Method and wireless communication system for GNB-UE bilateral control of artificial intelligence / machine learning model

Abstract

A method for user equipment (UE) device receiving artificial intelligence (AI) and / or machine learning (ML) model state information from at least one base station (gNB) within a wireless communication network for gNB-UE bilateral model control, in which, upon receipt of the artificial intelligence (AI) and / or machine learning (ML) model state information, the user equipment (UE) device receives the state information from the at least one base station (gNB) within the wireless communication network. The user equipment (UE) makes a determination that if artificial intelligence (AI) and / or machine learning (ML) model state information is activated, the user equipment (UE) maintains the same artificial intelligence (AI) and / or machine learning (ML) model state information; if the artificial intelligence (AI) and / or machine learning (ML) model state information is deactivated, the user equipment (UE) switches to an alternative mode of operation based on assistance information from a base station (gnB).

Description

Technical Field

[0001] This disclosure relates to gNb UE bilateral model control. Background Art

[0002] The 3rd Generation Partnership Project (3GPP) has standardized coverage enhancement (CE) for the Internet of Things (IoT) to connect devices to the cellular network under challenging radio conditions. CE is based on the principle of extended transmission time, which takes advantage of the fact that many IoT applications have relaxed requirements for data rate and latency, and for such applications, the coverage can be significantly improved through repeated transmissions. However, CE consumes a large amount of radio resources and should be carefully implemented for different applications.

[0003] The evolution of wireless communication technology to the 5th generation (5G) enables everything to be connected via the Internet. For example, a large amount of information related to human activities has been recorded and monitored through various types of IoT applications (e.g., health monitoring, smart home, intelligent transportation, industrial automation) and exchanged via the cellular network. Unfortunately, however, traditional cellular networks are not designed to accommodate such diverse IoT applications. Accordingly, significant efforts have been made to support emerging IoT scenarios in the cellular network, namely, massive machine-type communication (mMTC) or massive IoT (mIoT), which is one of the main use cases of 5G.

[0004] In the cellular network, an extremely large number of IoT devices are expected to be deployed, and by 2030, the number of connected devices will reach 500 billion. Each IoT device sporadically generates small packets to report sensed information to the IoT server via a base station (BS / gNB). In particular, the IoT device remains disconnected from the BS to reduce power consumption due to sporadic packet generation. This means that whenever a data packet is transmitted to the IoT server, each IoT device should perform a random access (RA) procedure to establish a connection with the BS. The RA procedure adopted in existing cellular systems such as LTE / LTE-A / 5G consists of a four-step handshake procedure. Since IoT devices are densely deployed in the cellular IoT network, simultaneous RA attempts in a certain RA time slot (or equivalently, the physical RA channel (PRACH)) may cause collision problems. The collision problems largely result in poor access performance on the device side (i.e., RA failure). Specifically, IoT devices may spend a large amount of time accessing the network, so the network cannot guarantee an acceptable end-to-end delay according to their access priorities.

[0005] In a cellular system, the connection between each IoT device and the BS is a prerequisite for data communication. To establish a connection, the device should perform a 4-step RA procedure. To sum up, the general description of the traditional RA procedure in cellular networks (e.g., LTE / LTE-A / 5G) is as follows:

[0006] · Step 1. Preamble transmission: Each IoT device randomly selects a single RA preamble from a set of available RA preambles and transmits it on the PRACH.

[0007] · Step 2. Random access response: The BS detects which preambles are active. In response to the detected preambles, the BS transmits random access response (RAR) messages, each message consisting of a RA preamble identifier (RAPID), a timing alignment (TA), an uplink grant (UG), and a temporary identifier. Each IoT device that transmitted a preamble in the first step waits for a RAR message containing the same RAPID. If there is a corresponding RAR message, each device uses the information within the message for subsequent steps (i.e., Step 3).

[0008] · Step 3. Scheduled transmission: Each IoT device transmits its scheduling message (e.g., connection request message) on the uplink resources assigned on the physical uplink shared channel (PUSCH), where the assigned uplink resources are indicated by the UG value contained in the RAR message received in the second step. To determine whether a resource conflict occurs on the used uplink resources, once the message of Step 3 is transmitted, each IoT device starts a contention resolution (CR) timer.

[0009] · Step 4. Acknowledgment: The BS back-transmits the identifier of the IoT device whose transmitted scheduling message is successfully decoded without any resource conflict. If each IoT device receives a correct acknowledgment (ACK) message before the CR timer expires, it considers the RA attempt as successful. Otherwise, it considers the RA attempt as failed and re-attempts the RA procedure in the next available RA slot after performing backoff.

[0010] US2019 / 0156247 describes techniques for performing accuracy-based dynamic experimentation and deployment of machine learning (ML) models. The inference traffic flowing to the ML model and the accuracy of the model are analyzed and used to ensure that better-performing models are executed more frequently via model selection. A prediction component can evaluate which model is more likely to be accurate for certain input data elements. Ensemble techniques can combine the inference results of multiple ML models in order to achieve an overall result better than any single model alone.

[0011] US2019 / 0095756 describes techniques for selecting a machine learning algorithm based on the performance prediction of a trained algorithm-specific regressor. In an embodiment, a computer derives meta-feature values from an inference dataset by deriving a corresponding meta-feature value from the inference dataset for each meta-feature. For each trainable algorithm and each regression meta-model respectively associated with the algorithm, a corresponding score is calculated by invoking the meta-model based on at least one of the following: a corresponding subset of meta-feature values, and / or hyperparameter values of a corresponding subset of hyperparameters of the algorithm. The (multiple) algorithms are selected based on their respective scores. Based on the inference dataset, the selected (multiple) algorithms can be invoked to obtain results. In an embodiment, the trained regressor is a uniquely configured artificial neural network. In an embodiment, the trained regressor is included in an algorithm-specific set. Techniques for optimally training the regressor and / or the set are also provided.

[0012] US2021 / 0328630 describes methods, systems, and devices for wireless communication, where a base station can develop many different prediction models for each of many different functions. The different functions can be used to determine various beamforming parameters for beamforming communication between a user equipment (UE) and the base station. The base station can provide the models to the UE, and then the UE can use these models to determine the values of one or more beamforming parameters. There may be multiple different models for the same function (e.g., a beam prediction function for identifying transmit / receive beams for communication), which can be provided by the base station to the UE and can be used based on specific channel conditions or the location of the UE. The UE or the base station can select which one of the multiple prediction models to use for communication.

[0013] WO 2020 / 245639 discloses a system for predicting faults in a cloud computing environment. The system includes a model selection module configured to: use offline data of the cloud computing environment to train multiple candidate models; and select one or more trained candidate models at least partially based on the respective training results of each trained candidate model. An online predictor is configured to predict faults in the cloud computing environment based on the selected trained candidate models and online data of the cloud computing environment.

[0014] KR 20200141835 discloses a device for generating a model that learns a dataset in a distributed environment. The device includes: a model generation unit that generates a local model for learning a local dataset; a model selection unit that selects a previously learned and stored global model; and a model association unit that associates the local model with the global model to generate a federated model. Thus, the performance of the learning model can be improved through optimal model generation.

[0015] US2019220758 describes a distributed system for training a classifier. The system includes machine learning (ML) workers and a parameter server (PS). The PS is configured for parallel processing to provide a model to each ML worker, receive model updates from each ML worker, and iteratively update the model using each model update. The PS includes: a gradient dataset associated with the corresponding ML worker for storing a model update identifier (delta-M-ID) indicating the computed model update and the corresponding model update; a global dataset that stores the delta-M-ID, the identifier of the ML worker (ML-worker-ID) that computed the model update, and a model version that marks the new model in the PS, where the new model is computed by merging the model update with the previous model in the PS; and a model download dataset that stores the ML-worker-ID and model version of each transmitted model.

