Communication method and communication apparatus

Abstract

The present application provides a communication method and a communication apparatus. In the present application, after sending first request information and before a network device issues instruction information for switching or deactivating a first AI function, a first measurement report is sent, wherein the first measurement report is a measurement report generated by a terminal device on the basis of the first AI function before sending the first request information; or, the first measurement report is a measurement report generated by using a predefined field on the basis of a measurement report format corresponding to the first AI function; or, the first measurement report is a measurement report generated by the terminal device on the basis of the first AI function, the first measurement report comprises a first content and / or a second content, and the credibility of the first content is higher than that of the second content; or, the first measurement report is a measurement report generated by the terminal device on the basis of a non-AI function. The present application solves the problem of how to use an AI function or an AI model when the AI function or the AI model are inapplicable to the terminal device.

Description

Communication method and communication device

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on August 11, 2023, with application number 202311018680.5 and application name “Communication Method and Communication Device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a communication method and a communication device. Background Art

[0003] Currently, artificial intelligence (AI) models or AI functions on user equipment (UE) are increasingly being applied in communications technology. However, due to the generalization issues of these AI models or AI functions, their applicability conditions or scenarios are limited, resulting in them being unsuitable in certain conditions or scenarios.

[0004] Therefore, when AI functions or AI models are not applicable, how UE should use AI functions or AI models becomes a technical problem that needs to be solved urgently.

[0005] Summary of the Invention

[0006] The present application provides a communication method and a communication device to solve the problem of how a terminal device uses an AI function or an AI model when the AI ​​function or AI model is not applicable.

[0007] In a first aspect, the present application provides a communication method, the method including: sending first request information, where the first request information is used to request a network device to switch or deactivate a first artificial intelligence (AI) function; sending a first measurement report, where the first measurement report is a measurement report generated by a terminal device based on the first AI function before sending the first request information, or the first measurement report is a measurement report generated using a predefined field based on a measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report includes first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0008] Among them, this method can be executed by the terminal device, and can also be executed by the chip system, hardware circuit and / or software module applied to the terminal device.

[0009] Among them, the first artificial intelligence (AI) function is an AI function that is currently not applicable in the terminal device.

[0010] Optionally, the terminal device can determine whether the current AI function of the terminal device is applicable based on the judgment conditions configured by the network device for the terminal device, or the terminal device can determine whether the current AI function is applicable based on its own status information, which includes speed, scene, location, cell, power, storage capacity and other information.

[0011] Optionally, the first request information may be a physical layer request, or a medium access control (MAC) layer request, or a radio resource control (RRC) layer request.

[0012] Optionally, the terminal device may periodically send the first request information. Accordingly, the network device periodically detects whether the first request information is present on the corresponding time domain resources or frequency domain resources. The period may be a predefined period or a period configured by the network device.

[0013] In this technical solution, the four first measurement reports sent by the terminal device correspond to the four default reporting modes of the terminal device. The terminal device can independently determine which measurement report to send, or determine it based on the configuration of the network device.

[0014] As an example, when the first measurement report is a measurement report generated based on the first AI function before sending the first request information, the default reporting method is to report an unupdated measurement report; when the first measurement report is a measurement report generated using a predefined field based on the measurement report format corresponding to the first AI function, the default reporting method is to report a measurement report filled with predefined fields; when the first measurement report is a measurement report generated based on the first AI function and the first measurement report contains first content and / or second content, the default reporting method is to report a measurement report that can distinguish the credibility of the content; when the first measurement report is a measurement report generated based on a non-AI function, the default reporting method is to report the measurement report corresponding to the non-AI function.

[0015] In this technical solution, when the terminal device determines that the first AI function is not applicable, it can send a first request message to the network device to request the network device to deactivate the first AI function, or switch the first AI function to other applicable AI functions, and send a first measurement report obtained based on a predefined default reporting method to the network device, so that the network device can quickly identify whether the first measurement report is a reliable measurement report, thereby improving efficiency; when using the first AI function to generate a measurement report, the content in the first measurement report is distinguished, thereby avoiding the terminal device using an inapplicable AI function to send an inaccurate measurement report to the network device, thereby avoiding power consumption of the terminal device and waste of AI resources.

[0016] In combination with the first aspect, in certain implementations of the first aspect, before sending the first request information, the method further includes: receiving first configuration information, where the first configuration information is used to indicate a preset duration, and the starting time of the preset duration is determined based on the sending time of the first request information; wherein the duration between the sending time of the first request information and the time when the terminal device receives the response information is greater than the preset duration, and the response information is used to instruct the terminal device to switch the first AI function to the second AI function or deactivate the first AI function; sending the first measurement report includes: sending the first measurement report before the end time of the preset duration.

[0017] Optionally, the first configuration information is also used to instruct the terminal device to autonomously determine that the current AI function is not applicable for subsequent operations after the end of the preset time period.

[0018] Optionally, one AI function may correspond to one or more preset durations, and one preset duration may also correspond to one or more AI functions.

[0019] In this implementation, the network device can configure a timer for each AI function in the terminal device. The timer duration is used to indicate the maximum response time between the network device receiving the first request information and issuing the response information. The timer duration can also be called the preset duration. The timer starts at the start time of the preset duration, and the timer stops at the end time of the preset duration.

[0020] In this implementation, the network device configures a preset duration for the terminal device so that after the preset duration ends, the terminal device does not need to wait for the network device to send a response message, but can independently determine that the current AI function is not applicable for subsequent operations to save power consumption.

[0021] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: after the end time of the preset duration, deactivating the first AI function and / or sending a second measurement report, where the second measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0022] In this implementation, the terminal device autonomously deactivates the first AI function after the end of the preset duration, and can also release the measurement configuration of the first AI function, such as one or more of the measurement resource configuration and the reporting configuration, to improve the utilization of AI resources.

[0023] Optionally, after the end of the preset time period, the terminal device can also adopt a fallback method (such as falling back to a traditional method) to replace the first AI function to generate and send the first measurement report, thereby avoiding power consumption waste of the terminal device. Among them, the traditional method is a non-AI function method.

[0024] In combination with the first aspect, in certain implementations of the first aspect, before receiving the first configuration information, the method further includes: sending first indication information, where the first indication information is used to indicate the maximum value of the preset duration supported by the terminal device.

[0025] In this implementation, the maximum value of the preset duration can be understood as the maximum duration that the terminal device expects to wait for switching or deactivating the first AI function, or the maximum duration that the terminal device allows to wait for the network device to switch / deactivate the first AI function after sending the first request information.

[0026] Optionally, the maximum value of the preset duration corresponding to all AI functions in the terminal device may be consistent or inconsistent, and no specific limitation is imposed here.

[0027] In this implementation, the network device can configure the preset duration to the terminal device based on the first indication information, thereby improving the rationality of setting the preset duration.

[0028] In combination with the first aspect, in certain implementations of the first aspect, before sending the first request information, the method further includes: sending second indication information, the second indication information including at least one of the following information: combination information of applicable conditions supported by the AI ​​characteristics of the terminal device, measurement information reporting method corresponding to the combination information of the applicable conditions, and measurement information reporting method supported by the terminal device.

[0029] In this implementation, the configuration of a default AI function that can serve as a fallback when the activated AI function in the terminal device is not applicable can be determined based on the combination of applicable conditions supported by the terminal device's AI features. It should be understood that the default AI function is more robust or generalizable than the activated AI function in the terminal device.

[0030] Among them, the corresponding configuration of the default AI function can be called the measurement information reporting method corresponding to the combination information of the applicable conditions supported by the AI ​​feature of the terminal device.

[0031] Optionally, the measurement information reporting method supported by the terminal device can be understood as the measurement information reporting method determined autonomously by the terminal device, including the default reporting method mentioned above, and also including the default AI function as fallback.

[0032] In this implementation, after the AI ​​function activated in the terminal device is no longer applicable, the first measurement report can be generated and determined according to the method indicated in the second indication information, thereby improving the flexibility of determining the first measurement report; and when the first measurement report is generated using the default AI function, it is more adapted to the AI ​​capability information of the terminal device, so that the accuracy of the first measurement report is higher.

[0033] In combination with the first aspect, in certain implementations of the first aspect, the first measurement report is generated based on a measurement information reporting method supported by the terminal device; wherein, the first measurement report also includes third indication information, and the third indication information is used to indicate the measurement information reporting method used by the terminal device when generating the first measurement report.

[0034] In this implementation, since the first measurement report is determined based on the measurement information reporting method autonomously determined by the terminal device, the specific measurement information reporting method used (such as the category or identifier of the aforementioned default reporting method, or the category or identifier of the default AI function) can be indicated in the first measurement report to avoid the problem of inconsistent understanding of the first measurement report by the network device and the terminal device.

[0035] In combination with the first aspect, in certain implementations of the first aspect, after sending the second indication information and before sending the first request information, the method further includes: receiving second configuration information, where the second configuration information is used to indicate a measurement information reporting method used by the terminal device when the first AI function fails; wherein the first measurement report is a measurement report determined based on the second configuration information.

[0036] In this implementation, after receiving the second indication information, the network device may configure, for the terminal device, based on the second indication information, a measurement information reporting method to be used when the first AI function fails. The failure of the first AI function may be understood as the first AI function no longer being applicable.

[0037] Optionally, the second configuration information may also be used to indicate whether the network device allows the terminal device to determine the first measurement report using an autonomously determined measurement information reporting method.

[0038] In combination with the first aspect, in certain implementations of the first aspect, the first request information is further used to indicate a time when the first AI function is deactivated.

[0039] In this implementation, the terminal device has the ability to predict that the currently applicable AI function may no longer be applicable within a certain period of time in the future.

[0040] The failure moment of the first AI function may be understood as the moment when the first AI function is no longer applicable, and may be the moment when the measurement configuration of the first AI function is no longer applicable, such as the moment when the measurement resource configuration is no longer applicable and / or the reporting configuration of the first AI function is no longer applicable.

[0041] In this implementation, after predicting that the first AI function may fail in the future, the terminal device may carry the failure time or failure duration of the first AI function in the first request information and send it to the network device.

[0042] In this implementation, after predicting that the first AI function may fail, the terminal device sends a first request message to the network device in advance, thereby reserving time for the network device to send a response message, thereby reducing the time waiting for switching or deactivating the first AI function.

[0043] In combination with the first aspect, in certain implementations of the first aspect, before sending the first request information, the method further includes: sending fourth indication information, where the fourth indication information is used to indicate the terminal device's ability to predict the failure of the AI ​​function, and the fourth indication information includes at least one of the following information: the terminal device has the ability to predict the failure of the AI ​​function; the expected failure time of the AI ​​function in the terminal device; and the confidence level of the expected failure time.

[0044] In this implementation, the terminal device has the ability to predict the failure of AI functions, which can be understood as the terminal device having the ability to predict that the currently applicable AI functions may no longer be applicable within a certain period of time in the future.

[0045] Optionally, the start time of the expected expiration time of the AI ​​function may be the time when the fourth indication information is sent, or a time offset from the time when the fourth indication information is sent; the end time of the expected expiration time of the AI ​​function may be the time when the AI ​​function fails, or a time offset from the time when the AI ​​function fails. The offset time may be set according to actual needs and is not specifically limited in this application.

[0046] Optionally, the expected failure time of an AI function can be represented by the time at which the AI ​​function fails. For example, if the terminal device predicts that the third AI function may fail in 5 ms, the expected failure time of the third AI function is 5 ms. For another example, if the terminal device predicts that the third AI function will fail at a certain time, the predicted failure time of the third AI function is that time.

[0047] Optionally, the expected failure time of an AI function can be expressed using a fixed duration level. For example, if the terminal device predicts that the third AI function may fail after 5ms, and 5ms falls within the range of the second fixed duration level [5ms, 8ms], then the expected failure time of the third AI function is the second fixed duration level.

[0048] Optionally, the fixed duration level may be predetermined by the terminal device and the network device through an agreement, or may be configured by the network device to the terminal device, and this application does not impose any specific restrictions on this.

[0049] Optionally, the expected failure time of an AI function can be represented by a characteristic value of a fixed duration level corresponding to the AI ​​function. The characteristic value of the fixed duration level can be the left endpoint, right endpoint, or average value of a fixed duration level interval, etc., without specific limitation. For example, if the terminal device predicts that the third AI function falls within the range of the second fixed duration level [5ms, 8ms], the expected failure time of the third AI function can be 5ms, 8ms, or 6.5ms.

[0050] Optionally, the fourth indication information is also used to indicate the time point when the AI ​​function is expected to fail. The time point can be expressed using absolute time (such as Greenwich Mean Time, etc.) or using frame number, subframe number, time slot number, or symbol number.

[0051] Optionally, the fourth indication information further indicates an AI function failure interruption time, which indicates the length of time the AI ​​function will be inactive from the moment the AI ​​function fails, after which the AI ​​function can be used again. For example, a 10s interruption time indicates that the AI ​​function is considered inactive for 10 seconds from the moment the AI ​​function fails. After 10 seconds, the AI ​​function is no longer inactive, i.e., the AI ​​function is still usable.

[0052] In this implementation, the confidence level of the expected failure time can be understood as the accuracy of the expected failure time.

[0053] Optionally, different confidence levels can be set for different expected failure times. For example, when the expected failure time is expressed as the expiration time of the AI ​​function, if the expected failure time is within 5 seconds, the accuracy of the expected failure time can be 90%, and if the expected failure time is within 10 seconds, the accuracy of the expected failure time can be 80%.

[0054] In combination with the first aspect, in certain implementations of the first aspect, the expected failure time is the expected failure time of the first AI function.

[0055] In this implementation, before the first AI function fails, the terminal device uses the first AI function to generate and report the first measurement report. After the first AI function fails, the terminal device can use the aforementioned default reporting method or the default AI function to replace the first AI function to determine and report the first measurement report, thereby avoiding the terminal device using an inappropriate AI function to send an inaccurate measurement report to the network device, thereby avoiding power consumption of the terminal device and waste of AI resources.

[0056] In combination with the first aspect, in some implementations of the first aspect, before sending the first request information, the method further includes: receiving third configuration information, where the third configuration information is used to indicate the time when the terminal device sends the first measurement report.

[0057] In this implementation, the first measurement report can be sent periodically or multiple times at the terminal device's discretion, which can result in the network device being unable to quickly identify whether the content in the first measurement report is reliable. Therefore, the network device can send third configuration information to the terminal device to limit the time when the first measurement report is sent.

