Methods and devices for reporting inference data generated by an inference model

By applying an inference data reporting criterion at the UE to filter out unreliable data, the patent addresses inefficiencies in wireless communication systems, ensuring only accurate data is transmitted to the RAN, thus optimizing network resource usage.

WO2026238797A1PCT designated stage Publication Date: 2026-11-19CONTINENTAL AUTOMOTIVE TECHNOLOGIES GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2026/063903
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-16
Filing Date
2026-05-13
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Existing wireless communication systems face inefficiencies and resource wastage due to inaccurate inference data reporting by user equipment (UE) executing AI/ML models, which can lead to inappropriate network actions.

Method used

Implementing an inference data reporting criterion at the UE to determine if the data is worth reporting to the radio access network (RAN), based on criteria such as confidence score, input data freshness, and model expiration, and skipping transmission if the criterion is not met.

Benefits of technology

Reduces communication resource usage by ensuring only reliable and relevant inference data is reported, thereby optimizing network operations and resource allocation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2026063903_19112026_PF_FP_ABST
    Figure EP2026063903_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The present disclosure relates to a method (40) for communicating in a wireless communication system, the method being implemented by a user equipment, UE, of the wireless communication system, wherein the UE comprises a communication unit configured to exchange data with a radio access network, RAN, of the wireless communication system, wherein the exchanged data comprises data related to an inference model to be executed by the UE to produce inference data to be reported to the RAN, the method comprises: - (S40) executing the inference model to produce inference data based on input data, - (S41) evaluating an inference data reporting criterion, - in response to the inference data reporting criterion not being verified: (S42) skipping transmitting the inference data to the RAN.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] 202502839

[0002] 1

[0003] DESCRIPTION TITLE: Methods and devices for reporting inference data generated by an inference model

[0004] TECHNICAL FIELD

[0005] The present disclosure relates to wireless communication systems and relates more specifically to methods and devices for enabling efficient usage of artificial intelligence (Al) / machine learning (ML) techniques in wireless communication systems.

[0006] BACKGROUND

[0007] Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, broadcasts, or other similar types of services. Although wireless communication systems have made great technological advancements over many years, challenges still exist, and there still exists a need for further improvements in wireless communications systems.

[0008] For example, 3GPP (Third Generation Partnership Project) is currently investigating AI / ML techniques in order to improve the performance of the air interface, among others.

[0009] More specifically, 3GPP is currently investigating several use cases in which AI / ML techniques are expected to improve the performance of the air interface. For example, the technical report TR 38.843 V18.0.0 discusses the following use cases:

[0010] channel state information (CSI) feedback enhancement: for example, spatial-frequency domain CSI compression using two-sided Al model (or possibly only a UE-sided Al model), time domain CSI prediction using UE-sided model, etc.,

[0011] beam management: for example, spatial-domain downlink (DL) beam prediction for Set A of beams based on measurement results of Set B of beams, temporal DL beam prediction for Set A of beams based on the historic measurement results of Set B of beams, etc.,

[0012] positioning accuracy enhancements: for example, direct AI / ML positioning, AI / ML assisted positioning, etc.

[0013] In some cases, a UE-sided AI / ML model may be used. When a UE performs model inference (i.e. , it executes the UE-sided AI / ML model), it can report the inference data to the network. This process is part of the broader framework for AI / ML model lifecycle management, which includes model training, deployment, inference, monitoring, and updating. In the technical report TR 38.843 V18.0.0, the signaling procedures for model inference reporting involve several key phases:

[0014] model inference initiation: the UE initiates the model inference process based on predefined triggers or conditions,

[0015] inference execution: the UE performs the inference using the deployed AI / ML model, reporting results: the UE reports the inference data to the network; this can include202502839

[0016] 2

[0017] various metrics and outcomes derived from the AI / ML model,

[0018] network actions: based on the reported inference data, the network may take specific actions, such as adjusting resource allocations or optimizing network parameters. However, AI / ML models for a given use case may be tailored towards and applicable to specific scenarios, locations, configurations, deployments, among other factors. In this regard, it is acknowledged that the execution of a given AI / ML model may fail to produce accurate inference data, if performed in a context for which the AI / ML model is not applicable. In such a case, the network may receive inaccurate inference data, based on which it might trigger inappropriate actions that will further waste resources of the wireless communication system.