[0016] US2022083911 describes a user equipment. The user equipment includes a processor configured to: download a main machine learning model to generate user recommendations related to the use of an application of the user equipment, use the main machine learning model and data related to one or more of the user of the user equipment or the interaction between the user and the user equipment to compute a model update of the main machine learning model, encode the computed model update using an ε-differential privacy mechanism, and transmit the ε-differentially private encoded model update.

[0017] US2022012601 describes a federated learning server and method. The federated learning server is configured to aggregate a plurality of received model updates to update a main machine learning model. Once a predefined threshold or interval of the received model updates is reached, a set of current hyperparameter values and corresponding validation set performance metrics obtained from the updated main machine learning model are sent to a hyperparameter optimization model. The optimization model uses the pairwise history of the hyperparameter values and corresponding performance metrics to infer the next set of optimal hyperparameters. The inferred hyperparameter values are sent to the federated learning server, which updates the main machine learning model with the updated set of hyperparameter values and redistributes the updated main machine learning model with the updated set of hyperparameter values. Depending on the application, hyperparameter optimization in the federated learning mode can be implemented to provide accurate personalized recommendations.

[0018] US2022198337 describes an information processing system. The information processing system obtains a training data set including input data and a label (the label being the true data of the input data), thereby training a machine learning model on the training data set; inputs test data into the machine learning model trained on the training data set; evaluates whether the performance of the machine learning model meets a predetermined condition based on the output of the machine learning model into which the test data is input; updates the training data set when it is evaluated that the performance of the machine learning model does not meet the predetermined condition; and retrains the machine learning model on the updated training data set. The information processing system repeats updating, retraining, and evaluating the data set in response to the evaluation.

[0019] WO 2021032496 describes a method performed by a first network entity in a communication network. The method includes training a model to obtain a local model update, the local model update including an update to the value of one or more parameters of the model, wherein training the model includes inputting training data into a machine learning algorithm. The method further includes applying a serialization function to the local model update to construct a serialized representation of the local model update, thereby removing information indicating the model structure, and transmitting the serialized representation of the local model update to an aggregator entity in the communication network.

[0020] US2022182263 describes an OAM core network. The OAM core network can receive requests for an ML / NN model and features associated with an ML / NN program. The OAM core network can determine the most recent updates to the ML / NN model and features based on the request and generate a response to the request that indicates the most recent updates to the ML / NN model and features. In aspects, a base station can initiate a request for an ML / NN model and features by transmitting a request for the ML / NN model and features to the OAM core network. The base station can receive the response generated by the OAM core network based on the transmitted request. In other aspects, a UE can initiate a request for an ML / NN model and features by transmitting a request to the base station, wherein the UE can receive the ML / NN model and features from the base station based on the transmitted request.

[0021] WO 2021252981 describes techniques for authenticating a user based on a machine learning model, including: receiving user authentication data associated with the user; generating an output from a neural network model based on the user authentication data; determining a distance between the output and an embedding vector associated with the user; comparing the determined distance with a distance threshold; and making an authentication decision based on the comparison.

[0022] The terminology of the work list includes a set of high-level descriptions regarding AI / ML model training, inference, validation, testing, UE-side (AI / ML) models, network-side (AI / ML) models, single-sided (AI / ML) models, bilateral (AI / ML) models, etc. In this application, common abbreviations are used, such as AI: Artificial Intelligence, ML: Machine Learning, and NR: New Radio.

[0023] Since the terminology definitions are still under discussion and subject to further modification, not all signaling aspects to support the above items have been specified yet.

[0024] Any potential impact of new or enhanced mechanisms using the above work list items to support AI / ML models on the standard is one of the key areas of investigation in the AI / ML research project.

[0025] In a bilateral AI / ML model, joint inference is performed between the UE and the gNB. Since the AI / ML models in the gNB and the UE are both executed collaboratively, if the joint inference stops unexpectedly before any pre-configured time or the end of the operation cycle and the gNB model is deactivated, and no UE-specific behavior is specified for such a situation, the UE may have an impact on the model performance.

[0026] For example, the deactivation of the model on the gNB side can be due to the following reasons. The link quality deteriorates, unable to meet the joint inference connection or joint inference operation of the gNB model, and no timers or durations are configured to support the bilateral AI / ML model phase, or the bilateral AI / ML model phase terminates prematurely due to gNB network issues. Summary of the Invention

[0027] Generally speaking, new signaling messages from the gNB and associated UE behavior specifications are proposed, where the UE determines the behavior of the model operation.

[0028] The main objective of this application is to provide a solution to the problems proposed.

[0029] The proposed idea can improve the reliability of the model performance and contribute to the required changes in gNB-UE signaling and UE behavior based on bilateral models.

[0030] The described problems are solved through the embodiments of this application.

[0031] A method for gNB-UE bilateral model control by receiving artificial intelligence (AI) and / or machine learning (ML) model status information from at least one base station (gNB) within a wireless communication network by a user equipment (UE) device is characteristic of a specific embodiment. After receiving the artificial intelligence (AI) and / or machine learning (ML) model status information, the user equipment (UE) makes the following determination: If the artificial intelligence (AI) and / or machine learning (ML) model status information is activated, the user equipment (UE) maintains the same artificial intelligence (AI) and / or machine learning (ML) model status information; if the artificial intelligence (AI) and / or machine learning (ML) model status information is deactivated, the user equipment (UE) switches to an alternative operation mode based on auxiliary information from the base station (gNB).

[0032] According to a specific embodiment, the method for gNB-UE bilateral model control by a user equipment (UE), wherein the artificial intelligence (AI) and / or machine learning (ML) model status information includes a model status indication from the base station (gNB), model transmission information, and a user equipment (UE) behavior index.

[0033] A method for gNB-UE bilateral model control by a base station (gNB) in a wireless communication network is characteristic of a specific embodiment, wherein the method includes:

[0034] Determine artificial intelligence (AI) and / or machine learning (ML) model status information,

[0035] Check the status of the artificial intelligence (AI) and / or machine learning (ML) model status information by the base station (gNB),

[0036] If the result of the check is that the base station (gNB) is not deactivated, the artificial intelligence (AI) and / or machine learning (ML) model status information remains the same artificial intelligence (AI) and / or machine learning (ML) model status information, and the artificial intelligence (AI) and / or machine learning (ML) model status information is used for further procedures,

[0037] If the result of the check is that the artificial intelligence (AI) and / or machine learning (ML) model status information is deactivated, transmit the artificial intelligence (AI) and / or machine learning (ML) model status information to at least one user equipment (UE) via the base station (gNB).

[0038] A particular embodiment features a method for gNB-UE bilateral model control by a base station (gNB) in a wireless communication network, wherein the artificial intelligence (AI) and / or machine learning (ML) model state information includes a model state indication from the base station (gNB), model transmission information, and a user equipment (UE) behavior index.

[0039] A particular embodiment features a method for gNB-UE bilateral model control between at least one user equipment (UE) and a base station (gNB) in a wireless communication network, the method including an artificial intelligence (AI) and / or machine learning (ML) model training phase and an artificial intelligence (AI) and / or machine learning (ML) model inference phase.