[0058] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: before receiving the third configuration information, sending fifth indication information, wherein the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

[0059] In this implementation, the terminal device sends the expected sending time of the first measurement report to the network device, so that the network device can send the third configuration information based on the fifth indication information, so that the terminal device and the network device reach an agreement on the sending time of the first measurement report.

[0060] In combination with the first aspect, in certain implementations of the first aspect, the time for sending the first measurement report indicated by the third configuration information includes at least one of the following moments: the time when the first request information is sent, the time when the first threshold condition is satisfied between the time when the first request information is sent, the time when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received, and the time when the second threshold condition is satisfied between the time when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received.

[0061] In a second aspect, the present application provides a communication method, the method including: receiving first request information, where the first request information is used to request a network device to switch or deactivate a first artificial intelligence (AI) function; receiving a first measurement report, where the first measurement report is a measurement report generated by a terminal device based on the first AI function before sending the first request information, or the first measurement report is a measurement report generated using a predefined field based on a measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report contains first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0062] This method can be executed by a network device, or by a chip system, hardware circuit and / or software module applied in the network device.

[0063] In this implementation, after the terminal device sends the first request information and the first measurement report, the network device receives the first request information and the first measurement report accordingly.

[0064] In combination with the second aspect, in certain implementations of the second aspect, before receiving the first request information, the method further includes: sending first configuration information, where the first configuration information is used to indicate a preset duration; wherein the duration between the time when the first request information is sent and the time when the terminal device receives the response information is greater than the preset duration, and the response information is used to instruct the terminal device to switch the first AI function to the second AI function or deactivate the first AI function.

[0065] Optionally, the first configuration information is also used to instruct the terminal device to autonomously determine that the current AI function is not applicable for subsequent operations after the end of the preset time period.

[0066] In this implementation, the network device can configure a preset duration for the terminal device so that after the preset duration ends, the terminal device does not need to wait for the network device to send a response message, but can independently determine that the current AI function is not applicable for subsequent operations to save power consumption.

[0067] In combination with the second aspect, in certain implementations of the second aspect, the method further includes: receiving a second measurement report, where the second measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0068] In this implementation, after the end of the preset duration or after deactivating the first AI function, the terminal device generates a second measurement report using a non-AI function. Accordingly, the network device receives the second measurement report.

[0069] In combination with the second aspect, in certain implementations of the second aspect, before sending the first configuration information, the method further includes: receiving first indication information, where the first indication information is used to indicate the maximum value of the preset duration supported by the terminal device.

[0070] Among them, the maximum value of the preset time length can be understood as the maximum value of the time length that the terminal device expects to wait for switching or deactivating the first AI function, or the maximum time length that the terminal device allows to wait for the network device to switch / deactivate the first AI function after sending the first request information.

[0071] In this implementation, the network device can configure the preset duration to the terminal device based on the first indication information, thereby improving the rationality of setting the preset duration.

[0072] In combination with the second aspect, in certain implementations of the second aspect, before receiving the first request information, the method further includes: receiving second indication information, the second indication information including at least one of the following information: combination information of applicable conditions supported by the AI ​​characteristics of the terminal device, measurement information reporting method corresponding to the combination information of the applicable conditions, and measurement information reporting method supported by the terminal device.

[0073] In this implementation, the terminal device may send to the network device a measurement information reporting method that may be used when the activated AI function in the terminal device fails. Accordingly, the network device receives the second indication information.

[0074] In combination with the second aspect, in certain implementations of the second aspect, the first measurement report is generated based on a measurement information reporting method supported by the terminal device; wherein, the first measurement report also includes third indication information, and the third indication information is used to indicate the measurement information reporting method used by the terminal device when generating the first measurement report.

[0075] In combination with the second aspect, in certain implementations of the second aspect, after receiving the second indication information and before receiving the first request information, the method further includes: sending second configuration information, where the second configuration information is used to indicate the measurement information reporting method used by the terminal device when the first AI function fails; wherein the first measurement report is a measurement report determined based on the second configuration information.

[0076] In this implementation, the network device can configure the measurement information reporting method used when the first AI function fails for the terminal device based on the second indication information, so that the terminal device can generate and report the first measurement report based on the second configuration information.

[0077] Optionally, the second configuration information is further used to indicate whether the terminal device can determine the first measurement report using an autonomously determined measurement information reporting method.

[0078] In combination with the second aspect, in certain implementations of the second aspect, the first request information is further used to indicate a time when the first AI function is invalidated.

[0079] In combination with the second aspect, in certain implementations of the second aspect, before receiving the first request information, the method also includes: receiving fourth indication information, the fourth indication information being used to indicate the ability information of the terminal device to predict the failure of the AI ​​function, the fourth indication information including at least one of the following information: the terminal device has the ability to predict the failure of the AI ​​function; the expected failure time of the AI ​​function in the terminal device; and the confidence level of the expected failure time.

[0080] In this implementation, the network device receives the fourth indication information sent by the terminal device.

[0081] In combination with the second aspect, in certain implementations of the second aspect, the expected failure time is the expected failure time of the first AI function.

[0082] In combination with the second aspect, in some implementations of the second aspect, before receiving the first request information, the method further includes: sending third configuration information, where the third configuration information is used to indicate the time when the terminal device sends the first measurement report.

[0083] In this implementation, since the first measurement report can be sent periodically or multiple times at the terminal device's discretion, the network device cannot quickly identify whether the content in the first measurement report is reliable. Therefore, the network device can send third configuration information to the terminal device to limit the time when the first measurement report is sent.

[0084] In combination with the second aspect, in certain implementations of the second aspect, the method further includes: before sending the third configuration information, receiving fifth indication information, wherein the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

[0085] In this implementation manner, the network device may configure the third configuration information based on the fifth indication information to improve the rationality of the configuration of the sending time of the first measurement report.

[0086] In combination with the second aspect, in certain implementations of the second aspect, the time for sending the first measurement report indicated by the third configuration information includes at least one of the following moments: the time when the first request information is sent, the moment when the first threshold condition is satisfied between the time when the first request information is sent, the moment when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received, and the moment when the second threshold condition is satisfied between the time when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received.

[0087] In a third aspect, the present application provides a communication device, which includes modules for implementing the method in the first aspect or any one of the implementations thereof, and each module can be implemented in the form of hardware and / or software.

[0088] For example, the device may include: a sending module. The sending module is used to send a first request message, where the first request message is used to request the network device to switch or deactivate a first artificial intelligence (AI) function; the sending module is also used to send a first measurement report, where the first measurement report is a measurement report generated by the terminal device based on the first AI function before sending the first request message, or the first measurement report is a measurement report generated using a predefined field based on a measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report contains first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0089] In combination with the third aspect, in certain implementations of the third aspect, the apparatus further includes: a receiving module. The receiving module is configured to receive first configuration information before sending the first request information, where the first configuration information is used to indicate a preset duration, and the starting time of the preset duration is determined based on the sending time of the first request information; wherein the duration between the sending time of the first request information and the time when the terminal device receives the response information is greater than the preset duration, and the response information is used to instruct the terminal device to switch the first AI function to the second AI function or deactivate the first AI function. The sending module is specifically configured to send the first measurement report before the end time of the preset duration.

[0090] In conjunction with the third aspect, in certain implementations of the third aspect, the apparatus further includes a processing module. The processing module is configured to deactivate the first AI function after the end of the preset duration. The sending module is further configured to send a second measurement report after the end of the preset duration, where the second measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0091] In combination with the third aspect, in certain implementations of the third aspect, the sending module is further used to send first indication information before receiving the first configuration information, and the first indication information is used to indicate the maximum value of the preset time length supported by the terminal device.

[0092] In combination with the third aspect, in certain implementations of the third aspect, the sending module is further used to send second indication information before sending the first request information, and the second indication information includes at least one of the following information: the combination information of the applicable conditions supported by the AI ​​characteristics of the terminal device, the measurement information reporting method corresponding to the combination information of the applicable conditions, and the measurement information reporting method supported by the terminal device.

[0093] In combination with the third aspect, in certain implementations of the third aspect, the first measurement report is generated based on a measurement information reporting method supported by the terminal device; wherein, the first measurement report also includes third indication information, and the third indication information is used to indicate the measurement information reporting method used by the terminal device when generating the first measurement report.

[0094] In combination with the third aspect, in certain implementations of the third aspect, the receiving module is further used to receive second configuration information after sending the second indication information and before sending the first request information, where the second configuration information is used to indicate the measurement information reporting method used by the terminal device when the first AI function fails; wherein the first measurement report is a measurement report determined based on the second configuration information.

[0095] In combination with the third aspect, in certain implementations of the third aspect, the first request information is further used to indicate a time when the first AI function is invalidated.

[0096] In combination with the third aspect, in certain implementations of the third aspect, the sending module is further used to send fourth indication information before sending the first request information, wherein the fourth indication information is used to indicate the ability information of the terminal device to predict the failure of the AI ​​function, and the fourth indication information includes at least one of the following information: the terminal device has the ability to predict the failure of the AI ​​function; the expected failure time of the AI ​​function in the terminal device; and the confidence level of the expected failure time.

[0097] In combination with the third aspect, in certain implementations of the third aspect, the expected failure time is the expected failure time of the first AI function.

[0098] In combination with the third aspect, in some implementations of the third aspect, the receiving module is further used to receive third configuration information before sending the first request information, and the third configuration information is used to indicate the time when the terminal device sends the first measurement report.

[0099] In combination with the third aspect, in certain implementations of the third aspect, the sending module is further used to send fifth indication information before receiving the third configuration information, and the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

[0100] In combination with the third aspect, in certain implementations of the third aspect, the time for sending the first measurement report indicated by the third configuration information includes at least one of the following moments: the time when the first request information is sent, the moment when the first threshold condition is satisfied between the time when the first request information is sent, the moment when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received, and the moment when the second threshold condition is satisfied between the time when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received.

[0101] In a fourth aspect, the present application provides a communication device, which includes modules for implementing the method in the second aspect or any one of the implementation methods thereof, and each module can be implemented in the form of hardware and / or software.

[0102] For example, the device may include: a receiving module. The receiving module is used to receive a first request message, where the first request message is used to request the network device to switch or deactivate a first artificial intelligence (AI) function. The receiving module is also used to receive a first measurement report, where the first measurement report is a measurement report generated by the terminal device based on the first AI function before sending the first request message, or the first measurement report is a measurement report generated using a predefined field based on the measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report contains first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0103] In conjunction with the fourth aspect, in certain implementations of the fourth aspect, the apparatus further includes a sending module. The sending module is configured to send first configuration information before receiving the first request information, where the first configuration information is used to indicate a preset duration; wherein the duration between the time when the first request information is sent and the time when the terminal device receives the response information is greater than the preset duration, and the response information is used to instruct the terminal device to switch the first AI function to the second AI function or to deactivate the first AI function.

[0104] In combination with the fourth aspect, in certain implementations of the fourth aspect, the receiving module is further used to receive a second measurement report, where the second measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0105] In combination with the fourth aspect, in certain implementations of the fourth aspect, the receiving module is further used to receive first indication information before sending the first configuration information, and the first indication information is used to indicate the maximum value of the preset duration supported by the terminal device.

[0106] In combination with the fourth aspect, in certain implementations of the fourth aspect, the receiving module is further used to receive second indication information before receiving the first request information, and the second indication information includes at least one of the following information: the combination information of the applicable conditions supported by the AI ​​characteristics of the terminal device, the measurement information reporting method corresponding to the combination information of the applicable conditions, and the measurement information reporting method supported by the terminal device.

[0107] In combination with the fourth aspect, in certain implementations of the fourth aspect, the first measurement report is generated based on a measurement information reporting method supported by the terminal device; wherein, the first measurement report also includes third indication information, and the third indication information is used to indicate the measurement information reporting method used by the terminal device when generating the first measurement report.

[0108] In combination with the fourth aspect, in certain implementations of the fourth aspect, the sending module is further used to send second configuration information after receiving the second indication information and before receiving the first request information, wherein the second configuration information is used to indicate the measurement information reporting method used by the terminal device when the first AI function fails; wherein the first measurement report is a measurement report determined based on the second configuration information.

[0109] In combination with the fourth aspect, in certain implementations of the fourth aspect, the first request information is further used to indicate a time when the first AI function is invalidated.

[0110] In combination with the fourth aspect, in certain implementations of the fourth aspect, the receiving module is further used to receive fourth indication information before receiving the first request information, wherein the fourth indication information is used to indicate the ability information of the terminal device to predict the failure of the AI ​​function, and the fourth indication information includes at least one of the following information: the terminal device has the ability to predict the failure of the AI ​​function; the expected failure time of the AI ​​function in the terminal device; and the confidence level of the expected failure time.

[0111] In combination with the fourth aspect, in certain implementations of the fourth aspect, the expected failure time is the expected failure time of the first AI function.

[0112] In combination with the fourth aspect, in certain implementations of the fourth aspect, the sending module is further used to send third configuration information before receiving the first request information, and the third configuration information is used to indicate the time when the terminal device sends the first measurement report.

[0113] In combination with the fourth aspect, in certain implementations of the fourth aspect, the receiving module is further used to receive fifth indication information before sending the third configuration information, and the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

[0114] In combination with the fourth aspect, in certain implementations of the fourth aspect, the time for sending the first measurement report indicated by the third configuration information includes at least one of the following moments: the time when the first request information is sent, the moment when the first threshold condition is satisfied between the time when the first request information is sent, the moment when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received, and the moment when the second threshold condition is satisfied between the time when the positive confirmation ACK or negative confirmation NACK corresponding to the first request information is received.

[0115] In a fifth aspect, the present application provides a communication device, comprising a processor, which may be coupled to a memory and configured to call program code in the memory to execute the method described in the first aspect or any possible implementation thereof. Optionally, the device further comprises a memory. Optionally, the device further comprises a communication interface, the processor being coupled to the communication interface.

[0116] Optionally, the apparatus may be a terminal device, or a chip system, a hardware circuit and / or a software module applied in the terminal device.

[0117] In a sixth aspect, the present application provides a communication device, comprising a processor, the processor being coupled to a memory and configured to call program code in the memory to execute the method described in the second aspect or any possible implementation thereof. Optionally, the device further comprises a memory. Optionally, the device further comprises a communication interface, the processor being coupled to the communication interface.

[0118] Optionally, the apparatus may be a network device, or a chip system, a hardware circuit and / or a software module applied in a network device.

[0119] In a seventh aspect, the present application provides a communication system, which includes the device in the third aspect or the fifth aspect, and the device in the fourth aspect or the sixth aspect.

