples, the sending comprises: sending a first indicator to the base station indicating if there are one or more models appropriate for the one or more network conditions and / or sending a second indicator to the base station providing an identifier associated with a respective one of the one or more user equipment side models. In examples, the sending comprises: sending the associated time period and the network coverage area information with the second indicator to the base station. In examples, the receiving comprises: subsequently receiving from the base station one or more commands for the selected one or more of the plurality of user equipment side models. According to an aspect, there is provided a method, comprising: sending, to an user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected; and receiving, from the user equipment, information indicating the selected one or more of the plurality of user equipment side models. In examples, the method comprises: keeping one or more network related conditions stable during said time period. In examples, the sending comprises: sending to the user equipment network coverage area information associated with the time period. In examples, the sending comprises: sending assistance information to the user equipment to be used when selecting one or more of the one or more user equipment side models. In examples, the receiving comprises: receiving from the user equipment a first indicator indicating if there are one or more models appropriate for the one or more network conditions and / or for receiving a second indicator comprising an identifier associated with a respective one of the one or more user equipment side models. In examples, the receiving comprises: receiving the associated time period and the network coverage area information with the second indicator from the user equipment. According to an aspect, there is provided an apparatus comprising: means for receiving information defining a time period from a base station; means for selecting one or more of a plurality of user equipment side models during the time period; and means for sending information to the base station indicating the selected one or more of the plurality of user equipment side models. According to an aspect, there is provided an apparatus comprising: means for sending, to a user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected; and means for receiving, from the user equipment, information indicating the selected one or more of the plurality of user equipment side models. In examples, the means for sending is for sending to the user equipment network coverage area information associated with the time period. In examples, the means for sending is for sending assistance information to the user equipment to be used when selecting one or more of the one or more user equipment side models. According to an aspect, there is provided a computer readable medium comprising instructions which, when executed by an apparatus, cause the apparatus to perform at least the following: receiving information from a base station; selecting one or more of a plurality of user equipment side models during the time period; and sending information to the base station indicating the selected one or more of the plurality of user equipment side models. According to an aspect, there is provided a computer readable medium comprising instructions which, when executed by an apparatus, cause the apparatus to perform at least the following: sending, to an user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected; and receiving, from the user equipment, information indicating the selected one or more of the plurality of user equipment side models. According to an aspect, there is provided a non-transitory computer readable medium comprising program instructions that, when executed by an apparatus, cause the apparatus to perform at least the method according to any of the preceding aspects. In the above, many different embodiments have been described. It should be appreciated that further embodiments may be provided by the combination of any two or more of the embodiments described above. DESCRIPTION OF FIGURES Some example embodiments will now be described, by way of non-limiting and illustrative example only, with reference to the accompanying Figures in which: Figure 1 shows a simple overview of some embodiments; Figures 2a and 2b show an example procedure of some embodiments Figure 3 shows a representation of a 5th generation communication system; Figure 4 shows a representation of an apparatus for the communication system of Figure 3 according to some example embodiments; Figure 5 shows a representation of an apparatus according to some example embodiments; Figure 6 shows a first method of some embodiments; Figure 7 shows a second method of some embodiments; Figure 8 shows a third method of some embodiments; and Figure 9 shows a fourth method of some embodiments. DETAILED DESCRIPTION Some embodiments relate to the use of artificial intelligence (Al) / machine learning (ML). AI / ML based models may be used in relation an air interface, that is the air interface between a user equipment and an access node. The use of these models may improve the perform of air-interface functions. Some embodiments will be described by way of example in relation to a 5G system where the access node is provided by a gNB. However, other embodiments may be provided in different systems. An AI / ML Model is a data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs. A user equipment may have a number of machine learning models. These may be a UE-side model or a two-sided model. A two-sided (AI / ML) model is a paired AI / ML model(s) over which joint inference is performed. Joint inference comprises AI / ML inference whose inference is performed jointly across the UE and the network, i.e, the first part of inference is firstly performed by UE and then the remaining part is performed by gNB, or vice versa. A UE-side (AI / ML) model is an AI / ML Model whose inference is performed entirely at the UE. Some embodiments relate to model identification where one or more of the models on the UE is selected or identified. Model identification is a process / method of identifying an AI / ML model. This may be for a common understanding between the network and the UE. Information regarding the AI / ML model may be shared during model identification. Model selection is a process for selecting an AI / ML model for activation among multiple models for the same AI / ML enabled feature. A UE-sided model may be associated with a set of one or more additional conditions. An additional condition may comprise or more of the following: training dataset information; site-related information (e.g., scenario, location / TRP (transmission reception point) / area information, beam direction, codebook information and / or the like); time related / Timestamp information gNB implementation related information; UE implementation related information; statistical information (e.g. delay spread, angular spread, LOS / NLOS (line of sight / non line of sight) data and / or the like); and / or speed and range of speeds information It should be appreciated that these conditions are by way of example only and in other embodiments one or more other conditions may alternatively or additionally be used. For a UE-sided model or a two-sided model, the model training may be mapped to a set of assumptions considered for one or more additional conditions such as described previously. A UE-sided model or two-sided model can be associated with one or more of the additional conditions. One or more additional conditions may be regarded as NW-sided additional conditions (assumptions for the NW-side model in the two-sided model). One or more additional conditions may be regarded as UE-sided additional conditions (assumptions on UE-side for the UE-sided model or the UE side model in the two-sided model). One or more additional conditions may be regarded as common additional conditions (assumptions which are generic for both UE and NW side models). However, describing these additional conditions for online model identification is not straight forward. This may an issue in the scenario where a model may be identified via over-the-air signaling. Some embodiments may address or mitigate this issue. Model identification is a process / method of identifying an AI / ML model for the common understanding between the NW and the UE. A model may be identified via over-the-air signaling. In some 5G standardization this is referred to a Type B. The model identification may be initiated by the UE, and the NW assists the remaining steps (if any) of