Artificial intelligence communication method and device, equipment and storage medium

Abstract

The invention relates to an artificial intelligence AI communication method and device, equipment and a storage medium, and the method comprises the steps: determining to carry out first operation on a first AI model or a first AI function, and enabling the first AI model or the first AI function to be deployed in first equipment; the first AI model or the first AI function is determined to be subjected to the related operation, so that the first equipment can select the proper AI model or AI function to execute the corresponding task, and the use accuracy of the AI model or the AI function is improved.

Description

Artificial intelligence communication method, device, equipment and storage medium Technical Field

[0001] The present disclosure relates to the field of communication technologies, and in particular to artificial intelligence (AI) communication methods, devices, equipment, and storage media. Background Art

[0002] The widespread adoption of 5G technology is bringing profound changes to every aspect of our lives. According to the International Telecommunication Union (ITU), 5G will permeate every aspect of future society, building a comprehensive, user-centric information ecosystem. 5G technology can support advanced services like mobile virtual reality, massive IoT devices, meet the demanding requirements of connected vehicles and industrial control, and provide a superior user experience in high-speed rail environments. As a representative of new infrastructure, 5G will be central to building the future information society.

[0003] In recent years, AI technology has achieved continuous breakthroughs in multiple fields. The continued development of fields such as intelligent voice and computer vision has not only brought a rich variety of applications to smart terminals, but has also found widespread application in education, transportation, home furnishings, healthcare, retail, security, and other fields. While bringing convenience to people's lives, it is also promoting industrial upgrading across various industries.

[0004] Summary of the Invention

[0005] To overcome the problems existing in the related art, the present disclosure provides an AI communication method, apparatus, device and storage medium.

[0006] According to a first aspect of an embodiment of the present disclosure, an AI communication method is provided, which is executed by a first device and includes: determining to perform a first operation on a first AI model or a first AI function, wherein the first AI model or the first AI function is deployed on the first device.

[0007] According to a second aspect of an embodiment of the present disclosure, an AI communication device is provided, comprising: a processing module for determining to perform a first operation on a first AI model or a first AI function, wherein the first AI model or the first AI function is deployed on a first device.

[0008] According to a third aspect of an embodiment of the present disclosure, an AI communication device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the first aspect and any one of the methods in the first aspect.

[0009] According to a fourth aspect of an embodiment of the present disclosure, a non-temporary computer-readable storage medium is provided, which, when instructions in the storage medium are executed by a processor of a first device, enables the first device to execute the first aspect and any one of the methods in the first aspect.

[0010] By determining to perform relevant operations on the first AI model or the first AI function, the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0011] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0013] Fig. 1 is a schematic diagram of a wireless communication system according to an exemplary embodiment.

[0014] Fig. 2 is a flow chart showing an AI communication method according to an exemplary embodiment.

[0015] Fig. 3 is a flow chart showing another AI communication method according to an exemplary embodiment.

[0016] Fig. 4 is a flow chart showing another AI communication method according to an exemplary embodiment.

[0017] Fig. 5 is a flow chart showing yet another AI communication method according to an exemplary embodiment.

[0018] Fig. 6 is a flow chart showing another AI communication method according to an exemplary embodiment.

[0019] FIG7 is a flowchart showing another AI communication method according to an exemplary embodiment.

[0020] Fig. 8 is a flow chart showing yet another AI communication method according to an exemplary embodiment.

[0021] FIG9 is a flow chart showing another AI communication method according to an exemplary embodiment.

[0022] FIG10 is a flow chart showing another AI communication method according to an exemplary embodiment.

[0023] FIG11 is a flow chart showing yet another AI communication method according to an exemplary embodiment.

[0024] FIG12 is a schematic diagram of an AI communication device according to an exemplary embodiment.

[0025] FIG13 is a schematic diagram of an AI communication device according to an exemplary embodiment.

[0026] FIG14 is a schematic diagram showing another AI communication device according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings 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 disclosure.