[0040] A particular embodiment features a method for gNB-UE bilateral model control between at least one user equipment (UE) and a base station (gNB) in a wireless communication network, the method including an artificial intelligence (AI) and / or machine learning (ML) model training phase and an artificial intelligence (AI) and / or machine learning (ML) model inference phase, wherein the artificial intelligence (AI) and / or machine learning (ML) model inference phase includes the following steps:

[0041] Transmitting a measurement report on artificial intelligence (AI) and / or machine learning (ML) and / or user equipment (UE) capabilities from the user equipment (UE) to the base station (gNB)

[0042] · Transmitting a parameter configuration update on artificial intelligence (AI) and / or machine learning (ML) from the base station (gNB) to at least one user equipment (UE)

[0043] · Performing artificial intelligence (AI) and / or machine learning (ML) model inference by the base station (gNB) and at least one user equipment (UE)

[0044] · Conducting model transmission and / or data exchange of inference outputs between the base station (gNB) and at least one user equipment (UE)

[0045] · Conducting a model performance monitoring feedback exchange between the base station (gNB) and at least one user equipment (UE).

[0046] A particular embodiment features a method for gNB-UE bilateral model control between at least one user equipment (UE) and a base station (gNB) in a wireless communication network, the method including an artificial intelligence (AI) and / or machine learning (ML) model inference phase, the phase including the following steps:

[0047] · Determining model state changes and user equipment (UE) behavior indices

[0048] · Generate model transmission information

[0049] · Create artificial intelligence (AI) and / or machine learning (ML) assisted information, where the artificial intelligence (AI) and / or machine learning (ML) assisted information includes model status indication, model transmission information, and user equipment (UE) behavior index

[0050] · Transmit the artificial intelligence (AI) and / or machine learning (ML) assisted information to at least one user equipment (UE),

[0051] · Based on the artificial intelligence (AI) and / or machine learning (ML) assisted information, the user equipment (UE) switches to an alternative operation mode

[0052] · Transmit a user equipment (UE) behavior update report from the user equipment (UE) to the base station (gNB).

[0053] A particular embodiment is characterized by an apparatus for gNB-UE bilateral model control between a user equipment (UE) and a base station (gNB) in a wireless communication network. The apparatus includes a wireless transceiver and a processor coupled to a memory storing computer program instructions, and the instructions are configured to implement the steps of claims 1 to 7.

[0054] A particular embodiment is characterized by a user equipment (UE) including the above apparatus.

[0055] A particular embodiment is characterized by a base station (gNB) including the above apparatus.

[0056] A particular embodiment of a method is implemented in a base station (gNB) in a wireless communication network, and is characterized by the following steps: receiving a random access (RA) request coverage enhancement (CE) and a notification to a user equipment (UE) in a connected mode,

[0057] The user equipment (UE) decides whether to switch to an alternative network based on overall network congestion, where the decision information is determined by the failure of the maximum random access (RA) procedure reaching the coverage enhancement (CE) level (k). In the case of the failure of the maximum random access (RA) procedure reaching the coverage enhancement (CE) level (k), it is checked whether a user equipment (UE)-initiated handover to the alternative network should be performed. If the result of the check is affirmative, the handover to the alternative network is performed. If the result of the check is negative, the user equipment (UE) remains in the same network and upgrades from the coverage enhancement (CE) level (k) to level (k + 1) for the random access (RA) procedure. Thus, the base station (gNB) configures whether the user equipment (UE) should switch to the alternative network.

[0058] A specific embodiment of a method performed by a base station (gNB) in a wireless communication network, characterized by the following steps:

[0059] Receiving a notification of random access (RA) request coverage enhancement (CE) to a user equipment (UE) in connected mode,

[0060] The user equipment (UE) decides whether to switch to an alternative network based on overall network congestion, wherein the decision information is determined by the failure of the maximum random access (RA) procedure reaching the coverage enhancement (CE) level (k). In the case of the failure of the maximum random access (RA) procedure reaching the coverage enhancement (CE) level (k), it is checked whether the user equipment (UE)-initiated handover to the alternative network should be performed. If the result of this check is affirmative, the handover to the alternative network is performed. If the result of this check is negative, the user equipment (UE) remains in the same network and upgrades from the coverage enhancement (CE) level (k) to level (k + 1) for the random access (RA) procedure. Thus, the base station (gNB) configures whether the user equipment (UE) should switch to the alternative network.

[0061] A specific embodiment is characterized by a wireless communication system for performing gNB-UE bilateral model control from a base station to a user equipment, wherein the base station includes a processor coupled to a memory in which computer program instructions are stored, the instructions being configured to implement the steps of claims 4 to 7. The user equipment (UE) includes a processor coupled to a memory in which computer program instructions are stored, the instructions being configured to implement the steps of claims 1 to 3.

[0062] A specific embodiment is characterized by a wireless communication system for performing energy-saving coverage enhancement. After the user equipment (UE) obtains handover decision information from at least one cell A to at least another cell B based on defined criteria, after performing the following steps, the user equipment (UE)-initiated handover is performed:

[0063] A user equipment (UE) sends a random access (RA) request with coverage enhancement (CE) level (k) to cell A. Cell A sends back the random access (RA) failure with coverage enhancement (CE) level (k) to the user equipment (UE) and waits for the user equipment (UE) to connect with a higher coverage enhancement (CE) level (k). The user equipment (UE) sends a random access (RA) request with coverage enhancement (CE) level (k) to cell B. And if the user equipment (UE) fails to perform random access (RA), cell A buffers the data of the user equipment (UE). Cell B sends back the random access (RA) response with coverage enhancement (CE) level (k) to the user equipment (UE). The user equipment (UE) and cell B connect. Cell B updates the details of the user equipment (UE). If cell A obtains the update about the user equipment (UE) from cell B, cell A transfers the data of the user equipment (UE) to the cell. If cell A does not obtain the update about the user equipment (UE) from other cells, cell A declares a radio link failure (RLF).

[0064] The information medium can be any entity or device capable of storing the program. For example, the medium can include a storage device such as a ROM (e.g., CD ROM or microelectronic circuit ROM), a FLASH memory, or any magnetic recording device (e.g., a hard disk drive).

[0065] Furthermore, the information medium can be a transmissible medium that can be transmitted via radio or by other means via a cable or an optical fiber, such as an electrical signal or an optical signal.

[0066] Alternatively, the information medium can be an integrated circuit containing the program, and the circuit is adapted to execute or for executing the method under discussion.

[0067] The advantages of the apparatus, the user equipment, the wireless system, the computer program, and the information medium are the same as the advantages presented with respect to the corresponding method according to any one of the above embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Other advantages and features of the present invention will become clearer upon reading the following description (given by way of simple illustrative and non - limiting examples) and the drawings, in which:

[0069] Figure 1 The behavior of the gNB is shown,

[0070] Figure 2 The behavior of the user equipment (UE) is shown,

[0071] Figure 3 The bilateral model inference signaling flow is shown,

[0072] Figure 4Shows a user equipment (UE) behavior signaling flow. Detailed implementation

[0073] The detailed implementation described below with reference to the accompanying drawings is intended as a description of various configurations and is not intended to represent the only configuration in which the concepts described herein can be practiced. The detailed implementation includes specific details to provide a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts can be practiced without these specific details. In particular, although terms from 3GPP 5G NR may be used in this disclosure to illustrate the embodiments herein, this should not be construed as limiting the scope of the present invention.

[0074] Some embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. However, other embodiments are also within the scope of the subject matter disclosed herein, and the disclosed subject matter should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0075] In general, all terms used herein should be interpreted according to their ordinary meaning in the relevant technical field, unless a different meaning is explicitly given and / or is implied from the context in which they are used. All references to an element, apparatus, component, manner, step, etc. should be construed openly as referring to at least one instance of the element, apparatus, component, manner, step, etc., unless otherwise explicitly stated. The steps of any method disclosed herein need not be performed in the exact order disclosed, unless a step is explicitly described as after or before another step and / or it is implicit that a step must be after or before another step. In appropriate cases, any feature of any embodiment disclosed herein can be applied to any other embodiment. Similarly, any advantage of any embodiment can be applied to any other embodiment, and vice versa. Other objects, features, and advantages of the appended embodiments will become apparent from the following description.

