Artificial intelligence assisted base station and beam joint prediction and handover

Abstract

The invention relates to artificial intelligence assisted base station and beam joint prediction and handover. A method for wireless communication is provided that includes utilizing an artificial intelligence (AI) model to pre-determine that a radio link failure (RLF) or handover will occur, where the pre-determination is based at least on radio link condition information related to a user equipment (UE), the radio link status information includes at least information reflecting a status of a radio link between the UE and a serving base station (SBS).

Description

Technical Field

[0001] The present disclosure generally relates to wireless communication technologies. More specifically, the present disclosure relates to an AI (Artificial Intelligence)-assisted target base station and its serving beam prediction scheme. Background Art

[0002] Benefiting from the great advantages of large-scale antenna arrays, massive bandwidth resources, and dense network deployments, the millimeter-wave band can support ultra-high-speed wireless communication of several Gbps and has become one of the key alternative technologies for next-generation mobile communications. However, due to the fact that both free-space propagation loss and molecular absorption attenuation increase with the increase in frequency, the millimeter-wave channel has a high propagation attenuation. How to solve this problem has become one of the key points in millimeter-wave communication research. Fortunately, the millimeter wave has the advantage of a short wavelength, which makes it easy to integrate large-scale antenna arrays on a limited panel, so as to use beamforming technology to achieve directional convergent transmission of energy and compensate for propagation attenuation. However, millimeter-wave radio frequency devices have high power consumption and high cost, which results in a small number of radio frequency links that can be equipped in millimeter-wave communication, and traditional digital beamforming is difficult to effectively implement. Therefore, millimeter-wave communication usually adopts an analog beamforming structure based on phase shifters, and each signal passes through phase shifters with different phases and is emitted by multiple antennas to form a beam for directional transmission. In an actual communication system, in order to reduce the complexity of beamforming, the beamforming vector is usually directly selected from a predefined beamforming codebook, which contains beams corresponding to different directions. To support the above hardware structure, millimeter-wave communication requires beam management, that is, to obtain and maintain the optimal beam with the maximum received power for the user equipment (UE) to achieve reliable transmission.

[0003] Inspired by the powerful adaptive-nonlinear fitting ability of deep learning, intelligent communication applying deep learning to wireless communication has become one of the focus directions of 6G. As one of the three typical scenarios of the NR air interface, 3GPP has carried out standardization discussions on deep learning-assisted wireless communication in beam management, where deep learning can play two roles. On the one hand, deep learning can effectively extract the non-linear features in beam management and improve beam management performance. On the other hand, deep learning can adaptively model complex high-dimensional wireless environments in a data-driven manner and timely adapt to environmental fluctuations through online learning, which helps to achieve efficient beam management based on specific environments.

[0004] At the same time, deep learning has significant advantages compared with three traditional beam management methods:

[0005] (1) In the case of limited beam directions, the most commonly used beam management scheme is beam search, that is, scanning and selecting the optimal beam with the highest received power. However, this often incurs a large overhead. In contrast, deep learning can extract the inherent non-linear features of beam measurement signals, thus effectively reducing the search overhead.

[0006] (2) A large amount of work estimates the optimal beam direction by leveraging the mathematical properties of channel models. However, this method highly depends on explicit channel prior assumptions, and its performance will severely degrade when the real scenario does not match the assumptions. In the face of the real wireless environment, deep learning can achieve end-to-end learning in a data-driven form without explicit prior knowledge.

[0007] (3) Traditional machine learning methods such as Support Vector Machines (SVM) have been used for beam management. However, the fitting capabilities of these methods are often limited. In contrast, deep learning relies on a vast number of learnable parameters and has a more powerful non-linear fitting ability. Summary of the Invention

[0008] A brief overview of the present disclosure is given in this section to provide a basic understanding of some aspects of the present disclosure. However, it should be understood that this overview is not an exhaustive overview of the present disclosure. It is not intended to identify the key or important parts of the present disclosure, nor is it intended to limit the scope of the present disclosure. Its purpose is merely to present some concepts of the present disclosure in a simplified form as a prelude to the more detailed description that follows.

[0009] According to one aspect of the present disclosure, there is provided a method for wireless communication, including: using an Artificial Intelligence (AI) model to predict that a Radio Link Failure (RLF) or handover will occur, wherein the prediction is at least based on wireless link condition information related to a User Equipment (UE), and the wireless link condition information at least includes information reflecting the condition of the wireless link between the UE and a Serving Base Station (SBS). According to another aspect of the present disclosure, there is provided a system for wireless communication, including one or more processors and a memory. The memory stores computer-readable program instructions, and when the computer-readable program instructions are executed by the one or more processors, the method as described above is executed.

[0010] According to another aspect of the present disclosure, there is provided a system for wireless communication, including one or more processors and a memory, the memory storing computer-readable program instructions, and the computer-readable program instructions, when executed by the one or more processors, cause the method as described above to be executed.

[0011] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer-readable program instructions, and the computer-readable program instructions, when executed by one or more processors, cause the method for wireless communication as described above to be executed.

[0012] According to another aspect of the present disclosure, there is provided a method executed by a user equipment (UE), including predicting a probability of occurrence of a radio link failure (RLF) or a handover by using a first artificial intelligence (AI) model, the prediction being at least based on radio link condition information related to the user equipment (UE), and the radio link condition information at least including information reflecting a condition of a radio link between the UE and a serving base station.

[0013] According to another aspect of the present disclosure, there is provided a user equipment (UE), including one or more processors and a memory, the memory storing computer-readable program instructions, and the computer-readable program instructions, when executed by the one or more processors, cause the method as described above to be executed.

[0014] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer-readable program instructions, and the computer-readable program instructions, when executed by one or more processors, cause the method executed by the user equipment (UE) as described above to be executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present disclosure can be better understood by referring to the detailed description given below in conjunction with the accompanying drawings, in which the same or similar reference numerals are used throughout the drawings to denote the same or similar elements. All the drawings, together with the following detailed description, are included in this specification and form a part of the specification, and are used to further illustrate the embodiments of the present disclosure and explain the principles and advantages of the present disclosure. Among them:

[0016] Figure 1 is a flowchart illustrating an exemplary conditional handover process according to the related art.

[0017] Figure 2 is a schematic diagram illustrating an exemplary beam failure recovery process according to the related art.

[0018] Figures 3A to 3D is a flowchart showing an exemplary method for wireless communication according to an embodiment of the present disclosure.

[0019] Figure 4 is a schematic diagram showing an exemplary LSTM structure according to an embodiment of the present disclosure.

[0020] Figure 5 is a schematic diagram showing the structure of an exemplary base station-beam switching joint prediction model according to an embodiment of the present disclosure.

[0021] Figure 6 is a flowchart showing an exemplary method for wireless communication according to an embodiment of the present disclosure.

[0022] Figure 7 is a diagram showing the structure of an exemplary base station-beam switching joint prediction model according to an embodiment of the present disclosure.

[0023] Figure 8 is a flowchart showing an exemplary method for wireless communication according to an embodiment of the present disclosure.

[0024] Figure 9 shows adjacent beam measurements according to an embodiment of the present disclosure.

[0025] Figure 10 shows probability-priority-based beam measurements according to an embodiment of the present disclosure.

[0026] Figure 11 is a flowchart showing an exemplary method for BS-beam switching joint prediction based on downlink BFR measurements according to an embodiment of the present disclosure.

[0027] Figure 12 is a flowchart showing an exemplary method for BS-beam switching joint prediction based on uplink BFR measurements according to an embodiment of the present disclosure.

[0028] Figure 13 is a flowchart showing an exemplary UE-side beam pattern feedback method according to an embodiment of the present disclosure.

[0029] Figure 14 shows an example of UE-side beam absolute angle calculation according to an embodiment of the present disclosure.

[0030] Figure 15 shows the influence of UE self-rotation on the UE beam angle.

[0031] Figure 16 is a flowchart showing a base station-beam switching method according to an embodiment of the present disclosure.

[0032] Figure 17 is a flowchart showing a base station-beam switching method according to an embodiment of the present disclosure.

[0033] Figure 18is a flowchart showing an exemplary method of collecting training data according to an embodiment of the present disclosure.

[0034] Figure 19 Shows a reduced prediction output according to an embodiment of the present disclosure.

[0035] Figure 20 is a flowchart showing a method of activating an AI-assisted RLF / handover prediction service according to an embodiment of the present disclosure.

[0036] Figure 21 is a flowchart showing a method of activating and deactivating a BS-beam switching joint prediction model for a UE according to an embodiment of the present disclosure.

[0037] Figure 22 Shows an example of triggering model update based on a relative threshold according to an embodiment of the present disclosure.

[0038] Figure 23 Shows an exemplary wireless channel environment according to an embodiment of the present disclosure.

[0039] Figure 24 Shows the normalized beam gain performance of BS-beam switching joint prediction according to an embodiment of the present disclosure.

[0040] Figure 25 Shows the curve of the accuracy rate of BS prediction varying with the number of base stations according to an embodiment of the present disclosure.

[0041] Figures 26A - 26C Block diagrams respectively showing exemplary configurations of base stations according to some embodiments of the present disclosure.

[0042] Figure 27A and Figure 27B Flowcharts respectively showing exemplary methods for wireless communication according to an embodiment of the present disclosure.

[0043] Figure 27C and Figure 27D Block diagrams respectively showing exemplary configurations of user equipment according to an embodiment of the present disclosure.

[0044] Figure 28 is a block diagram showing a first example of a schematic configuration of a base station to which the technology of the present disclosure can be applied.

[0045] Figure 29 is a block diagram showing a second example of a schematic configuration of a base station to which the technology of the present disclosure can be applied.

[0046] Figure 30 is a block diagram showing an example of a schematic configuration of a smart phone to which the technology of the present disclosure can be applied.

[0047] Figure 31It is a block diagram showing an example of the schematic configuration of an in-vehicle navigation device to which the technology of the present disclosure can be applied.

[0048] The features and aspects of the present disclosure will be clearly understood by reading the following detailed description with reference to the accompanying drawings. Detailed Description of the Embodiment

[0049] Hereinafter, various exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. For clarity and conciseness, not all implementation manners of the embodiments are described in this specification. However, it should be noted that many implementation-specific settings can be made according to specific requirements when implementing the embodiments of the present disclosure so as to achieve the specific goals of the developer. In addition, it should also be understood that although the development work may be complex and time-consuming, for those skilled in the art who benefit from the present disclosure content, such development disclosure is merely a routine task.

[0050] In addition, it should also be noted that in order to avoid obscuring the present disclosure with unnecessary details, only the processing steps and / or device structures closely related to the technical solution of the present disclosure are shown in the drawings. The following description of the exemplary embodiments is merely illustrative and is not intended to be any limitation on the present disclosure and its applications.

[0051] The meanings of some abbreviations appearing in the present disclosure are listed below first.

[0052] 3GPP (3rd Generation Partnership Project) Third Generation Partnership Project

[0053] AoA (Angle of Arrival) Angle of Arrival

[0054] AoD (Angle of Departure) Angle of Departure

[0055] BF (Beam Failure) Beam Failure

[0056] BFR (Beam Failure Recovery) Beam Failure Recovery

[0057] BS (Base Station) Base Station

[0058] CSI (Channel State Information) Channel State Information

[0059] CSI-RS (Channel State Information Reference Signal) Channel State Information Reference Signal

[0060] CNN (Convolutional Neural Network) Convolutional Neural Network

[0061] MIB (Master Information Block) Master Information Block

[0062] NR (New Radio) New Radio

[0063] PBCH (Physical Broadcasting Channel) Physical Broadcasting Channel

[0064] PBS (Predicted Base Station) Predicted Handover Base Station (specifically refers to the accessible base station predicted by the model in this disclosure)

[0065] PCI (Physical Cell Index) Physical Cell Index

[0066] PSS (Primary Synchronization Signal) Primary Synchronization Signal

[0067] SBS (Serving Base Station) Serving Base Station (specifically refers to the serving base station before handover in this disclosure)

[0068] SSB (Synchronization Signal Block) Synchronization Signal Block

[0069] SSS (Secondary Synchronization Signal) Secondary Synchronization Signal

[0070] SRS (Sounding Reference Signal) Sounding Reference Signal

[0071] SVM (Support Vector Machine) Support Vector Machine

[0072] RLF (Radio Link Failure) Radio Link Failure

[0073] RSRP (Reference Signal Receiving Power) Reference Signal Receiving Power

[0074] RRC (Radio Resource Control) Radio Resource Control

[0075] TBS (Target Base Station) Target Access Base Station (specifically refers to the access base station after RRC reconstruction of the UE in this disclosure)

[0076] UE (User Equipment) User Equipment (user)

[0077] ULA (Uniform Linear Array) Uniform Linear Array

[0078] Although deep learning has brought significant performance gains in beam management, existing deep learning-assisted beam management methods often only focus on beam prediction within a single base station (BS). In high-speed mobile scenarios, in addition to beam switches within a single BS, handovers between BSs also occur frequently. In the traditional BS handover process, the user equipment listens to the reference signal received power (RSRP) of adjacent BSs and sends a measurement report to the serving BS. After the serving BS requests and obtains the handover confirmation from the target BS, it notifies the UE of the handover configuration through an RRC reconfiguration message, and then the UE immediately performs the handover. In Release 16, conditional handover (CHO) is further introduced. At this time, the target BS prepares for conditional handover, and the UE determines the actual trigger time of the handover according to the preconfigured conditions, as Figure 1 shown. This can avoid the situation of UE handover failure caused by changes in the radio link state during the signaling interaction between the UE and the source base station and the signaling interaction between the source base station and the target base station, thus improving the robustness of the user during the handover process. However, using the existing conditional handover mechanism, in high-frequency scenarios such as the millimeter-wave band, the UE needs to sequentially search for a relatively large number of alternative beams sequentially sent by the target base station to determine the beam that meets the handover conditions, and there is still a certain beam scanning delay and overhead.

[0079] In addition, the existing NR standard uses radio link failure (RLF) to define the event that the communication quality of the current serving BS (SBS) does not meet the UE's requirements. When RLF is triggered, the UE performs RRC reconstruction and listens to the synchronization signal and PBCH block (SSB) broadcast by surrounding BSs to search for accessible BSs and their optimal beams, but this will bring a large RRC reconstruction delay and overhead.

[0080] To reduce the above overhead, the present disclosure contemplates using deep learning to predict RLF or handover and jointly predict the BS and its beam handover. Specifically, since deep learning can extract high-dimensional radio environment features around the BS and UE (the high-dimensional radio environment features are the features contained in the surrounding scattering environment, including the position and scattering properties of scatterers (such as buildings, walls, trees, etc.). Generally, the more numerous and diverse the number and type of scatterers contained in an environment, the higher the overall dimension of the corresponding radio environment features), it can establish a mapping from the link-related information of the serving base station and neighboring base stations to the user location, which provides the feasibility for predicting the accessible BS and its appropriate serving beam. Based on this prediction, the predicted accessible predicted BS (PBS) and UE only need to measure a small number of predicted high-quality beams to complete the BS handover, thus significantly reducing the handover overhead.

[0081] Generally speaking, the present disclosure relates to first pre-judging whether a radio link failure (RLF) or handover (e.g., RLF or handover-related events, including A3 events) will occur by using artificial intelligence (AI) at least based on the link condition information related to the UE, and then, in the case where it is pre-judged that an RLF or handover will occur, predicting a target BS-beam set by using AI at least based on the link condition information related to the UE for performing subsequent handover. An AI model can be implemented, which can include one or more AI sub-models deployed on the serving base station or deployed on the serving base station and the UE. Through these one or more AI sub-models, at least one of the pre-judgment of RLF or handover and the prediction of the target BS-beam set is achieved.

[0082] To implement the above pre-judgment of RLF or handover and / or prediction of the target BS-beam set, it is necessary to obtain radio link information related to the UE as the prediction input. In the NR standard, considering that the RRC reconstruction overhead after RLF triggering is relatively large, and at the same time the main reason for the reduction of the UE communication quality in the mobile scenario is beam misalignment rather than leaving the SBS cell range, the existing NR standard defines a beam failure (BF) event to avoid frequent RLF triggering, as Figure 2 shown. Specifically, as Figure 2 shown, the UE detects beam failure and determines new candidate beams. Subsequently, the UE sends a beam failure recovery (BFR) request to the BS through random access and listens for the reply of the BS to the BFR request. When the UE successfully finds a candidate beam that meets its own communication requirements and receives the reply of the BS to the BFR request, the BFR process is completed without triggering RLF and a complex RRC reconstruction process.