[0028] The communication method involved in the present disclosure can be applied to the wireless communication system 100 shown in Figure 1. The network system may include a network device 110 and a terminal 120. It will be understood that the wireless communication system shown in Figure 1 is merely for illustrative purposes, and the wireless communication system may also include other network devices, such as core network devices, wireless relay devices, and wireless backhaul devices, which are not shown in Figure 1. The embodiments of the present disclosure do not limit the number of network devices and terminals included in the wireless communication system.

[0029] It can be further understood that the wireless communication system of the embodiment of the present disclosure is a network that provides wireless communication functions. The wireless communication system can adopt different communication technologies, such as code division multiple access (CDMA), wideband code division multiple access (WCDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal frequency-division multiple access (OFDMA), single carrier-frequency division multiple access (SC-FDMA), and carrier sense multiple access with collision avoidance. According to factors such as the capacity, rate, and latency of different networks, the network can be divided into 2G (English: generation) network, 3G network, 4G network or future evolution network, such as the 5th generation wireless communication system (5G) network. The 5G network can also be referred to as NR. For the convenience of description, the present disclosure sometimes refers to the wireless communication network as simply a network.

[0030] Furthermore, the network device 110 involved in the present disclosure may also be referred to as a wireless access network device. The wireless access network device may be: a base station, an evolved node B (eNB), a home base station, an access point (AP) in a wireless fidelity (WIFI) system, a wireless relay node, a wireless backhaul node, a transmission point (TP) or a TRP, etc. It may also be a gNB in ​​an NR system, or it may also be a component or part of a base station. When it is a vehicle-to-everything (V2X) communication system, the network device may also be an on-board device. It should be understood that in the embodiments of the present disclosure, the specific technology and specific device form adopted by the network device are not limited.

[0031] Furthermore, the terminal 120 involved in the present disclosure may also be referred to as a terminal device, user equipment (UE), mobile station (MS), mobile terminal (MT), etc., which is a device that provides voice and / or data connectivity to users. For example, the terminal may be a handheld device with wireless connection function, a vehicle-mounted device, etc. At present, some examples of terminals are: smart phones (mobile phones), pocket personal computers (PPCs), handheld computers, personal digital assistants (PDAs), laptops, tablet computers, wearable devices, or vehicle-mounted devices, etc. In addition, when it is a vehicle-to-everything (V2X) communication system, the terminal device may also be a vehicle-mounted device. It should be understood that the embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the terminal.

[0032] In the embodiments of the present disclosure, the network device 110 and the terminal 120 can use any feasible wireless communication technology to achieve mutual data transmission. The transmission channel corresponding to the transmission of data from the network device 110 to the terminal 120 is called a downlink channel (DL), and the transmission channel corresponding to the transmission of data from the terminal 120 to the network device 110 is called an uplink channel (UL). It is understood that the network device involved in the embodiments of the present disclosure can be a base station. Of course, the network device can also be any other possible network device, and the terminal can be any possible terminal, and this disclosure is not limited thereto.

[0033] The widespread application of 5G technology has brought about tremendous changes in all aspects of people's lives. According to the ITU's vision, 5G will penetrate all areas of future society and build a comprehensive information ecosystem centered on users. Among them, 5G user experience rates can reach 100Mbit / s to 1Gbit / s, which can support ultimate service experiences such as mobile virtual reality; 5G peak rates can reach 10Gbit / s to 20Gbit / s, and traffic density can reach 10Mbit / s / m2, which can support the future growth of mobile service traffic by more than a thousand times; 5G connection density can reach 1 million / m 2 , effectively supporting massive numbers of IoT devices; 5G transmission latency can reach milliseconds, meeting the stringent requirements of the Internet of Vehicles and industrial control; 5G can support mobile speeds of 500 km / h, providing a good user experience in high-speed rail environments. As a representative of new infrastructure, 5G will reshape the future information society.

[0034] In recent years, AI technology has achieved continuous breakthroughs in numerous fields. The continued development of fields like intelligent voice and computer vision has not only brought a rich variety of applications to mobile devices, but has also found widespread application in education, transportation, home furnishings, healthcare, retail, security, and other fields. While bringing convenience to people's lives, it is also promoting industrial upgrading across various industries. AI technology is also rapidly interpenetrating with other disciplines. Its development integrates knowledge from different disciplines while also providing new directions and methods for their development.