[0019] SUMMARY

[0020] The present disclosure aims at improving the situation. In particular, the present disclosure aims to address at least some of the limitations of the prior art discussed above. In particular, the present disclosure aims to propose a solution for reducing the amount of communication resources used for reporting inference data, in particular for reporting inference data that may be inaccurate.

[0021] For that purpose, it is proposed to report inference data, obtained by executing the inference model, only when an inference data reporting criterion is verified. Such an inference data reporting criterion aims at evaluating whether the inference data obtained is worth reporting to the radio access network (RAN). For example, the inference data may not be worth reporting to the RAN if e.g. a confidence score associated with the inference data is too low and / or if the amount of (fresh) input data processed by the inference model is too low and / or if the inference model has expired, etc.

[0022] According to a first aspect, the present disclosure relates to a method for communicating in a wireless communication system, the method being implemented by a user equipment, UE, of the wireless communication system, wherein the UE comprises a communication unit configured to exchange data with a radio access network, RAN, of the wireless communication system, wherein the exchanged data comprises data related to an inference model to be executed by the UE to produce inference data to be reported to the RAN, wherein the method comprises:

[0023] executing the inference model to produce inference data based on input data, evaluating an inference data reporting criterion,

[0024] in response to the inference data reporting criterion not being verified: skipping transmitting the inference data to the RAN.

[0025] In some embodiments, the method according to the first aspect can further comprise one or more of the following optional features, considered either alone or in any technically possible combination.

[0026] In some embodiments of the method according to the first aspect, evaluating the202502839

[0027] 3

[0028] inference data reporting criterion comprises evaluating whether a confidence score criterion is verified, based on a confidence score associated with the inference data.

[0029] In some embodiments of the method according to the first aspect, evaluating the confidence score criterion comprises comparing the confidence score to a confidence score threshold, wherein the confidence score criterion is verified if the confidence score is greater than the confidence score threshold.

[0030] In some embodiments of the method according to the first aspect, evaluating the inference data reporting criterion comprises evaluating whether an input data criterion is verified, based on the input data processed by the inference model.

[0031] In some embodiments of the method according to the first aspect, evaluating the input data criterion comprises comparing the amount of input data processed to an input data threshold, wherein the input data criterion is verified if the amount of input data processed is greater than the input data threshold.

[0032] In some embodiments of the method according to the first aspect, evaluating the inference data reporting criterion comprises evaluating whether a completion time criterion is verified, based on a completion time of the execution of the inference model.

[0033] In some embodiments of the method according to the first aspect, evaluating the completion time criterion comprises comparing the completion time of the execution of the inference model to a completion time threshold, wherein the completion time criterion is verified if the completion time of the execution of the inference model is lower than the completion time threshold.

[0034] In some embodiments, the method according to the first aspect comprises: receiving, from the RAN, configuration data for configuring the inference data reporting criterion,

[0035] configuring the inference data reporting criterion based on the received configuration data.

[0036] In some embodiments of the method according to the first aspect, the configuration data includes one or more thresholds.

[0037] In some embodiments of the method according to the first aspect, the configuration data is received by the UE in a signaling message addressed specifically to said UE or to a group of UEs which includes said UE, and / or in system information broadcasted by the RAN.

[0038] In some embodiments, the method according to the first aspect comprises transmitting the inference data to the RAN in response to the inference data reporting criterion being verified.

[0039] In some embodiments of the method according to the first aspect, the inference model is used for at least one among:

[0040] channel state information, CSI, compression,

[0041] CSI prediction,202502839

[0042] 4

[0043] downlink, DL, beam prediction,

[0044] UE positioning.

[0045] According to a second aspect, the present disclosure relates to a user equipment comprising at least one memory and at least one processor configured to carry out a method according to any one of the embodiments of the first aspect.

[0046] According to a third aspect, the present disclosure relates to a method for communicating in a wireless communication system, the method being implemented by a base station, BS, of a radio access network, RAN, of the wireless communication system, wherein the BS is configured to exchange data with a user equipment, UE, wherein the exchanged data comprises data related to an inference model to be executed by the UE to produce inference data to be reported to the BS, the method comprises:

[0047] transmitting, to the UE, configuration data for configuring an inference data reporting criterion at said UE,

[0048] receiving inference data from the UE.