[0083] Traditional BFR has two deficiencies: (1) BFR is usually triggered when the service quality is insufficient to support basic communication requirements. If BFR fails, RLF is directly triggered, and the UE cannot communicate with the current serving BS for the serving BS to provide assistance in a timely manner; (2) Limited by the timing of triggering BFR, the number of measurable beams may be small, making it difficult to ensure that the newly measured candidate beams are the optimal options. Therefore, in some embodiments of the present disclosure, an artificial intelligence model can be used to predict RLF, so as to trigger the measurement of corresponding candidate beams in advance when the UE can still communicate with the current serving base station and report the measurement results to the current serving base station. On the other hand, since the embodiments of the present disclosure perform candidate beam measurement in advance, the measurement range can be increased, thereby improving the quality of the measurement results. According to other embodiments of the present disclosure, the measurement results of the candidate beams triggered in advance according to the AI prediction can be used as the input for predicting the target BS-beam set.

[0084] In addition, in the face of a complex wireless environment, it is difficult for traditional methods to accurately model the relationship between the CSI information of the SBS and BS-beam switching; at the same time, the total number of alternative beams of all adjacent BSs around the SBS is large, making it difficult to ensure accurate prediction. Some embodiments of the present disclosure use deep learning to adaptively model the wireless environment characteristics around the SBS, which can achieve accurate prediction; and can reduce the output space according to the training set labels to improve the prediction performance.

[0085] In the present disclosure, the serving base station (SBS) refers to the serving BS before BS handover. The prediction model according to the embodiments of the present disclosure can be deployed in the SBS to fully extract the wireless environment characteristics around the SBS and improve the prediction performance.

[0086] The predicted handover BS (PBS) refers to the predicted accessible BS, which may include one or more BSs. When the PBS includes multiple BSs, BS handover attempts can be made sequentially according to the prediction priority until the handover is successful or the upper limit of the preset number of BSs is reached.

[0087] The target access base station (TBS) refers to the access BS after the UE successfully performs BS handover in some embodiments. When BS handover (or RRC reconstruction when RLF is triggered) adopts traditional methods, the TBS refers to the BS accessed after the traditional BS handover process; when BS handover adopts the methods in some embodiments of the present disclosure, the TBS refers to the BS actually switched to in the predicted PBS. In other embodiments, the TBS is the BS that the UE reconnects to after successfully performing beam failure recovery, that is, it may also be the original serving base station.

[0088] ·Judgment of RLF / handover and prediction of target BS-beam

[0089] Figure 3AIt is a flowchart showing an exemplary method 3100 for wireless communication according to an embodiment of the present disclosure.

[0090] As Figure 3A shown, the method 3100 for wireless communication may include operation 3101, in which it is predicted that a radio link failure (RLF) or handover will occur, where the prediction is at least based on radio link condition information related to a user equipment (UE), and the radio link condition information at least includes information reflecting the condition of the radio link between the UE and a serving base station (SBS).

[0091] In some embodiments, this prediction operation may be performed by the UE. The UE may make a prediction on whether an RLF or handover will occur according to statistical information on the reception of downlink signals and comparison with a predetermined threshold (the predetermined threshold is looser than the relevant thresholds for existing RLF events or A3 events, etc.).

[0092] In some embodiments, this prediction operation may be performed by the serving base station. The serving base station may make a prediction on whether an RLF or handover will occur according to statistical information on the reception of uplink signals and comparison with a predetermined threshold (the predetermined threshold is looser than the relevant thresholds for existing RLF events or A3 events, etc.).

[0093] The statistical information may be statistics of signal strengths such as RSRP, RSRQ, RSSI, etc., or may be statistics of signal error rates such as BLER, BER, etc.

[0094] In short, the prediction can be made in advance by setting conditions looser than existing RLF events or A3 events.

[0095] On the one hand, predicting in advance that an RLF or handover will occur can enable the UE to trigger corresponding candidate beam measurements in advance when it can still communicate data with the current serving base station and report the measurement results to the current serving base station.

[0096] On the other hand, candidate beam measurements can be made in advance, so the measurement range can be increased, thereby improving the quality of the measurement results. In the case where target BS-beam prediction is to be performed subsequently, high-quality candidate beam measurement results can provide high-quality input for target BS-beam prediction.

[0097] Figure 3B It is a flowchart showing an exemplary method 3200 for wireless communication according to an embodiment of the present disclosure.

[0098] As Figure 3BAs shown, a method 3200 for wireless communication may include operation 3201, in which an artificial intelligence (AI) model is used to predict that a radio link failure (RLF) or handover will occur, where the prediction is at least based on radio link condition information related to a user equipment (UE), and the radio link condition information at least includes information reflecting the condition of a radio link between the UE and a serving base station (SBS).

[0099] In some embodiments, the AI model may be arranged at the UE, and the UE may send the prediction result obtained by using the AI model or information related to the prediction result (such as the probability of predicted RLF or handover occurrence) to the serving base station.

[0100] In some embodiments, the UE may use the AI model to predict the probability of RLF or handover occurrence. When the probability of RLF or handover occurrence is greater than a predetermined threshold, the UE may predict that RLF or handover will occur.

[0101] In some other embodiments, the UE may also use the AI model to directly output a prediction result, that is, whether RLF or handover will occur.

[0102] In some other embodiments, the AI model may be arranged at the serving base station. The serving base station may use the AI model to predict the probability of RLF or handover occurrence, and compare the probability of RLF or handover occurrence with a predetermined threshold to predict whether RLF or handover will occur.

[0103] In still some other embodiments, the serving base station may use the AI model to directly output a prediction result.

[0104] Since the AI model can establish a connection between a complex radio environment and BS-beam handover, the prediction of AI-assisted RLF or handover allows for accurately predicting in advance the probability of occurrence and / or occurrence time of RLF or handover, etc., based on radio link condition information related to the UE, and making a response earlier to better ensure the stability of communication.

[0105] Predicting in advance that RLF or handover will occur can enable the UE to trigger corresponding candidate beam measurements in advance when it can still communicate with the current serving base station and report the measurement results to the current serving base station. Moreover, performing candidate beam measurements in advance can increase the measurement range, thereby improving the quality of the measurement results. In the case of subsequent target BS-beam prediction, high-quality candidate beam measurement results can provide high-quality inputs for target BS-beam prediction.

[0106] Figure 3C is a flowchart showing an exemplary method 3300 for wireless communication according to an embodiment of the present disclosure.

[0107] As Figure 3C shown, a method 3300 for wireless communication may include an operation 3301, in which an artificial intelligence (AI) model is utilized to predict a target base station and a beam of the target base station for serving the user equipment, where the prediction is at least based on wireless link condition information related to the user equipment (UE), and the wireless link condition information at least includes information reflecting the condition of the wireless link between the UE and the serving base station (SBS).

[0108] The AI model may be arranged at the serving base station. In some embodiments, the AI model may be used to assist the serving base station in predicting the target base station and the beam of the target base station for serving the user equipment, for example, for configuring conditional handover.

[0109] The operation 3201 may be performed in response to a pre-judgment that an RLF or a handover is about to occur. The pre-judgment that an RLF or a handover is about to occur may be obtained according to various methods combined Figure 3A or Figure 3B described.

[0110] Utilizing the AI model can accurately establish the connection between the complex wireless environment and the BS-beam handover. Therefore, the AI-assisted BS-beam prediction can allow for accurately predicting in advance the target base station and the serving beam of the target base station for the UE based on the wireless link condition information related to the UE, thereby better ensuring the stability of communication.

[0111] Figure 3D is a flowchart showing an exemplary method 3400 for wireless communication according to an embodiment of the present disclosure.

[0112] As Figure 3D shown, a method 3400 for wireless communication may include an operation 3401, in which an artificial intelligence (AI) model is utilized to pre-judge that a radio link failure (RLF) or a handover will occur, where the pre-judgment is at least based on wireless link condition information related to the user equipment (UE), and the wireless link condition information at least includes information reflecting the condition of the wireless link between the UE and the serving base station (SBS).

[0113] As Figure 3D shown, a method 3400 for wireless communication may include an operation 3402, in which, in response to the pre-judgment, the AI model is utilized to predict a target base station and a beam of the target base station for serving the user equipment for configuring conditional handover, where the prediction is at least based on the wireless link condition information related to the UE.

[0114] The AI model is an AI-based system model that can implement RLF or handover prediction and target BS-beam prediction. The AI model may include one or more AI sub-models arranged at the UE and the serving base station. In some embodiments, the AI model may only include one or more AI sub-models arranged at the serving base station.

[0115] The AI sub-model arranged at the UE can be used for predicting RLF or handover. The AI sub-model arranged at the serving base station can be used for predicting RLF or handover and / or predicting the target BS-beam.

[0116] The output of the AI sub-model for predicting RLF or handover may include at least one of the occurrence probability and occurrence time of RLF or handover. In some embodiments, the output of the AI sub-model for predicting RLF or handover may be the prediction result of RLF or handover, that is, whether RLF or handover will occur. The output of the AI sub-model for predicting the target BS-beam may include the predicted target base station-beam set.

[0117] The input for predicting RLF or handover and the input for predicting the target BS-beam may be the same or different. For example, the input for predicting the target BS-beam may include, in addition to the input for predicting RLF or handover, additional wireless link condition information obtained through supplementary measurements, etc., after the prediction.

[0118] The output of the AI model may include a target base station-beam set, which includes one or more target base station-beam pairs, and each pair indicates a predicted target base station and a beam of the target base station used to serve the user equipment.

[0119] The above only shows some embodiments according to the present disclosure. Those skilled in the art can conceive of various other embodiments under the teaching of the present disclosure. For example, in the case where the UE or the serving base station predicts that RLF or handover will occur, the serving base station can also configure conditional handover or assist beam recovery using traditional methods according to the measurement results of the serving base station and neighboring base stations reported by the UE, instead of configuring handover based on the BS-beam set predicted using the AI model, etc.

[0120] ·BS-beam handover joint prediction model

[0121] An AI-based model that can implement the prediction of RLF / handover and target BS-beam prediction can be realized, hereinafter referred to as the BS-beam handover joint prediction model. The BS-beam handover joint prediction model may include one or more AI sub-models. These AI sub-models can be deployed at the serving base station, or deployed at the serving base station and the UE.

[0122] Each AI sub-model included in the BS-beam switching joint prediction model has an input that includes at least radio link condition information related to the UE.

[0123] The output of each AI sub-model for target BS-beam prediction in the BS-beam switching joint prediction model may include the target base station and its set of candidate beams for serving the UE. The output of each AI sub-model for pre-judging RLF / handover in the BS-beam switching joint prediction model may include the occurrence probability and / or occurrence time of RLF or handover. In the case where the BS-beam switching joint prediction model consists of only one AI sub-model, the output of this AI sub-model may include at least one of the occurrence probability, occurrence time, and target BS-beam set of RLF or handover.

[0124] The BS-beam switching joint prediction model of the present disclosure supports predicting the occurrence probability and occurrence time of RLF / handover, etc., based on radio link information over a past period of time. Specifically, the input of this model may be radio link condition information over a past period of time, including but not limited to the UE's past periodic or aperiodic downlink beam measurement results for the serving BS and / or neighboring BSs (such as beam IDs and the corresponding RSRP, RSSI, etc. of the beams), CSI measurements for the serving BS (such as PMI, RI, CQI), and so on. The output may include the occurrence probability and occurrence time of RLF / handover, the predicted BS-beam set, and so on.

[0125] As an embodiment of the artificial intelligence model of the present disclosure, a recurrent neural network (RNN), such as a long short-term memory network (LSTM), may specifically be adopted.

[0126] Figure 4 is a schematic diagram showing the basic structure of an LSTM. As Figure 4 shown, the input information of the LSTM at the current time t includes the output of the LSTM at the previous time and the input of the LSTM at the current time. Therefore, the LSTM can sequentially learn radio link information over a past period of time to accurately model the user's movement state.

[0127] RLF or handover is often caused by user movement. For example, due to user movement, the direct path to the current serving base station is blocked, and the communication quality of the current serving base station is significantly reduced, resulting in RLF. Therefore, effectively modeling and extracting the user's movement state helps to more accurately predict RLF or handover. Further, the wireless link information of the user in the past period of time can reflect the user's movement characteristics. Therefore, a neural network suitable for processing time series is used to extract the user's movement characteristics, that is, a recurrent neural network (including long short-term memory network LSTM, etc.).

[0128] The input of the LSTM can be the wireless link condition information related to the UE input in the form of a time series over a period of time (or the wireless link condition features extracted after preprocessing according to the wireless link condition information related to the UE input in the form of a time series over a period of time). The input can be input in the form of a time series, including but not limited to: for the wireless link condition information measured at equal time intervals, the input is the wireless link condition information forming a sequence from front to back in time order; for the wireless link condition information measured at unequal time intervals, the input can be the wireless link condition information forming a sequence from front to back in time order and the corresponding time information (such as the normalized measurement interval).

[0129] The output of the LSTM can be the prediction result of the occurrence of RLF or handover (or the prediction features of the occurrence of RLF or handover, and these features form the prediction result after passing through the output layer), including but not limited to the occurrence probability, occurrence time, handover target base station and beam of RLF or handover.

[0130] Figure 5 It is a schematic diagram showing the structure of an exemplary base station-beam handover joint prediction model 500 according to an embodiment of the present disclosure. In some embodiments, the base station-beam handover joint prediction model 500 may include a first AI sub-model 501 deployed in the serving base station.

[0131] As Figure 5 shown, for the base station-beam handover joint prediction model 500 implemented by the LSTM structure, assume that the past periodic CSI measurement results H1, H2,..., H t are used for prediction (that is, used as the input of the AI model), H tIt represents the CSI measurement result of the t-th period. In one example, the CSI measurement result H is obtained by restoring the downlink CSI measurement result based on the UE's measurement of the CSI-RS of the current serving base station and reporting it. For example, it is the downlink channel matrix restored according to the PMI and RI included in the CSI. In another example, especially applicable to a TDD system where the uplink and downlink channels are reciprocal, the CSI measurement result is the uplink CSI measurement result obtained by the current serving base station measuring the SRS sent by the UE, that is, the uplink channel matrix. After each CSI is input, first, a preprocessing layer is used to perform preliminary feature extraction on the CSI. The implementation forms of the preprocessing layer in this disclosure include Convolutional Neural Networks (CNN), Fully-Connected Neural Networks (FCNN), etc. Subsequently, an LSTM is used to sequentially extract features from the CSI sequence to model the user movement information. Finally, an output layer is used to obtain the final result, which may include the predicted RLF / handover probability, the predicted RLF / handover time, and the predicted target BS-beam.

[0132] When it is determined that RLF / handover occurs because the predicted RLF / handover probability exceeds a certain preset threshold, the target BS and beam for handover can be directly predicted according to the output layer (but not limited to this). In Figure 5 the embodiment, an AI sub-model 501 is adopted, and after predicting RLF or handover and before predicting the target BS-beam, no additional measurement is performed. Then, the inputs for determining the handover event and predicting the handover BS-beam are the same as the input for predicting the handover probability itself. The above multiple prediction targets can be implemented using a single neural network structure. This neural network can use the same radio link-related information related to the UE as the input, and finally output multiple targets at the output layer. The advantage of such a design is that it can significantly reduce the overall computational overhead of the prediction model.

[0133] Figure 5 In the embodiment, the implementation form of the output layer is a fully-connected layer (assuming that the output of the LSTM at the t-th moment is x t , the fully-connected layer can be written as y t = Wx t + b t , where W and b are the weight matrix and the bias vector, and yt is the output of the fully-connected layer).

[0134] The specific values of the weight matrix W and the bias vector b are learned by the AI model, but their dimensions can usually be predefined in advance to meet the matrix operation dimension constraints of the output y = Wx + b. x is the output feature vector of the LSTM layer. Assume that the dimension of the feature vector output by the LSTM at the t-th layer is N t×1, the output is the probability p (scalar) of whether the predicted handover event occurs, and the dimension of the corresponding weight matrix W of this output is N t ×1, the dimension of the bias vector is 1×1. Correspondingly, if the output is the base station-beam probability of handover, there are M alternative base station-beam combinations, so the output is a vector of M×1, and the dimension of the corresponding weight matrix W of this output is N t ×M, the dimension of the bias vector is M×1.

