WO2025168141A1 - Electronic device, method, and computer program product - Google Patents

Abstract

The present disclosure relates to an electronic device, a method, and a computer program product. The provided electronic device comprises: a processor; and a memory comprising computer program codes, wherein when being executed by the processor, the computer program codes cause the electronic device to perform operations, the operations including: determining that there is self-blocking caused by the posture of a user operating a user equipment (UE) to signal reception of the UE; by means of an artificial intelligence (AI) model, predicting, from received signal power of the UE, beam self-blocking information associated with a beam set of the UE; and on the basis of the beam self-blocking information, switching a beam used by the UE.

Description

Electronic device, method and computer program product

CROSS-REFERENCE TO RELATED APPLICATIONS This disclosure claims priority to Chinese invention patent application 202410176243.4, filed on February 8, 2024, entitled “ELECTRONIC DEVICE, METHOD AND COMPUTER PROGRAM PRODUCT,” the disclosure of which is hereby incorporated by reference in its entirety.

Technical Field

The present disclosure relates generally to the field of wireless communications, and more particularly to electronic devices, methods, and computer program products for beam prediction based on artificial intelligence (AI) models.

Background Art

In current wireless communication systems, massive antenna technologies such as multiple-input multiple-output (MIMO) are widely used. In these systems, both base stations and terminal devices have multiple antennas, and beamforming can be used to form spatial beams with narrow directivity to provide strong power coverage in a specific direction, thereby counteracting the large path loss in high-frequency channels. A set of beams with different transmission directions is used to achieve cell coverage. In order to improve the reception quality of beam signals, base stations and terminal devices need to select beams that match the direction of the wireless channel as much as possible. Traditionally, base stations and terminal devices can select and manage beams through beam training.

With the development of AI technology, AI models such as neural networks are being applied to beam management to achieve better performance or lower overhead. AI-based beam prediction is a method that uses AI to predict wireless signal beams. These methods typically employ machine learning techniques such as deep learning. By learning and training large amounts of historical data, they automatically extract features and patterns from the data and build a prediction model. This method has the advantage of automatically adapting to varying environments and channel conditions without manual parameter adjustment and exhibiting good generalization capabilities. Automatically learning and optimizing the beam prediction model improves prediction accuracy and stability.

The industry is still exploring the development and application of AI models for beam prediction. Therefore, there is a need to improve the beam prediction methods based on AI models to enhance their applicability and performance.

Summary of the Invention

The present disclosure provides multiple aspects. By applying one or more aspects of the present disclosure, the above needs can be met.

A brief overview of the present disclosure is provided below 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 key or important parts of the present disclosure, nor is it intended to limit the scope of the present disclosure. Its purpose is simply to present certain concepts of the present disclosure in a simplified form as a prelude to the more detailed description that will be given later.

According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory comprising computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations, the operations comprising: determining that there is self-occlusion of signal reception of a user equipment (UE) caused by a posture of a user operating the UE; predicting beam self-occlusion information associated with a beam set of the UE from the received signal power of the UE through an artificial intelligence (AI) model; and switching the beam used by the UE based on the beam self-occlusion information.

According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory comprising computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations, the operations comprising: receiving a self-occlusion status report from a user equipment (UE), the self-occlusion status report indicating an impact of self-occlusion caused by a user's posture of operating the UE on a base station's transmit beam, and based on beam self-occlusion information predicted by the UE using an artificial intelligence (AI) model; and determining, based on the self-occlusion status report, a plurality of transmit beams for use in beam training between the base station and the UE.

According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory comprising computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations, the operations comprising: preparing a training data set comprising input data and output data, wherein the input data comprises the received signal power of a user equipment (UE) associated with a plurality of postures of a user operating the UE, and the output data comprises beam self-occlusion information of a beam set of the UE associated with the plurality of postures; and training an artificial intelligence (AI) model on the training set to determine parameters of the AI model.

According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations, the operations comprising: receiving information from a base station about an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction; activating the AI model according to the activation period; and performing beam prediction using the activated AI model based on measurements of a beam management reference signal sent by the base station according to the activation period.

According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory, comprising computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations, the operations comprising: determining an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction; sending the determined activation period to the UE; and sending a beam management reference signal according to the activation period for the AI model to perform beam prediction.

According to another aspect of the present disclosure, a method is provided, including operations performed by any of the above electronic devices.

According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing executable instructions is provided. When the executable instructions are executed, the operations performed by any electronic device as described above are implemented.

BRIEF DESCRIPTION OF THE DRAWINGS

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

FIG1 schematically illustrates a beam training process in a wireless communication system;

Figures 2 and 3 schematically illustrate situations where wireless signals sent by a base station are subject to external shielding and self-shielding;

FIG4 illustrates a process of beam management according to the first embodiment of the present disclosure;

5A and 5B show the RSRP changes caused by external shading and self-shading, respectively;

FIG6 schematically illustrates auxiliary information used to determine self-occlusion;

FIG7 shows a schematic diagram of an AI model according to the first embodiment;

FIG8 shows a flow chart for switching beams according to the first embodiment;

FIG9 shows a training method for an AI model according to the first embodiment;

FIG10 shows four exemplary gestures and their blocking effects on the beams of various antenna panels;

FIG11 shows the wireless channel environment in the simulation;

12A-12C show simulation results;

FIG13 shows a block diagram of an electronic device according to the first embodiment;

Figure 14 shows a comparison between traditional beam management and various beam prediction use cases;

15 is a flowchart showing a model activation process according to the second embodiment;

16 and 17 show block diagrams of electronic devices according to a second embodiment;

FIG18 shows an example block diagram of a computer that may be implemented as a user device or a control device according to the present disclosure;

FIG19 illustrates a first example of a schematic configuration of a base station according to the present disclosure;

FIG20 illustrates a second example of a schematic configuration of a base station according to the present disclosure;

FIG21 illustrates a schematic configuration example of a smartphone according to the present disclosure;

FIG22 illustrates a schematic configuration example of a car navigation device according to the present disclosure.

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

Various exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. The following description of the exemplary embodiments is merely illustrative and is not intended to limit the present disclosure and its applications. For the sake of clarity and conciseness, not all features of the embodiments are described in this specification. However, it should be noted that when implementing the embodiments of the present disclosure, many implementation-specific settings can be made according to specific needs, such as to comply with those restrictions related to equipment and services, and these restrictions may vary depending on the implementation.

In addition, it should be noted that in order to avoid obscuring the present disclosure due to unnecessary details, some drawings only show processing steps and/or equipment structures that are closely related to at least the technical content of the present disclosure, while in other drawings, in order to facilitate a better understanding of the present disclosure, existing processing steps and/or equipment structures are additionally shown.

For the purpose of convenience of explanation, one or more aspects of the present disclosure may be described below in the context of 5G New Radio (NR). However, it should be noted that this is not a limitation on the scope of application of the present disclosure, and one or more aspects of the present disclosure may also be applied to already commonly used wireless communication systems such as 4G LTE/LTE-A, or various wireless communication systems to be developed in the future. The architectures, entities, functions, processes, etc. mentioned in the following description are not limited to those in the NR communication system, but may be found in other communication standards.

【Overview】

In wireless communication systems such as 4G LTE or 5G NR, base stations and terminal devices (also referred to as "user equipment", hereinafter referred to as "UE") can apply technologies such as Massive MIMO. In order to support the application of MIMO technology, both base stations and UEs have many antennas, such as dozens, hundreds or even thousands of antennas. The antennas are arranged into one or more antenna arrays in a specific form. An antenna array can be composed of antenna elements in a whole row, a whole column, multiple rows, or multiple columns, thereby forming an independently configurable transceiver unit (TXRU). By configuring the amplitude parameters and/or phase parameters of the antenna elements that make up the TXRU, the antenna pattern of the TXRU is adjusted, and the electromagnetic wave radiation emitted by all antenna elements in the antenna array forms a narrower beam pointing to a specific spatial direction, that is, beamforming is achieved.

It should be noted that the term "base station" used in this disclosure is an example of a control device on the network side and has the full breadth of its usual meaning. For example, in addition to the gNB and ng-eNB specified in the 5G communication standard, depending on the scenario in which the technical solution of this disclosure is applied, the "base station" may also be, for example, an eNB in an LTE communication system, a transmit receive point (TRP), a remote radio head (RRH), a wireless access point (AP), a drone control tower, or a communication device that performs similar functions. The following sections will describe in detail the application examples of base stations.

In this disclosure, the term "UE" has its full, common meaning and encompasses various terminal devices or in-vehicle equipment that communicate with a base station. For example, a UE can be a terminal device or component thereof, such as a mobile phone, laptop, tablet, in-vehicle communication device, or drone. The following sections describe detailed application examples of UEs.

The following briefly describes the process by which a base station or UE uses an antenna array for data transmission. First, the baseband signal representing the user data stream is mapped to m radio frequency links (m ≥ 1) through digital precoding. Each radio frequency link up-converts the baseband signal to obtain a radio frequency signal and transmits the radio frequency signal to the corresponding antenna array. A set of analog beamforming parameters is applied to the antenna elements in the antenna array according to the transmission direction. The analog beamforming parameters may, for example, include phase setting parameters and/or amplitude setting parameters for the antennas in the antenna array. Based on the corresponding analog beamforming parameters, the electromagnetic radiation emitted by all antennas in the antenna array forms a desired beam in space (hereinafter also referred to as the "transmit beam"). Reception using the antenna array is an inverse process: analog beamforming parameters associated with a specific direction are applied to the antennas in the antenna array so that the antenna array optimally receives the beam signal in that direction (hereinafter also referred to as the "receive beam"), and user data is recovered through demodulation and decoding. The base station or UE may pre-store a beamforming codebook containing beamforming parameters for generating a limited number of beams.

In order to improve transmission performance, the base station and UE need to select a transmitting beam or receiving beam from their available beams that matches the channel direction as much as possible. That is, at the transmitting end, the transmitting beam is aligned with the channel departure angle, and at the receiving end, the receiving beam is aligned with the channel arrival angle.

Traditionally, base stations and UEs can select beams through beam training. Beam training generally includes processes such as beam measurement, beam reporting, and beam indication. The beam training process in a wireless communication system is briefly described with reference to Figure 1. As shown in the figure, the base station 1000 can use n _{t_DL} (n _{t_DL} ≥ 1) downlink transmission beams with different directions, and the UE 1002 can use n _{r_DL} (n _{r_DL} ≥ 1) downlink reception beams with different directions. Similarly, in the uplink direction, the base station 1000 and the UE 1004 can also use several reception beams and transmission beams with different directions (not shown), respectively. It should be understood that the number and coverage of the beams shown in Figure 1 are merely exemplary.