[0049] In some embodiments, the method according to the third aspect can further comprise one or more of the following optional features, considered either alone or in any technically possible combination.

[0050] In some embodiments of the method according to the third aspect, the configuration data includes one or more thresholds.

[0051] In some embodiments of the method according to the third aspect, the one or more thresholds include at least one among:

[0052] a confidence score threshold,

[0053] an input data threshold,

[0054] a completion time threshold.

[0055] In some embodiments of the method according to the third aspect, the configuration data is transmitted to the UE in a signaling message addressed specifically to said UE or to a group of UEs which includes said UE, and / or in system information broadcasted by the BS.

[0056] According to a fourth aspect, the present disclosure relates to a base station comprising at least one memory and at least one processor configured to carry out a method according to any one of the embodiments of the third aspect.

[0057] According to a fifth aspect, the present disclosure relates to a wireless communication system comprising at least one base station according to any one of the embodiments of the present disclosure and at least one user equipment according to any one of the embodiments of the present disclosure.

[0058] According to a sixth aspect, the present disclosure relates to a computer program product comprising instructions which, when executed by at least one processor, configure said at least one processor to carry out a method for communicating according to any one of the202502839

[0059] 5

[0060] embodiments of the present disclosure. The computer program product can use any programming language, and can be in the form of source code, object code, or in any intermediate form between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0061] According to a seventh aspect, the present disclosure relates to a (non-transitory) computer-readable storage medium comprising instructions which, when executed by at least one processor, configure said at least one processor to carry out a method for communicating according to any one of the embodiments of the present disclosure.

[0062] BRIEF DESCRIPTION OF FIGURES

[0063] The invention will be better understood upon reading the following description, given as an example that is in no way limiting, and made in reference to the figures which show:

[0064] Figure 1: a schematic representation of a wireless communication system,

[0065] Figure 2: a schematic representation of an example of a UE,

[0066] Figure 3: a schematic representation of an example of a BS,

[0067] Figures 4 and 5: flow charts illustrating examples of methods for communicating implemented by a UE and a BS, respectively.

[0068] In these figures, references identical from one figure to another designate identical or analogous elements. For reasons of clarity, the elements shown are not to scale, unless explicitly stated otherwise.

[0069] Also, the order of steps or phases of any methods disclosed herein, in particular in the figures, is provided for illustration purposes only and is not meant to limit the present disclosure which may be applied with the same steps or phases executed in a different order and / or with all or part of the steps or phases executed in parallel and / or jointly.

[0070] DETAILED DESCRIPTION

[0071] The detailed description set forth below, with reference to the figures, is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. For instance, although 3GPP terminology, from e.g., 5G NR, may be used in this disclosure to exemplify embodiments herein, this should not be seen as limiting the scope of the present disclosure. In particular, the present disclosure can be applied to future wireless communication systems such as 6G wireless communication systems.

[0072] Generally, all terms used herein are to be interpreted according to their ordinary meaning in the relevant technical field, unless a different meaning is clearly given and / or is implied from the context in which it is used. All references to a / an / the element, apparatus, component, means, step, etc. are to be interpreted openly as referring to at least one instance202502839

[0073] 6

[0074] of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. Also, the order of steps of any methods disclosed herein, in particular in the figures, is provided only for illustration purposes and is not meant to limit the present disclosure which may be applied with the same steps executed in a different order and / or with all or part of the steps executed in parallel or jointly, unless a step is explicitly described as necessarily following or preceding another step and / or where it is implicit that a step must follow or precede another step. Also, in a figure, steps represented surrounded by a dashed line are to be considered as optional for the embodiment represented in this figure. Any feature of any of the embodiments disclosed herein may be applied to any other embodiment, wherever appropriate. Likewise, any advantage of any of the embodiments may apply to any other embodiments, and vice versa.