[0135] For the case where the AI sub-model outputs different results simultaneously, multiple output layers are usually used, and each output layer corresponds to a different result. For example, when simultaneously outputting the handover probability, handover time, and BS-beam set, three output layers can be adopted. The input of each output layer is the LSTM output feature vector. Assume that the dimension of the feature vector output by the t-th layer of LSTM is N t ×1. For the handover probability output (scalar), the dimensions of the weight matrix W and the bias vector b are N t ×1 and 1×1 respectively; for the handover time output (scalar), the dimensions of the weight matrix W and the bias vector b are N t ×1 and 1×1 respectively; for the BS-beam set output, assume that there are M alternative base station-beam combinations, so the output is a vector of M×1, and the dimension of the corresponding weight matrix W of this output is N t ×M, the dimension of the bias vector is M×1.

[0136] The output layer itself is a part of the AI model, and its training process is consistent with that of the entire AI model, that is, based on the loss between the prediction result and the true label (the smaller the difference between the prediction result and the true label, the smaller the loss; the larger the difference between the prediction result and the true label, the larger the loss), the parameters of the output layer are optimized using, for example, the gradient backpropagation algorithm.

[0137] As the above exemplary CSI-based model input, it can also be replaced by beam measurement results, such as the beam ID of the serving base station measured in each period and indicators such as RSRP and BLER reflecting its strength or quality. In addition, the model input can also be the neighbor cell ID providing coverage by adjacent base stations, neighbor cell beam ID, and indicators such as RSRP reflecting its strength measured during the mobility measurement process. In some other embodiments, the input data of the model 500 may include real-time geographical environment and wireless environment information of at least one of the high-precision map and the electromagnetic map, the location and movement characteristics of the UE, or sensor data of at least one of the UE's accelerometer and barometer. These data can essentially also reflect the wireless link conditions related to the UE.

[0138] As Figure 5As shown, the input of the base station-beam switching joint prediction model 500 at time t can be the radio link condition information related to the UE at that time, and the output at that time can include the predicted RLF / handover probability, the predicted RLF / handover time, and / or the predicted target BS-beam set.

[0139] In some embodiments, the serving base station can predict whether an RLF or handover will occur based on this output. For example, the serving base station can compare the predicted RLF / handover probability with a pre-set threshold to predict whether an RLF or handover will occur. In the case of predicting that an RLF or handover will occur, the serving base station can use the predicted BS-beam set for subsequent handovers.

[0140] In some embodiments, the predicted BS-beam set can include multiple BS-beams. In some embodiments, the prediction result can provide the UE with a prioritized BS handover process. For example, the predicted priority of the BS handover for the UE can be BS2, BS3, BS1, BS4. Then the UE can preferentially access BS2 and preferentially access the beam with higher predicted quality within BS2. When the handover to BS2 fails, continue the handover attempt to BS3 according to the priority, and so on, until the handover is successful or the upper limit of the pre-set BS handover overhead is reached.

[0141] Those skilled in the art can understand that the output layer of the base station-beam switching joint prediction model 500 (the first AI sub-model 501) can be designed as needed. In some embodiments, the output can be the prediction result of RLF or handover and / or the predicted occurrence time of RLF or handover, and the predicted BS-beam set.

[0142] Figure 6 is a flowchart showing an exemplary method 600 for wireless communication according to an embodiment of the present disclosure.

[0143] This method can be applicable to the case where the base station-beam switching joint prediction model described in combination with Figure 4 and Figure 5 includes the first sub-AI model deployed at the serving base station. In this case, a trained AI sub-model can complete the operations of predicting the occurrence probability of RLF / handover and predicting the target BS-beam. The serving base station can use this AI sub-model to complete RLF / handover prediction and target BS-beam prediction. In some embodiments, this AI sub-model can also predict the occurrence time of RLF / handover.

[0144] As Figure 6As shown, method 600 may include operation 601, in which, at least based on wireless link condition information related to the UE, the serving base station uses a first AI sub-model to predict the occurrence probability of RLF or handover and the target base station-beam set.

[0145] The wireless link condition information related to the UE may include at least one of the wireless link condition information related to the UE received from the UE or the wireless link condition information related to the UE measured by the serving base station.

[0146] Method 600 may further include operation 602, in which the serving base station pre-judges that RLF or handover will occur based on the comparison between the occurrence probability of RLF or handover and a predetermined threshold.

[0147] Method 600 may further include operation 603, in which, in response to the pre-judgment, the serving base station uses the target base station-beam set for configuring conditional handover.

[0148] In some embodiments, the serving base station may directly output the pre-judgment result of RLF or handover using the first AI sub-model. In some embodiments, the first AI sub-model may further indicate the occurrence time of the predicted RLF or handover. By adopting a trained AI model (which can accurately model the relationship between the wireless environment characteristics of the SBS and the BS-beam handover), accurate pre-judgment of RLF or handover and accurate prediction of the target BS-beam can be achieved.

[0149] By setting one AI sub-model at the serving base station to implement both the pre-judgment of RLF / handover and the prediction of the target BS-beam, and thus achieving multiple prediction outputs with one neural network structure, the overall computational overhead of the prediction model can be reduced.

[0150] Figure 7 FIG. is a diagram showing the structure of an exemplary base station-beam handover joint prediction model 700 according to an embodiment of the present disclosure.

[0151] As shown in FIG. 700, the base station-beam handover joint prediction model 700 may include a second AI sub-model 701 and a third AI sub-model 702.

[0152] The second AI sub-model 701 and the third AI sub-model 702 may both be deployed on the serving base station.

[0153] The internal structures of the second AI sub-model 701 and the third AI sub-model 702 may be similar to Figure 5 the structure of the first AI sub-model 501 shown in.

[0154] The input of the second AI sub-model 701 may be the wireless link condition information related to the UE, and the output may be the occurrence probability of RLF or handover.

[0155] The serving base station can anticipate an upcoming RLF or handover based on the occurrence probability of RLF or handover predicted by the second AI sub-model 701, and collect additional UE-related radio link condition information.

[0156] It can be understood that in some embodiments, the second sub-model 701 can also directly output the anticipation result of RLF or handover. In some embodiments, the second sub-model 701 can also output the predicted occurrence time of RLF or handover.

[0157] The UE-related radio link condition information and the additional radio link condition information collected after the anticipation are included together in the input of the third AI model 702, and the output of the third AI model 702 can be the target BS-beam set.

[0158] The UE-related radio link condition information and the additional radio link condition information collected after the anticipation can be different types of information.

[0159] For example, the UE-related radio link condition information for anticipation can include mobility measurement results, real-time geographical and radio environment information including at least one of a high-precision map and an electromagnetic map, the location and movement characteristics of the UE, or data of sensors including at least one of an accelerometer and a barometer of the UE, etc., while the additional radio link condition information collected after the anticipation for the third AI sub-model can include CSI.

[0160] It can be understood that when training the second AI sub-model and the third AI sub-model, the training data used can also be different. For example, the training data for training the second AI sub-model can include mobility measurement results, real-time geographical and radio environment information including at least one of a high-precision map and an electromagnetic map, the location and movement characteristics of the UE, or data of sensors including at least one of an accelerometer and a barometer of the UE, etc., while the training data for training the third AI sub-model can also include CSI.

[0161] It can be understood that in some embodiments, the UE-related radio link condition information for anticipation can include CSI, while the additional radio link condition information collected after the anticipation for the third AI sub-model can include mobility measurement results, real-time geographical and radio environment information including at least one of a high-precision map and an electromagnetic map, the location and movement characteristics of the UE, or data of sensors including at least one of an accelerometer and a barometer of the UE, etc. In the training phase, the training data for training the third AI sub-model and the third AI sub-model can also be correspondingly different.

[0162] Figure 8It is a flowchart showing an exemplary method 800 for wireless communication according to an embodiment of the present disclosure.

[0163] This method can be applicable to the case where the base station-beam switching joint prediction model 700 described in Figure 7 includes two AI sub-models (i.e., the second AI sub-model 701 and the third AI sub-model 703) deployed at the serving base station. In this case, the two trained AI sub-models can respectively implement the operations of RLF / handover probability prediction and target BS-beam prediction. The serving base station can use the second AI sub-model 701 to complete the RLF / handover pre-judgment and use the third AI sub-model 702 to complete the target BS-beam prediction.

[0164] As Figure 8 shown, method 800 may include operation 801, at which, based on the radio link condition information related to the UE, the serving base station uses the second AI sub-model, such as Figure 7 the second AI sub-model 701, to predict the occurrence probability of RLF or handover.

[0165] Method 800 may further include operation 802, at which the serving base station pre-judges that RLF or handover will occur based on the comparison between the occurrence probability of RLF or handover and a predetermined threshold.

[0166] Method 800 may further include operation 803, in response to this pre-judgment, the serving base station may collect supplementary radio link condition information related to the UE. As described above, the supplementary radio link condition information related to the UE collected after the pre-judgment may be the same type of information or different types of information as the radio link condition information related to the UE collected before the pre-judgment.

[0167] Method 800 may further include operation 804, at which, in response to this pre-judgment, based on the radio link condition information related to the UE and the supplementary radio link condition information related to the UE, the serving base station uses the third AI sub-model, such as Figure 7 the third AI sub-model 702, to predict the target base station-beam set for configuring conditional handover.

[0168] It can be understood that in some embodiments, the second AI sub-model 701 may directly output the pre-judgment result of RLF or handover at operation 801. In this case, operation 802 may be omitted.

[0169] After pre-judging RLF or handover, additional measurement events are performed for handover, and BS-beam prediction is performed based on the additional measurement results, which can allow the predicted BS-beam set to be more accurate. This is particularly advantageous for the case where the UE is moving at a relatively high speed.

[0170] The above Figures 5 - 8 all describe the situation where the AI sub - model included in the base station - beam switching joint prediction model is only deployed at the serving base station. In some cases, the AI sub - model can also be deployed at the UE as part of the base station - beam switching joint prediction model.

[0171] In some embodiments, an AI sub - model (referred to herein as the fourth AI sub - model) can be deployed at the UE. The structure of the fourth AI sub - model can be similar to Figure 5 the structure of the first AI sub - model 501 shown in

[0172] The UE can predict the occurrence probability of RLF or handover using the fourth AI sub - model based on the radio link condition information related to the UE. In some embodiments, the fourth AI sub - model can also output the occurrence time of RLF or handover. The UE can send the predicted occurrence probability and / or occurrence time of RLF or handover to the serving base station.

[0173] In some embodiments, the serving base station can include the occurrence probability of RLF or handover predicted by the UE received from the UE in its own input for predicting the occurrence probability of RLF or handover and predicting the target base station - beam using the AI sub - model. For example, in Figure 6 and Figure 8 the method of

[0174] By arranging the fourth AI sub - model at the UE, the RLF or handover pre - judgment result on the UE side can be obtained as part of the input information for the RLF or handover pre - judgment and / or the target BS - beam at the serving base station. On the one hand, the amount of information transmitted from the UE to the serving base station can be reduced, and on the other hand, the accuracy of the pre - judgment and / or prediction at the serving base station can be improved.

[0175] In some other embodiments, the base station - beam switching joint prediction model can include an AI sub - model deployed at the UE (referred to as the fifth AI sub - model) and an AI sub - model deployed at the serving base station (referred to as the sixth AI sub - model). The structures of the fifth AI sub - model and the sixth AI sub - model can be similar to Figure 5 the structure of the first AI sub - model 501 shown in

[0176] In this case, the UE can predict the occurrence probability and / or occurrence time of RLF or handover based at least on the radio link condition information related to the UE by using the fifth AI sub-model. The UE can send the predicted occurrence probability and / or occurrence time of RLF or handover to the serving base station. The serving base station can pre-judge the occurrence of RLF or handover based on the comparison between the predicted occurrence probability of RLF or handover received from the UE and a predetermined threshold. In response to this pre-judgment, the serving base station can use the sixth AI sub-model to predict the target base station-beam set for configuring conditional handover based at least on the radio link condition information related to the UE.

[0177] As described above, the AI sub-model deployed on the UE side itself can predict the probability and time of RLF / handover according to the channel quality information of the serving BS and / or neighboring BSs in the past period of time. When the predicted probability is greater than a certain preset threshold, it pre-judges the occurrence of the event and feeds back one or more of the event type, predicted probability, and time of the event related to RLF / handover to the serving BS. The serving BS can directly use the probability fed back by the UE for pre-judgment; it can also use this probability as an input to further predict the occurrence probability of RLF / handover and the target BS-beam handover.

[0178] In some embodiments, the fifth AI sub-model can also directly output the pre-judgment result and / or occurrence time of RLF or handover. The sixth AI sub-model can start predicting the target BS-beam set in response to the pre-judgment result from the fifth AI sub-model.

[0179] Making predictions on the user side can reduce the overhead of the UE feeding back radio link condition information to the serving BS. Taking the above embodiments as an example, the UE does not have to feed back the channel quality information of the serving BS and neighboring BSs in the past period of time, but only needs to feed back the predicted probability, or even only the pre-judgment result.

[0180] The above illustrates the pre-judgment of RLF or handover and the prediction of the target BS-beam by one or more AI sub-models. Those skilled in the art can implement various variations according to the teachings of the present disclosure. For example, in some embodiments, only the AI sub-model for implementing the pre-judgment of RLF or handover can be used, and the conditional handover or beam recovery after the pre-judgment is still carried out according to the traditional method. In some embodiments, only the AI sub-model for implementing the prediction of the target BS-beam can be used, and the pre-judgment of RLF or handover does not use AI assistance.

[0181] In some embodiments, the handover or RLF-related event may include an A3 event. The A3 event defined in the 3GPP NR standard refers to: within a preset time threshold, the difference between the communication quality of the SBS and the communication quality of the neighboring base station satisfies a preset relationship. Assume that the RSRP of the SBS is P S , and the RSRP of a certain neighboring base station is P N, that is, satisfying

[0182] P N >P S +P th

[0183] where P th is a preset threshold, usually a positive value or a negative value with a small absolute value, indicating that the communication quality of the neighboring base station is better than or close to the communication quality of the SBS.

[0184] In some embodiments, for an A3 event, the radio link status information related to the UE may include information reflecting the radio link status between the UE and the serving base station and the neighboring base stations. The traditional A3 event is triggered by the UE measuring the communication quality of the serving BS and the neighboring BS. The embodiments of the present disclosure can predict the probability and occurrence time of the A3 event based on the radio link information in the past period of time. The prediction model can predict the probability and occurrence time of the A3 event based on the radio link information of the serving BS and / or the neighboring BS in the past period of time. When the predicted probability is greater than a certain threshold, it is determined that the A3 event will occur.

[0185] The determination of the occurrence of the A3 event can be used to further determine RLF or handover. In some embodiments, the determination of RLF / handover can be activated according to the determination that the A3 event is about to occur. The input of the AI model for the determination of RLF / handover may include the predicted probability and / or the predicted occurrence time of the A3 event.

[0186] In some other embodiments, at least one of the RLF or handover prediction or target BS-beam prediction of the embodiments of the present disclosure can be performed based on the traditional A3 event. For example, when the A3 event is triggered, the UE can send a measurement report (RSRP or reference signal received quality (RSRQ)) of the communication quality of the SBS and the neighboring base stations to the SBS. The SBS can predict the target base station and its serving beam according to the measurement report and other radio link related information (for example, the CSI measurement, beam measurement results, etc. of the UE in the past period of time). In other words, the measurement results after the A3 event is triggered can be used as part of the radio link status information related to the UE, and thus used as at least part of the input for the target BS-beam prediction.

[0187] · Radio link status information

[0188] In some embodiments, the radio link status information related to the UE may include, in addition to the information reflecting the status of the radio link between the UE and the serving base station, the information reflecting the status of the radio link between the UE and one or more neighboring base stations.

[0189] In some embodiments, the radio link condition information related to the UE may include the radio link condition information measured by the UE or the radio link condition information measured by the serving base station. The serving base station may receive the radio link condition information measured by the UE from the UE.

[0190] In some embodiments, the radio link condition information related to the UE may include, but is not limited to: beam measurement results, CSI measurement results, mobility measurement results, real-time geographical environment and radio environment information including at least one of a high-precision map or an electromagnetic map, the location and movement characteristics of the UE, and data of sensors including at least one of an accelerometer or a barometer of the UE. The beam measurement results include, but are not limited to, BFR measurement results.

[0191] The target handover prediction of the base station and the beam depends on the user location and the radio environment. The location and movement characteristics of the UE, the radio link measurement results, and the sensor information (such as an accelerometer reflecting acceleration, a barometer reflecting altitude, etc.) can all reflect the current or future location information of the user. The high-precision map and the electromagnetic map (the mapping of different locations and radio link conditions, such as the mapping of different locations and CSI) can reflect the radio environment information. Those skilled in the art can understand that any information related to the condition of the radio link of the UE and capable of assisting in the determination of the UE location and the radio environment can be included in the radio link condition information related to the UE.