[0035] In Release 18 of the 3rd Generation Partnership Project (3GPP), a research project on artificial intelligence (AI) technology in the radio air interface was established within Radio Access Network (RAN) 1. This project aims to investigate how to integrate AI technology into the radio air interface and explore how AI can assist in improving transmission technologies within the radio air interface.

[0036] For example, in wireless AI research, application cases of artificial intelligence include:

[0037] -AI-based channel state information (CSI) enhancement

[0038] -AI-based beam management

[0039] -AI-based positioning

[0040] In some embodiments, AI models may also exist in multiple types, such as unilateral models and bilateral models. A unilateral model means that all parts of the AI ​​model are deployed on the same device. It can be deployed entirely on the terminal or entirely on the network device. A bilateral model means that part of the AI ​​model is deployed on the terminal, while the other part is deployed on the network device. In other words, different parts of the same AI model can be deployed on different devices. In this case, the AI ​​model needs to be jointly completed by the terminal and the network device during the reasoning process. Of course, in the case of a bilateral model, although part of the same AI model is deployed on the terminal or network device, for the terminal or network device, the part of the AI ​​model deployed can be considered as one AI model. That is, in the case of a bilateral model, the terminal deploys AI model 1 and the network device deploys AI model 2, and AI model 1 and AI model 2 can be considered to belong to the same AI model.

[0041] For terminals or network devices, it is possible to deploy multiple AI models or AI functionalities. For a terminal or a network device, the multiple AI models or AI functionalities deployed may be used for different instances, such as CSI enhancement, positioning, etc. For a terminal or a network device, the multiple AI models or AI functionalities deployed may also be used for different scenarios of the same instance. For example, all are used for CSI enhancement, but some are used for macro cells and some for micro cells.

[0042] As for terminals or network devices, how to choose to activate, deactivate or switch an AI model or AI function is still unclear.

[0043] Therefore, the present disclosure provides AI communication methods, devices, equipment, and storage media, which, by determining relevant operations on a first AI model or a first AI function, enable the first device to select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0044] FIG2 is a flow chart of an AI communication method according to an exemplary embodiment. As shown in FIG2 , the method is executed by a first device and may include the following steps:

[0045] In step S11 , it is determined to perform a first operation on a first AI model or a first AI function.

[0046] In some embodiments, the first device may determine to perform a first operation on a first AI model or a first AI function, wherein the first AI model or the first AI function is deployed on the first device.

[0047] In some embodiments, there may be the following scenarios associated with a model: model activation, model switching, and / or model deactivation.

[0048] For example, model activation can be understood as the transition from a state processed by a non-AI model to a state processed by an AI model.

[0049] For another example, model switching can be understood as switching from one AI model to another.

[0050] For another example, model deactivation can be understood as a way of falling back from AI model processing to non-AI model processing, such as traditional processing that does not utilize AI models.

[0051] In some embodiments, the first operation may be an activation operation. The first device may determine to activate the first AI model or the first AI function. It will be understood that after the first AI model or the first AI function is activated, the first device may use the first AI model or the first AI function to perform corresponding tasks. For example, if the first AI model or the first AI function is an AI model or AI function for CSI compression, then after the first AI model or the first AI function is activated, the first device may use the first AI model or the first AI function to perform tasks related to CSI compression.

[0052] In some embodiments, the first operation may be a deactivation operation. The first device may determine to deactivate the first AI model or the first AI function. It will be understood that after the first AI model or the first AI function is deactivated, the first device will not be able to use the first AI model or the first AI function to perform corresponding tasks. For example, if the first AI model or the first AI function is an AI model or AI function for CSI compression, then after the first AI model or the first AI function is deactivated, the first device will no longer be able to use the first AI model or the first AI function to perform tasks related to CSI compression.

[0053] In some embodiments, the first operation may be an AI switching operation. The first device may determine to de-switch the first AI model or the first AI function. The AI ​​switching operation may be used to indicate switching the AI ​​model, and the AI ​​switching operation may also be used to indicate switching the AI ​​function. For example, the first device may determine to switch the currently used AI model to the first AI model, or the first device may determine to switch the currently used AI function to the first AI function. It will be understood that after switching to the first AI model or the first AI function, the first device may use the first AI model or the first AI function to perform the corresponding task. For example, the first AI model or the first AI function is an AI model or AI function for CSI compression. After switching to the first AI model or the first AI function is activated, the first device may use the first AI model or the first AI function to perform tasks related to CSI compression.