The base station 1000 and the UE 1002 traverse all transmit beam-receive beam combinations by scanning beams. Taking downlink beam scanning as an example, first, the base station 1000 sends different downlink reference signals, such as non-zero power CSI-RS (NZP-CSI-RS) resources or SSB resources, to the UE 1002 through its n _{t_DL} transmit beams according to the downlink scanning period. The UE 1002 receives each transmit beam through its n _{r_DL} receive beams and measures the beam signal. For example, the UE 1002 can measure the downlink reference signal carried in each transmit beam, that is, obtain a total of n _{t_DL} ×n _{r_DL} measurement results. For example, the UE 1002 can measure the reference signal received power (L1-RSRP), reference signal received quality (L1-RSRQ), signal to interference plus noise ratio (L1-SINR), etc. of the physical layer (L1).

UE 1002 then reports the beam measurement results to base station 1000. To reduce the amount of reported data, UE 1004 can be configured to report only the measurement results of some transmit beams (e.g., only Nr < n _{t_DL} , where Nr is pre-configured by base station 1000) and the identification information of the associated reference signals. Based on the reported beam measurement results, base station 1000 can select the best transmit beam from the transmit beams reported by UE 1002 for downlink transmission with UE 1002. In addition, base station 1000 indicates the reference signal corresponding to the best transmit beam to UE 1002, so that UE 1002 can determine the best receive beam corresponding to the reference signal during the beam scanning process. This achieves alignment of the transmit beam and the receive beam.

To reduce overhead, a two-stage beam training approach can be considered: first, a wide beam search is performed, followed by a narrow beam search within the coverage area of the selected wide beam. Context-based beam search has also been proposed. However, traditional beam selection methods essentially search through all possible beam pairs, which is both expensive and time-consuming.

With the development of AI technology, AI models are being used in beamforming management to achieve better performance due to their powerful feature extraction capabilities. Many studies have explored using neural networks to learn beam information from observation data to estimate the optimal beam. The signals received by all available beams in the beamforming codebook are directly used as input to the neural network. Results show that compared with traditional beamforming training, the advantages of AI models for beamforming are reflected in total training time slots and spectral efficiency. By leveraging the feature extraction capabilities of deep learning to assist beam management, it is possible to reduce measurement overhead and latency without relying on prior assumptions about the channel model.

However, the implementation and standardization of AI-based beam prediction solutions is still ongoing, and many issues remain to be addressed. The present disclosure provides several improvements. Exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[First embodiment]

Despite bringing significant performance gains in beam management, existing AI-assisted beam prediction methods often only consider the free propagation of beams in wireless space under ideal conditions, and do not consider the impact of user behavior on the beams.

The first embodiment of the present disclosure will discuss the problem of self-blocking (Self Blockage) of the beam signal by the user himself. As used in the present disclosure, "self-blocking" refers to the blocking of the UE's beam reception caused by the posture of the user when operating the UE. Typically, the user can hold the UE with various gestures. If his/her hand happens to cover the antenna panel, the direct path, reflection path, scattering path, etc. of the beam signal may all be blocked, resulting in severe attenuation of the received signal power, as shown in the lower part of Figure 2. It should be noted that the following may mainly use the user's hand or gesture as an example to discuss self-blocking, but the present disclosure is not limited to this. The posture of the user operating the UE may not be limited to gestures, and the part causing self-blocking may be any other body part.

In contrast, "external obstruction" in this disclosure refers to obstruction of beam signals by objects other than the user. For example, buildings, trees, or mountains may block the direct path between the base station and the UE, causing signal loss, as shown in the upper part of Figure 2. When external obstruction occurs, although the direct path is blocked, other reception paths may still exist, such as reflection paths, scattered paths, and diffraction paths. Figure 3 schematically illustrates the situation where the wireless signal transmitted by the base station is subject to external obstruction and self-obstruction.

For carrier frequencies below Sub-6GHz, the impact of human body occlusion is not yet obvious. However, higher frequencies such as millimeter waves are more sensitive to human body occlusion, so it becomes important to consider this self-occlusion factor and understand its impact on system performance. Taking into account that different users have different self-occlusion postures, according to the first embodiment of the present disclosure, a neural network is deployed on the UE side to learn the intrinsic correlation between the user's behavioral habits and the occlusion of the beam, so as to recommend the best beam under user self-occlusion.

Figure 4 illustrates the process of beam management according to the first embodiment of the present disclosure. As shown in Figure 4, the process may start by determining whether there is self-occlusion on the UE side (step S11). Both external occlusion and self-occlusion may result in a reduction in the received signal power. It is necessary to distinguish whether the current occlusion is external occlusion or self-occlusion, because for external occlusion, due to its lack of regularity, the beam used is generally switched through the traditional beam failure recovery (BFR) process. Self-occlusion, as a special occlusion situation, is usually related to the user's behavioral habits. Therefore, the beam failure caused by user self-occlusion can be quickly recovered by utilizing the user's behavioral patterns, for example, by using the UE-side beam switching process to be described later. When beam failure is caused by external occlusion and a beam recovery method for self-occlusion is used, the effect may be unsatisfactory and may even cause a greater recovery delay.

According to the first embodiment, the determination of whether self-obstruction exists can be based on the UE's received signal power, such as the reference signal received power (RSRP). This determination can be triggered by a decrease in the UE's received signal power. For example, when the UE's RSRP drops below a predetermined threshold, step S11 in Figure 4 can be triggered. In step S11, the presence of self-obstruction can be determined by extracting features from the UE's RSRP to see if a predetermined feature appears.

In one example, after the RSRP falls below a threshold, the change in RSRP over a period of time can be monitored to detect whether there is power jitter. When external shielding occurs, although the power of the received signal decreases due to the obstruction of the direct path, because there are other receiving paths, such as reflection paths, scattering paths, diffraction paths, etc., the received signal is the vector superposition of the signals on these different paths at the receiving end, which reflects the fluctuation phenomenon in the power of the received signal, that is, power jitter. Figure 5A shows a schematic diagram of the change in RSRP caused by external shielding. As for self-shielding, when the user blocks the antenna panel of the receiving end, since all signal paths are blocked, the received power drops sharply, and there is no power jitter at this time. Figure 5B shows a schematic diagram of the change in RSRP caused by self-shielding. Therefore, whether external shielding or self-shielding occurs can be distinguished by whether there is jitter in power.

In another example, in addition to received signal power, other information can also be used to assist in distinguishing external obstruction from self-obstruction. This auxiliary information includes information detected by various sensors on the UE, such as touchscreen information, gyroscope information, camera information, and infrared sensor information. For example, a touch sensor or infrared sensor equipped on the UE can detect contact between the user's hand and the UE, thereby helping to determine the user's gesture and whether the antenna panel is obstructed. For example, the UE's gyroscope can obtain real-time UE deflection angle information. If the UE's deflection angle and received signal power show a certain change pattern, the change pattern can be used to determine whether it is self-obstruction, as shown in the right part of Figure 6. For another example, the front camera can capture an image of a human face to determine whether the UE is in portrait or landscape orientation. As shown in the left part of Figure 6, some base stations may even be equipped with cameras. If the UE is detected in the image captured by the base station camera, it means that there is no obstruction in the direct path between the base station and the UE. If the UE can obtain this information from the base station and the UE's received signal power is low (for example, below a certain threshold), it can be determined that self-obstruction has occurred.

In addition, although step S11 is shown in Figure 4 as being separate from the beam prediction step S12 to be described later, as shown in the dotted box in the figure, these two steps can be implemented together. That is, the received signal power (optionally, the above-mentioned auxiliary information that can indicate the user's operation gesture) can be input into an AI model such as a neural network, allowing the AI model to learn the changing pattern of the input information and thus determine external occlusion or self-occlusion. In this example, a threshold value such as RSRP can be set. When the RSRP drops below the threshold, the AI model prediction is triggered, and information about whether there is self-occlusion can be reflected in the output of the AI model.

When it is determined in step S11 that self-occlusion exists, as shown in FIG4 , the trained AI model can be used to perform beam prediction (step S12). Specifically, the AI model according to this embodiment is deployed on the UE side to learn the impact of the user's behavioral habits on the beam, so as to recommend the optimal beam under self-occlusion by mining the inherent laws of user behavior. As mentioned above, both external occlusion and user self-occlusion will affect the beam, and the occurrence of external occlusion is random, sudden, and irregular. On the contrary, for user self-occlusion, there are stronger rules to follow. This is because people's behavioral habits themselves are regular, so the impact of user behavior on the beam also becomes regular. For example, how the user holds the UE, how the current gesture of holding the UE changes, and so on. Therefore, this makes it feasible to learn the impact of user behavior patterns on the beam.

AI models can be implemented as neural networks with various architectures, including but not limited to convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory networks (LSTMs). A neural network is a mathematical model that simulates biological nervous systems and consists of multiple interconnected neurons that process input data and generate output signals. Neural networks learn and recognize patterns by adjusting connection weights and transfer functions between neurons. The structure of a neural network can be divided into an input layer, a hidden layer, and an output layer. The hidden layer can have multiple layers and is a key component of the neural network. In addition, a neural network may also include one or more of a batch normalization layer, an activation function, a pooling layer, and a fully connected layer.

Figure 7 shows a schematic diagram of an AI model according to this embodiment. As shown in the figure, the AI model for beam prediction can accept the received signal power of the UE as input, such as RSRP measurement results. Optionally, if the AI model also takes into account the function of determining whether there is self-occlusion (i.e., executing step S11 in Figure 4), it can also accept auxiliary information such as gyroscope information, touch screen information, camera information, infrared sensor information, etc., which may include user operation posture information. When the input is only RSRP, the AI model is a single-input network; when the input also contains auxiliary information, the AI model is a multi-input network. The multi-input network is a parallel connection of a single-input network structure, that is, each single-input network is used to extract the features of each type of information separately, and finally the information is fused in a cascade form. The input information can be extracted or spliced into a feature vector of a specific dimension suitable for the AI model.

As a result of the prediction, the AI model can output beam self-occlusion information, which is intended to describe the effect of self-occlusion on the UE's beam. The beam self-occlusion information may include state information indicating the self-occlusion status of all available beams of the UE, expressed as I∈{index, flag}, where index represents the beam index, and flag indicates whether there is self-occlusion for the beam, for example, ‘1’ indicates that there is self-occlusion, and ‘0’ indicates that there is no self-occlusion. The prediction of the state information I can be performed for all beams on the UE side, such as those included in the beamforming codebook. In addition, the beam index can be a beam number used internally by the UE, or it can be a reference signal identifier. The state information I can reflect whether there is self-occlusion and which beams are affected by self-occlusion.

Additionally or alternatively, the beam self-obstruction information may include priority information Ξ indicating a beam selection priority under self-obstruction. The priority information Ξ may give a priority ranking of a predetermined number of beams (e.g., 1, 2, 3, 4). The priority information Ξ may be output as a beam switching order under the current self-obstruction, such as Ξ = {index1, index2, index3, index4}, which represents four UE beams with descending priority.