[0192] When the primary cell BS and the secondary cell BS simultaneously serve the UE, the measurement results of both the primary cell and the secondary cell can be used as prediction inputs, that is, the radio link condition information related to the UE.

[0193] In some embodiments, the BFR measurement may be triggered when the predetermined communication quality threshold is not met and / or when it is predicted that RLF / handover will occur. The communication quality threshold Th1 and the alternative set of BFR success thresholds Th2 for triggering the BFR measurement can be configured by the RRC. The serving base station may send the above threshold set to the UE through the RRC Reconfiguration message, and the UE configures the thresholds by itself according to the alternative set.

[0194] The communication quality threshold Th1 may be a threshold that can guarantee the basic communication requirements but has the potential possibility of RLF or handover. For example, the RSRP threshold can support the basic communication requirements but cannot support high-speed services.

[0195] When a beam that reaches the BFR success threshold Th2 is found, it is considered that the subsequent BFR will be successful, and there is no need to perform the BS-beam handover joint prediction.

[0196] The BFR measurement triggered by the SBS predicting that RLF / handover will occur will be able to support the prediction of the target BS-beam.

[0197] The BFR measurement according to the embodiments of the present disclosure starts earlier than the traditional BFR measurement, so it can include additional beam measurements, thereby increasing the success probability of the BFR to find a better beam and collecting reliable prediction inputs at the same time.

[0198] In some embodiments, adjacent beam measurements can be performed, as Figure 9 shown. In adjacent beam measurements, a set of beams that are adjacent to the current serving beam in terms of angle are measured. As Figure 9 shown, the beams that can be measured include the serving beam and the beams adjacent to the serving beam. This provides additional beam measurements compared to traditional BRF measurements.

[0199] In other embodiments, beam measurements based on probability priority can be performed, as Figure 10 shown. In beam measurements based on probability priority, a certain number of beams are selected for measurement from largest to smallest according to the probability that each beam becomes the target beam obtained by prediction. This method can be based on a deep learning-assisted BFR process, that is, the probability that each alternative beam becomes the target beam is predicted by a deep learning model, and this probability reflects the priority of beam measurement. The serving base station can use an AI model (referred to as the seventh AI sub-model) to maintain its own optimal beam. In the training phase, its input can be the beam measurement information (such as RSRP) of the serving base station itself in the past period of time, and the label can be the optimal beam index at present and in the future (obtained by traditional beam management methods), so as to train this model; when the training ends and starts to predict the optimal beam index at present and in the future, the output of the model can usually be the probability that each alternative beam becomes the optimal beam. The beam with the highest predicted probability is selected as the predicted optimal beam. The predicted probability information can reflect the priority of beam quality, that is, the higher the probability, the better the beam quality is considered, so priority measurements can be performed according to the probability from high to low. As Figure 10 shown, beam 6 can be measured first, then beam 5, and then beam 3, etc.

[0200] In some embodiments, both adjacent beam measurements and beam measurements based on probability priority can be performed.

[0201] Figure 11 FIG.

[0202] As Figure 11 shown, method 1100 may include operation 1101, in which the BS configures CSI-RS resources for downlink BFR measurement.

[0203] Method 1100 may further include operation 1102, in which the BS transmits CSI-RS and the UE performs downlink BFR measurement.

[0204] For example, when the BFR fails to find a beam that meets the preset communication quality requirements of the UE, operation 1103-0015 is performed.

[0205] Method 1100 may further include operation 1103, in which the UE feeds back the downlink BFR measurement result to the BS.

[0206] Method 1100 may further include operation 1104, in which the UE feeds back the UE receive beam pattern to the BS.

[0207] Method 1100 may further include operation 1105, in which the BS performs BS-beam switching joint prediction.

[0208] In this embodiment, the beam measurement result may include the downlink BFR measurement result and the UE receive beam pattern. The beam measurement result may be included in the radio link condition information associated with the UE for BS-beam switching joint prediction. At least based on the radio link condition information, pre-judgment of RLF or handover and target BS-beam prediction may be performed, as described above.

[0209] Figure 12 It is a flowchart showing an exemplary method 1200 for BS-beam switching joint prediction based on uplink BFR measurement according to an embodiment of the present disclosure.

[0210] As Figure 12 shown, method 1200 may include operation 1201, in which the BS configures SRS resources for uplink BFR measurement.

[0211] Method 1200 may include operation 1202, in which the BS notifies the UE of the configured SRS resources.

[0212] Method 1200 may include operation 1203, in which the UE transmits SRS and the BS performs uplink BFR measurement.

[0213] For example, when the BFR fails to find a beam that meets the preset communication quality requirements of the BS, operations 1204-1205 are performed.

[0214] Method 1200 may include operation 1204, in which the UE feeds back the UE transmit beam pattern to the BS.

[0215] Method 1200 may include operation 1205, in which the BS performs BS-beam switching joint prediction.

[0216] In this embodiment, the beam measurement result may include an uplink BFR measurement result and a UE transmission beam pattern. The beam measurement result may be included in the radio link condition information associated with the UE for BS-beam switching joint prediction. At least based on the radio link condition information, pre-judgment of ELF or handover and target BS-beam prediction may be performed as described above.

[0217] In some embodiments, to improve the BS-beam switching joint prediction performance, the present disclosure contemplates an improved UE-side beam pattern feedback method, which may feedback UE-side beam absolute angle information. Figure 13 FIG. 1300 is a flowchart illustrating an exemplary UE-side beam pattern feedback method according to an embodiment of the present disclosure.

[0218] As Figure 13 shown, method 1300 may include the following operations:

[0219] Operation 1301, in which the BS configures the granularity of UE absolute beam angle feedback.

[0220] Operation 1302, in which the BS notifies the UE of the granularity of UE absolute beam angle feedback.

[0221] Operation 1303, in which the UE feedbacks the absolute beam angle to the BS according to the configured granularity.

[0222] Figure 14 FIG. 1304 illustrates an example of UE-side beam absolute angle calculation according to an embodiment of the present disclosure.

[0223] As Figure 14 shown, the UE rotation angle is θ UE = 14.5°, the UE-side relative beam angle is θ UE,beam = {-30°, -15°, 0°, 15°, 30°}, then the UE-side absolute beam angle is θ = θ UE + θ UE,beam = {-15.5°, -0.5°, 14.5°, 29.5°, 44.5°}. Assuming that the BS sets the feedback interval to -90° to 90° and the quantization feedback with a granularity of 4° (i.e., the sampling points are {-90°, -86°, …, 86°, 90°}), then the quantized UE-side absolute beam angle feedback is {-14°, -2°, 14°, 30°, 46°}.

[0224] Performing UE-side absolute beam angle feedback is to improve the BS-beam switching joint prediction performance because the model needs to more accurately model the radio environment around the BS and the UE and infer user location characteristics from the radio link information.

[0225] Figure 15Illustrate the impact of the UE's own rotation on the UE beam angle. As Figure 15 shown, for the BS, its beam direction is fixed. Therefore, the measurement results corresponding to different beams can accurately reflect the wireless environment around the BS. However, for the UE, its beam direction is affected by the rotation angle of the UE itself. When the UE rotation angle is different, the absolute angles corresponding to the same beam are different. Therefore, the measurement results of the same beam cannot accurately model the wireless environment around the UE. To solve this problem, the present disclosure proposes that the UE feeds back the absolute angle information of the beam to the BS, so that the prediction model deployed at the BS can extract the wireless environment characteristics around the UE based on the absolute angle information of the beam and the measurement results, improving the prediction performance.

[0226] In some embodiments, methods for reducing the feedback overhead of the UE-side beam absolute angle information can be considered. For example, the feedback methods for the absolute angle information may include, but are not limited to: (1) The UE feeds back the UE's own rotation angle and the relative beam angle information; (2) The UE feeds back the calculated set of significant absolute angles (for example, several absolute angles with the strongest received power), and so on.

[0227] The absolute angle feedback method according to the embodiments of the present disclosure helps to accurately model the wireless environment around the UE and improve the prediction performance. Feeding back the absolute angle at a certain granularity can reduce the feedback overhead of the UE absolute beam angle.

[0228] ·BS - Beam Switching

[0229] As described above, the target BS-beam set predicted by using the base station-beam switching joint prediction model according to the embodiments of the present disclosure can be used for subsequent handovers, such as conditional handovers.

[0230] Figure 16 is a flowchart showing the base station-beam switching method 16000 according to the embodiments of the present disclosure. As Figure 16 shown, the method 16000 may include operation S1601, in which one or more candidate target base stations are selected by the serving base station based on the target BS-beam set. Preferably, multiple candidate target base stations are selected to improve the UE handover success rate.

[0231] The method 16000 may further include operation S1602, in which the serving base station sends a handover request to the candidate target base stations among the one or more candidate target base stations.

[0232] The method 16000 may further include operation S1603. If the handover is agreed, the candidate target base station performs access control and radio resource allocation, and feeds back an acknowledgment and relevant configuration information to the serving base station.

[0233] Method 16000 may further include operation S1604, where the serving base station sends handover configuration to the UE via an RRC Reconfiguration message, and the handover configuration includes handover execution conditions of candidate target cells and configuration information of candidate target cells including target beam information.

[0234] Figure 17 is a flowchart showing a base station-beam handover method 17000 according to an embodiment of the present disclosure.

[0235] As Figure 17 shown, method 17000 includes the following operations:

[0236] Operation S1701, the serving base station SBS notifies the predicted handover base station PBS of the number of CSI-RS or SSB resources required for UE access, or notifies the PBS of the target BS-beam prediction result of the SBS for the PBS to determine the number of CSI-RS or SSB resources configured by itself. For example, the target BS-beam prediction result may include the probability of each alternative beam of the PBS predicted by the SBS.

[0237] S1702, the PBS configures CSI-RS or SSB resources for UE access, and the PBS subsequently uses these CSI-RS or SSB resources to send reference signals to the UE in the predicted high-quality beam direction.

[0238] S1703, the PBS notifies the SBS of the configured CSI-RS or SSB resources (when the PBS does not configure CSI-RS or SSB resources, the PBS does not provide access services for this UE).

[0239] S1704, the SBS notifies the UE of the CSI-RS or SSB resources configured by the PBS. Optionally, the SBS also notifies the UE of the predicted UE receiving beam direction corresponding to the transmission beam of the PBS.

[0240] S1705, the UE measures the beam reception quality of the PBS on the CSI-RS or SSB resources configured by the PBS (the metrics may include but are not limited to RSRP).

[0241] S1706, the UE feeds back its decision to access the PBS to the SBS.

[0242] S1707, when the UE decides to access the PBS, the SBS notifies the UE of the necessary access information of the PBS via an RRC Reconfiguration message, including but not limited to the physical layer cell identity PCI and the master information block MIB of the PBS.

[0243] S1708, the UE initiates an uplink random access request to the PBS.

[0244] In some embodiments, step S1706 may not need to be performed, and S1707 may be incorporated into step S1704. That is, in S1704, the SBS also notifies the UE of the necessary access information of the PBS through the RRC Reconfiguration message, including but not limited to the physical layer cell identity (PCI) of the PBS and the master information block (MIB). In this case, when the UE decides to access the PBS, the UE initiates an uplink random access request to the PBS.

[0245] Taking RLF as an example, in the traditional RRC reconstruction process relying on SSB broadcasting, the SSB itself contains information such as the cell ID (carried by PSS / SSS, and PSS / SSS has a synchronization function), MIB (carried by PBCH), etc. for access. When using the prediction result to assist handover, the UE does not need to obtain the above information through the traditional process. Using the SBS as the medium for transmitting the PBS access reference signal resources and information to assist the UE in accessing the PBS; at the same time, assuming that adjacent BSs are basically synchronized, the UE can directly receive the CSI-RS or SSB of the PBS without performing a dedicated synchronization process and beam scanning process.

[0246] In some embodiments, when the PBS includes multiple BSs, the SBS may repeatedly execute S1701 - S1706 from high to low according to the predicted priority of the PBS until the UE decides to access the PBS or the number of attempts to access the PBS reaches a preset upper limit.

[0247] · Training of the BS-beam switching joint prediction model

[0248] The present disclosure also contemplates collecting a training data set for training the BS-beam switching joint prediction model. The input in the training data set may be the link condition information related to the UE over a period of time, and the output may be the event type, event occurrence time, and target handover base station-beam of the RLF or handover-related event of the UE, where the training data set is obtained through the traditional base station handover or RRC reconstruction process after the occurrence of RLF.

[0249] Figure 18 is a flowchart showing an exemplary method 18000 for collecting training data according to an embodiment of the present disclosure. As Figure 18 shown, method 1800 may include the following operations:

[0250] S1801, the SBS obtains wireless link condition information.

[0251] S1802, when RLF or handover occurs, the UE accesses the new base station using the traditional RRC reconstruction or base station handover process.

[0252] S1803, the UE notifies the PCI of the previous serving BS (SBS) to the newly accessed target serving BS (TBS).

[0253] S1804, the TBS notifies the SBS of its own PCI and serving beam through the Xn interface.

[0254] In order to better optimize the service for the AI model, during the AI model training data collection phase, in addition to the TBS feeding back the index of the TBS and the serving beam to the SBS, the service quality information (such as the RSRP of the serving beam) can also be further fed back to optimize the AI prediction model.

[0255] As described above, during the model training data collection phase, the SBS obtains the prediction input, and the TBS accessed by the UE and the serving beam label and the event occurrence time obtained through the traditional RRC reconstruction or BS handover process are fed back to the SBS, so that the model can learn the mapping between the prediction input and the output. It should be noted that when the prediction model is deployed in the SBS, after the UE accesses the TBS using the traditional RRC reconstruction process, the TBS cannot know the PCI of the SBS, so it cannot notify the SBS of the TBS-related information for model training. Therefore, the UE feeds back the PCI of the SBS to the TBS, so that the TBS can notify the SBS of its own PCI and serving beam.

[0256] In step S1804, the TBS can also notify the SBS of the occurrence time of RLF and handover through the Xn interface.

[0257] It can be understood that for BS handover, the handover BS knows the time when the UE performs BS handover, so the handover time can be included in the data transmission from the handover BS to the serving BS through the Xn interface as the training data output. For RLF, the handover BS does not know the true time when the UE has an RLF because the possible waiting time for RRC reconstruction is very long and the time of RRC reconstruction is not equal to the time of RLF. Therefore, for the occurrence time of RLF, the UE can report the occurrence time of RLF to the handover BS, and then the handover BS notifies the serving BS deploying the AI prediction model through the Xn interface.

[0258] As mentioned before, BS-beam handover joint prediction relies on accurate modeling of the surrounding radio environment. However, traditional methods are difficult to adaptively model the high-dimensional radio environment characteristics. Therefore, deep learning is considered to implement this prediction. Further, there may be many adjacent BSs around the SBS. Selecting the target beam, such as the optimal beam, from all the alternative beams of these adjacent BSs will bring a large prediction output space. However, when the UE moves to the edge area of the SBS cell and may trigger RLF, only the beams of a small number of adjacent BSs may be the target beams, such as Figure 19As shown in the figure. Therefore, in order to reduce the prediction output space, the prediction output of deep learning can only consider the BS-beam set that appears most frequently in the training dataset and whose total proportion is greater than a predetermined threshold (e.g., 99% or 99.9%).

[0259] The BS-beam switching joint prediction model can adopt a deep learning model. In some embodiments, the prediction output of the BS-beam switching joint prediction model can only include the base station-beam set that appears most frequently in the training dataset and whose total proportion is greater than a predetermined threshold. For example, assume that during the training data collection phase, the total number of tags obtained by the serving BS includes J beams from N neighboring BSs (assuming each neighboring BS has M alternative beams, for a total of MN alternative beams), where J is less than MN. The reason why the number of beams J in the tag is less than or equal to the total number of alternative beams MN is that only the beams adjacent to the serving BS can become handover beams (as Figure 19 shown). Further, we sort the occurrence frequencies of the J beams in the tag from high to low, and only select the top K beams with the highest frequencies as the prediction output tags. The preferred criterion for K is that the total frequency of the K beams accounts for a proportion greater than a predetermined threshold in all tags, such as 99% or 99.9%, so as to further reduce the prediction output space and reduce the output burden of the AI model with almost no reduction in training samples.