[0054] The present disclosure determines to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0055] In the AI ​​communication method provided in the embodiment of the present disclosure, FIG3 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG3, the method may further include the following steps:

[0056] In step S21 , a first data set is determined.

[0057] In some embodiments, the first device may determine a first data set, wherein the first data set may be a data set used for testing an AI model.

[0058] For example, the first dataset can be a pre-set dataset. For example, it can be a dataset for CSI compression. The first dataset can be used to monitor or assist in testing the AI ​​model. It can be input into the AI ​​model for corresponding testing. It will be understood that the testing in this disclosure may include monitoring the AI ​​model or AI function and / or assisting in testing the AI ​​model or AI function.

[0059] Of course, one or more first data sets can be determined. Among them, different first data sets can be used for different application cases. For example, they include a first data set for CSI compression, a first data set for beam management, and so on. Of course, different first data sets can also be different application scenarios in the same application case. For example, multiple first data sets can all be first data sets for CSI compression, but the difference is that some first data sets are applicable to macro cell scenarios, and some first data sets are applicable to micro cell scenarios. Alternatively, some first data sets are applicable to first type terminals, such as normal terminals; some first data sets are applicable to second type terminals, such as low-capability (redcap) terminals, and so on.

[0060] It is understood that the present disclosure does not limit the number of first data sets, nor the cases and scenarios in which each first data set is applied. The first device may determine an appropriate first data set based on actual conditions, and the present disclosure does not limit this.

[0061] In step S22 , the first AI model or the AI ​​model associated with the first AI function is used, and an AI operation is performed based on the first data set to obtain a first result.

[0062] In some embodiments, the first device may use the first AI model or the AI ​​model associated with the first AI function to perform a corresponding AI operation based on the first data set determined in S21 to obtain a corresponding first result.

[0063] Among them, the AI ​​operation can be considered as an operation that processes, analyzes, transforms, etc. an AI model or an AI model associated with an AI function to obtain a corresponding first result. For example, the first data set is input into the AI ​​model for calculation, analysis, etc., and then the result output by the AI ​​model is obtained. It can be understood that the AI ​​model may include an AI model for beam prediction, an AI model for CSI compression, an AI model for beam management, an AI model for positioning, and so on. The AI ​​function may include an AI function for beam prediction, an AI function for CSI compression, an AI function for beam management, an AI function for positioning, and so on. The present disclosure does not limit the specific AI models and AI functions.

[0064] It is understood that for different AI models, AI models corresponding to AI functions, etc., the first output result is the result of the corresponding AI model. For example, if the AI ​​model is an AI model for beam management, the actual output result is related to beam management, such as the predicted preferred beam, the predicted reference signal received power of a beam, etc.

[0065] For example, after determining the first data set, the first device may input the first data set into a corresponding AI model or an AI model associated with an AI function to perform a corresponding AI operation, thereby obtaining an AI result output by the AI ​​model or the AI ​​model associated with the AI ​​function based on the first data set.

[0066] It can be understood that the AI ​​model deployed on the first device or the AI ​​model associated with the AI ​​function, the AI ​​result output can be called the first result.

[0067] Taking the first data set as a CSI compression data set as an example, the CSI compression data set can be input into a first AI model, or into an AI model associated with a first AI function. It will be understood that in the above process, the first AI model or first AI function can be an inactivated AI model or AI function. Furthermore, the first AI model or first AI function is an AI model or AI function for CSI compression, and the first data set and the first AI model or first AI function can be considered to correspond to the same application instance. Alternatively, the first AI model or first AI function is not an AI model or AI function for CSI compression, and the first data set and the first AI model or first AI function can be considered to correspond to different application instances. The first AI model and the AI ​​model associated with the first AI function take the first data set as input and obtain an output AI result.