[0076] In some embodiments, the more general term "network node" may be used, which may correspond to any type of radio network node or any network node that communicates with a UE (directly or via another node) and / or communicates with another network node. Examples of network nodes are NodeB, MeNB, eNB, network nodes belonging to MCG or SCG, base station (BS), multi-standard radio (MSR) radio nodes (such as MSR BS, eNodeB, gNodeB), network controller, radio network controller (RNC), base station controller (BSC), relay, donor node controlled relay, base transceiver station (BTS), access point (AP), transmission point, transmission node, RRU, RRH, nodes in a distributed antenna system (DAS), core network nodes (such as mobile switching center (MSC), mobility management entity (MME), etc.), operation and maintenance (O&M), operation support system (OSS), self-optimizing network (SON), positioning node (such as evolved serving mobile location center (E-SMLC)), minimized drive test (MDT), test equipment (physical node or software), etc.

[0077] In some embodiments, the non-limiting terms user equipment (UE) or wireless device may be used, and the term may refer to any type of wireless device that communicates with network nodes and / or another UE in a cellular or mobile communication system. Examples of UEs are target devices, device-to-device (D2D) UEs, machine type UEs or UEs capable of machine-to-machine (M2M) communication, PDAs, PADs, tablet computers, mobile terminals, smart phones, laptop embedded equipment (LEE), laptop mounted equipment (LME), USB dongles, Ml category UEs, M2 category UEs, ProSe UEs, V2V UEs, V2X UEs, etc.

[0078] In addition, terms such as base station / gNodeB and UE should be regarded as non-restrictive and, in particular, do not imply a certain hierarchical relationship between the two; generally, "gNodeB" can be regarded as device 1, "UE" can be regarded as device 2, and these two devices communicate with each other through a certain radio channel. And hereinafter, the transmitter or receiver may be a gNodeB (gNB) or a UE.

[0079] Narrowband Internet of Things (NB-IoT) is a low-power wide-area network (LPWAN) radio technology standard developed by 3GPP for cellular devices and services. The specification was frozen in June 2016 in 3GPP Release 13 (LTE Advanced Pro). Other 3GPP IoT technologies include eMTC (Enhanced Machine-Type Communication) and EC-GSM-IoT. NB-IoT focuses particularly on indoor coverage, low cost, long battery life, and high connection density. NB-IoT uses a subset of the LTE standard but limits the bandwidth to a single narrowband of 200 kHz. It uses OFDM modulation for downlink communication and SC-FDMA for uplink communication. When operating in the licensed spectrum, NB-IoT has no duty cycle limit and will better serve IoT applications that require more frequent communication.

[0080] Coverage enhancement is to provide reliable connections with extended coverage, and NB-IoT introduces retransmission, repetition, and power ramping schemes as well as CE groups during both the RACH procedure and the data transmission procedure. Retransmission means repeating the RA request in the absence of an RA response. Repetition means repeating the preamble to increase the detection probability at the gNB. Power ramping means increasing the transmit power level to compensate for higher path loss.

[0081] A wireless communication system can include one or more base stations, one or more UEs, and a core network. In some examples, the wireless communication system can be a Long-Term Evolution (LTE) network, an LTE-Advanced (LTE-A) network, an LTE-A Pro network, or a New Radio (NR) network. In some examples, the wireless communication system can support enhanced broadband communication, ultra-reliable (e.g., mission-critical) communication, low-latency communication, communication using low-cost and low-complexity devices, or any combination thereof.

[0082] As those skilled in the art will understand, aspects of the embodiments can be embodied as a system, apparatus, method, or program product. Thus, the embodiments can take the form of a complete hardware embodiment, a complete software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and hardware aspects.

[0083] For example, the disclosed embodiments can be implemented as hardware circuits that include custom very-large-scale integration ("VLSI") circuits or gate arrays, off-the-shelf semiconductors (such as logic chips, transistors, or other discrete components). The disclosed embodiments can also be implemented in programmable hardware devices such as field-programmable gate arrays, programmable array logic, programmable logic devices, etc. As another example, the disclosed embodiments can include one or more physical or logical blocks of executable code, which can be organized, for example, as objects, programs, or functions.

[0084] In addition, an embodiment may take the form of a program product, which is embodied in one or more computer-readable storage devices storing machine-readable code, computer-readable code, and / or program code (hereinafter referred to as code). The storage device may be tangible, non-transitory, and / or non-transmissive. The storage device may not embody a signal. In a particular embodiment, the storage device only uses a signal to access the code.

[0085] Any combination of one or more computer-readable media may be utilized. The computer-readable media may be a computer-readable storage medium. The computer-readable storage medium may be a storage device storing the code. The storage device may be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, holographic, micro-mechanical, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.

[0086] More specific examples (a non-exhaustive list) of the storage device will include the following: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (“RAM”), a read-only memory (“ROM”), an erasable programmable read-only memory (“EPROM”), or a flash memory, a portable compact disc read-only memory (“CD-ROM”), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer-readable storage medium may be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0087] The code for performing the operations of the embodiment may be of any number of lines and may be written in any combination of one or more programming languages, including object-oriented programming languages such as Python, Ruby, Java, Smalltalk, C++, etc., and conventional procedural programming languages such as the “C” programming language, and / or machine languages such as assembly language. The code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (“LAN”), a wireless LAN (“WLAN”), or a wide area network (“WAN”), or may be connected to an external computer (e.g., using an Internet service provider (“ISP”) through the Internet).

[0088] In addition, the described features, structures, or characteristics of the embodiments can be combined in any suitable manner. In the following description, numerous specific details are provided, such as examples of programming, software modules, user selections, network transactions, database queries, database structures, hardware modules, hardware circuits, hardware chips, etc., to provide a thorough understanding of the embodiments. However, those skilled in the relevant art will recognize that the embodiments can be practiced without one or more of the specific details or using other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the embodiments. References throughout the specification to "one embodiment", "an embodiment", or similar language mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, unless otherwise explicitly specified, the phrases "one embodiment", "an embodiment", and similar language that appear throughout the specification may, but do not necessarily, all refer to the same embodiment, but rather mean "one or more embodiments, but not all embodiments". Unless otherwise explicitly specified, the terms "including" and "comprising" and their variants mean "including but not limited to". Unless otherwise explicitly specified, the list of enumerated items does not imply that any or all of the items are mutually exclusive. Unless otherwise explicitly specified, the terms "a" and "an" and "the" also refer to "one or more".

[0089] Aspects of the embodiments are described below with reference to schematic flowcharts and / or schematic block diagrams of methods, apparatuses, systems, and program products according to the embodiments. It should be understood that each block of the schematic flowcharts and / or schematic block diagrams, and combinations of blocks in the schematic flowcharts and / or schematic block diagrams, can be implemented by code. This code can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device create means for implementing the functions / actions specified in the flowchart and / or block diagram.

[0090] The code can also be stored in a storage device that can direct a computer, other programmable data processing device, or other device to work in a particular manner, such that the instructions stored in the storage device produce an article of manufacture including instructions for implementing the functions / actions specified in the flowchart and / or block diagram.

[0091] The code can also be loaded onto a computer, other programmable data processing apparatus, or other devices, causing a series of operational steps to be performed on the computer, other programmable apparatus, or other devices, so as to generate a computer-implemented process, whereby the code executed on the computer or other programmable apparatus provides a process for implementing the functions / actions specified in the flowchart and / or block diagram.

[0092] The flowcharts and / or blocks in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatuses, systems, methods, and program products according to various embodiments. In this regard, each block in the flowchart and / or block diagram may represent a module, segment, or portion of code that includes one or more executable instructions for implementing the specified logical function(s).