[0075] Figure 1 represents schematically an example of wireless communication system, which may be for example a 5G NR wireless communication system. More specifically, figure 1 represents a RAN of the wireless communication system, which is used to exchange data with UEs 20 via radio signals. For example, the RAN may send data to the UEs 20 (downlink, DL), for instance data received from a core network (CN, not represented in the figures). The RAN may also receive data from the UEs 20 (uplink, UL), which data may be forwarded to the CN.

[0076] In the example illustrated by figure 1, the RAN comprises one base station, BS, 30. Of course, the RAN may comprise more than one BS 30 to increase the radio coverage of the wireless communication system. Each of these BSs may be referred to as NB, eNodeB (or eNB), gNodeB (or gNB, in the case of a 5G NR wireless communication system), an access point or the like, depending on the wireless communication standard(s) implemented.

[0077] Figure 2 represents schematically an example of a wireless device 25 suitable for implementing any method, discussed in the present disclosure, performed at a UE 20. Basically, the wireless device 25 corresponds to an apparatus that provides wireless connectivity with the RAN of the wireless communication system, and that can be used to exchange data with said RAN. Such a wireless device 25 may be included in a UE 20, as illustrated by figure 2. The UE 20 may for instance be a cellular phone, a wireless modem, a handheld device, a laptop computer, a headset, smart glasses, a vehicle (manned or unmanned), etc., or any other equipment that may execute applications that need to exchange data with remote recipients, via the wireless device 25.

[0078] As illustrated by figure 2, the wireless device 25 comprises one or more processors 250 and one or more memories 251. The one or more processors 250 may include for instance a central processing unit (CPU), a digital signal processor (DSP), a graphics processing unit (GPU), a neural processing unit (NPU), a field-programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc. The one or more memories 251 may include any type of computer readable volatile and non-volatile memories (magnetic hard disk, solid-state disk, optical disk, electronic memory, etc.). The one or more memories 251 may store a computer202502839

[0079] 7

[0080] program product 252, in the form of a set of program-code instructions to be executed by the one or more processors 250 to implement all or part of the steps of a method 40 for communicating, performed at a UE 20, according to any one of the embodiments disclosed herein.

[0081] As illustrated by figure 2, the wireless device 25 comprises also a wireless communication unit 253 adapted to exchange data with BSs 30 of the RAN using radio signals. The wireless communication unit 253 may implement one or more wireless communication protocols, and may for instance be a 3G, 4G, 5G, 6G, WiFi, WiMax, etc. transceiver or the like. In some examples, the wireless communication unit 253 comprises a 5G NR wireless communication unit.

[0082] Figure 3 represents schematically an example of a BS 30 suitable to implement any method, discussed in the present disclosure, performed by the RAN.

[0083] As illustrated by figure 3, the BS 30 comprises one or more processors 300 and one or more memories 301. The one or more processors 300 may include for instance a CPU, a DSP, a GPU, an NPU, an FPGA, an ASIC, etc. The one or more memories 301 may include any type of computer readable volatile and non-volatile memories (magnetic hard disk, solid-state disk, optical disk, electronic memory, etc.). The one or more memories 301 may store a computer program product 302, in the form of a set of program-code instructions to be executed by the one or more processors 300 to implement all or part of the steps of a method 50 for communicating, performed at the RAN’s side, according to any one of the embodiments disclosed herein.

[0084] As illustrated by figure 3, the BS 30 comprises also a wireless communication unit 303, configured to exchange data with UEs 20 using radio signals, and more specifically with wireless communication units 253 of wireless devices 25 included in these UEs 20. The wireless communication unit 303 may for instance be a 3G, 4G, 5G, 6G, WiFi, WiMax, etc. transceiver or the like. In some examples, the wireless communication unit 303 of the BS 30 comprises a 5G NR transceiver.

[0085] As illustrated by figure 3, the BS 30 may comprise also, in some examples, a network communication unit 304, configured to exchange data with other base stations of the RAN and / or with the CN. The network communication unit 304 may support one or more suitable communication protocols, which may be wired (including optical) and / or wireless.

[0086] As discussed above, the present disclosure aims at proposing a solution for reducing the amount of communication resources used for reporting inference data obtained by the UE 20 executing an inference model.