[0068] It is understood that each AI model or AI model associated with an AI function outputs an AI result. In other words, each AI result is associated with a corresponding AI model.

[0069] In some embodiments, determining to perform the first operation on the first AI model or the first AI function in S11 may further include the following steps:

[0070] In step S23 , it is determined to perform a first operation on the first AI model or the first AI function based on the first result.

[0071] In some embodiments, the first device may determine to perform a first operation on the first AI model or the first AI function based on the first result obtained in S22.

[0072] For example, when the first device obtains the first result, it may determine whether to perform the first operation on the first AI model or the first AI function based on the first result. For example, if the first result satisfies certain conditions, the first device determines to perform the first operation on the first AI model or the first AI function. For another example, if the first result does not satisfy certain conditions, the first device determines to perform the first operation on the first AI model or the first AI function. For another example, if the first device obtains the first result, it directly determines to perform the first operation on the first AI model or the first AI function.

[0073] For another example, the first device may send the first result to the second device when obtaining the first result. This allows the second device to perform an AI operation on the first result based on the AI ​​model or AI function deployed on the second device to obtain a second result. The second device may instruct the first device to perform a first operation on the first AI model or the first AI function based on the second result. For example, if the second result meets certain conditions, the second device instructs the first device to perform the first operation on the first AI model or the first AI function. For another example, if the second result does not meet certain conditions, the second device instructs the first device to perform the first operation on the first AI model or the first AI function. For another example, if the second device obtains the second result, it directly instructs the first device to perform the first operation on the first AI model or the first AI function. The first device may determine to perform the first operation on the first AI model or the first AI function based on the instruction of the second device.

[0074] For another example, the first device may send the first result to the second device upon obtaining the first result. This allows the second device to perform an AI operation on the first result based on the AI ​​model or AI function deployed on the second device to obtain a second result. The second device may feed back the second result to the first device. The first device determines to perform a first operation on the first AI model or the first AI function based on the first result and / or the second result. For example, the first result satisfies some conditions corresponding to the first result, and / or the second result satisfies some conditions corresponding to the second result. The first device determines to perform the first operation on the first AI model or the first AI function. For another example, the first result does not satisfy some conditions corresponding to the first result, and / or the second result does not satisfy some conditions corresponding to the second result. The first device determines to perform the first operation on the first AI model or the first AI function. For another example, the first device may directly determine to perform the first operation on the first AI model or the first AI function upon determining the first result and / or the second result.

[0075] In some embodiments, the first operation may be one or more of an activation operation, a deactivation operation, and an AI switching operation on the first AI model or the first AI function.

[0076] The present disclosure determines whether to perform relevant operations on the first AI model or the first AI function through the first data set, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0077] In the communication method provided by the embodiment of the present disclosure, determining to perform a first operation on the first AI model or the first AI function based on the first result in S23 may include: determining to perform the first operation on the first AI model or the first AI function when the first result satisfies a first condition.

[0078] In some embodiments, when the first result obtained in S22 satisfies the first condition, the first device may determine to perform a first operation on the first AI model or the first AI function.

[0079] For example, the first device determines, based on the first result obtained in S22, that the first result satisfies the first condition. In this case, the first device may determine to perform a first operation on the first AI model or the first AI function, such as activating the first AI model or the first AI function, deactivating the first AI model or the first AI function, and / or switching the first AI model or the first AI function.

[0080] For example, if a first result satisfies a first condition, the first device, based on preconfigured rules, independently determines to perform a first operation on the first AI model or first AI function. In this case, the first device may send the first result to the second device so that the second device can obtain a second result based on the first result. The second device then sends the second result to the first device. One possible way for the first result to satisfy the first condition is for the second result to satisfy the second condition. For example, in a positioning application, the second result indicates the location. When the location satisfies a pre-set location condition, the first result is considered to satisfy the first condition. Of course, another possible way for the first result to satisfy the first condition is for the first and second results to satisfy the second condition. For another example, in a CSI compression application, the first device may determine the CSI compression error based on the first and second results. If the error reaches a certain error threshold, the first result may be considered to satisfy the first condition. For example, if the error is less than or equal to the first error threshold, the first AI model or first AI function may be activated; for another example, if the error is greater than or equal to the second error threshold, the first AI model or first AI function may be switched. Different error thresholds can be set depending on the situation and are not limited in this disclosure.