Optionally, the beam self-occlusion information may also include an output self-occlusion duration T, which indicates how long the current self-occlusion state lasts. The self-occlusion duration T actually predicts the time the user maintains the current operating posture. During this period of time, the beam self-occlusion situation of the UE may not change much, so to a certain extent it can also be regarded as the validity period of the above-mentioned state information I or priority information Ξ. In addition, the duration T may also be related to the reactivation of the AI model, that is, the time of the next activation of the AI model can be determined based on the duration T to predict whether the self-occlusion state of the UE beam has changed. In one example, the duration T can take several predefined time values. In addition, an upper limit and a lower limit can be configured for the duration T, and may vary for different UEs.

Optionally, the beam self-occlusion information may also include future self-occlusion state information Γ∈{index, flag}. State information Γ is similar to state information I, but differs in that state information I predicts the self-occlusion state at the current time t (up to the subsequent time t+T), while state information Γ predicts the self-occlusion state after the future time t+T, reflecting changes in the self-occlusion state caused by changes in the user's operating posture.

It should be noted that although four possible beam self-occlusion information are illustrated above, the AI mode may also output other forms of beam self-occlusion information as needed.

Subsequently, as shown in Figure 4, the UE can switch the beam it uses based on the beam self-occlusion information output by the AI model (step S13). The switching is because the user's current behavior has affected the UE's signal reception performance, that is, the beam currently used by the UE is no longer the optimal reception beam.

In one example, the UE may switch beams based on priority information Ξ predicted by the AI model. Since the priority information Ξ recommends the priority of beam usage, the UE may directly switch the current receive beam to the beam with the highest priority according to the priority information Ξ. In another example, the UE may switch beams based on the state information I predicted by the AI model, for example, switching to a beam indicated in the state information I that has no self-occlusion. The dwell time of the switched receive beam may be the self-occlusion duration T predicted by the AI model, but may not be limited thereto. Thus, the UE can quickly switch to the receive beam recommended by the AI model without going through the traditional beam failure recovery process, which helps to reduce signaling overhead and latency. In addition, in scenarios where the uplink channel and the downlink channel are symmetrical, such as time division duplex (TDD), the UE may also switch the transmit beam used for uplink transmission to reduce or avoid the impact of self-occlusion on the transmit beam.

However, in some cases, simply switching the UE's beam may not be the optimal solution because, under the influence of self-obstruction, the base station's transmit beam may not be aligned with the UE's receive beam recommended by the AI model. Therefore, according to this embodiment, the UE can interact with the base station to determine the optimal beam pair under the current self-obstruction.

Figure 8 shows a flowchart for switching beams according to this embodiment. The process shown in the figure can occur after step S12 of Figure 4, or after the UE's received beam, which has been switched as described above, still does not meet the requirements. As shown in the figure, in step S21, the UE sends a self-occlusion status report to the base station. The self-occlusion status report can be generated based on the beam self-occlusion information predicted by the AI model and indicates the impact of user-induced self-occlusion on the base station's transmit beam.

Based on the beam self-occlusion status information or beam priority information predicted by the AI model, the UE can indicate the desired base station transmit beam in the self-occlusion status report. This can be achieved in combination with previous beam training results or beam prediction results. It is hoped that the beam signal sent by the base station can be received by one or more beams on the UE side that are not affected by or less affected by self-occlusion (for example, a beam indicated as not self-occluded in the status information, or a beam indicated as having the highest priority in the priority information). On the contrary, it is not desirable that the base station's transmit beam can only be received by the self-occluded UE beam.

In one example, the UE may suggest the desired base station transmit beam range through a self-occlusion status report. For example, based on the reference signal resource set configured by the base station for beam scanning, the available transmit beams of the base station may be divided into several parts, and the UE transmits information indicating one of the parts. For example, the base station scans sequentially according to the beam sequence {1, 2, ..., 8, 9}, then '01' represents the first part {1, 2, 3}, '10' represents the second part {4, 5, 6}, '11' represents the third part {7, 8, 9}, and when all receive beams are self-occluded, the information may take the value '00'. If historical beam training or beam prediction shows that the base station transmit beam that best matches the UE's highest priority receive beam is in the second part, the UE reports '10' in the self-occlusion status report. However, it should be noted that the number of bits or representation of the information indicating the beam range may not be limited to this.

In another example, the UE may indicate a specific base station transmit beam index in the self-obstruction status report. For example, the UE may determine which base station transmit beams correspond to the UE's receive beams that are not affected by or less affected by self-obstruction, and report the indexes of these transmit beams, such as reference signal identifiers, or transmission configuration indication (TCI) status that references the reference signal identifiers.

The self-obstruction status report is triggered, meaning it is reported only when the UE predicts self-obstruction. The self-obstruction status report is placed, for example, in uplink control information (UCI). The UE can request uplink PUSCH resources from the base station to transmit this information, and the base station can allocate resources through DCI signaling.

Subsequently, in step S22, the UE can determine one or more receive beams to be scanned in a later beam training. These receive beams are beams that are not self-occluded or have high priority as predicted by the AI model. In step S23, the base station can determine one or more transmit beams to be scanned. As described above, the base station receives a self-occlusion status report from the UE, which indicates the desired transmit beam range or index, from which the base station can determine the transmit beam to be scanned.

In step S24, the UE and base station may perform beam training. The specific process has been described with reference to Figure 1 and will not be repeated here. As a result of beam training, the optimal receive beam for the UE and the optimal transmit beam for the base station under self-obstruction conditions may be determined. In steps S25 and S26, the UE and base station may respectively switch their beams. Similarly, in TDD scenarios, for example, the UE may also switch the transmit beam used for uplink transmission.

Through the process shown in FIG8 , compared with traditional beam failure recovery, the beam search space can be effectively reduced and the efficiency of beam recovery can be improved.

Figure 9 shows a training method for an AI model according to this embodiment. As shown in the figure, the training method generally includes preparing a training data set (step S31) and training a model on the training data set (step S32), wherein the training data set includes input data and output data.

According to this embodiment, the model training phase includes a general training phase and a specific training phase. For the general training phase, the input data of the training data set includes RSRP (optionally, other auxiliary information), and the output data includes beam self-occlusion information, such as self-occlusion status information, priority information, or self-occlusion duration. Training data sets can be collected from multiple users for a variety of operating postures. Figure 10 shows four exemplary gestures and the occlusion effects on the beams of each antenna panel. In order to obtain a relatively pure data set of received power changes under gesture occlusion, it can be considered to be performed in a weak signal environment, such as in a microwave darkroom, to shield the influence of other signals. The test user can make various operating gestures, measure the received signal power under the gesture as input data, and collect beam self-occlusion information as output data. In addition, the training data set can also include input data and output data associated with gestures that do not cause self-occlusion.

Subsequently, in step S32, training is performed based on the cross entropy loss function until the AI model converges. In order to make the computational complexity of the operation feasible in practice, the training step is based on an iterative process, such as a stochastic gradient descent (SGD) algorithm. To this end, the weights of the neural network are initialized (e.g., randomly) at the beginning. The input data of the training data set is input into the neural network to obtain the corresponding output, such as the predicted beam self-occlusion information, and the value of the loss function is calculated based on the difference between the predicted result and the actual beam self-occlusion information. The weights and biases of the neural network are updated according to the gradient information of the loss function. By repeating the above steps until a preset number of iterations is reached or the loss function value is lower than a preset threshold.

The AI model can be trained and provided by the device vendor or mobile operator. The device vendor can pre-configure the trained model on the UE. Alternatively, the UE can download the model from the device vendor or mobile operator's server over the network.

The specific training phase is for different users. For users who use UE, a unique user data set is collected according to their behavioral habits, and the parameters of the AI model are fine-tuned based on this user data. The collection of unique user data sets involves user privacy. It is very important to collect data without leaking user privacy. The collection method is such as the mobile phone camera. Some of the postures of holding the phone can be observed through the front and rear cameras. In addition, it can be obtained through, for example, the temperature sensor on the UE; for example, the UE can specify the user's holding gestures, and the gestures at this time can be recorded to form a unique data set. The unique data set only helps to fine-tune the network during fine-tuning, so the amount of data required is much smaller than the general data set.

According to this embodiment, lifecycle management of the AI model can be achieved. The prediction accuracy of the AI model is monitored, and when the prediction accuracy is lower than a predetermined threshold, the model can be updated based on the collected training data as described above.

In one example, the update of the AI model may utilize the current state information I and the future state information Γ after a duration T predicted by the AI model. For example, the update process may include:

1) For time t, record the state information Γ _t at the time after the duration T (i.e., t+T) predicted by the AI model;

2) At time (t+T), using the AI model to predict state information I _t+T ;

3) Compare the state information Γ _t and the state information I _t+T .

Repeat the above steps multiple times and calculate the AI model's prediction accuracy. For example, if Γ _t = It _{t + T} occurs M times out of N, the prediction accuracy can be calculated as M/N. When the prediction accuracy falls below a certain threshold, retrain the AI model to update its parameters. During this update process, the AI model update trigger does not require external assistance and is determined solely by the model's adjacent prediction values.

The following describes the simulation of beam prediction according to this embodiment. Figure 11 shows the simulated wireless channel environment. The Saleh-Valenzuela channel model is used to measure the actual channel environment based on characteristics such as attenuation, delay, angle of arrival (AoA), and angle of departure (AoD) of each path. The number of available beams for each UE is four. Other simulation parameters are shown in Table 1.

Table 1 Channel simulation parameters

For the prediction model, a convolutional neural network is used here. Its specific structure and parameters are shown in Table 2, where _fi 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 edge padding size of the convolution layer, respectively, BatchNorm refers to batch normalization, AvgPooling refers to average pooling, ReLU refers to the ReLU activation function, and _Nout refers to the output size.

Table 2 Structure and parameters of the prediction model

In the training dataset, N _out =5, indicating the four gestures shown in FIG10 and a gesture without self-occlusion. Table 3 shows the gestures and the beam attenuation corresponding to the gesture occlusion.

Table 3 Attenuation corresponding to each gesture

The simulation results are shown in Figures 12A, 12B, and 12C. Figure 12A shows the evolution of the loss function during training, showing that the prediction network gradually converges with increasing iterations. Figure 12B shows the model's predicted normalized beam gain performance in the presence of user self-occlusion, calculated as the gain ratio between the predicted optimal beam and the actual optimal beam. The simulation graph shows that as the iteration number increases, the predicted optimal beam achieves an average normalized beam gain of approximately 97%, indicating nearly perfect beam alignment.

Figure 12C shows the model's prediction accuracy in the presence of user self-occlusion, calculated as the ratio of the predicted optimal beam to the actual optimal beam. As can be seen, the optimal beam prediction accuracy reaches approximately 97% after 300 training iterations, nearly finding the optimal beam. Therefore, the solution of this embodiment can recover the optimal beam even in the presence of self-occlusion.