[0120] In an eighth aspect, the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the method as described in the first aspect, the second aspect, or any possible implementation thereof.

[0121] In a ninth aspect, the present application provides a computer-readable medium storing program code for execution by a device, wherein the program code includes a method for executing the method described in the first aspect, the second aspect, or any possible implementation thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0122] FIG1 is a schematic diagram of an application scenario provided by an embodiment of the present application;

[0123] FIG2 is a schematic flow chart of a communication method provided in Example 1 of the present application;

[0124] FIG2a is a schematic diagram showing an embodiment of the present application for determining whether a prediction result is accurate;

[0125] FIG3 is a schematic flow chart of a communication method provided in Example 2 of the present application;

[0126] FIG3a is a schematic diagram of applicable conditions of an AI feature provided in one embodiment of the present application;

[0127] FIG3 b is a schematic diagram of configuring an AI function according to an embodiment of the present application;

[0128] FIG3c is a schematic diagram of configuring an AI function according to an embodiment of the present application;

[0129] FIG4 is a schematic flow chart of a communication method provided in Example 3 of the present application;

[0130] FIG5 is a schematic flow chart of a communication method provided in Example 4 of the present application;

[0131] FIG6 is a schematic diagram illustrating a combination of information that explicitly indicates applicable conditions provided by an embodiment of the present application;

[0132] FIG7 is a schematic flow chart of a communication method provided in Example 5 of the present application;

[0133] FIG8 is a schematic flow chart of a communication method provided in Example 6 of the present application;

[0134] FIG9 is a schematic flow chart of a communication method provided in Example 7 of the present application;

[0135] FIG10 is a schematic flow chart of a communication method provided in Example 8 of the present application;

[0136] FIG11 is a schematic structural diagram of a communication device provided by one embodiment of the present application;

[0137] FIG12 is a schematic structural diagram of a communication device provided in yet another embodiment of the present application.

[0138] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0139] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0140] Currently, version 18 of the fifth-generation (5G) communication system has basically determined the basic framework of the lifecycle management (LCM) of air interface artificial intelligence (AI), as well as three air interface AI use cases / features: AI channel state indicator (CSI) feedback, AI beam management, and AI positioning. Considering that network equipment and terminal equipment can only present the concept of logical models to the outside world when using AI to implement the above use cases, the granularity of AI lifecycle management can be divided into two granularities: AI functionality and AI model. Among them, AI functionality, as an AI feature / feature group enabled by one / multiple configurations, has a coarser granularity than the AI ​​model.

[0141] Taking the use of AI functions on terminal devices as an example, given the generalization issues when using AI functions, AI functions are not used in all situations. As a result, the applicable or available AI functions on the terminal device side may change dynamically as the terminal device changes. Changes in terminal devices include external changes such as the terminal device's location, the scene it is in, and the environment it is in, as well as internal changes such as the terminal device's memory, power, or other hardware conditions.

[0142] As an example, if under specific conditions, the performance of the AI ​​function can meet the preset requirements, then the AI ​​function can be considered applicable under the conditions; otherwise, the AI ​​function can be considered inapplicable under the conditions. For example, when the applicable scenario of an AI function is an indoor scenario, the AI ​​function is applicable in indoor scenarios but not in outdoor scenarios. For another example, an AI function is only applicable when the power level of the terminal device is greater than or equal to a power threshold. When the power level of the terminal device is less than the power threshold, the AI ​​function is no longer applicable.

[0143] It should be understood that AI models also have generalization issues, so the applicable or available AI functions on the terminal device side will also change dynamically with changes in the terminal device.

[0144] Moreover, the AI ​​model or AI function on the terminal device side needs to be controlled by the network side. For example, the activation, deactivation, and switching of the AI ​​model or AI function on the terminal device side all need to be used under the control of the network device. Therefore, if the AI ​​function or AI model on the terminal device side is not applicable, the terminal device can report relevant information to the network device, and the network device can deactivate the no longer applicable AI function or AI model based on the relevant information, or switch to or activate the applicable AI function or AI model.

[0145] However, there is a delay between when the AI ​​model or AI function on the terminal device becomes inapplicable and when the network device officially issues instructions for switching, activating, or deactivating the AI ​​model or AI function. How the terminal device should use the AI ​​function or AI model during this delay has become a pressing technical issue.

[0146] In view of this, the embodiments of the present application provide a communication method and a communication device to solve the problem of how the terminal device uses the AI ​​function or AI model within the delay time. The delay time is the period after the terminal device sends a request message to the network device to switch or deactivate the currently unapplicable AI function or AI model, and before receiving the formal instruction message from the network device to switch, activate, or deactivate the AI ​​model or AI function. The communication method and communication device provided in this application are based on the same technical concept. Since the principles of solving the problem by the method and the device are similar, the implementation of the method and the device can refer to each other, and the repeated parts will not be repeated.

[0147] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0148] The technical solutions provided in the embodiments of the present application can be applied to various communication systems. For example, they can be applied to 5G communication systems, such as 5G new radio (NR) systems, and can be applied to various application scenarios of 5G communication systems, such as application scenarios using air interface AI models or AI functions in 5G cellular network systems. The technical solutions provided in the embodiments of the present application can also be applied to various communication systems that will evolve in the future, such as the sixth generation (6G) communication system, which is not specifically limited in this application.

[0149] Figure 1 is a schematic diagram of an application scenario provided by an embodiment of the present application. As shown in Figure 1 , a communication network 100 includes a network device 101 and a terminal device 102 . The network device 101 can provide wireless access to one or more terminal devices 102 within the coverage area of ​​the network device 101 .

[0150] The network device 101 may be a node in a radio access network (RAN), which may also be referred to as a base station or a RAN node (or device). Examples of the network device 101 include a next generation nodeB (gNB), a next generation evolved nodeB (Ng-eNB), a transmission reception point (TRP), an evolved nodeB (eNB), a radio network controller (RNC), a nodeB (NB), a base station controller (BSC), a base transceiver station (BTS), a home base station (e.g., a home evolved nodeB or home nodeB, HNB), a base band unit (BBU), or a wireless fidelity (Wi-Fi) access point (AP). The network device 101 may also be a satellite, which may also be referred to as a high-altitude platform, a high-altitude aircraft, or a satellite base station. The network device 101 may also be other devices having network device functions. For example, the network device 101 may also be a device that functions as a network device in device-to-device (D2D) communication, Internet of Vehicles (IoV), or machine-to-machine (M2M) communication. The network device 101 may also be any possible network device in a future communication system, and this application does not specifically limit this.

[0151] Terminal device 102, also known as user equipment (UE), mobile station (MS), or mobile terminal (MT), provides voice and / or data connectivity to users. Terminal device 102 includes handheld devices with wireless connectivity and in-vehicle devices, also known as on-board units (OBUs). The terminal device 102 can be: a mobile phone, a tablet computer, a laptop computer, a PDA, a mobile internet device (MID), a wearable device (such as a smart watch, a smart bracelet, a pedometer, etc.), an in-vehicle device (such as a car, a bicycle, an electric car, an airplane, a ship, a train, a high-speed train, etc.), a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a smart home device (such as a refrigerator, a television, an air conditioner, an electric meter, etc.), an intelligent robot, workshop equipment, a wireless terminal in unmanned driving, a wireless terminal in remote medical surgery, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, or a wireless terminal in a smart home, a flying device (such as an intelligent robot, a hot air balloon, a drone), etc. The terminal device 102 may also be another device having terminal device functions. For example, the terminal device 102 may also be a device that functions as a terminal device in device-to-device (D2D) communication, Internet of Vehicles (IoV), or machine-to-machine (M2M) communication. In particular, when communication is performed between network devices, a network device that functions as a terminal device may also be considered a terminal device, and this application does not specifically limit this.

[0152] Figure 2 is a schematic flow chart of a communication method provided in Embodiment 1 of the present application, which is applicable to the communication system shown in Figure 1. As shown in Figure 2, the method may include S210 and S220.

[0153] S210: Send first request information, where the first request information is used to request the network device to switch or deactivate a first AI function.

[0154] In this embodiment, if the terminal device determines that the current AI function is not applicable, the terminal device may send a first request message to the network device, requesting the network device to deactivate the currently unapplicable AI function or switch to another applicable AI function. Accordingly, the network device receives the first request message. The currently unapplicable AI function may be referred to as the first AI function.

[0155] Optionally, the terminal device can determine whether the current AI function of the terminal device is applicable based on the judgment conditions configured by the network device for the terminal device. Among them, the judgment conditions include the conditions that need to be met when each AI function fails. For example, when the AI ​​function is applicable or fails, the threshold range that the terminal device's speed needs to meet, the type of scenario in which the terminal device is located, the current reference signal receive power (reference signal receive power, RSRP), signal to interference plus noise ratio (signal to interference plus noise ratio, SINR), reference signal receive quality (reference signal receive quality, RSRQ) and other signal quality conditions that need to be met, the conditions that the channel characteristics need to meet (such as the threshold range of power delay profile (PDP) delay, the threshold range of channel impulse response (CIR) delay, etc.), etc.

[0156] It should be noted that the failure of AI function can be understood as the inapplicability of AI function.

[0157] Optionally, the terminal device can determine whether the current AI function is applicable based on its own status information, including speed, scenario, location, power, storage capacity, and other information. For example, when a terminal device uses an AI function for time-domain prediction, if the terminal device's speed falls within the failure speed threshold range corresponding to the AI ​​function, this indicates that the AI ​​function can no longer support longer measurement sample intervals, longer prediction sample intervals, or larger prediction windows. In another example, the terminal device can determine the delay distribution characteristics based on the channel's PDP spectrum, thereby determining the terminal device's current scenario. For example, when the delay distribution is concentrated in an area with low delay, the terminal device can be determined to be in a near-line-of-sight (LOS) scenario or a near-open-air scenario, thereby determining whether the current AI function is applicable. In another example, the terminal device can determine whether the current AI function is applicable based on physical conditions such as the terminal device's current power and storage capacity. If the power required by the current AI function exceeds the current power of the terminal device, the current AI function can be considered no longer applicable.

[0158] It should be noted that a measurement sample can be understood as one or a group of actual measurement results, and a predicted sample can be understood as one or a group of predicted measurement results or other predicted reporting quantities determined based on the predicted measurement results. As an example, the actual measurement result of a reference signal (RS) at a specific time can be understood as a measurement sample, such as the RSRP obtained based on the measurement of one RS; or the actual measurement result of multiple RSs at a specific time can also be understood as a measurement sample. As an example, the predicted measurement result of an RS at a specific time in the future can be understood as a predicted sample, such as the RSRP that may be measured based on the RS at that point in the future; or the predicted channel information determined based on the predicted measurement results of one or more RSs at a specific time in the future can also be understood as a predicted sample. Among them, if the predicted channel information is determined by the predicted measurement results of one or more RSs, there is no need to measure the actual measurement results of one or more RSs at that point in the future, and then determine the channel information at that point in time based on the actual measurement results, thereby saving resources.

[0159] As an example, the first request information may be a physical layer request. For example, the terminal device sends the first request information as uplink control information (UCI) carried on a physical uplink control channel (PUCCH) or a physical uplink shared channel (PUSCH) to the network device.

[0160] Among them, the network device can configure the terminal device to allow the terminal device to report the request information, and allocate the corresponding UCI time domain resources and frequency domain resources. Among them, the time domain resources include the period, the number of occupied time domain units (such as the number of radio frames, subframes, time slots, symbols, etc.), the offset in the time domain position (such as the offset based on radio frames, subframes, time slots, symbols, etc.), etc., and the frequency domain resources include the bandwidth part (bandwidth part, BWP) for sending the first request information, the position and number of resource blocks (resource block, RB), the position and number of resource elements (resource element, RE), the position and number of subcarriers, etc. Optionally, the network device can configure the PUCCH resource identifier used for the reporting request for the terminal device. Optionally, the network device can configure the number of reporting request information for the terminal device, such as the maximum number of reporting times.

[0161] Optionally, the terminal device may periodically send the first request information. Accordingly, the network device periodically detects whether the first request information is present on the corresponding time domain resources or frequency domain resources. The period may be a predefined period or a period configured by the network device.

[0162] Optionally, the network device may determine whether there is a first request message by detecting a change in energy (such as signal strength) on the corresponding time domain resource or frequency domain resource. For example, when the signal strength on the frequency domain resource exceeds a threshold range, it indicates that there is a first request message on the frequency domain resource.

[0163] Optionally, the network device may also determine whether the terminal device has sent a request message requesting switching or deactivation by determining an identifier in the demodulated or decoded first request message.

[0164] As an example, a network device can determine whether a terminal device has sent a request message based on the length of an identifier. For example, if the length of the identifier is greater than a length threshold, it indicates that the terminal device has sent a request message; if the length of the identifier is less than or equal to the length threshold, it indicates that the terminal device has not sent a request message. For another example, if the length of the identifier is less than or equal to the length threshold, it indicates that the terminal device has sent a request message; if the length of the identifier is greater than the length threshold, it indicates that the terminal device has not sent a request message.

[0165] As another example, the length of the identifier can be 1 bit. In this example, it can be considered that when the identifier is 1, it indicates that the terminal device has sent the request information, and when the identifier is 0, it indicates that the terminal device has not sent the request information. It can also be considered that when the identifier is 0, it indicates that the terminal device has sent the request information, and when the identifier is 1, it indicates that the terminal device has not sent the request information.

[0166] As an example, the first request information may be a medium access control (MAC) layer request. For example, the first request information may be sent by configuring a MAC control element (MAC-CE) signaling specifically for the first request information, or by reusing other existing MAC-CE signaling to send the first request information. For example, the MAC-CE header may carry an identifier indicating that the MAC-CE reports the first request information, and the data unit of the MAC-CE may carry an AI function identifier associated with the first request information, such as an AI function identifier requested to be deactivated, or an AI function identifier requested to be switched and a target AI function identifier to be switched to.

[0167] Optionally, the network device may configure a trigger condition for the terminal device to send the first request information. When the trigger condition is met, the terminal device may trigger MAC-CE signaling to send the first request information; or, based on the implementation of the terminal device, the network device may determine whether the terminal device can autonomously trigger applicable MAC-CE signaling to send the first request information. The trigger condition may be a judgment condition configured by the network device for the terminal device to determine whether the AI ​​function is applicable, or other conditions configured for the network device, such as a periodic trigger condition, which is not specifically limited in this application.

