Artificial intelligence communication method and device and medium

Abstract

The invention provides an artificial intelligence communication method and device and a medium, relates to the technical field of communication, and is applied to mobile communication between a network device and a terminal device based on the AI technology, and the method comprises the following steps: the terminal device reports AI operation indication information to the network device, the AI operation indication information is used for representing whether the terminal equipment is currently suitable for executing the AI operation or not. According to the invention, the terminal equipment actively reports the AI operation indication information to the network equipment, so that the monitoring of the network equipment on the AI operation feasibility information is enhanced, the success rate and the high efficiency of the terminal equipment for executing the AI operation are improved, and the air interface interaction efficiency and the artificial intelligence communication effect are further improved.

Description

Technical Field

[0001] The present disclosure relates at least to the field of communication technologies, and particularly to an artificial intelligence communication method, a terminal device, a network device, and a computer-readable storage medium. Background Art

[0002] In recent years, a large number of application cases of the integration of AI (Artificial Intelligence) and wireless communication systems have emerged. When introducing AI technology into a wireless communication system, a possible working mode is that AI-related operations (including but not limited to model training, model inference, etc.) are executed in a terminal device.

[0003] Currently, during the mobile communication process between a network device and a terminal device based on AI technology, there is a lack of monitoring of the feasibility information of AI operations by the network device, which affects the success rate and efficiency of the terminal device in executing AI operations. Summary of the Invention

[0004] The technical problem to be solved by the present disclosure is to provide an artificial intelligence communication method, a terminal device, a network device, and a computer-readable storage medium to solve the problem that the network device cannot obtain AI operation-related information of the terminal device in view of the above deficiencies.

[0005] In a first aspect, the present disclosure provides an artificial intelligence communication method, which includes:

[0006] The terminal device reports artificial intelligence (AI) operation indication information to the network device, and the AI operation indication information is used to indicate whether the terminal device is currently suitable for executing AI operations.

[0007] In a second aspect, the present disclosure provides an artificial intelligence communication method, which includes:

[0008] The network device receives the artificial intelligence (AI) operation indication information reported by the terminal device, and the AI operation indication information is used to indicate whether the terminal device is currently suitable for executing AI operations.

[0009] In a third aspect, the present disclosure provides a terminal device, including:

[0010] A first sending module, configured to report artificial intelligence (AI) operation indication information to the network device, and the AI operation indication information is used to indicate whether the terminal device is currently suitable for executing AI operations.

[0011] In a fourth aspect, the present disclosure provides a network device, including:

[0012] A first receiving module, configured to receive AI operation instruction information reported by a terminal device, where the AI operation instruction information is used to indicate whether the terminal device is currently suitable for performing an AI operation.

[0013] In a fifth aspect, the present disclosure provides a computer-readable storage medium storing a computer program, which, when run by a processor, implements the artificial intelligence communication method as described above.

[0014] The present disclosure provides an artificial intelligence communication method, a terminal device, a network device, and a computer-readable storage medium. The terminal device actively sends AI operation instruction information to the network device to indicate whether the terminal device is currently suitable for performing an AI operation, enhancing the network device's monitoring of the feasibility information of AI operations, improving the success rate and efficiency of the terminal device in performing AI operations, and thus improving the air interface interaction efficiency and artificial intelligence communication effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flowchart of an artificial intelligence communication method according to an embodiment of the present disclosure;

[0016] Figure 2 is a first interaction flowchart of an artificial intelligence communication method according to an embodiment of the present disclosure;

[0017] Figure 3 is a second interaction flowchart of an artificial intelligence communication method according to an embodiment of the present disclosure;

[0018] Figure 4 is a third interaction flowchart of an artificial intelligence communication method according to an embodiment of the present disclosure;

[0019] Figure 5 is a fourth interaction flowchart of an artificial intelligence communication method according to an embodiment of the present disclosure;

[0020] Figure 6 is a flowchart of another artificial intelligence communication method according to an embodiment of the present disclosure;

[0021] Figure 7 is a schematic structural diagram of a terminal device according to an embodiment of the present disclosure;

[0022] Figure 8 is a schematic structural diagram of a network device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] To enable those skilled in the art to better understand the technical solutions of the present disclosure, the following will further describe in detail the embodiments of the present disclosure in conjunction with the accompanying drawings.