13 is a block diagram showing an electronic device 100 according to this embodiment. The electronic device 100 may be implemented as a UE or a component thereof.

As shown in FIG13 , electronic device 100 includes processing circuitry 101. Processing circuitry 101 includes at least a determination unit 102, a prediction unit 103, and a switching unit 104. Processing circuitry 101 may be configured to perform the process shown in FIG4 . Processing circuitry 101 may refer to various implementations of digital circuitry, analog circuitry, or mixed-signal (a combination of analog and digital signals) circuitry that performs functions in a UE.

Determining unit 102 is configured to determine whether self-occlusion exists due to the user's gesture on the UE's signal reception, i.e., to execute step S11 in FIG. 4 . In one example, determining unit 102 may determine whether self-occlusion exists by detecting whether jitter occurs in the received signal power. In another example, determining unit 102 may determine whether self-occlusion exists using an AI model.

Prediction unit 103 is configured to predict beam self-occlusion information associated with the UE's beam set based on the UE's received signal power using an AI model, i.e., execute step S12 in FIG4 . The beam self-occlusion information may include self-occlusion status information, beam selection priority information, and optionally, the duration of the self-occlusion or future self-occlusion status information.

Switching unit 104 is configured to switch the beam used by the UE based on the beam self-blocking information, i.e., execute step S13 in Figure 4. In one example, the UE can switch the current beam to the beam with the highest priority based on the priority information predicted by the AI model. In another example, the UE can switch to the appropriate reception protection by performing beam training based on the beam self-blocking information with the base station.

The electronic device 100 may further include a communication unit 105. The communication unit 105 may be configured to communicate with a base station under the control of the processing circuit 101. In one example, the communication unit 105 may be implemented as a transceiver, including communication components such as an antenna array and/or a radio frequency link. The communication unit 105 is depicted with a dashed line because it may also be located outside the electronic device 100.

The electronic device 100 may further include a memory 106. The memory 106 may store various data and instructions, such as programs and data used for the operation of the electronic device 100, various data generated by the processing circuit 101, various control signals or service data sent or received by the communication unit 105, etc. The memory 106 is drawn with a dotted line because it may be located within the processing circuit 101 or outside the electronic device 100.

[Second embodiment]

The second embodiment of the present disclosure relates to activation management of an AI model for beam prediction. It should be understood that the AI models discussed in this embodiment include but are not limited to the AI model for beam prediction in the first embodiment above.

Currently, the industry is experimenting with deploying AI models on the UE or network side to predict downlink transmit beams. The motivation for using AI for beam management is to reduce the frequency of UE reference signal measurements and the latency overhead incurred by these measurements. However, the effectiveness of this AI-based beam management requires further evaluation. The following discussion will explore the use cases of beam prediction models.

Figure 14 shows a comparison between traditional beam management and various beam prediction use cases. As previously described with reference to Figure 1, in the traditional scheme, the base station and the UE can use two-stage beam scanning to search for the best beam pair. The base station can periodically or aperiodically send a reference signal for beam management (hereinafter referred to as a "beam management reference signal"), such as an SSB or CSI-RS, and after determining the approximate direction angle of the UE using a wide beam, perform a finer-grained beam scanning using a narrow beam only for that angle. As shown in Figure 14, it is assumed that in each time period of T0-T1, T1-T2, T2-T3, and T3-T4, the beam set A needs to be scanned and measured.

Use case 1 involves implementing beam prediction in the spatial domain. That is, in each time period, the AI model predicts the optimal beam based only on a subset B of the beam set A. As can be seen in the figure, this clearly reduces measurement overhead. However, the question is whether the saved measurement overhead is worth it. In other words, during traditional beam monitoring and beam failure recovery, the beam set that needs to be measured each time is not the entire set. The performance of existing beam management mechanisms is better than the prediction performance of AI models, and the model's prediction performance needs to be monitored during use. If the model's reduced measurement overhead is close to that of existing mechanisms, then using traditional beam management mechanisms may be more popular.

Furthermore, because spatial-only beam prediction has no concept of time—that is, it does not consider the time-domain characteristics of the channel—such a model cannot output any time information, such as the beam's dwell time. The model can wait until a beam failure occurs before restarting, but the link's transmission quality is impaired after a beam failure, which may cause the model to have difficulty with the required input. While the UE can still measure periodic SSBs, the SSB index must be demodulated to obtain it. Therefore, it is desirable to activate the model's spatial beam prediction at a better time than when a beam failure occurs. Furthermore, when using AI for beam management, beam switching should be based on model predictions rather than traditional mechanisms; otherwise, the spatial beam prediction model becomes unnecessary. As can be clearly seen in the figure, activation of the spatial beam prediction model can be triggered each time a reference signal measurement is instructed. However, current simulation results show that this method of triggering the spatial beam prediction model only reduces measurement overhead in a single operation while ensuring performance.

Use case 2 refers to time-domain beam prediction. Unlike the spatial-domain beam prediction in use case 1, the model used for time-domain beam prediction can collect time-related information. That is, the model output can include the predicted dwell time of the candidate beams. The essence of the time-domain beam prediction model is to fully measure beam set A but extend the measurement period of the beam management reference signal (BMRRS) to reduce measurement overhead. The model can be reactivated after the dwell time expires. However, for the UE-side model, the UE also requires time to measure, collect, and filter the model input data. If the BMRRS is still monitored during the model's output period—for example, as shown in Figure 14, the UE still needs to measure the BMRRS during the time periods T1-T2 or T3-T4—then the model's significance is greatly reduced. Therefore, for time-domain beam prediction, the timing and period of model activation also need to be studied.

Use case 3 involves joint spatial and temporal beam prediction. This model performs both spatial and temporal beam prediction. This reduces the number of beams measured in a single measurement, for example, by measuring only a subset of beam set A. This also extends the measurement period for beam management reference signals. The model outputs candidate beams predicted for multiple future moments, including the current moment. In this case, the same challenges as in use cases 1 and 2 persist.

The second embodiment of the present disclosure aims to reasonably manage the activation and deactivation of the beam prediction model so as to minimize the measurement overhead while ensuring the prediction performance of the model.

Figure 15 is a flowchart illustrating the model activation process according to this embodiment. The process may begin at step S41, where the base station determines an activation period for the UE-side beam prediction model. The beam prediction model may utilize, for example, a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory network (LSTM), etc. Its structure may include an input layer, multiple hidden layers, and an output layer. The hidden layers may utilize different types of neural network layers, such as convolutional layers and recurrent layers.

Step S41 typically occurs after the UE-side AI model is selected. Through pre-configuration or online download, the UE can install an AI model trained by the equipment vendor or mobile operator, or even multiple AI models. When multiple AI models are available, the UE needs to select which model to use and report the selection result to the base station. This stage is called model identification. For the AI model determined on the UE side, the base station can determine its activation period by considering various factors.

In one example, the activation period of the AI model needs to take into account the capabilities of the UE, such as the time required to measure reference signals and collect and filter model input data, or the speed of running the AI model, etc. In another example, the parameters of the model itself also need to be taken into consideration. For example, when using an AI model for speculation, the period of input and output in the time domain varies from model to model. The UE can report this information as UE capabilities to the base station, for example, through RRC signaling. However, preferably, the UE can determine the minimum activation period that the UE can support based on its capabilities and/or model parameters, and report it to the base station. For example, when selecting an AI model, the UE can determine the minimum activation period corresponding to the model and send it to the base station as an RRC parameter during the model identification phase. The activation period determined by the base station for the AI model should not be shorter than this minimum activation period.

The stability of the wireless channel can also affect the activation period of the beam prediction model. For example, for UEs with relatively fixed locations (such as terminal sensors or actuators in automated factories) or UEs with relatively stable mobility (such as terminal devices on a smoothly moving high-speed train), the spatial and temporal characteristics of the channel between them and the base station are also relatively stable. In other words, such UEs switch beams with a lower frequency or greater regularity, and the model can use a longer activation period. The UE can report information about its location or mobility attributes to the base station through, for example, RRC signaling.

Alternatively, channel characteristics can be perceived through reference signal measurements. This is particularly useful in scenarios where channel conditions are less stable, as the wireless channel fluctuates rapidly, requiring more frequent channel monitoring and prediction of appropriate beams. For example, in a scenario where only the spatial beam prediction module is deployed, changes in channel characteristics in the time domain and a high frequency of beam switching shorten the model's activation period. Conversely, if channel characteristics fluctuate less, the model's activation period increases.

In one example, the UE can measure a downlink reference signal, such as a CSI-RS or a demodulation reference signal (DMRS), and report the measurement results to the base station through the UCI. In another example, the base station can also directly measure an uplink reference signal sent by the UE, such as a sounding reference signal (SRS). Based on the attributes reported by the UE or based on the measurement of the downlink reference signal or the uplink reference signal, the base station can evaluate the time domain variation characteristics of the channel to determine the corresponding AI model activation period.

Subsequently, in step S42, the base station sends the determined activation period to the UE. The base station may send the activation period via RRC parameters. This can reduce signaling overhead for models with relatively stable activation periods. Alternatively, the base station may send the activation period via dynamic control signaling, such as MAC CE or DCI, which is particularly suitable for situations where the activation period changes frequently.

In step S43, the base station may send beam management reference signals, such as CSI-RS, to the UE according to the determined activation period for the UE to measure as model input. The base station may send the beam management reference signal only once in each activation period. Depending on whether the AI model on the UE side can perform spatial beam prediction, the base station may use the full set of scanning beams or a subset of them to send the reference signal.

In step S44, the UE activates its AI model according to the activation period received in step S42. In step S43, the UE receives and measures the beam management reference signal and, based on the measurement results, extracts feature data to be input into the AI model, thereby implementing beam prediction. As shown in the dashed box in the figure, steps S43 and S44 can be repeatedly performed according to the activation period.

The above describes how the base station controls the activation of the AI model on the UE side through the activation period. However, there may be scenarios where the AI model needs to be activated for beam prediction even before the activation period. According to this embodiment, some specific activation conditions are considered to trigger the activation of the model.

1) Start model monitoring

Model monitoring monitors the prediction capabilities of AI models, for example, by comparing predictions with traditional beam training results. Therefore, when model monitoring is enabled, both the AI model mechanism and the traditional mechanism coexist. Furthermore, model monitoring typically calculates the accuracy of multiple beam predictions, so the model monitoring period must be longer than the model activation period. Model monitoring can be enabled periodically, configured on the network side.

When model monitoring needs to be activated, the base station can send an activation command for the AI model to the UE to instruct the UE to activate the AI model for beam prediction. The activation command can include the value of the model monitoring period or indicate the number of activation periods spanned by the model monitoring.

