Electronic device and method for user equipment side, electronic device and method for network side, computer-readable storage medium, and computer program product

Abstract

The present disclosure provides an electronic device and a method for a user equipment side, an electronic device and a method for a network side, a computer-readable storage medium, and a computer program product. The electronic device for the user equipment side comprises: at least one processor, and at least one memory comprising computer program code, wherein the at least one memory and the computer program code are configured to, by means of the at least one processor, enable the electronic device to: on the basis of a measurement result obtained by measuring a reference signal used for artificial intelligence model failure detection, determine a performance metric of the currently used first artificial intelligence model; and when the performance metric meets a predetermined condition, determine that an artificial intelligence model failure has occurred, and execute a rollback to the working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model.

Description

Electronic device and method for user equipment side and network side, computer readable storage medium and computer program product

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on February 6, 2024, with application number 202410171390.2 and invention name “Electronic device and method for user equipment side and network side, computer-readable storage medium and computer program product”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of wireless communications, and more specifically, to failure detection and recovery techniques for artificial intelligence models in wireless communications. More specifically, the present application relates to electronic devices and methods for user equipment (UE), electronic devices and methods for network, computer-readable storage media, and computer program products. Background Art

[0003] Using artificial intelligence (AI) to replace traditional physical layer processes to achieve or enhance certain performance aspects is a hot topic within 3GPP RAN1. Existing use cases include AI-based beam management, AI-based positioning, and AI-based channel state information (CSI) enhancement. Discussions focus on the following model aspects: model training, model inference, and model monitoring. Summary of the Invention

[0004] A brief overview of the present disclosure is provided below to provide a basic understanding of certain aspects of the present disclosure. 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 aspects of the present disclosure, nor is it intended to limit the scope of the present disclosure. Its purpose is simply to present certain concepts in a simplified form as a prelude to the more detailed description discussed later.

[0005] According to one aspect of the present disclosure, an electronic device for a user device side is provided, comprising: at least one processor; and at least one memory including computer program code, wherein the at least one memory and the computer program code are configured to enable the electronic device to execute, through the at least one processor: determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring a reference signal for artificial intelligence model failure detection; and determining that an artificial intelligence model failure has occurred, when the performance metric meets a predetermined condition, and executing a fallback to a working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model.

[0006] According to another aspect of the present disclosure, an electronic device for a network side is provided, comprising: at least one processor; and at least one memory, comprising computer program code, wherein the at least one memory and the computer program code are configured to enable the electronic device to execute, through the at least one processor: generate a configuration of a reference signal for artificial intelligence model failure detection, and provide the configuration to a user device, wherein the user device determines a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring the reference signal; and when the performance metric meets a predetermined condition, determines that an artificial intelligence model failure has occurred, and the user device executes a fallback to a working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model.

[0007] According to one aspect of the present disclosure, a method for an electronic device on a user equipment side is provided, comprising: determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring a reference signal used for artificial intelligence model failure detection; and determining that an artificial intelligence model failure has occurred if the performance metric meets a predetermined condition, and executing a fallback to a non-artificial intelligence model operating mechanism or a switch from the first artificial intelligence model to a second artificial intelligence model.

[0008] According to another aspect of the present disclosure, a method for an electronic device on a network side is provided, including: generating a configuration of a reference signal for artificial intelligence model failure detection, and providing the configuration to a user device, the user device determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring the reference signal; and when the performance metric meets a predetermined condition, determining that an artificial intelligence model failure has occurred, and the user device executing a fallback to a working mechanism of a non-artificial intelligence model or switching from the first artificial intelligence model to a second artificial intelligence model.

[0009] According to other aspects of the present disclosure, a computer program code and a computer program product for implementing the above method, as well as a computer-readable storage medium having the computer program code for implementing the above method recorded thereon, are also provided.

[0010] The electronic device and method according to the embodiments of the present application can detect and recover from failure of an artificial intelligence (AI) model by measuring a reference signal.

[0011] These and other advantages of the present disclosure will become more apparent through the following detailed description of the preferred embodiments of the present disclosure in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to further illustrate the above and other advantages and features of the present disclosure, the following is a further detailed description of the specific embodiments of the present disclosure in conjunction with the accompanying drawings. The drawings, together with the detailed description below, are included in this specification and form a part of this specification. Elements with the same function and structure are represented by the same reference numerals. It should be understood that these drawings only depict typical examples of the present disclosure and should not be regarded as limiting the scope of the present disclosure. In the drawings:

[0013] FIG1 is a block diagram showing functional modules of an electronic device on a user equipment side according to an embodiment of the present application;

[0014] FIG2 is a block diagram showing functional modules of an electronic device on a user equipment side according to an embodiment of the present application;

[0015] FIG3 is a schematic diagram showing an information flow when a UE feeds back an ACK;

[0016] FIG4 is a schematic diagram showing an information flow when a UE feeds back a NACK;

[0017] FIG5 is a block diagram showing functional modules of an electronic device for a network side according to another embodiment of the present application;

[0018] FIG6 shows a flowchart of a method for an electronic device on a user equipment side according to an embodiment of the present application;

[0019] FIG7 shows a flowchart of a method for an electronic device on a network side according to another embodiment of the present application;

[0020] FIG8 is a block diagram illustrating a first example of a schematic configuration of an eNB or gNB to which the technology of the present disclosure may be applied;

[0021] FIG9 is a block diagram illustrating a second example of a schematic configuration of an eNB or gNB to which the technology of the present disclosure may be applied;

[0022] FIG10 is a block diagram showing an example of a schematic configuration of a smartphone to which the technology of the present disclosure can be applied;

[0023] FIG11 is a block diagram showing an example of a schematic configuration of a car navigation device to which the technology of the present disclosure can be applied; and

[0024] FIG12 is a block diagram of an exemplary structure of a general-purpose personal computer in which the method and / or apparatus and / or system according to the embodiments of the present disclosure may be implemented. DETAILED DESCRIPTION

[0025] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings. For the sake of clarity and conciseness, not all features of an actual implementation are described in this specification. However, it should be understood that in the process of developing any such actual implementation, many implementation-specific decisions must be made in order to achieve the developer's specific goals, such as compliance with system and business-related constraints, which may vary from implementation to implementation. In addition, it should be understood that although the development work may be very complex and time-consuming, it is a routine task for those skilled in the art who benefit from the contents of this disclosure.

[0026] It is also necessary to explain here that, in order to avoid obscuring the present disclosure due to unnecessary details, the accompanying drawings only show the device structure and / or processing steps that are closely related to the solution according to the present disclosure, while other details that are not closely related to the present disclosure are omitted.

[0027] <First embodiment>

[0028] As mentioned above, for example, beam management can be performed based on an AI model, and the AI ​​model can be deployed on at least one of the user equipment (UE) and the network (NW) side. The AI ​​model needs to be monitored during use to ensure its normal operation. This embodiment provides a technology for failure detection and recovery of the AI ​​model running on the UE side. It should be understood that although this embodiment is described for the scenario where AI is applied to beam management, it is not limited thereto, but can be appropriately or partially applied to other scenarios.

[0029] Figure 1 shows a functional module block diagram of an electronic device 100 for the user equipment side according to this embodiment. As shown in Figure 1, the electronic device 100 includes: a first determination unit 101, configured to determine the performance metric of the currently used first AI model based on a measurement result obtained by measuring a reference signal for AI model failure detection; and a second determination unit 102, configured to determine that an AI model failure has occurred when the performance metric meets a predetermined condition, and perform a fallback to a non-AI model working mechanism or a switch from the first AI model to the second AI model.

[0030] The first determination unit 101 and the second determination unit 102 can be implemented by one or more processing circuits and at least one memory. The processing circuit can be implemented as a chip, a processor, etc., and the at least one memory can be any form of storage device such as RAM, ROM, flash memory, etc. The at least one memory is used to store computer program code and data required for the processing circuit to perform processing. It should be understood that the various functional units in the electronic device shown in Figure 1 are merely logical modules divided according to the specific functions they implement, and are not intended to limit specific implementation methods. In addition, the above description is also applicable to the other functional units mentioned later and will not be repeated here.