[0024] It is understood that the specific embodiments and drawings described herein are only for explaining the present disclosure, rather than limiting the present disclosure.

[0025] It is understood that, without conflict, the embodiments in the present disclosure and the features in the embodiments may be combined with each other.

[0026] It is understood that, for the convenience of description, only the parts related to the present disclosure are shown in the drawings of the present disclosure, and the parts unrelated to the present disclosure are not shown in the drawings.

[0027] It is understood that each unit and module involved in the embodiments of the present disclosure may correspond to only one entity structure, or may be composed of multiple entity structures. Alternatively, multiple units and modules may also be integrated into one entity structure.

[0028] It is understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present disclosure may occur in an order different from that marked in the drawings.

[0029] It is understood that in the flowcharts and block diagrams of the present disclosure, the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the embodiments of the present disclosure are shown. Among them, each block in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified function. Moreover, each block or combination of blocks in the block diagram and flowchart may be implemented by a hardware-based system for implementing the specified function, or may be implemented by a combination of hardware and computer instructions.

[0030] It is understood that the units and modules involved in the embodiments of the present disclosure may be implemented in software or in hardware. For example, the units and modules may be located in the processor.

[0031] Embodiment 1:

[0032] As Figure 1 shown, the present disclosure provides an artificial intelligence communication method, and the method

[0033] includes:

[0034] S11. The terminal device reports AI operation indication information to the network device, and the AI operation indication information is used to indicate whether the terminal device is currently suitable for performing an AI operation.

[0035] Specifically, benefiting from the advantages of artificial intelligence (AI) technology in dealing with non-linear and complex problems, introducing AI technology into wireless communication systems is expected to break through the performance bottleneck of wireless communication technology. Introducing artificial intelligence into wireless communication systems is a future development trend of wireless technology. However, both model training and model inference in artificial intelligence technology require certain computing resources and generate corresponding energy consumption. Moreover, due to the dependence of artificial intelligence models on data, the model generalization ability is limited, and different scenarios may require different models to complete inference tasks. Therefore, device status, hardware conditions, and application scenarios will all affect the effect of the device when performing artificial intelligence operations, especially crucial for terminal devices. Terminal devices not only have limited computing resources and power, but also have many and rapidly changing application scenarios due to their strong mobility. Sometimes, the status or the scenario of the terminal device is not suitable for the AI operation that needs to be executed currently. For this problem, there is currently no relevant solution.

[0036] In this embodiment, by the terminal device actively reporting artificial intelligence (AI) operation indication information to the network device to indicate whether the terminal device is currently suitable for executing an AI operation, the monitoring of the feasibility information of the AI operation by the network device is enhanced, the success rate and efficiency of the terminal device executing the AI operation are improved, and thus the air interface interaction efficiency and the artificial intelligence communication effect are improved.

[0037] The information content of the AI operation indication information in this embodiment includes relevant information for indicating whether the terminal device is suitable for executing an AI operation, enabling the network device to judge the AI capability matching situation and then judge whether the status or the scenario of the terminal device when executing the AI operation is suitable for continuing to execute the AI operation, so as to timely understand whether the terminal device meets the requirements such as computing resources, energy consumption, and applicable scenarios required for the AI operation it undertakes, and improve the application effect of AI technology in wireless communication systems.

[0038] A more specific example is Figure 2 As shown, this method includes Step 1: The terminal device sends (i.e., reports) AI operation indication information to the network device. The AI operation indication information is specifically indication information for executable AI operations to instruct the network device to evaluate whether the terminal device is suitable for executing an AI operation. The network device can be at least one of a base station, a core network, a core network element, a network management system, etc. In a mobile communication system, when the terminal device reports indication information for executable AI operations to the network device, it will cause the network device to evaluate whether the terminal device is suitable for executing an AI operation.