[0081] In some embodiments, the first result satisfying the first condition may be that the first result is greater than or equal to a first threshold. Alternatively, the first result satisfying the first condition may be that the first result is less than or equal to the first threshold. In this case, the first result may be a specific value, such as a CSI compression ratio, data error, etc.

[0082] In some embodiments, the first result satisfying the first condition may be that the first result is a first parameter. The first result satisfying the first condition may also be that the first result is not the first parameter. In this case, the first result may be a specific parameter, such as a numerical value, a position, or an indication of a parameter. It can be understood that the first result being a position may be in an application instance scenario where positioning is performed. The first result indicating a parameter may, for example, indicate a beam, and this may be an application instance scenario where beam management is performed. The first result being a numerical value may be the reference signal received power predicted by a beam, etc., or may be a certain angle when positioning is performed, or may be a compression parameter during CSI compression.

[0083] It is clear that the above is only an exemplary description, and the present disclosure does not limit the specific first result and the specific first condition.

[0084] Of course, it should be understood that the above examples are merely illustrative descriptions and the present disclosure does not limit them.

[0085] For another example, the first device may send the first result to the second device, so that the second device can obtain the second result based on the first result. Furthermore, if the second result satisfies the second condition, the first device may send information to the first device indicating that a first operation should be performed on the first AI model or the first AI function. This allows the first device to determine whether to perform the first operation on the first AI model or the first AI function. It will be understood that in this case, one possible way for the first result to satisfy the first condition is when the second result satisfies the second condition.

[0086] It is understood that the first AI model or the first AI function may be the AI ​​model or AI function corresponding to the terminal accurately performing the first operation. For the first device, since the first AI model or the first AI function has been deployed on the first device, the first device can use the first AI model or the first AI function regardless of whether the first AI model or the first AI function has been activated. The purpose of activation is to determine which AI model or AI function to use to perform the related task when the related task is subsequently executed.

[0087] The present disclosure determines whether to perform relevant operations on the first AI model or the first AI function when the first result satisfies the first condition, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0088] In the AI ​​communication method provided in the embodiments of the present disclosure, FIG4 is a flowchart of another AI communication method according to an exemplary embodiment. As shown in FIG4 , determining to perform a first operation on the first AI model or the first AI function in S11 or S23 may further include the following steps:

[0089] In step S31, first information sent by a second device is received.

[0090] In some embodiments, a first device may receive first information sent by a second device. The first information may be used to instruct a first operation to be performed on a first AI model or a first AI function. Alternatively, the first information may be used to instruct a first operation not to be performed on the first AI model or the first AI function.

[0091] For example, after a first device inputs a first data set into a first AI model or an AI model associated with a first AI function for processing and obtains a first result, it may send the first result to a second device. After receiving the first result, the second device may input the first result into a second AI model or an AI model associated with the second AI function to obtain a second result. Based on the first result and / or the second result, the second device may determine whether to allow the first device to perform a first operation on the first AI model or the first AI function, and send first information indicating the determination. For example, if the second device, based on the first result and / or the second result, determines that the first device should perform a first operation on the first AI model or the first AI function, the second device may send first information to the first device indicating that the first operation should be performed on the first AI model or the first AI function. For another example, if the second device, based on the first result and / or the second result, determines not to allow the first device to perform the first operation on the first AI model or the first AI function, the second device may send first information to the first device indicating that the first operation should not be performed on the first AI model or the first AI function.

[0092] For another example, the first device may directly receive the first information sent by the second device. If the first information indicates that a first operation should be performed on the first AI model or the first AI function, the first device determines to perform the first operation on the first AI model or the first AI function. Alternatively, if the first information indicates that the first operation should not be performed on the first AI model or the first AI function, the first device determines not to perform the first operation on the first AI model or the first AI function.

[0093] The present disclosure uses the first information sent by the second device to indicate whether to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0094] In the AI ​​communication method provided in the embodiment of the present disclosure, FIG5 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG5 , the method may further include the following steps:

[0095] In step S41, the second information is sent.