2) Model performance testing

The base station may need to test the performance of the AI model on the UE side. At this time, the base station can send an activation command to the UE to activate the corresponding model for performance testing. In one example, when there are multiple candidate models on the UE side, one needs to be selected for use. The base station can manage the models on the UE side. The UE may not be able to support the simultaneous activation of all models, so the base station can indicate the order in which these models should be activated in the activation command. The activation order can be determined by the base station according to the information received in the model identification phase or the scenario to which the model is applicable.

3) Beam failure occurs

Beam failure will inevitably cause the model to re-predict. In the traditional beam management mechanism, the base station configures the beam management reference signal set for the UE to find candidate beams, but only when a beam failure occurs will the new candidate beam be reported to the base station through the physical random access channel (PRACH). The question that needs to be considered is whether it is necessary to configure this set when using AI for beam management. If this reference signal set is always configured, the beam switching can be based entirely on the traditional mechanism. If this set is not configured, after the beam fails, the UE needs to periodically measure the SSB, but as mentioned earlier, the SSB index is obtained after demodulation, and the model should also support wide beams as input.

In this scenario, the UE can monitor its beam signal quality. If it falls below a predetermined threshold, it indicates a beam failure. Even if the activation period has not yet arrived, the UE can immediately activate the AI model for beam prediction. In this case, the AI model needs to be able to scan the SSB wide beam as input in the idle state to quickly recover the beam after a beam failure. Otherwise, the traditional beam failure recovery mechanism is still used.

4) Radio link quality deteriorates. Existing link monitoring mechanisms can detect degradation in link quality, and it's possible to restore it by switching beams. However, consideration may be given to the extent to which the link quality deteriorates. If the link quality deteriorates enough to trigger a link failure, the UE returns to the idle state, requiring SSB measurements to activate the AI model for beam prediction. This places demands on the model's capabilities and requires a longer recovery time, making it undesirable.

According to this embodiment, beam switching can be considered before link failure occurs. A threshold can be set, and when the UE detects that the link quality has dropped below this threshold, the model is activated for prediction. It is important to note that this threshold should be set to trigger before link failure occurs. Assuming that the link failure judgment threshold is link quality threshold th1, and the trigger model activates link quality threshold th2, when th2 > th1, it is possible to avoid link failure in advance by switching beams, allowing the system to continuously maintain the optimal link matching state.

Fig. 16 shows a block diagram of an electronic device 200 according to this embodiment. The electronic device 200 may be implemented as a UE or a component thereof.

As shown in FIG16 , electronic device 200 includes processing circuitry 201. Processing circuitry 201 includes at least a receiving unit 202, an activation unit 203, and a prediction unit 204. Processing circuitry 201 may be configured to perform the operations shown in FIG15 . Processing circuitry 201 may refer to various implementations of digital circuitry, analog circuitry, or mixed-signal (a combination of analog and digital signals) circuitry that performs functions in a UE.

The receiving unit 202 is configured to receive information about the activation period of the beam prediction model of the UE from the base station, that is, to execute step S42. The activation period can be carried by RRC signaling or dynamic control signaling such as MAC CE or DCI.

The activation unit 203 is configured to activate its beam prediction model according to the activation period received by the receiving unit 202, that is, to execute step S44.

The prediction unit 204 is configured to perform beam prediction using the activated AI model based on the UE's measurement of the beam management reference signal (e.g., CSI-RS) sent by the base station according to the activation period. The AI model can perform beam prediction in the spatial domain, beam prediction in the time domain, or beam prediction in both the spatial and time domains.

The electronic device 200 may further include a communication unit 205. The communication unit 205 may be configured to communicate with a base station (e.g., the electronic device 300 described below) under the control of the processing circuit 201. In one example, the communication unit 205 may be implemented as a transmitter or a transceiver, including communication components such as an antenna array and/or a radio frequency link. The communication unit 205 is depicted with a dashed line because it may also be located outside the electronic device 200.

The electronic device 200 may further include a memory 206. The memory 206 may store various data and instructions, programs and data for the operation of the electronic device 200, various data generated by the processing circuit 201, data to be transmitted by the communication unit 205, etc. The memory 206 is drawn with a dotted line because it may also be located within the processing circuit 201 or outside the electronic device 200.

Fig. 17 shows a block diagram of an electronic device 300 according to this embodiment. The electronic device 300 may be implemented as a base station or a component thereof.

As shown in FIG17 , electronic device 300 includes processing circuitry 301. Processing circuitry 301 includes at least a determining unit 302 and a transmitting unit 303. Processing circuitry 301 can be configured to perform the operations shown in FIG15 . Processing circuitry 301 can refer to various implementations of digital circuitry, analog circuitry, or mixed-signal (a combination of analog and digital signals) circuitry that performs functions in a base station device.

The determining unit 302 is configured to determine the activation period of the beam prediction model on the UE side, that is, to execute step S41 in Figure 15. The determining unit 302 may determine the activation period based on the minimum activation period and/or channel status information reported by the UE.

The sending unit 303 is configured to send the determined activation period to the UE, i.e., execute step S42 in Figure 15. The activation period can be carried by RRC signaling or dynamic control signaling such as MAC CE or DCI.

In addition, the sending unit 303 is also configured to send a beam management reference signal (such as CSI-RS) to the UE according to the activation period for the AI model on the UE side to perform beam prediction, that is, execute step S43 in Figure 15.

The electronic device 300 may further include a communication unit 305. The communication unit 305 may be configured to communicate with a UE (e.g., the electronic device 200 described above) under the control of the processing circuit 301. In one example, the communication unit 305 may be implemented as a transmitter or a transceiver, including communication components such as an antenna array and/or a radio frequency link. The communication unit 305 is depicted with a dashed line because it may also be located outside the electronic device 300.

The electronic device 300 may further include a memory 306. The memory 306 may store various data and instructions, programs and data for the operation of the electronic device 300, various data generated by the processing circuit 301, data to be transmitted by the communication unit 305, etc. The memory 306 is drawn with a dotted line because it may also be located within the processing circuit 301 or outside the electronic device 300.

Various aspects of the embodiments of the present disclosure have been described in detail above, but it should be noted that the above description of the structure, arrangement, type, quantity, etc. of the antenna array shown, ports, reference signals, communication equipment, communication methods, etc. is not intended to limit the aspects of the present disclosure to these specific examples.

It should be understood that the various units of the electronic devices 100, 200 and 300 described in the above embodiments are merely logical modules divided according to the specific functions they implement, and are not intended to limit specific implementation methods. In actual implementation, the above units can be implemented as independent physical entities, or can also be implemented by a single entity (for example, a processor (CPU)).
or DSP, etc.), integrated circuits, etc.) to achieve this.

It should be understood that the processing circuits 101, 201, and 301 described in the above embodiments may include, for example, circuits such as integrated circuits (ICs), application-specific integrated circuits (ASICs), portions or circuits of a separate processor core, the entire processor core, a separate processor, a programmable hardware device such as a field programmable gate array (FPGA), and/or a system including multiple processors. The memories 106, 206, and 306 may be volatile memories and/or non-volatile memories. For example, the memories may include, but are not limited to, random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), read-only memory (ROM), and flash memory.

It should be understood that the various units of the electronic devices 100, 200 and 300 described in the above embodiments are merely logical modules divided according to the specific functions they implement, and are not intended to limit specific implementation methods. In actual implementation, the above units can be implemented as independent physical entities, or can also be implemented by a single entity (for example, a processor (CPU)).
or DSP, etc.), integrated circuits, etc.) to achieve this.

[Exemplary Implementation of the Present Disclosure]

According to the embodiments of the present disclosure, various implementations of the concepts of the present disclosure may be conceived, including but not limited to the following exemplary examples (EE):

EE1. An electronic device comprising:

processor; and

a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising:

Determining whether there is self-occlusion of a signal received by a user equipment (UE) caused by a posture of a user operating the UE;

Predicting, by an artificial intelligence (AI) model, beam self-occlusion information associated with a receive beam set of the UE from a received signal power of the UE; and

Based on the beam self-blocking information, the receiving beam used by the UE is switched.

EE2. The electronic device according to EE1, wherein determining the presence of self-occlusion comprises:

After the received signal power drops below a threshold, detecting whether the received signal power of the UE within a time period exhibits a predefined characteristic, wherein the predefined characteristic includes no jitter; and

When the predefined feature is not detected, it is determined that self-occlusion exists.

EE3. The electronic device according to EE1, wherein determining the presence of self-occlusion comprises:

The presence of self-blocking is determined by inputting the received signal power of the UE within a time period into the AI model.

EE4. The electronic device according to EE3, wherein the operation further comprises:

In addition to the received signal power of the UE, auxiliary information related to the user's posture is also input into the AI model, and the auxiliary information includes at least one of the following: touch screen information, gyroscope information, camera information, and infrared sensor information.

EE5. The electronic device according to EE1, wherein the beam self-blocking information includes priority information indicating a selection priority of a reception beam in the reception beam set under the self-blocking, and

Wherein, switching the receiving beam used by the UE is based on the priority information.

EE6. The electronic device according to EE1 or EE5, wherein the beam self-obstruction information includes state information indicating a self-obstruction state of each receive beam in the receive beam set.

EE7. The electronic device according to EE6, wherein the operation further comprises:

generating, based on the status information, a self-obstruction status report indicating an effect of the self-obstruction on a transmit beam of the base station;

Sending the self-shading status report to a base station.

EE8. The electronic device according to EE7, wherein the operation further comprises:

Switching the receive beam used by the UE by performing beam training between the multiple receive beams determined by the UE based on the status information and the multiple transmit beams determined by the base station based on the self-obstruction status report;

The self-blocking status report indicates the expected base station transmission beam range or the expected base station transmission beam index.

EE9. The electronic device according to EE1, wherein the operation further comprises:

detecting a received signal power of the UE; and

In a case where the received signal power of the UE is lower than a predetermined threshold, determining whether self-blocking exists is performed.

EE10. The electronic device according to EE1, wherein the operation further comprises:

receiving configuration information about the activation period of the AI model from a base station; and

The AI model is activated according to the activation cycle.

EE11. The electronic device according to EE1, wherein the beam self-occlusion information further includes a duration T of the self-occlusion, and wherein the operation further includes:

After the duration T, the AI model is reactivated.

EE12. The electronic device according to EE11, wherein the beam self-occlusion information includes state information I t and state information Γ _t respectively indicating a self-occlusion state of each receive beam in the receive beam set at a current time t and a time (t+T) after a duration T, and wherein the operation further comprises:

At time t, record the state information Γt predicted by the AI model;

At time (t+T), the AI model is used to predict state information I _t+T ;

Calculating the prediction accuracy of the AI model by comparing the state information Γt and the state information I _t+T ; and

When the prediction accuracy is lower than a predetermined threshold, the AI model is updated.