[0031] The electronic device 100 can be set on the UE side. It should also be noted that the electronic device 100 can be implemented at the chip level, or it can also be implemented at the device level. For example, the electronic device 100 can work as the UE itself, and can also include external devices such as memory, transceiver (not shown in the figure), etc. The memory can be used to store programs and related data information that the UE needs to execute to implement various functions. The transceiver may include one or more communication interfaces to support communication with different devices (for example, other UEs, base stations, etc.), and the implementation form of the transceiver is not specifically limited here.

[0032] The ordinal numbers such as first and second in the present disclosure are only for the purpose of distinction and do not represent any meaning of order in time, space or priority unless otherwise specified.

[0033] AI models may include but are not limited to CNN, RNN, LSTM, etc., and their structure may include, for example, an input layer, multiple hidden layers and an output layer. Different types of neural network layers such as convolutional layers and recurrent layers may be used in the hidden layers.

[0034] For example, the reference signal used for AI model failure detection may be a channel state information reference signal (CSI-RS) or a synchronization signal block (SSB). These reference signals may be configured by the network side. Accordingly, as shown in FIG2 , the electronic device 100 may further include a communication unit 103 configured to receive a configuration for a reference signal from the network side. For example, in the case where the AI ​​model is used for beam management, the configuration of the reference signal may be performed by adding AImodelFailure for artificial intelligence model failure detection in the purpose field of RadioLinkMonitoringRS.

[0035] Among them, RadioLinkMonitoringRS is an existing information element (IE) for configuring a reference signal for link failure detection and beam failure detection, as shown below. Its specific use is determined by purpose.

[0036] In this embodiment, the configuration of the reference signal for AI model failure detection can reuse this IE, for example, by adding AImodelFailure in the purpose field. The reference signal for AI model failure detection can be periodic, aperiodic, or semi-persistent.

[0037] The UE measures the reference signal according to the configuration of the reference signal obtained from the network side. The measurement results include but are not limited to signal strength, channel state related information, positioning feature related information, etc. The first determination unit 101 determines the performance metric (Performance Metric) of the currently used AI model (i.e., the first AI model) based on the measurement results. The performance metric may include one or more of the following: the prediction accuracy of the AI ​​model, the difference between the predicted value and the true value of the AI ​​model, the input and output data distribution characteristics, and the link quality.

[0038] For example, in AI-based beam management, the prediction result of the AI ​​model can be the reference signal received power (RSRP) and identification (ID) of a group of candidate beams. For example, according to the prediction of the AI ​​model, the group of candidate beams is the optimal beam. The network side (such as the base station, gNB) sends the corresponding beam according to the ID of the group of candidate beams. The UE measures the RSRP of the beam sent by the network side and calculates the difference between the RSRP of the measured beam and the RSRP predicted by the AI ​​model. The difference can be used as a performance metric. For example, the smaller the difference, the better the performance of the AI ​​model. It should be understood that the difference is only given here as an example, but the present application is not limited to this.

[0039] Therefore, the performance metric reflects the performance of the AI ​​model, and thus whether the AI ​​model fails can be detected based on the performance metric.

[0040] In the first example, the UE side monitors the AI ​​model, that is, the UE determines whether an AI model failure has occurred. For example, the second determination unit 102 determines whether the performance metric meets a predetermined condition, and determines that an AI model failure has occurred if the predetermined condition is met. The second determination unit 102 can be configured to: start a timer and a counter when an AI model failure event is detected for the first time, and add 1 to the counter each time an AI model failure event is subsequently detected, wherein the detection of an AI model failure event includes a performance metric being lower than a first threshold; and before the timer expires, if the count value of the counter is greater than a predetermined maximum count, determine that the performance metric meets the predetermined condition and determine that an AI model failure has occurred.

[0041] The timer and counter can be defined in higher layers of the UE, such as the MAC layer. When the MAC layer receives an indication of an AI model failure event, the timer starts timing and the counter starts counting. If enough AI model failure events are received before the timer expires (the count value is greater than a predetermined maximum count), an AI model failure is triggered.

[0042] The second determination unit 102 is further configured to, after determining that an AI model failure has occurred, execute a fallback to a working mechanism of a non-AI model, and measure a performance metric of a candidate AI model in the candidate AI model list to determine a second AI model, wherein the performance metric of the second AI model is higher than a second threshold.

[0043] That is, after determining that an AI model failure has occurred, the second determination unit 102 performs fallback and switching to the second AI model. Here, fallback refers to falling back to a traditional working mechanism that does not adopt an AI model. For example, fallback includes the UE re-establishing a connection with the network side through a random access process. Specifically, the UE falls back to the random access state and reports the candidate beam to re-establish the connection. After the fallback, the UE can also switch the AI ​​model, that is, find a second AI model suitable for switching to and switch to the second AI model. For example, the UE can evaluate the AI ​​model in the candidate AI model list to determine whether it can be used as the second AI model. The candidate AI model list can, for example, be included in the field reported by the UE capability. In other words, the candidate AI models in the candidate AI model list are known to both the UE and the network side.

[0044] If the candidate AI model list is not empty, the second determination unit 102 measures the candidate AI models and finds an AI model whose performance metric meets the requirements (higher than a second threshold) as the second AI model. The UE can deactivate the current first AI model and switch to the second AI model. For example, the communication unit 103 can notify the network side of the second AI model being switched to via uplink control information (UCI), such as the model ID. Alternatively, the UE can also notify the network side of the information of the second AI model before performing the switch to seek confirmation from the network side.

[0045] When the list of candidate AI models is empty, there are two possibilities for the AI ​​model on the UE side: one is that there is no candidate AI model; the other is that there is a candidate AI model but it is in a transparent state, that is, the network side is unaware of the existence of the candidate AI model. The AI ​​model that is transparent to the network side may belong to one of the following situations: in the function-based identification, there are multiple AI models under the same function; AI models that are updated or fine-tuned online in real time according to scene changes; unidentified and unactivated AI models. For these AI models, the network side cannot give clear instructions such as model switching or selection during the management process, nor can it clearly determine which of the two possibilities the candidate AI model belongs to, so the UE side needs to feedback confirmation information.

[0046] For example, the communication unit 103 is further configured to: receive an indication regarding model switching from the network via a connection; in response to the indication, send an ACK to the network if a second AI model transparent to the network exists, and send a NACK to the network if such a second AI model does not exist; if the second AI model exists, send a switching success report to the network after successfully switching to the second AI model. On the other hand, if the second AI model does not exist, the second determination unit 102 causes the UE to remain in the non-AI model operating mechanism. Specifically, after the UE and the network reestablish a connection, the network needs to inquire whether to switch models. In other words, the network awaits feedback from the UE before making a decision. An ACK from the UE indicates that a candidate AI model transparent to the network exists on the UE. After evaluating the performance of the candidate AI models and finding a second AI model that meets the requirements, the UE switches to the second AI model and notifies the network of the successful switch. A NACK from the UE indicates that no candidate AI model exists on the UE that could be a possible switching target. Furthermore, the communication unit 103 is further configured to send a request for additional conditions to the network after the UE successfully switches to the second AI model. For example, additional conditions include conditions that need to be considered during use, such as scenarios, locations, and data sets, in addition to the parameters of the second AI model itself. The additional conditions here include, for example, network-side beam patterns. The request for additional conditions can be sent separately or together with the handover success report, which is not restrictive. For ease of understanding, Figures 3 and 4 show schematic diagrams of the information flow when the UE feeds back ACK and the UE feeds back NACK, respectively.

[0047] The above describes an example of performing fallback before model switching when an AI model failure occurs, but the present invention is not limited to this. For example, it is also possible to measure the candidate AI model and determine the second AI model before determining that an AI model failure has occurred, so that the UE can directly switch to the second AI model when an AI model failure occurs. In this case, the UE needs to have the ability to simultaneously run or assist the network side in running two or more AI models that implement the same function.

[0048] For example, the second determining unit 102 is configured to: upon starting a timer, begin measuring the performance metrics of candidate AI models in the candidate AI model list to determine a second AI model, wherein the performance metrics of the second AI model are greater than a second threshold; and upon determining that an AI model failure has occurred, perform a switch to the second AI model. Similarly, the candidate AI model list may be included in a field reported by the UE, for example.

[0049] Since there is no fallback process in this case, in order to report to the network the second AI model to which the UE has switched, the communication unit 103 can be configured to notify the network of the second AI model to which it has switched through a beam failure recovery (BFR) process during the random access process. In addition, a mapping relationship can be set between the random access preamble sequence and the identifier of the AI ​​model, so that the network implicitly determines the second AI model.