[0260] More specifically, a specific [PCI-beam] pairing can represent a possible output. For example, [Cell 7 - Beam 2], [Cell 9 - Beam 2], and [Cell 7 - Beam 6] represent three different possible outputs. During the training phase, a large amount of training data can be collected, and each training data contains a specific [PCI-beam] pairing. For a serving base station, among all the surrounding neighboring base stations, only a few beams may become the handover beams of the current serving base station. Therefore, we can use the [PCI-beam] pairing as different outputs of the training samples. For example, assume that we do not consider the six [PCI-beam] pairings in all the current training data: [Cell 7 - Beam 2], [Cell 7 - Beam 4], [Cell 7 - Beam 6], [Cell 9 - Beam 1], [Cell 9 - Beam 2], and [Cell 3 - Beam 5], with proportions of 30%, 20%, 20%, 20%, 9.9%, and 0.1% respectively, and the proportion threshold is set to 99%. Then, first, we find the set of pairings whose sum of proportions is greater than the threshold in descending order of proportion. We find that the top 5 already reach 99%. Then, we exclude the [PCI-beam] pairing with a probability of 0.1%, do not include the radio link status information corresponding to the [PCI-beam] pairing with a probability of 0.1% in the input, and the output is a 5x1 vector, where each element represents the probability of determining that the handover should be to the corresponding [PCI-beam] pairing.

[0261] As described above, the BS-beam switching joint prediction model may include one or more AI sub-models deployed on the serving base station and / or the UE. For example, for the case where the BS-beam switching joint prediction model 500 shown in Figure 5 includes one AI sub-model 501 deployed on the serving base station, the training data set of the AI model may include: the input is the radio link information between the UE and the serving base station and / or neighboring base stations over a period of time, and the output is the type of event that occurs at the UE, the time, and the optimal handover BS and beam.

[0262] For example, for the case where the BS-beam switching joint prediction model 700 shown in Figure 7 includes two AI sub-models deployed on the serving base station, two training data sets may be used. The training data set for the second AI sub-model 701 may include: the input may be the radio link information between the UE and the serving base station and / or neighboring base stations over a period of time, and the output may be the type of event that occurs at the UE, the time, for training the RLF or handover prediction model. The training data set for the third AI sub-model 702 may include: the input may be the radio link information between the UE and the serving base station and / or neighboring base stations over a period of time and possible additional measurements (such as the BFR measurement results obtained after the A3 event is triggered), and the output may be the optimal handover BS and beam, for training the BS / beam prediction model.

[0263] Those skilled in the art can construct different training data sets as needed to train each AI sub-model included in the BS-beam switching joint prediction model in the present disclosure.

[0264] ·Enabling of AI-assisted services and activation and deactivation of the BS-beam switching joint prediction model for the UE

[0265] The present disclosure can be used for UEs with low tolerance for latency (such as in the ultra-reliable low-latency communication uRLLC scenario). The present disclosure considers the conditions for providing AI-assisted services, and the characteristics of the conditions may be that the UE feeds back its own tolerance for latency to the serving BS, and the serving BS decides whether to provide / activate / enable the AI-assisted services based on the tolerance for latency. The AI-assisted services may include BS-beam switching joint prediction and other services according to the embodiments of the present disclosure.

[0266] Figure 20 is a flowchart showing a method 20000 for activating an AI-assisted RLF / handover prediction service according to an embodiment of the present disclosure. As shown in Figure 20 shown, the method 20000 may include operation S2001, in which, for the BS configured with AI-assisted services, the RRC configures a tolerance for latency set for the user to feed back the tolerance for latency.

[0267] Method 20000 may further include operation S2002, in which, after the UE accesses the BS, the serving BS notifies the UE of the tolerated delay set information through an RRCReconfiguration message.

[0268] Method 20000 may further include operation S2003, in which the UE feeds back its own tolerated delay to the BS.

[0269] Method 20000 may further include operation S2004, in which, when the tolerated delay of the UE changes, the UE notifies the BS of the changed tolerated delay.

[0270] Method 20000 may further include operation S2005, in which the BS activates or deactivates the AI-assisted RLF / handover prediction service based on the UE feedback or the updated tolerated delay.

[0271] In some embodiments, when the tolerated delay of the UE is less than the preset threshold of the BS, the BS may activate the AI-assisted service. When the tolerated delay of the UE is greater than the preset threshold of the BS, the BS may deactivate the AI-assisted service.

[0272] For the case where the BS serves multiple UEs, if the tolerated delay of at least one UE fed back is less than the preset threshold of the BS, the AI-assisted service may be enabled at the BS (i.e., the AI function is enabled). For the UEs whose tolerated delays meet the requirements, according to the specific situation, for each of these UEs, an AI model such as the BS-beam switching joint prediction model according to the embodiments of the present disclosure and possibly other supported AI models may be activated / applied.

[0273] Figure 21 It is a flowchart showing a method 21000 for activating and deactivating the BS-beam switching joint prediction model for a UE according to an embodiment of the present disclosure.

[0274] As Figure 21 shown, method 21000 includes the following operations:

[0275] Operation S2101, in which, for example, when the communication quality of a UE is lower than a threshold, the SBS activates the BS-beam switching joint prediction model for the UE. Activating the BS-beam switching joint prediction model for the UE may also be based on, for example, the anticipation that an A3 event is about to occur.

[0276] Operation S2102, in which, after the handover is completed, the SBS deactivates the BS-beam switching joint prediction model for the UE. After the handover is completed, the SBS may simultaneously deactivate other AI models for the UE, including but not limited to one or more AI models for CSI feedback, in-beam management within the TBS, wireless positioning, etc.

[0277] Operation S2103: When the UE completes handover and accesses the TBS, when the communication quality of the UE is lower than the threshold, the TBS activates the BS-beam switching joint prediction model for this UE. When the UE completes handover and accesses the TBS, the TBS can activate other AI models for this UE, including but not limited to one or more AI models for CSI feedback, beam management within the TBS, wireless positioning, etc.

[0278] · Update and optimization of the BS-beam switching joint prediction model

[0279] In a real scenario, the wireless environment is constantly changing, and the model prediction performance often gradually decreases over time. This disclosure considers triggering model updates when necessary to ensure prediction accuracy.

[0280] In some embodiments, online updates of the BS-beam switching joint prediction model are provided. When the prediction accuracy metrics (including but not limited to the success rate of BS handover under a predetermined overhead) do not meet the predetermined threshold, model online update is activated, that is, the process described above is used to collect training data to update the model. The above-mentioned predetermined threshold includes but not limited to a fixed threshold, a relative threshold compared to the prediction accuracy when the model is initially deployed. Figure 18 The above-mentioned predetermined threshold includes but not limited to a fixed threshold, a relative threshold compared to the prediction accuracy when the model is initially deployed.

[0281] For example, the success rate of BS handover can be given by S1706 of the BS-beam switching process as shown in Figure 17 . When the UE decides to access the PBS, it is regarded as a successful handover; when the UE decides not to access the PBS under the preset handover overhead (for example, measuring no more than 8 beams or attempting to access no more than 2 PBSs), it is regarded as a failed handover.

[0282] Figure 22 An example of triggering model update based on a relative threshold according to an embodiment of the present disclosure is shown. When the prediction model starts to be deployed, it is considered that the prediction model can match the current environment. At this time, the initial handover success rate r0 = 89% reflects the upper limit of the model's ability in the current wireless environment. As the wireless environment fluctuates, the prediction accuracy of the model gradually decreases. Set the relative threshold as r th = 0.9×r0 = 80.1%. When the monitored handover success rate is less than the above threshold within a certain period of time, trigger model update.

[0283] Traditional fixed thresholds are difficult to adapt to the differences in the prediction accuracy of the model itself in different environments. Compared with fixed thresholds, this relative threshold-based mechanism that uses the prediction accuracy at the initial deployment moment of the model (least affected by environmental changes) as a reference can better adapt to different wireless environments.

[0284] After the user switches to TBS, if they switch again or experience RLF within a preset time threshold, TBS can notify this event to the SBS through the Xn interface. This time threshold is usually short to evaluate whether ping-pong effects or non-optimal handovers occur. This failure feedback can be used to optimize the prediction of RLF / handoff events, PBS prediction, and access strategies in the AI model (e.g., reinforcement learning can adjust strategies based on the above feedback) to reduce the impact of ping-pong effects and the like.

[0285] · Simulation

[0286] Consider the Figure 23 wireless channel environment shown in (see A. Alkhateeb, “DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications,” in Proc. ITA Workshop, San Diego, CA, USA, Feb. 2019, pp. 1–8). The Saleh-Valenzuela channel model is used to calculate the CSI based on the characteristics of each path's attenuation, delay, angle of arrival (AoA), angle of departure (AoD), etc. The SBS index of the UE is 5 (as shown by the white dashed circle), and the indices of the adjacent 4 BSs are 2, 11, 14, 18 (as shown by the white solid circles). The number of alternative beams for each BS is 32, and the adjacent BSs together contain 128 alternative beams. Other simulation parameters are shown in Table 1.

[0287] Parameter Value SBS Index 5 Adjacent BS Index 2,11,14,18 Center Frequency 28 GHz Number of BS Antennas 32 (ULA) Number of UE Antennas 1 Number of BS Candidate Beams 32 Bandwidth 50 MHz Noise Figure 5 dB

[0288] Table 1 Channel simulation parameters

[0289] For the BS-beam handover joint prediction model, we implement it using a convolutional neural network (CNN). Its specific structure and parameters are shown in Table 2, where f i and f o represent the number of input and output feature channels respectively, (a, b, c) represent the convolution kernel size, downsampling step size, and padding size of the convolutional layer, BatchNorm refers to batch normalization, AvgPooling refers to average pooling, ReLU refers to the ReLU activation function, dropout refers to the random dropout strategy to suppress overfitting, and N out refers to the output size. In the training dataset, only 62 out of the 128 alternative beams of all adjacent BSs are likely to be switched to. To effectively reduce the prediction output space, N out = 62 in the simulation.

[0290]

[0291] Table 2 Prediction Model Structure and Parameters

[0292] Figure 24 Fig. shows the normalized beam gain performance of BS-beam switching joint prediction according to an embodiment of the present disclosure, which is calculated as the gain ratio between the predicted optimal beam and the actual optimal beam. This figure further considers a certain number of beams with the highest measurement prediction probability (as the abscissa), selects the beam with the highest received power as the predicted optimal beam and evaluates its normalized beam gain. It can be seen that the predicted optimal beam can reach an average normalized beam gain of 64%; further considering the measurement of the 5 highest probability beams, the proposed model can reach a normalized beam gain of 93%, that is, almost perfect beam alignment.

[0293] Figure 25 Fig. shows the curve of the accuracy rate of BS prediction according to an embodiment of the present disclosure varying with the number of base stations. This accuracy rate is calculated as the proportion that the predicted optimal BS is the actual optimal BS (target beam), and the actual optimal BS is defined as the TBS corresponding to the optimal BS-beam switching. Since the prediction result is the beam probability, we sum the beam probabilities of each BS as the BS probability. Similar to Figure 24 Fig., this figure further considers a certain number of BSs with the highest measurement prediction probability as the abscissa. It can be seen that the prediction accuracy rate of the optimal BS is 78.6%, and when searching for two BSs, there is a 96% probability of finding the optimal BS.

[0294] Finally, for RLF events, we evaluate the delay comparison between the traditional RRC reconstruction scheme and the proposed scheme. Assume that the SSB broadcast period of adjacent BSs is τ SSB = 80 ms, and RLF may be triggered at any time. For the traditional RRC reconstruction process, the average delay for monitoring the TBS to broadcast the SSB is τ SSB / 2 = 40 ms, and 32 alternative beams need to be measured to find the optimal beam. For the proposed scheme, as shown in Figure 24 Fig., almost perfect beam alignment can be achieved when only measuring 5 beams. Therefore, the proposed scheme can significantly reduce the delay of BS switching.

[0295] · Serving base station

[0296] Figure 26A Fig. shows a block diagram of an exemplary configuration of base station 2610 according to some embodiments of the present disclosure. Base station 2610 can implement a serving base station according to an embodiment of the present disclosure.

[0297] As shown in the figure, the base station 2610 may include a processing circuit 2613. The processing circuit 2613 may include a pre-judgment unit 2611. The pre-judgment unit 2611 is configured to perform operations related to the pre-judgment of RLF or handover. In some embodiments, the pre-judgment unit may be configured to pre-judge that a radio link failure (RLF) or handover will occur at least based on radio link condition information related to a user equipment (UE), as Figure 3A shown in operation 3101. In some embodiments, the pre-judgment unit may be configured to pre-judge that a radio link failure (RLF) or handover will occur at least based on radio link condition information related to a user equipment (UE), and use an AI model to pre-judge, as Figure 3B shown in operation 3201 or Figure 3D shown in operation 3401.

[0298] As shown in the figure, the processing circuit 2613 may further include a prediction unit 2612. The prediction unit 2612 is configured to perform operations related to the prediction of a target BS-beam set. In some embodiments, the prediction unit 2612 may be configured to predict, at least based on radio link condition information related to a user equipment (UE), the target base station and the beam used by the target base station to serve the user equipment by using an AI model, as Figure 3C shown in operation 3301, or as Figure 3D shown in operation 3402.

[0299] Those skilled in the art can understand that the processing circuit 2613 may include more or fewer units than Figure 26A shown.

[0300] For example, in some embodiments, the processing circuit 2613 may not include the pre-judgment unit 2611, and the UE may perform operations related to the pre-judgment of RLF or handover. In some embodiments, the base station 2610 may not include the prediction unit 2611. After making a pre-judgment of RLF or handover, the base station 2610 may be configured for conditional handover or assisted beam recovery according to conventional methods.

[0301] Figure 26B The block diagram shows an exemplary configuration of a base station 2620 according to some embodiments of the present disclosure. The base station 2620 may implement a serving base station according to the embodiments of the present disclosure.

[0302] As Figure 26B shown, the base station 2620 may include a processing circuit 2624. The processing circuit 2624 may include a first unit 2621, which is configured to predict the occurrence probability of RLF or handover and a target base station-beam set at least based on radio link condition information related to the UE by using a first AI sub-model.

[0303] As Figure 26BAs shown, the processing circuit 2624 may further include a second unit 2622 configured to predict that an RLF or a handover will occur based on a comparison between the occurrence probability of the RLF or the handover and a predetermined threshold.

[0304] As Figure 26B shown, the processing circuit 2624 may further include a third unit 2623 configured to use the target base station-beam set for conditional handover configuration in response to the prediction.

[0305] Figure 26C A block diagram showing an exemplary configuration of a base station 2630 according to some embodiments of the present disclosure. The base station 2630 may implement a serving base station according to an embodiment of the present disclosure.

[0306] As Figure 26C shown, the base station 2630 may include a processing circuit 2635. The processing circuit 2635 may include a first unit 2631 configured to predict the occurrence probability of an RLF or a handover using a second AI sub-model based on radio link condition information related to the UE.

[0307] As Figure 26C shown, the processing circuit 2635 may further include a second unit 2632 configured to predict that an RLF or a handover will occur based on a comparison between the occurrence probability of the RLF or the handover and a predetermined threshold.

[0308] As Figure 26C shown, the processing circuit 2635 may further include a third unit 2633 configured to, in response to the prediction, the serving base station may collect supplementary radio link condition information related to the UE.

[0309] As Figure 26C shown, the processing circuit 2635 may further include a fourth unit 2634 configured to, in response to the prediction, predict a target base station-beam set for conditional handover configuration based on the radio link condition information related to the UE and the supplementary radio link condition information related to the UE using a third AI sub-model.

[0310] Those skilled in the art can understand that Figures 26A - 26C only some exemplary unit configurations of the base station are shown, and those skilled in the art can implement various corresponding units according to the operations to be performed by the base station. More specifically, generally speaking, a base station includes a processor (processing circuit) and a memory (not shown), and the memory stores computer-readable instructions that contain corresponding instructions for various operations to be performed by the base station. For example, for a serving base station according to an embodiment of the present disclosure, the computer-readable instructions may include, but are not limited to, for example, for performing Figure 3A operation 3101 in Figure 3B operation 3201 inFigure 3C Operation 3301 in Figure 3D Operations 3401 and 3402 in Figure 6 Operations 601 - 603 in Figure 8 Operations 801 - 804 in Figures 11 - 13 Figures 16 - 18 , and Figures 20 - 21 Instructions for operations such as the operations executed by SBS in, etc., the processor can implement the functions of the units corresponding to these operations by reading and executing the corresponding instructions stored in the memory for various operations to be performed by the base station.

