Technologies for reporting applicable artificial intelligence functionality or model
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-02-05
- Publication Date
- 2026-08-13
Smart Images

Figure CN2025075842_13082026_PF_FP_ABST
Abstract
Description
TECHNOLOGIES FOR REPORTING APPLICABLE ARTIFICIAL INTELLIGENCE FUNCTIONALITY OR MODELTECHNICAL FIELD
[0001] This application relates generally to communication networks and, in particular, to reporting applicability of configured artificial intelligence (AI) features.BACKGROUND
[0002] Third Generation Partnership Project (3GPP) Technical Specifications (TSs) define standards for wireless networks. These TSs describe aspects related to user plane and control plane signaling over the networks.BRIEF DESCRIPTION OF THE DRAWINGS
[0003] FIG. 1 illustrates a network environment in accordance with some embodiments.
[0004] FIG. 2 illustrates a signaling diagram in accordance with some embodiments.
[0005] FIG. 3 illustrates another signaling diagram in accordance with some embodiments.
[0006] FIG. 4 illustrates another signaling diagram in accordance with some embodiments.
[0007] FIG. 5 illustrates an operation flow / algorithmic structure in accordance with some embodiments.
[0008] FIG. 6 illustrates another operational flow / algorithmic structure in accordance with some embodiments.
[0009] FIG. 7 illustrates a user equipment in accordance with some embodiments.
[0010] FIG. 8 illustrates a network node in accordance with some embodiments.DETAILED DESCRIPTION
[0011] The following detailed description refers to the accompanying drawings. The same reference numbers may be used in different drawings to identify the same or similar elements. In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular structures, architectures, interfaces, and techniques to provide a thorough understanding of the various aspects of various embodiments. However, it will be apparent to those skilled in the art having the benefit of the present disclosure that the various aspects of the various embodiments may be practiced in other examples that depart from these specific details. In certain instances, descriptions of well-known devices, circuits, and methods are omitted so as not to obscure the description of the various embodiments with unnecessary detail. For the purposes of the present document, the phrases “A / B” and “A or B” mean (A) , (B) , or (A and B) ; and the phrase “based on A” means “based at least in part on A, ” for example, it could be “based solely on A” or it could be “based in part on A. ”
[0012] The following is a glossary of terms that may be used in this disclosure.
[0013] The term “circuitry, ” as used herein, refers to, is part of, or includes hardware components that are configured to provide the described functionality. The hardware components may include an electronic circuit, a logic circuit, a processor (shared, dedicated, or group) or memory (shared, dedicated, or group) , an application-specific integrated circuit (ASIC) , a field-programmable device (FPD) (e.g., a field-programmable gate array (FPGA) , a programmable logic device (PLD) , a complex PLD (CPLD) , a high-capacity PLD (HCPLD) , a structured ASIC, or a programmable system-on-a-chip (SoC) ) , or a digital signal processor (DSP) . In some embodiments, the circuitry may execute one or more software or firmware programs to provide at least some of the described functionality. The term “circuitry” may also refer to a combination of one or more hardware elements (or a combination of circuits used in an electrical or electronic system) with the program code used to carry out the functionality of that program code. In these embodiments, the combination of hardware elements and program code may be referred to as a particular type of circuitry.
[0014] The term “processor circuitry, ” as used herein, refers to, is part of, or includes circuitry capable of sequentially and automatically carrying out a sequence of arithmetic or logical operations, recording, storing, or transferring digital data. The term “processor circuitry” may refer to an application processor, baseband processor, central processing unit (CPU) , graphics processing unit, single-core processor, dual-core processor, triple-core processor, quad-core processor, or any other device capable of executing or otherwise operating computer-executable instructions, such as program code, software modules, or functional processes.
[0015] The term “interface circuitry, ” as used herein, refers to, is part of, or includes circuitry that enables the exchange of information between two or more components or devices. The term “interface circuitry” may refer to one or more hardware interfaces, for example, buses, I / O interfaces, peripheral component interfaces, and network interface cards.
[0016] The term “user equipment” or “UE” as used herein refers to a device with radio communication capabilities that may allow a user to access network resources in a communications network. The term “user equipment” or “UE” may be considered synonymous to, and may be referred to as, client, mobile, mobile device, mobile terminal, user terminal, mobile unit, mobile station, mobile user, subscriber, user, remote station, access agent, user agent, receiver, radio equipment, reconfigurable radio equipment, or reconfigurable mobile device. Furthermore, the term “user equipment” or “UE” may include any type of wireless / wired device or any computing device, including a wireless communications interface.
[0017] The term “computer system, ” as used herein, refers to any type of interconnected electronic devices, computer devices, or components thereof. Additionally, the term “computer system” or “system” may refer to various components of a computer that are communicatively coupled with one another. Furthermore, the term “computer system” or “system” may refer to multiple computer devices or multiple computing systems that are communicatively coupled with one another and configured to share computing or networking resources.
[0018] The term “resource” as used herein refers to a physical or virtual device, a physical or virtual component or asset within a computing or network environment, or a physical or virtual component within, accessible by, or available to a device or component. Resources could include, but are not limited to, memory space / usage, processor / CPU time, processor / CPU usage, processor and accelerator loads, hardware time or usage, electrical power, input / output operations, ports or network sockets, channel / link allocations, throughput, or workload units. A “hardware resource” may refer to compute, storage, or networking resources provided by physical hardware elements. A “virtualized resource” may refer to compute, storage, or networking resources provided by virtualization infrastructure to an application, device, or system. The term “communication resource” may refer to resources that are accessible by, or available to, computer devices / systems for transferring information over a channel of a communication network. For example, communication resources may include, but are not limited to, time / frequency resources, code resources, modulation resources, etc. The term “system resources” may refer to any kind of shared entities to provide services and may include computing or network resources. System resources may be considered as a set of coherent functions, network data objects, or services accessible through a server where such system resources reside on a single host or multiple hosts and are clearly identifiable.
[0019] The term “channel, ” as used herein, refers to any transmission medium, either tangible or intangible, that is used to communicate data or a data stream. The term “channel” may be synonymous with or equivalent to “communications channel, ” “data communications channel, ” “transmission channel, ” “data transmission channel, ” “access channel, ” “data access channel, ” “link, ” “data link, ” “carrier, ” “radio-frequency carrier, ” or any other like term denoting a pathway or medium through which data is communicated. Additionally, the term “link, ” as used herein, refers to a connection between two devices for the purpose of transmitting and receiving information.
[0020] The terms “instantiate, ” “instantiation, ” and the like as used herein refers to the creation of an instance. An “instance” also refers to a concrete occurrence of an object, which may occur, for example, during the execution of program code.
[0021] The term “connected” may mean that two or more elements at a common communication protocol layer have an established signaling relationship with one another over a communication channel, link, interface, or reference point.
[0022] The term “network element, ” as used herein, refers to physical or virtualized equipment or infrastructure used to provide wired or wireless communication network services. The term “network element” may be considered synonymous with or referred to as a networked computer, networking hardware, network equipment, network node, or a virtualized network function.
[0023] The term “information element” refers to a structural element containing one or more fields. The term “field” refers to individual contents of an information element or a data element that contains content. An information element may include one or more additional information elements.
[0024] FIG. 1 illustrates a network environment 100 in accordance with some embodiments. The network environment 100 may include a UE 104 communicatively coupled with a base station 108 of a radio access network (RAN) 110. The UE 104 and the base station 108 may communicate over air interfaces compatible with 3GPP TSs, such as those that define a Fifth Generation (5G) new radio (NR) system or a later system. The base station 108 may provide user plane and control plane protocol terminations toward the UE 104.
[0025] Operations described herein as associated with devices of the network environment 100 (for example, the UE 104 and the base station 108) may be fully, substantially, or partially performed by processor circuitry of the device.
[0026] The network environment 100 may further include a core network 112. For example, the core network 112 may comprise a 5th Generation Core network (5GC) or a later generation core network. The core network 112 may be coupled to the base station 108 via a fiber optic or wireless backhaul. The core network 112 may provide functions for the UE 104 via the base station 108. These functions may include managing subscriber profile information, subscriber location, authentication of services, or switching functions for voice and data sessions. The core network 112, RAN 110, and RAN 110 may collectively be referred to as network 102.
[0027] The network environment 100 may further include a data network 120. Data network 120 may include a system of interconnected nodes that facilitate data transmission between UE 104 and various application servers and other service providers. The base station 108 and the core network 112 may route application data between the UE 104 and external data network 120 or application servers. These application servers host web applications, cloud storage, and multimedia streaming services, which communicate with the UE 104 via standardized protocols and interfaces defined by 3GPP, ensuring secure and efficient data exchange.
[0028] The UE 104 may be equipped with functionalities for training, deploying, or using AI models. These capabilities may be enabled through a combination of local data collection, network-assisted data exchange, or communication protocols optimized for AI operations. AI operations, as used herein, includes operations utilizing machine learning (ML) , artificial neural networks, deep learning, generative pre-trained transformers (GPTs) , or large language models (LLMs) . The UE 104 may be configured to use AI features for tasks such as optimizing application performance, improving network resource utilization, or enhancing the user experience. For example, the UE 104 may predict network conditions and dynamically adjust its communication parameters to maintain optimal performance. This may involve the collection of data from the device itself, such as sensor inputs or application usage patterns, or data received from the network, including signal strength or traffic load information.
[0029] The base station 108 or network 102 may enable AI functionalities on the UE 104. They may facilitate the transfer of AI models to the UE 104, either by providing pre-trained models or by assisting in the training process through the exchange of relevant data. These operations may occur through control plane signaling or user plane data exchanges. The base station 108 may also support periodic updates to AI models stored on the UE 104, ensuring the models remain accurate and effective as network conditions change.
[0030] The core network 112 may provide support for AI-enabled operations on the UE 104 by managing the collection and distribution of data required for AI model training and inference. This may include access to network performance metrics, subscriber behavior patterns, or other relevant datasets. The core network 112 may host centralized AI model repositories, which the UE 104 can access to download updated models or retrieve inference results. These repositories may allow AI models to be consistently available and aligned with the latest network configurations and capabilities.