the model identification. The model may be assigned with a model ID during the model identification. In some 5G standardization this is referred to Type B1. The model identification may be initiated by the NW, and UE responds (if applicable) for the remaining steps (if any) of the model identification. The model may be assigned with a model ID during the model identification. In some 5G standardization this is referred to Type B2. Reference is made to Figure 1 which shows a schematic overview of some embodiments. A UE 300 has three ML models, model 1, model 2, and model 3. The models are stored on the UE in one or more memories. The UE may have two or more models. In some embodiments, the UE will have more than three models. The UE 300 communicates with an access node, for example a gNB 100 via an air interface. In some embodiments, the access node, for example, the gNB can enquire about UE capabilities. The gNB may send an enquiry to the UE and / or a user data store in the network for information about the UE capabilities. These capabilities may relate ML / AI support. The user data store in the network may be provided by an AMF (Access and Mobility Management Function) and / or one or more other network functions of the network. After receiving the information about the UE capabilities, the gNB may cause the AMF to update the information which it has about the UE to include the information about the UE capabilities. A UE that supports an ML feature may report the associated one or more conditions to the gNB. These conditions may be provided in the information about the UE capabilities, (e.g., BM-Case1 - that is a ML-based beam prediction in spatial domain case) The UE may indicate that it supports online model identification in the information about the UE capabilities. The UE may indicate one or conditions for the online model identification the information about the UE capabilities. The one or more conditions for online model identification may include a preferred minimum time duration that UE wishes to be configured with to support model identification. The one or more conditions may include required assistance on DL RS (downlink reference signal) transmission, associated functionalities or conditions for the model identification. The UE may receive information about a model identification time period in which the UE is identify the UE sided model or UE part of the two-sided model that supports a required ML-feature. This defines a time duration for model identification. This model identification time period may be provided as a duration or length of time. Alternatively a signal may be received from the gNB to start or initiate the model identification time period and a further signal may be received from the gNB terminate or end the model identification time period, The UE can assume that the one or more NW-additional conditions remain stable during the time duration for model identification. The UE may select one or more UE-models that suit or are appropriate for the observed NW-additional conditions on the UE side during this time duration for model identification. The UE may be configured with information about the area for which the time duration for model identification is applicable. The information about the area may comprises information about one or more of: one or more cells; one or more TRPs (transmission reception point); one or more PCIs (physical cell identity); and / or any other suitable information defining the area. The U E may assume that the one or more NW additional conditions for the area remain stable during the time duration for model identification. The UE may select one or more UE-model that suit the observed NW-additional conditions for the area. From the gNB's side, the NW-sided additional conditions may be kept stable, as far as possible, for the area during the time duration for model identification. For example, the network may defer making any changes, where possible during this duration, which may result in the NW-additional conditions changing. For example, for the BM-case1, beam codebooks and / or antenna configurations / settings associated with the area may not be changed during the time duration for model identification. The time duration for model identification may be configured by the gNB. The time duration for model identification may be associated with: one or more NW-configured functionality(ies); one or more UE-reported condition(s); and / or configured independently for model identification. In some embodiments, the time duration for model identification may be defined in advance. The time duration for model identification may be defined in one or more standards. The UE may use a fixed time duration for model identification or where there are more than one fixed time durations for model identification, the UE may use one of those fixed time duration. In some embodiments, the UE may signal the fixed time duration for model identification to the gNB. In some embodiments, when there are more than one fixed time durations for model identification, the UE may use more than one fixed time duration for model identification. The UE can assume that the one or more NW-additional conditions remain stable during one fixed time duration for model identification, and it is not necessarily the case that the same NW-additional conditions are applicable in other fixed time durations for model identification. The UE may select one or more UE-models that suit or are appropriate for the observed NW-additional conditions across more than one fixed time duration for model identification. One or more DCI (downlink control information) or MAC-CE (media access control -control elements) messages may comprise information to one or more of start model identification and / or terminating the model identification. The DCI or MAC-CE messages may be sent to the UE by the gNB. This may be used where the UE may have variation in the time duration for model identification depending on when the NW wishes to keep the NW-additional conditions unchanged or not. In the time duration for model identification, the UE may determine, monitor, and / or assess the applicability of a UE-sided ML model(s) or the UE side of the two-sided ML model. This may be for a given scenario, area, and / or site-specific setting that the gNB plans to use for the ML feature. Model assessment and monitoring may be performed by the UE while supporting an ML functionality. The UE may be configured to or indicated to operate with this functionality during time duration for model identification. This may allow the UE to assess the performance of different UE-sided AI / ML models while supporting the required ML functionality. Alternatively or additionally, the model assessment and monitoring may be performed by the UE without supporting an ML functionality. During the time duration for model identification, the UE is not expected to operate with the ML functionality but can still assess the performance of different UE-sided AI / ML models. The model assessment and / or monitoring may be conducted by the UE with support from the NW or without support from the NW. A logical AI / ML model may refer to a model that is identified and assigned a model ID. A physical AI / ML model may refer to an actual implementation of such a model. In some embodiments, the UE may have N number of ML model (logical models) which are developed assuming different NW / UE / additional conditions. These may be common-additional conditions. In other words, N different ML models may be associated with N different assumptions on additional conditions. However, all these N models may be related to the same functionality. For example the N models may relate to a same RRC (radio resource control configuration that is configured by the gNB. The UE may monitor channel characteristics, beam measurements and / or one or more other aspects based on received signals. In the BM-case 1 example, the UE may monitor the propagation environment to provide channel characteristics. The UE may use one or more of L1-RSRP (layer 1 reference signal received power), L1-SINR (layer 1 signal to interference noise ratio) and / or beam arrival directions to provide beam measurements. The UE may make use of one or more of DL RSs (downlink reference signals), PDSCH (physical downlink shared channel), the PDCCH (physical downlink control channel) and / or the like to monitor one or more required aspects. When an adequate level of monitoring is available at the UE (within the time duration for model identification), the UE may be able to map these observations to NW-additional conditions. As the UE knows which NW-additional conditions that are