[0093] It should also be noted that in some alternative implementations, the functions noted in the blocks may not occur in the order noted in the drawings. For example, two blocks shown in succession may in fact be executed substantially in parallel, or the blocks may sometimes be executed in the reverse order, depending on the functions involved. Other steps and methods may be conceived that are equivalent in function, logic, or effect to one or more blocks or portions thereof shown in the drawings.

[0094] Although various arrow types and line types may be employed in the flowcharts and / or block diagrams, it should be understood that they do not limit the scope of the corresponding embodiments. In fact, certain arrows or other connectors may be used only to indicate the logical flow of the depicted embodiments. For example, an arrow may indicate a waiting or monitoring period of unspecified duration between enumerated steps of the depicted embodiment. It should also be noted that each block of the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a system based on dedicated hardware that performs the specified functions or actions, or by a combination of dedicated hardware and code.

[0095] The description of an element in each drawing may refer to an element in a subsequent drawing. In all the drawings, the same reference numerals refer to the same elements, including alternative embodiments of the same elements.

[0096] Figure 1 The gNB behavior is shown. For the UE(s) participating in bilateral AI / ML model operation, the gNB sends a deactivation message (as a model state indication) of the gNB-based AI / ML model via PDCCH / MACCE (UE-specific) and system information (for all UEs).

[0097] Specifically, 1 bit is used for model state indication:

[0098] “0” -> active, and

[0099] "1" -> Deactivate.

[0100] Additionally, model transfer information (e.g., configuration parameters of the AI / ML model based on the gNB) is sent to the (multiple) UEs together with the deactivation message - optional.

[0101] The configurable criteria for determining the deactivation of the AI / ML model in the gNB are set by the network.

[0102] In the case of event triggering, the triggering mechanism for gNB model deactivation can be based on pre-configured KPIs, such as model performance level, model drift, etc. and any threshold specifications.

[0103] Base stations can be scattered throughout a geographical area to form a wireless communication system and can be devices in different forms or with different capabilities. The base stations and UEs can communicate wirelessly via one or more communication links. Each base station can provide a coverage area over which the UEs and the base station can establish one or more communication links. The coverage area can be an example of a geographical area over which the base station and the UEs can support signal communication according to one or more radio access technologies.

[0104] UEs can be scattered throughout the coverage area of the wireless communication system, and each UE can be stationary or mobile, or stationary and mobile at different times. UEs can be devices in different forms or with different capabilities. The UEs described herein can be capable of communicating with various types of devices, such as other UEs, base stations, or network equipment (e.g., core network nodes, relay devices, integrated access and backhaul (IAB) nodes, or other network equipment).

[0105] Base stations can communicate with the core network, or with each other, or with the core network and with each other. For example, a base station can interface with the core network via one or more backhaul links (e.g., via S1, N2, N3, or other interfaces). Base stations can communicate directly (e.g., directly between base stations), or indirectly (e.g., via the core network), or directly and indirectly with each other via backhaul links (e.g., via X2, Xn, or other interfaces). In some examples, the backhaul link can be or can include one or more wireless links.

[0106] One or more of the base stations described herein can include or can be referred to by those of ordinary skill in the art as a base transceiver station, radio base station, access point, radio transceiver, NodeB, eNodeB (eNB), next-generation NodeB or giga-NodeB (either of which can be referred to as gNB), home NodeB, home eNodeB, or other suitable terms.

[0107] A UE may include or may be referred to as a mobile device, wireless device, remote device, handheld device, or subscriber device, or some other suitable term, where the "device" may also be referred to as a unit, station, terminal, or client, etc. The UE may also include or may be referred to as a personal electronic device, such as a cellular phone, personal digital assistant (PDA), tablet computer, laptop computer, or personal computer. In some examples, the UE may include or be referred to as a wireless local loop (WLL) station, Internet of Things (IoT) device, Internet of Everything (IoE) device, or machine type communication (MTC) device, etc., which may be implemented in various objects (such as appliances or vehicles, meters, etc.).

[0108] As Figure 1 shown, the UE described herein may be capable of communicating with various types of devices, such as other UEs that may sometimes act as relays, as well as base stations and network equipment (including macro eNBs or gNBs, small cell eNBs or gNBs, or relay base stations, etc.).

[0109] The UE and the base station may wirelessly communicate with each other via one or more communication links on one or more carriers. The term "carrier" may refer to a set of radio spectrum resources having a defined physical layer structure for supporting a communication link. For example, a carrier for a communication link may include a portion (e.g., bandwidth part (BWP)) of a radio spectrum band that operates one or more physical layer channels according to a given radio access technology (e.g., LTE, LTE-A, LTE-A Pro, NR). Each physical layer channel may carry acquisition signaling (e.g., synchronization signals, system information), control signaling for coordinating carrier operation, user data, or other signaling. A wireless communication system may use carrier aggregation or multi-carrier operation to support communication with the UE. According to a carrier aggregation configuration, the UE may be configured with multiple downlink component carriers and one or more uplink component carriers. Carrier aggregation may be used with frequency division duplex (FDD) and time division duplex (TDD) component carriers.

[0110] The signal waveform transmitted on a carrier can be composed of multiple subcarriers (e.g., using multi-carrier modulation (MCM) techniques such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform spread OFDM (DFT-S-OFDM)). In a system adopting MCM techniques, a resource element can be composed of a symbol period (e.g., the duration of a modulated symbol) and a subcarrier, where the symbol period and the subcarrier spacing are negatively correlated. The number of bits carried by each resource element can depend on the modulation scheme (e.g., the order of the modulation scheme, the coding rate of the modulation scheme, or both). Therefore, the more resource elements the UE receives and the higher the order of the modulation scheme, the higher the data rate of the UE may be. Wireless communication resources can refer to a combination of radio spectrum resources, time resources, and space resources (e.g., spatial layers or beams), and using multiple spatial layers can further increase the data rate or data integrity of the communication with the UE.

[0111] The time intervals of the base station or the UE can be expressed as multiples of a basic time unit, which can refer to, for example, the sampling period Ts = 1 / (Δfmax·Nf) seconds, where Δfmax can represent the maximum supported subcarrier spacing, and Nf can represent the maximum supported discrete Fourier transform (DFT) size. The time intervals of the communication resources can be organized according to radio frames each having a specified duration (e.g., 10 milliseconds (ms)). Each radio frame can be identified by a system frame number (SFN) (e.g., ranging from 0 to 1023).

[0112] Each frame can include a plurality of consecutively numbered subframes or time slots, and each subframe or time slot can have the same duration. In some examples, a frame can be divided (e.g., in the time domain) into subframes, and each subframe can be further divided into a plurality of time slots.

[0113] Alternatively, each frame can include a variable number of time slots, and the number of time slots can depend on the subcarrier spacing. Each time slot can include a plurality of symbol periods (e.g., depending on the length of the cyclic prefix appended to each symbol period). In some wireless communication systems 100, a time slot can be further divided into a plurality of mini-slots each containing one or more symbols. Excluding the cyclic prefix, each symbol period can contain one or more (e.g., Nf) sampling periods. The duration of the symbol period can depend on the subcarrier spacing or the operating frequency band.

[0114] A subframe, a slot, a mini-slot, or a symbol can be the smallest scheduling unit of a wireless communication system (e.g., in the time domain) and can be referred to as a transmission time interval (TTI). In some examples, the TTI duration (e.g., the number of symbol periods in a TTI) can be variable. Additionally or alternatively, the smallest scheduling unit of wireless communication system 100 can be dynamically selected (e.g., in a burst of shortened TTIs (sTTIs)).