[0050] During the above-described AI model failure detection and recovery process, various parameters are required. For example, the communication unit 103 may be further configured to receive a configuration for AI model failure detection and recovery from the network. The configuration may include, for example, one or more of the following: the number of events required to trigger an AI model failure, a specified time duration for counting AI model failure events, a first threshold for estimating the performance level of a first AI model, a list of candidate AI models, and a second threshold for estimating the performance level of a second AI model.

[0051] Specifically, the following new IEs can be defined for AI model failure detection and recovery: AImodelFailureInstanceMaxCount, the number of events required to trigger AI model failure. For example, when AImodelFailureInstanceMaxCount = 2, if two AI model failure events are detected, it is considered that an AI model failure has occurred; AImodelFailureInstanceMaxTimer, the specified time length for counting AI model failure events, is used to limit the expiration time of the timer; PerformanceMetric-Threshold, a first threshold for estimating the performance level of the first AI model, that is, the threshold of the performance metric of the current AI model. If the performance metric is lower than the threshold, it is considered that an AI model failure event has occurred; candidate AI model list candidateAImodelRSList, which records the selectable candidate AI models that can be selected and switched; PerformanceMetric-ThresholdMFR, a second threshold for estimating the performance level of the second AI model, that is, the threshold of the performance metric of the AI ​​model (second AI model) that continues to perform beam management next.

[0052] In the second example, the network side monitors the AI ​​model, that is, the network side determines whether an AI model failure occurs.

[0053] The communication unit 103 may be configured to provide a performance indication based on the performance metric of the first AI model to the network side, so that the network side determines whether an AI model failure has occurred based on the performance indication. The second determination unit 102 determines that an AI model failure has occurred based on the indication from the network side. The performance indication here may include one of the following: a performance metric, and an indication of whether the performance metric is above a first threshold.

[0054] As described in detail in the second embodiment below, a timer and counter are set on the network side to count the frequency of AI model failure events, thereby determining whether an AI model failure has occurred. When the network side determines that an AI model failure has occurred, the network side issues an indication to the UE. Based on this indication, the UE determines that an AI model failure has occurred and executes a fallback to a non-AI model operating mechanism and switches from the first AI model to the second AI model.

[0055] For example, the fallback includes the UE re-establishing a connection with the network side through a random access process. After the fallback, the UE switches the AI ​​model, that is, searches for a second AI model suitable for switching to and switches to the second AI model.

[0056] If the candidate AI model list is not empty, the second determination unit 102 measures the candidate AI models and finds an AI model whose performance metric meets the requirements (higher than a second threshold) as the second AI model. The UE can deactivate the current first AI model and switch to the second AI model. For example, the communication unit 103 can notify the network side of the switched second AI model via UCI, such as notifying information such as the model ID. Alternatively, the UE can also notify the network side of information about the second AI model before performing the switch to seek confirmation from the network side.

[0057] When the candidate AI model list is empty, there are two possibilities for the AI ​​model on the UE side: one is that there is no candidate AI model; the other is that there is a candidate AI model that is transparent to the network side. In this case, as described with reference to Figures 3 and 4, the communication unit 103 can also be configured to: receive an indication of model switching from the network side via a connection; in response to the indication, send an ACK to the network side if there is a second AI model that is transparent to the network side, and send a NACK to the network side if there is no such second AI model; if there is a second AI model, send a switching success report to the network side after successfully switching to the second AI model. On the other hand, if there is no second AI model, the second determination unit 102 keeps the UE under the working mechanism of the non-AI model. In addition, the communication unit 103 is further configured to send a request for additional conditions to the network side after the UE successfully switches to the second AI model, where the additional conditions include, for example, the network side beam pattern. The relevant specific details have been given in the description of the first example and will not be repeated here.

[0058] In addition, the network side can also be configured to issue an indication to the UE when the timer starts, so that the UE can measure the performance metrics of the candidate AI model before the AI ​​model failure occurs to determine the second AI model. In this way, when an indication of the failure of the AI ​​model is received from the network side, the switching can be performed immediately.

[0059] For example, the second determination unit 102 is configured to: upon receiving an indication of timer start from the network side, start measuring the performance metrics of the candidate AI models in the candidate AI model list to determine the second AI model, wherein the performance metric of the second AI model is higher than the second threshold; and after determining that an AI model failure occurs, execute switching to the second AI model.

[0060] Since there is no fallback process in this case, in order to report to the network side the second AI model to which the UE switches, the communication unit 103 can be configured to notify the network side of the second AI model switched to through the beam failure recovery (BFR) process in the random access process. In addition, it can also be set that there is a mapping relationship between the random access preamble sequence and the identifier of the AI ​​model, so that the network side implicitly determines the second AI model. The relevant specific details have been given in the description of the first example and will not be repeated here.

[0061] In summary, the electronic device 100 according to this embodiment can detect and recover from failure of the AI ​​model by measuring the reference signal.

[0062] <Second embodiment>

[0063] Figure 5 shows a functional module block diagram of an electronic device 200 for the network side according to this embodiment. As shown in Figure 5, the electronic device 200 includes: a generation unit 201, configured to generate a configuration of a reference signal for AI model failure detection; a communication unit 202, configured to provide the configuration to the UE, and the UE determines the performance metric of the currently used first AI model based on the measurement result obtained by measuring the reference signal; and a determination unit 203, configured to determine that an AI model failure has occurred when the performance metric meets a predetermined condition, and the UE performs a fallback to a non-AI model working mechanism or switches from the first AI model to the second AI model.

[0064] The generation unit 201, communication unit 202, and determination unit 203 may be implemented by one or more processing circuits and at least one memory. The processing circuit may be implemented as a chip, a processor, or the like, and the at least one memory may be any form of storage device, such as RAM, ROM, or flash memory. The at least one memory is used, for example, to store computer program code and data required for the processing circuit to perform processing. Furthermore, it should be understood that the various functional units in the electronic device shown in FIG5 are merely logical modules divided according to the specific functions they implement, and are not intended to limit specific implementations.

[0065] The electronic device 200 can be set on the network side, such as the base station side. It should also be noted that the electronic device 200 can be implemented at the chip level, or it can also be implemented at the device level. For example, the electronic device 200 can work as the base station itself, and can also include external devices such as memory and transceiver (not shown in the figure). The memory can be used to store programs and related data information that need to be executed by the base station to implement various functions. The transceiver may include one or more communication interfaces to support communication with different devices (e.g., UE, other base stations, etc.), and the implementation form of the transceiver is not specifically limited here.

[0066] In this embodiment, the electronic device 200 is located on the network side, and the UE executes the AI ​​model. As described in the first embodiment, the reference signal used for AI model failure detection can be CSI-RS or SSB. When the AI ​​model is used for beam management, the configuration of the reference signal can be performed by adding AImodelFailure for AI model failure detection in the purpose field of RadioLinkMonitoringRS. The relevant detailed description has been given in the first embodiment and will not be repeated here.

[0067] The UE measures the reference signal based on the above configuration and determines a performance metric of the first AI model (current AI model) based on the measurement result. The UE can determine whether an AI model failure has occurred based on the performance metric, as described in the first embodiment. The network side can determine that an AI model failure has occurred based on the UE's indication, and the UE performs a fallback or AI model switching.

[0068] In addition, the network side may also determine whether an AI model failure has occurred. The communication unit 202 may also be configured to obtain a performance indicator based on a performance metric from the UE, and the determination unit 203 may determine whether an AI model failure has occurred based on the performance indicator.

[0069] The performance indication may include one of the following: a performance metric, and whether the performance metric is above a first threshold.

[0070] The determining unit 203 may be configured to: start a timer and a counter when the performance indicator indicates that the performance metric is below a first threshold, and increment the counter by 1 each time the performance metric subsequently falls below the first threshold; and if the count value of the counter is greater than a predetermined maximum count before the timer expires, determine that the performance metric satisfies a predetermined condition and determine that an AI model failure has occurred. In other words, the number of occurrences of the AI ​​model failure event (the performance metric falls below the first threshold) is counted.