assumed for a given ML model, the UE can select a ML model by comparing the observed NW-additional conditions to the additional conditions associated with a ML model. In some embodiments, the UE may employ an internal model monitoring procedure and may identify at least one model that can be applied to the scenario, area, and / or sitespecific setting that the gNB plans to use for the ML feature. This may be transparent to the gNB, For example, in the BM-case 1, if the UE measures a prediction beam set and measurement beam set for a given PCI / TRP or cell, the UE may be able to monitor the performance of N ML models. In some embodiments, the UE may monitor the performance on less than N of the ML models. In some embodiments, the UE may monitor 2 or more of the N ML models. The UE evaluate the performance at the UE side. When the UE has evaluated, the UE can identify which one of these N models are suited for considered NW-additional conditions in the time duration for model identification. This UE may provide an assessment outcome. In some embodiments, the UE, the gNB may provide monitoring resources and required assistance for the model monitoring procedure performed by the UE. In the BM-case 1, for example, the UE may get dedicated DL RS and assistance information to monitor ML models such that UE can evaluate the performance of ML models with the support of the NW. During or after the time duration for model identification, the UE may report one or more of the following: A first indicator to report whether a model (new or old) suits the NW-additional conditions considered during the time duration for model identification. The same first indicator may also be used to report the absence of a model that suits the NW-additional conditions considered during the time duration for model identification. The first indicator may be conveyed using one field (e.g., with a single bit). Other embodiments may use one or more fields. The field may comprise one or more bits. The first indicator may indicate whether there is a new identification of the UE-sided model; and / or A second indicator to report a model-ID of a respective model that is being identified by the UE. This may be a new model or a matching model-ID that was identified before with a model-ID. A new model applicable for the NW-additional conditions during the time duration for model identification may be reported by the UE. The UE may indicate a model-ID from a pre-defined / configured bitfield. This may be done in an incremental order of assigning newmodel-IDs or in any other suitable way. When the pre-defined / configured bit field used for the second indicator is full, the UE may replace an existing model-ID (from the already identified list) and report the new model that suits the NW-additional conditions used during the time duration for model identification. When an older model (that is previously reported to the NW) is applicable for the NW-additional conditions in the time duration for model identification, the UE can report the already identified model-ID. In some embodiments, the UE may indicate changes with the first indicator, (e.g., the first indicator may have two bits, one for new model indication, one for older model replacement, one bit for an existing model, and one for a no model identification where there is no suitable model. The UE report can be an NW-initiated report. For example, when the time duration for model identification is over, there can be a gNB-initiated scheduled report. The UE report can be an UE-initiated report. For example, the UE triggers a MAC-CE message to report the first and / or second indicators. In some embodiments, the UE may report additional information associated with the second indicator. The additional information may comprise associated cells and / or TRPs and / or PCIs and / or area information. In some embodiments, when UE identifies a model considering more than one fixed time duration for model identification, the UE may report additional information associated with the second indicator. The additional information may comprise associated fixed time durations for model identification, associated cells and / or TRPs and / or PCIs and / or area information. After receiving the UE report that indicates the first and / or second indicators, the gNB determines whether there is any new model identification or not. If there is new model identification with a new model-ID or with an older-model-ID, the gNB may associate the model-ID with the set of assumptions used by the cells and / or TRPs and / or PCIs and / or area and time duration information. For example, based on the information that gNB has, it can associate Model-ID_x to a set of conditions and a set of conditions known at the network. In the BM-case 1 example, the conditions are time information_K, cell_L, beam codebook_M, antenna configuration_N, TRP_implementation_P, and / or the like. It should be noted that the UE may be supporting single cell measurements or multiple Cell / TRP / PCI measurements. That means the model identification may be per cell and / or TRP and / or PCI or per group of cells and / or TRPs and / or PCIs within the time duration for model identification. This latter option may reduce the need to perform a model identification procedure with the TRP or cell changes. The gNB may send the Model-ID and optionally time information associated with the UE’s model identification process to for, example, the AMF. A set of information (for example gNB implementation-related information, such as the beam codebook used by the gNB, and antenna panel / configuration used by the gNB.) may be stored within the NW, for example the gNB, with the model-ID associated with them. For example, the gNB may report to the the AMF that Model-ID_x associated to a set of conditions, time information_K, Cell_L, and TRP_M. The AMF may gather more than one model-ID and related NW-additional conditions and conditions. Depending on the level of information that the gNB has the gNBs may operate with model-ID LCM towards the UE. When a detach / reattach of a UE to the network occurs, the AMF may flush the model-IDs fetched from gNBs and online model identification may start again for a UE. This may be performed after UE reports new capability information. The NW-additional conditions may be referred as additional conditions in DL direction. This may be the case where the UE is mainly observing these additional conditions based on DL channels / signals. From NW perspective, a model-ID identified during the time duration for model identification by the above may be used to identify the NW or common-additional conditions (or additional conditions in DL direction) of a ML model implicitly. The model-ID may be considered by gNB and UE for signalling purposes to align with additional conditions that the NW plans to use the ML feature. The Model -ID may not select an exact physical model at the UE. In general, if the UE can maintain many ML models for the same NW / common additional conditions, the model-ID reported by the UE can refer to one or more ML models at the UE side. If there are more than one ML model associated with a model-ID, there may be other additional selection methods used by the UE to select the ML model out of multiple ML models depending on the used UE-sided additional conditions considered at a given time. Reference is made to the procedure illustrated in Figures 2a and 2b. As referenced 1, the UE starts an initial attach where an ML capable UE reports the conditions associated with the UE ML-feature to the gNB. The initial attach may be for example, as a result of the UE being powered up. The UE may reports that the UE can support online model identification. The UE may report related parameters / conditions when supporting online model identification. The gNB may send the supported ML feature and conditions to the AMF. As referenced 2, the gNB determines one or more functionalities (e.g. configurations to enable a ML-feature) based on the reported conditions. The gNB may determine a suitable time duration for model identification. This may be based on the NW-planned variations of the NW-additional conditions and / or parameters and / or conditions of the UE associated with the online model identification. As referenced 3, the UE receives a configuration for online model identification. This may be received in an RRC configuration or reconfiguration message. The configuration for online model identification may provide the time duration for model identification. The configuration for online model identification may provide parameters such as which one or more cells, one or more TRPs, and / or one or more