[0031] The data network 120, or server 125, may connect the UE 104 to external AI service providers and application servers that may provide the models to the UE 104 and host advanced models or perform computation-intensive tasks on behalf of the UE 104. For example, the UE 104 can offload large-scale AI model training or inference tasks to these external servers, reducing the processing burden on the device. The data network may also facilitate real-time application programming interface (API) interactions between the UE 104 and third-party AI platforms, enabling features such as image recognition, natural language processing, or predictive analytics.
[0032] The configuration of the UE 104 to support AI features may involve integrating these various components of the network environment 100. The base station 108 may provide the physical and logical connection to the RAN 110, the core network 112 may manage data flow and resources, and the data network 120 or server 125 may extend connectivity to external services. Together, these components may enable the UE 104 to collect, process, and use data for AI applications.
[0033] The UE 104, the network 102, the data network 120, or the server 125, may include an AI / ML module, e.g., AI module 130 to execute or control the operation of AI feature. In some examples, structured data, unstructured data, sensor data, streaming data, user-provided data, historical data, multimodal data, or categorical and numerical data inputs can be fed to the AI module 130. The AI module 130 can include one or more learning-based and / or non-learning-based models, e.g., AI model 135, for perceiving, synthesizing, and inferring information. Persons skilled in the art will appreciate that the AI module 135 can include any suitable number of processes to generate predictive analytics, classifications, or recommendations based on input sensor data, user-provided data, or historical data.
[0034] Persons of ordinary skill in the art will appreciate that the AI module 130 can include any suitable machine learning models that are well-known or widely available such as regression techniques, classification techniques, neural networks, and deep learning networks. In instances where the AI module 130 comprises a machine-learning based model (e.g., the AI model 135) , it can be trained to generate predictive analytics, classifications, or recommendations based on input sensor data, user-provided data, or historical data using one or more well-known or widely available training techniques such as supervised leaming, semi-supervised learning, unsupervised learning, and / or reinforcement learning techniques. The training data can include the aforementioned structured data, unstructured data, sensor data, streaming data, user-provided data, historical data, multimodal data, or categorical and numerical data inputs.
[0035] In some embodiments, the UE 104 may provide its capabilities in performing AI operations to the network 102. Network 102 may configure the UE 104 with one or more AI features, e.g., via radio resource control (RRC) signaling. The UE 104 may determine whether the AI feature is suitable or relevant to its current operational conditions or context. Once the network 102 configures the UE 104 with an AI feature, the UE 104 may evaluate the applicability of the feature based on various criteria, which may include environmental factors, network conditions, device capabilities, and the specific requirements of the AI feature itself.
[0036] In one example, the UE 104 may determine a “positive applicability. ” Positive applicability may indicate that the AI feature is deemed relevant and can be utilized by the UE 104 under the current conditions (e.g., conditions of the UE 104, network 102, or the AI model 135) . For example, the UE 104 may determine that the feature aligns with its available resources (e.g., processing power, memory, battery life) , the network's performance (e.g., latency, bandwidth, beam management) , and the data or model required for the feature to operate (e.g., sensor data or network-provided inputs, or model availability) . When the applicability is positive, the UE 104 may activate the AI feature and utilize it to perform tasks such as optimizing communication parameters, improving application performance, or enhancing user experience.
[0037] In one example, the UE 104 may determine a “negative applicability. ” Negative applicability may signify that the AI feature is not suitable for the current context or cannot be effectively executed. This determination could result from a variety of factors, such as insufficient device resources, suboptimal network conditions, mismatch between UE 104 or network 102 configurations and the AI model 135 for the AI feature, or the absence of necessary model or data inputs. For instance, if the feature requires high computational power that the UE 104 cannot provide due to battery constraints or if the network latency exceeds the threshold necessary for the feature's operation, the UE 104 may determine negative applicability. In such cases, the AI feature may remain inactive, and the UE may either defer its use or request assistance from the network, such as offloading the feature to a more capable external server.
[0038] The determination process of applicability of the AI feature may involve the UE 104 analyzing static or dynamic parameters or configurations. Static parameters may include the predefined requirements of the AI feature and the UE's hardware and software capabilities. Dynamic parameters may involve real-time conditions, such as current network metrics, device state, and environmental factors. The UE 104 may also consider policy rules or configuration settings provided by the network 102 to guide its determination process. Once the determination is made, the UE 104 may communicate its determination to the network 102.
[0039] The UE 104 may generate an applicability report for transmission to the network 102 to indicate whether the AI feature is applicable. The applicability report may indicate the positive or negative applicability. In some instances, when the applicability report indicates a negative applicability, the UE 104 may include, in the applicability report, a cause or condition for the negative applicability of the AI feature. The UE 104 may include the applicability report in RRC signaling or UE assistance information (UAI) signaling.
[0040] In some embodiments, negative applicability may be caused by a network condition, a UE condition, or a condition of the AI model 135. The network condition may include a configuration, parameter, or state at the network 102 that may not be known to the UE 104. In some embodiments, UE 104 may detect a network condition when there is a mismatch between an identifier (ID) associated with the network 102 and an ID associated with the UE 104 or the AI feature or AI model 135. For example, the ID of a beam management configuration set by the network 102 may be different from the ID of a beam management configuration associated with the AI feature or AI model 135.
[0041] The UE condition may include a UE configuration or parameter of the UE 104 that may not be known to the network 102. In some embodiments, the configuration or parameters may include UE speed, UE location, antenna placement in the UE 104, battery status of the UE 104, or memory or storage capacity of the UE 104.
[0042] The model availability condition may indicate whether UE 104 has the AI model 135 available in the device or a new model needs to be downloaded from server 125 or data network 120 to the device, e.g., UE 104. For example, the AI model 135 may be used for beam management, positioning, channel state information (CSI) prediction, or CSI compression, and the AI model 135 may be stored external to the UE 104 device, e.g., at network 102, data network 120, or server 125. In some instances, the applicability report may include a time indicating a duration that UE 104 may need to make the model available, e.g., by downloading it from the server 125.
[0043] When the UE 104 reports “negative applicability” for a configured AI feature and includes the cause or condition, it provides actionable feedback to the network 102 to enhance service performance or resource efficiency. The network 102 can use this information to address specific issues, such as adjusting the feature's configuration to align with the UE's capabilities, improving network conditions like latency or bandwidth, or offioading tasks to extemal servers 125. This feedback can also help identify recurring issues across multiple UEs, guiding resource optimization, feature redesign, or infrastructure upgrades. By leveraging these reports, the network 102 can make immediate adjustments and long-term improvements to ensure better applicability of AI features and more efficient use of resources.
[0044] FIG. 2 illustrates a signaling diagram 200 in accordance with some embodiments. The signaling diagram 200 is an example of signaling for configuring, managing, and reporting of AI-enabled features. The signaling diagram 200 includes six operations that define the processes and interactions between the network 102 and the UE 104.
[0045] At 220, the network 102 may initiate the process by sending a UE capability inquiry message to the UE 104. This message may query the UE 104 to determine which AI-enabled features and functionalities it supports. These features may include the ability to execute AI models (e.g., AI model 135) locally, perform inference tasks, or handle specific configurations. Through this inquiry, the network 102 may gain detailed information on the UE's static capabilities, which may include hardware support, software features, and other relevant limitations. Additionally, the network 102 may obtain a baseline for determining which AI functionalities or features can potentially be configured. The inquiry may also include specific fields indicating the types of AI functionalities or features of interest to the network. The operations at 220 may be referred to as Step 1. In some embodiments, steps may be in order, but are not required to be in other embodiments.
[0046] At 225, in response to the UE capability inquiry, the UE 104 may send a UE capability information message. This message may provide a detailed list of supported AI functionalities or features, including supported configurations. For instance, the UE 104 may report its ability to support AI models for beam management, mobility prediction, or other specific use cases. The UE 104 may also include information about supported network-side additional conditions, such as associated IDs or configurations it can process. In one example, when the number of associated IDs are smaller than a threshold value, the UE 104 may include the associated IDs in the report. When the number of associated IDs exceeds the threshold, the UE may feedback the associated IDs with the applicability report, for example, as described at 235 below. The UE capability information may allow the network 102 to filter and identify the AI functionalities or features that align with the UE's capabilities and the network's intended operations. The UE capability information may establish a foundation for the network 102 to proceed with configuring specific functionalities in subsequent signaling. The operations at 2205 may be referred to as Step 2.
[0047] At 230, after evaluating the UE's capabilities, the network 102 may send a configuration message to configure the AI functionalities or features on the UE 104. In some instances, the network 102 may send an RRC reconfiguration message to configure the AI functionalities or features on the UE 104. This message may specify the functionalities to be enabled, configured, or evaluated, along with associated parameters. The configuration may include: network-side additional conditions, inference configurations, or applicability reporting instructions.
[0048] In one example, the configuration message may include network-side additional conditions. For example, the configuration message may include associated IDs that define the conditions under which a model (e.g., AI model 135) was trained or should be inferred. The additional network-side conditions may provide consistency between training and inference.
[0049] In one example, the configuration message may include inference configurations. The network 102 may provide details such as beam resource configurations, measurement parameters, or other functionality-specific configurations.
[0050] In one example, the configuration message may include applicability reporting instructions. The network 102 may configure the UE 104 to report applicable, positive applicability, or non-applicable, negative applicability functionalities. The configuration message may configure the UE 104 to report the applicability report periodically or aperiodically (e.g., upon specific triggers such as changes in UE 104 conditions) . In some instances, the configuration message may also include filtering information to limit the scope of applicability reporting, thereby reducing signaling overhead. The operations at 230 may be referred to as Step 3.
[0051] In some instances, the configuration message, e.g., the RRC reconfiguration message, may include an indication of permission for UE assistance information (UAI) reporting. The network 102 may explicitly or implicitly allow the UE 104 to provide the applicability report via UAI reporting. For example, the UE 104 may use the “other configuration” field in the UAI for applicability reporting.
[0052] In some instances, the applicability report may be based on one or both of the following: 1) one or more channel state information (CSI) report configurations for inference reporting or 2) one or more sets of inference-related parameters dedicated to the applicability report.