[0168] As an example, the first request information may be a radio resource control (RRC) layer request. For example, the first request information may be sent by configuring a dedicated RRC layer signaling specifically for the first request information, or by adding / configuring a field for carrying the first request information in the existing RRC layer signaling, or by reusing other existing RRC layer signaling to send the first request information, where the existing RRC layer signaling includes RRC reconnection request information, UE capability update, UE assistance information (UAI) and other signaling. For example, a new field is added to the existing RRC configuration of UAI to carry the first request information, and the network device may further configure and enable the field in the UAI configuration of RRC, so that the terminal device can trigger the sending of UAI and carry the first request in the newly added field.

[0169] Optionally, if the first request information is used to request deactivation of the first AI function, the first request information may carry an identifier of the first AI function in addition to the request information.

[0170] Optionally, if the first request information is used to request switching to the first AI function, the first request information may carry, in addition to the request information, an identifier of the first AI function and an identifier of a second AI function to be switched to.

[0171] S220: Send a first measurement report, where the first measurement report is a measurement report generated by the terminal device based on the first AI function before sending the first request information, or the first measurement report is a measurement report generated using a predefined field based on the measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report contains first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

[0172] Optionally, the measurement report may be a CSI report, a positioning report, a mobility management report, etc., wherein the CSI report may be a CSI feedback report or a CSI beam management report, which is not specifically limited in this application.

[0173] In this embodiment, after sending the first request information and within the delay period before receiving the network device's formal instruction information to switch and deactivate the first AI function, the terminal device may send a first measurement report obtained based on a predefined default reporting method to the network device, and no longer use the first AI function. Accordingly, the network device receives the first measurement report.

[0174] As an example, the first measurement report is a first AI report generated based on the first AI function before sending the first request information.

[0175] In this example, the terminal device is not required to update the required first AI report, so that the terminal device no longer updates the AI ​​report corresponding to the first AI function before receiving the switching or deactivation indication information, thereby saving power consumption and AI resources.

[0176] As an example, the first measurement report is a measurement report generated using a predefined field based on a measurement report format corresponding to the first AI function. The predefined field may be predefined by a protocol or defined by a network device configuration.

[0177] Optionally, the report format of the measurement report generated using the predefined fields may be inconsistent with the report format of the first AI report, and this application does not specifically limit this. For example, the length, field position, field meaning, and other contents of the measurement report generated using the predefined fields are inconsistent with the length, field position, field meaning, and other contents configured in the first AI report. Optionally, the predefined fields may be different from normal content. For example, the first field in the measurement report is used to represent the measured RSRP result, which is a numerical value, while the predefined field may be a meaningless or invalid numerical value or string. When the accuracy of the RSRP result measured in the first field is low, the predefined field may be used to fill the first field.

[0178] As an example, the first measurement report is a first AI report generated based on a first AI function, and the first measurement report includes first content and second content, and the credibility of the first content is higher than the credibility of the second content. For example, the first content may be a reliable result predicted based on the first AI function, and the second content may be an unreliable result predicted based on the first AI function, wherein the credibility can be understood as accuracy, or the probability that the predicted result is correct. For another example, the first content may be identification information for identifying a reliable result in the first measurement report, and the second content may be identification information for identifying an unreliable result in the first measurement report, or the first content may be identification information for identifying an unreliable result in the first measurement report, and the second content may be identification information for identifying a reliable result in the first measurement report, wherein the credibility can be understood as the accuracy of the identification, or the probability that the identification is correct.

[0179] In this example, when the first measurement report contains only one of the first content and the second content, the first content or the second content may be an identification bit, which is used to identify whether the information in the first measurement report is reliable information or unreliable information, wherein the credibility can be understood as the accuracy of the identification.

[0180] Optionally, in the configuration of the first measurement report, the content reported by the terminal device may include multiple prediction results. For example, in CSI prediction, the first measurement report includes actual measurement channel results or predicted channel results at multiple time points, and in beam management, the first measurement report includes predicted beam measurement results at multiple time points. Among them, because the first measurement report is determined based on the first AI function, the first measurement report will contain accurate prediction results and inaccurate prediction results. For example, the accurate prediction result may include a prediction result in which the duration between the prediction moment and the reporting moment is less than or equal to the duration threshold, and the inaccurate prediction result may include a prediction result in which the duration between the prediction moment and the reporting moment is greater than the duration threshold. Among them, the reporting moment is the sending moment of the first measurement report, and the predicted moment is the moment corresponding to the prediction result generated by the first AI function based on the measured information. For example, when the first AI function can predict the channel information at a certain moment in the future based on the measurement results of one or more RSs, the predicted moment corresponding to the predicted channel information can be understood as the future moment.

[0181] Exemplarily, Figure 2a is a schematic diagram for determining whether a prediction result is accurate, provided by an embodiment of the present application. As shown in Figure 2a, the first measurement report contains four prediction results. Among them, T1, T2, T3, and T4 are the prediction moments corresponding to prediction result 1, prediction result 2, prediction result 3, and prediction result 4, respectively. Ts is the sending time of the first measurement report, and the duration between Ts and T0 is a preset duration threshold. Among them, if the duration between T1, T2 and Ts is less than the duration threshold, prediction result 1 and prediction result 2 are considered to be accurate prediction results, and if the duration between T3, T4 and Ts is greater than the duration threshold, prediction result 3 and prediction result 4 are considered to be inaccurate prediction results. Therefore, after the terminal device generates the first measurement report based on the first AI function, it can identify in the first measurement report that prediction result 1 and prediction result 2 are reliable results, and prediction result 3 and prediction result 4 are unreliable results.

[0182] Optionally, a predefined field may be used to replace an inaccurate result in the first measurement report, so as to facilitate the network device to identify reliable information in the first measurement report and improve efficiency.

[0183] In this example, even when some configurations of the first AI function are not applicable, some accurate information can still be reported to the network device, thereby improving the robustness of this technical solution.

[0184] As an example, the first measurement report is a measurement report generated based on a non-AI function. For example, in CSI feedback, the measurement report generated by the non-AI function may be a report determined based on the R15 to R18 codebook. For example, in beam management, the measurement report generated by the non-AI function may be a report determined by actual measurement (for example, traversing all beams). For example, in positioning, the measurement report generated by the non-AI function may be a report determined by traditional positioning methods such as round trip time (RTT) and time difference of arrival (TDOA). For example, in mobility management, the report generated by the non-AI function may be a report determined by actual measurement. This application does not limit the specific method of the non-AI function.

[0185] It should be understood that the four first measurement reports sent by the terminal device in this embodiment correspond to the four default reporting methods of the terminal device.

[0186] In this embodiment, when the terminal device determines that the first AI function is not applicable, it can send a first request message to the network device to request the network device to deactivate the first AI function, or switch the first AI function to other applicable AI functions, and send a first measurement report obtained based on a predefined default reporting method to the network device, so that the network device can quickly identify whether the first measurement report is a reliable measurement report, thereby improving efficiency; when using the first AI function to generate a measurement report, the content in the first measurement report is distinguished, thereby avoiding the terminal device using an inapplicable AI function to send an inaccurate measurement report to the network device, thereby avoiding power consumption of the terminal device and waste of AI resources.

[0187] Figure 3 is a schematic flow chart of a communication method provided in Embodiment 2 of the present application. Figure 3 may include S210 and S220, and may also include S201, S202, and S203.

[0188] S201, sending AI capability information of the terminal device.

[0189] In this embodiment, the AI ​​capability information includes AI features and applicable conditions supported by the configuration under the corresponding feature.

[0190] Among them, AI features mainly refer to AI features or use cases supported by the protocol, such as AI CSI compression feedback features in the spatial-frequency domain, AI CSI prediction features, AI spatial-domain beam prediction features, AI time-domain beam prediction features, AI direct positioning features, and AI indirect positioning features.

[0191] Each AI feature or configuration under each AI feature can have corresponding applicable condition information. For example, when a terminal device implements a certain AI feature or implements measurement / reporting under a certain AI feature, it needs to use the input conditions, output conditions, applicable scenarios, and other dependent information.

[0192] As an example, in the time domain prediction related features (such as AI time domain beam prediction or AI CSI prediction), the applicable conditions may include: the number of supported input measurement samples, the number of supported inference output samples, the supported input sample time interval, the supported output sample time interval, the supported predicted time of the inference output sample, the supported speed (such as speed range, speed value), the supported scene environment (such as indoor, outdoor, dense, sparse, etc.), the supported input sample types (such as RSRP, SINR, RSRQ, CIR or PDP and other signal measurement values, channel information, RS identity (ID), etc.), the supported output sample types (such as RSRP, SINR, CIR or PDP and other predicted signal measurement values, predicted channel information (such as precoding matrix indicator (PMI), rank indicator (RI), channel quality indicator (CQI), etc.), the predicted RS ID, etc.), supported measurement beam patterns, the association between the measurement set and the prediction set, supported preprocessing methods, supported post-processing methods, supported measurement reference signal types, supported prediction reference signal types, supported number of measurement ports for reference signals, supported time-frequency domain configurations of reference signals, etc.

[0193] For example, Figure 3a is a schematic diagram illustrating the applicable conditions of an AI feature provided in one embodiment of the present application. As shown in Figure 3a, the AI ​​feature supports N input samples, an input sample interval of X ms (milliseconds), and RS input samples; it supports M output samples, an output sample interval of Y ms, and the output samples are channel information.

[0194] As an example, in the relevant features of spatial prediction (such as AI spatial beam prediction), the applicable conditions may include: the size of the measurement set, the size of the prediction set, the supported measurement reference signal type, the supported prediction reference signal type, the association relationship between the measurement set and the prediction set, the measured beam mode (pattern), the supported input sample type (such as RSRP, SINR, RSRQ, CIR or PDP signal measurement values, channel information, RS ID, etc.), the supported output sample type (such as RSRP, SINR, RSRQ, CIR or PDP, etc. predicted signal measurement values, predicted channel information (such as PMI, RI, CQI, etc.), predicted RS ID, etc.), supported preprocessing methods, supported post-processing methods, supported number of reference signal measurement ports, supported time-frequency domain configuration of reference signals, etc.

[0195] As an example, in the related features of spatial domain compression (such as spatial-frequency domain AI CSI feedback compression), the applicable conditions may include: supported input sample types (such as RSRP, SINR, RSRQ, CIR or PDP signal measurement values, channel information, RS ID, etc.), channel characteristic vectors, etc., supported measurement reference signal types, supported number of measurement ports of reference signals, supported time-frequency domain configuration of reference signals, supported output sample types (such as RSRP, SINR, CIR or PDP predicted signal measurement values, channel information (such as PMI, RI, CQI, etc.), predicted RS ID, etc.), supported quantization methods (such as vector quantization, scalar quantization, etc.), supported number of quantization bits, supported preprocessing methods, supported post-processing methods, etc.

[0196] As an example, in the relevant features of positioning, the applicable conditions may include: the types of supported input samples, such as signal measurement values ​​such as CIR or PDP; the types of supported output samples, such as LOS identifier, non-line of sight (NLOS) identifier, time of arrival (TOA) information, location information, etc., the number of sampling points of supported input samples, the number of truncated sampling points of supported input samples, the sampling rate of supported input samples, the number of transmit receive points (TRP) supported for measurement, the supported timing error range, the supported synchronization error range, the supported number of measurement ports of reference signals, the supported time-frequency domain configuration of reference signals, etc.

[0197] Optionally, the AI ​​capability information can be carried in existing UE capability information (UE_CAP_INFO) and other signaling, and sent to the network device together with other UE capability information after the terminal device accesses the network, for example, through non-access stratum (NAS) signaling.

[0198] Optionally, a separate dedicated AI capability signaling similar to UE_CAP_INFO signaling, such as UE_AI_CAP_INFO, may be configured for the AI ​​capability information to send the AI ​​capability information.

[0199] S202: Receive the AI ​​function configured by the network device.

[0200] In this embodiment, the network device may configure a group of AI functions for each AI feature based on the AI ​​capability information reported by the terminal device, and activate one of the AI ​​functions in the group of AI functions.

[0201] When configuring each AI feature, network devices can combine multiple AI functions based on the applicable conditions supported by each AI feature. Therefore, multiple AI functions can be configured under the same AI feature.

[0202] For example, Figure 3b is a schematic diagram of configuring an AI function provided by an embodiment of the present application. As shown in (a) in Figure 3b, assuming that the applicable conditions supported by the AI ​​feature include: the number of supported input measurement samples is 1, 2 or 4, the number of supported inference output samples is 1 or 2, and the time interval of supported output samples is 2ms or 4ms, then the AI ​​function configured by the network device for the AI ​​feature can be a combination of the above three types of applicable conditions. As shown in (b) in Figure 3b, the AI ​​functions that can be configured under the AI ​​feature include {1,1,2ms}, {1,1,4ms}, ..., {1,2,2ms}, etc. Among them, {1,2,2ms} can be understood as an AI function that inputs 1 measurement sample, outputs 2 inference samples, and the time interval between the two output inference samples is 2ms. The "Number of supported input measurement samples" parameter in this applicable condition indicates that when a terminal device implements this AI feature, it requires one, two, or four measurement results as input samples. Therefore, the network device can configure one or more reference signals for the terminal device so that the terminal device can complete one, two, or four measurements, thereby enabling the terminal device to implement the input of this AI feature. The "Number of supported inference output samples" parameter in this applicable condition indicates that when a terminal device implements this AI feature, it can obtain one or two output results obtained through AI inference. Therefore, the network device can configure one or more reporting configurations for the terminal device so that the terminal device can report one or two prediction results obtained through AI inference to the network device. The "Time interval of supported output samples" parameter in this applicable condition indicates that when a terminal device implements this AI feature, the interval between inference output results is 2ms or 4ms. Therefore, in the reporting configuration configured for the terminal device, the network device can configure the interval between predicted reporting results or the prediction time of each prediction result to be 2ms or 4ms.

[0203] Optionally, the applicable conditions supported by the AI ​​feature may also include a supported input sample interval, such as 2ms or 4ms. The input sample interval parameter indicates that when a terminal device implements the AI ​​feature, the interval between measurement results or measurement signals may be 2ms or 4ms. Therefore, when a network device configures a reference signal for the terminal device, the interval between reference signals may be configured to be 2ms or 4ms.

[0204] In one achievable manner, configuration of the AI ​​function may be manifested as configuration of AI-related measurement configuration, such as configuration of AI-related measurement resource configuration and / or reporting configuration.

[0205] As an example, AI-related configurations can be identified in the measurement resource configuration and / or reporting configuration in a certain measurement configuration, so that when the terminal device uses the measurement configuration, it can determine the AI-related measurement resource configuration and / or reporting configuration based on the identification information to implement the configuration of the AI ​​function, thereby enabling AI-based features such as AI CSI feedback, AI beam management, AI positioning, etc.