[0039] In an implementation manner, the terminal device reporting AI operation indication information to the network device specifically includes:

[0040] The terminal device sends terminal assistance information UAI including the AI operation indication information to the network device, so that the network device can obtain the AI operation indication information from the UAI.

[0041] Specifically, the terminal device can actively send terminal assistance information (UE Assistance Information, UAI) to the network device according to 3GPP standards.

[0042] In this embodiment, the AI operation indication information used for the network device to evaluate whether the terminal device is suitable for performing AI operations can be included in the UAI. The format of the AI operation indication information can specifically be a 1-bit indication or a list of indication information.

[0043] In an implementation manner, the AI operation indication information specifically includes the terminal state information and / or AI application conditions of the terminal device, so that the network device can evaluate whether the terminal device is currently suitable for performing AI operations according to the AI operation indication information.

[0044] Specifically, during the process of the terminal device performing AI operations, changes in factors such as the state or the scenario of the terminal device may cause the terminal device to be unsuitable for performing AI operations. If the network device cannot obtain the information related to the AI operations of the terminal device, it cannot adjust the AI operation control strategy in a timely manner, which may lead to problems such as AI operation failures or deterioration of the performance of the terminal device.

[0045] In this embodiment, the information content of the AI operation indication information is used for the network device to evaluate whether the terminal device is suitable for performing AI operations. Specifically, it can be an indication information for executable AI operations, or an AI operation feasibility indication information, that is, Figure 3 Indication for feasibility of AI operations sent independently as shown, or it can be information included in the UAI. The network device timely evaluates whether the terminal device is suitable for performing AI operations according to the AI operation indication information, and then can adjust the AI operation control strategy in a timely manner based on this to avoid situations such as AI operation failures or deterioration of the performance of the terminal device. The AI operation indication information can be the terminal state information and / or AI application conditions of the terminal device, or other information that may characterize whether the terminal device is currently suitable for performing AI operations.

[0046] In an implementation manner, the AI application conditions specifically include the signal-to-noise ratio and / or reference signal received power of the terminal device.

[0047] In an implementation manner, the terminal state information specifically includes the current terminal battery information.

[0048] In one embodiment, the current terminal power information specifically includes the remaining power value E of the terminal and / or the remaining power percentage r of the terminal currently.

[0049] In one embodiment, the E is obtained by measuring with a fuel gauge chip or by collecting the current battery voltage and current of the terminal and then looking up a table for calculation, and is used for the network device to evaluate that the terminal device is currently suitable for executing the first AI operation in response to the E being greater than the energy consumption required for the AI task to be sent.

[0050] In one embodiment, the r is calculated based on the total power of the terminal device and the E, and is used for the network device to evaluate that the terminal device is currently not suitable for executing the second AI operation in response to the r being less than a preset threshold.

[0051] Specifically, a possible problem scenario is that when the terminal device executes the AI model training task with insufficient power, it will accelerate the terminal device to enter the shutdown state, and the training task will also terminate accordingly. This not only reduces the air interface interaction efficiency but also affects the artificial intelligence communication effect. In addition, in some scenarios, the terminal device no longer has the condition to execute the AI application, such as reaching the signal-to-noise ratio (SINR, Signal to Interference plus Noise Ratio) threshold or occurring the measurement results or events that meet the cell handover.