[0311] · User Equipment

[0312] Figure 27A The flowchart of an exemplary method 2710 for wireless communication according to an embodiment of the present disclosure is shown. This method can be executed by a user equipment according to an embodiment of the present disclosure.

[0313] As Figure 27A shown, method 2710 may include operation 2711, in which the UE uses a first artificial intelligence (AI) model to predict the occurrence probability of a radio link failure (RLF) or a handover, and the prediction is at least based on radio link condition information related to the user equipment (UE), and the radio link condition information at least includes information reflecting the condition of the radio link between the UE and the serving base station. The first AI model may be similar to the aforementioned fifth AI sub - model.

[0314] As Figure 27A shown, method 2710 may include operation 2712, in which (A) the UE sends the predicted occurrence probability of RLF or handover to the serving base station, or (B) the UE pre - judges that an RLF or handover will occur based on the comparison between the occurrence probability of RLF or handover and a predetermined threshold, and in response to this pre - judgment, notifies the serving base station that an RLF or handover is pre - judged to occur.

[0315] As Figure 27A shown, method 2710 may include operation 2713, in which the UE receives a conditional handover configuration from the serving base station, and the conditional handover configuration is based on the target base station predicted by the serving base station using a second AI model in response to the pre - judgment notification received from the UE and the beam of the target base station for serving the user equipment, where the prediction of the serving base station is at least based on the radio link condition information related to the UE. The second AI model is similar to the aforementioned sixth AI sub - model.

[0316] Figure 27B The flowchart of an exemplary method 2720 for wireless communication according to an embodiment of the present disclosure is shown. This method can be executed by a user equipment according to an embodiment of the present disclosure.

[0317] As shown Figure 27B in FIG. 2720, method 2720 may include operation 2721, in which the UE receives CSI-RS or SSB resources configured by the PBS from the SBS.

[0318] Method 2720 may further include operation 2722, in which the UE measures the beam reception quality of the PBS or SSB on the CSI-RS or SSB resources configured by the PBS.

[0319] Method 2720 may further include operation 2723, in which the UE initiates an uplink random access request to the PBS when the UE decides to access the PBS.

[0320] Figure 27C FIG. 2730 is a block diagram showing an exemplary configuration of a user equipment 2730 according to an embodiment of the present disclosure.

[0321] As shown Figure 27C in FIG. 2730, the user equipment 2730 may include a processing circuit 2734. The processing circuit 2734 may include a prediction unit 2731. The prediction unit 2731 is configured to perform operations related to the prediction of RLF or handover. In some embodiments, the prediction unit may be configured to predict the occurrence of a radio link failure (RLF) or handover based at least on radio link condition information related to the user equipment (UE), as shown Figure 3A in operation 3101. In some embodiments, the prediction unit may be configured to predict the occurrence of a radio link failure (RLF) or handover based at least on radio link condition information related to the user equipment (UE) using an AI model, as shown Figure 3B in operation 3201.

[0322] In some embodiments, the prediction unit 2731 may be configured to predict the probability of occurrence of a radio link failure (RLF) or handover. In other embodiments, the prediction unit 2731 may be configured to predict the probability of occurrence of a radio link failure (RLF) or handover and predict the occurrence of RLF or handover based on a comparison of the probability of occurrence of RLF or handover with a predetermined threshold.

[0323] As shown Figure 27C in FIG. 2730, the processing circuit 2734 may further include a transmission unit 2732. The transmission unit 2732 may be configured to send the probability of occurrence of RLF or handover predicted by the prediction unit 2731 to the serving base station, or may be configured to send the prediction result of the prediction unit 2731 to the serving base station.

[0324] As shown Figure 27CAs shown, the processing circuit 2734 may further include a receiving unit 2733. The receiving unit 2733 may be configured to receive a conditional handover configuration for the serving base station, where the conditional handover configuration is based on a target base station predicted by the serving base station using a second AI model in response to the predicted notification received from the UE, and a beam used by the target base station to serve the user equipment, where the prediction of the serving base station is at least based on the radio link status information related to the UE.

[0325] Figure 27D A block diagram showing an exemplary configuration of a user equipment 2740 according to an embodiment of the present disclosure.

[0326] As Figure 27D shown, the user equipment 2740 may include a processing circuit 2744. The processing circuit 2744 may include a first unit 2741 configured to receive CSI-RS or SSB resources configured for PBS from the SBS.

[0327] The processing circuit 2744 may further include a second unit 2742 configured to measure the beam reception quality of PBS or SSB on the CSI-RS or SSB resources configured for PBS.

[0328] The processing circuit 2744 may further include a third unit 2743 configured to initiate an uplink random access request to the PBS when the UE decides to access the PBS.

[0329] Those skilled in the art can understand that Figures 27A - 27D only some exemplary operations and unit configurations of the user equipment are shown, and those skilled in the art can implement various corresponding units according to the operations to be performed by the user equipment. More specifically, generally speaking, the user equipment includes a processor (processing circuit) and a memory (not shown), and the memory stores computer-readable instructions that contain corresponding instructions for various operations to be performed by the user equipment. For example, for the user equipment according to an embodiment of the present disclosure, the computer-readable instructions may include, but are not limited to, for example, instructions for performing Figure 3A operation 3101 in Figure 3B operation 3201 in Figure 3C operation 3301 in Figure 3D operation 3401 in Figures 11 - 13 , Figures 16 - 18 , Figures 20 - 21 and Figures 27A - 27B operations performed by the UE in

[0330] Those skilled in the art can understand that the structures and deployments of the first AI sub-model to the seventh AI sub-model mentioned in this disclosure are schematically shown for illustrative purposes and do not represent a limitation to this disclosure. Those skilled in the art can design and deploy AI models with different structures and different numbers according to needs to implement the base station-beam switching joint prediction model. The expressions "first" to "seventh" are only for clear distinction and do not represent any order or relevance.

[0331] Next, an electronic device and a communication method according to some embodiments of the present disclosure will be described.

[0332]

Exemplary Implementations of the Present Disclosure

[0333] According to the embodiments of the present disclosure, various implementation manners for implementing the concepts of the present disclosure can be conceived, including but not limited to:

[0334] 1. A method for wireless communication, including:

[0335] Using an artificial intelligence (AI) model to predict that a radio link failure (RLF) or a handover will occur, where the prediction is at least based on radio link status information related to a user equipment (UE), and the radio link status information at least includes information reflecting the status of the radio link between the UE and a serving base station (SBS).

[0336] 2. The method according to item 1, further including:

[0337] In response to the prediction, using the AI model to predict a target base station and a beam used by the target base station to serve the user equipment for configuring a conditional handover, where the prediction is at least based on the radio link status information related to the UE.

[0338] 3. The method according to item 2, where the output of the AI model includes a target base station-beam set, and the target base station-beam set includes one or more target base station-beam pairings, and each pairing indicates a predicted target base station and a beam used by the target base station to serve the user equipment.

[0339] 4. The method according to item 1, where the radio link status information further includes information reflecting the status of the radio link between the UE and one or more neighboring base stations.

[0340] 5. The method according to item 3, where the radio link status information includes at least one of the following:

[0341] Beam measurement results;

[0342] Channel state information (CSI) measurement results;

[0343] Mobility measurement results;

[0344] Real-time geographical environment and radio environment information including at least one of a high-precision map and an electromagnetic map;

[0345] The location and movement characteristics of the UE; or

[0346] Data of sensors including at least one of an accelerometer and a barometer of the UE.

[0347] 6. The method according to item 5, wherein the AI model includes one or more AI sub-models, and the input of each AI sub-model in the AI model includes at least radio link condition information related to the UE.

[0348] 7. The method according to item 6, wherein the AI model includes a first AI sub-model deployed at the serving base station, and the method further includes:

[0349] Based at least on radio link condition information related to the UE, the serving base station uses the first AI sub-model to predict the occurrence probability of RLF or handover and the target base station-beam set;

[0350] The serving base station pre-judges that RLF or handover will occur based on the comparison between the occurrence probability of RLF or handover and a predetermined threshold; and

[0351] In response to the pre-judgment, the serving base station uses the target base station-beam set for configuring conditional handover.

[0352] 8. The method according to item 6, wherein the AI model includes a second AI sub-model and a third AI sub-model deployed at the serving base station, and the method further includes:

[0353] Based on radio link condition information related to the UE, the serving base station uses the second AI sub-model to predict the occurrence probability of RLF or handover;

[0354] The serving base station pre-judges that RLF or handover will occur based on the comparison between the occurrence probability of RLF or handover and a predetermined threshold; and

[0355] In response to the pre-judgment, based on the radio link condition information related to the UE and the supplementary radio link condition information related to the UE, the serving base station uses the third AI sub-model to predict the target base station-beam set for configuring conditional handover.

[0356] 9. The method according to item 7 or 8, wherein the AI model further includes a fourth AI sub-model deployed at the UE, and the method further includes:

[0357] At least based on the radio link condition information related to the UE, the UE uses the fourth AI sub-model to predict the occurrence probability of RLF or handover;

[0358] The UE sends the predicted occurrence probability of RLF or handover to the serving base station; and

[0359] Wherein the occurrence probability of RLF or handover of the UE is included in the radio link condition information related to the UE.

[0360] 10. The method according to item 6, wherein the AI model includes a fifth AI sub-model deployed at the UE and a sixth AI sub-model deployed at the serving base station, and the method further includes:

[0361] At least based on the radio link condition information related to the UE, the UE uses the fifth AI sub-model to predict the occurrence probability of RLF or handover;

[0362] The UE sends the predicted occurrence probability of RLF or handover to the serving base station;

[0363] The serving base station pre-judges that RLF or handover will occur based on the comparison between the predicted occurrence probability of RLF or handover received from the UE and a predetermined threshold; and

[0364] In response to the pre-judgment, at least based on the radio link condition information related to the UE, the serving base station uses the sixth AI sub-model to predict the target base station-beam set for configuring conditional handover.

[0365] 11. The method according to item 5, wherein the beam measurement is triggered based on at least one of being lower than a predetermined communication quality threshold and in response to the pre-judgment that RLF or handover will occur,

[0366] Wherein the beam measurement is based on at least one of the following:

[0367] (1) Adjacent beam measurement, wherein the beams adjacent to the current serving beam in terms of angle are measured; or

[0368] (2) Probability-priority-based beam measurement, wherein a number of beams selected based on the magnitude of the probability of each beam becoming the optimal beam are measured.

[0369] 12. The method according to item 5, further including:

[0370] The serving base station configures the channel state information reference signal (CSI-RS) resources for downlink beam failure recovery (BFR) measurement;

[0371] The serving base station sends CSI-RS for the UE to perform downlink BFR measurement;

[0372] The UE sends the downlink BFR measurement result to the serving base station; and

[0373] The UE sends the UE receive beam pattern to the serving base station,

[0374] wherein the beam measurement result includes the downlink BFR measurement result and the UE receive beam pattern.

[0375] 13. The method according to item 5, further comprising:

[0376] The serving base station configures the sounding reference signal (SRS) resources for uplink BFR measurement;

[0377] The serving base station notifies the UE of the configured SRS resources;

[0378] The UE sends the SRS;

[0379] The serving base station performs uplink BFR measurement to obtain the uplink BFR measurement result; and

[0380] The UE feeds back the UE transmit beam pattern to the serving base station,

[0381] wherein the beam measurement result includes the uplink BFR measurement result and the UE transmit beam pattern.

[0382] 14. The method according to item 12 or 13, further comprising: the UE feeds back the absolute angle information of the UE-side beam.

[0383] 15. The method according to item 14, further comprising:

[0384] The serving base station configures the granularity of the absolute beam angle feedback for the UE;

[0385] The serving base station notifies the UE of the granularity of the absolute beam angle feedback; and

[0386] The UE feeds back the absolute beam angle to the serving base station according to the configured granularity.

[0387] 16. The method according to item 3, further comprising:

[0388] The serving base station selects one or more candidate target base stations based on the predicted target base station-beam set;

[0389] The serving base station sends a handover request to the candidate target base station among the one or more candidate target base stations;

[0390] If the handover is agreed, the candidate target base station performs access control and radio resource allocation, and feeds back an acknowledgment and relevant configuration information to the serving base station; and

[0391] The serving base station sends the switching configuration to the UE through an RRC reconfiguration message, where the switching configuration includes the switching execution conditions of the candidate target cell and the configuration information of the candidate target cell including the target beam information.

[0392] 17. The method according to item 16, further comprising:

[0393] The SBS notifies a predicted switching base station (PBS), i.e., a candidate target base station, of the number of CSI-RS or SSB resources required for UE access, or notifies the PBS of the prediction result of the SBS for the PBS to decide the number of CSI-RS or SSB resources to be configured;

[0394] CSI-RS or SSB resources configured by PBS for UE access;

[0395] The PBS notifies the SBS of the configured CSI-RS or SSB resources;

[0396] The SBS notifies the UE of the CSI-RS or SSB resources configured by the PBS;

[0397] The UE measures the beam reception quality of the PBS or SSB on the CSI-RS or SSB resources configured by the PBS; and

[0398] When the UE decides to access the PBS, the UE initiates an uplink random access request to the PBS.

[0399] 18. The method of item 17, wherein there are multiple PBSs, and switching attempts are performed sequentially for the multiple PBSs according to the predicted priorities until the switching is successful or a preset upper limit of the number of base stations is reached.

[0400] 19. The method according to claim 17, further comprising:

[0401] After the user equipment is switched to the PBS, if an RLF or a switching-related event occurs again within a predetermined time threshold, the PBS notifies the SBS of the event via the Xn interface.

[0402] 20. The method according to claim 2, further comprising:

[0403] The serving base station activates the AI-assisted service based on the delay tolerance information fed back by the UE.

[0404] The serving base station activates the AI ​​model for the UE based on the communication quality of the UE; and

[0405] After the UE is switched from the serving base station to another base station, the AI ​​model is deactivated for the UE.

[0406] 21. The method according to item 1 further includes collecting a training data set for training the AI model, where the input in the training data set includes link condition information related to the UE over a period of time, and the output includes the event type of the RLF or handover-related event of the UE and the target handover base station-beam, and the training data set is obtained through the traditional base station handover or the RRC reconstruction process after the occurrence of RLF.

[0407] 22. The method according to item 21, wherein the AI model adopts a deep learning model, and the output of the AI model includes a set of base station-beams with the highest occurrence frequency in the training data set and a total frequency greater than a predetermined threshold.

[0408] 23. The method according to item 21 further includes:

[0409] In response to the prediction accuracy of the AI model not meeting a predetermined accuracy threshold, collecting an online training data set for training the AI model to perform an online update on the AI model, where the predetermined accuracy threshold includes a relative threshold compared to the prediction accuracy when the AI model is initially deployed.

[0410] 24. A system for wireless communication includes:

[0411] One or more processors,

[0412] A memory that stores computer-readable program instructions, and when the computer-readable program instructions are executed by the one or more processors, the method according to any one of items 1-23 is executed.

[0413] 25. A non-transitory computer-readable storage medium stores computer-readable program instructions, and when the computer-readable program instructions are executed by one or more processors, the method for wireless communication according to any one of items 1-23 is executed.

[0414] 26. A method executed by a user equipment (UE) includes:

[0415] Using a first artificial intelligence (AI) model to predict the occurrence probability of a radio link failure (RLF) or a handover, where the prediction is at least based on radio link condition information related to the user equipment (UE), and the radio link condition information at least includes information reflecting the condition of the radio link between the UE and the serving base station.

[0416] 27. The method according to item 26 further includes:

[0417] Sending the predicted occurrence probability of the RLF or the handover to the serving base station.

[0418] 28. The method according to item 27 further includes:

[0419] Predict RLF or handover based on the comparison between the occurrence probability of RLF or handover and a predetermined threshold;

[0420] In response to the prediction, notify the serving base station that RLF or handover is predicted to occur.

[0421] 29. The method according to item 28, further comprising:

[0422] Receive a conditional handover configuration from the serving base station, where the conditional handover configuration is based on the target base station predicted by the serving base station using a second AI model in response to the predicted notification received from the UE and the beam used by the target base station to serve the user equipment, and the prediction of the serving base station is at least based on the radio link condition information related to the UE.

[0423] 30. The method according to item 28, wherein the radio link condition information includes at least one of the following:

[0424] Beam measurement results;

[0425] Channel state information (CSI) measurement results;

[0426] Mobility measurement results;

[0427] Real-time geographical environment and radio environment information including at least one of a high-precision map and an electromagnetic map;

[0428] The location and movement characteristics of the UE; or

[0429] Data of sensors including at least one of the accelerometer and barometer of the UE.