[0206] As an example, the identification information in the measurement resource configuration or reporting configuration may be a Boolean value, and when the Boolean value is true, it indicates that the measurement resource configuration or reporting configuration can be used for AI. Among them, the AI ​​reporting amount can also be used in the reporting configuration to indicate that the reporting configuration can be applied to AI. For example, in the reporting configuration of CSI feedback, the reporting amount can be the compressed result or index of the PMI or CSI using AI; in the reporting configuration of beam management, the reporting amount can be the beam index or predicted measurement result using AI, and the predicted measurement result can include the predicted beam quality (such as predicted RSRP). In the reporting configuration of positioning, the reporting amount may include the predicted positioning result (such as predicted location information), or the predicted positioning parameters (such as the predicted LOS identifier, TOA information, etc.).

[0207] For example, Figure 3c is a schematic diagram of a configuration AI function provided by an embodiment of the present application. The CSI measurement configuration shown in (a) of Figure 3c includes 2 CSI measurement resource configurations and 1 CSI reporting configuration, and identification information is used in each CSI measurement resource configuration and CSI reporting configuration to indicate whether the configuration is related to AI. Among them, the Boolean value in CSI measurement resource configuration 1 is true, and CSI measurement resource configuration 1 contains multiple channel state information reference signal (CSIRS) resources, which means that the measurement results of multiple CSIRS resources in CSI measurement resource configuration 1 can be used for AI, the reporting amount in CSI reporting configuration 1 is AIPMI, and the Boolean value in CSI reporting configuration 1 is true, which means that CSI reporting configuration 1 is related to AI.

[0208] Referring to (a) in Figure 3c, if the Boolean values ​​in CSI measurement resource configuration 1 and CSI measurement resource configuration 2 are both true, and the reporting amount of CSI reporting configuration 1 is AIPMI, the terminal device can generate a CSI measurement configuration for the AI ​​function based on CSI measurement resource configuration 1, CSI measurement resource configuration 2, and CSI reporting configuration 1 to enable AI CSI feedback. It should be understood that as long as the CSI measurement configuration contains identification information indicating that the measurement resource configuration or reporting configuration can be used for AI, it can be considered that the measurement configuration implements the configuration of the AI ​​function.

[0209] In one possible implementation, the corresponding measurement configuration, such as measurement resource configuration and reporting configuration, can be directly configured for the AI ​​function. As shown in (b) of Figure 3c, the AI ​​CSI measurement configuration includes AI CSI measurement resource configuration and AI CSI reporting configuration.

[0210] It should be understood that the measurement resource configuration and reporting configuration in this implementation can both be used for AI. Therefore, when using this approach to implement AI function configuration, the measurement resource configuration and / or reporting configuration may not include identification information indicating AI-related information.

[0211] Optionally, in the above configuration, the identifier of the AI ​​model or AI function can be further configured or associated.

[0212] Optionally, in the above configuration, an association identifier of the AI ​​model or AI function can be further configured. The association identifier can be used to indicate an additional condition on the network side. For example, in beam management, the additional conditions on the network side may include parameters, shapes, etc. of the beam of the network device, which are not limited here. The association identifier can correspond to the additional conditions on the network side, that is, when the association identifiers are the same, the terminal device can assume that the additional conditions on the network side are the same. It should be understood that the terminal device can determine the additional conditions on the network side based on the association identifier of the AI ​​model or AI function, thereby determining the AI ​​model or AI function that the terminal device can train / use.

[0213] For example, in AI beam management, multiple CSIRS resources can be configured to form an AI CSI measurement resource configuration, and the reporting amount of the AI ​​CSI reporting configuration can be configured to be None (i.e., no reporting is required), and an association identifier #1 can be configured in the AI ​​CSI measurement configuration, indicating that the measurement results under the AI ​​CSI measurement resource configuration can be used for training, reasoning, or monitoring of AI function #1 or AI model #1.

[0214] Optionally, the identifier of the AI ​​model or AI function, or the associated identifier of the AI ​​model or AI function, may be global (e.g., globally unique) or local (e.g., valid only within a certain region).

[0215] In this embodiment, the measurement resource configuration and the reporting configuration can be combined into a configuration of an AI function. Specifically, the configuration of the AI ​​function can be achieved by associating a set of measurement resource configurations and reporting configurations. For example, the purpose of the AI ​​function is to predict the channel information once in the future (such as 10ms later) based on the measurement results of 3 historical CSIRS (the measurement time of each CSIRS can be the same or different). Among them, the measurement resource configuration of the AI ​​function includes 3 CSIRS resources, each CSIRS resource has different or the same time-frequency domain, port and other configurations, the reporting configuration is channel information, and further indicates that the channel information is the channel information reported 10ms later. Therefore, the terminal device can obtain the channel information 10ms later by measuring the measurement results of the 3 CSIRS resources historically, so there is no need to measure the CSIRS after 10ms to determine the channel information after 10ms.

[0216] Optionally, the same AI function may correspond to multiple measurement resource configurations or multiple reporting configurations, which is not specifically limited in this application.

[0217] As an example, the network device can configure the AI ​​function and / or activate the AI ​​function for the terminal device through at least one of RRC signaling, MAC-CE signaling, downlink control information (DCI) or NAS signaling. Taking RRC signaling as an example, the network device can add the corresponding field of the measurement resource configuration of the AI ​​function in the measurement resource configuration of the RS to identify that the RS resource can be used for the input of the AI ​​function; or, add the corresponding field of the reporting configuration of the AI ​​function in the reporting configuration of the CSI or positioning to identify that the report is associated with the AI ​​function or that the report needs to be determined based on the AI ​​function. Therefore, the network device can carry the information for configuring the AI ​​function and the information for activating the AI ​​function in the above-mentioned signaling, and send it to the terminal device. Accordingly, the terminal device receives the configuration information and activation information of the AI ​​function.

[0218] Optionally, the configuration and activation of the AI ​​function can be implemented through one signaling or through a combination of multiple signalings, and this application does not impose any specific restrictions on this. For example, the network device configures a group of AI functions for an AI feature of the terminal device through RRC signaling, selects a subset of the group of AI functions through MAC-CE signaling, and dynamically activates one or more AI functions in the subset through DCI signaling. For another example, the network device configures a group of AI functions for an AI feature of the terminal device through RRC signaling, and activates one or more AI functions in the group of AI functions through MAC-CE signaling. For another example, the network device configures a group of AI functions for an AI feature of the terminal device through RRC signaling, and activates one or more AI functions in the group of AI functions through RRC signaling.

[0219] Optionally, the configuration and activation of the AI ​​function can be simultaneous. For example, when configuring the AI ​​function, one of the AI ​​functions is activated by default. The AI ​​function activated by default can be determined by a default rule between the network device and the terminal device (such as the maximum or minimum value of the ID of the AI ​​function). For another example, the AI ​​function is configured through a first RRC signaling and activated through a second RRC signaling. The first RRC signaling and the second RRC signaling can be the same RRC signaling or different RRC signaling.

[0220] Optionally, one or more AI functions may be activated simultaneously.

[0221] In this embodiment, the terminal device can perform measurement based on the measurement resource configuration in the measurement configuration corresponding to the activated AI function, and report based on the reporting configuration in the measurement configuration corresponding to the AI ​​function.

[0222] S210: Send first request information, where the first request information is used to request the network device to switch or deactivate a first AI function.

[0223] Optionally, the judgment conditions for determining whether the AI ​​function is applicable, configured by the network device for the terminal device, can be carried in the configuration information of the AI ​​function and delivered together.

[0224] S220: Send a first measurement report.

[0225] The specific implementation of S210 and S220 can refer to the relevant content in Example 1 and will not be repeated here.

[0226] S203: Receive response information sent by the network device, where the response information is used to instruct the terminal device to deactivate the first AI function or switch the first AI function to the second AI function.

[0227] In this embodiment, after receiving the first request information, the network device may respond to the first request information and send response information to the terminal device. Correspondingly, the terminal device receives the response information.

[0228] Optionally, the response information may include an instruction to deactivate the first AI function, or an instruction to switch the first AI function to a second AI function. After receiving the response information, the terminal device no longer uses a predefined default reporting method to report the first measurement report, or the terminal device reports the measurement report according to the measurement resource configuration and reporting configuration corresponding to the second AI function indicated in the response information.

[0229] Optionally, the response information may include an instruction to deactivate the first AI function, and may also include an instruction to activate other AI functions, so that the terminal device can report the measurement report based on the measurement resource configuration and reporting configuration corresponding to the other AI functions.

[0230] Optionally, the response information can be carried in the DCI information for indication. For example, a dedicated field is set in the DCI information or a reserved field in an existing field in the DCI information is used to carry the identifier of the first AI function to be deactivated, or a dedicated field is set in the DCI or a reserved field in an existing field is used to carry a deactivation indication, indicating that all activated AI functions in the current terminal device are deactivated. For another example, a dedicated field is set in the DCI information or a reserved field in an existing field is used to carry the identifier of the second AI function to be switched to. Optionally, the identifier of the first AI function to be switched can also be carried to indicate switching from the first AI function to the second AI function.

[0231] Optionally, the response information can be carried in the MAC-CE information for indication. For example, a MAC-CE for deactivation is used, and the MAC-CE header or data unit indicates that the MAC-CE is a deactivation indication, and the data unit of the MAC-CE carries the identifier of the first AI function to be deactivated. For another example, a MAC-CE for AI function switching is used, and the MAC-CE header or data unit indicates that the MAC-CE is an AI function switching indication, and the data unit of the MAC-CE carries the identifier of the second AI function to be switched to. Optionally, the identifier of the first AI function currently to be switched can also be carried to indicate switching from the first AI function to the second AI function.

[0232] Optionally, the response information can be carried in RRC signaling for indication. For example, RRC signaling for deactivation is used, and the signaling carries the identifier of the first AI function to be deactivated. For another example, RRC signaling for AI function switching is used, and the RRC signaling carries the identifier of the second AI function to be switched to. Optionally, the RRC signaling can also carry the identifier of the first AI function currently to be switched to indicate switching from the first AI function to the second AI function.

[0233] It should be understood that if the network device does not receive the first request information or does not respond to the first request information, the network device will not send a response information to the terminal device, causing the first AI function to remain activated for a long time. The terminal device may not be suitable for using the first AI function, resulting in the first AI function occupying the terminal device's AI resources for a long time, resulting in resource waste. If the network device takes too long to send a response information, it will also cause a certain degree of resource waste.

[0234] Therefore, a preset time duration can be agreed upon between the terminal device and the network device. After the preset time duration has expired, the terminal device no longer needs to wait for the network device to send a response message, but can adopt autonomous decision-making or fallback to handle the failure of the first AI function to save resources.

[0235] Based on the above description, Figure 4 is a schematic flow chart of a communication method provided in Embodiment 3 of the present application. As shown in Figure 4 , the method may include S410 to S450.

[0236] S410: Send first indication information, where the first indication information is used to indicate the maximum value of the preset duration supported by the terminal device.

[0237] In this embodiment, the maximum value of the preset duration can be understood as the maximum delay time that the terminal device expects the network device to respond to the terminal device deactivating / switching the first AI function, or the maximum time that the terminal device is allowed to wait for the network device to switch / deactivate the first AI function after sending the first request information.

[0238] Optionally, the first indication information may be sent to the network device together with the AI ​​capability information of the terminal device. The specific implementation of reporting the AI ​​capability information may be referred to in S201 and will not be described in detail here.

[0239] Optionally, the first indication information may be sent to the network device separately, for example, after the terminal device receives the AI ​​function configured by the network device, it may be sent to the network device.

[0240] Optionally, the maximum value of the preset duration corresponding to all AI functions in the terminal device may be consistent or inconsistent, and this application does not impose any specific restrictions on this.

[0241] S420: Receive first configuration information, where the first configuration information is used to indicate a preset duration.

[0242] In this embodiment, after receiving the first indication information, the network device may configure a timer for each AI function in the terminal device based on the first indication information. The timer duration is used to indicate the maximum response time between the network device receiving the first request information and issuing the response information. The timer duration may also be referred to as a preset duration. The start time of the timer is the start time of the preset duration, and the stop time of the timer is the end time of the preset duration.

[0243] The first configuration information is also used to instruct the terminal device to autonomously determine the deactivation operation of the first AI function after the timer stops.

[0244] Optionally, one AI function may correspond to one or more timers, and one timer may correspond to one or more AI functions.

[0245] Optionally, the first configuration information may be sent to the terminal device together with the AI ​​function configured by the network device for the terminal device. The specific implementation of the network device configuring the AI ​​function for the terminal device can be found in S202 and will not be described in detail here.

[0246] Optionally, the first configuration information may also be sent to the terminal device separately, and this application does not make any specific limitation on this.

[0247] Optionally, the timer duration can be bound to the measurement configuration corresponding to each AI function, such as the measurement resource configuration or reporting configuration, or it can be configured separately and associated with the corresponding AI function. For example, the timer duration can be configured in the measurement resource configuration of the AI ​​function, or in the reporting configuration of the AI ​​function. For another example, the timer duration can be configured in the AI ​​function configuration. For another example, after the AI ​​function is configured, the timer can be configured and associated with the AI ​​function.

[0248] Optionally, the unit of the timer duration can be a time unit, such as hours, minutes, seconds, milliseconds, microseconds, nanoseconds, etc., or a time unit of the communication system, such as time slots, symbols, frames, subframes, and time units Tc and Ts, etc. This application does not impose any restrictions on this.

[0249] Optionally, the activation of a timer configured for an AI function on a network device can be synchronized with the activation of the AI ​​function. For example, when a network device activates an AI function, the timer corresponding to the AI ​​function is also activated. It should be noted that activating a timer is different from starting a timer; a timer can only start running when it is in the activated state.

[0250] Optionally, the timer may be predefined by the protocol. In this case, S410 and S420 may not be executed, and S430, S440 and S450 may be directly executed.

[0251] Optionally, if the timer is predefined by the protocol, the protocol may define a unified duration for all AI functions in the terminal device, or may define different durations for different AI functions.

[0252] Optionally, if the timer is predefined by the protocol, the timer can be used only for AI functions that meet certain conditions. For example, the timer can be used only when the number of RSs measured by the AI ​​function meets a first threshold, or when the number of prediction results output by the AI ​​function meets a second threshold. The first and second thresholds can be set according to actual needs. For AI functions that do not meet the conditions, the timer is considered not to be used.