[0052] In this embodiment, the AI operation instruction information specifically includes the first information, and the first information can be the terminal status information and / or the application conditions of the AI; among them, the terminal status information includes but is not limited to the current terminal power information, and the current remaining power value E of the terminal and the current remaining power percentage r of the terminal can be used to represent the current terminal power; the current remaining power E of the terminal battery can be calculated by a fuel gauge chip, or a current remaining power value E of the terminal battery can be obtained by collecting the battery voltage and current and looking up a table for calculation; and by obtaining the total power set by the terminal device at the factory, the current remaining power percentage r of the terminal is calculated; among them, the AI application conditions can be the signal-to-noise ratio threshold or the measurement results or events that meet the cell handover. E and r can be used alone or in combination to provide an evaluation basis for the network device to evaluate whether the terminal device is suitable for executing the AI operation. For example, when the remaining power percentage r of the terminal is less than the preset threshold, the terminal device is not suitable for continuing the AI-based communication method, and the network device sends an AI capability deactivation instruction to the terminal device to make the terminal device stop executing the first AI operation currently being executed; when the remaining power percentage r of the terminal is greater than or equal to the preset threshold and the remaining power E is greater than the energy consumption required for the AI task to be sent by the network device, the network device sends the AI operation-related configuration information to the terminal to make the terminal device start executing the second AI operation.

[0053] In a specific example as shown in Figure 2 after the network device receives the indication information for performing AI operations sent by the terminal device (such as the Indication for feasibility of AI operations shown in Figure 3 ), the network device does not send any feedback or configuration information to the terminal device. For example, when the evaluation result indicates that the terminal device can continue to perform the current AI operation being executed, or when the evaluation result indicates that the terminal device is not suitable for performing the to-be-issued AI task, the network device does not need to send any new instructions to the terminal device.

[0054] In one embodiment, after the terminal device sends the AI operation indication information to the network device, the method further includes:

[0055] The terminal device receives an AI deactivation instruction from the network device and stops the third AI operation according to the AI deactivation instruction. The AI deactivation instruction is sent by the network device in response to evaluating that the terminal device is currently not suitable for performing the third AI operation based on the AI operation indication information; or,

[0056] The terminal device receives an AI task configuration instruction from the network device and performs a fourth AI operation according to the AI task configuration instruction. The AI task configuration instruction is sent by the network device in response to evaluating that the terminal device is currently suitable for performing the fourth AI operation based on the AI operation indication information.

[0057] In one embodiment, the AI task execution instruction includes an AI task configuration instruction and an AI model training or application instruction;

[0058] Performing the fourth AI operation according to the AI task configuration instruction specifically includes:

[0059] The terminal device obtains AI task configuration information according to the AI task configuration instruction, and performs AI model training or application according to the AI model training or application instruction and the AI task configuration information.

[0060] Specifically, different AI operations require different computing resources, consume different amounts of energy, and have different applicable scenarios. For example, in solutions such as AI-based CSI (Channel State Information) feedback, AI-based beam management, and AI-based positioning, some or all of the AI operations may be performed by the terminal device, and the adaptability of the terminal device needs to be evaluated in combination with the specific AI task.

[0061] In this embodiment, after the network device receives the indication information reported by the terminal and adapted to perform AI operations, it evaluates whether the terminal device is suitable to start or continue with the AI-based communication method, and accordingly sends an AI operation indication to the terminal device. The AI operation indication includes, but is not limited to, AI function control instructions and AI configuration information. The AI configuration information includes, but is not limited to, data sets, AI model IDs (Identities), and AI model parameters. The AI function control instructions include, but are not limited to, model training, model retraining, model inference, AI function deactivation, model deactivation, model activation, model update, etc.