[0096] In some embodiments, the first device sends second information to the second device, wherein the second information is used to indicate the first result.

[0097] In some embodiments, one possible implementation of the first result satisfying the first condition may be that the second result satisfies the second condition. The second result is obtained by the second device using an AI model associated with a second AI model or a second AI function based on the first result, and the second AI model or second AI function is deployed on the second device.

[0098] For example, a first device inputs a first data set into a first AI model or an AI model associated with a first AI function for processing, obtaining a first result. The first device may send the first result to a second device. Upon receiving the first result, the second device may input the first result into a second AI model or an AI model associated with a second AI function, obtaining a second result. The second AI model or the second AI function is deployed on the second device. If the second result meets a second condition, the second device may send first information to the first device to instruct the first device to perform relevant operations on the first AI model or the first AI function.

[0099] Taking the application example of CSI compression, the second device can determine the CSI compression error based on the second result and the first result. If the error reaches a certain error threshold, the second result can be considered to meet the second condition. For example, if the error is less than or equal to the first error threshold, the second device can send first information to the first device to instruct the first operation to be performed on the first AI model or first AI function. For another example, if the error is greater than or equal to the second error threshold, the first information may not be sent to the first device.

[0100] It can be understood that the different error thresholds in the above exemplary descriptions can be set according to circumstances, and this disclosure does not limit them.

[0101] The present disclosure sends second information to a second device so that the second device determines whether to perform relevant operations on the first AI model or the first AI function based on the second information, thereby enabling the first device to select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0102] In the AI ​​communication method provided by the embodiments of the present disclosure, the first AI model and the second AI model have an associated relationship; and / or the first AI function and the second AI function have an associated relationship.

[0103] In some embodiments, a first AI model deployed on a first device is associated with a second AI model deployed on a second device.

[0104] For example, the first AI model deployed on the first device can be a model combination (pair) with the second AI model deployed on the second device. For example, the first AI model and the second AI model can collaboratively process an AI task. Taking the application example of CSI compression as an example, the first AI model can be an encoder for compressing CSI information, and the second AI model can be a decoder for decompressing CSI information. The first AI model and the second AI model can be considered as bilateral models, that is, deployed on the first device and the second device respectively.

[0105] In some embodiments, a first AI function deployed on a first device is associated with a second AI function deployed on a second device.

[0106] For example, a first AI function deployed on a first device can form a function pair with a second AI function deployed on a second device. For example, the first AI function and the second AI function can collaboratively process an AI task. Taking the application example of CSI compression as an example, the AI ​​model corresponding to the first AI function can be an encoder for compressing CSI information, and the AI ​​model corresponding to the second AI function can be a decoder for decompressing CSI information. The AI ​​model corresponding to the first AI function and the AI ​​model corresponding to the second AI function can be considered as bilateral models, that is, they are deployed on the first device and the second device respectively.

[0107] The present disclosure is applicable to the case of a bilateral model, so that the second device and the first device can collaboratively process and determine whether to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0108] In the AI ​​communication method provided by the embodiments of the present disclosure, the first AI model and the second AI model have the same combination identifier; and / or the first AI function and the second AI function have the same combination identifier.

[0109] In some embodiments, the first AI model and the second AI model that are a pair may have the same combination identifier.

[0110] In some embodiments, the identifier may include an identity (ID) or an index (index)

[0111] For example, the first AI model and the second AI model may have the same pair ID.

[0112] In some embodiments, the first AI function and the second AI function that are a pair may have the same combination identifier.

[0113] For example, the first AI function and the second AI function may have the same pair ID.

[0114] The present disclosure is applicable to the case of a bilateral model, so that the second device and the first device can collaboratively process and determine whether to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0115] In the AI ​​communication method provided by the embodiments of the present disclosure, the first AI model and the second AI model are determined by negotiation between the first device and the second device; or, the first AI function and the second AI function are determined by negotiation between the first device and the second device; or, the first AI model and the second AI model are determined by the first device; or, the first AI function and the second AI function are determined by the first device; or, the first AI model and the second AI model are determined by the second device; or, the first AI function and the second AI function are determined by the second device.