[0087] For that purpose, it is proposed to report inference data, obtained by the UE 20 executing the inference model, only when an inference data reporting criterion is verified. Such an inference data reporting criterion aims at evaluating whether the inference data obtained is202502839

[0088] 8

[0089] worth reporting to a BS 30 of the RAN. For example, the inference data may not be worth reporting to a BS 30 of the RAN if e.g. a confidence score associated with the inference data is too low and / or if the amount of (fresh) input data processed by the inference model is too low and / or if the inference model has expired, etc., as discussed below.

[0090] We now present non-limitative examples of methods for communicating in a wireless communication system.

[0091] Figure 4 represents a diagram showing the main steps of an exemplary embodiment of a method 40 for communicating, which is implemented by a UE 20. Figure 5 represents a diagram showing the main steps of an exemplary embodiment of a method 50 for communicating, which may be implemented by a BS 30 of the RAN when the UE 20 implements the method 40 for communicating illustrated by figure 4.

[0092] As illustrated by figure 4, the method 40 for communicating comprises a step S40 of executing an inference model to produce inference data based on input data.

[0093] It should be noted that the present disclosure may be applied to any type of inference model used at a UE 20 to perform inference (i.e. , to produce inference data that may be transmitted to the RAN), and the choice of a specific type of inference model corresponds to a specific but non-limitative embodiment of the present disclosure. In some embodiments, the inference model is a machine learning model, for example previously trained to perform a specific type of inference. Again, the present disclosure is not limited to a specific type of inference and may be applied to many different inference types. For example, the inference model may be used to perform inference for any one of the use cases discussed in the technical report TR 38.843 V18.0.0, e.g.:

[0094] CSI compression,

[0095] CSI prediction,

[0096] DL beam prediction,

[0097] UE positioning, etc.

[0098] As illustrated by figure 4, the method 40 for communicating comprises a step S41 of evaluating an inference data reporting criterion, after step S40. As discussed above, the inference data reporting criterion aims at evaluating whether the inference data obtained by the UE 20 is worth reporting to the RAN, and detailed examples are provided hereinbelow.

[0099] If the inference data reporting criterion is not verified (reference S41a in figure 4), the method 40 for communicating comprises a step S42 of skipping the transmission of the obtained inference data to the RAN. In other words, the obtained inference data is not transmitted to the RAN, thereby saving communication resources.

[0100] Conversely, if the inference data reporting criterion is verified (reference S41b in figure 4), the method 40 for communicating comprises a step S43 of transmitting the obtained inference data to the RAN.202502839

[0101] 9

[0102] The inference data reporting criterion may be predefined (e.g., specified in a standard, hard coded in a non-volatile memory of the UE 20, etc.) and / or preconfigured by using configuration data that may be provided via the RAN. In the non-limitative example illustrated by figure 4, the method 40 for communicating comprises an optional step S44 of receiving, from the RAN, configuration data which is used by the UE 20 for configuring the inference data reporting criterion. For example, if different types of criteria can be evaluated in the inference data reporting criterion (see examples below), the configuration data may indicate the type of criterion to be considered, or the types of criteria to be considered if multiple criteria are to be evaluated. Also, the evaluation of the inference data reporting criterion may rely on the comparison of one or more parameter values to respective thresholds. In such a case, the configuration data may include values for the one or more thresholds to be used.

[0103] It should be noted that different types of signaling messages may be considered for transmitting the configuration data to the UE 20, and the choice of a specific type corresponds to a specific but non-limitative embodiment of the present disclosure. For example, the configuration data is received by the UE 20 in a signaling message addressed specifically to said UE (e.g., in a radio resource control, RRC, reconfiguration message, etc.) or to a group of UEs which includes said UE 20, and / or in system information broadcasted by the RAN.

[0104] We now provide three non-limitative examples of different types of criteria that may be evaluated in the inference data reporting criterion, namely:

[0105] a confidence score criterion,

[0106] an input data criterion,

[0107] a completion time criterion.

[0108] It should be noted that the evaluation of the inference data reporting criterion may comprise the evaluation of a single type of criterion, or the cumulative evaluation of multiple types of criteria. If multiple types of criteria are evaluated, then the inference data reporting criterion may be considered verified if at least one evaluated criterion is verified, or if all the evaluated criteria are verified, etc.