[0062] Another more specific example is as follows Figure 3 shown, including Step 1: The terminal device sends indication information indicating that AI operations can be performed to the network device (specifically, it can be, for example, Figure 5 the independently sent Indication for feasibility of AI operations as shown, or it can be the information included in the UAI) to indicate to the network device to evaluate whether the terminal device is adapted to perform AI operations, and Step 2: The network device sends configuration information to the terminal device, specifically, it can be an AI operation indication (specifically, it can be, for example, Figure 5The independently sent Indication of AI operations (which can also be included in other information) is used to indicate whether the terminal device performs AI operations. Among them, step two is sent by the network device to the terminal device after the network device evaluates the matching degree between the AI task to be sent and the terminal device. The evaluation result at least includes that the terminal device is suitable for performing AI operations and not suitable for performing AI operations. The sent Indication of AI operations also correspondingly includes an AI deactivation instruction and an AI task execution instruction, which are used to indicate the terminal device to stop performing the third AI operation or start performing the fourth AI operation. It can be understood that these two instructions cannot appear simultaneously for the same AI operation, but may appear simultaneously for different AI operations. For example, the terminal device sends terminal power status information to the network device. The network device evaluates whether the terminal device is suitable for the AI-based communication method based on the terminal power status information sent by the terminal device, and evaluates the matching degree between the AI task to be sent and the terminal status. For example, when the remaining battery percentage r of the terminal < 10%, the terminal device is not suitable for continuing the AI-based communication method, and the network device sends an AI capability deactivation instruction to the terminal device; when the remaining battery percentage r >= 10% and the remaining power E is greater than the energy consumption required for the AI task to be sent by the network device, the network device sends AI task-related configuration information to the terminal, which can enable the terminal device to obtain AI task configurations such as data sets, AI models, and AI model parameters, and then perform AI task operations such as model training, model retraining, model inference, model activation, and model update.

[0063] Embodiment 2:

[0064] As Figure 6 shown, the present disclosure provides an artificial intelligence communication method, and the method includes:

[0065] S21. The network device receives AI operation indication information reported by the terminal device, and the AI operation indication information is used to characterize whether the terminal device is currently suitable for performing AI operations.

[0066] In an embodiment, the network device receiving the AI operation indication information reported by the terminal device specifically includes:

[0067] The network device receives the terminal auxiliary information UAI sent by the terminal device and including the AI operation indication information, and obtains the AI operation indication information from the UAI.

[0068] In an embodiment, the AI operation indication information specifically includes the terminal status information and / or AI application conditions of the terminal device;

[0069] After the network device receives the AI operation instruction information sent by the terminal device, the method further includes:

[0070] The network device evaluates whether the terminal device is currently suitable for performing the AI operation according to the AI operation indication information.

[0071] In one implementation, the AI application condition specifically includes a signal-to-noise ratio and / or a reference signal receiving power of the terminal device.

[0072] In one implementation, the network device evaluates whether the terminal device is currently adapted to perform the AI operation according to the first information in the AI operation indication information, specifically including:

[0073] The network device parses the AI operation instruction information to specifically include the current available remaining power value E of the terminal device and / or the current remaining power percentage r of the terminal;

[0074] In response to E being greater than the energy consumption required for the AI task to be sent, the network device evaluates that the terminal device is currently suitable for performing the first AI operation, and / or, in response to r being less than a preset threshold, the network device evaluates that the terminal device is currently not suitable for performing the second AI operation.

[0075] In one embodiment, after the network device evaluates whether the terminal device is currently adapted to perform the AI operation according to the AI operation indication information, the method further includes:

[0076] The network device sends an AI operation instruction to the terminal device,

[0077] The AI operation instruction specifically includes an AI deactivation instruction and / or an AI task execution instruction.

[0078] The AI deactivation instruction is sent by the network device in response to evaluating, according to the AI operation instruction information, that the terminal device is currently not suitable for performing the third AI operation, and is used to instruct the terminal device to stop the third AI operation.

[0079] The AI task execution instruction is sent by the network device in response to evaluating, according to the AI operation indication information, that the terminal device is currently adapted to perform a fourth AI operation, and is used to instruct the terminal device to perform the fourth AI operation.

[0080] In one embodiment, the AI task execution instructions include AI task configuration instructions and AI model training or application instructions.

[0081] The AI task configuration instruction is used to instruct the terminal device to obtain AI task configuration information.

[0082] The AI model training or application instruction is used to instruct the terminal device to perform AI model training or application according to the AI task configuration information.

[0083] Embodiment 3:

[0084] like Figure 7 As shown, the present disclosure provides a terminal device, including:

[0085] The first sending module 11 is used to report artificial intelligence AI operation indication information to the network device, and the AI operation indication information is used to indicate whether the terminal device is currently suitable for performing AI operations.