[0430] 31. The method according to item 30, wherein the beam measurement is triggered based on at least one of below a predetermined communication quality threshold and in response to the prediction of RLF or handover occurrence,

[0431] wherein the beam measurement is based on at least one of the following:

[0432] (1) Adjacent beam measurement, where beams adjacent to the current serving beam in terms of angle are measured; or

[0433] (2) Beam measurement based on probability priority, where a number of beams selected based on the magnitude of the probability of each beam becoming the optimal beam are measured.

[0434] 32. The method according to item 31, further comprising:

[0435] Receive, from a serving base station, CSI-RS resources configured by the serving base station for downlink beam failure recovery (BFR) measurement for the UE to perform downlink BFR measurement;

[0436] Send the downlink BFR measurement result to the serving base station; and

[0437] Send the UE receive beam pattern to the serving base station,

[0438] wherein the beam measurement result includes the downlink BFR measurement result and the UE receive beam pattern.

[0439] 33. The method according to item 31, further comprising:

[0440] Receive, from a serving base station, sounding reference signal (SRS) resources configured by the serving base station for uplink BFR measurement;

[0441] Send SRS for the serving base station to perform uplink BFR measurement to obtain an uplink BFR measurement result;

[0442] Feedback the UE transmit beam pattern to the serving base station,

[0443] wherein the beam measurement result includes the uplink BFR measurement result and the UE transmit beam pattern.

[0444] 34. The method according to item 32 or 33, further comprising:

[0445] Feedback the absolute angle information of the UE-side beam.

[0446] 35. The method according to item 32 or 33, further comprising:

[0447] Receive, from a serving base station, the granularity of the absolute beam angle feedback configured for the UE; and

[0448] Feedback the absolute beam angle to the serving base station according to the configured granularity.

[0449] 36. The method according to item 29, wherein the conditional handover configuration is sent by the serving base station to the UE through an RRC reconfiguration message, and the conditional handover configuration includes the handover execution condition of the candidate target cell and the configuration information of the candidate target cell including the target beam information.

[0450] 37. The method according to item 36, further comprising:

[0451] Receive CSI-RS or SSB resources configured by PBS from SBS;

[0452] Measure the beam reception quality of PBS or SSB on the CSI-RS or SSB resources configured by PBS; and

[0453] When the UE decides to access the PBS, it initiates an uplink random access request to the PBS.

[0454] 38. A user equipment (UE), comprising

[0455] one or more processors,

[0456] A memory storing computer-readable program instructions which, when executed by the one or more processors, cause the method described in any one of items 26-37 to be performed.

[0457] 39. A non-transitory computer-readable storage medium storing computer-readable program instructions which, when executed by one or more processors, cause the method performed by a user equipment (UE) as described in any one of items 26-37 to be executed.

[0458] Application examples of the present disclosure

[0459] The techniques described in this disclosure can be applied to a variety of products.

[0460] For example, the electronic device according to the embodiments of the present disclosure may be implemented as various base stations or installed in a base station, or may be implemented as various user equipments or installed in various user equipments.

[0461] The communication method according to the embodiments of the present disclosure can be implemented by various base stations or user equipment; the methods and operations according to the embodiments of the present disclosure can be embodied as computer-executable instructions, stored in a non-temporary computer-readable storage medium, and can be executed by various base stations or user equipment to implement one or more functions described above.

[0462] The functions of the various elements disclosed herein can be implemented using circuits or processing circuits including general-purpose processors, special-purpose processors, integrated circuits, AISCs ("Application Specific Integrated Circuits"), conventional circuits, and / or combinations thereof, which are configured or programmed to perform the disclosed functions. Processors are considered circuits or processing circuits because processors include transistors and other circuits. In the present disclosure, circuits, units, or components are hardware that performs the functions described or is programmed to perform the functions described. Hardware can be any hardware that is disclosed herein or otherwise known and is programmed or configured to perform the functions described. When the hardware is a processor (which is considered a circuit), the circuit, unit, or component is a combination of hardware and software, and the software is used to configure the hardware and / or processor.

[0463] The technology according to the embodiments of the present disclosure can be made into various computer program products, which can be used in various base stations or user equipments to implement one or more functions described above.

[0464] The base station described in the present disclosure can be implemented as any type of base station. Preferably, it is a macro gNB and an ng-eNB defined in the 5G NR standard of 3GPP. The gNB can be a gNB that covers a cell smaller than a macro cell, such as a pico gNB, a femto gNB, and a home (femto) gNB. Instead, the base station can be implemented as any other type of base station, such as a NodeB, an eNodeB, and a base transceiver station (BTS), or a network-side infrastructure in a next-generation communication standard. The base station can also include: a main body configured to control wireless communication, and one or more remote radio heads (RRHs), wireless relay stations, drone towers, control nodes in an automated factory, etc., which are arranged in places different from the main body.

[0465] The user equipment can be implemented as a mobile terminal (such as a smart phone, a tablet personal computer (PC), a notebook PC, a portable game terminal, a portable / dongle-type mobile router, and a digital imaging device) or a vehicle-mounted terminal (such as an automotive navigation device). The user equipment can also be implemented as a terminal that performs machine-to-machine (M2M) communication (also referred to as a machine type communication (MTC) terminal), a drone, sensors and actuators in an automated factory, etc. In addition, the user equipment can be a wireless communication module (such as an integrated circuit module including a single wafer) installed on each of the above terminals.

[0466] Examples of the base station and the user equipment to which the technology of the present disclosure can be applied are briefly introduced below.

[0467] It should be understood that the term "base station" used in the present disclosure has the full breadth of its ordinary meaning and at least includes a wireless communication station used as part of a wireless communication system or a radio system to facilitate communication. In D2D, M2M, and V2V communication scenarios, a logical entity having a control function for communication can also be referred to as a base station. In a cognitive radio communication scenario, a logical entity that plays a role in spectrum coordination can also be referred to as a base station. In an automated factory, a logical entity that provides a network control function can be referred to as a base station.

[0468] The first application example of the base station

[0469] Figure 30 is a block diagram showing a first example of the schematic configuration of a base station to which the technology of the present disclosure can be applied. In Figure 30 it, the base station can be implemented as a gNB 1400. The gNB 1400 includes a plurality of antennas 1410 and a base station device 1420. The base station device 1420 and each antenna 1410 can be connected to each other via RF cables.

[0470] Antenna 1410 includes a plurality of antenna elements, such as a plurality of antenna arrays for massive MIMO. Antenna 1410 can be arranged, for example, as an antenna array matrix and is used for the base station device 1420 to transmit and receive wireless signals. For example, the plurality of antennas 1410 can be compatible with multiple frequency bands used by gNB 1400.

[0471] The base station device 1420 includes a controller 1421, a memory 1422, a network interface 1423, and a wireless communication interface 1425.

[0472] The controller 1421 can be, for example, a CPU or a DSP, and operates various functions of the higher layers of the base station device 1420. For example, the controller 1421 generates data packets based on the data in the signals processed by the wireless communication interface 1425 and transmits the generated packets via the network interface 1423. The controller 1421 can bundle data from multiple baseband processors to generate a bundled packet and transmit the generated bundled packet. The controller 1421 can have a logical function to execute controls such as radio resource control, radio bearer control, mobility management, admission control, and scheduling. This control can be executed in combination with nearby gNBs or core network nodes. The memory 1422 includes a RAM and a ROM, and stores programs executed by the controller 1421 and various types of control data (such as a terminal list, transmission power data, and scheduling data).

[0473] The network interface 1423 is a communication interface for connecting the base station device 1420 to the core network 1424 (e.g., a 5G core network). The controller 1421 can communicate with core network nodes or other gNBs via the network interface 1423. In this case, gNB 1400 can be connected to core network nodes or other gNBs through logical interfaces such as the NG interface and the Xn interface. The network interface 1423 can also be a wired communication interface or a wireless communication interface for a wireless backhaul line. If the network interface 1423 is a wireless communication interface, it can use a higher frequency band for wireless communication compared to the frequency band used by the wireless communication interface 1425.

[0474] The wireless communication interface 1425 supports any cellular communication scheme (such as 5G NR) and provides a wireless connection to a terminal in the cell of the gNB 1400 via the antenna 1410. The wireless communication interface 1425 generally may include, for example, a baseband (BB) processor 1426 and an RF circuit 1427. The BB processor 1426 may perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and perform various types of signal processing of each layer (such as the physical layer, MAC layer, RLC layer, PDCP layer, SDAP layer). Instead of the controller 1421, the BB processor 1426 may have a part or all of the above logical functions. The BB processor 1426 may be a memory storing a communication control program, or a module including a processor configured to execute the program and related circuits. The update program may change the functions of the BB processor 1426. The module may be a card or blade inserted into a slot of the base station device 1420. Alternatively, the module may also be a chip mounted on the card or blade. At the same time, the RF circuit 1427 may include, for example, mixers, filters, and amplifiers, and transmit and receive wireless signals via the antenna 1410. Although Figure 30 an example of one RF circuit 1427 connected to one antenna 1410 is shown, the present disclosure is not limited to this illustration, but one RF circuit 1427 may be connected to multiple antennas 1410 simultaneously.

[0475] As Figure 30 shown, the wireless communication interface 1425 may include multiple BB processors 1426. For example, multiple BB processors 1426 may be compatible with multiple frequency bands used by the gNB 1400. As Figure 30 shown, the wireless communication interface 1425 may include multiple RF circuits 1427. For example, multiple RF circuits 1427 may be compatible with multiple antenna elements. Although Figure 30 an example in which the wireless communication interface 1425 includes multiple BB processors 1426 and multiple RF circuits 1427 is shown, the wireless communication interface 1425 may also include a single BB processor 1426 or a single RF circuit 1427.

[0476] Combined with Figures 26A - 26C one or more units (such as the prediction unit 2611, prediction unit 2612, first unit 2621, second unit 2622, third unit 2623, first unit 2631, second unit 2632, third unit 2633, fourth unit 2634, etc.) included in the processing circuits 2613, 2624, or 2635 described Figure 30 may be implemented in the controller 1421 of the gNB 1400 shown in Figure 3A The operation 3101 in Figure 3B The operation 3201 inFigure 3C Operation 3301 in Figure 3D operations 3401 and 3402 in Figure 6 operations 601 - 603 in Figure 8 operations 801 - 804 in Figures 11 - 13 , Figures 16 - 18 , and Figures 20 - 21 at least a part of the operations performed by the BS in Figure 30 can be implemented by the controller 1421 and / or the BB processor 1426 of the gNB 1400 shown in

[0477] For example, the gNB1400 includes a part (e.g., the BB processor 1426) or the whole of the radio communication interface 1425, and / or includes a module of the controller 1421, and one or more components can be implemented in the module. In this case, the module can store a program for allowing the processor to act as one or more components (in other words, a program for allowing the processor to execute the operations of one or more components), and can execute the program. As another example, a program for allowing the processor to act as one or more components can be installed in the gNB1400, and the radio communication interface 1425 (e.g., the BB processor 1426) and / or the controller 1421 can execute the program. As described above, as a device including one or more components, the gNB1400, the base station device 1420 or the module can be provided, and a program for allowing the processor to act as one or more components can be provided. In addition, a readable medium in which the program is recorded can be provided.

[0478] Second application example of the base station

[0479] Figure 29 is a block diagram showing a second example of the schematic configuration of a base station to which the technology of the present disclosure can be applied. In Figure 29 , the base station is shown as the gNB 1530. The gNB 1530 includes a plurality of antennas 1540, a base station device 1550, and an RRH 1560. The RRH1560 and each antenna 1540 can be connected to each other via an RF cable. The base station device 1550 and the RRH 1560 can be connected to each other via a high-speed line such as an optical fiber cable.

[0480] The antenna 1540 includes a plurality of antenna elements, such as a plurality of antenna arrays for massive MIMO. The antenna 1540 can be arranged, for example, in an antenna array matrix, and is used for the base station device 1550 to transmit and receive radio signals. For example, the plurality of antennas 1540 can be compatible with a plurality of frequency bands used by the gNB 1530.

[0481] The base station device 1550 includes a controller 1551, a memory 1552, a network interface 1553, a wireless communication interface 1555, and a connection interface 1557. The controller 1551, the memory 1552, and the network interface 1553 are the same as the controller 1421, the memory 1422, and the network interface 1423 described with reference to Figure 29 The same.

[0482] The wireless communication interface 1555 supports any cellular communication scheme (such as 5G NR) and provides wireless communication to terminals located in the sector corresponding to the RRH 1560 via the RRH 1560 and the antenna 1540. The wireless communication interface 1555 may generally include, for example, a BB processor 1556. Except that the BB processor 1556 is connected to the RF circuit 1564 of the RRH 1560 via the connection interface 1557, the BB processor 1556 is the same as the BB processor 1426 described with reference to Figure 14 The same. As Figure 29 As shown, the wireless communication interface 1555 may include a plurality of BB processors 1556. For example, the plurality of BB processors 1556 may be compatible with multiple frequency bands used by the gNB 1530. Although Figure 29 An example in which the wireless communication interface 1555 includes a plurality of BB processors 1556 is shown, but the wireless communication interface 1555 may also include a single BB processor 1556.

[0483] The connection interface 1557 is an interface for connecting the base station device 1550 (wireless communication interface 1555) to the RRH 1560. The connection interface 1557 may also be a communication module for communication in the above-mentioned high-speed line for connecting the base station device 1550 (wireless communication interface 1555) to the RRH 1560.

[0484] The RRH 1560 includes a connection interface 1561 and a wireless communication interface 1563.

[0485] The connection interface 1561 is an interface for connecting the RRH 1560 (wireless communication interface 1563) to the base station device 1550. The connection interface 1561 may also be a communication module for communication in the above-mentioned high-speed line.

[0486] The wireless communication interface 1563 transmits and receives wireless signals via the antenna 1540. The wireless communication interface 1563 may generally include, for example, an RF circuit 1564. The RF circuit 1564 may include, for example, a mixer, a filter, and an amplifier, and transmits and receives wireless signals via the antenna 1540. Although Figure 29An example is shown where an RF circuit 1564 is connected to an antenna 1540. However, the present disclosure is not limited to this illustration, and an RF circuit 1564 can be connected to multiple antennas 1540 simultaneously.

[0487] As Figure 29 shown, the wireless communication interface 1563 can include multiple RF circuits 1564. For example, multiple RF circuits 1564 can support multiple antenna elements. Although Figure 29 an example is shown where the wireless communication interface 1563 includes multiple RF circuits 1564, the wireless communication interface 1563 can also include a single RF circuit 1564.

[0488] Combined Figures 26A - 26C with one or more components included in the processing circuits 2613, 2624, or 2635 described (such as the pre-judgment unit 2611, prediction unit 2612, first unit 2621, second unit 2622, third unit 2623, first unit 2631, second unit 2632, third unit 2633, fourth unit 2634, etc.), at least a part can be implemented in Figure 29 the controller 1552 in the gNB 1500 shown in Figure 3A Operation 3101 in Figure 3B Operation 3201 in Figure 3C Operation 3301 in Figure 3D Operations 3401 and 3402 in Figure 6 Operations 601 - 603 in Figure 8 Operations 801 - 804 in Figures 11 - 13 , Figures 16 - 18 , and Figures 20 - 21 at least a part of the operations performed by the BS in Figure 30 can be implemented by the controller 1552 and / or the BB processor 1556 of the gNB 1500 shown in

[0489] For example, the gNB 1500 includes a portion (e.g., BB processor 1526) or the entirety of the wireless communication interface 1525, and / or a module including the controller 1521, and one or more components may be implemented in the module. In this case, the module may store a program for allowing the processor to function as one or more components (in other words, a program for allowing the processor to perform the operation of one or more components), and the program may be executed. As another example, a program for allowing the processor to function as one or more components may be installed in the gNB 1500, and the wireless communication interface 1525 (e.g., BB processor 1526) and / or the controller 1521 may execute the program. As described above, as an apparatus including one or more components, the gNB 1500, the base station device 1520, or a module may be provided, and a program for allowing the processor to function as one or more components may be provided. In addition, a readable medium in which the program is recorded may be provided.

[0490] First application example of user equipment

[0491] Figure 30 16 is a block diagram showing an example of a schematic configuration of a smartphone 1600 to which the technology of the present disclosure can be applied.