[0109] In one example, evaluating whether the inference data reporting criterion is verified comprises evaluating whether a confidence score criterion is verified, based on a confidence score associated with the inference data. Indeed, when the inference model is executed, it is possible to determine a confidence score associated with the inference data produced by the inference model. Such a confidence score is representative of the reliability of the inference data, and may correspond e.g. to a probability that the inference data is accurate, an estimated error level affecting the inference data, etc. Different types of confidence scores may be used, and the choice of a specific type corresponds to a specific but non-limitative embodiment of the present disclosure. Confidence scores are commonly provided by inference machine learning models. For example, evaluating the confidence score criterion comprises comparing the202502839

[0110] 10

[0111] confidence score associated with inference data to a preconfigured confidence score threshold. Assuming in a non-limitative manner that the confidence score increases with reliability, the confidence score criterion is for example verified if the confidence score is greater than or equal to the confidence score threshold, since unreliable inference data may not be worth reporting to the RAN. If the confidence score is e.g. a percentage, the confidence score threshold is for example greater than 60%, or greater than 70%.

[0112] Alternatively, or in combination thereof, evaluating whether the inference data reporting criterion is verified may comprise evaluating whether an input data criterion is verified, based on the input data processed by the inference model. For example, in some cases, the inference model may be executed based on the input data present in a buffer of the UE 20. However, in some cases, the input data in the buffer may be updated (refreshed), in all or part, less frequently than the execution of the inference model which may lead to the inference model being executed based on almost the same input data as during the previous execution. In such a case, the inference data obtained might be substantially the same as previously, such that it might not be worth reporting to the RAN. For example, evaluating the input data criterion comprises comparing the amount of (e.g., fresh, i.e. , not processed previously) input data processed to a preconfigured input data threshold. In such a case, the input data criterion is for example verified if the amount of input data is greater than or equal to the input data threshold.

[0113] Alternatively, or in combination thereof, evaluating whether the inference data reporting criterion is verified may comprise evaluating whether a completion time criterion is verified, based on a completion time of the execution of the inference model. For example, evaluating the completion time criterion comprises comparing the completion time of the execution of the inference model to a preconfigured completion time threshold. In some cases, the completion time threshold may correspond to an upper bound for the completion time of the execution of the inference model, and the completion time criterion is for example verified if the completion time of the execution of the inference model is lower than or equal to the completion time threshold. For example, the upper bound may correspond to an expiration of the inference model (e.g., expiration date, maximum duration during which it may be executed), or to a timer after the expiration of which a handover from a first cell to a second cell will take place (e.g., T310 timer, etc.). Indeed, the inference data might not be worth reporting to the RAN if obtained by using an expired inference model, or if obtained after handover to the second cell based on input data obtained for the first cell. In other cases, the completion time threshold may correspond to a lower bound for the completion time of the execution of the inference model, and the completion time criterion is for example verified if the completion time of the execution of the inference model is greater than or equal to the completion time threshold. For example, the lower bound may correspond to a duration ensuring that updated (fresh) input data is obtained for performing the inference.202502839

[0114] 11

[0115] As discussed above, figure 5 represents a diagram showing the main steps of an exemplary embodiment of a method 50 for communicating, which may be implemented by a BS 30 when the UE 20 implements the method 40 for communicating illustrated by figure 4.

[0116] As illustrated by figure 5, the method 50 for exchanging data comprises an optional step S50 of transmitting, to the UE 20, configuration data for configuring the inference data reporting criterion at said UE 20.

[0117] As discussed above, the configuration data may comprise one or more types of criteria to be used, one or more thresholds to be used (e.g., confidence score threshold, input data threshold, completion time threshold, etc.), etc.

[0118] In some examples, the configuration data is transmitted to the UE 20 in a signaling message addressed specifically to said UE 20 (e.g., an RRC reconfiguration message) or to a group of UEs which includes said UE 20, and / or in system information broadcasted by the BS.

[0119] As illustrated by figure 5, the method 50 for communicating comprises a step S51 of receiving inference data from the UE 20, only when the inference data reporting criterion is verified at the UE 20.