PCIs can be considered for online model identification. The configuration for online model identification may provide an indication of resources that the UE can use to measure / monitor background ML models on the UE side. The configuration for online model identification may be dependent on or relate to the one or more functionalities. As referenced 4, the time duration for model identification is started at the UE and the gNB. As referenced 5, the gNB identifies and does not change the NW-sided additional conditions for the associated one or more cells, one or more TRPs, and / or one or more PCIs. As referenced 6, the and the UE starts the assessing and / or monitoring ML models. As referenced 7, the UE performs a ML inference operation based on functionality LCM, and additionally with model-ID-LCM with models which are known by the start time of model identification time duration. An AI / ML model needs to be developed, deployed, and managed during the entire lifecycle—a process known as AI / ML model life cycle management (LCM). In functionality-based LCM, a network indicates activation and / or deactivation and / or fallback / switching of AI / ML functionality via signalling. A functionality refers to an AI / ML-enabled feature or feature group facilitated by a configuration. The configuration(s) is(are) supported based on conditions indicated by UE capability. In model-lD-based LCM, models are identified at the network, and network / UE may activate / deactivate / select / switch individual AI / ML models via model ID. A model may be associated with specific conditions associated with UE capability of an AI / ML-enabled feature or feature group and additional conditions In a 3GPP system, the signalling may be provided RRC signalling, MAC-CE, and / or DOI. As referenced 8, the gNB may assist the UE with the transmission of DL RS or assistance information to the UE. This assistance information may be used for model assessment and / or monitoring. The DL RSs may allow the UE to evaluate and / or observe NW additional conditions. As referenced 9, the time duration for model identification ends. As referenced 10, based on the assessment or monitoring that the UE does during the model identification time duration, the UE determine which ML model that suits the observed NW-additional conditions. This may be a new model to be identified to the gNB or a model which has previously been identified to the gNB. As referenced 11, after the end of the model identification time duration, the gNB can un-freeze the NW-additional conditions. For example, the gNB can to a different set of NW-additional conditions than the NW-additional conditions which applied during the model identification time duration. The timing of the part of the procedure referenced 10 and 11 can take place in either order or overlap in time. As referenced 12, the UE may assign a new model ID for the determined ML model(e.g., Model-ID_X). As referenced 13, the UE communicates the outcome of the model identification process to the gNB. The UE may provide a report which may include an indication of whether a new model is identified or not (e.g., new model identified) or if has the identity of a model which has previously been identified to the gNB. The report may specifies the assigned model-ID (e.g., Model-ID_X). The report may comprise the first and / or second indicators which have been described previously. As referenced 14, the UE may store or update the information for the determined model-ID and the associated time duration. This information may include the UE-additional conditions, observed NW-additional conditions, and / or other additional conditions. The UE can map the same model-ID to specific conditions and additional conditions. Examples of additional conditions may time information, cell information, observed NW-additional conditions, and other UE-additional conditions (e.g. UE- antenna configuration, UE implementation related, and / or the like. As referenced 15, after receiving the UE report the gNB evaluates whether any new model identifications have occurred. The report may contain the first and / or second indicators. If there are new model identifications, either with a new model-ID or an older model-ID, the gNB may associate the model-ID with a set of additional conditions relevant to the one or more cells, one ormoreTRPs, one or more PCIs, and / or one or more areas. The gNB may associate the model-ID with the time duration information. For example, based on available information, the gNB can map a Model-ID (e.g., Model-ID_X) to specific conditions and additional conditions, Examples of additional conditions may time information, cell information, beam codebook, antenna configuration, TRP implementation, and / or the like. As referenced 16, the gNB sends to the AMF, the Model-ID along with the associated time information related to the UE's model identification process. A complete set of information, including proprietary details or additional conditions that cannot be explicitly specified, may be stored within at the NW side, for example at the gNB. This may be associated with the respective model-ID. As a result of this approach, the AMF can accumulate multiple model-IDs along with their corresponding NW-additional conditions and conditions. The extent of information available to gNBs may vary. As referenced 17, the gNB may operate using model-ID-LCM approach towards the UE based on the information that the gNB has. This may be as previously described. Reference is made to Figure 6 which shows a method of some embodiments. This method may be performed by an apparatus. The apparatus may a user equipment. The apparatus may comprise suitable means, such as circuitry for providing the method. Alternatively or additionally, the apparatus may comprise at least one processor and at least one memory storing instructions that, when executed by the at least one processor cause the apparatus at least to provide the method below. Alternatively or additionally, the apparatus may be such as discussed in relation to Figure 5. The method may be provided by computer program code or computer executable instructions. The method may comprise as referenced A1, receiving information from a base station defining a time period. The method may comprise as referenced A2, selecting one or more of the plurality of user equipment side models during the time period. The method may comprise as referenced A3, sending information to the base station indicating the selected one or more of the plurality of user equipment side models. It should be appreciated that the method outlined in Figure 6 may be modified to include any of the previously described features. Reference is made to Figure 7 which shows another method of some embodiments. This method may be performed by an apparatus. The apparatus may in a base station or be a base station. The apparatus may comprise suitable means, such as circuitry for providing the method. Alternatively or additionally, the apparatus may comprise at least one processor and at least one memory storing instructions that, when executed by the at least one processor cause the apparatus at least to provide the method below. Alternatively or additionally, the apparatus may be such as discussed in relation to Figure 4. The method may be provided by computer program code or computer executable instructions. The method may comprise as referenced B1, sending, to an user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected. The method may comprise as referenced B2, receiving, from the user equipment, information indicating the selected one or more of the plurality of user equipment side models. It should be appreciated that the method outlined in Figure 7 may be modified to include any of the previously described features. Reference is made to Figure 8 which shows a method of some embodiments. This method may be performed by an apparatus. The apparatus may a user equipment. The apparatus may comprise suitable means, such as circuitry for providing the method. Alternatively or additionally, the apparatus may comprise at least one processor and at least one memory storing instructions that, when executed by the at least one processor cause the apparatus at least to provide the method below. Alternatively or additionally, the apparatus may be such as discussed in relation to Figure 5. The method may be provided by computer program code or computer executable instructions. The method may comprise as referenced C1, receiving information, from a base station, about a time duration for machine learning model identification. The method may comprise as referenced C2, performing, in the time duration