EE13. An electronic device comprising:

processor; and

a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising:

Receiving a self-occlusion status report from a user equipment (UE), the self-occlusion status report indicating an impact of self-occlusion caused by a user's posture of operating the UE on a transmit beam of a base station and based on beam self-occlusion information predicted by the UE using an artificial intelligence (AI) model; and

Based on the self-obstruction status report, a plurality of transmit beams are determined for beam training between the base station and the UE.

EE14. The electronic device according to EE13, wherein the operation further comprises:

Sending configuration information about the activation period of the AI model to the UE.

EE15. An electronic device comprising:

processor; and

a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising:

Preparing a training data set including input data and output data, wherein the input data includes received signal power of a user equipment (UE) associated with multiple postures of a user operating the UE, and the output data includes beam self-occlusion information of a receive beam set of the UE associated with the multiple postures; and

An artificial intelligence (AI) model is trained on the training set to determine parameters of the AI model.

EE16. The electronic device according to EE15, wherein the beam self-occlusion information includes at least one of the following:

priority information indicating a selection priority of a reception beam in the reception beam set under self-occlusion caused by a user's gesture;

Status information indicating a self-occlusion status of each receive beam in the receive beam set;

the duration of the self-occlusion; and

State information of a self-occlusion state of each receive beam in the receive beam set at a time after the duration.

EE17. The electronic device according to EE15, wherein the operations further comprise:

Collecting personalized characteristic data related to specific user's behavior habits; and

The AI model is trained using the personalized feature data to fine-tune the parameters of the AI model.

EE18. A method comprising:

Determining whether there is self-occlusion of a signal received by a user equipment (UE) caused by a posture of a user operating the UE;

Predicting beam self-occlusion information associated with the beam set of the UE from the received signal power of the UE through an artificial intelligence (AI) model; and

Based on the beam self-blocking information, the beam used by the UE is switched.

EE19. A method comprising:

receiving a self-occlusion status report from a user equipment (UE), the self-occlusion status report indicating an impact of self-occlusion caused by a user's posture operating the UE on a transmit beam of a base station and based on beam self-occlusion information predicted by the UE using an artificial intelligence (AI) model;

Based on the self-obstruction status report, a plurality of transmit beams are determined for beam training between the base station and the UE.

EE20. A method comprising:

Preparing a training data set including input data and output data, wherein the input data includes received signal power of a user equipment (UE) associated with multiple postures of a user operating the UE, and the output data includes beam self-occlusion information of a beam set of the UE associated with the multiple postures; and

An artificial intelligence (AI) model is trained on the training set to determine parameters of the AI model.

EE21. An electronic device comprising:

processor; and

a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising:

receiving, from a base station, information about an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction;

activating the AI model according to the activation cycle; and

Based on the measurement of the beam management reference signal sent by the base station according to the activation period, beam prediction is performed using the activated AI model.

EE22. The electronic device according to EE21, wherein the information about the activation period of the AI model is included in radio control resource (RRC) signaling or dynamic control signaling.

EE23. The electronic device according to EE21, wherein the operation further comprises:

Determining a minimum activation period of the AI model based on the capabilities of the UE and parameters of the AI model; and

The minimum activation period is reported to a base station, wherein the activation period is not less than the minimum activation period.

EE24. An electronic device according to EE23, wherein the minimum activation period is reported during the model identification phase of the AI model.

EE25. The electronic device according to EE21, wherein the operation further comprises:

An uplink reference signal or a measurement of a downlink reference signal is sent to a base station, wherein the activation period is determined by the base station based on the measurement of the uplink reference signal or the downlink reference signal.

EE26. The electronic device according to EE21, wherein the AI model is configured to perform one of the following:

Beam prediction in the airspace;

Beam prediction in the time domain;

Beam prediction in spatial and temporal domains.

EE27. The electronic device according to EE21, wherein the operation further comprises:

receiving an activation command for the AI model from a base station; and

According to the activation command, the AI model is activated for beam prediction.

EE28. The electronic device according to EE27, wherein the activation command indicates the start of model monitoring of the AI model and a model monitoring period, and the model monitoring period is greater than the activation period; or

The activation command indicates the start of the performance test of the AI model.

EE29. The electronic device according to EE21, wherein the operation further comprises:

monitoring the beam signal quality of the UE; and

In response to monitoring a beam failure, the AI model is activated for beam prediction, wherein the activated AI model uses a measurement of a beam signal of a synchronization signal block (SSB) as input.

EE30. The electronic device according to EE21, wherein the operation further comprises:

monitoring a radio link quality of the UE; and

In response to monitoring that the radio link quality of the UE is lower than a predetermined threshold, activating the AI model for beam prediction, wherein the predetermined threshold is higher than a decision threshold for radio link failure.

EE31. The electronic device according to EE21, wherein the operation further comprises:

The prediction result is reported to the base station only when the prediction result of the AI model in the current activation cycle is inconsistent with the prediction result of the previous activation cycle.

EE32. An electronic device comprising:

processor; and

a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising:

determining an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction;

sending information about the determined activation period to the UE; and

According to the activation period, a beam management reference signal is sent for the AI model to perform beam prediction.

EE33. The electronic device according to EE32, wherein the information about the activation period of the AI model is included in radio control resource (RRC) signaling or dynamic control signaling.

EE34. The electronic device according to EE32, wherein the operation further comprises:

receiving information about a minimum activation period of the AI model from the UE; and

The activation period is determined based on the minimum activation period, wherein the activation period is not less than the minimum activation period.

EE35. The electronic device according to EE32, wherein the minimum activation period is received during an identification phase of the AI model.

EE36. The electronic device according to EE32, wherein the operation further comprises:

receiving an uplink reference signal or a measurement of a downlink reference signal from the UE; and

The activation period is determined based on measurement of an uplink reference signal or a downlink reference signal.

EE37. The electronic device according to EE32, wherein the AI model is configured to perform one of the following:

Beam prediction in the airspace;

Beam prediction in the time domain;

Beam prediction in spatial and temporal domains.

EE38. The electronic device according to EE32, wherein the operation further comprises:

An activation command regarding the AI model is sent to the UE, wherein the UE activates the AI model for beam prediction in response to the activation command.

EE39. The electronic device according to EE32, wherein the activation command indicates the start of model monitoring of the AI model and a model monitoring period, and the model monitoring period is greater than the activation period; or

The activation command indicates the start of the performance test of the AI model.

EE40. A method comprising:

receiving, from a base station, information about an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction;

activating the AI model according to the activation cycle; and

Based on the measurement of the beam management reference signal sent by the base station according to the activation period, beam prediction is performed using the activated AI model.

EE41. A method comprising:

determining an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction;

sending the determined activation period to the UE; and

According to the activation period, a beam management reference signal is sent for the AI model to perform beam prediction.

EE42. A computer program product comprising executable instructions, which, when executed, cause an electronic device to perform the method as described in any one of EE18-EE20 and EE40-EE41.

[Application Examples of the Present Disclosure]

FIG18 shows an example block diagram of a computer that can be implemented as a sending device, a relay device, or a receiving device according to an embodiment of the present disclosure.

In FIG18 , a central processing unit (CPU) 1301 executes various processes according to a program stored in a read-only memory (ROM) 1302 or a program loaded from a storage unit 1308 into a random access memory (RAM) 1303. RAM 1303 also stores data required when CPU 1301 executes various processes, etc., as needed.

The CPU 1301, the ROM 1302, and the RAM 1303 are connected to each other via a bus 1304. An input/output interface 1305 is also connected to the bus 1304.

The following components are connected to the input/output interface 1305: an input section 1306 including a keyboard, a mouse, etc.; an output section 1307 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1308 including a hard disk, etc.; and a communication section 1309 including a network interface card such as a LAN card, a modem, etc. The communication section 1309 performs communication processing via a network such as the Internet.

A drive 1310 is also connected to the input/output interface 1305 as needed. A removable medium 1311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is mounted on the drive 1310 as needed so that a computer program read therefrom is installed in the storage section 1308 as needed.

In the case of realizing the above-described series of processing by software, a program constituting the software is installed from a network such as the Internet or a storage medium such as the removable medium 1311 .

It should be understood by those skilled in the art that such storage medium is not limited to the removable medium 1311 shown in FIG. 18 , in which the program is stored and which is distributed separately from the device to provide the program to the user. Examples of the removable medium 1311 include magnetic disks (including floppy disks (registered trademark)), optical disks (including compact disk read-only memories (CD-ROMs) and digital versatile disks (DVDs)), magneto-optical disks (including minidiscs (MDs) (registered trademark)), and semiconductor memories. Alternatively, the storage medium may be the ROM 1302, a hard disk included in the storage portion 1308, or the like, in which the program is stored and distributed to the user together with the device containing them.

In the server 1300 shown in FIG. 18 , the processing circuit 101 described with reference to FIG. 13 , the processing circuit 201 described with reference to FIG. 16 , or the processing circuit 301 described with reference to FIG. 17 may be implemented by the processor 701 .

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

For example, the electronic device 300 according to an embodiment of the present disclosure may be implemented as various base stations or installed in a base station, and the electronic device 100 or 200 may be implemented as various user equipments or installed in various user equipments.

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.

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.

The base station referred to in this disclosure can be implemented as any type of base station, preferably, such as the macro gNB and ng-eNB defined in the 3GPP 5GNR standard. The gNB can be a gNB that covers a cell smaller than a macro cell, such as a pico gNB, micro gNB, and home (femto) gNB. Alternatively, the base station can be implemented as any other type of base station, such as a NodeB, eNodeB, and base transceiver station (BTS). The base station may also include: a main body configured to control wireless communications and one or more remote radio heads (RRHs) located separately from the main body, wireless relay stations, drone towers, control nodes in automated factories, etc.

The user equipment can be implemented as a mobile terminal (such as a smartphone, a tablet personal computer (PC), a notebook PC, a portable game terminal, a portable/dongle-type mobile router, and a digital camera) or an in-vehicle terminal (such as a car navigation device). The user equipment can also be implemented as a terminal that performs machine-to-machine (M2M) communication (also known as a machine-type communication (MTC) terminal), a drone, a sensor and actuator 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 chip) installed on each of the above terminals.

First application example of base station

FIG19 is a block diagram illustrating a first example of a schematic configuration of a base station to which the techniques of this disclosure can be applied. In FIG19 , the base station may be implemented as gNB 1400. gNB 1400 includes multiple antennas 1410 and a base station device 1420. Base station device 1420 and each antenna 1410 may be connected to each other via an RF cable. In one implementation, gNB 1400 (or base station device 1420) may correspond to the electronic device 300 for a receiving device described above.

Antenna 1410 includes multiple antenna elements. Antenna 1410 can be arranged, for example, in an antenna array matrix and used by base station device 1420 to transmit and receive wireless signals. For example, multiple antennas 1410 can be compatible with multiple frequency bands used by gNB 1400.