[0115] Physical channels can be multiplexed on a carrier according to various techniques. Physical control channels and physical data channels can be multiplexed on a downlink carrier, for example, using one or more of time division multiplexing (TDM) techniques, frequency division multiplexing (FDM) techniques, or hybrid TDM-FDM techniques. The control region of a physical control channel (e.g., a control resource set (CORESET)) can be defined by a plurality of symbol periods and can extend across the system bandwidth of the carrier or a subset of the system bandwidth. One or more control regions (e.g., CORESETs) can be configured for a group of UEs. For example, one or more of the UEs can monitor or search a control region for control information according to one or more search space sets, and each search space set can include one or more control channel candidates at one or more aggregation levels arranged in a cascaded manner. The aggregation level of a control channel candidate can refer to the number of control channel resources (e.g., control channel elements (CCEs)) associated with the encoded information of a control information format with a given payload size. The search space set can include: a common search space set configured to send control information to a plurality of UEs, and a UE-specific search space set for sending control information to a specific UE.

[0116] In some examples, a base station can be mobile and thus provide communication coverage for a mobile geographic coverage area. In some examples, different geographic coverage areas associated with different technologies can overlap, but different geographic coverage areas can be supported by the same base station. In other examples, overlapping geographic coverage areas associated with different technologies can be supported by different base stations. A wireless communication system can include, for example, a heterogeneous network in which different types of base stations use the same or different radio access technologies to provide coverage for various geographic coverage areas.

[0117] A wireless communication system can be configured to support ultra-reliable communication or low-latency communication, or various combinations thereof. For example, a wireless communication system can be configured to support ultra-reliable low-latency communication (URLLC) or mission-critical communication. A UE can be designed to support ultra-reliability, low latency, or critical functions (e.g., mission-critical functions). Ultra-reliable communication can include private communication or group communication and can be supported by one or more mission-critical services such as mission-critical push-to-talk (MCPTT), mission-critical video (MCVideo), or mission-critical data (MCData). Support for mission-critical functions can include service prioritization, and mission-critical services can be used for public safety or general commercial applications. The terms ultra-reliable, low-latency, mission-critical, and ultra-reliable low-latency may be used interchangeably herein.

[0118] In some examples, a UE may also be able to communicate directly with other UEs over a device-to-device (D2D) communication link (e.g., using peer-to-peer (P2P) or D2D protocols). One or more UEs utilizing D2D communication may be within the geographic coverage area of a base station. Other UEs in such a group may be outside the geographic coverage area of the base station or may not be able to receive transmissions from the base station. In some examples, a group of UEs communicating via D2D communication may utilize a one-to-many (1:M) system in which each UE transmits to every other UE in the group. In some examples, the base station facilitates scheduling of resources for D2D communication. In other cases, D2D communication is performed between UEs without involving the base station.

[0119] The core network can provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. The core network can be an evolved packet core (EPC) or a 5G core (5GC), which can include: at least one control plane entity that manages access and mobility (e.g., a mobility management entity (MME), an access and mobility management function (AMF)), and at least one user plane entity that routes packets or interconnects to external networks (e.g., a serving gateway (S-GW), a packet data network (PDN) gateway (P-GW), or a user plane function (UPF)). The control plane entity can manage non-access stratum (NAS) functions such as mobility, authentication, and bearer management of UEs served by a base station associated with the core network. User IP packets can be transported through the user plane entity, which can provide IP address allocation and other functions. The user plane entity can be connected to a network operator IP service. The operator IP service can include access to the Internet, an intranet(s), an IP multimedia subsystem (IMS), or packet-switched streaming services.

[0120] Some network devices (such as base stations) may include sub-components (such as access network entities, which may be an example of an access node controller (ANC)). Each access network entity may communicate with a UE via one or more other access network transmission entities, which may be referred to as radio heads, intelligent radio heads, or transmit / receive points (TRPs). Each access network transmission entity may include one or more antenna panels. In some configurations, various functions of each access network entity or base station may be distributed across various network devices (e.g., radio heads and ANCs), or consolidated into a single network device (e.g., a base station).

[0121] A wireless communication system may operate using one or more frequency bands (generally in the range of 300 megahertz (MHz) to 300 gigahertz (GHz)). Generally, the region from 300 MHz to 3 GHz is referred to as the ultra-high frequency (UHF) region or the decimeter band because the length of the wavelength is in the range of approximately one decimeter to one meter. UHF waves may be blocked or redirected by buildings and environmental features, but for macro cells, these waves may penetrate structures sufficiently to provide service to UEs located indoors. Compared to transmissions at smaller frequencies and longer wavelengths using the high frequency (HF) or very high frequency (VHF) portions of the spectrum below 300 MHz, transmissions of UHF waves may be associated with smaller antennas and shorter ranges (e.g., less than 100 kilometers).

[0122] A wireless communication system may utilize both licensed radio spectrum bands and unlicensed radio spectrum bands. For example, a wireless communication system may employ licensed-assisted access (LAA), LTE-unlicensed (LTE-U) radio access technology, or NR technology in an unlicensed band (such as the 5 GHz industrial, scientific, and medical (ISM) band). When operating in an unlicensed radio spectrum band, devices (such as base stations and UEs) may employ carrier sensing for collision detection and avoidance. In some examples, operations in the unlicensed band may be based on a carrier aggregation configuration (e.g., LAA) combined with a component carrier operating in a licensed band. Operations in the unlicensed spectrum may include downlink transmissions, uplink transmissions, peer-to-peer (P2P) transmissions, or device-to-device (D2D) transmissions, etc.

[0123] A base station or a UE may be equipped with multiple antennas, which can be used to adopt techniques such as transmit diversity, receive diversity, multiple-input multiple-output (MIMO) communication, or beamforming. The antennas of the base station or the UE may be located within one or more antenna arrays or antenna panels, which can support MIMO operation, or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays may be co-located at an antenna assembly (such as an antenna tower). In some examples, the antennas or antenna arrays associated with the base station may be located in different geographical locations. The base station may have an antenna array with multiple rows and columns of antenna ports, and the base station may use these antenna ports to support beamforming for communication with the UE. Similarly, the UE may have one or more antenna arrays that can support various MIMO or beamforming operations. Additionally or alternatively, the antenna panel may support radio frequency beamforming for signals transmitted via the antenna ports.

[0124] Beamforming (which may also be referred to as spatial filtering, directional transmission, or directional reception) is a signal processing technique that can be used at a transmitting device or a receiving device (e.g., a base station, a UE) to shape or steer an antenna beam (e.g., a transmit beam, a receive beam) along a spatial path between the transmitting device and the receiving device. Beamforming can be achieved by combining signals transmitted via the antenna elements of an antenna array such that some signals propagating in a particular orientation relative to the antenna array experience constructive interference while other signals experience destructive interference.

[0125] The adjustment of the signals transmitted via the antenna elements may include: the transmitting device or the receiving device applying an amplitude offset, a phase offset, or both an amplitude offset and a phase offset to the signals carried via the antenna elements associated with the device. The adjustment associated with each of the antenna elements may be defined by a set of beamforming weights associated with a particular orientation (e.g., relative to the antenna array of the transmitting device or the receiving device, or relative to some other orientation).

[0126] A wireless communication system includes a base station, a UE, a satellite, and a core network. In some examples, the wireless communication system may be an LTE network, an LTE-A network, an LTE-A Pro network, or an NR network. In some cases, the wireless communication system may support enhanced broadband communication, ultra-reliable (e.g., mission-critical) communication, low-latency communication, or communication using low-cost and low-complexity devices.