[0071] Alternatively, determining unit 203 may be configured to: start a timer and a counter when the performance indicator indicates that the performance metric is below a first threshold, and increment the counter by 1 each subsequent time the performance metric exceeds the first threshold; and if the counter value is less than a predetermined value before the timer expires, determine that the performance metric satisfies a predetermined condition and determine that an AI model failure has occurred. That is, the number of times the performance metric exceeds the first threshold is counted. This is because, when the AI ​​model is used for beam management, a deterioration in the performance metric indicates poor quality of the currently serving beam, which may affect the UE's reporting of a performance metric below the first threshold. To avoid this risk, the performance metric may be reported when it exceeds the first threshold, and a counter may count the number of reports. When the count value reaches the predetermined value but the timer has not expired, the timer may be stopped, determining that no AI model failure has occurred. If the count value does not reach the predetermined value when the timer expires, it indicates that the overall performance metric is low, and an AI model failure is determined to have occurred.

[0072] After the determination unit 203 determines that an AI model failure has occurred, the communication unit 202 may notify the UE of the AI ​​model failure. As previously mentioned, the UE may perform a fallback, which includes the UE reestablishing a connection with the network side through a random access procedure.

[0073] After fallback, the UE can also switch the AI ​​model, that is, find a suitable second AI model to switch to and switch. For example, the UE can measure the performance metrics of the candidate AI models in the candidate AI model list to determine the second AI model. The candidate AI model list can be included in the field reported by the UE capability. In other words, the candidate AI models in the candidate AI model list are known to both the UE and the network.

[0074] If the candidate AI model list is not empty, the UE measures the candidate AI models therein and finds an AI model whose performance metric meets the requirements (higher than the second threshold) as the second AI model. The UE can deactivate the current first AI model and switch to the second AI model. For example, since a connection has been established between the UE and the network side at this time, the communication unit 202 can obtain information about the second AI model to be switched to, such as the model ID, from the UE through uplink control information (UCI). Alternatively, the UE can also notify the network side of the information of the second AI model before performing the switch to seek confirmation from the network side.

[0075] When the candidate AI model list is empty, there are two possible AI models on the UE side: one is that there is no candidate AI model; the other is that a candidate AI model exists but is transparent to the network side, that is, the network side is unaware of the existence of the candidate AI model. For these AI models, the network side cannot provide clear instructions during the management process, such as model switching or selection, nor can it clearly determine which of the two possibilities the candidate AI model belongs to. Therefore, the UE side needs to feedback confirmation information.

[0076] For example, the communication unit 202 is further configured to: send an indication regarding model switching to the UE via a connection; and receive an ACK or NACK sent by the UE in response to the indication, wherein the ACK indicates the existence of a second AI model that is transparent to the network side, i.e., an AI model that meets the requirements exists as the second AI model among the candidate AI models that are transparent to the network, and the NACK indicates the absence of the second AI model. The communication unit 202 is further configured to, upon receiving the ACK, further receive a handover success report from the UE, indicating that the UE has successfully switched to the second AI model; and upon receiving the NACK, to learn that the UE remains under the non-AI model operating mechanism. In addition, upon receiving the ACK, the communication unit 202 is further configured to receive a request for additional conditions from the UE. The additional conditions include, for example, the network-side beam pattern. The request for additional conditions can be sent separately or together with the handover success report, which is not restrictive. The specific process has been described with reference to Figures 3 and 4 in the first embodiment and will not be repeated here.

[0077] In addition, corresponding to the first embodiment, the UE can determine the second AI model before determining that an AI model failure has occurred. When the timer and counter are defined on the UE side, for example, when the timer is started, the performance metric of the candidate AI model can be measured to determine a second AI model that meets the requirements (the performance metric of the second AI model is higher than the second threshold). When the timer and counter are defined on the network side, the communication unit 202 can send an indication to the UE when the timer is started, so that the UE measures the performance metric of the candidate AI model in response to the indication, thereby determining the second AI model.

[0078] In this case, when it is determined that an AI model failure has occurred, the UE can directly switch to the second AI model without first performing a fallback. Since the high-quality link connection is not restored through the fallback process, the UE can report the information of the second AI model to the network side through the random access process. For example, the communication unit 202 can obtain information about the second AI model switched to from the UE through the BFR process in the random access process. For example, there is a mapping relationship between the random access preamble sequence and the identifier of the AI ​​model, and the determination unit 203 can implicitly determine the second AI model based on this mapping relationship.

[0079] In addition, the communication unit 202 may also be configured to send a configuration for AI model failure detection and recovery to the UE. The configuration may include, for example, one or more of the following: the number of events required to trigger an AI model failure, a specified event length for counting AI model failure events, a first threshold for estimating the performance level of a first AI model, a list of candidate AI models, and a second threshold for estimating the performance level of a second AI model. The relevant details have been provided in the first embodiment and will not be repeated here.

[0080] In summary, the electronic device 200 according to this embodiment can configure and assist the UE to detect and recover from AI model failure by measuring the reference signal.

[0081] <Third embodiment>

[0082] In the process of describing the electronic device for the user device side and the electronic device for the network side in the above embodiments, it is obvious that some processing or methods are also disclosed. Below, an overview of these methods is given without repeating some of the details discussed above, but it should be noted that although these methods are disclosed in the process of describing the electronic device for the user device side and the electronic device for the network side, these methods do not necessarily adopt the components described or are not necessarily performed by those components. For example, the embodiments of the electronic device for the user device side and the electronic device for the network side can be partially or completely implemented using hardware and / or firmware, and the methods discussed below can be completely implemented by a computer-executable program, although these methods can also adopt the hardware and / or firmware of the electronic device for the user device side and the electronic device for the network side.

[0083] Figure 6 shows a flowchart of a method for an electronic device on the user device side according to an embodiment of the present application, the method including: determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring a reference signal for artificial intelligence model failure detection (S11); and when the performance metric meets a predetermined condition, determining that an artificial intelligence model failure has occurred, and performing a fallback to a working mechanism of a non-artificial intelligence model or switching from the first artificial intelligence model to the second artificial intelligence model (S12).

[0084] Performance metrics include, for example, one or more of the following: prediction accuracy of the artificial intelligence model, the difference between the predicted value of the artificial intelligence model and the true value, input and output data distribution characteristics, and link quality.

[0085] For example, step S12 may include: starting a timer and a counter when an artificial intelligence model failure event is detected for the first time, and adding 1 to the counter each time an artificial intelligence model failure event is subsequently detected, wherein the detection of an artificial intelligence model failure event includes a performance metric being lower than a first threshold; and before the timer expires, if the count value of the counter is greater than a predetermined maximum count, judging that the performance metric meets a predetermined condition and determining that an artificial intelligence model failure has occurred.

[0086] Step S12 may also include: after determining that an artificial intelligence model failure has occurred, executing a fallback to a working mechanism of a non-artificial intelligence model; and measuring the performance metric of the candidate artificial intelligence models in the candidate artificial intelligence model list to determine a second artificial intelligence model, wherein the performance metric of the second artificial intelligence model is higher than a second threshold. The candidate artificial intelligence model list may be included in the field reported by the user device capability.

[0087] Fallback includes, for example, the UE re-establishing a connection with the network side through a random access procedure.

[0088] The above method also includes: notifying the network side of the second artificial intelligence model switched to through uplink control information.

[0089] Alternatively, the above method may also include: receiving an indication about model switching from the network side via the above connection; in response to the indication, sending ACK to the network side if there is the second artificial intelligence model that is transparent to the network side, and sending NACK to the network side if there is no second artificial intelligence model; if there is a second artificial intelligence model, sending a switching success report to the network side after successfully switching to the second artificial intelligence model; and if there is no second artificial intelligence model, keeping the UE under the working mechanism of the non-artificial intelligence model.

[0090] After successfully switching to the second artificial intelligence model, a request for additional conditions may be sent to the network side. The additional conditions may include, for example, a network-side beam pattern.

[0091] As another example, step S12 may include: when the timer starts, starting to measure the performance metrics of the candidate artificial intelligence models in the candidate artificial intelligence model list to determine a second artificial intelligence model, wherein the performance metric of the second artificial intelligence model is higher than a second threshold; and after determining that an artificial intelligence model failure occurs, executing a switch to the second artificial intelligence model.