[0053] In some instances, the configuration message may include parameters for beam management use cases. For example, the configuration message may include time instances for measurements or time instances for predictions.
[0054] At 235, after receiving the configuration, the UE 104 may evaluate the applicability of the configured AI features or functionalities and send a report back to the network 102. The UE 104 may consider multiple factors to evaluate the applicability of the configured AI features or functionalities, such as network-side additional conditions, UE-side additional conditions, model availability, or dynamic environmental factors.
[0055] In one example, the UE 104 may consider network-side additional conditions to evaluate the applicability of the configured AI features or functionalities. The UE 104 may check whether the trained model (e.g., AI model 135) matches the signaled network-side additional condition. For example, the UE 104 may check whether the configured conditions (e.g., associated IDs) are met.
[0056] In one example, the UE 104 may consider UE-side additional conditions to evaluate the applicability of the configured AI features or functionalities. The UE 104 may check whether the trained model (e.g., AI model 135) matches the current UE-side additional condition. For example, the UE 104 may evaluate internal factors such as processing power, memory availability, or battery level and determine whether configured AI features or functionality are applicable.
[0057] In one example, the UE 104 may consider model availability to evaluate the applicability of the configured AI features or functionalities. The UE 104 may determine whether the required trained AI models are available and ready for inference.
[0058] In one example, the UE 104 may consider dynamic environmental factors to evaluate the applicability of the configured AI features or functionalities. The UE 104 may account for changing network conditions, such as signal strength or latency, in evaluating the applicability of configured AI features or functionalities.
[0059] The UE's applicability report may specify whether each configured functionality is applicable, e.g., positive applicability, or not, e.g., negative applicability. For functionalities deemed non-applicable, the UE 104 may include additional information, such as the reason, cause, or condition leading to the negative applicability. This feedback may provide actionable insights to the network, enabling the network 102 to refine the configuration or address specific limitations. The operations at 235 may be referred to as Step 4.
[0060] In some embodiments, the UE 104 may use RRC signaling to send the applicability report to the network 102. For example, the UE 104 may include the applicability report in the RRC reconfiguration complete message. Additionally or alternatively, UE 104 may use UE assistance information to report the applicability report.
[0061] At 240, based on the applicable functionality report from the UE 104, the network 102 may send an updated reconfiguration message to refine the configuration at 230, e.g., step 3. The network 102 may send an RRC reconfiguration message to refine the AI features or functionalities configurations. In some instances, e.g., in reactive reporting, the network 102 may dynamically adapt the configuration based on real-time feedback. The updated configuration may include: updated inference configurations, deactivation of non-applicable features or functionalities, or activation of features or functionalities or alternate features or functionalities.
[0062] In one example, the network 102 may update the inference configuration. The reconfiguration message may include adjustments to model parameters, resource allocations, or measurement settings to better align with the UE's conditions.
[0063] In one example, the network 102 may deactivate non-applicable features or functionalities. If a feature or functionality is determined to be non-applicable, the network 102 may deactivate it. Deactivating the non-applicable features or functionalities may conserve resourc es.
[0064] In one example, the network 102 may activate alternate functionalities or previously configured features or functionalities. The network 102 may configure alternative functionalities that are more suitable for the UE's current condition. In some embodiments, the network 102 may explicitly or implicitly activate the AI features or functionalities that are applicable, e.g., associated with a positive applicability report. The operations at 220 may be referred to as Step 5.
[0065] At 245, the configured AI features or functionalities are activated, deactivated, or monitored. The activation or deactivation may be initiated by the UE 104 or by the network 102. The operations at 245 may be referred to as Step 6.
[0066] In one example, for features or functionalities that are deemed applicable, the UE 104 may execute the AI models and perform inference tasks based on the provided configuration.
[0067] In one example, the UE 104 or the network 102 may monitor the performance of the configured features or functionalities. The UE 104 or the network 102 may collect data and evaluate metrics such as accuracy, latency, or resource utilization to evaluate and monitor the performance of applicable features or functionalities.
[0068] In one example, if the performance metrics indicate suboptimum operation, the network 102 or the UE 104 may dynamically adapt by initiating further configuration updates or deactivating the feature or functionality.
[0069] In one example, ifthe functionality is no longer needed or becomes non-applicable due to changing conditions, it is deactivated. The operation in step 6 may ensure that the AI features or functionalities operate as intended while allowing for ongoing adjustments to maintain their effectiveness and efficiency.
[0070] In some embodiments, when step 5 is not performed and step 3 includes all inference related configuration for the AI feature, the UE 104 may send RRC Reconfiguration Complete message indicating whether the AI feature is applicable or not applicable. The RRC Reconfiguration Complete message may indicate the cause of non-applicability when the AI feature is not applicable. The RRC Reconfiguration Complete message may include applicability status for each inference related configuration. For each inference related configuration that is indicated as non-applicable, the UE 104 may include the cause associated with the non-applicability determination. When the AI feature is deemed non-applicable due to the model availability, the UE 104 may indicate the time duration needed for the model (e.g., AI model 135) to become available.
[0071] In some embodiments, when step 3 includes partial inference related configuration, e.g., partial configuration that are suitable for UE 104 to check applicability, the UE 104 may check whether the AI feature is applicable. The UE 104 may send RRC Reconfiguration Complete message indicating whether the model is applicable. If the AI model is deemed non-applicable, the cause associated with non-applicability may be included in the RRC Reconfiguration Complete message. The network 102 may send the remaining inference related configuration in step. In addition, the network 102 may configure UAI in step 5, allowing the UE 104 to report any change in applicability, e.g., from non-applicable to applicable or from applicable to non-applicable. The UE 104 may send UAI update. For example, ifthe remaining configuration or change in condition make the AI model applicable, the UE 104 may send UAI to indicate that the AI feature is applicable.
[0072] In some embodiments, when any of the conditions or causes in step 3 does not match, or the UE 104 is configured to report adaptability based on both RRC Reconfiguration Complete and UAI, in step 4, the UE 104 may send RRC Reconfiguration Complete message to indicate model applicability is false along with an indication of the cause of inapplicability of the model.
[0073] In some embodiments, when any of the conditions or causes in step 3 does not match, and the UE 104 is configured by the network 102 to use UAI for applicability reporting, UE 104 may only report applicability in UAI and may not include applicability in RRC Reconfiguration Complete.
[0074] In some instances, reporting positive applicability of an AI feature may indicate activation of that feature or reporting negative applicability of the AI feature may indicate deactivation of the feature. In some instances, the network 102 may explicitly or implicitly activate or deactivate the AI feature, e.g., via RRC Reconfiguration message.
[0075] In some embodiments, after RRC Reconfiguration Message or UAI confirmation of the applicability of the AI feature, or after the network 102 activating the AI feature, the UE 104 may further send UAI to indicate that an AI feature that was previously applicable is no longer applicable. The UE 104 may include detailed reasons or conditions that are causing the previously applicable AI feature becoming inapplicable. For example, UE-side changes in condition may cause a previously applicable AI feature become inapplicable, e.g., the AI feature does not support UE speed for CSI prediction, AI feature is not trained for the new scenarios (for example, move to a new site within a cell for AI based direct positioning) , or there are insufficient resources (e.g., memory or battery) for AI feature. For example, in AI-based direct positioning, the UE 104 may directly generates the positioning based on measured channel. The model may be trained as site specific model, and thus, when the UE 104 moves to a new site, or the site has changed (e.g., environmental or structural changes, for example, large blocking materials or structures being moved, etc. ) , the performance of the site-specific trained model may degrade.
[0076] In some embodiments, the UE 104 may report to the network 102 whether an AI feature is activated or deactivated. For example, for each configured AI feature, the UE 104 may include in the applicability report, e.g., via RRC Reconfiguration Complete or UAI, an indication of whether that AI feature is activated or deactivated.
[0077] FIG. 3 illustrates another signaling diagram 300 in accordance with some embodiments. Signaling diagram 300 is an example of reporting non-applicable cause in a configuration complete message, e.g., RRC reconfiguration complete.
[0078] At 320, signaling diagram 300 includes transmission, by the network 102 to the UE 104, of a UE capability inquiry, for example, step 1 as described in signaling diagram 200 at 220.
[0079] At 325, signaling diagram 300 includes transmission, by the UE 104 to the network 102, of a UE response to the UE capability inquiry, for example, step 2 as described in signaling diagram 200 at 225. The UE response may include UE features and capabilities related to the AI features or functionalities.
[0080] At 330, signaling diagram 300 includes transmission, by the network 102 to the UE 102, of configuration message, for example, step 3 as described in signaling diagram 200 at 225. The configuration message may include parameters associated with the AI feature, including assisted information used in data collection. The configuration message may be an RRC connection reconfiguration message.
[0081] At 335, the UE 104 may determine whether the UE-side model is trained for the configured AI feature or functionality. Additionally, or alternatively, the UE 104 may determine whether the UE-side model is ready for inference.
[0082] The UE 104 may determine a UE-side model is trained for the configured AI feature and is ready for inference when 1) the UE 104 determines that all UE-side additional conditions are met, 2) the UE 104 determines that all network-side additional conditions are met, or 3) the UE 104 determines that the model is on the device, compiled, and available to run inference.
[0083] In one example, determining that all UE-side additional conditions are met may include confirming that the UE 104 has sufficient processing power, memory, or battery resources to support the AI functionality; or necessary sensor data or other inputs are available and accurate to use the trained model in making inferences. Additionally, or alternatively, it may involve determining that the scenario or site aligns with the requirements of a site-specific model or that the UE's speed or Doppler shift is suitable for making inferences with the trained model, e.g., AI-based CSI prediction. Conversely, the UE may determine that conditions are not met if resource constraints, misaligned scenarios, or unsuitable mobility states prevent the effective execution of the AI functionality, e.g., making inferences using the trained model of the AI feature.