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

The controller 1421 may be, for example, a CPU or DSP, and operates various higher-layer functions of the base station device 1420. For example, the controller 1421 may include the processing circuit 301 described above, or control various components of the base station device 300. For example, the controller 1421 generates data packets based on data in the signal processed by the wireless communication interface 1425 and transmits the generated packets via the network interface 1423. The controller 1421 may bundle data from multiple baseband processors to generate bundled packets and transmit the generated bundled packets. The controller 1421 may have logic functions for performing control such as radio resource control, radio bearer control, mobility management, admission control, and scheduling. This control may be performed in conjunction with nearby gNBs or core network nodes. The memory 1422 includes RAM and ROM and stores programs executed by the controller 1421 and various types of control data (such as terminal lists, transmission power data, and scheduling data).

Network interface 1423 is a communication interface for connecting base station device 1420 to core network 1424 (e.g., a 5G core network). Controller 1421 can communicate with a core network node or another gNB via network interface 1423. In this case, gNB 1400 and the core network node or other gNB can be connected to each other via logical interfaces (such as NG interfaces and Xn interfaces). Network interface 1423 can also be a wired communication interface or a wireless communication interface for wireless backhaul. If network interface 1423 is a wireless communication interface, network interface 1423 can use a higher frequency band for wireless communication than the frequency band used by wireless communication interface 1425.

The wireless communication interface 1425 supports any cellular communication scheme (such as 5G NR) and provides wireless connectivity to terminals located in the cell of the gNB 1400 via the antenna 1410. The wireless communication interface 1425 may typically include, for example, a baseband (BB) processor 1426 and RF circuitry 1427. The BB processor 1426 can perform various signal processing functions, such as encoding/decoding, modulation/demodulation, and multiplexing/demultiplexing, and can also perform various types of signal processing at various layers (e.g., the physical layer, MAC layer, RLC layer, PDCP layer, and SDAP layer). In place of the controller 1421, the BB processor 1426 may have some or all of the aforementioned logical functions. The BB processor 1426 may be a memory that stores communication control programs, or a module including a processor configured to execute programs and associated circuitry. Program updates can modify the functionality of the BB processor 1426. This module may be a card or blade inserted into a slot in the base station device 1420. Alternatively, it may be a chip mounted on the card or blade. Meanwhile, the RF circuit 1427 may include, for example, a mixer, a filter, and an amplifier, and transmits and receives wireless signals via the antenna 1410. Although FIG19 shows an example in which one RF circuit 1427 is connected to one antenna 1410, the present disclosure is not limited to this illustration, and one RF circuit 1427 may be connected to multiple antennas 1410 at the same time.

As shown in FIG19 , 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 gNB 1400 . As shown in FIG19 , wireless communication interface 1425 may include multiple RF circuits 1427 . For example, multiple RF circuits 1427 may be compatible with multiple antenna elements. Although FIG19 illustrates an example in which wireless communication interface 1425 includes multiple BB processors 1426 and multiple RF circuits 1427 , wireless communication interface 1425 may also include a single BB processor 1426 or a single RF circuit 1427 .

In the gNB 1400 shown in FIG19 , one or more units included in the processing circuit 301 described with reference to FIG17 may be implemented in the wireless communication interface 825. Alternatively, at least a portion of these components may be implemented in the controller 821. For example, the gNB 1400 may include a portion (e.g., the BB processor 1426) or the entirety of the wireless communication interface 1425, and/or a module including the controller 1421, 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 function as one or more components (in other words, a program that allows the processor to perform the operations of one or more components) and may execute the program. As another example, the program that allows the processor to function as one or more components may be installed in the gNB 1400, and the wireless communication interface 1425 (e.g., the BB processor 1426) and/or the controller 1421 may execute the program. As described above, the gNB 1400, base station device 1420, or module may be provided as a device including one or more components, and the program that allows the processor to function as one or more components may also be provided. In addition, a readable medium having the program recorded therein can be provided.

Second application example of base station

FIG20 is a block diagram illustrating a second example of a schematic configuration of a base station to which the technology of the present disclosure can be applied. In FIG20 , the base station is illustrated as gNB 1530. gNB 1530 includes multiple antennas 1540, a base station device 1550, and an RRH 1560. The RRH 1560 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. In one implementation, the gNB 1530 (or base station device 1550) herein may correspond to the electronic device 300 for the receiving device described above.

Antenna 1540 includes multiple antenna elements. Antenna 1540 can be arranged, for example, in an antenna array matrix and used by base station device 1550 to transmit and receive wireless signals. For example, multiple antennas 1540 can be compatible with multiple frequency bands used by gNB 1530.

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. Controller 1551, memory 1552, and network interface 1553 are the same as controller 1421, memory 1422, and network interface 1423 described with reference to FIG.

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. The BB processor 1556 is the same as the BB processor 1426 described with reference to FIG. 19 , except that the BB processor 1556 is connected to the RF circuit 1564 of the RRH 1560 via the connection interface 1557. As shown in FIG. 20 , the wireless communication interface 1555 may include multiple BB processors 1556. For example, the multiple BB processors 1556 may be compatible with multiple frequency bands used by the gNB 1530. Although FIG. 20 shows an example in which the wireless communication interface 1555 includes multiple BB processors 1556, the wireless communication interface 1555 may also include a single BB processor 1556.

Connection interface 1557 is an interface for connecting base station device 1550 (wireless communication interface 1555) to RRH 1560. Connection interface 1557 may also be a communication module for connecting base station device 1550 (wireless communication interface 1555) to RRH 1560 for communication via the aforementioned high-speed line.

RRH 1560 includes a connection interface 1561 and a wireless communication interface 1563.

Connection interface 1561 is an interface for connecting RRH 1560 (wireless communication interface 1563) to base station device 1550. Connection interface 1561 can also be a communication module for communication in the above-mentioned high-speed line.

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 FIG. 20 illustrates an example in which one RF circuit 1564 is connected to one antenna 1540, the present disclosure is not limited to this illustration, and one RF circuit 1564 may be connected to multiple antennas 1540 simultaneously.

As shown in FIG20 , the wireless communication interface 1563 may include multiple RF circuits 1564. For example, multiple RF circuits 1564 may support multiple antenna elements. Although FIG20 shows an example in which the wireless communication interface 1563 includes multiple RF circuits 1564, the wireless communication interface 1563 may also include a single RF circuit 1564.

In the gNB 1500 shown in FIG. 20 , one or more units included in the processing circuit 301 described with reference to FIG. 17 may be implemented in the wireless communication interface 1525. Alternatively, at least a portion of these components may be implemented in the controller 1521. For example, the gNB 1500 may include a portion (e.g., the 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 that allows the processor to function as one or more components (in other words, a program that allows the processor to perform the operations of one or more components) and may execute the program. As another example, the program that allows the processor to function as one or more components may be installed in the gNB 1500, and the wireless communication interface 1525 (e.g., the BB processor 1526) and/or the controller 1521 may execute the program. As described above, the gNB 1500, base station device 1520, or module may be provided as a device including one or more components, and the program that allows the processor to function as one or more components may also be provided. In addition, a readable medium having the program recorded therein can be provided.

First application example of user equipment

21 is a block diagram illustrating an example of a schematic configuration of a smartphone 1600 to which the technology of the present disclosure may be applied. In one example, the smartphone 1600 may be implemented as the electronic device 100 or 200 described in the present disclosure.

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.

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 the processing circuit 101 described with reference to FIG13 or the processing circuit 201 described with reference to FIG16. The memory 1602 includes RAM and ROM, and stores data and programs executed by the processor 1601 to implement the communication method described above. 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.

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 gyroscope sensor, a geomagnetic sensor, and an acceleration sensor. The microphone 1608 converts the sound input to the smartphone 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 the 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 the output image of the smartphone 1600. The speaker 1611 converts the audio signal output from the smartphone 1600 into sound.

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. As shown in FIG. 21 , the wireless communication interface 1612 may include multiple BB processors 1613 and multiple RF circuits 1614. Although FIG. 21 shows an example in which the wireless communication interface 1612 includes multiple BB processors 1613 and multiple RF circuits 1614, the wireless communication interface 1612 may also include a single BB processor 1613 or a single RF circuit 1614.

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

Each of the antenna switches 1615 switches the 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 .

The antenna 1616 includes a plurality of antenna elements. The antenna 1616 may be arranged in an antenna array matrix, for example, and is used for the wireless communication interface 1612 to transmit and receive wireless signals. The smartphone 1600 may include one or more antenna panels (not shown).

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.

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. The battery 1618 supplies power to the various blocks of the smartphone 1600 shown in FIG. 21 via feeders, which are partially shown as dashed lines in the figure. The auxiliary controller 1619 operates the minimum necessary functions of the smartphone 1600, for example, in sleep mode.

In the smartphone 1600 shown in FIG21 , one or more components included in the processing circuit 101 described with reference to FIG13 or the processing circuit 201 described with reference to FIG16 may be implemented in the wireless communication interface 1612. Alternatively, at least a portion of these components may be implemented in the processor 1601 or the auxiliary controller 1619. As an example, the smartphone 1600 includes a portion (e.g., the BB processor 1613) or the entirety 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 function as one or more components (in other words, a program for allowing the processor to perform the operations of one or more components) and may execute the program. As another example, a program for allowing the processor to function as one or more components may be installed in the smartphone 1600, and the wireless communication interface 1612 (e.g., the BB processor 1613), the processor 1601, and/or the auxiliary controller 1619 may execute the program. As described above, the smartphone 1600 or module may be provided as a device including one or more components, and a program for allowing a processor to function as one or more components may be provided. In addition, a readable medium having the program recorded therein may be provided.

Second application example of user equipment

Figure 22 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 can be implemented as the electronic device 100 described with reference to Figure 13 or the electronic device 200 described with reference to Figure 16. 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. In one example, the car navigation device 1720 can be implemented as the electronic device 100 or 200 described in the present disclosure.

The processor 1721 may be, for example, a CPU or an 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 programs executed by the processor 1721.

The GPS module 1724 uses GPS signals received from GPS satellites to measure the position (such as latitude, longitude, and altitude) of the car navigation device 1720. The sensor 1725 may include a group of sensors such as a gyroscope sensor, a geomagnetic sensor, and an air pressure sensor. The data interface 1726 is connected to, for example, the vehicle network 1741 via a terminal not shown, and obtains data generated by the vehicle (such as vehicle speed data).

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

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. As shown in FIG. 22 , the wireless communication interface 1733 may include multiple BB processors 1734 and multiple RF circuits 1735. Although FIG. 22 shows an example in which the wireless communication interface 1733 includes multiple BB processors 1734 and multiple RF circuits 1735, the wireless communication interface 1733 may also include a single BB processor 1734 or a single RF circuit 1735.

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.

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.