[0092] For example, the network side can be notified of the second artificial intelligence model switched to through the beam failure recovery process during the random access process. In the case where there is a mapping relationship between the random access preamble sequence and the identifier of the artificial intelligence model, the network side can implicitly determine the second artificial intelligence model.

[0093] In addition, step S12 may further include: providing a performance indication based on the performance metric of the first artificial intelligence model to the network side, so that the network side determines whether an artificial intelligence model failure has occurred based on the performance indication; and determining that an artificial intelligence model failure has occurred based on the indication from the network side. The performance indication includes, for example, one of the following: the performance metric, and an indication of whether the performance metric is above a first threshold.

[0094] Although not shown in the figure, the method may further include: receiving a configuration for the reference signal from the network side, and when the first artificial intelligence model and the second artificial intelligence model are used for beam management, the configuration of the reference signal is performed by adding AImodelFailure for artificial intelligence model failure detection in the purpose field of RadioLinkMonitoringRS.

[0095] The method may further include: receiving a configuration for artificial intelligence model failure detection and recovery from a network side, wherein the configuration includes one or more of the following: the number of events required to trigger artificial intelligence model failure, a specified time length for counting artificial intelligence model failure events, a first threshold for evaluating the performance level of the first artificial intelligence model, a list of candidate artificial intelligence models, and a second threshold for evaluating the performance level of the second artificial intelligence model.

[0096] The above method corresponds to the electronic device 100 of the first embodiment. The relevant specific details can be referred to the first embodiment and will not be repeated here.

[0097] Figure 7 shows a flowchart of a method for an electronic device on the network side according to an embodiment of the present application, the method including: generating a configuration of a reference signal for artificial intelligence model failure detection (S21); providing the configuration to a user device, the user device determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring the reference signal (S22); and when the performance metric meets a predetermined condition, determining that an artificial intelligence model failure has occurred, and the user device executing a fallback to a working mechanism of a non-artificial intelligence model or switching from the first artificial intelligence model to a second artificial intelligence model (S23).

[0098] For example, the first artificial intelligence model and the second artificial intelligence model are used for beam management, and the configuration of the reference signal is performed by adding AImodelFailure for artificial intelligence model failure detection in the purpose field of RadioLinkMonitoringRS.

[0099] Step S23 may include: obtaining a performance indication based on a performance metric from the UE; and determining whether an artificial intelligence model failure has occurred based on the performance indication. The performance indication may include one of the following: a performance metric, and an indication of whether the performance metric is above a first threshold.

[0100] For example, when the performance indicator indicates that the performance metric is lower than the first threshold, the timer and counter are started, and the counter is incremented by 1 each subsequent time the performance metric is lower than the first threshold; and before the timer expires, if the count value of the counter is greater than a predetermined maximum count, it is judged that the performance metric meets the predetermined condition and it is determined that an artificial intelligence model failure has occurred.

[0101] Alternatively, a timer and a counter are started when the performance indicator indicates that the performance metric is lower than a first threshold, and the counter is incremented by 1 each subsequent time the performance metric is higher than the first threshold; and before the timer expires, if the count value of the counter is less than a predetermined value, it is judged that the performance metric meets the predetermined condition and it is determined that an artificial intelligence model failure has occurred.

[0102] Falling back may include the UE re-establishing a connection with the network side through a random access process. Information about the second artificial intelligence model switched to may be obtained from the UE through uplink control information.

[0103] Alternatively, the method further includes: sending an indication regarding model switching to the UE via the connection; receiving an ACK or NACK sent by the UE in response to the indication, wherein the ACK indicates the presence of a second artificial intelligence model transparent to the network side, and the NACK indicates the absence of such a second artificial intelligence model; upon receiving the ACK, further receiving a switching success report from the UE, the switching success report indicating that the UE successfully switched to the second artificial intelligence model; and upon receiving the NACK, knowing that the UE remains in the non-artificial intelligence model operating mechanism. In addition, upon receiving the ACK, a request for additional conditions is also received from the UE. The additional conditions include, for example, a network-side beam pattern.

[0104] In the case where the AI ​​model switching is performed directly without performing fallback, the above method may further include: obtaining information about the second AI model switched to from the UE through a beam failure recovery process during the random access process. For example, a mapping relationship exists between the random access preamble sequence and the identifier of the AI ​​model, so that the electronic device on the network side can implicitly determine the second AI model.

[0105] The above method corresponds to the electronic device 200 in the second embodiment. The relevant specific details can be referred to the second embodiment and will not be repeated here.

[0106] Note that the above methods can be used in combination or individually.

[0107] The technology of the present disclosure can be applied to various products.

[0108] The electronic device 100 can be implemented as various user devices. The user device 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 device can also be implemented as a terminal that performs machine-to-machine (M2M) communication (also known as a machine type communication (MTC) terminal). In addition, the user device can be a wireless communication module (such as an integrated circuit module including a single chip) installed on each of the above terminals.

[0109] For example, the electronic device 200 can be implemented as various base stations. The base station can be implemented as any type of evolved Node B (eNB) or gNB (5G base station). eNBs include, for example, macro eNBs and small eNBs. Small eNBs can be eNBs that cover cells smaller than macro cells, such as pico eNBs, micro eNBs, and home (femto) eNBs. Similar situations can also be encountered for gNBs. Alternatively, the base station can be implemented as any other type of base station, such as a NodeB and a base transceiver station (BTS). The base station may include: a main body (also referred to as a base station device) configured to control wireless communications; and one or more remote radio heads (RRHs) located at a different location from the main body. In addition, various types of user equipment can operate as a base station by temporarily or semi-permanently performing base station functions.

[0110] [Application examples for base stations]

[0111] (First application example)

[0112] FIG8 is a block diagram illustrating a first example of a schematic configuration of an eNB or gNB to which the techniques of this disclosure can be applied. Note that the following description uses an eNB as an example, but is equally applicable to a gNB. eNB 800 includes one or more antennas 810 and a base station device 820. Base station device 820 and each antenna 810 can be connected to each other via an RF cable.

[0113] Each of the antennas 810 includes a single or multiple antenna elements (such as multiple antenna elements included in a multiple-input multiple-output (MIMO) antenna) and is used for base station device 820 to transmit and receive wireless signals. As shown in FIG8 , eNB 800 may include multiple antennas 810. For example, multiple antennas 810 may be compatible with multiple frequency bands used by eNB 800. Although FIG8 shows an example in which eNB 800 includes multiple antennas 810, eNB 800 may also include a single antenna 810.

[0114] The base station device 820 includes a controller 821 , a memory 822 , a network interface 823 , and a wireless communication interface 825 .

[0115] The controller 821 may be, for example, a CPU or a DSP, and operates various functions of the higher layers of the base station device 820. For example, the controller 821 generates data packets based on the data in the signal processed by the wireless communication interface 825, and transmits the generated packets via the network interface 823. The controller 821 may bundle data from multiple baseband processors to generate bundled packets, and transmit the generated bundled packets. The controller 821 may have logic functions for performing the following controls: the control may be radio resource control, radio bearer control, mobility management, admission control, and scheduling. The control may be performed in conjunction with a nearby eNB or core network node. The memory 822 includes RAM and ROM, and stores programs executed by the controller 821 and various types of control data (such as a terminal list, transmission power data, and scheduling data).

[0116] The network interface 823 is a communication interface for connecting the base station device 820 to the core network 824. The controller 821 can communicate with the core network node or another eNB via the network interface 823. In this case, the eNB 800 and the core network node or other eNBs can be connected to each other through a logical interface (such as an S1 interface and an X2 interface). The network interface 823 can also be a wired communication interface or a wireless communication interface for a wireless backhaul line. If the network interface 823 is a wireless communication interface, the network interface 823 can use a higher frequency band for wireless communication than the frequency band used by the wireless communication interface 825.

[0117] The wireless communication interface 825 supports any cellular communication scheme, such as Long Term Evolution (LTE) and LTE-Advanced, and provides wireless connectivity to terminals located in the cell of the eNB 800 via the antenna 810. The wireless communication interface 825 may typically include, for example, a baseband (BB) processor 826 and RF circuitry 827. The BB processor 826 can perform various signal processing functions, such as encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and performs various types of signal processing for layers such as Layer 1 (L1), Medium Access Control (MAC), Radio Link Control (RLC), and Packet Data Convergence Protocol (PDCP). In place of the controller 821, the BB processor 826 may have some or all of the aforementioned logical functions. The BB processor 826 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 826. This module may be a card or blade inserted into a slot in the base station device 820. Alternatively, the module may be a chip mounted on the card or blade. Meanwhile, the RF circuit 827 may include, for example, a mixer, a filter, and an amplifier, and transmit and receive wireless signals via the antenna 810 .