an identification of one or more user equipment side machine learning models from a plurality of user equipment side machine learning models. It should be appreciated that the method outlined in Figure 8 may be modified to include any of the previously described features. Reference is made to Figure 9 which shows another method of some embodiments. This method may be performed by an apparatus. The apparatus may in a base station or be a base station. The apparatus may comprise suitable means, such as circuitry for providing the method. Alternatively or additionally, the apparatus may comprise at least one processor and at least one memory storing instructions that, when executed by the at least one processor cause the apparatus at least to provide the method below. Alternatively or additionally, the apparatus may be such as discussed in relation to Figure 4. The method may be provided by computer program code or computer executable instructions. The method may comprise as referenced D1, sending information, to a user equipment, about a time duration for machine learning model identification. The method may comprise as referenced D2, receiving one or more reports, from the user equipment, about the identification of one or more user equipment side machine learning models from a plurality of user equipment side machine learning models. It should be appreciated that the method outlined in Figure 9 may be modified to include any of the previously described features. According to the model identification definition, the process of model identification pinpoints how the NW and UE have a common understanding of “conditions” and "additional conditions" associated with an ML model. RAN1 agreed that the “conditions” associated with an ML model can be recognized by the functionality framework, where functionalities are configured based on the UE-reported conditions (conditions are reported in the UE-capability report). On the other hand, there are no exact agreements on additional conditions. However, several discussions took place on at least the following additional conditions. Example for Additional conditions: o Training dataset information o Site-related information (e.g., scenario, location / TRP / area information, beam direction / codebook information) o Time-related / Timestamp information o gNB implementation information (explicit or implicit details for specific gNB implementation details) o UE implementation information (explicit or implicit details for specific UE implementation details) o Statistical information (like delay spread, angular spread, and LOS / NLOS data) o Speed and range of speed information For a UE-sided model (also for a two-sided model), the model training may be associated with a set of assumptions (e.g., practical deployment assumptions) considered for additional conditions listed above, thus a trained UE-sided model can be associated with certain assumptions considered for additional conditions. It is feasible to categorize additional conditions as NW-sided additional conditions (assumptions for the NW-side), UE-sided additional conditions (assumptions on the UE-side), and common additional conditions (assumptions that are generic for both UE and NW-side). Via model identification, NW and UE shall gain a common understanding of the model-related additional conditions, and certain steps may be required to capture the over-the-air signaling linked with online model identification. However, by looking at the list of example additional conditions above, it should be clear that accurately describing these additional conditions (for online model identification) in the specifications will be challenging. Observation: Additional conditions associated with a Model can be categorized as NW-sided additional conditions, UE-sided additional conditions, and common additional conditions. o NW-sided additional conditions may refer to the NW assumptions that the ML model assumed when training the model (or generating the training dataset). These assumptions may not be known at the UE in the inference stage. o UE-sided additional conditions may refer to the UE assumptions that the ML model assumed when training the model (or generating the training dataset). These assumptions may not be known at the gNB in the inference stage. o Common additional conditions may refer to the common assumption that the ML model assumed when training the model (or generating the training dataset). These assumptions may be known at the gNB and UE in the inference stage. Observation: Specifying details on additional conditions can be a challenging task. Observation: For model identification, i.e., for identifying additional conditions for an ML model, it may be easier to consider implicit mutual understanding on associated NW-sided and UE-sided additional conditions than defining explicit details for additional conditions. As mentioned in the earlier section, the NW shall know whether the UE can support online model identification Types (that means identifying additional conditions associated with a Model), and it is expected that the UE may indicate if it supports online model identification and conditions for the online model identification. Here, conditions for online model identification may include a preferred minimum time duration that UE wishes to be configured with to support model identification. Also, conditions may further include required assistance on DL RS transmission, associated functionalities, or conditions for model identification. The UE shall receive a time duration for model identification from the gNB to identify the UE-sided model(s). These time durations for model identification could be sent by the gNB based on configuration such as time duration and sending indications on when the start and stop online model identification. The purpose of the time duration for model identification is to make sure that the UE can assume that the NW-sided additional conditions remain stable during the time duration for model identification, and the UE may be able to select UE-model(s) that suit the observed NW-additional conditions on the UE side during this time duration. It should be feasible to configure, details of cells / TRPs / PCIs / Area information where the model identification time duration is applicable. The UE can assume that the NW-additional conditions for those cells / TRPs / PCIs / Area remain stable during the time duration for model identification, and the UE may be able to select UE-model(s) that suit the observed NW-additional conditions for those cells / TRPs / PCIs / Area. In some embodiments, for online identification of UE-sided models, the gNB can configure or indicate a time duration that UE can assume to perform model identification. • Further information on cells / TRPs / PCIs / Area information where the model identification time duration is applicable can also be provided to the UE. It should be noted that, from the gNB's perspective, cells / TRPs / PCIs / Areas that are associated with this model identification time duration or indications are supposed to keep the "NW-sided additional conditions" stable for those cells / sites / area / TRPs. Also, as functionality identification and functionality LCM work as baseline operations for ML features, the model identification process may be linked to functionality or a set of functionalities associated with the ML model. With that, the associated conditions for the online model identification can be known. As the NW deployments can be different from one another, the NW may also have other considerations that they shall assume when deciding NW-sided additional conditions. In some sense, more freedom in configuring different time durations may be needed. In that perspective, compared to configuring fixed time durations via RRC, it may be feasible to send dynamic signaling to control the model identification time durations via DCI and MAC-CE signaling. In some embodiments, the indication(s) of initializing the model identification and terminating the model identification may be sent via DCI or MAC-CE messages, where the UE may have variable sizes of time durations for model identifications depending on when NW wishes to keep static NW-additional conditions or not. Within the time duration for model identification defined by the NW, the UE shall determine, monitor, and assess the applicability of the UE-sided ML model(s) for a given scenario, cell, or site-specific setting that the gNB plans to use for the ML feature. In one option, the Model assessment and monitoring may be performed by the UE while supporting an ML functionality. The UE may be configured or indicated to operate with this functionality