[0084] In one example, determining that all network-side additional conditions are met may include determining that, for example, Set A and Set B configurations for beam management are aligned, an associated ID provided by the network 102 matches the trained model and inference configurations, or time instances for measurements and predictions are synchronized. Additional conditions may also involve consistent signal properties, such as transmission power and antenna configurations, or alignment with specific deployment scenarios, such as Urban Micro or Urban Macro. These conditions, evaluated by the UE 104 during applicability reporting, ensure that the configured AI features can function effectively under the network's provided parameters. Conversely, the UE 104 may determine that not all network-side additional conditions are met if, for instance, the provided Set A and Set B configurations are inconsistent with the AI model's training data, the associated ID does not align with the inference configuration, or the deployment scenario differs from the conditions under which the AI model was trained. These conditions, evaluated by the UE during applicability reporting, ensure that the configured AI features can function effectively under the network's provided parameters. Set A and Set B, above, refer to beam resource configurations used in beam management, where Set A may represent a larger set of candidate beams for initial evaluation, and Set B may be a refined subset of beams selected for further measurements or predictions.
[0085] In one example, determining that model availability conditions are met may include determining that the required AI model is present on the device, trained, compiled, and available to run inference, ensuring compatibility with the current network-provided configurations such as associated IDs, Set A and Set B configurations, or other inference parameters. This may also involve verifying that the model is up-to-date and trained to reflect the latest network conditions or site-specific requirements. Conversely, the UE 104 may determine that model availability conditions are not met ifthe model is not stored or partially stored on the device, not compiled, outdated, or otherwise unavailable to perform inference under the provided configurations, thereby preventing the effective execution of the AI functionality.
[0086] When the UE 104 evaluates the applicability of a configured AI feature or functionality, it may use a combination of network-provided configurations, internal conditions, and model readiness to determine whether the feature or functionality can be executed. The UE 104 may communicate the results of this evaluation to the network through a reporting mechanism. In some embodiments, based on whether the operation at step 5, described in FIG 2. at 240, is performed, the following two options may be used for transmission of the applicability report.
[0087] In Option 1, step 5 is not performed, e.g., the network 102 does not send the reconfiguration message to reconfigure or adjust the configuration of the AI feature. In some instances, step 5 may be unnecessary when the initial configuration in step 3 fully includes all required parameters (e.g., CSI-Report configuration, associated IDs, etc. ) and the UE 104 confirms applicability in step 4 without the need for refinement.
[0088] At 340, the UE 104 may generate and send a message indicating the applicability status of the AI feature or functionality, along with any relevant details regarding non-applicability, if applicable. For example, the UE 104 may use RRC signaling to report applicability. The UE 104 may use the RRC Reconfiguration Complete message to convey AI feature applicability information to the network 102.
[0089] In some embodiments, the RRC Reconfiguration Complete message may be sent within 16 milliseconds (ms) . This message may include the applicability status of each configured inference-related parameter. If the UE 104 determines that an inference-related configuration is non-applicable, the message may identify the condition or cause of the non-applicability. The causes or conditions are categorized as follows: Cause 1: not all UE-side additional conditions are met (e.g., insufficient processing power, memory, or battery resources) ; Cause 2: not all network-side additional conditions are met (e.g., mismatches in Set A / Set B configurations or associated IDs) ; and Cause 3: the required AI model is not available (e.g., the model is not present, compiled, or ready for inference on the device) .
[0090] In some embodiments, for configurations marked as non-applicable, the UE 104 may include an estimate of how long it would take for the functionality to become applicable. For example, if the network-side and UE-side conditions are satisfied, but the model transfer from an external server 120 or 125 is still ongoing, the UE 104 may indicate the expected time for the model to become available.
[0091] When all conditions (for example, UE-side, network-side, and model availability) are met, the UE 104 may report positive applicability. The UE 104 may report positive applicability in the RRC Reconfiguration Complete message. For instance, ifthe UE has sufficient resources, the provided Set A and Set B configurations align with the AI model training, and the model is compiled and ready for inference, the UE 104 may confirm the functionality as applicable. Conversely, if one or more conditions are not met, the UE 104 may report negative applicability and specifies the cause. For example, ifthe model is missing on the device (Cause 3) , the UE 104 may indicate negative applicability along with the cause in the message. In some instances, the UE 104 may also provide an estimate for when the model will become available, enabling the network to adjust its configurations or wait for the functionality to become applicable.
[0092] By structuring the RRC Reconfiguration Complete message in this way, the UE 104 may provide actionable insights to the network, allowing it to refine configurations, address resource constraints, or manage AI functionality activation more effectively. This reporting mechanism may enable efficient communication between the UE 104 and the network 102, improving the adaptability of AI functionalities in dynamic environments.
[0093] In Option 2, step 5 is performed, e.g., the network 102 sends the reconfiguration message to reconfigure or adjust the configuration of the AI feature. This step may be triggered after the UE 104 sends an applicability report using the RRC Reconfiguration Complete message in step 4. In some embodiments, the initial applicability report may be based on a partial inference-related configuration provided in step 3. The partial configuration may include enough information for the UE 104 to determine whether the AI functionality is applicable. Ifthe UE 104 determines that the AI functionality is applicable, the network 102 proceeds by sending the remaining inference-related configuration in step 5. This additional configuration may complete the configuration of the AI functionality and make the AI feature ready for execution by the UE 104.
[0094] If the applicability report in step 4 indicates that the AI functionality is non-applicable, e.g., negative applicability, the UE 104 may include the cause of non-applicability in the RRC Reconfiguration Complete message. The causes are categorized and described above.
[0095] Based on the applicability report, if the network 102 determines that adjustments to the configuration are necessary, it may send an updated RRC Reconfiguration message in step 5.This message may include additional inference-related parameters or instructions to address the identified causes of non-applicability.
[0096] After receiving the updated configuration in step 5, the UE 104 may reevaluate the applicability of the AI functionality. The UE 104 may check all three conditions (UE-side, network-side, and model availability) against the new configuration and generate a second applicability report. If the updated configuration resolves the causes of non-applicability, the UE 104 may send a positive applicability report, indicating that the AI functionality is now applicable. However, if one or more conditions remain unfulfilled, the UE 104 may send another negative applicability report specifying the unresolved causes. This second applicability report may be transmitted using UE assistance information (UAI) to provide the network 102 with the updated status.
[0097] In some embodiments, the use of UAI for applicability reporting may be configured in step 5, e.g., via RRC Reconfiguration message, to allow the UE 102 to report any subsequent changes in applicability. For instance, if the AI functionality transitions from applicable to non-applicable due to Cause 2 (e.g., a change in network-side conditions) or Cause 3 (e.g., the model becoming unavailable) , the UE can dynamically notify the network using UAI. Similarly, ifthe UE 104 initially provides no indication of applicability in step 4 (neither positive nor negative) , the UAI can be used to provide further updates once the applicability status is determined.
[0098] This iterative process may enable the network 102 and the UE 104 to maintain a dynamic and adaptive interaction to configure and execute AI features and functionalities. By using step 5 to refine configurations and leveraging UAI for ongoing updates, the framework may support continuous optimization or refinement of AI features or functionality in response to changing conditions.
[0099] The following is applicable to both Option 1 and Option 2.
[0100] At 345, once the AI feature or functionality is deemed to be positively applicable, the UE 104 may proceed with executing the intended inference using the UE-side model. This determination is based on the UE confirming that all necessary conditions are met, including the availability of the AI model, the satisfaction of network-side additional conditions (e.g., associated ID, Set A / Set B configurations, or inference parameters) , and the fulfillment of UE-side additional conditions such as processing power, memory, and battery resources. The UE-side model, which has been compiled and prepared for inference, is then used to process the relevant data inputs, such as CSI measurements or sensor data, to generate the desired outputs. This inference process may involve real-time predictions, optimizations, or decision-making tasks, depending on the configured AI functionality.
[0101] At 350, following the completion of the inference process, the UE 104 may report the results of the inference back to the network 102. This inference report can include details such as the outputs of the AI model, performance metrics (e.g., latency, accuracy, or resource utilization) , and any conditions or observations that may affect the functionality's ongoing operation. For example, ifthe inference involves beam prediction for beam management, the UE 104 may report the predicted beam indices or related measurements. The report may also serve as feedback for the network 102 to evaluate the effectiveness of the configured AI functionality, enabling further refinements or adjustments to the feature. This reporting mechanism may allow the network 102 to remain informed about the AI functionality's operation and take proactive steps to optimize its performance or address any emerging issues.
[0102] FIG. 4 illustrates another signaling diagram 400 in accordance with some embodiments. Signaling diagram 400 is an example of reporting a non-applicable cause in a configuration complete message, e.g., RRC reconfiguration complete and a UAI.
[0103] At 420, signaling diagram 400 includes transmission, by the network 102 to the UE 104, of a UE capability inquiry, for example, step 1 as described in signaling diagram 200 at 220.
[0104] At 425, signaling diagram 400 includes transmission, by the UE 104 to the network 102, of a UE response to the UE capability inquiry, for example, step 2 as described in signaling diagram 200 at 225. The UE response may include UE features and capabilities related to the AI features or functionalities.
[0105] At 430, signaling diagram 400 includes transmission, by the network 102 to the UE 102, of configuration message, for example, step 3 as described in signaling diagram 200 at 225. The configuration message may include parameters associated with the AI feature, including assisted information used in data collection. The configuration message may be an RRC connection reconfiguration message.
[0106] At 435, the UE 104 may determine whether the UE-side model is trained for the configured AI feature or functionality, for example, as described in signaling diagram 300 at 335.
[0107] In some embodiments, both RRC signaling and UAI may be utilized for applicability reporting to provide a more dynamic and responsive framework for AI functionality management. In step 3, the network 102 (e.g., the base station 108) may send an RRC Reconfiguration message to the UE 104, including the necessary inference-related configurations and instructions for applicability determination. If any of the conditions for applicability (network-side additional conditions, UE-side additional conditions, or model availability) do not match, or if the network explicitly configures the use of UAI for applicability reporting in step 3, the UE 104 may follow a dual-layered approach involving both RRC Reconfiguration Complete and UAI updates.