[0118] As shown in FIG8 , the wireless communication interface 825 may include multiple BB processors 826. For example, multiple BB processors 826 may be compatible with multiple frequency bands used by the eNB 800. As shown in FIG8 , the wireless communication interface 825 may include multiple RF circuits 827. For example, multiple RF circuits 827 may be compatible with multiple antenna elements. Although FIG8 illustrates an example in which the wireless communication interface 825 includes multiple BB processors 826 and multiple RF circuits 827, the wireless communication interface 825 may also include a single BB processor 826 or a single RF circuit 827.

[0119] In the eNB 800 shown in FIG8 , the communication unit 103 and transceiver of the electronic device 100 may be implemented by the wireless communication interface 825. At least a portion of the functionality may also be implemented by the controller 821. For example, the controller 821 may detect and recover from AI model failures by measuring reference signals by executing the functions of the first determination unit 101, the second determination unit 102, and the communication unit 103.

[0120] (Second application example)

[0121] FIG9 is a block diagram illustrating a second example of a schematic configuration of an eNB or gNB to which the techniques of this disclosure can be applied. Note that similarly, the following description uses an eNB as an example, but is equally applicable to a gNB. An eNB 830 includes one or more antennas 840, a base station device 850, and an RRH 860. The RRH 860 and each antenna 840 can be connected to each other via an RF cable. The base station device 850 and the RRH 860 can be connected to each other via a high-speed line such as an optical fiber cable.

[0122] Each of the antennas 840 includes a single or multiple antenna elements (such as multiple antenna elements included in a MIMO antenna) and is used for RRH 860 to transmit and receive wireless signals. As shown in FIG9 , eNB 830 may include multiple antennas 840. For example, multiple antennas 840 may be compatible with multiple frequency bands used by eNB 830. Although FIG9 shows an example in which eNB 830 includes multiple antennas 840, eNB 830 may also include a single antenna 840.

[0123] Base station device 850 includes a controller 851, a memory 852, a network interface 853, a wireless communication interface 855, and a connection interface 857. Controller 851, memory 852, and network interface 853 are the same as controller 821, memory 822, and network interface 823 described with reference to FIG.

[0124] The wireless communication interface 855 supports any cellular communication scheme (such as LTE and LTE-Advanced) and provides wireless communication to terminals located in the sector corresponding to the RRH 860 via the RRH 860 and the antenna 840. The wireless communication interface 855 may generally include, for example, a BB processor 856. The BB processor 856 is the same as the BB processor 826 described with reference to FIG8 , except that the BB processor 856 is connected to the RF circuit 864 of the RRH 860 via the connection interface 857. As shown in FIG9 , the wireless communication interface 855 may include multiple BB processors 856. For example, the multiple BB processors 856 may be compatible with multiple frequency bands used by the eNB 830. Although FIG9 shows an example in which the wireless communication interface 855 includes multiple BB processors 856, the wireless communication interface 855 may also include a single BB processor 856.

[0125] The connection interface 857 is an interface for connecting the base station device 850 (wireless communication interface 855) to the RRH 860. The connection interface 857 may also be a communication module for connecting the base station device 850 (wireless communication interface 855) to the RRH 860 for communication in the high-speed line.

[0126] The RRH 860 includes a connection interface 861 and a wireless communication interface 863 .

[0127] The connection interface 861 is an interface for connecting the RRH 860 (wireless communication interface 863) to the base station device 850. The connection interface 861 may also be a communication module for communication in the above-mentioned high-speed line.

[0128] The wireless communication interface 863 transmits and receives wireless signals via the antenna 840. The wireless communication interface 863 may generally include, for example, an RF circuit 864. The RF circuit 864 may include, for example, a mixer, a filter, and an amplifier, and transmits and receives wireless signals via the antenna 840. As shown in FIG9 , the wireless communication interface 863 may include multiple RF circuits 864. For example, the multiple RF circuits 864 may support multiple antenna elements. Although FIG9 shows an example in which the wireless communication interface 863 includes multiple RF circuits 864, the wireless communication interface 863 may also include a single RF circuit 864.

[0129] In the eNB 830 shown in FIG9 , the communication unit 103 and the transceiver of the electronic device 100 may be implemented by the wireless communication interface 855 and / or the wireless communication interface 863. At least a portion of the functionality may also be implemented by the controller 851. For example, the controller 851 may detect and recover from AI model failures by measuring reference signals by executing the functions of the first determination unit 101, the second determination unit 102, and the communication unit 103.

[0130] [Application examples on user devices]

[0131] (First application example)

[0132] 10 is a block diagram showing an example of a schematic configuration of a smartphone 900 to which the technology of the present disclosure can be applied. The smartphone 900 includes a processor 901, a memory 902, a storage device 903, an external connection interface 904, a camera 906, a sensor 907, a microphone 908, an input device 909, a display device 910, a speaker 911, a wireless communication interface 912, one or more antenna switches 915, one or more antennas 916, a bus 917, a battery 918, and an auxiliary controller 919.

[0133] The processor 901 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 900. The memory 902 includes RAM and ROM, and stores data and programs executed by the processor 901. The storage device 903 may include storage media such as semiconductor memories and hard disks. The external connection interface 904 is an interface for connecting external devices (such as memory cards and universal serial bus (USB) devices) to the smartphone 900.

[0134] The camera 906 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 907 may include a group of sensors such as a measurement sensor, a gyroscope sensor, a geomagnetic sensor, and an acceleration sensor. The microphone 908 converts the sound input to the smartphone 900 into an audio signal. The input device 909 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 910, and receives an operation or information input from the user. The display device 910 includes a screen such as a liquid crystal display (LCD) and an organic light emitting diode (OLED) display and displays an output image of the smartphone 900. The speaker 911 converts the audio signal output from the smartphone 900 into sound.

[0135] The wireless communication interface 912 supports any cellular communication scheme (such as LTE and LTE-Advanced) and performs wireless communications. The wireless communication interface 912 may typically include, for example, a BB processor 913 and an RF circuit 914. The BB processor 913 may perform, for example, encoding / decoding, modulation / demodulation, and multiplexing / demultiplexing, and may also perform various types of signal processing for wireless communications. Meanwhile, the RF circuit 914 may include, for example, mixers, filters, and amplifiers, and transmit and receive wireless signals via an antenna 916. Note that while the figure shows a scenario where one RF link is connected to one antenna, this is merely illustrative, and also encompasses scenarios where one RF link is connected to multiple antennas via multiple phase shifters. The wireless communication interface 912 may be a chip module on which the BB processor 913 and the RF circuit 914 are integrated. As shown in FIG10 , the wireless communication interface 912 may include multiple BB processors 913 and multiple RF circuits 914. While FIG10 illustrates an example in which the wireless communication interface 912 includes multiple BB processors 913 and multiple RF circuits 914, the wireless communication interface 912 may also include a single BB processor 913 or a single RF circuit 914.

[0136] In addition, in addition to the cellular communication scheme, the wireless communication interface 912 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 912 may include a BB processor 913 and an RF circuit 914 for each wireless communication scheme.

[0137] Each of the antenna switches 915 switches a connection destination of the antenna 916 between a plurality of circuits (eg, circuits for different wireless communication schemes) included in the wireless communication interface 912 .

[0138] Each of the antennas 916 includes a single or multiple antenna elements (such as multiple antenna elements included in a MIMO antenna) and is used for transmitting and receiving wireless signals via the wireless communication interface 912. As shown in FIG10 , the smartphone 900 may include multiple antennas 916. Although FIG10 shows an example in which the smartphone 900 includes multiple antennas 916, the smartphone 900 may also include a single antenna 916.

[0139] In addition, the smartphone 900 may include an antenna 916 for each wireless communication scheme. In this case, the antenna switch 915 may be omitted from the configuration of the smartphone 900.