during this time duration, allowing it to assess the performance of different UE-sided AI / ML models while supporting the functionality. In another option, the Model assessment and monitoring may also be performed by the UE without supporting an ML functionality. During this time duration, the UE is not expected to operate with the functionality but can still assess the performance of different UE-sided AI / ML models. In some embodiments, during the model identification time duration, the UE shall be able to assess / monitor different ML models to select / identify a suitable model for the NW-sided additional conditions. Consider the following options depending on UE capabilities, • Option 1: parallelly supporting the ML feature based on a functionality framework. • Option 2: ML feature is not supported during the model identification time duration. Also, it may be feasible to assume that model assessment and monitoring may be conducted by the UE with or without additional support from the NW. In one example, for BM-casel, the UE may have N number of ML models (logical models) which are developed assuming different NW / UE / common-additional conditions. In other words, N different ML models may be associated with N different assumptions on additional conditions. However, all these N models may be related to the same functionality (same RRC configuration that gNB configures). The UE may monitor channel characteristics (propagation environment), beam measurements (L1-RSRP, L1-SINR, beam arrival directions, etc.....), and other aspects based on the DL RSs / PDSCH / PDCCH that UE receives from the gNB. When a good level of monitoring is available at the UE (within the model identification time duration), the UE may be able to map these observations to NW-additional conditions. As the UE knows which NW-additional conditions it assumes for a given ML model, the UE can select an ML model by comparing the observed NW-additional conditions to the additional conditions associated with an ML model. In another option, the gNB may provide monitoring resources and required assistance for the UE's model monitoring procedure (somewhat visible to the gNB) and may identify at least one model that can be applied to the scenario, cell, or site-specific setting that the gNB plans to use for the ML feature. For the same example assumed before, the UE may get dedicated DL RS and assistance information to monitor background ML models such that UE can evaluate the performance of ML models with the support of the NW. When the UE has a monitoring outcome, the UE can identify which one of these N models is suited for considered NW-setting in the model identification time duration. In some embodiments, to enable assessment / monitoring of different ML models to select / identify a suitable model for the NW-sided additional conditions, the NW may provide further assistance by transmitting required monitoring resources (DL RSs) and sending assistance info. During or after the time duration for model identification, the UE may report an indicator to report whether a model (new or old) suits the NW-additional conditions considered during the time duration for model identification. The same indicator can also report the absence of a model that suits the NW-additional conditions considered during the time duration for model identification. This indicator may be conveyed using the same field (e.g., with a single bit in reporting). In some embodiments, after or during the model identification time duration, the UE may report an indicator highlighting whether there is a new model identification or not for the associated model identification time duration. Additionally, the UE can also report another indicator, mainly reporting a model ID, that is being identified by the UE (a new Model) or report a matching model ID that was identified before with a model ID. A new model applicable for the NW-additional conditions during the model identification time duration can be reported by the UE. Here, the UE is expected to indicate a model ID from the pre-defined / configured bit field in an incremental order of assigning new model IDs. When an older model is applicable for the NW-additional conditions during the model identification time duration, the UE can report the already identified model ID. When a new model is applicable for the NW-additional conditions during the model identification time duration, the UE can report a new model ID. In this case, the UE is expected to indicate a model ID from the pre-defined / configured bit field in an incremental order of assigning new model IDs. When the pre-defined / configured bit field used for the second indicator is full, the UE may replace an existing model ID (from the already identified list) and report the new model that suits the NW-additional conditions used during the time duration for model identification. The UE report can be an NW-initiated report (e.g., when the model identification time duration is over, there can be a gNB-initiated scheduled report) or a UE-initiated report (e.g., the UE triggers a MAC-CE message to report the first and second indicators). In some embodiments, after or during the model identification time duration, the UE may report a second indicator identifying a model ID for the model identification time duration. After receiving the UE report(s), the gNB determines whether there is any new model identification or not. If there is new model identification with a new model ID or with an older model ID, the gNB may associate the model ID with the set of assumptions used by the cells / TRPs / PCIs / Area and time duration information. For example, based on the information that gNB has, it can associate Model-ID_x to a set of “conditions” and a set of “additional conditions”, where additional conditions are Time information_K, Cell_L, Beam codebook_M, Antenna Config_N, TRP_implementation_P, etc. From the NW perspective, Model-ID identified by the above can only be used to identify the NW or common-additional conditions (or additional conditions in DL direction) of an ML model in implicit means. The model ID is mainly considered by gNB and UE for signaling purposes to align with additional conditions that NW plans to use the ML feature. It does not always select an exact physical model at the UE. In general, if the UE can maintain many ML models for the same NW / common additional conditions, the model ID reported by the UE can refer to one or more ML models on the UE side. If there is more than one ML model associated with a model ID, there may be other additional selection methods used by the UE to select the ML model out of multiple ML models depending on the used UE-sided additional conditions considered at a given time. In some embodiments, for model-lD-based LCM or functionality-LCM that uses model-ID, the identified model-lD can be reused by the NW and UE for signaling purposes, where model-l D refers to the additional conditions assumed during the corresponding model identification time duration. Some embodiments may be provided in a 5G scenario. A 5th generation communication system (5GS), an access network and a core network (5GC) thereof, and communication devices are briefly explained with reference to Figures 3, 4 and 5. Figure 3 shows a schematic representation of a 5G communication system (5GS). The 5GS may comprise a user equipment (UE) 300, an access network such as a 5G radio access network (5G-RAN) or next generation radio access network (NG-RAN), a 5G core network (5GC), and one or more application functions. An application function may be deployed in the 5GS as trusted application function or may be deployed or host on one or more application servers of the data network. Such application functions are untrusted application functions. The 5GS connects the UE to a data network the access network and the 5GC (e.g., a UPF of the 5GC). The 5G-RAN may comprise one or more radio access nodes, such as gNodeB (gNB) 100. A gNB may include one or more gNodeB (gNB) distributed units connected to one or more gNodeB (gNB) centralized units . The 5GC may comprise the following network functions: Network Slice Selection Function (NSSF); Network Exposure Function; Network Repository Function (NRF); Policy Control Function (PCF); Unified Data Management (UDM); Application Function (AF); Authentication Server Function (AUSF); an Access and Mobility Management Function (AMF); and Session Management Function (SMF), and a user plane function (UPF). Figure 3 also shows the various interfaces (N1, N2 etc.) that may be implemented between the various elements of the system. Figure 4 illustrates an example of an apparatus 200. The apparatus may