[0253] Optionally, when the network device configures the timer in an autonomously determined manner instead of based on the first indication information, S410 may not be executed, and S420, S430, S440 and S450 may be directly executed.

[0254] S430: Send first request information, where the first request information is used to request the network device to switch or deactivate the first AI function.

[0255] In this embodiment, optionally, after sending the first request information, the terminal device may immediately start the timer corresponding to the first AI function, or start the timer corresponding to the first AI function after a period of time after the first request information is sent. This application does not impose any specific restrictions on this.

[0256] Optionally, the offset duration may be pre-defined by the protocol, may be configured by the network device, or may be determined by the network device based on the expected offset duration reported by the terminal device. This application does not impose any specific restrictions on this.

[0257] It should be understood that the duration of the timer is less than or equal to the duration between the time when the first request information is sent and the time when the terminal device receives the response information.

[0258] In addition to the above description, the specific implementation of S430 can be found in S210 and will not be repeated here.

[0259] S440: Send a first measurement report.

[0260] In this embodiment, the terminal device sends the first measurement report using the default reporting method described in S220 before the timer stops, which is not repeated here.

[0261] S450: Deactivate the first AI function and / or send a second measurement report.

[0262] In this embodiment, after the timer expires, the terminal device uses an autonomous decision-making method to deactivate the first AI function or uses a fallback method to report the second measurement report, without the network device instructing the deactivation or fallback. The second measurement report can be determined based on other AI functions or based on traditional or non-AI methods, which is not limited in this application.

[0263] Optionally, after the terminal device deactivates the first AI function, the measurement configuration of the first AI function, such as the measurement resource configuration and / or the reporting configuration, can be synchronously released to improve the utilization of AI resources. If the measurement resource configuration or reporting configuration corresponding to the first AI function is also associated with other AI functions that have not expired, the measurement resource configuration or reporting configuration corresponding to the first AI function will not be released. For example, if the measurement resource configuration of the first AI function is the first measurement resource configuration, the reporting configuration is the first reporting configuration, and the measurement resource configuration of the second AI function is also the first measurement resource configuration, and the reporting configuration is the second reporting configuration, then when the first AI function is deactivated and the second AI function is still activated, only the first reporting configuration is released without releasing the first measurement resource configuration.

[0264] Optionally, the network device and the terminal device may determine, through protocol pre-definition or network device configuration, whether the terminal device can use a fallback method (such as falling back to a traditional method) to replace the first AI function. The traditional method is a non-AI function method.

[0265] In this embodiment, by configuring a timer for the AI ​​function, after the timer stops, the terminal device no longer needs to wait for the network device to send a response message, but can use autonomous decision-making or fallback to handle the failure of the first AI function, thereby avoiding waste of AI resources and power consumption of the terminal device.

[0266] In one possible method, if the terminal device receives a response message sent by the network device before the timer stops, the terminal device deactivates the first AI function according to the instruction of the response information, or switches to the second AI function, and generates a second measurement report according to other AI functions or other reporting methods indicated in the response information, and sends it to the network device.

[0267] Based on the methods shown in Example 1 and Example 2, when the first AI function fails, the terminal device may also use the default AI function shown in Example 4 to generate a first measurement report and send it to the network device.

[0268] For example, Figure 5 is a schematic flow chart of a communication method provided in Embodiment 4 of the present application. As shown in Figure 5 , the method may include S510 to S550.

[0269] S510: Send second indication information, where the second indication information includes combination information of applicable conditions supported by the AI ​​characteristics of the terminal device and / or a measurement information reporting method corresponding to the combination information of applicable conditions.

[0270] In the embodiments of the present application, the combination information of applicable conditions can be understood as the applicable conditions that the AI ​​feature needs to support when configuring a robust or generalized AI function. It should be understood that the combination information of applicable conditions can include different types of applicable conditions, such as supported input samples, supported output samples, etc. The applicable conditions under each type can include one or more applicable conditions. For example, the applicable conditions of the supported input sample type can include applicable conditions that support one input sample, or can include applicable conditions that support one or two input samples.

[0271] Optionally, the second indication information may be sent to the network device together with the AI ​​capability information of the terminal device. The specific implementation of reporting the AI ​​capability information may be referred to in S201 and will not be described in detail here.

[0272] As an example, when reporting AI capability information, the AI ​​feature's supported applicable conditions can explicitly indicate the combination of applicable conditions supported by the AI ​​feature. For example, in the AI ​​spatial beam prediction feature, the applicable conditions supported by the AI ​​feature include: the number of beams supported for measurement is [2, 4, 8], and the range of the number of beams in the predicted beam set is [16, 64]. In this example, the number of beams measured is 2, and the range of the number of beams in the predicted beam set is 16 as the combined applicable condition information.

[0273] It should be understood that the applicable conditions supported by the AI ​​feature may include combined information of the applicable conditions supported by the AI ​​feature, that is, the combined information of the applicable conditions may be a subset of the applicable conditions reported by the terminal device.

[0274] Optionally, the applicable conditions supported by the AI ​​feature may be reported in the first list, and the combination information of the applicable conditions supported by the AI ​​feature may be reported in the second list, thereby explicitly indicating the combination information.

[0275] Optionally, the applicable conditions supported by the AI ​​feature may be reported in the first list. Furthermore, the applicable conditions in the combined information of the applicable conditions supported by the AI ​​feature may be reported in the first list. Optionally, an identifier may be set before the combined information to identify the portion as combined information.

[0276] Optionally, the combination information of the applicable conditions supported by the AI ​​feature may be identified in the reported applicable conditions supported by the AI ​​feature, thereby explicitly indicating the combination information.

[0277] Taking the AI ​​time-domain prediction beam feature as an example, Figure 6 is a schematic diagram illustrating a combination of information that explicitly indicates applicable conditions, provided by an embodiment of the present application. (a) in Figure 6 shows the applicable conditions supported by the AI ​​feature reported by the terminal device, including the supported input sample time window of [5ms, 10ms], the supported number of time-domain input measurement samples of [1, 2, 4], and the supported number of time-domain inference output measurement samples of [1, 2].

[0278] As shown in Figure 6(b), among the applicable conditions supported by the AI ​​feature, identification information is added to the three conditions (e.g., 5ms in the time window, 1 in the number of input measurement samples, and 1 in the number of inference output measurement samples) to identify these three conditions as applicable conditions in the combined information of applicable conditions supported by the AI ​​feature. Optionally, the identification information can be 1 bit long. Optionally, the identification information can be a Boolean representation of true / false. For easier distinction, in Figure 6(b), the 5ms in the time window, the 1 in the number of input measurement samples, and the 1 in the number of inference output measurement samples are bolded.

[0279] As another example, when reporting AI capability information, the applicable conditions in the combination information of the applicable conditions supported by the AI ​​feature can be indicated implicitly or according to predefined rules in the applicable conditions supported by the AI ​​feature. For example, in the AI ​​spatial beam prediction feature, the applicable conditions supported by the AI ​​feature include: the number of beams supported for measurement is [2, 4, 8], and the range of beam sets supported for prediction is [16, 64]. The combination information of the applicable conditions supported by the AI ​​feature can be a parameter at a specific position in each type of applicable condition (such as the first parameter), such as using 2 in the number of measured beams and 16 in the range of predicted beam sets as the combination information of the applicable conditions in this example, thereby implicitly indicating the combination information.

[0280] Optionally, the applicable conditions in the combination information of applicable conditions supported by the AI ​​feature can be determined in a manner predefined by the protocol. For example, the condition with the lowest or most relaxed capability requirement among the applicable conditions supported by the AI ​​feature can be used as the combination information, such as the condition with the smallest parameter value among all applicable conditions as the applicable condition in the combination information of applicable conditions.

[0281] Optionally, the AI ​​feature may support multiple combinations of applicable conditions. For example, each type of applicable condition supported by the AI ​​feature may include more than one condition that can be combined. For example, the number of beams supported for measurement in the AI ​​spatial beam prediction feature may include two and four conditions that can be combined.

[0282] Optionally, after determining the combination information of the applicable conditions supported by the AI ​​feature, it can be determined based on the combination information that when the activated AI function in the terminal device is not applicable, the corresponding configuration of the default AI function can be used as a fallback solution, so that when the first AI function fails, the terminal device can use the default AI function to generate and report a first measurement report. The corresponding configuration of the default AI function can be referred to as the configuration of the measurement information reporting method corresponding to the combination information of the applicable conditions supported by the AI ​​feature of the terminal device.

[0283] Optionally, the identifier of the measurement information reporting mode may be an identifier of a default AI function.

[0284] Therefore, the second indication information may also include the corresponding configuration of the default AI function as fallback, as shown in (c) in Figure 6.

[0285] Optionally, the second indication information may also be sent to the network device separately, and this application does not impose any specific restrictions on this.

[0286] It should be understood that the generalization or robustness of the default AI function determined by the combination of applicable conditions supported by the AI ​​feature is stronger than the AI ​​function activated in the terminal device. In the case where the activated AI function is not applicable, the default AI function can still be applied.

[0287] S520: Receive second configuration information, where the second configuration information is used to indicate a measurement information reporting method used by the terminal device when the first AI function fails.

[0288] In this embodiment, after receiving the second indication information, the network device may configure a default AI function for at least one AI function in the terminal device according to the second indication information, and activate the corresponding default AI function.

[0289] Optionally, the second configuration information can be sent to the terminal device together with the AI ​​function configured by the network device for the terminal device, and when a certain AI function is activated, the default AI function corresponding to the AI ​​function is enabled at the same time or after a period of time. The interval time can be set according to actual needs.

[0290] In this embodiment, to distinguish it from the default AI function, the AI ​​function configured by the network device for the terminal device is referred to as the normal AI function. The enabling method of the default AI function is the same as the activation method of the normal AI function in S202, and will not be repeated here.

[0291] It should be noted that the AI ​​function that is the default AI function can also be used as one of the configured normal AI functions. For example, a network device configures and activates a first AI function and a second AI function for a terminal device, and uses the first AI function as the default AI function for the second AI function. When the second AI function fails, the terminal device performs corresponding measurement reporting based on the configuration of the first AI function. For example, at the time of reporting the second AI function, the terminal device reports a first measurement report determined based on the first AI function. This does not affect the reporting of the first AI function, which is originally configured as a normal AI function by the terminal device.

[0292] Optionally, the normal AI function may be associated with one or more default AI functions. When the normal AI function fails, the specific default AI function to be used may be configured by the network device or determined autonomously by the terminal device.

[0293] Optionally, the normal AI function may semi-statically or dynamically adjust the default AI function associated with the normal AI function. For example, the default AI function associated with the normal AI function may be reconfigured via RRC signaling, or the default AI function associated with the normal AI function may be selected via MAC-CE signaling; or the default AI function associated with the normal AI function may be selected via DCI dynamic signaling.

[0294] S530: Send first request information, where the first request information is used to request the network device to switch or deactivate the first AI function.

[0295] Optionally, when the first AI function is associated with multiple default AI functions and the second configuration information does not indicate which default AI function to use when the first AI function fails, the terminal device can carry the identifier of the default AI function to be used, or the identifier of the default AI function selected by the terminal device independently, in the first request information and send it to the network device.

[0296] The specific implementation method of sending the first request information can be found in S210 and will not be repeated here.

[0297] S540: Send a first measurement report.

[0298] In this embodiment, the terminal device may generate and send a first measurement report to the network device based on the default AI function when the first AI function indicated in the second configuration information fails.

[0299] Optionally, when the second configuration information does not indicate which default AI function to use when the first AI function fails, and the network device does not respond to the first request information, the terminal device can generate and send a first measurement report based on the autonomously selected default AI function, and indicate the identifier of the default AI function autonomously selected by the terminal device in the first measurement report.

[0300] S550: Receive response information sent by the network device, where the response information is used to instruct the terminal device to deactivate the first AI function or switch the first AI function to the second AI function.

[0301] The specific implementation of S550 can be found in S203 and will not be described in detail here.

[0302] In this embodiment, the default AI function determined based on the combined information of the applicable conditions supported by the AI ​​feature reported by the terminal device is more adapted to the capabilities of the terminal device; and, compared with the default reporting method of implementation one and implementation two, the first measurement report generated using the default AI function is more accurate.

[0303] Optionally, when the terminal device in the third embodiment adopts the fallback method, it can fall back to the default AI function determined in the fourth embodiment, thereby generating and sending a second measurement report based on the default AI function.

[0304] Based on the first, second and fourth embodiments, the fifth embodiment provides a method in which a terminal device generates and sends a first measurement report using an autonomously determined measurement information reporting method.

[0305] For example, Figure 7 is a schematic flow chart of a communication method provided in Embodiment 5 of the present application. As shown in Figure 7 , the method may include S710 to S750.

[0306] S710: Send second indication information, where the second indication information includes a measurement information reporting method supported by the terminal device.

[0307] Optionally, the measurement information reporting method supported by the terminal device can be understood as the measurement information reporting method independently determined by the terminal device, including the default reporting method shown in S220, and also including the default AI function shown in S510.

[0308] Optionally, the second indication information includes measurement information reporting methods supported by the terminal device and identifiers of each measurement information reporting method.

[0309] Optionally, the second indication information further includes AI capability information (such as applicable conditions) corresponding to the default AI function, so that the network device knows how the default AI function is autonomously determined.

[0310] Optionally, the second indication information may be sent to the network device together with the AI ​​capability information of the terminal device. The specific implementation of reporting the AI ​​capability information may be referred to in S201 and will not be described in detail here.

[0311] S720: Receive second configuration information, where the second configuration information is used to indicate a measurement information reporting method used by the terminal device when the first AI function fails.

[0312] In this embodiment, after receiving the second indication information, the network device can determine, based on the second indication information and the implementation of the terminal device, whether the terminal device can use the autonomously determined measurement information reporting method to generate and report the first measurement report.

[0313] Optionally, the network device may set an identifier in the corresponding configuration of the AI ​​function configured for the terminal device to identify whether the terminal device is allowed to generate and report the first measurement report using an autonomously determined measurement information reporting method after the AI ​​function fails.

[0314] S730: Send first request information, where the first request information is used to request the network device to switch or deactivate the first AI function.

[0315] In this embodiment, the specific implementation method of sending the first request information can be found in S210 and will not be repeated here.

[0316] S740: Send a first measurement report.

[0317] In this embodiment, the first measurement report is generated based on the measurement information reporting method supported by the terminal device.

[0318] Optionally, the first measurement report further includes third indication information, where the third indication information is used to indicate the specific measurement information reporting method used by the terminal device when generating the first measurement report.