[0086] In one implementation, the first sending module 11 specifically includes:

[0087] The UAI sending unit is used to send terminal auxiliary information UAI containing the AI operation indication information to the network device, so that the network device obtains the AI operation indication information from the UAI.

[0088] In one embodiment, the AI operation indication information specifically includes terminal status information and / or AI application conditions of the terminal device, so that the network device evaluates whether the terminal device is currently suitable for performing AI operations based on the AI operation indication information.

[0089] In one implementation, the AI application condition specifically includes a signal-to-noise ratio and / or a reference signal receiving power of the terminal device.

[0090] In one implementation, the terminal status information specifically includes current terminal power information.

[0091] In one implementation, the current terminal power information specifically includes the current available remaining power value E of the terminal and / or the current remaining power percentage r of the terminal.

[0092] In one embodiment, the E is obtained by measuring with a power meter chip, or by collecting the current battery voltage and the current battery current of the terminal and then looking up the table for calculation, and is used by the network device to evaluate that the terminal device is currently suitable for executing the first AI operation in response to the E being greater than the energy consumption required for the AI task to be sent.

[0093] In one implementation, the r is calculated based on the total power of the terminal device and the E, and is used by the network device to evaluate that the terminal device is currently not suitable for performing the second AI operation in response to the r being less than a preset threshold.

[0094] In one embodiment, the terminal device further includes a second receiving module, which is configured to: after sending the AI operation instruction information to the network device:

[0095] Receive an AI deactivation instruction from a network device, and stop the third AI operation according to the AI deactivation instruction. The AI deactivation instruction is sent by the network device in response to an assessment that the terminal device is currently not suitable for performing the third AI operation based on the AI operation indication information; or,

[0096] Receive an AI task configuration instruction from a network device, and execute a fourth AI operation according to the AI task configuration instruction. The AI task configuration instruction is sent by the network device in response to an assessment that the terminal device is currently suitable for performing the fourth AI operation based on the AI operation indication information.

[0097] In one embodiment, the AI task execution instruction includes an AI task configuration instruction and an AI model training or application instruction;

[0098] Executing the fourth AI operation according to the AI task configuration instruction specifically includes:

[0099] The terminal device obtains AI task configuration information according to the AI task configuration instruction, and performs AI model training or application according to the AI model training or application instruction and the AI task configuration information.

[0100] Example 4:

[0101] As Figure 8 shown, the present disclosure provides a network device, including:

[0102] A first receiving module 21, configured to receive artificial intelligence (AI) operation indication information reported by a terminal device, where the AI operation indication information is used to characterize whether the terminal device is currently suitable for performing an AI operation.

[0103] In one embodiment, the first receiving module 21 is specifically configured to:

[0104] Receive terminal auxiliary information UAI sent by the terminal device and containing the AI operation indication information, and obtain the AI operation indication information from the UAI.

[0105] In one embodiment, the AI operation indication information specifically includes the terminal status information of the terminal device and / or AI application conditions;

[0106] The network device further includes:

[0107] An evaluation module, connected to the first receiving module 21, configured to evaluate whether the terminal device is currently suitable for performing an AI operation according to the AI operation indication information.

[0108] In one embodiment, the AI application conditions specifically include the signal-to-noise ratio and / or reference signal received power of the terminal device.

[0109] In one embodiment, the evaluation module specifically includes:

[0110] A parsing unit, configured to parse the AI operation instruction information, and obtain the first information specifically including the current available remaining power value E of the terminal device and / or the current remaining percentage r of the terminal power;

[0111] An evaluation unit, connected to the parsing unit, is configured to evaluate that the terminal device is currently adapted to execute a first AI operation in response to the E being greater than the energy consumption required for the to-be-issued AI task, and / or, the network device evaluates that the terminal device is currently not adapted to execute a second AI operation in response to the r being less than a preset threshold.