be a gNB or part of gNB. The apparatus 200 may comprise at least one random access memory (RAM) 211a, at least on read only memory (ROM) 211b, at least one processor 212, 213 and an network interface 214. The at least one processor 212, 213 may be coupled to the RAM 211a and the ROM 211b. The at least one processor 212, 213 may be configured to execute an appropriate software code 215. Execution of the software code 215 may for example may cause the apparatus to perform operations for controlling the gNB. The software code 215 may be stored in the ROM 211b. Figure 5 illustrates an example of a communication device 300, such as the UE illustrated on Figure 1,2 and 3. The communication device 300 may be provided by any device capable of sending and receiving radio signals. Non-limiting examples of a communication device 300 comprise a user equipment, a mobile station (MS) or mobile device such as a mobile phone or what is known as a ’smart phone’, a computer provided with a wireless interface card or other wireless interface facility (e.g., USB dongle), a personal data assistant (PDA) or a tablet provided with wireless communication capabilities, a machine-type communications (MTC) device, an Internet of things (loT) type communication device or any combinations of these or the like. The communication device 300 may comprise a transceiver for transmitting and / or receiving, for example, wireless signals carrying communications, for example radio signals. The communications may be one or more of voice, electronic mail (email), text messages, multimedia data, machine data and so on. The communication device 300 may receive wireless signals (e.g., radio signals) over an air or radio interface 307 via appropriate apparatus for receiving and may transmit wireless signals via appropriate apparatus for transmitting radio signals. In Figure 5 transceiver is designated schematically by block 306. The transceiver 306 may comprise, for example, a radio part and associated antenna arrangement. The antenna arrangement may be arranged internally or externally to the mobile device and may comprise one or more antenna elements. The antenna arrangement may be a multi-input multi output (MIMO) antenna. The communication device 300 may be provided with at least one processor 301, at least one memory ROM 302a, at least one RAM 302b and other possible components 303 for use in software and hardware aided execution of tasks it is designed to perform, including control of access to and communications with access networks (e.g., the 5G-RAN or NG-RAN illustrated in Figure 3) and other communication devices. The at least one processor 301 is coupled to the RAM 302b and the ROM 302a. The at least one processor 301 may be configured to execute an appropriate software code 308. The software code 308 may for example allow to perform one or more operations of the communication device. The software code 308 may be stored in the ROM 302a. The processor, the ROM, and the RAM, the transceiver and other circuitry of the communication device (e.g., a modem) can be provided on a circuit board, in chipsets, or in a system on chip. The circuit board, chipsets or system on chip is denoted by reference 304. The communication device 300 may optionally have a user interface such as key pad 305, touch sensitive screen or pad, combinations thereof or the like. Optionally one or more of a display, a speaker and a microphone may be provided depending on the type of communication device. It is noted that whilst some embodiments have been described in relation to 5G networks, similar principles can be applied in relation to other networks and communication systems. Therefore, although certain embodiments were described above by way of example with reference to certain example architectures for wireless networks, technologies and standards, embodiments may be applied to any other suitable forms of communication systems than those illustrated and described herein. It is also noted herein that while the above describes example embodiments, there are several variations and modifications which may be made to the disclosed solution without departing from the scope of the present invention. As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements. In general, the various embodiments may be implemented in hardware or special purpose circuitry, software, logic or any combination thereof. Some aspects of the disclosure may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device, although the disclosure is not limited thereto. While various aspects of the disclosure may be illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that these blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof. As used herein, the term “circuitry” may refer to one or more or all of the following: (a) hardware-only circuit implementations (such as implementations in only analog and / or digital circuitry) and (b) combinations of hardware circuits and software, such as (as applicable): (i) a combination of analog and / or digital hardware circuit(s) with software / firmware and (ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a mobile phone or server, to perform various functions) and (c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation.” This definition of circuitry applies to all uses of this term herein, including in any claims. As a further example, as used herein, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device. The embodiments of this disclosure may be implemented by computer software executable by a data processor of the mobile device, such as in the processor entity, or by hardware, or by a combination of software and hardware. Computer software or program, also called program product, including software routines, applets and / or macros, may be stored in any apparatus-readable data storage medium and they comprise program instructions to perform particular tasks. A computer program product may comprise one or more computerexecutable components which, when the program is run, are configured to carry out embodiments. The one or more computer-executable components may be at least one software code or portions of it. Further in this regard it should be noted that any blocks of the logic flow as in the Figures may represent program steps, or interconnected logic circuits, blocks and functions, or a combination of program steps and logic circuits, blocks and functions. The software may be stored on such physical media as memory chips, or memory blocks implemented within the processor, magnetic media such as hard disk or floppy disks, and optical media such as for example DVD and the data variants thereof, CD. The physical media is a non-transitory media. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM). The memory may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as semiconductor based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory. The data processors may be of any type suitable to the local technical environment, and may comprise one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASIC), FPGA, gate level circuits and processors based on multi core processor architecture, as non-limiting examples. Various example embodiments of the disclosure may be practiced in various components such as integrated circuit modules. The design of integrated circuits is by and large a highly automated process. Complex and powerful software tools are available for converting a logic level design into a semiconductor circuit design ready to be etched and formed on a semiconductor substrate. The scope of protection sought for various example embodiments of the disclosure is set out by the independent claims. The example embodiments and features thereof, if any, described in this disclosure that do not fall under the scope of the independent claims are to be interpreted as examples useful for understanding various example embodiments of the 5 disclosure. The foregoing description has provided, by way of non-limiting and illustrative examples, a full and informative description of the various example embodiments of this disclosure. However, various modifications and adaptations may become apparent to those skilled in the relevant arts in view of the foregoing description, when read in conjunction with 10 the accompanying drawings and the claims. However, all such and similar modifications of the teachings will still fall within the various example embodiments of the disclosure as set forth in the claims. By way of non-limiting and illustrative example, there is a further example embodiment comprising a combination of one or more example embodiments with any of the other example embodiments previously discussed.