[0140] The bus 917 connects the processor 901, the memory 902, the storage device 903, the external connection interface 904, the camera 906, the sensor 907, the microphone 908, the input device 909, the display device 910, the speaker 911, the wireless communication interface 912, and the auxiliary controller 919. The battery 918 supplies power to the various blocks of the smartphone 900 shown in FIG10 via feeders, which are partially shown as dashed lines in the figure. The auxiliary controller 919 operates the minimum necessary functions of the smartphone 900, for example, in sleep mode.

[0141] In the smartphone 900 shown in FIG10 , the communication unit 202 and the transceiver of the electronic device 200 may be implemented by the wireless communication interface 912. At least a portion of the functionality may also be implemented by the processor 901 or the auxiliary controller 919. For example, the processor 901 or the auxiliary controller 919 may configure and assist the UE in detecting and recovering from AI model failures by measuring reference signals by executing the functions of the generation unit 201, the communication unit 202, and the determination unit 203.

[0142] (Second application example)

[0143] 11 is a block diagram showing an example of a schematic configuration of a car navigation device 920 to which the technology of the present disclosure can be applied. The car navigation device 920 includes a processor 921, a memory 922, a global positioning system (GPS) module 924, a sensor 925, a data interface 926, a content player 927, a storage medium interface 928, an input device 929, a display device 930, a speaker 931, a wireless communication interface 933, one or more antenna switches 936, one or more antennas 937, and a battery 938.

[0144] The processor 921 may be, for example, a CPU or an SoC, and controls a navigation function and other functions of the car navigation apparatus 920. The memory 922 includes a RAM and a ROM, and stores data and programs executed by the processor 921.

[0145] The GPS module 924 measures the position (such as latitude, longitude, and altitude) of the car navigation device 920 using GPS signals received from GPS satellites. The sensor 925 may include a group of sensors such as a gyroscope sensor, a geomagnetic sensor, and an air pressure sensor. The data interface 926 is connected to, for example, the in-vehicle network 941 via an unillustrated terminal and acquires data generated by the vehicle (such as vehicle speed data).

[0146] The content player 927 reproduces content stored in a storage medium (such as a CD or DVD) inserted into the storage medium interface 928. The input device 929 includes, for example, a touch sensor, button, or switch configured to detect a touch on the screen of the display device 930, and receives an operation or information input from the user. The display device 930 includes a screen such as an LCD or OLED display and displays an image of a navigation function or reproduced content. The speaker 931 outputs the sound of the navigation function or the reproduced content.

[0147] The wireless communication interface 933 supports any cellular communication scheme (such as LTE and LTE-Advanced) and performs wireless communication. The wireless communication interface 933 may generally include, for example, a BB processor 934 and an RF circuit 935. The BB processor 934 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 935 may include, for example, a mixer, a filter, and an amplifier, and transmit and receive wireless signals via an antenna 937. The wireless communication interface 933 may also be a chip module on which the BB processor 934 and the RF circuit 935 are integrated. As shown in Figure 11, the wireless communication interface 933 may include multiple BB processors 934 and multiple RF circuits 935. Although Figure 11 shows an example in which the wireless communication interface 933 includes multiple BB processors 934 and multiple RF circuits 935, the wireless communication interface 933 may also include a single BB processor 934 or a single RF circuit 935.

[0148] In addition, in addition to the cellular communication scheme, the wireless communication interface 933 can support other types of wireless communication schemes, such as a short-range wireless communication scheme, a near field communication scheme, and a wireless LAN scheme. In this case, for each wireless communication scheme, the wireless communication interface 933 can include a BB processor 934 and an RF circuit 935.

[0149] Each of the antenna switches 936 switches a connection destination of the antenna 937 between a plurality of circuits included in the wireless communication interface 933 , such as circuits for different wireless communication schemes.

[0150] Each of the antennas 937 includes a single or multiple antenna elements (such as multiple antenna elements included in a MIMO antenna) and is used for transmitting and receiving wireless signals via the wireless communication interface 933. As shown in FIG11, the car navigation device 920 may include multiple antennas 937. Although FIG11 shows an example in which the car navigation device 920 includes multiple antennas 937, the car navigation device 920 may also include a single antenna 937.

[0151] Furthermore, the car navigation device 920 may include an antenna 937 for each wireless communication scheme. In this case, the antenna switch 936 may be omitted from the configuration of the car navigation device 920.

[0152] The battery 938 supplies power to the respective blocks of the car navigation device 920 shown in Fig. 11 via a feeder line, which is partially shown as a dotted line in the figure. The battery 938 accumulates the power supplied from the vehicle.

[0153] In the car navigation device 920 shown in FIG11 , the communication unit 202 and transceiver of the electronic device 200 may be implemented by the wireless communication interface 933. At least a portion of the functionality may also be implemented by the processor 921. For example, the processor 921 may configure and assist the UE in detecting and recovering from AI model failures by measuring reference signals by executing the functions of the generation unit 201, the communication unit 202, and the determination unit 203.

[0154] The technology of the present disclosure can also be implemented as an in-vehicle system (or vehicle) 940 including a car navigation device 920, an in-vehicle network 941, and one or more blocks of a vehicle module 942. The vehicle module 942 generates vehicle data (such as vehicle speed, engine speed, and fault information) and outputs the generated data to the in-vehicle network 941.

[0155] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be pointed out that for those skilled in the art, it is understandable that all or any steps or components of the methods and devices of the present disclosure can be implemented in any computing device (including a processor, storage medium, etc.) or a network of computing devices in the form of hardware, firmware, software, or a combination thereof. This can be achieved by those skilled in the art using their basic circuit design knowledge or basic programming skills after reading the description of the present disclosure.

[0156] Furthermore, the present disclosure also provides a program product storing machine-readable instruction codes. When the instruction codes are read and executed by a machine, the method according to the embodiment of the present disclosure can be executed.

[0157] Accordingly, the storage medium for carrying the program product storing the machine-readable instruction code is also included in the disclosure of the present invention, including but not limited to a floppy disk, an optical disk, a magneto-optical disk, a memory card, a memory stick, and the like.

[0158] When the present disclosure is implemented through software or firmware, the programs constituting the software are installed from a storage medium or a network to a computer with a dedicated hardware structure (such as the general-purpose computer 1200 shown in Figure 12). When various programs are installed on the computer, it can perform various functions, etc.

[0159] In FIG12 , a central processing unit (CPU) 1201 executes various processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage section 1208 to a random access memory (RAM) 1203. In the RAM 1203, data required when the CPU 1201 executes various processes, etc., is also stored as needed. The CPU 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output interface 1205 is also connected to the bus 1204.

[0160] The following components are connected to the input / output interface 1205: an input section 1206 (including a keyboard, a mouse, etc.), an output section 1207 (including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and speakers, etc.), a storage section 1208 (including a hard disk, etc.), and a communication section 1209 (including a network interface card such as a LAN card, a modem, etc.). The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 may also be connected to the input / output interface 1205 as needed. A removable medium 1211 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is installed in the drive 1210 as needed, so that a computer program read therefrom is installed in the storage section 1208 as needed.

[0161] 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 1211 .

[0162] It should be understood by those skilled in the art that such storage media is not limited to the removable medium 1211 shown in FIG12 , which stores the program therein and is distributed separately from the device to provide the program to the user. Examples of the removable medium 1211 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 1202, a hard disk included in the storage section 1208, or the like, in which the program is stored and distributed to the user together with the device containing them.

[0163] It should also be noted that in the apparatus, method, and system of the present disclosure, each component or step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure. Furthermore, the steps of performing the above series of processes can naturally be performed in chronological order according to the order of description, but do not necessarily need to be performed in chronological order. Certain steps can be performed in parallel or independently of each other.

[0164] Finally, it should be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. Furthermore, in the absence of further limitations, an element defined by the phrase "comprises a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0165] Although the embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, it should be understood that the embodiments described above are merely illustrative of the present disclosure and are not intended to limit the present disclosure. Those skilled in the art will appreciate that various modifications and variations can be made to the above embodiments without departing from the spirit and scope of the present disclosure. Therefore, the scope of the present disclosure is solely defined by the appended claims and their equivalents.

Claims

1. An electronic device for a user device, comprising: at least one processor; as well as at least one memory including computer program code, wherein the at least one memory and the computer program code are configured to, through the at least one processor, cause the electronic device to execute: determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring a reference signal for artificial intelligence model failure detection; as well as When the performance metric satisfies a predetermined condition, it is determined that an artificial intelligence model failure has occurred, and a fallback to a working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model is executed.