[0108] In some embodiments, in step 4 at 440, the UE 104 may send an RRC Reconfiguration Complete message indicating that the AI model applicability is false (non-applicable) . Along with this indication, the UE 104 may include the specific cause of non-applicability to assist the network in understanding the underlying issue. The causes may be those described above, e.g., Cause 1-3. In addition causes may include: Cause A, indicating that trained model does not match the signaled network-side additional condition; Cause B, indicating that the trained model does not match current UE-side additional condition; and Cause C, indicating that the trained model is available on server and require additional time to download. In some embodiments, Cause 2 may include Cause A, or Cause A may also include Cause 2; Cause B may include Cause 1, or Cause 1 may also include Cause B; or Cause C may include Cause 3, or Cause 3 may include Cause C.
[0109] In some embodiments, when Cause A is signaled, no further UAI updates are needed until the network 102 can send new configuration (e.g., new associated ID or new assisted information) . The mismatch lies in the network-side conditions, which can only be resolved by the network. In such cases, the network 102 may send new configurations (e.g., updated associated IDs or assisted information) to address the issue. This may eliminate unnecessary uplink signaling, as the network is responsible for taking corrective actions. In some embodiments, the UE 104 may provide, in a UAI, associated IDs of received data collection configuration for beam management and CSI prediction, or required assistance data information element (IE) of UE-based Downlink Time Difference of Arrival (DL-TDOA) or UE-based Downlink Time Difference of Arrival (DL-AoD) for positioning use case to assist the network 102 to reconfigure the AI feature.
[0110] For Cause B or Cause C, the UE 104 may provide additional context through UAI to assist the network 102 in reconfiguring the AI feature. For Cause 2, the UE 102 may include more granular information, such as whether the limitation is due to insufficient memory, battery, or processing power. For Cause 3, the UE 104 may indicate the estimated time required to download and prepare the model for inference. This provides the network with actionable insights to manage the reconfiguration process efficiently.
[0111] In cases where UAI is used for applicability reporting as configured in step 3, in some instances, the UE 104 may not include applicability information in the RRC Reconfiguration Complete message. Instead, the UE 104 may report applicability status and related information exclusively through UAI. For example, the UE 104 may send UAI updates with: associated ID (s) of received data collection configurations for beam management (BM) and CSI prediction or assistance data information elements (IEs) required for positioning use cases, such as UE-based Downlink Time Difference of Arrival (DL-TDOA) or Downlink Angle of Departure (DL-AoD) , which can support the network in preparing refined configurations for positioning-related AI functionalities.
[0112] Further, UAI updates may allow the UE 104 to report when conditions change. For example, if UE-side additional conditions improve (e.g., additional memory becomes available) , the UE 104 may report that the AI functionality is now applicable, or when the model becomes available after being downloaded and prepared, the UE 104 may update the network via UAI.
[0113] The network 102 may evaluate the UAI updates and decide whether to activate the model. If activation is deemed necessary, the network may send another RRC Reconfiguration message with the inference configuration to activate the AI functionality.
[0114] For periodic reporting, activation may require an additional RRC reconfiguration message specifying the conditions for periodic operation. For aperiodic or semi-persistent use cases, activation may be triggered via Downlink Control Information (DCI) or Medium Access Control (MAC) Control Element (CE) signaling.
[0115] By combining RRC signaling and UAI, this approach enables a robust framework for applicability reporting, allowing the UE 104 and network 102 to dynamically adapt to changing conditions while optimizing signaling overhead and operational efficiency.
[0116] Combining RRC signaling for initial applicability reporting and UAI for follow-up assistance information may provide a flexible and efficient mechanism for managing AI functionalities. The RRC Reconfiguration Complete message may provide immediate feedback to the network 102 regarding the current applicability status, while UAI may enable the UE 104 to supply additional context to support timely reconfiguration by the network 102. This dual-layered approach may enhance the adaptability of AI functionalities in dynamic network environments while reducing signaling overhead.
[0117] In some embodiments, at 445, the network 102 may send additional configuration to enable the UE 104 with UAI report. The configuration may be delivered through an RRC Reconfiguration message, where the network specifies the criteria and triggers for UAI reporting. This may include defining aperiodic reporting, event-based reporting (e.g., changes in network-side additional conditions or UE-side constraints) , or periodic updates. The network 102 may also configure the UAI framework to include specific information elements, such as applicability statuses, associated IDs, or assistance data for positioning use cases, ensuring the reports are tailored to the network’s operational needs. This may allow the UE 104 to dynamically adapt its reporting behavior to provide the network with timely and actionable insights.
[0118] To resolve negative applicability, in some embodiments, the UE 104 may download a new AI model to address conditions preventing the AI functionality from being applicable.
[0119] At 450, the UE 104 may send an inquiry to server 125, requesting the necessary AI model. This inquiry may include specific parameters such as the associated ID received from the network 102, the type of model required, and any additional configuration details provided in the earlier RRC Reconfiguration message. The server 125, which hosts the AI models, may process the request and provide the appropriate model that aligns with the network’s configurations and the UE’s operational constraints.
[0120] At 455, once the UE 104 receives the model, it may verify its compatibility with the previously signaled network-side additional conditions (e.g., associated IDs, Set A / Set B configurations) ; and check its readiness for execution by ensuring the model is compiled and available for inference. If the model satisfies the requirements, the UE may update its applicability status. This updated status may then be reported back to the network using UAI or other signaling mechanisms, confirming that the AI functionality is now positively applicable.
[0121] At 460, in some embodiments, after the UE 104 indicates to the network 102 that the model is ready, the network 102 may send an RRC Reconfiguration message with the remaining inference-related configurations required for the AI feature. This message may serve to finalize the setup for the AI functionality by providing detailed parameters necessary for inference execution.
[0122] At 465, the UE 104 may report the results of the inference back to the network 102, for example, as described in signaling diagram 300 at 350.
[0123] FIG. 5 illustrates an operation flow / algorithmic structure 500 in accordance with some embodiments. The operation flow / algorithmic structure 500 may be performed or implemented by a UE such as, for example, the UE 104 or UE 700; or components thereof, for example, baseband processor circuitry 704A.
[0124] The operation flow / algorithmic structure 500 may include, at 510, processing a configuration of an AI feature. The configuration may be provided through a message such as RRC Reconfiguration. The configuration may include parameters like CSI-ReportConfig for managing channel state information, associated IDs for aligning training and inference, and Set A / Set B configurations for beam management. It may also specify time instances for measurements or predictions, ensuring proper alignment with network operations.
[0125] Examples of AI features may include beam prediction for efficient beam management, positioning enhancements using DL-TDOA or DL-AoD for precise location inference, and traffic load prediction to optimize resource allocation. The configuration may define reporting mechanisms (e.g., aperiodic, periodic, or event-triggered) and resource requirements (e.g., memory, bandwidth) .
[0126] Additionally, the configuration may outline operational constraints such as latency limits or fallback conditions for deactivating the feature if applicability changes. These details ensure the AI functionality is tailored to both the UE’s capabilities and the network’s requirements, enabling seamless execution and adaptability.
[0127] The operation flow / algorithmic structure 500 may include, at 520, determining a condition associated with the AI feature. This may involve evaluating whether the necessary conditions for executing the AI functionality are met, specifically focusing on scenarios where one or more conditions fail to align with the requirements. For example, the UE may determine that UE-side conditions such as insufficient processing power, memory, or battery resources prevent the feature from operating effectively. Similarly, the operation flow 500 may identify that network-side conditions are not satisfied, such as mismatched associated IDs or inconsistencies in Set A and Set B configurations, which disrupt alignment between the training and inference environments.
[0128] Additionally, the operation flow may determine that the AI model required for the feature is unavailable, for instance, ifthe model is not stored, compiled, or ready to run inference on the device. Other dynamic factors, such as changes in network conditions (e.g., latency, bandwidth) or environmental constraints (e.g., mobility state or Doppler shift) , may also contribute to unmet conditions.
[0129] The operation flow / algorithmic structure 500 may include, at 530, determining the negative applicability of the AI feature. The operation flow 500 may determine that the AI functionality cannot be activated or executed due to unmet conditions identified in the previous step. Negative applicability is determined when one or more conditions-such as UE-side, network-side, or model availability-are not satisfied. For example, the UE may determine negative applicability if it lacks sufficient processing power, memory, or battery resources or if the associated ID or Set A / Set B configurations provided by the network do not align with the AI model’s training environment.
[0130] Additionally, negative applicability may be determined ifthe required AI model is unavailable-for instance, if it is not stored, compiled, or ready for inference. Dynamic factors such as poor network conditions, excessive latency, or mobility issues (e.g., high Doppler shift) may also contribute to this determination. By identifying and classifying the specific reasons for negative applicability, the UE prepares the groundwork for generating a detailed applicability report to communicate the issue to the network, enabling corrective actions or reconfigurations.
[0131] The operation flow / algorithmic structure 500 may include, at 540, generating an applicability report indicating the negative applicability and the condition. When the operation flow 500 determines that one or more conditions required for the AI feature are not met, it may generate a report to communicate the negative applicability to the network. The report may specify the applicability status as ‘negative’a nd include detailed information about the condition (s) that caused the negative applicability.
[0132] For instance, ifthe negative applicability arises due to UE-side conditions, such as insufficient resources (e.g., low or insufficient battery or memory) , the report may explicitly or implicitly identify this as the cause. If the issue lies with network-side conditions, such as a mismatch in associated IDs or inconsistencies in Set A and Set B configurations, the report may highlight these misalignments. Similarly, if the AI model required for the feature is unavailable (e.g., missing, not compiled, or not ready for inference) , the report includes this as a cause.
[0133] The report may also include additional context, such as an estimated time for the condition to resolve. For example, the UE might indicate how long it would take to download and prepare the required model ifthe issue is related to model availability. The applicability report is generated in a structured format and transmitted to the network via signaling mechanisms such as RRC Reconfiguration Complete or UAI, ensuring that the network is informed not only of the negative applicability but also of the specific reasons behind it.
[0134] In some embodiment, the operation 500 may configure the UE 104 with partial configuration of the AI feature. The operation 500 may include generating, by the UE 104 and for transmission to the network 102, an RRC Reconfiguration Complete message indicating a positive applicability of the AI feature. In response, the operation 500 may include processing a remaining configuration of the AI feature received by the UE 104 from the network 102.
[0135] In some embodiments, the operation 500 may include processing a reconfiguration message to de-configure or deactivate the AI feature.