Claims

1. An apparatus comprising:means for receiving information defining a time period from a base station;means for selecting one or more of a plurality of user equipment side models during the time period; andmeans for sending information to the base station indicating the selected one or more of the plurality of user equipment side models.

2. The apparatus as claimed in claim 1, wherein one or more network related conditions are stable during said time period.

3. The apparatus of claim 1 or 2, wherein the means for selecting is for selecting the one or more of the plurality of user equipment side models based on the one or more network related conditions.

4. The apparatus of claim 1, 2 or 3, wherein the means for receiving information is for receiving information defining a time duration in which the one or more of the plurality of user equipment side models is to be selected.

5. The apparatus as claimed in any preceding claim, wherein the means for receiving information is for receiving a first message from the base station which starts the time period and / or receiving a second message from the base station which ends the time period.

6. The apparatus as claimed in any preceding claim, comprising means for sending to the base station information indicating a capability of the user equipment for supporting user equipment side model selection.

7. The apparatus as claimed in claim 6, comprising the means for sending is for sending to the base station information indicating one or more capability conditions associated with the capability of the user equipment for supporting user equipment side model selection.

8. The apparatus as claimed in claim 7, wherein the one or more capability conditions associated with the capability of the user equipment comprise one or more of:a minimum length of the time period; required assistance from the base station; and / or one or more machine learning enabled features.

9. The apparatus as claimed in any preceding claim, wherein the means for receiving is for receiving from the base station network coverage area information associated with the time period.

10. The apparatus as claimed in claim 9, wherein the network coverage area information comprises one or more of:one or more cells; one or more transmission reception points; one or more physical cell identifiers; and / or area information.

11. The apparatus as claimed in any preceding claim, wherein the means for selecting of the one or more user equipment side models is for assessing a performance of one or more user equipment side models and selecting based on the performance of one or more of the one or more user equipment side models.

12. The apparatus as claimed in any preceding claim, wherein the means for receiving is for receiving assistance information from the base station and using the assistance information when selecting one or more of the one or more user equipment side models.

13. The apparatus as claimed in claim 12, wherein the assistance information comprises one or more of: one or more reference signals; and information about one or more of the network related conditions.

14. The apparatus as claimed in any preceding claim, comprising means for determining one or more of the network related conditions.

15. The apparatus as claimed in any preceding claims, wherein the means for selecting one or more of the plurality of user equipment side models during the time period is for comparing one or more network conditions associated with a respective user equipment side model and one or more current network conditions.

16. The apparatus as claimed in any preceding claim, comprising means for assigning an identity to a respective one of the selected one or more user equipment side models and sending the assigned identity to the base station.

17. The apparatus as claimed in any preceding claim, wherein the means for sending is for sending a first indicator to the base station indicating if there are one or more models appropriate for the one or more network conditions and / or for sending a second indicator tothe base station providing an identifier associated with a respective one of the one or more user equipment side models.

18. The apparatus as claimed in claim 17, wherein the means for sending is for sending the associated time period and the network coverage area information with the second indicator to the base station.

19. The apparatus as claimed in any preceding claim, wherein the means for receiving is for subsequently receiving from the base station one or more commands for the selected one or more of the plurality of user equipment side models20. An apparatus, comprising:means for sending, to a user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected; andmeans for receiving, from the user equipment, information indicating the selected one or more of the plurality of user equipment side models.

21. The apparatus as claimed in claim 20, further comprising means for keeping one or more network related conditions stable during said time period.

22. The apparatus as claimed in claim 20 or 21, wherein the means for receiving is for receiving from the user equipment a first indicator indicating if there are one or more models appropriate for the one or more network conditions and / or for receiving a second indicator comprising an identifier associated with a respective one of the one or more user equipment side models.

23. The apparatus as claimed in claim 22, wherein the means for receiving is for receiving the associated time period and network coverage area information with the second indicator from the user equipment.

24. A method, comprising:receiving information from a base station defining a time period;selecting one or more of a plurality of user equipment side models during the time period; andsending information to the base station indicating the selected one or more of the plurality of user equipment side models.

25. A method, comprising:sending, to an user equipment, information defining a time period in which one or more of a plurality of user equipment side models is to be selected; andreceiving, from the user equipment, information indicating the selected one or more of5 the plurality of user equipment side models.

Images