[0319] Optionally, the third indication information of the terminal device may include an identifier of a specific measurement information reporting method used by the terminal device when generating the first measurement report.

[0320] S750: Receive response information sent by the network device, where the response information is used to instruct the terminal device to deactivate the first AI function or switch the first AI function to the second AI function.

[0321] In this embodiment, the specific implementation of S750 can refer to S550 and will not be repeated here.

[0322] In this embodiment, the terminal device may be allowed to report the first measurement report using an autonomously determined measurement information reporting method when the first AI function fails, so that the terminal device still has the ability to report information within the delay time.

[0323] Based on the above embodiments, embodiment 6 provides a method in which a terminal device has the ability to predict in advance the failure of an activated AI function in the terminal device, so as to request a network device to switch to an unsuitable AI function in advance.

[0324] For example, Figure 8 is a schematic flow chart of a communication method provided in Embodiment 6 of the present application. As shown in Figure 8 , the method may include S810 to S850.

[0325] S810: Send fourth indication information, where the fourth indication information is used to indicate the terminal device's ability to predict AI function failure.

[0326] In this embodiment, the fourth indication information may include at least one of the following information: whether the terminal device has the ability to predict the failure of the AI ​​function, the expected failure time of the AI ​​function in the terminal device, and the confidence level of the expected failure time.

[0327] In this embodiment, AI function failure can be understood as the AI ​​function being unapplicable. The terminal device has the ability to predict AI function failure, which can be understood as the terminal device has the ability to predict that the currently applicable AI function may no longer be applicable within a certain period of time in the future.

[0328] In this embodiment, the expected failure time of the AI ​​function can optionally be the specific failure moment of the AI ​​function. For example, when the terminal device predicts that the third AI function may fail after 5ms, the expected failure time of the third AI function is 5ms. Optionally, the expected failure time of the AI ​​function indicates that the AI ​​function fails after a first duration starting from a first moment. For example, the first moment is the moment when the fourth indication information is sent, or a moment offset from the moment when the fourth indication information is sent, or a predefined moment. The first duration can be a time value such as 5ms or 10ms, or a predefined time value.

[0329] Optionally, the expected failure duration of the AI ​​function may be a fixed duration level corresponding to the AI ​​function. For example, if the terminal device predicts that the third AI function may fail after 5ms, and 5ms falls within the range of the second fixed duration level [5ms, 8ms], then the expected failure duration of the third AI function is the second fixed duration level.

[0330] Optionally, the fixed duration level may be predetermined by the terminal device and the network device through an agreement, or may be configured by the network device to the terminal device, and this application does not impose any specific restrictions on this.

[0331] Optionally, the expected failure time of the AI ​​function may be a characteristic value of a fixed duration level corresponding to the AI ​​function. The characteristic value of the fixed duration level may be the left endpoint, right endpoint, or average value of a fixed duration level interval, and this application does not impose specific restrictions on this. For example, when the terminal device predicts that the third AI function falls within the range of the second fixed duration level [5ms, 8ms], the expected failure time of the third AI function may be 5ms, 8ms, or 6.5ms.

[0332] Optionally, the fourth indication information may further include the expected failure duration of the AI ​​function predicted by the terminal device, and the description method thereof, for example, by a duration level or by a specific duration value.

[0333] In this embodiment, the confidence level of the expected failure time can be understood as the accuracy of the expected failure time, indicating the probability that the AI ​​function actually fails within the expected failure time.

[0334] Optionally, different confidence levels can be set for different expected failure times. For example, if the expected failure time is within 5 seconds, the accuracy of the expected failure time can be 90%; if the expected failure time is within 10 seconds, the accuracy of the expected failure time can be 80%.

[0335] Optionally, the fourth indication information may further include an expected expiration time point, which may be represented by an absolute time (eg Greenwich Mean Time, etc.), or by a frame number, a subframe number, a time slot number, or a symbol number.

[0336] Optionally, the fourth indication information may further include an AI function failure interruption time, which indicates the length of time the AI ​​function will be inoperative from the moment the AI ​​function fails, after which the AI ​​function can be used again. For example, if the interruption time is 10 seconds, the AI ​​function is considered inoperative for 10 seconds from the moment the AI ​​function fails. After 10 seconds, the AI ​​function is no longer inoperative, i.e., the AI ​​function is still usable.

[0337] Optionally, the fourth indication information may be sent to the network device together with the AI ​​capability information of the terminal device. Optionally, the ability to predict failure may be associated with the AI ​​capability information. For example, in the reported AI capability information, it may be explicitly indicated that in the time domain prediction where the length of the input measurement sample is 5 and the length of the output reasoning measurement sample is 3, the accuracy when the expected failure time is 10s is 80%, and the accuracy when the expected failure time is 20s is 50%. Among them, the specific implementation method of reporting the AI ​​capability information can be referred to S201 and will not be repeated here.

[0338] Optionally, the fourth indication information may be sent to the network device after the terminal device receives the AI ​​function configuration issued by the network device. In this case, different confidence levels may be set for the expected failure time of each AI function. For example, if the predicted expected failure time of the third AI function is 80% accurate, the predicted expected failure time of the fourth AI function may be 90% accurate.

[0339] S820: Receive the AI ​​function configured by the network device.

[0340] In this embodiment, the specific implementation method of the network device configuring or activating the AI ​​function for the terminal device can be specifically referred to S202 or S520, which will not be repeated here.

[0341] In this embodiment, if the fourth indication information is sent to the network device together with the AI ​​capability information of the terminal device, the network device can, based on the fourth indication information, increase the expected failure duration corresponding to the AI ​​function in the corresponding configuration of the AI ​​function configured for the terminal device.

[0342] S830: Send first request information, where the first request information is used to request the network device to switch or deactivate the first AI function.

[0343] In this embodiment, when the terminal device predicts that the first AI function may fail, the terminal device sends a first request message to the network device in advance to instruct the network device to switch or deactivate the first AI function in advance to reduce latency.

[0344] Optionally, the first request information may carry an expected expiration time of the first AI function. The starting time of the expected expiration time may be the time when the first request information is sent, or a time with a first offset time from the time when the first request information is sent, and the ending time of the expected expiration time may be the time when the first AI function fails, or a time with a second offset time from the time when the first AI function fails. The first offset time and the second offset time may be set according to actual needs. Alternatively, the first offset time and the second offset time may be predefined values, which are not limited in this application.

[0345] S840: Send a first measurement report.

[0346] In this embodiment, before the first AI function fails, the terminal device uses the first AI function to generate a first measurement report and sends it to the network device; after the first AI function fails and no response information sent by the network device is received, the first measurement report can be determined by referring to the reporting method shown in any of the aforementioned embodiments and sent to the network device.

[0347] S850: Receive response information sent by the network device, where the response information is used to instruct the terminal device to deactivate the first AI function or switch the first AI function to the second AI function.

[0348] In this embodiment, the specific implementation of S850 can refer to S550 and will not be repeated here.

[0349] It should be noted that if the terminal device receives a response message sent by the network device before the first AI function fails, the terminal device will use the AI ​​function indicated by the response message to generate and report the first measurement report after receiving the response message, and will no longer use the first AI function.

[0350] In this embodiment, when the terminal device predicts that the first AI function may fail, it sends a request to the network device in advance to switch / deactivate the first AI function. This allows time for the network device to send a response message, thereby reducing latency and avoiding waste of AI resources and power consumption on the terminal device. In this embodiment, the terminal device may achieve seamless switching from the first AI function to other AI functions.

[0351] Embodiment 7 provides a method for scheduling a sending time of a first measurement report.

[0352] For example, Figure 9 is a schematic flow chart of a communication method provided in Embodiment 7 of the present application. As shown in Figure 9 , the method may include S910 to S950.

[0353] S910: Send fifth indication information, where the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

[0354] In this embodiment, the fifth indication information can be sent to the network device together with the AI ​​capability information of the terminal device. The specific implementation method of reporting the AI ​​capability information can be referred to S201 and will not be repeated here.

[0355] Optionally, the time when the terminal device expects to send the first measurement report may include: the time when the first request information is sent, or sending the first measurement report after receiving the positive confirmation (ACK) or negative confirmation (ACK) issued by the network device in response to the first request information.

[0356] Optionally, the time when the terminal device expects to send the first measurement report may also include an offset value, which is used to start sending the first measurement report after an offset period of time at the time of sending the first request information, or at the time of receiving the ACK / NACK sent by the network device in response to the first request information.

[0357] S920: Receive fourth configuration information, where the fourth configuration information is used to indicate the time when the terminal device sends the first measurement report.

[0358] In this embodiment, after receiving the fifth indication information, the network device may generate fourth configuration information based on the time indicated in the fifth indication information and send it to the terminal device.

[0359] Optionally, the fourth configuration information may be sent to the terminal device together with the AI ​​function configured by the network device for the terminal device. Optionally, the network device may add a time for starting to send the first measurement report configured for the terminal device to the AI ​​function configured for the terminal device.

[0360] Optionally, the moment for sending the first measurement report indicated by the fourth configuration information includes at least one of the following moments: the moment for sending the first request information, the moment when the first threshold condition is satisfied between the moment for sending the first request information, the moment when the ACK or NACK corresponding to the first request information is received, and the moment when the second threshold condition is satisfied between the moment for receiving the ACK or NACK corresponding to the first request information.

[0361] Among them, the specific implementation method of the network device configuring or activating the AI ​​function for the terminal device can be specifically referred to S202 or S520, which will not be repeated here.

[0362] Optionally, when the network device generates the fourth configuration information in an autonomously determined manner instead of based on the time indicated in the fifth indication information, S910 may not be executed, and S920 to S950 may be directly executed.

[0363] S930: Send first request information, where the first request information is used to request the network device to switch or deactivate the first AI function.

[0364] In this embodiment, the specific implementation of S930 can refer to S210 and will not be repeated here.

[0365] S940: Send a first measurement report.

[0366] In this embodiment, after sending the first request information, the terminal device may send the first measurement report according to the time indicated by the fourth configuration information.

[0367] Optionally, after sending the first request information, the terminal device may also send the first measurement report according to the predetermined sending time of the protocol. In this case, the terminal device does not need to send the fifth indication information to the network device, nor does it need to receive the fourth configuration information sent by the network device, so as to save resources.

[0368] As an example, the terminal device sends a first measurement report according to a sending time predefined by the protocol, including at least one of the following situations: if the first request information is a physical layer request (such as UCI), the terminal device may send the first measurement report after sending the first request information, and this method is applicable to a new radio (NR) network system. If the first request information is a physical layer request (such as UCI), the terminal device may send the first measurement report after receiving an uplink ACK indication corresponding to the UCI, and this method is applicable to a long term evolution (LTE) communication system or a future communication system with an uplink hybrid automatic repeat-request (HARQ) indication. If the first request information is a MAC layer / RRC layer request (such as MAC-CE signaling or RRC signaling), the terminal device may send the first measurement report after sending the first request information, or send the first measurement report after receiving an ACK corresponding to the MAC-CE signaling or RRC signaling.

[0369] Optionally, the terminal device may also send the first measurement report after a period of time offset from a protocol-predefined sending time. The offset duration may be determined autonomously by the terminal device and sent to the network device; may be predefined by the protocol, or may be configured by the network device, and this application does not specifically limit this.

[0370] Optionally, the sending time indicated in the fourth configuration information may also include a sending time predefined in the above protocol.

[0371] Optionally, if the first measurement report can be sent at multiple times, the terminal device can send the first measurement report multiple times at multiple sending times.

[0372] Optionally, if there are multiple sending times for the first measurement report, the above-mentioned sending time may be a specific sending time of the first measurement report, such as the sending time of the first measurement report generated by the default AI function.

[0373] S950: Receive response information sent by the network device, where the response information is used to instruct the terminal device to deactivate the first AI function or switch the first AI function to the second AI function.

[0374] In this embodiment, the specific implementation of S950 can refer to S550 and will not be repeated here.

[0375] In this embodiment, by agreeing on the sending time of the first measurement report, the terminal device can send multiple first measurement reports at multiple times without causing the network device to be unable to quickly identify and understand the multiple first measurement reports.

[0376] It should be understood that Examples 3 to 7 should also include the methods shown in S201 and S202.

[0377] The methods shown in the above embodiments are applicable to AI functions and are also applicable to AI models. When applicable to AI models, the relevant configurations / instructions based on AI functions in the above embodiments can be expanded to configurations / instructions based on the granularity of AI models. For example, the preset duration in Example 3 can be expanded to have the same preset duration for different AI models, or different AI models have different preset durations, and each different preset duration is associated with a different model ID.

[0378] For example, Figure 10 is a schematic flow chart of a communication method provided in Embodiment 8 of the present application. As shown in Figure 10 , the method may include S1010 to S1050.

[0379] S1010: Send AI capability information of the terminal device. The AI ​​capability information includes the AI ​​model supported by the terminal device and the ID of the AI ​​model.

[0380] In this embodiment, the AI ​​capability information reported by the terminal device may also carry the specific AI model supported by the terminal device and the ID corresponding to the AI ​​model. The ID corresponding to the AI ​​model may be a globally unique model identifier.

[0381] Among them, the specific implementation method of reporting AI capability information can be found in S201 and will not be repeated here.

[0382] S1020: Receive an AI function configured on a network device, and associate the AI ​​function with a corresponding AI model.

[0383] In this embodiment, the specific implementation method of the network device configuring or activating the AI ​​function for the terminal device can be specifically referred to S202 or S520, which will not be repeated here.

[0384] When configuring AI functions for terminal devices, network devices can also bind corresponding AI models and AI model IDs to each AI function. For example, in AI CSI feedback, for different CSI measurement configurations, such as CSI measurement resource configuration or CSI report configuration, the AI ​​model associated with the CSI measurement resource configuration or CSI report configuration, as well as the AI ​​model ID, is further configured.

[0385] In another possible implementation, the network device does not configure the AI ​​function, but directly configures AI-related measurement configurations, such as measurement resource configuration and reporting configuration, and configures the associated AI model and the ID of the AI ​​model in the measurement resource configuration and / or reporting configuration.

[0386] S1030: Send a first request message, where the first request message is used to request the network device to switch or deactivate the first AI model.

[0387] In this embodiment, when the terminal device determines that the current first AI model is invalid, it sends a first request message to the network device to switch or deactivate the first AI model. The first request message may carry an identifier of the first AI model.

[0388] Among them, the specific implementation method of determining whether the AI ​​model is invalid can be found in S210 and will not be repeated here.