[0136] FIG. 6 illustrates an operational flow / algorithmic structure 600 in accordance with some embodiments. The operation flow / algorithmic structure 600 may be performed or implemented by a base station such as, for example, the base station 108 or the base station 800; or components thereof, for example, baseband processor circuitry 804A.
[0137] The operation flow / algorithmic structure 600 may include, at 610, generating a configuration. The configuration to be transmitted to the UE 104. The configuration may indicate an AI feature. For example, the configuration may be included in an RRC Reconfiguration message.
[0138] In some embodiments, the operation flow 600 may include a configuration associated with UAI. For example, the configuration may enable the UE 104 with applicability reporting via UAI signaling.
[0139] The operation flow / algorithmic structure 600 may include, at 620, processing an applicability report. The applicability report may be received via RRC Reconfiguration Complete message or via a UAI message. The applicability report may include an indication of positive applicability or negative applicability associated with an AI feature.
[0140] When the applicability report indicates negative applicability of an AI feature, the applicability report may also include an indication of one or more causes or conditions associated with the negative applicability of the AI feature. For example, the applicability report may indicate the negative applicability is due to one or more of: 1) network-side condition (e.g., the trained model not matching the network-side conditions) , 2) UE-side condition (e.g., the trained model not matching current UE-side conditions) , or 3) model availability condition (e.g., the trained model not being available at the device) .
[0141] In some embodiments, the configuration may include a partial configuration of the AI feature. In response to the applicability report, the operation flow 600 may include generating or transmitting the remaining configuration of the AI feature to the UE 104.
[0142] In some embodiments, the operation flow 600 may include generating or transmitting a message indicating an activation, deactivation, or de-configuration of an AI feature. The message may be generated or transmitted based on a received applicability report, or based on determining a change in the network-side or UE-side conditions.
[0143] In some embodiments, the operation flow 600 may include receiving or processing a message indicating an activation or deactivation of an AI feature. For example, for each configured AI feature, the applicability report may indicate whether that AI feature is activated or deactivated.
[0144] FIG. 7 illustrates a UE 700 in accordance with some embodiments. The UE 700 may be similar to and substantially interchangeable with the UE 104.
[0145] The UE 700 may be any mobile or non-mobile computing device, such as, for example, mobile phones, computers, tablets, industrial wireless sensors (for example, microphones, carbon dioxide sensors, pressure sensors, humidity sensors, thermometers, motion sensors, accelerometers, laser scanners, fluid level sensors, inventory sensors, electric voltage / current meters, or actuators) , video surveillance / monitoring devices (for example, cameras or video cameras) , wearable devices (for example, a smartwatch) , or Intemet-of-things devices.
[0146] The UE 700 may include processors 704, RF interface circuitry 708, memory / storage 712, user interface 716, sensors 720, driver circuitry 722, power management integrated circuit (PMIC) 724, antenna 726, and battery 728. The components of the UE 700 may be implemented as integrated circuits (ICs) , portions thereof, discrete electronic devices, or other modules, logic, hardware, software, firmware, or a combination thereof. The block diagram of FIG. 7 is intended to show a high-level view of some of the components of the UE 700. However, some of the components shown may be omitted, additional components may be present, and different arrangements of the components shown may occur in other implementations.
[0147] The components of the UE 700 may be coupled with various other components over one or more interconnects 732, which may represent any type of interface, input / output, bus (local, system, or expansion) , transmission line, trace, or optical connection that allows various circuit components (on common or different chips or chipsets) to interact with one another.
[0148] The processors 704 may include processor circuitry such as, for example, baseband processor circuitry (BB) 704A, central processor unit circuitry (CPU) 704B, and graphics processor unit circuitry (GPU) 704C. The processors 704 may include any type of circuitry or processor circuitry that executes or otherwise operates computer-executable instructions, such as program code, software modules, or functional processes from memory / storage 712 to cause the UE 700 to perform operations as described herein. The processors 704 may also include interface circuitry 704D to communicatively couple the processor circuitry with one or more other components of the UE 700.
[0149] In some embodiments, the baseband processor circuitry 704A may access a communication protocol stack 736 in the memory / storage 712 to communicate over a 3GPP-compatible network. In general, the baseband processor circuitry 704A may access the communication protocol stack 736 to: perform user plane functions at a PHY layer, MAC layer, RLC layer, PDCP layer, SDAP layer, and PDU layer; and perform control plane functions at a PHY layer, MAC layer, RLC layer, PDCP layer, RRC layer, and a NAS layer. In some embodiments, the PHY layer operations may additionally / alternatively be performed by the components of the RF interface circuitry 708.
[0150] The baseband processor circuitry 704A may generate or process baseband signals or waveforms that carry information in 3GPP-compatible networks. In some embodiments, the waveforms for NR may be based on cyclic prefix OFDM (CP-OFDM) in the uplink or downlink, and discrete Fourier transform spread OFDM (DFT-S-OFDM) in the uplink.
[0151] The memory / storage 712 may include one or more non-transitory, computer-readable media that includes instructions (for example, communication protocol stack 736) that may be executed by one or more of the processors 704 to cause the UE 700 to perform various operations described herein.
[0152] The memory / storage 712 includes any type of volatile or non-volatile memory that may be distributed throughout the UE 700. In some embodiments, some of the memory / storage 712 may be located on the processors 704 themselves (for example, memory / storage 712 may be part of a chipset that corresponds to the baseband processor circuitry 704A) , while other memory / storage 712 is external to the processors 704 but accessible thereto via a memory interface. The memory / storage 712 may include any suitable volatile or non-volatile memory such as, but not limited to, dynamic random access memory (DRAM) , static random access memory (SRAM) , erasable programmable read only memory (EPROM) , electrically erasable programmable read only memory (EEPROM) , Flash memory, solid-state memory, or any other type of memory device technology.
[0153] The RF interface circuitry 708 may include transceiver circuitry and a radio frequency front module (RFEM) that allows the UE 700 to communicate with other devices over a radio access network. The RF interface circuitry 708 may include various elements arranged in transmit or receive paths. These elements may include, for example, switches, mixers, amplifiers, filters, synthesizer circuitry, and control circuitry.
[0154] In the receive path, the RFEM may receive a radiated signal from an air interface via antenna 726 and proceed to filter and amplify (with a low-noise amplifier) the signal. The signal may be provided to a receiver of the transceiver that down-converts the RF signal into a baseband signal that is provided to the baseband processor of the processors 704.
[0155] In the transmit path, the transmitter of the transceiver up-converts the baseband signal received from the baseband processor and provides the RF signal to the RFEM. The RFEM may amplify the RF signal through a power amplifier prior to the signal being radiated across the air interface via the antenna 726.
[0156] In various embodiments, the RF interface circuitry 708 may be configured to transmit / receive signals in a manner compatible with NR access technologies.
[0157] The antenna 726 may include antenna elements to convert electrical signals into radio waves to travel through the air and to convert received radio waves into electrical signals. The antenna elements may be arranged into one or more antenna panels. The antenna 726 may have antenna panels that are omnidirectional, directional, or a combination thereof to enable beamforming and multiple input, multiple output communications. The antenna 726 may include microstrip antennas, printed antennas fabricated on the surface of one or more printed circuit boards, patch antennas, or phased array antennas. The antenna 726 may have one or more panels designed for specific frequency bands including bands in FR1 or FR2.
[0158] The user interface 716 includes various input / output (I / O) devices designed to enable user interaction with the UE 700. The user interface 716 includes input device circuitry and output device circuitry. Input device circuitry includes any physical or virtual means for accepting an input including, inter alia, one or more physical or virtual buttons (for example, a reset button) , a physical keyboard, keypad, mouse, touchpad, touchscreen, microphones, scanner, headset, or the like. The output device circuitry includes any physical or virtual means for showing information or otherwise conveying information, such as sensor readings, actuator position (s) , or other like information. Output device circuitry may include any number or combinations of audio or visual display, including, inter alia, one or more simple visual outputs / indicators (for example, binary status indicators such as light emitting diodes (LEDs) and multi-character visual outputs, or more complex outputs such as display devices or touchscreens (for example, liquid crystal displays (LCDs) , LED displays, quantum dot displays, and projectors) , with the output of characters, graphics, multimedia objects, and the like being generated or produced from the operation of the UE 700.
[0159] The sensors 720 may include devices, modules, or subsystems whose purpose is to detect events or changes in their environment and send the information (sensor data) about the detected events to some other device, module, or subsystem. Examples of such sensors include inertia measurement units comprising accelerometers, gyroscopes, or magnetometers; microelectromechanical systems or nanoelectromechanical systems comprising 3-axis accelerometers, 3-axis gyroscopes, or magnetometers; level sensors; flow sensors; temperature sensors (for example, thermistors) ; pressure sensors; barometric pressure sensors; gravimeters; altimeters; image capture devices (for example, cameras or lensless apertures) ; light detection and ranging sensors; proximity sensors (for example, infrared radiation detector and the like) ; depth sensors; ambient light sensors; ultrasonic transceivers; and microphones or other like audio capture devices.
[0160] The driver circuitry 722 may include software and hardware elements that operate to control particular devices that are embedded in the UE 700, attached to the UE 700, or otherwise communicatively coupled with the UE 700. The driver circuitry 722 may include individual drivers allowing other components to interact with or control various input / output (I / O) devices that may be present within or connected to the UE 700. For example, driver circuitry 722 may include a display driver to control and allow access to a display device, a touchscreen driver to control and allow access to a touchscreen interface, sensor drivers to obtain sensor readings of sensors 720, and control and allow access to sensors 720, drivers to obtain actuator positions of electro-mechanic components or control and allow access to the electro-mechanic components, a camera driver to control and allow access to an embedded image capture device, audio drivers to control and allow access to one or more audio devices.
[0161] The PMIC 724 may manage power provided to various components of the UE 700. In particular, with respect to the processors 704, the PMIC 724 may control power-source selection, voltage scaling, battery charging, or DC-to-DC conversion.