2. The electronic device according to claim 1, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: Upon first detecting an artificial intelligence model failure event, starting a timer and a counter, and incrementing the counter by 1 each time the artificial intelligence model failure event is subsequently detected, wherein the detection of the artificial intelligence model failure event includes the performance metric being lower than a first threshold; and If the count value of the counter is greater than a predetermined maximum count before the timer expires, it is determined that the performance metric meets a predetermined condition and that the artificial intelligence model failure has occurred.

3. The electronic device according to claim 2, wherein: The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: After determining that the artificial intelligence model fails, executing a fallback to the working mechanism of the non-artificial intelligence model; as well as The performance metric of the candidate artificial intelligence models in the candidate artificial intelligence model list is measured to determine the second artificial intelligence model, wherein the performance metric of the second artificial intelligence model is higher than a second threshold.

4. The electronic device according to claim 3, wherein The fallback includes the user equipment re-establishing a connection with the network side through a random access process.

5. The electronic device according to claim 4, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: receiving an instruction regarding model switching from the network side via the connection; In response to the indication, sending an ACK to the network side if the second artificial intelligence model transparent to the network side exists, and sending a NACK to the network side if the second artificial intelligence model does not exist; In the case where the second artificial intelligence model exists, sending a switching success report to the network side after successfully switching to the second artificial intelligence model; as well as In the absence of the second artificial intelligence model, the user equipment is kept in a non-artificial intelligence model working mechanism. The electronic device according to claim 5 , wherein: The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: After successfully switching to the second artificial intelligence model, a request for additional conditions is also sent to the network side.

7. The electronic device according to claim 6, wherein: The additional conditions include a network-side beam pattern.

8. The electronic device according to claim 4, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: The second artificial intelligence model switched to is notified to the network side through uplink control information.

9. The electronic device according to claim 2, wherein: The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: When the timer starts, start measuring the performance metric of the candidate artificial intelligence models in the candidate artificial intelligence model list to determine the second artificial intelligence model, wherein the performance metric of the second artificial intelligence model is higher than a second threshold; and After determining that the artificial intelligence model fails, switching to the second artificial intelligence model is executed.

10. The electronic device according to claim 9, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: The second artificial intelligence model switched to is notified to the network side through a beam failure recovery process in a random access process.

11. The electronic device according to claim 9, wherein There is a mapping relationship between the random access preamble sequence and the identifier of the artificial intelligence model, so that the network side implicitly determines the second artificial intelligence model.

12. The electronic device according to claim 3 or 9, wherein: The candidate artificial intelligence model list is included in the field reported by the user equipment capability.

13. The electronic device according to claim 1, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: providing a performance indication based on the performance metric of the first artificial intelligence model to a network side, so that the network side determines whether an artificial intelligence model failure has occurred based on the performance indication; as well as An artificial intelligence model failure is determined to have occurred based on an indication from the network side.

14. The electronic device according to claim 13, wherein: The performance indication comprises one of: the performance metric, an indication of whether the performance metric is above a first threshold.

15. The electronic device according to claim 1, wherein The performance metrics include one or more of the following: prediction accuracy of the artificial intelligence model, the difference between the predicted value and the true value of the artificial intelligence model, input and output data distribution characteristics, and link quality.

16. The electronic device according to claim 1, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: receiving a configuration for the reference signal from a network side, The first artificial intelligence model and the second artificial intelligence model are used for beam management, and the configuration of the reference signal is performed by adding AImodelFailure for artificial intelligence model failure detection in the purpose field of RadioLinkMonitoringRS.

17. The electronic device according to claim 1, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: Receive configurations for AI model failure detection and recovery from the network side, The configuration includes one or more of the following: the number of events required to trigger the failure of an artificial intelligence model, a specified time length for counting artificial intelligence model failure events, a first threshold for evaluating the performance level of the first artificial intelligence model, a list of candidate artificial intelligence models, and a second threshold for evaluating the performance level of the second artificial intelligence model.

18. An electronic device for a network side, comprising: at least one processor; as well as at least one memory including computer program code, wherein the at least one memory and the computer program code are configured to, through the at least one processor, cause the electronic device to execute: generating a configuration of a reference signal for artificial intelligence model failure detection, and providing the configuration to a user equipment, wherein the user equipment determines a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring the reference signal; as well as When the performance metric satisfies a predetermined condition, it is determined that an artificial intelligence model failure has occurred, and the user equipment performs a fallback to a working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model.

19. The electronic device according to claim 18, wherein The first artificial intelligence model and the second artificial intelligence model are used for beam management, and the configuration of the reference signal is performed by adding AImodelFailure for artificial intelligence model failure detection in the purpose field of RadioLinkMonitoringRS.

20. The electronic device according to claim 18, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: Obtaining a performance indication based on the performance metric from the user equipment; and A determination is made as to whether an artificial intelligence model failure has occurred based on the performance indication.

21. The electronic device according to claim 20, wherein The performance indication comprises one of: the performance metric, an indication of whether the performance metric is above a first threshold.

22. The electronic device according to claim 21, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: starting a timer and a counter when the performance indicator indicates that the performance metric is below the first threshold, and incrementing the counter by 1 each subsequent time the performance metric is below the first threshold; as well as If the count value of the counter is greater than a predetermined maximum count before the timer expires, it is determined that the performance metric meets a predetermined condition and that the artificial intelligence model failure has occurred.

23. The electronic device according to claim 21, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: starting a timer and a counter when the performance indicator indicates that the performance metric is below the first threshold, and incrementing the counter by 1 each subsequent time the performance metric is above the first threshold; as well as If the count value of the counter is less than a predetermined value before the timer expires, it is determined that the performance metric meets the predetermined condition and that the artificial intelligence model failure has occurred.

24. The electronic device according to claim 18, wherein The fallback includes the user equipment re-establishing a connection with the network side through a random access process.

25. The electronic device according to claim 24, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: sending an indication of model switching to the user equipment via the connection; receiving, from the user equipment, an ACK or a NACK sent by the user equipment in response to the indication, wherein the ACK indicates the existence of the second artificial intelligence model that is transparent to the network side, and the NACK indicates the absence of the second artificial intelligence model; Upon receiving the ACK, further receiving a handover success report from the user equipment, wherein the handover success report indicates that the user equipment has successfully switched to the second artificial intelligence model; and When the NACK is received, it is known that the user equipment is still operating under the non-artificial intelligence model.

26. The electronic device according to claim 25, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: In case the ACK is received, a request for additional conditions is also received from the user equipment.

27. The electronic device according to claim 26, wherein The additional conditions include a network-side beam pattern.

28. The electronic device according to claim 18, wherein The at least one memory and the computer program code are further configured to, through the at least one processor, cause the electronic device to: Acquiring information of the second artificial intelligence model switched to from the user equipment through a beam failure recovery process in a random access process; or The information of the second artificial intelligence model switched to is obtained from the user equipment through uplink control information.

29. The electronic device according to claim 28, wherein There is a mapping relationship between the random access preamble sequence and the identifier of the artificial intelligence model. The at least one memory and the computer program code are further configured to enable the electronic device to implicitly determine the second artificial intelligence model through the at least one processor.

30. A method for an electronic device on a user equipment side, comprising: determining a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring a reference signal for artificial intelligence model failure detection; as well as When the performance metric satisfies a predetermined condition, it is determined that an artificial intelligence model failure has occurred, and a fallback to a working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model is executed.

31. A method for an electronic device on a network side, comprising: generating a configuration of a reference signal for artificial intelligence model failure detection, and providing the configuration to a user equipment, wherein the user equipment determines a performance metric of a currently used first artificial intelligence model based on a measurement result obtained by measuring the reference signal; as well as When the performance metric satisfies a predetermined condition, it is determined that an artificial intelligence model failure has occurred, and a fallback to a working mechanism of a non-artificial intelligence model or a switch from the first artificial intelligence model to a second artificial intelligence model is executed.

32. A computer-readable storage medium having computer-executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method according to claim 30 or 31.

33. A computer program product comprising a computer program / instructions, wherein: When the computer program / instructions are executed by a processor, the steps of the method according to claim 30 or 31 are implemented.

Images