[0492] The smart phone 1600 includes a processor 1601, a memory 1602, a storage device 1603, an external connection interface 1604, a camera 1606, a sensor 1607, a microphone 1608, an input device 1609, a display device 1610, a speaker 1611, a wireless communication interface 1612, one or more antenna switches 1615, one or more antennas 1616, a bus 1617, a battery 1618 and an auxiliary controller 1619.

[0493] The processor 1601 may be, for example, a CPU or a system on a chip (SoC), and controls the functions of the application layer and other layers of the smartphone 1600. The processor 1601 may include or function as any of the processing circuits 1001, 2001, 3001, 4001 described with reference to the drawings. The memory 1602 includes a RAM and a ROM, and stores data and a program executed by the processor 1601. The storage device 1603 may include a storage medium such as a semiconductor memory and a hard disk. The external connection interface 1604 is an interface for connecting an external device (such as a memory card and a universal serial bus (USB) device) to the smartphone 1600.

[0494] The camera 1606 includes an image sensor such as a charge coupled device (CCD) and a complementary metal oxide semiconductor (CMOS), and generates a captured image. The sensor 1607 may include a group of sensors such as a measurement sensor, a gyro sensor, a geomagnetic sensor, and an acceleration sensor. The microphone 1608 converts the sound input to the smart phone 1600 into an audio signal. The input device 1609 includes, for example, a touch sensor, a keypad, a keyboard, a button, or a switch configured to detect a touch on the screen of the display device 1610, and receives an operation or information input from a user. The display device 1610 includes a screen such as a liquid crystal display (LCD) and an organic light emitting diode (OLED) display, and displays an output image of the smart phone 1600. The speaker 1611 converts an audio signal output from the smart phone 1600 into sound.

[0495] The wireless communication interface 1612 supports any cellular communication scheme (such as 4G LTE or 5G NR, etc.) and performs wireless communication. The wireless communication interface 1612 may generally include, for example, a BB processor 1613 and an RF circuit 1614. The BB processor 1613 may perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and perform various types of signal processing for wireless communication. Meanwhile, the RF circuit 1614 may include, for example, a mixer, a filter, and an amplifier, and transmit and receive wireless signals via an antenna 1616. The wireless communication interface 1612 may be a chip module on which the BB processor 1613 and the RF circuit 1614 are integrated. Figure 30 As shown, the wireless communication interface 1612 may include multiple BB processors 1613 and multiple RF circuits 1614. Figure 30 An example is shown in which the wireless communication interface 1612 includes a plurality of BB processors 1613 and a plurality of RF circuits 1614 , but the wireless communication interface 1612 may also include a single BB processor 1613 or a single RF circuit 1614 .

[0496] In addition, in addition to the cellular communication scheme, the wireless communication interface 1612 can support other types of wireless communication schemes, such as short-range wireless communication schemes, near field communication schemes, and wireless local area network (LAN) schemes. In this case, the wireless communication interface 1612 can include a BB processor 1613 and an RF circuit 1614 for each wireless communication scheme.

[0497] Each of the antenna switches 1615 switches a connection destination of the antenna 1616 between a plurality of circuits (eg, circuits for different wireless communication schemes) included in the wireless communication interface 1612 .

[0498] Antenna 1616 includes multiple antenna elements, such as multiple antenna arrays for massive MIMO. Antenna 1616 can be arranged in an antenna array matrix, for example, and used for wireless communication interface 1612 to transmit and receive wireless signals. Smartphone 1600 may include one or more antenna panels (not shown).

[0499] In addition, the smartphone 1600 may include an antenna 1616 for each wireless communication scheme. In this case, the antenna switch 1615 may be omitted from the configuration of the smartphone 1600.

[0500] The bus 1617 connects the processor 1601, the memory 1602, the storage device 1603, the external connection interface 1604, the camera 1606, the sensor 1607, the microphone 1608, the input device 1609, the display device 1610, the speaker 1611, the wireless communication interface 1612, and the auxiliary controller 1619 to each other. Figure 30 The various blocks of the smartphone 1600 shown are supplied with power, the feed lines being partially shown as dashed lines in the figure. The auxiliary controller 1619 operates the minimum necessary functions of the smartphone 1600, for example in a sleep mode.

[0501] Combination Figures 27C - 27D At least a portion of one or more units (such as the prediction unit 2731, the sending unit 2732, the receiving unit 2733, the first unit 2741, the second unit 2742, the third unit 2743, etc.) included in the processing circuits 2734 and 2744 described above may be implemented in Figure 30 In the BB processor in the wireless communication interface 1612 in the smart phone 1600 shown in FIG. Figures 27C - 27D One or more units (eg, the prediction unit 2731, the sending unit 2732, the receiving unit 2733, the first unit 2741, the second unit 2742, the third unit 2743, etc.) included in the processing circuits 2734 and 2744 described above may be implemented in Figure 30 In the smart phone 1600 shown in FIG. 1 , in the processor 1601 or the auxiliary controller 1619 . Figure 3A Operation 3101, Figure 3B Operation 3201, Figure 3C Operation 3301, Figure 3D Operation 3401, Figures 11 - 13 , Figures 16 - 18 , Figures 20 - 21 and Figures 27A - 27B At least part of the operations performed by the UE in Figure 30 The processor 1601 and / or the auxiliary controller 1619 and / or the BB processor 1613 in the smart phone 1600 shown in FIG.

[0502] As an example, the smart phone 1600 includes a part (e.g., BB processor 1613) or the whole of the wireless communication interface 1612, and / or a module including the processor 1601 and / or the auxiliary controller 1619, and one or more components may be implemented in the module. In this case, the module may store a program that allows the processor to play the role of one or more components (in other words, a program for allowing the processor to perform the operation of one or more components), and the program may be executed. As another example, a program for allowing the processor to play the role of one or more components may be installed in the smart phone 1600, and the wireless communication interface 1612 (e.g., BB processor 1613), the processor 1601 and / or the auxiliary controller 1619 may execute the program. As described above, as a device including one or more components, a smart phone 1600 or a module may be provided, and a program for allowing the processor to play the role of one or more components may be provided. In addition, a readable medium in which the program is recorded may be provided.

[0503] Second application example of user equipment

[0504] Figure 31 17 is a block diagram showing an example of a schematic configuration of a car navigation device 1720 to which the technology of the present disclosure can be applied. The car navigation device 1720 includes a processor 1721, a memory 1722, a global positioning system (GPS) module 1724, a sensor 1725, a data interface 1726, a content player 1727, a storage medium interface 1728, an input device 1729, a display device 1730, a speaker 1731, a wireless communication interface 1733, one or more antenna switches 1736, one or more antennas 1737, and a battery 1738.

[0505] The processor 1721 may be, for example, a CPU or a SoC, and controls a navigation function and other functions of the car navigation device 1720. The memory 1722 includes a RAM and a ROM, and stores data and a program executed by the processor 1721.

[0506] The GPS module 1724 measures the position (such as latitude, longitude and altitude) of the car navigation device 1720 using GPS signals received from GPS satellites. The sensor 1725 may include a group of sensors such as a gyro sensor, a geomagnetic sensor and an air pressure sensor. The data interface 1726 is connected to, for example, the vehicle network 1741 via an unshown terminal and acquires data (such as vehicle speed data) generated by the vehicle.

[0507] The content player 1727 reproduces content stored in a storage medium such as a CD and a DVD, which is inserted into the storage medium interface 1728. The input device 1729 includes, for example, a touch sensor, a button, or a switch configured to detect a touch on the screen of the display device 1730, and receives an operation or information input from a user. The display device 1730 includes a screen such as an LCD or an OLED display, and displays an image of a navigation function or reproduced content. The speaker 1731 outputs a sound of a navigation function or reproduced content.

[0508] The wireless communication interface 1733 supports any cellular communication scheme (such as 4G LTE or 5G NR) and performs wireless communication. The wireless communication interface 1733 may generally include, for example, a BB processor 1734 and an RF circuit 1735. The BB processor 1734 may perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and perform various types of signal processing for wireless communication. Meanwhile, the RF circuit 1735 may include, for example, a mixer, a filter, and an amplifier, and transmit and receive wireless signals via an antenna 1737. The wireless communication interface 1733 may also be a chip module on which the BB processor 1734 and the RF circuit 1735 are integrated. Figure 31 As shown, the wireless communication interface 1733 may include multiple BB processors 1734 and multiple RF circuits 1735. Figure 31 An example is shown in which the wireless communication interface 1733 includes a plurality of BB processors 1734 and a plurality of RF circuits 1735 , but the wireless communication interface 1733 may also include a single BB processor 1734 or a single RF circuit 1735 .

[0509] In addition, in addition to the cellular communication scheme, the wireless communication interface 1733 can support other types of wireless communication schemes, such as short-range wireless communication schemes, near field communication schemes, and wireless LAN schemes. In this case, for each wireless communication scheme, the wireless communication interface 1733 can include a BB processor 1734 and an RF circuit 1735.

[0510] Each of the antenna switches 1736 switches a connection destination of the antenna 1737 between a plurality of circuits included in the wireless communication interface 1733 , such as circuits for different wireless communication schemes.

[0511] The antenna 1737 includes multiple antenna elements, such as multiple antenna arrays for massive MIMO. The antenna 1737 can be arranged in an antenna array matrix, for example, and is used for the wireless communication interface 1733 to transmit and receive wireless signals.

[0512] In addition, the car navigation device 1720 may include an antenna 1737 for each wireless communication scheme. In this case, the antenna switch 1736 may be omitted from the configuration of the car navigation device 1720.

[0513] Battery 1738 is fed via a feed line to Figure 31 The respective blocks of the illustrated car navigation device 1720 are supplied with electric power, and feed lines are partially illustrated as dotted lines in the figure. The battery 1738 accumulates electric power supplied from the vehicle.

[0514] Combination Figures 27C - 27D At least a portion of one or more units (such as the prediction unit 2731, the sending unit 2732, the receiving unit 2733, the first unit 2741, the second unit 2742, the third unit 2743, etc.) included in the processing circuits 2734 and 2744 described above may be implemented in Figure 31 In the BB processor 1734 in the wireless communication interface 1733 in the car navigation device 1720 shown in FIG. Alternatively, in combination Figures 27C - 27D One or more units (eg, the prediction unit 2731, the sending unit 2732, the receiving unit 2733, the first unit 2741, the second unit 2742, the third unit 2743, etc.) included in the processing circuits 2734 and 2744 described above may be implemented in Figure 31 In the processor 1721 in the car navigation device 1720 shown in FIG. Figure 3A Operation 3101, Figure 3B Operation 3201, Figure 3C Operation 3301, Figure 3D Operation 3401, Figures 11 - 13 , Figures 16 - 18 , Figures 20 - 21 and Figures 27A - 27B At least part of the operations performed by the UE in Figure 31 The BB processor 1734 and / or the processor 1721 in the car navigation device 1720 shown in FIG.

[0515] As an example, the car navigation device 1720 includes a part (e.g., BB processor 1734) or the whole of the wireless communication interface 1733, and / or a module including a processor 1721, and one or more components may be implemented in the module. In this case, the module may store a program (in other words, a program for allowing the processor to perform the operation of one or more components) that allows the processing to function as one or more components, and the program may be executed. As another example, a program for allowing the processor to function as one or more components may be installed in the car navigation device 1720, and the wireless communication interface 1733 (e.g., BB processor 1734) and / or the processor 1721 may execute the program. As described above, as a device including one or more components, the car navigation device 1720 or a module may be provided, and a program for allowing the processor to function as one or more components may be provided. In addition, a readable medium in which the program is recorded may be provided.

[0516] The technology of the present disclosure may also be implemented as an in-vehicle system (or vehicle) 1740 including a car navigation device 1720, an in-vehicle network 1741, and one or more blocks in a vehicle module 1742. The vehicle module 1742 generates vehicle data (such as vehicle speed, engine speed, and fault information), and outputs the generated data to the in-vehicle network 1741.

[0517] The exemplary embodiments of the present disclosure are described above with reference to the accompanying drawings, but the present disclosure is certainly not limited to the above examples. Those skilled in the art may obtain various changes and modifications within the scope of the appended claims, and it should be understood that these changes and modifications will naturally fall within the technical scope of the present disclosure.

[0518] For example, a plurality of functions included in one unit in the above embodiments may be implemented by separate devices. Alternatively, a plurality of functions implemented by a plurality of units in the above embodiments may be implemented by separate devices, respectively. In addition, one of the above functions may be implemented by a plurality of units. Needless to say, such a configuration is included in the technical scope of the present disclosure.

[0519] In this specification, the steps described in the flowchart include not only the processing performed in time series in the order described, but also the processing performed in parallel or individually rather than necessarily in time series. In addition, even in the steps processed in time series, it goes without saying that the order can be appropriately changed.

[0520] Although the present disclosure and its advantages have been described in detail, it should be understood that various changes, substitutions and transformations can be made without departing from the spirit and scope of the present disclosure as defined by the appended claims. Moreover, the terms "including", "comprising" or any other variants of the embodiments of the present disclosure are intended to cover non-exclusive inclusions, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the statement "including one..." do not exclude the presence of other identical elements in the process, method, article or equipment including the elements.

Claims

1. A method for wireless communication, comprising: Using an artificial intelligence (AI) model to predict that a radio link failure (RLF) or handover will occur, wherein the prediction is at least based on radio link condition information related to a user equipment (UE), and the radio link condition information at least includes information reflecting the condition of the radio link between the UE and a serving base station (SBS).

2. The method according to claim 1, further comprising: In response to the prediction, using the AI model to predict a target base station and a beam of the target base station for serving the user equipment for configuring a conditional handover, wherein the prediction is at least based on the radio link condition information related to the UE.

3. The method according to claim 2, wherein the output of the AI model includes a target base station-beam set, and the target base station-beam set includes one or more target base station-beam pairings, and each pairing indicates a predicted target base station and a beam of the target base station for serving the user equipment.

4. The method according to claim 1, wherein, The radio link condition information further includes information reflecting the condition of the radio link between the UE and one or more neighboring base stations.

5. The method according to claim 3, wherein, The radio link condition information includes at least one of the following: Beam measurement results; Channel state information (CSI) measurement results; Mobility measurement results; Real-time geographical environment and radio environment information including at least one of a high-precision map and an electromagnetic map; The location and movement characteristics of the UE; or Data of sensors including at least one of an accelerometer and a barometer of the UE.

6. The method according to claim 5, wherein, The AI model includes one or more AI sub-models, and the input of each AI sub-model in the AI model at least includes radio link condition information related to the UE.

7. The method according to claim 6, wherein, The AI model includes a first AI sub-model deployed at a serving base station, and the method further comprises: Based at least on radio link condition information related to the UE, the serving base station uses the first AI sub-model to predict the occurrence probability of RLF or handover and the target base station-beam set; The serving base station predicts that RLF or handover will occur based on a comparison of the occurrence probability of RLF or handover with a predetermined threshold; and In response to the prediction, the serving base station uses the target base station-beam set for configuring a conditional handover.

8. The method according to claim 6, wherein The AI model includes a second AI sub-model and a third AI sub-model deployed at a serving base station, and the method further comprises: Based on radio link condition information related to the UE, the serving base station uses the second AI sub-model to predict the occurrence probability of RLF or handover; The serving base station predicts that RLF or handover will occur based on a comparison of the occurrence probability of RLF or handover with a predetermined threshold; and In response to the prediction, based on the radio link condition information related to the UE and supplementary radio link condition information related to the UE, the serving base station uses the third AI sub-model to predict the target base station-beam set for configuring a conditional handover.

9. The method according to claim 7 or 8, wherein The AI model further includes a fourth AI sub-model deployed at the UE, and the method further comprises: Based at least on radio link condition information related to the UE, the UE uses the fourth AI sub-model to predict the occurrence probability of RLF or handover; The UE sends the predicted occurrence probability of RLF or handover to the serving base station; and wherein the occurrence probability of RLF or handover of the UE is included in the radio link condition information related to the UE.

10. The method according to claim 6, wherein the AI model includes a fifth AI sub-model deployed at the UE and a sixth AI sub-model deployed at the serving base station, and the method further includes: The UE predicts the occurrence probability of RLF or handover by using the fifth AI sub-model at least based on the radio link condition information related to the UE; The UE sends the predicted occurrence probability of RLF or handover to the serving base station; The serving base station pre-judges that RLF or handover will occur based on the comparison between the predicted occurrence probability of RLF or handover received from the UE and a predetermined threshold; and In response to the pre-judgment, the serving base station predicts a target base station-beam set for configuring conditional handover by using the sixth AI sub-model at least based on the radio link condition information related to the UE.

Images