[0112] In one embodiment, the network device further includes a second sending module, configured to, after evaluating whether the terminal device is currently adapted to execute an AI operation according to the AI operation instruction information:

[0113] Send an AI operation instruction to the terminal device,

[0114] The AI operation instruction specifically includes an AI deactivation instruction and / or an AI task execution instruction,

[0115] The AI deactivation instruction is sent by the network device in response to evaluating that the terminal device is currently not adapted to execute a third AI operation according to the AI operation instruction information, and is used to instruct the terminal device to stop the third AI operation.

[0116] The AI task execution instruction is sent by the network device in response to evaluating that the terminal device is currently adapted to execute a fourth AI operation according to the AI operation instruction information, and is used to instruct the terminal device to execute the fourth AI operation.

[0117] In one embodiment, the AI task execution instruction includes an AI task configuration instruction and an AI model training or application instruction.

[0118] The AI task configuration instruction is used to instruct the terminal device to obtain AI task configuration information.

[0119] The AI model training or application instruction is used to instruct the terminal device to perform AI model training or application according to the AI task configuration information.

[0120] Example 5:

[0121] Embodiment 5 of the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is run by a processor, it implements the artificial intelligence communication method as described in Embodiment 1 or 2, or implements the device as described in Embodiment 3 or 4.

[0122] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, computer program modules, or other data. The computer-readable storage medium includes, but is not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), digital versatile discs (DVDs) or other optical disc storage, magnetic cassettes, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0123] In addition, the present disclosure may also provide a computer device, including a memory and a processor. A computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the artificial intelligence communication method described in Embodiment 1 or 2. The computer device may be the device described in Embodiment 3 or 4.

[0124] Wherein, the memory is connected to the processor. The memory can adopt flash memory or read-only memory or other memories, and the processor can adopt a central processing unit or a single-chip microcomputer.

[0125] Embodiments 1-5 of the present disclosure provide an artificial intelligence communication method, a terminal device, a network device, and a computer-readable storage medium. By actively sending AI operation instruction information from the terminal device to the network device to indicate whether the terminal device is currently suitable for performing AI operations, the monitoring of the feasibility information of AI operations by the network device is enhanced, the success rate and efficiency of the terminal device in performing AI operations are improved, and thus the air interface interaction efficiency and artificial intelligence communication effect are improved.

[0126] It can be understood that the above embodiments are merely exemplary embodiments adopted to illustrate the principle of the present disclosure. However, the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present disclosure, and these modifications and improvements are also regarded as the protection scope of the present disclosure.

Claims

1. An artificial intelligence communication method, characterized in that, The method includes: The terminal device reports artificial intelligence (AI) operation indication information to the network device, where the AI operation indication information is used to indicate whether the terminal device is currently suitable for performing AI operations.

2. The method according to claim 1, wherein The AI operation indication information specifically includes the terminal state information and / or AI application conditions of the terminal device, so that the network device can evaluate whether the terminal device is currently suitable for performing AI operations according to the AI operation indication information.

3. The method according to claim 2, wherein The AI application conditions specifically include the signal-to-noise ratio and / or reference signal received power of the terminal device.

4. The method according to claim 2, wherein The terminal state information specifically includes the current terminal power information.

5. The method according to claim 4, wherein The current terminal power information specifically includes the remaining power value E of the terminal and / or the remaining percentage r of the terminal power.

6. The method according to claim 5, wherein The E is obtained by measuring with a fuel gauge chip or by calculating through a look-up table after collecting the current battery voltage and current of the terminal, and is used for the network device to evaluate that the terminal device is currently suitable for performing the first AI operation in response to the E being greater than the energy consumption required for the to-be-downloaded AI task.

7. The method according to claim 5, characterized in that The r is calculated according to the total power of the terminal device and the E, and is used for the network device to evaluate that the terminal device is currently not suitable for performing the second AI operation in response to the r being less than a preset threshold.