[0127] The wireless communication system may also include one or more satellites. The satellites may communicate with base stations (also referred to as gateways in NTN) and UEs (or other high-altitude or terrestrial communication devices). The satellites may be any suitable type of communication satellite configured to relay communications between different end nodes in the wireless communication system. Examples of satellites may include space satellites, balloons, airships, airplanes, drones, unmanned aerial vehicles, etc. In some examples, the satellites may be in geosynchronous or geostationary orbits, low Earth orbits, or medium Earth orbits. The satellites may be multi-beam satellites configured to serve multiple service beam coverage areas in a predetermined geographical service area. The satellites may be at any distance from the Earth's surface.

[0128] In some cases, a cell may be provided or established by a satellite that is part of a non-terrestrial network. In some cases, the satellite may perform the functions of a base station, act as a bent-pipe satellite, or may act as a regenerative satellite, or a combination of the above. In other cases, the satellite may be an example of an intelligent satellite or a satellite with intelligence. For example, an intelligent satellite may be configured to perform more functions than a regenerative satellite (e.g., may be configured to perform specific algorithms in addition to those used in regenerative satellites, be configured to be reprogrammed, etc.). A bent-pipe transponder or satellite may be configured to receive signals from a ground station and transmit these signals to different ground stations. In some cases, the bent-pipe transponder or satellite may amplify the signals or convert them from uplink frequencies to downlink frequencies. Like a bent-pipe transponder or satellite, a regenerative transponder or satellite may be configured to relay signals, but may also use on-board processing to perform other functions. Examples of these other functions may include demodulating received signals, decoding received signals, re-encoding signals to be transmitted, or modulating signals to be transmitted, or a combination of the above. For example, a bent-pipe satellite (e.g., a satellite) may receive signals from a base station and may relay the signals to a UE or a base station, or vice versa. According to various aspects of the present disclosure, coverage enhancement for communication with satellites may be provided through repeated communications, which may reduce the BLER and enhance communication reliability.

[0129] The UE may include a UE communication manager. The UE communication manager may receive an indication of repeated activation of communication from a base station or a satellite. The UE communication manager may determine resources for the communication including the repetition in response to the repeated indication. When receiving communication, the UE communication manager may buffer signals from resources including multiple transmissions and attempt to decode the associated communication. In some cases, when transmitting an uplink message to a base station or a satellite, the UE communication manager may prepare multiple repetitions of the communication based on the configured repetition. Additionally, in some cases, a frequency bandwidth smaller than the full channel bandwidth may be used to transmit the uplink message in order to provide a higher power density and enhance the likelihood of receiving the uplink message.

[0130] The base station may include a base station communication manager. The base station communication manager may configure a coverage enhancement scheme at one or more UEs, where multiple repetitions of communication are provided to help reduce the BLER. The repetition may be provided by bundling time slots of the PDCCH in consecutive or non-consecutive time slots, or by providing multiple repetitions within the same time slot.

[0131] Coverage Enhancement: CE is achieved through repetition in time, retransmission, and power boost in in-band and guard-band operating modes. The coverage extension feature of NB-IoT is very convenient when sensors are located in remote or challenging areas. Reliable coverage enhancement is achieved through repeated transmission of data and control signaling. Each transmission may be configured to be repeated a specified number of times in order to achieve a higher chance of success at the desired coverage level.

[0132] The repetition count value is proportional to the maximum coupling loss (MCL). The repetition count will improve the SNR at the receiver. An increase in the repetition count results in an increase in power consumption. A large repetition count results in a higher latency. CE will cause radio resource waste (in the case of a higher level of CE, the UE occupies the channel for a longer duration). CE is suitable for applications that are not sensitive to latency (applications that can tolerate a transmission delay of 10 seconds). Avoiding CE upgrade using any alternative can improve energy efficiency and reduce radio resource waste.

[0133] Figure 2 UE behavior is shown. When the UE receives a deactivation signal from the gNB, new signaling is needed to define the behavior of the UE.

[0134] Specifically, 2 bits are used for the UE behavior index:

[0135] "00" (UE behavior #1) -> Configuration model from the gNB

[0136] "01" (UE behavior #2) -> UE-side model

[0137] “10” (UE behavior #3) -> Fallback (non-AI / ML operation)

[0138] “11” (UE behavior #4) -> Reserved.

[0139] The gNB can configure the same behavior for all UEs (via system information), and / or configure UE-specific behavior via dedicated RRC messages (e.g., RRC reconfiguration messages). In a periodic manner, the gNB can signal a deactivation message based on a specific time period X (ms) or a specific time slot Y, where the values of X and Y are indicated in the system information message or the dedicated RRC message.

[0140] Figure 3 Shows the bilateral model inference signaling flow, and Figure 4 shows the user equipment (UE) behavior signaling flow.

[0141] When the bilateral model operation of joint inference crashes due to model deactivation and no auxiliary information is received from the gNB, the UE can also automatically switch to the UE-side model or fallback mode.

[0142] Then the UE behavior update report is transmitted to the gNB.

[0143] In the absence of gNB signaling, the mode switch in the UE is determined based on criteria using pre-configured thresholds and KPIs (such as UE model performance, UE AI / ML capability status, etc.).

[0144] During the joint inference phase, joint model parameter updates can be performed between the gNB and the UE through online federated learning.

[0145] In a bilateral model using separate models on the network side and the UE side, the network-side model can be located in the serving cell and / or the cloud computing server.

[0146] Through the associated serving cell, the connected neighbor cells can be configured to assist the gNB-UE joint inference in the bilateral model, such as inference calculation or model transmission information. These assisting neighbor cells can provide support for some UEs in motion during the joint inference phase based on the communication between the serving cell and the neighbor cells.

[0147] In a first additional embodiment, the gNB sends a 1-bit deactivation message of the gNB-based AI / ML model via PDCCH / MAC CE (UE-specific) and system information (for all UEs).

[0148] In a second additional embodiment, the gNB sends 2-bit UE behavior configuration information when sending a deactivation message to define the UE behavior for all UEs (via system information) and / or define UE-specific behavior via a dedicated RRC message (e.g., an RRC reconfiguration message).

[0149] In a third additional embodiment, the gNB sends model transmission information (e.g., configuration parameters of the gNB-based AI / ML model) to the (multiple) UEs when sending a deactivation message.

[0150] In a fourth additional embodiment, the gNB may signal a deactivation message to a specific time period X (ms) or a specific time slot Y, where the values of X and Y are indicated in a system information message or a dedicated RRC message.

[0151] In a fifth additional embodiment, when no auxiliary information is received from the gNB, the mode switch in the UE can also be determined based on criteria using pre-configured thresholds and KPIs (e.g., UE model performance, UE AI / ML capability status, etc.), so that the UE can also automatically switch to the UE-side model or the fallback mode.

[0152] In a sixth additional embodiment, during the joint inference phase, joint model parameter updates can be performed between the gNB and the UE through online federated learning.

[0153] In a seventh additional embodiment, the connected neighbor cells can be configured to assist the gNB-UE joint inference in the bilateral model to support the inference calculation or model transmission information (including mobile UEs) during the joint inference phase.

[0154] The core aspect of the new method is to create the following mechanism: where in the case of reaching the maximum RA procedure failure, the UE decides whether to upgrade to a higher level of CE or switch the network (TN to NTN, or NTN to TN). When the UE (NB-IoT) initially connects to TN, it can consider leveraging NTN before upgrading to a higher level of CE (CE 1, CE 2). When the UE (NB-IoT) initially connects to NTN, it can consider leveraging TN before upgrading to a higher level of CE (CE 1, CE 2).