The antenna 1737 includes a plurality of antenna elements and may be arranged in an antenna array matrix, for example, and is used by the wireless communication interface 1733 to transmit and receive wireless signals.

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.

The battery 1738 supplies power to the respective blocks of the car navigation device 1720 shown in Fig. 22 via a feeder line, which is partially shown as a dotted line in the figure. The battery 1738 accumulates the power supplied from the vehicle.

In the car navigation device 1720 shown in FIG. 22 , one or more components included in the processing circuit 101 described with reference to FIG. 13 or the processing circuit 201 described with reference to FIG. 16 may be implemented in the wireless communication interface 1733. Alternatively, at least a portion of these components may be implemented in the processor 1721. As an example, the car navigation device 1720 includes a portion (e.g., the BB processor 1734) or the entirety of the wireless communication interface 1733, and/or a module including the processor 1721, 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 function as one or more components (in other words, a program for allowing the processor to perform the operations of one or more components) and may execute the program. 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., the BB processor 1734) and/or the processor 1721 may execute the program. As described above, the car navigation device 1720 or a module may be provided as a device including one or more components, and a program for allowing the processor to function as one or more components may be provided. In addition, a readable medium having the program recorded therein can be provided.

The technology of the present disclosure can 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 of 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.

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.

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.

In this specification, the steps described in the flowchart include not only processing executed in time series in the order described, but also processing executed 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 changed as appropriate.

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 "comprises," "comprising," or any other variations thereof in the embodiments of the present disclosure are intended to cover non-exclusive inclusions, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

Claims (42)

An electronic device, comprising: processor; and a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising: Determining whether there is self-occlusion of a signal received by a user equipment (UE) caused by a posture of a user operating the UE; Predicting beam self-occlusion information associated with the beam set of the UE from the received signal power of the UE through an artificial intelligence (AI) model; and Based on the beam self-blocking information, the beam used by the UE is switched. The electronic device according to claim 1, wherein determining the presence of self-occlusion comprises: After the received signal power drops below a threshold, detecting whether the received signal power of the UE within a time period exhibits a predefined characteristic, wherein the predefined characteristic includes jitter; and When the predefined feature is not detected, it is determined that self-occlusion exists. The electronic device according to claim 1, wherein determining the presence of self-occlusion comprises: After the received signal power drops below a threshold, the received signal power of the UE within a time period is input into the AI model to determine whether self-blocking exists. The electronic device according to claim 3, wherein the operations further comprise: In addition to the received signal power of the UE, auxiliary information related to the user's posture is also input into the AI model, and the auxiliary information includes at least one of the following: touch screen information, gyroscope information, camera information, and infrared sensor information. The electronic device according to claim 1, wherein the beam self-occlusion information includes priority information indicating a selection priority of a beam in the beam set under the self-occlusion, and The switching of the beam used by the UE is based on the priority information. The electronic device according to claim 1 or 5, wherein the beam self-occlusion information includes state information indicating a self-occlusion state of each beam in the beam set. The electronic device according to claim 6, wherein the operations further comprise: generating, based on the status information, a self-obstruction status report indicating an effect of the self-obstruction on a transmit beam of the base station; Sending the self-shading status report to a base station. The electronic device according to claim 7, wherein the operations further comprise: Switching the receive beam used by the UE by performing beam training between the multiple receive beams determined by the UE based on the status information and the multiple transmit beams determined by the base station based on the self-obstruction status report; The self-blocking status report indicates the expected base station transmission beam range or the expected base station transmission beam index. The electronic device according to claim 1, wherein the operations further comprise: detecting a received signal power of the UE; and In a case where the received signal power of the UE is lower than a predetermined threshold, determining whether self-blocking exists is performed. The electronic device according to claim 1, wherein the operations further comprise: receiving configuration information about the activation period of the AI model from a base station; and The AI model is activated according to the activation cycle. The electronic device of claim 1, wherein the beam self-occlusion information further includes a duration T of the self-occlusion, and wherein the operations further comprise: After the duration T, the AI model is reactivated. The electronic device according to claim 11, wherein the beam self-occlusion information includes state information It and state information Γt respectively indicating the self-occlusion state of each beam in the beam set at a current time t and a time ( _t +T) after a duration _T , and wherein the operation further comprises: At time t, record the state information Γ _t predicted by the AI model; At time (t+T), the AI model is used to predict state information I _t+T ; Calculating the prediction accuracy of the AI model by comparing the state information Γ _t and the state information I _t+T ; and When the prediction accuracy is lower than a predetermined threshold, the AI model is updated. An electronic device, comprising: processor; and a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising: Receiving a self-occlusion status report from a user equipment (UE), the self-occlusion status report indicating an impact of self-occlusion caused by a user's posture of operating the UE on a transmit beam of a base station and based on beam self-occlusion information predicted by the UE using an artificial intelligence (AI) model; and Based on the self-obstruction status report, a plurality of transmit beams are determined for beam training between the base station and the UE. The electronic device according to claim 13, wherein the operations further comprise: Sending configuration information about the activation period of the AI model to the UE. An electronic device, comprising: processor; and a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising: Preparing a training data set including input data and output data, wherein the input data includes received signal power of a user equipment (UE) associated with multiple postures of a user operating the UE, and the output data includes beam self-occlusion information of a beam set of the UE associated with the multiple postures; and An artificial intelligence (AI) model is trained on the training set to determine parameters of the AI model. The electronic device according to claim 15, wherein the beam self-occlusion information comprises at least one of the following: priority information indicating a selection priority of beams in the set of beams under self-occlusion caused by a user's gesture; state information indicating a self-occlusion state of each beam in the set of beams; the duration of the self-occlusion; and State information of a self-occlusion state of each beam in the set of beams at a time instant after the duration. The electronic device according to claim 15, wherein the operations further comprise: Collecting personalized characteristic data related to specific user's behavior habits; and The AI model is trained using the personalized feature data to fine-tune the parameters of the AI model. A method comprising: Determining whether there is self-occlusion of a signal received by a user equipment (UE) caused by a posture of a user operating the UE; Predicting beam self-occlusion information associated with the beam set of the UE from the received signal power of the UE through an artificial intelligence (AI) model; and Based on the beam self-blocking information, the beam used by the UE is switched. A method comprising: receiving a self-occlusion status report from a user equipment (UE), the self-occlusion status report indicating an impact of self-occlusion caused by a user's posture operating the UE on a transmit beam of a base station and based on beam self-occlusion information predicted by the UE using an artificial intelligence (AI) model; Based on the self-obstruction status report, a plurality of transmit beams are determined for beam training between the base station and the UE. A method comprising: Preparing a training data set including input data and output data, wherein the input data includes received signal power of a user equipment (UE) associated with multiple postures of a user operating the UE, and the output data includes beam self-occlusion information of a beam set of the UE associated with the multiple postures; and An artificial intelligence (AI) model is trained on the training set to determine parameters of the AI model. An electronic device, comprising: processor; and a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising: receiving, from a base station, an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction; activating the AI model according to the activation cycle; and Based on the measurement of the beam management reference signal sent by the base station according to the activation period, beam prediction is performed using the activated AI model. The electronic device of claim 21, wherein the information about the activation period of the AI model is included in radio control resource (RRC) signaling or dynamic control signaling. The electronic device according to claim 21, wherein the operations further comprise: Determining a minimum activation period of the AI model based on the capabilities of the UE and parameters of the AI model; and The minimum activation period is reported to a base station, wherein the activation period is not less than the minimum activation period. The electronic device according to claim 23, wherein the minimum activation period is reported during a model identification phase of the AI model. The electronic device according to claim 21, wherein the operations further comprise: An uplink reference signal or a measurement of a downlink reference signal is sent to a base station, wherein the activation period is determined by the base station based on the measurement of the uplink reference signal or the downlink reference signal. The electronic device of claim 21, wherein the AI model is configured to perform one of the following: Beam prediction in the airspace; Beam prediction in the time domain; Beam prediction in spatial and temporal domains. The electronic device according to claim 21, wherein the operations further comprise: receiving an activation command for the AI model from a base station; and According to the activation command, the AI model is activated for beam prediction. The electronic device according to claim 27, wherein the activation command indicates the start of model monitoring of the AI model and a model monitoring period, the model monitoring period being greater than the activation period; or The activation command indicates the start of the performance test of the AI model. The electronic device according to claim 21, wherein the operations further comprise: monitoring the beam signal quality of the UE; and In response to monitoring a beam failure, the AI model is activated for beam prediction, wherein the activated AI model uses a measurement of a beam signal of a synchronization signal block (SSB) as input. The electronic device according to claim 21, wherein the operations further comprise: monitoring a radio link quality of the UE; and In response to monitoring that the radio link quality of the UE is lower than a predetermined threshold, activating the AI model for beam prediction, wherein the predetermined threshold is higher than a decision threshold for radio link failure. The electronic device according to claim 21, wherein the operations further comprise: The prediction result is reported to the base station only when the prediction result of the AI model in the current activation cycle is inconsistent with the prediction result of the previous activation cycle. An electronic device, comprising: processor; and a memory including computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising: determining an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction; sending the determined activation period to the UE; and According to the activation period, a beam management reference signal is sent for the AI model to perform beam prediction. The electronic device of claim 32, wherein the information about the activation period of the AI model is included in radio control resource (RRC) signaling or dynamic control signaling. The electronic device of claim 32, wherein the operations further comprise: receiving information about a minimum activation period of the AI model from the UE; and The activation period is determined based on the minimum activation period, wherein the activation period is not less than the minimum activation period. The electronic device of claim 32, wherein the minimum activation period is received during an identification phase of the AI model. The electronic device of claim 32, wherein the operations further comprise: receiving an uplink reference signal or a measurement of a downlink reference signal from the UE; and The activation period is determined based on measurement of an uplink reference signal or a downlink reference signal. The electronic device of claim 32, wherein the AI model is configured to perform one of the following: Beam prediction in the airspace; Beam prediction in the time domain; Beam prediction in spatial and temporal domains. The electronic device of claim 32, wherein the operations further comprise: An activation command regarding the AI model is sent to the UE, wherein the UE activates the AI model for beam prediction in response to the activation command. The electronic device according to claim 32, wherein the activation command indicates the start of model monitoring of the AI model and a model monitoring period, the model monitoring period being greater than the activation period; or The activation command indicates the start of the performance test of the AI model. A method comprising: receiving, from a base station, information about an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction; activating the AI model according to the activation cycle; and Based on the measurement of the beam management reference signal sent by the base station according to the activation period, beam prediction is performed using the activated AI model. A method comprising: determining an activation period of an artificial intelligence (AI) model used by a user equipment (UE) for beam prediction; sending the determined activation period to the UE; and According to the activation period, a beam management reference signal is sent for the AI model to perform beam prediction. A computer program product comprising executable instructions which, when executed, cause an electronic device to perform the method of any one of claims 18-20 and 40-41.