[0162] A battery 728 may power the UE 700, although in some examples, the UE 700 may be mounted deployed in a fixed location and may have a power supply coupled to an electrical grid. The battery 728 may be a lithium-ion battery, a metal-air battery, such as a zinc-air battery, an aluminum-air battery, a lithium-air battery, and the like. In some implementations, such as in vehicle-based applications, the battery 728 may be a typical lead-acid automotive battery.
[0163] FIG. 8 illustrates a network device 800 in accordance with some embodiments. The network device 800 may be similar to and substantially interchangeable with base station 108.
[0164] The network device 800 may include processors 804, RF interface circuitry 808 (if implemented as a base station) , core network (CN) interface circuitry 814, memory / storage circuitry 812, and antenna structure 826.
[0165] The components of the network device 800 may be coupled with various other components over one or more interconnects 828.
[0166] The processors 804, RF interface circuitry 808, memory / storage circuitry 812 (including communication protocol stack 810) , antenna structure 826, and interconnects 828 may be similar to like-named elements shown and described with respect to FIG. 7.
[0167] The processors 804 may include processor circuitry such as, for example, baseband processor circuitry (BB) 804A, central processor unit circuitry (CPU) 804B, and graphics processor unit circuitry (GPU) 804C. The processors 804 may include any type of circuitry or processor circuitry that executes or otherwise operates computer-executable instructions, such as program code, software modules, or functional processes from memory / storage circuitry 812 to cause the UE 700 to perform operations as described herein. The processors 804 may also include interface circuitry 804D to communicatively couple the processor circuitry with one or more other components of the network device 800.
[0168] The CN interface circuitry 814 may provide connectivity to a core network, for example, a 5th Generation Core network (5GC) using a 5GC-compatible network interface protocol such as carrier Ethernet protocols or some other suitable protocol. Network connectivity may be provided to / from the network device 800 via a fiber optic or wireless backhaul. The CN interface circuitry 814 may include one or more dedicated processors or FPGAs to communicate using one or more of the aforementioned protocols. In some implementations, the CN interface circuitry 814 may include multiple controllers to provide connectivity to other networks using the same or different protocols.
[0169] It is well understood that the use of personally identifiable information should follow privacy policies and practices generally recognized as meeting or exceeding industry or governmental requirements for maintaining users’ privacy. In particular, personally identifiable information data should be managed and handled so as to minimize risks of unintentional or unauthorized access or use, and the nature of authorized use should be clearly indicated to users.
[0170] For one or more embodiments, at least one of the components set forth in one or more of the preceding figures may be configured to perform one or more operations, techniques, processes, or methods as set forth in the example section below. For example, the baseband circuitry described above in connection with one or more of the preceding figures may be configured to operate according to one or more of the examples set forth below. For another example, circuitry associated with a UE, base station, or network element described above in connection with one or more of the preceding figures may be configured to operate according to one or more of the examples set forth below in the example section.
[0171] Some embodiments described herein can include use of leaming and / or non-learning-based process (es) . The use can include collecting, pre-processing, encoding, labeling, organizing, analyzing, recommending and / or generating data. Entities that collect, share, and / or otherwise utilize user data should provide transparency and / or obtain user consent when collecting such data. The present disclosure recognizes that the use of the data in the AI / ML function processes can be used to benefit users.
[0172] For example, the data can be used to train models that can be deployed to improve performance, accuracy, and / or functionality of applications and / or services. Accordingly, the use of the data enables the AI / ML function processes to adapt and / or optimize operations to provide more personalized, efficient, and / or enhanced user experiences. Such adaptation and / or optimization can include tailoring content, recommendations, and / or interactions to individual users, as well as streamlining processes, and / or enabling more intuitive interfaces. Further beneficial uses of the data in the AI / ML function processes are also contemplated by the present disclosure.
[0173] The present disclosure contemplates that, in some embodiments, data used by AI / ML function processes includes publicly available data. To protect user privacy, data may be anonymized, aggregated, and / or otherwise processed to remove or to the degree possible limit any individual identification. As discussed herein, entities that collect, share, and / or otherwise utilize such data should obtain user consent prior to and / or provide transparency when collecting such data. Furthermore, the present disclosure contemplates that the entities responsible for the use of data, including, but not limited to data used in association with AI / ML function processes, should attempt to comply with well-established privacy policies and / or privacy practices.EXAMPLES
[0174] In the following sections, further exemplary embodiments are provided.
[0175] Example 1 includes a method including: processing a configuration indicating an artificial intelligence (AI) feature; determining a condition associated with the AI feature; determining based on the condition, a negative applicability of the AI feature; and generating an applicability report indicating the negative applicability of the feature and the condition.
[0176] Example 2 includes the method of example 1 or some other examples herein, wherein the condition is a network condition causing the AI feature not applicable, the network condition includes a mismatch between a first identifier and a second identifier.
[0177] Example 3 includes the method of examples 1 or 2 or some other examples herein, wherein the condition is a user equipment (UE) condition causing the AI feature not applicable, the UE condition includes an antenna placement, or a speed of the UE.
[0178] Example 4 includes the method of any of examples 1-3 or some other examples herein, wherein the condition is a model availability condition causing the AI feature not applicable, the model availability condition includes a model that is not fully downloaded by a user equipment (UE) .
[0179] Example 5 includes the method of any of examples 1-4 or some other examples herein, wherein the applicability report is included in a radio resource control (RRC) configuration complete message or a user equipment assistance information (UAI) message.
[0180] Example 6 includes the method of any of examples 1-5 or some other examples herein, wherein the applicability report includes a duration associated with the negative applicability of the feature.
[0181] Example 7 includes the method of any of examples 1-6 or some other examples herein, wherein the configuration is a partial configuration of the AI feature, and the method further includes: generating a radio resource control (RRC) reconfiguration complete message indicating a positive applicability of the AI feature; and processing a remaining configuration of the AI feature.
[0182] Example 8 includes the method of any of examples 1-7 or some other examples herein, wherein the configuration is a first configuration, the condition is associated with a user equipment or a model, the applicability report is included in a radio resource control (RRC) reconfiguration complete message, and the method further includes: processing a second configuration to enable user equipment assistance information (UAI) reporting; processing information associated with a model of the AI feature received from a server; generating a UAI report to indicate that the model is ready; and generating a report including an inference associated with the AI feature.
[0183] Example 9 includes the method of any of examples 1-8 or some other examples herein, wherein the applicability report indicates a mismatch between a model associated with the AI feature and a network condition, a mismatch between a model associated with the AI feature and a user equipment condition, or a duration associated with availability of a model.
[0184] Example 10 includes the method of any of examples 1-9 or some other examples herein, wherein the condition is associated with a network, and the method further includes: generating a user equipment assistance information (UAI) report indicating that a data collection configuration associated with the AI feature is received.
[0185] Example 11 includes the method of any of examples 1-10 or some other examples herein, wherein the condition is associated with a user equipment (UE) , and the method further includes: generating a UE assistance information (UAI) report indicating insufficient memory or insufficient battery at the UE.
[0186] Example 12 includes the method of any of examples 1-11 or some other examples herein, wherein the condition is associated with a model associated with the AI feature, and the method further includes: generating a user equipment assistance information (UAI) report indicating a duration for downloading the model from a server, or an indication that the model is ready for an inference associated with the AI feature; and processing a message to activate the AI feature or the model.
[0187] Example 13 includes the method of any of examples 1-12 or some other examples herein, wherein configuration is a first configuration, and the method further includes: processing a second configuration of a report corresponding to an inference associated with the AI feature.
[0188] Example 14 includes the method of any of examples 1-13 or some other examples herein, wherein the report is an aperiodic report, and the method further includes: processing a downlink control information or a medium access control (MAC) control element (CE) to activate the AI feature.
[0189] Example 15 includes the method of any of examples 1-14 or some other examples herein, wherein the report is a periodic report, and the method further includes: processing a radio resource control (RRC) message to activate the AI feature.
[0190] Example 16 includes the method of any of examples 1-15 or some other examples herein, wherein the report is a channel state information (CSI) report.
[0191] Example 17 includes the method of any of examples 1-16 or some other examples herein, wherein configuration indicates using user equipment assistance information (UAI) for the applicability report.
[0192] Example 18 includes the method of any of examples 1-17 or some other examples herein, wherein the applicability report may indicate whether the AI feature or a model associated with the AI feature is activated.
[0193] Example 19 includes the method of any of examples 1-18 or some other examples herein, further includes: processing a reconfiguration to de-configure the AI feature.
[0194] Example 20 includes a method including: generating a configuration for transmission to a user equipment (UE) , the configuration indicating an artificial intelligence (AI) feature; and processing an applicability report received from the UE, the applicability report indicating a negative applicability of the feature and a condition associated with the negative applicability.
[0195] Example 21 includes the method of example 20 or some other examples herein, wherein the condition is a network condition causing the AI feature not applicable, the network condition includes a mismatch between a first identifier and a second identifier.
[0196] Example 22 includes the method of examples 20 or 21 or some other examples herein, wherein the condition is a user equipment (UE) condition causing the AI feature not applicable, the UE condition includes an antenna placement, or a speed of the UE.
[0197] Example 23 includes the method of any of examples 20-22 or some other examples herein, wherein the condition is a model availability condition causing the AI feature not applicable, the model availability condition includes a model that is not fully downloaded by a user equipment (UE) .
[0198] Example 24 includes the method of any of examples 20-23 or some other examples herein, wherein the applicability report is included in a radio resource control (RRC) configuration complete message or a user equipment assistance information (UAI) message.
[0199] Example 25 includes the method of any of examples 20-24 or some other examples herein, wherein the applicability report includes a duration associated with the negative applicability of the feature.
[0200] Example 26 includes the method of any of examples 20-25 or some other examples herein, wherein the configuration is a partial configuration of the AI feature, and the method further includes: processing a radio resource control (RRC) reconfiguration complete message indicating a positive applicability of the AI feature; and generating a remaining configuration of the AI feature for transmission to the UE.
[0201] Example 27 includes the method of any of examples 20-25 or some other examples herein, wherein the configuration is a first configuration, the condition is associated with a user equipment or a model, the applicability report is included in a radio resource control (RRC) reconfiguration complete message, and the method further includes: generating a second configuration to enable user equipment assistance information (UAI) reporting; processing a UAI report indicating that the model is ready; and processing a report including an inference associated with the AI feature.