[0155] A method is related to a wireless communication network, in which, in the case of reaching the maximum random access (RA) procedure failure, a user equipment (UE) decides whether to upgrade to a higher level of coverage enhancement (CE) or switch networks. The user equipment (UE) is in the idle mode. The user equipment (UE) decides whether to switch to an alternative network based on the overall network congestion, where this information is indicated by the network through the failure of the maximum random access (RA) procedure to reach the coverage enhancement (CE) level (k). It performs cell selection on the alternative network, verifies the coverage of the cell, and if the coverage of the cell is sufficient, then the user equipment (UE) switches to the alternative network and executes the random access (RA) procedure. If the coverage of the cell is insufficient, then the user equipment (UE) remains in the same network and upgrades from the coverage enhancement (CE) level (k) to level (k + 1) for the random access (RA) procedure.

[0156] The UE switches to the alternative network only when the coverage of the alternative network is good. If the coverage of the alternative network is poor, the UE will remain in the same network and upgrade the CE level.

[0157] A very important aspect of this new method is the low cost. Both LTE-M and NB-IoT have the advantages of low module cost and low operation and service prices. This is ensured through a simple chipset design (which eliminates unnecessary LTE functions) and increased production volume.

[0158] All the described solutions are based on LTE. NB-IoT and LTE-M are based on LTE and can be integrated into the existing LTE infrastructure via software upgrade. Since both NB-IoT and LTE-M can be provided in the GSM and LTE spectrums, no additional spectrum authorization is required.

[0159] Different from sensors that cannot transmit over long distances, both NB-IoT and LTE-M work via plug-and-play. The sensors are directly connected to the NB-IoT and / or LTE-M network without the need to install a local network or gateway.

[0160] Another very significant aspect of the described solutions is that they are extremely secure. NB-IoT and LTE-M are globally standardized technologies and use LTE security mechanisms according to 3GPP.

[0161] A wireless communication system can be configured to share available system resources and provide various telecommunication services (e.g., telephony, video, data, messaging, broadcasting, etc.) based on multiple access technologies that support communication with multiple users (e.g., CDMA systems, TDMA systems, FDMA systems, OFDMA systems, SC-FDMA systems, TD-SCDMA systems, etc.). In many cases, common protocols that facilitate communication with wireless devices are adopted in various telecommunication standards. For example, communication methods associated with eMBB, mMTC, and URLLC can be incorporated into the 5G NR telecommunication standard, while other aspects can be incorporated into the 4G LTE standard. Since mobile broadband technology is part of an ongoing evolution, further improvements in mobile broadband are still useful for continuing to advance the development of such technologies.

[0162] The performance of an AI / ML model or an additional neural network (NN) model can be based on multiple criteria, such as feature selection, model selection, the number of samples, etc. Feature selection can correspond to the input parameters of the ML / NN model used for training and testing the ML / NN model. Model selection can correspond to determining the model to be executed from multiple models (e.g., based on model complexity or optimization parameters). The number of samples can correspond to the number of observations of one or more features.

Claims

1. A method for receiving artificial intelligence (AI) and / or machine learning (ML) model state information from at least one base station (gNB) in a wireless communication network by a user equipment (UE) device for gNB-UE bilateral model control, It is characterized in that After receiving the artificial intelligence (AI) and / or machine learning (ML) model state information, the user equipment (UE) makes the following determinations: If the artificial intelligence (AI) and / or machine learning (ML) model state information is activated, the user equipment (UE) maintains the same artificial intelligence (AI) and / or machine learning (ML) model state information, If the artificial intelligence (AI) and / or machine learning (ML) model state information is disabled, the user equipment (UE) switches to an alternative operation mode based on assistance information from a base station (gnB).

2. The method for gNB-UE dual-side model control by user equipment (UE) according to claim 1, in, The artificial intelligence (AI) and / or machine learning (ML) model status information includes a model status indication from a base station (gNB), model transmission information, and a user equipment (UE) behavior index.

3. A method for gNB-UE bilateral model control by a base station (gNB) in a wireless communication network, in, The method includes: Determine artificial intelligence (AI) and / or machine learning (ML) model state information, The base station (gNB) checks the status of the artificial intelligence (AI) and / or machine learning (ML) model status information, If the result of the check is that the base station (gNB) is not deactivated, the artificial intelligence (AI) and / or machine learning (ML) model state information remains the same artificial intelligence (AI) and / or machine learning (ML) model state information, and the artificial intelligence (AI) and / or machine learning (ML) model state information is used for further procedures, If a result of the check is that the artificial intelligence (AI) and / or machine learning (ML) model status information is deactivated, the artificial intelligence (AI) and / or machine learning (ML) model status information is transmitted to at least one user equipment (UE) via the base station (gNB).

4. A method for gNB-UE bilateral model control by a base station (gNB) in a wireless communication network, in, The artificial intelligence (AI) and / or machine learning (ML) model status information includes a model status indication from a base station (gNB), model transmission information, and a user equipment (UE) behavior index.

5. A method for performing gNB-UE bilateral model control between at least one user equipment (UE) and a base station (gNB) in a wireless communication network, the method comprising an artificial intelligence (AI) and / or machine learning (ML) model training phase and an artificial intelligence (AI) and / or machine learning (ML) model inference phase.

6. The method for gNB-UE bilateral model control between at least one user equipment (UE) and a base station (gNB) in a wireless communication network according to claim 5, comprising an artificial intelligence (AI) and / or machine learning (ML) model training phase and an artificial intelligence (AI) and / or machine learning (ML) model inference phase, in, The artificial intelligence (AI) and / or machine learning (ML) model inference phase includes the following steps: transmitting measurement reports on artificial intelligence (AI) and / or machine learning (ML) and / or user equipment (UE) capabilities from the user equipment (UE) to the base station (gNB), transmitting parameter configuration updates regarding artificial intelligence (AI) and / or machine learning (ML) from the base station (gNB) to at least one user equipment (UE), Artificial intelligence (AI) and / or machine learning (ML) model inference is performed by the base station (gNB) and at least one user equipment (UE), Performing data exchange of model transmission and / or inference output between a base station (gNB) and at least one user equipment (UE), Model performance monitoring feedback exchange is performed between a base station (gNB) and at least one user equipment (UE).

7. A method for performing gNB-UE bilateral model control between at least one user equipment (UE) and a base station (gNB) in a wireless communication network, the method comprising an artificial intelligence (AI) and / or machine learning (ML) model inference phase, the phase The following steps are involved: Determine model state changes and user equipment (UE) behavior indices, Generate model transmission information, Creating artificial intelligence (AI) and / or machine learning (ML) auxiliary information, wherein the artificial intelligence (AI) and / or machine learning (ML) auxiliary information includes a model state indication, model transmission information, and a user equipment (UE) behavior index, transmitting the artificial intelligence (AI) and / or machine learning (ML) assistance information to at least one user equipment (UE), Switching, by the user equipment (UE), to an alternative operating mode based on the artificial intelligence (AI) and / or machine learning (ML) assistance information, A user equipment (UE) behavior update report is transmitted from the user equipment (UE) to the base station (gND).

8. An apparatus for performing gNB-UE bilateral model control between a user equipment (UE) and a base station (gNB) in a wireless communication network, the apparatus comprising a wireless transceiver, a processor coupled to a memory storing computer program instructions therein, the instructions being configured to implement the steps of the method of claims 1 to 7.

9. A user equipment comprising the apparatus according to claim 8.

10. A base station (gNB), comprising the apparatus according to claim 8.

11. A wireless communication system for performing gNB-UE bilateral model control from a base station to a user equipment, in, The base station comprises a processor coupled to a memory having computer program instructions stored therein, the instructions being configured to implement the steps of the method of claims 4 to 7, wherein the user equipment (UE) comprises a processor coupled to a memory having computer program instructions stored therein, the instructions being configured to implement the steps of the method of claims 1 to 3.

Images