8. The method according to claim 2, wherein After the terminal device reports the AI operation indication information to the network device, the method further includes: The terminal device receives an AI deactivation instruction from the network device and stops the third AI operation according to the AI deactivation instruction, where the AI deactivation instruction is sent by the network device in response to evaluating that the terminal device is currently not suitable for performing the third AI operation according to the AI operation indication information; or, The terminal device receives an AI task configuration instruction from the network device and performs a fourth AI operation according to the AI task configuration instruction, where the AI task configuration instruction is sent by the network device in response to evaluating that the terminal device is currently suitable for performing the fourth AI operation according to the AI operation indication information.

9. The method according to claim 8, wherein The AI task execution instruction includes an AI task configuration instruction and an AI model training or application instruction; Performing the fourth AI operation according to the AI task configuration instruction specifically includes: The terminal device obtains AI task configuration information according to the AI task configuration instruction, and performs AI model training or application according to the AI model training or application instruction and the AI task configuration information.

10. The method according to any one of claims 1-9, characterized in that, The terminal device reporting the AI operation indication information to the network device specifically includes: The terminal device sends terminal auxiliary information UAI including the AI operation indication information to the network device, so that the network device can obtain the AI operation indication information from the UAI.

11. An artificial intelligence communication method, characterized in that, The method includes: The network device receives the artificial intelligence (AI) operation indication information reported by the terminal device, where the AI operation indication information is used to indicate whether the terminal device is currently suitable for performing AI operations.

12. The method according to claim 11, wherein The AI operation indication information specifically includes the terminal state information and / or AI application conditions of the terminal device; After the network device receives the AI operation indication information reported by the terminal device, the method further includes: The network device evaluates whether the terminal device is currently suitable for performing AI operations according to the AI operation indication information.

13. The method according to claim 12, wherein The AI application conditions specifically include the signal-to-noise ratio and / or reference signal received power of the terminal device.

14. The method according to claim 12, wherein The network device evaluates whether the terminal device is currently suitable for performing AI operations according to the AI operation indication information, specifically including: The network device parses out that the AI operation indication information specifically includes the current remaining power value E of the terminal of the terminal device and / or the current remaining power percentage r of the terminal; The network device evaluates that the terminal device is currently suitable for performing the first AI operation in response to the E being greater than the energy consumption required for the to-be-issued AI task, and / or, the network device evaluates that the terminal device is currently not suitable for performing the second AI operation in response to the r being less than the preset threshold.

15. The method according to claim 12, characterized in that, After the network device evaluates whether the terminal device is currently suitable for performing AI operations according to the AI operation indication information, the method further includes: The network device sends an AI operation indication to the terminal device, The AI operation indication specifically includes an AI deactivation instruction and / or an AI task execution instruction, The AI deactivation instruction is sent by the network device in response to evaluating that the terminal device is currently not suitable for performing the third AI operation according to the AI operation indication information, and is used to instruct the terminal device to stop the third AI operation, The AI task execution instruction is sent by the network device in response to evaluating that the terminal device is currently suitable for performing the fourth AI operation according to the AI operation indication information, and is used to instruct the terminal device to perform the fourth AI operation.

16. The method according to claim 15, wherein The AI task execution instruction includes an AI task configuration instruction and an AI model training or application instruction, The AI task configuration instruction is used to instruct the terminal device to obtain AI task configuration information, The AI model training or application instruction is used to instruct the terminal device to perform AI model training or application according to the AI task configuration information.

17. The method according to any one of claims 11 - 16, characterized in that, The network device receives the AI operation indication information reported by the terminal device, specifically including: The network device receives the terminal auxiliary information UAI sent by the terminal device and containing the AI operation indication information, and obtains the AI operation indication information from the UAI.

18. A terminal device, characterized in that, Including: The first sending module is used to report artificial intelligence AI operation indication information to the network device, and the AI operation indication information is used to represent whether the terminal device is currently suitable for performing AI operations.

19. A network device, characterized in that, Including: The first receiving module is used to receive the artificial intelligence AI operation indication information from the terminal device, and the AI operation indication information is used to represent whether the terminal device is currently suitable for performing AI operations.

20. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is run by a processor, it implements the artificial intelligence communication method according to any one of claims 1-10 or 11-17.

Images