[0389] S1040: Send a first measurement report.

[0390] In this embodiment, after sending the first request information and before receiving the response information sent by the network device, the terminal device can use any reporting method described in the above embodiment to determine the first measurement report and send it to the network device.

[0391] S1050: Receive response information sent by the network device, where the response information is used to instruct the terminal device to deactivate the first AI model or switch the first AI model to the second AI model.

[0392] In this embodiment, after receiving the response information, the terminal device uses the new AI model to report the measurement report.

[0393] In this embodiment, the AI ​​management granularity in the method shown in the above embodiment is refined to the AI ​​model to expand the scope of application of the technical solution of this application.

[0394] Figure 11 is a schematic diagram of the structure of a communication device provided in one embodiment of the present application. The device 1100 shown in Figure 11 can be used to implement the method performed by a terminal device or network device in any of the aforementioned embodiments. As shown in Figure 11, the device 1100 of this embodiment may include: a sending module 1110, a receiving module 1120, and a processing module 1130.

[0395] When the apparatus 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in FIG. 2 , the sending module 1110 may be used to implement the operations performed by the terminal device in S210 and S220 .

[0396] When the apparatus 1100 is used to implement the various steps / operations performed by the network device in the method shown in FIG. 2 , the receiving module 1120 may be used to implement the operations performed by the network device in S210 and S220 .

[0397] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 3, the sending module 1110 can be used to implement the operations performed by the terminal device in S201, S210 and S220, and the receiving module 1120 can be used to implement the operations performed by the terminal device in S202 and S203.

[0398] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 3, the sending module 1110 can be used to implement the operations performed by the network device in S202 and S203, and the receiving module 1120 can be used to implement the operations performed by the network device in S201, S210 and S220.

[0399] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 4, the sending module 1110 can be used to implement the operations performed by the terminal device in S410, S430, S440 and S450, the receiving module 1120 can be used to implement the operation performed by the terminal device in S420, and the processing module 1130 is used to execute S450.

[0400] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 4, the sending module 1110 can be used to implement the operations performed by the network device in S420; the receiving module 1120 can be used to implement the operations performed by the network device in S410, S430, S440 and S450.

[0401] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 5, the sending module 1110 can be used to implement the operations performed by the terminal device in S510, S530 and S540, and the receiving module 1120 can be used to implement the operations performed by the terminal device in S520 and S550.

[0402] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 5, the sending module 1110 can be used to implement the operations performed by the network device in S520 and S550; the receiving module 1120 can be used to implement the operations performed by the network device in S510, S530 and S540.

[0403] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 7, the sending module 1110 can be used to implement the operations performed by the terminal device in S710, S730 and S740, and the receiving module 1120 can be used to implement the operations performed by the terminal device in S720 and S750.

[0404] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 7, the sending module 1110 can be used to implement the operations performed by the network device in S720 and S750; the receiving module 1120 can be used to implement the operations performed by the network device in S710, S730 and S740.

[0405] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 8, the sending module 1110 can be used to implement the operations performed by the terminal device in S810, S830 and S840, and the receiving module 1120 can be used to implement the operations performed by the terminal device in S820 and S850.

[0406] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 8, the sending module 1110 can be used to implement the operations performed by the network device in S820 and S850, and the receiving module 1120 can be used to implement the operations performed by the network device in S810, S830 and S840.

[0407] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 9, the sending module 1110 can be used to implement the operations performed by the terminal device in S910, S930 and S940, and the receiving module 1120 can be used to implement the operations performed by the terminal device in S920 and S950.

[0408] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 9, the sending module 1110 can be used to implement the operations performed by the network device in S920 and S950, and the receiving module 1120 can be used to implement the operations performed by the network device in S910, S930 and S940.

[0409] When the device 1100 is used to implement the various steps / operations performed by the terminal device in the method shown in Figure 10, the sending module 1110 can be used to implement the operations performed by the terminal device in S1010, S1030 and S1040, and the receiving module 1120 can be used to implement the operations performed by the terminal device in S1020 and S1050.

[0410] When the device 1100 is used to implement the various steps / operations performed by the network device in the method shown in Figure 10, the sending module 1110 can be used to implement the operations performed by the network device in S1020 and S1050, and the receiving module 1120 can be used to implement the operations performed by the network device in S1010, S1030 and S1040.

[0411] Figure 12 is a schematic diagram of the structure of a communication device provided by another embodiment of the present application. As shown in Figure 12, the device 1200 is used to implement the method executed by the terminal device or network device in any of the above embodiments.

[0412] As shown in Figure 12, the apparatus 1200 of this embodiment includes a memory 1210, a processor 1220, a communication interface 1230, and a bus 1240. The memory 1210, the processor 1220, and the communication interface 1230 are connected to each other via the bus 1240.

[0413] The memory 1210 may be a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1210 may store programs. When the program stored in the memory 1210 is executed by the processor 1220, the processor 1220 is configured to execute the steps performed by the terminal device or the network device in any of the aforementioned embodiments.

[0414] The processor 1220 may be a general-purpose central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits to execute related programs.

[0415] The processor 1220 may also be an integrated circuit chip with signal processing capabilities. In the implementation process, the various related steps in the embodiment of the present application may be completed by hardware integrated logic circuits in the processor 1220 or software instructions.

[0416] The processor 1220 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The processor 1220 may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor.

[0417] The steps of the method disclosed in conjunction with the embodiments of the present application can be directly implemented as being executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in memory 1210, and processor 1220 reads the information in memory 1210 and, in conjunction with its hardware, completes the functions required to be performed by the units included in the device of the present application.

[0418] The communication interface 1230 may use, but is not limited to, a transceiver or other transceiver device to implement communication between the apparatus 1200 and other devices or apparatuses.

[0419] The bus 1240 may include a path for transmitting information between the various components of the device 1200 (eg, the memory 1210 , the processor 1220 , and the communication interface 1230 ).

[0420] Some embodiments of the present application also provide a computer program product that, when executed on a processor, can implement the method implemented by the terminal device or network device in any of the above-mentioned embodiments. Some embodiments of the present application also provide a computer-readable storage medium that contains computer instructions that, when executed on a processor, can implement the method implemented by the terminal device or network device in any of the above-mentioned embodiments.

[0421] It should be noted that the modules or components shown in the above embodiments may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASICs), one or more microprocessors (digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module is implemented by a processing element calling a program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program code, such as a controller. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0422] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, software modules or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0423] The term "plurality" in this article refers to two or more. The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the previous and next associated objects are in an "or" relationship; in the formula, the character " / " indicates that the previous and next associated objects are in a "division" relationship. In addition, it should be understood that in the description of this application, words such as "first" and "second" are only used for the purpose of distinguishing the description, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order.

[0424] It will be understood that the various numerical numbers involved in the embodiments of the present application are merely distinctions for the convenience of description and are not intended to limit the scope of the embodiments of the present application.

[0425] It can be understood that in the embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

Claims

1. A communication method, characterized in that: The method comprises: Sending first request information, where the first request information is used to request the network device to switch or deactivate the first artificial intelligence AI function; Send a first measurement report, where the first measurement report is a measurement report generated by the terminal device based on the first AI function before sending the first request information, or the first measurement report is a measurement report generated using a predefined field based on the measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report contains first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

2. The method according to claim 1, characterized in that Before sending the first request information, the method further includes: Receive first configuration information, where the first configuration information is used to indicate a preset duration, where a start time of the preset duration is determined based on a sending time of the first request information; The duration between the time when the first request information is sent and the time when the terminal device receives the response information is greater than the preset duration, and the response information is used to instruct the terminal device to switch the first AI function to the second AI function or to deactivate the first AI function; The sending of the first measurement report includes: Before the end time of the preset duration, the first measurement report is sent.

3. The method according to claim 2, characterized in that The method further comprises: After the end time of the preset duration, the first AI function is deactivated and / or a second measurement report is sent, where the second measurement report is a measurement report generated by the terminal device based on a non-AI function.

4. The method according to claim 2 or 3, characterized in that: Before receiving the first configuration information, the method further includes: Send first indication information, where the first indication information is used to indicate the maximum value of the preset duration supported by the terminal device.

5. The method according to any one of claims 1 to 4, characterized in that Before sending the first request information, the method further includes: Send second indication information, wherein the second indication information includes at least one of the following information: combination information of applicable conditions supported by the AI ​​characteristics of the terminal device, a measurement information reporting method corresponding to the combination information of the applicable conditions, and a measurement information reporting method supported by the terminal device.

6. The method according to claim 5, characterized in that The first measurement report is generated based on a measurement information reporting method supported by the terminal device; The first measurement report also includes third indication information, and the third indication information is used to indicate the measurement information reporting method used by the terminal device when generating the first measurement report.

7. The method according to claim 5, characterized in that After sending the second indication information and before sending the first request information, the method further includes: receiving second configuration information, where the second configuration information is used to indicate a measurement information reporting method used by the terminal device when the first AI function fails; The first measurement report is a measurement report determined based on the second configuration information.

8. The method according to claim 1, characterized in that The first request information is further used to indicate a time when the first AI function is invalid.

9. The method according to claim 8, characterized in that Before sending the first request information, the method further includes: Sending fourth indication information, where the fourth indication information is used to indicate capability information of the terminal device to predict failure of the AI ​​function, and the fourth indication information includes at least one of the following information: The terminal device has the ability to predict the failure of AI functions; The expected failure time of the AI ​​function in the terminal device; The confidence level in the expected failure time.

10. The method according to claim 9, characterized in that The expected failure time is the expected failure time of the first AI function.

11. The method according to claim 1, characterized in that: Before sending the first request information, the method further includes: Receive third configuration information, where the third configuration information is used to indicate a time when the terminal device sends the first measurement report.

12. The method according to claim 11, characterized in that The method further comprises: Before receiving the third configuration information, fifth indication information is sent, where the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

13. The method according to claim 11 or 12, characterized in that: The time for sending the first measurement report indicated by the third configuration information includes at least one of the following time points: a time point for sending the first request information, a time point for satisfying a first threshold condition between the time point for sending the first request information, a time point for receiving a positive confirmation ACK or a negative confirmation NACK corresponding to the first request information, and a time point for satisfying a second threshold condition between the time point for receiving a positive confirmation ACK or a negative confirmation NACK corresponding to the first request information.

14. A communication method, characterized in that: The method comprises: Receiving first request information, where the first request information is used to request a network device to switch or deactivate a first artificial intelligence AI function; A first measurement report is received, where the first measurement report is a measurement report generated by the terminal device based on the first AI function before sending the first request information, or the first measurement report is a measurement report generated using a predefined field based on the measurement report format corresponding to the first AI function, or the first measurement report is a measurement report generated by the terminal device based on the first AI function and the first measurement report contains first content and / or second content, the credibility of the first content is higher than the credibility of the second content, or the first measurement report is a measurement report generated by the terminal device based on a non-AI function.

15. The method according to claim 14, characterized in that Before receiving the first request information, the method further includes: Sending first configuration information, where the first configuration information is used to indicate a preset duration; The duration between the time when the first request information is sent and the time when the terminal device receives the response information is greater than the preset duration, and the response information is used to instruct the terminal device to switch the first AI function to the second AI function or deactivate the first AI function.

16. The method according to claim 15, characterized in that The method further comprises: Receive a second measurement report, where the second measurement report is a measurement report generated by the terminal device based on a non-AI function.

17. The method according to claim 15 or 16, characterized in that Before sending the first configuration information, the method further includes: Receive first indication information, where the first indication information is used to indicate the maximum value of the preset duration supported by the terminal device.

18. The method according to any one of claims 14 to 17, characterized in that Before receiving the first request information, the method further includes: Receive second indication information, where the second indication information includes at least one of the following information: combination information of applicable conditions supported by the AI ​​characteristics of the terminal device, a measurement information reporting method corresponding to the combination information of the applicable conditions, and a measurement information reporting method supported by the terminal device.

19. The method according to claim 18, characterized in that The first measurement report is generated based on a measurement information reporting method supported by the terminal device; The first measurement report also includes third indication information, and the third indication information is used to indicate the measurement information reporting method used by the terminal device when generating the first measurement report.

20. The method according to claim 18, characterized in that After receiving the second indication information and before receiving the first request information, the method further includes: Sending second configuration information, where the second configuration information is used to indicate a measurement information reporting method used by the terminal device when the first AI function fails; The first measurement report is a measurement report determined based on the second configuration information.

21. The method according to claim 14, characterized in that The first request information is further used to indicate a time when the first AI function is invalid.

22. The method according to claim 21, characterized in that Before receiving the first request information, the method further includes: Receive fourth indication information, where the fourth indication information is used to indicate capability information of the terminal device to predict failure of the AI ​​function, and the fourth indication information includes at least one of the following information: The terminal device has the ability to predict the failure of AI functions; The expected failure time of the AI ​​function in the terminal device; The confidence level in the expected failure time.

23. The method according to claim 22, characterized in that The expected failure time is the expected failure time of the first AI function.

24. The method according to claim 14, characterized in that Before receiving the first request information, the method further includes: Send third configuration information, where the third configuration information is used to indicate a time when the terminal device sends the first measurement report.

25. The method according to claim 24, characterized in that The method further comprises: Before sending the third configuration information, fifth indication information is received, where the fifth indication information is used to indicate the time when the terminal device expects to send the first measurement report after the AI ​​function fails.

26. The method according to claim 24 or 25, characterized in that The time for sending the first measurement report indicated by the third configuration information includes at least one of the following time points: a time point for sending the first request information, a time point for satisfying a first threshold condition between the time point for sending the first request information, a time point for receiving a positive confirmation ACK or a negative confirmation NACK corresponding to the first request information, and a time point for satisfying a second threshold condition between the time point for receiving a positive confirmation ACK or a negative confirmation NACK corresponding to the first request information.

27. A communication device, characterized in that: The method comprises various functional modules for implementing the method according to any one of claims 1 to 13 or any one of claims 14 to 26.

28. A communication device, characterized in that: include: A processor, the processor is coupled to a memory, the memory is used to store a computer program, when the processor calls the computer program, the device executes the method according to any one of claims 1 to 13 or any one of claims 14 to 26.

29. A computer program product, characterized in that The method comprises a computer program code, which, when executed on a computer, causes the computer to implement the method as claimed in any one of claims 1 to 13 or any one of claims 14 to 26.

30. A computer readable medium, characterized in that The computer-readable medium stores a program code for computer execution, the program code comprising instructions for executing the method of any one of claims 1 to 13 or any one of claims 14 to 26.

Images