[0202] Example 28 includes the method of any of examples 20-27 or some other examples herein, wherein the applicability report indicates a mismatch between a model associated with the AI feature and a network condition, a mismatch between a model associated with the AI feature and a user equipment condition, or a duration associated with availability of a model.
[0203] Example 29 includes the method of any of examples 20-28 or some other examples herein, wherein the condition is associated with a network, and the method further includes: processing a user equipment assistance information (UAI) report indicating that a data collection configuration associated with the AI feature is received.
[0204] Example 30 includes the method of any of examples 20-29 or some other examples herein, wherein the condition is associated with a user equipment (UE) , and the method further includes: processing a UE assistance information (UAI) report indicating insufficient memory or insufficient battery at the UE.
[0205] Example 31 includes the method of any of examples 20-30 or some other examples herein, wherein the condition is associated with a model associated with the AI feature, and the method further includes: processing a user equipment assistance information (UAI) report indicating a duration for downloading the model from a server, or an indication that the model is ready for an inference associated with the AI feature; and generating a message to activate the AI feature or the model.
[0206] Example 32 includes the method of any of examples 20-31 or some other examples herein, wherein configuration is a first configuration, and the method further includes: generating a second configuration of a report corresponding to an inference associated with the AI feature.
[0207] Example 33 includes the method of any of examples 20-32 or some other examples herein, wherein the report is an aperiodic report, and the method further includes: generating a downlink control information or a medium access control (MAC) control element (CE) to activate the AI feature.
[0208] Example 34 includes the method of any of examples 20-33 or some other examples herein, wherein the report is a periodic report, and the method further includes: generating a radio resource control (RRC) message to activate the AI feature.
[0209] Example 35 includes the method of any of examples 20-34 or some other examples herein, wherein the report is a channel state information (CSI) report.
[0210] Example 36 includes the method of any of examples 20-35 or some other examples herein, wherein configuration indicates using user equipment assistance information (UAI) for the applicability report.
[0211] Example 37 includes the method of any of examples 20-36 or some other examples herein, wherein the applicability report may indicate whether the AI feature or a model associated with the AI feature is activated.
[0212] Example 38 includes the method of any of examples 20-37 or some other examples herein, further includes: generating a reconfiguration to de-configure the AI feature.
[0213] Another example may include an apparatus comprising means to perform one or more elements of a method described in or related to any of examples 1-38, or any other method or process described herein.
[0214] Another example may include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of a method described in or related to any of examples 1-38, or any other method or process described herein.
[0215] Another example may include an apparatus comprising logic, modules, or circuitry to perform one or more elements of a method described in or related to any of examples 1-38, or any other method or process described herein.
[0216] Another example may include a method, technique, or process as described in or related to any of examples 1-38, or portions or parts thereof.
[0217] Another example may include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform the method, techniques, or process as described in or related to any of examples 1-38, or portions thereof.
[0218] Another example may include a signal as described in or related to any of examples 1-38, or portions or parts thereof.
[0219] Another example may include a datagram, information element, packet, frame, segment, PDU, or message as described in or related to any of examples 1-38, or portions or parts thereof, or otherwise described in the present disclosure.
[0220] Another example may include a signal encoded with data as described in or related to any of examples 1-38, or portions or parts thereof, or otherwise described in the present disclosure.
[0221] Another example may include a signal encoded with a datagram, IE, packet, frame, segment, PDU, or message as described in or related to any of examples 1-38, or portions or parts thereof, or otherwise described in the present disclosure.
[0222] Another example may include an electromagnetic signal carrying computer-readable instructions, wherein execution of the computer-readable instructions by one or more processors is to cause the one or more processors to perform the method, techniques, or process as described in or related to any of examples 1-38, or portions thereof.
[0223] Another example may include a computer program comprising instructions, wherein execution of the program by a processing element is to cause the processing element to carry out the method, techniques, or process as described in or related to any of examples 1-38, or portions thereof.
[0224] Another example may include a signal in a wireless network as shown and described herein.
[0225] Another example may include a method of communicating in a wireless network, as shown and described herein.
[0226] Another example may include a system for providing wireless communication, as shown and described herein.
[0227] Another example may include a device for providing wireless communication, as shown and described herein.
[0228] Unless explicitly stated otherwise, any of the above-described examples may be combined with any other example (or combination of examples) . The foregoing description of one or more implementations provides illustration and description but is not intended to be exhaustive or to limit the scope of embodiments to the precise form disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from the practice of various embodiments.
[0229] Although the embodiments above have been described in considerable detail, numerous variations and modifications will become apparent to those skilled in the art once the above disclosure is fully appreciated. It is intended that the following claims be interpreted to embrace all such variations and modifications.
Claims
1.A method comprising:processing a configuration indicating an artificial intelligence (AI) feature;determining a condition associated with the AI feature;determining based on the condition, a negative applicability of the AI feature; andgenerating an applicability report indicating the negative applicability of the feature and the condition.2.The method of claim 1, wherein the condition is:a network condition causing the AI feature not to be applicable, the network condition includes a mismatch between a first identifier and a second identifier;a user equipment (UE) condition causing the AI feature not to be applicable, the UE condition includes an antenna placement, or a speed of the UE; ora model availability condition causing the AI feature not to be applicable, the model availability condition includes a model that is not fully downloaded by a user equipment (UE) .3.The method of claim 1 or 2, wherein the applicability report is included in a radio resource control (RRC) configuration complete message or a user equipment assistance information (UAI) message.4.The method of claim 1 or 2, wherein the applicability report includes a duration associated with the negative applicability of the feature.5.The method of claim 1 or 2, wherein the configuration is a partial configuration of the AI feature, and the method further comprises:generating a radio resource control (RRC) reconfiguration complete message indicating a positive applicability of the AI feature; andprocessing a remaining configuration of the AI feature.6.The method of claim 1 or 2, wherein the configuration is a first configuration, the condition is associated with a user equipment or a model, the applicability report is included in a radio resource control (RRC) reconfiguration complete message, and the method further comprises:processing a second configuration to enable user equipment assistance information (UAI) reporting;processing information associated with a model of the AI feature received from a server;generating a UAI report to indicate that the model is ready; andgenerating a report including an inference associated with the AI feature.7.The method of claim 1 or 2, wherein the applicability report indicates a mismatch between a model associated with the AI feature and a network condition, a mismatch between a model associated with the AI feature and a user equipment condition, or a duration associated with availability of a model.8.The method of claim 7, wherein:the condition is associated with a network, and the method further comprises generating a user equipment assistance information (UAI) report indicating that an inference configuration associated with the AI feature is received;the condition is associated with a user equipment (UE) , and the method further comprises generating a UE assistance information (UAI) report indicating insufficient memory or insufficient battery at the UE; orthe condition is associated with a model associated with the AI feature, and the method further comprises generating a user equipment assistance information (UAI) report indicating a duration for downloading the model from a server, or an indication that the model is ready for an inference associated with the AI feature; and processing a message to activate the AI feature or the model.9.The method of claim 1 or 2, wherein configuration is a first configuration, and the method further comprises:processing a second configuration of a report corresponding to an inference associated with the AI feature.10.The method of claim 9, wherein the report is an aperiodic report, and the method further comprises:processing a downlink control information or a medium access control (MAC) control element (CE) to activate the AI feature.11.The method of claim 9, wherein the report is a periodic report, and the method further comprises:processing a radio resource control (RRC) message to activate the AI feature.12.The method of claim 1 or 2, wherein configuration indicates using user equipment assistance information (UAI) for the applicability report.13.The method of claim 1 or 2, wherein the applicability report may indicate whether the AI feature or a model associated with the AI feature is activated.14.The method of claim 1 or 2, further comprises:processing a reconfiguration to de-configure the AI feature.15.A method comprising:generating a configuration for transmission to a user equipment (UE) , the configuration indicating an artificial intelligence (AI) feature; andprocessing an applicability report received from the UE, the applicability report indicating a negative applicability of the feature and a condition associated with the negative applicability.16.The method of claim 15, wherein the condition is:a network condition causing the AI feature not to be applicable, the network condition includes a mismatch between a first identifier and a second identifier;a user equipment (UE) condition causing the AI feature not applicable, the UE condition includes an antenna placement, or a speed of the UE; ora model availability condition causing the AI feature not applicable, the model availability condition includes a model that is not fully downloaded by a user equipment (UE) .17.The method of claim 15 or 16, wherein the applicability report includes a duration associated with the negative applicability of the feature.18.The method of claim 15 or 16, wherein the configuration is a partial configuration of the AI feature, and the method further comprises:processing a radio resource control (RRC) reconfiguration complete message indicating a positive applicability of the AI feature; andgenerating a remaining configuration of the AI feature for transmission to the UE.19.The method of claim 15 or 16, wherein the configuration is a first configuration, the condition is associated with a user equipment or a model, the applicability report is included in a radio resource control (RRC) reconfiguration complete message, and the method further comprises:generating a second configuration to enable user equipment assistance information (UAI) reporting;processing a UAI report indicating that the model is ready; andprocessing a report including an inference associated with the AI feature.20.The method of claim 15 or 16, wherein the condition is associated with a model of the AI feature, and the method further comprises:processing a user equipment assistance information (UAI) report indicating a duration for downloading the model from a server, or an indication that the model is ready for an inference associated with the AI feature; andgenerating a message to activate the AI feature or the model.21.The method of claim 15 or 16, wherein configuration is a first configuration, and the method further comprises:generating a second configuration of a report corresponding to an inference associated with the AI feature.22.The method of claim 21, wherein the report is an aperiodic report, and the method further comprises:generating a downlink control information or a medium access control (MAC) control element (CE) to activate the AI feature.23.The method of claim 21, wherein the report is a periodic report, and the method further comprises:generating a radio resource control (RRC) message to activate the AI feature.24.The method of claim 15 or 16, wherein configuration indicates using user equipment assistance information (UAI) for the applicability report.25.The method of claim 15 or 16, wherein the applicability report indicates whether the AI feature or a model associated with the AI feature is activated, and the method further comprises:generating a reconfiguration to de-configure the AI feature.