[0120] It is emphasized that the present disclosure is not limited to the above exemplary embodiments. Variants of the above exemplary embodiments are also within the scope of the present disclosure.

Claims

20250283912CLAIMS1. A method (40) for communicating in a wireless communication system, the method being implemented by a user equipment, UE (20), of the wireless communication system, wherein the UE comprises a communication unit configured to exchange data with a radio access network, RAN, of the wireless communication system, wherein the exchanged data comprises data related to an inference model to be executed by the UE to produce inference data to be reported to the RAN, the method comprises:(540) executing the inference model to produce inference data based on input data, (541) evaluating an inference data reporting criterion,in response to the inference data reporting criterion not being verified: (S42) skipping transmitting the inference data to the RAN.

2. The method (40) according to claim 1, wherein evaluating the inference data reporting criterion comprises evaluating whether a confidence score criterion is verified, based on a confidence score associated with the inference data.

3. The method (40) according to claim 2, wherein evaluating the confidence score criterion comprises comparing the confidence score to a confidence score threshold, wherein the confidence score criterion is verified if the confidence score is greater than the confidence score threshold.

4. The method (40) according to any one of the preceding claims, wherein evaluating the inference data reporting criterion comprises evaluating whether an input data criterion is verified, based on the input data processed by the inference model.

5. The method (40) according to claim 4, wherein evaluating the input data criterion comprises comparing the amount of input data processed to an input data threshold, wherein the input data criterion is verified if the amount of input data processed is greater than the input data threshold.

6. The method (40) according to any one of the preceding claims, wherein evaluating the inference data reporting criterion comprises evaluating whether a completion time criterion is verified, based on a completion time of the execution of the inference model.

7. The method (40) according to claim 6, wherein evaluating the completion time criterion comprises comparing the completion time of the execution of the inference model to a completion time threshold, wherein the completion time criterion is verified if the completion time of the execution of the inference model is lower than the completion time threshold.

8. The method (40) according to any one of the preceding claims, comprising:receiving, from the RAN, configuration data for configuring the inference data reporting criterion,configuring the inference data reporting criterion based on the received configuration20250283913data.

9. The method (40) according to claim 8, wherein the configuration data includes one or more thresholds.

10. The method (40) according to any one of claims 8 to 9, wherein the configuration data is received by the UE in a signaling message addressed specifically to said UE or to a group of UEs which includes said UE, and / or in system information broadcasted by the RAN.

11. The method (40) according to any one of the preceding claims, comprising (S43) transmitting the inference data to the RAN in response to the inference data reporting criterion being verified.

12. The method (40) according to any one of the preceding claims, wherein the inference model is used for at least one among:channel state information, CSI, compression,CSI prediction,downlink, DL, beam prediction,UE positioning.

13. A user equipment (20) comprising at least one memory and at least one processor configured to carry out a method (40) according to any one of the preceding claims.

14. A method (50) for communicating in a wireless communication system, the method being implemented by a base station, BS (30), of a radio access network, RAN, of the wireless communication system, wherein the BS is configured to exchange data with a user equipment, UE (20), wherein the exchanged data comprises data related to an inference model to be executed by the UE to produce inference data to be reported to the BS, the method comprises:(550) transmitting, to the UE, configuration data for configuring an inference data reporting criterion at said UE,(551) receiving inference data from the UE.

15. The method (50) according to claim 14, wherein the configuration data includes one or more thresholds.

16. The method (50) according to claim 15, wherein the one or more thresholds include at least one among:a confidence score threshold,an input data threshold,a completion time threshold.

17. The method (50) according to any one of claims 14 to 16, wherein the configuration data is transmitted to the UE in a signaling message addressed specifically to said UE or to a group of UEs which includes said UE, and / or in system information broadcasted by the BS.

18. A base station (30) comprising at least one memory and at least one processor configured to carry out a method (50) according to any one of claims 14 to 17.2025028391419. A wireless communication system comprising at least one base station (30) according to claim 18 and at least one user equipment (20) according to claim 13.

20. A computer program product (252, 302) comprising instructions which, when executed by at least one processor, configure said at least one processor to carry out a method (40) according to any one of claims 1 to 12 or a method (50) according to any one of claims 14 to 17.