[0116] In some embodiments, the first AI model and the second AI model may be determined by negotiation between the first device and the second device.

[0117] For example, a first device and a second device negotiate to monitor a first AI model and a second AI model. The first device uses a first data set to monitor the first AI model and sends a first detection result to the second device. The second device then detects the second AI model based on the first result and obtains a second result. This allows the first device to determine whether to perform a first operation on the first AI model.

[0118] For example, the first device and the second device can be negotiated and determined in advance through corresponding signaling. For example, the first device initiates a request to the second device to monitor the first AI model. The second device can inform the first device whether it is allowed to monitor the first AI model. For another example, the first device can initiate a request to the second device to ask which X AI models its own device is allowed to monitor. The second device determines which Y AI models out of X can be monitored based on its own situation. Wherein, X and Y are positive integers, and Y is less than or equal to X. It can be understood that the first device and the second device in the above exemplary description of the negotiation can be replaced with each other. In addition, the present disclosure is not limited to the above negotiation process, and a suitable negotiation method can be selected according to actual conditions to determine the first AI model and the second AI model.

[0119] In some embodiments, the first AI function and the second AI function are determined by negotiation between the first device and the second device.

[0120] For example, a first device and a second device negotiate to monitor a first AI function and a second AI function. The first device uses a first data set to monitor the first AI function and sends a first detection result to the second device. The second device then detects the second AI function based on the first result and obtains a second result. This allows the first device to determine whether to perform a first operation on the first AI function.

[0121] For example, the first device and the second device can be negotiated and determined in advance through corresponding signaling. For example, the first device initiates a request to the second device to monitor the first AI function. The second device can inform the first device whether it is allowed to monitor the first AI function. For another example, the first device can initiate a request to the second device to ask which X AI functions its own device is allowed to monitor. The second device determines which Y AI functions out of X can be monitored based on its own situation. Among them, X and Y are positive integers, and Y is less than or equal to X. It can be understood that the first device and the second device in the above exemplary description of the negotiation can be replaced with each other. In addition, the present disclosure is not limited to the above negotiation process, and an appropriate negotiation method can be selected according to actual conditions to determine the first AI function and the second AI function.

[0122] In some embodiments, the first AI model and the second AI model are determined by the first device.

[0123] For example, a first device determines to monitor a first AI model and a second AI model. The first device uses a first data set to monitor the first AI model and sends a first detection result to a second device. The second device then detects the second AI model based on the first result and obtains a second result. This allows the first device to determine whether to perform a first operation on the first AI model.

[0124] In some embodiments, the first AI function and the second AI function are determined by the first device.

[0125] For example, a first device determines to monitor a first AI function and a second AI function. The first device uses a first data set to monitor the first AI function and sends a first detection result to a second device. The second device then detects the second AI function based on the first result and obtains a second result. This allows the first device to determine whether to perform a first operation on the first AI function.

[0126] In some embodiments, the first AI model and the second AI model are determined by the second device.

[0127] For example, the second device determines to monitor the first AI model and the second AI model. In this case, the second device can send an instruction to the first device to instruct it to monitor the first AI model. Based on the instruction, the first device determines to monitor the first AI model using the first data set and sends the first detection result to the second device. The second device detects the second AI model based on the first result and obtains a second result. This allows the first device to determine whether to perform the first operation on the first AI model.

[0128] In some embodiments, the first AI function and the second AI function are determined by the second device.

[0129] For example, the second device determines to monitor a first AI function and a second AI function. In this case, the second device can send an instruction to the first device to instruct it to monitor the first AI function. Based on the instruction, the first device determines to monitor the first AI function using a first data set and sends a first detection result to the second device. The second device then detects the second AI function based on the first result and obtains a second result. This allows the first device to determine whether to perform the first operation on the first AI function.

[0130] The present disclosure provides multiple ways to instruct monitoring of a first AI model or a first AI function. Based on the monitoring results, a determination is then made as to whether to perform relevant operations on the first AI model or the first AI function. This allows the first device to select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0131] In the AI ​​communication method provided by the embodiment of the present disclosure, the second AI model is an AI model that has been activated by the second device.

[0132] In some embodiments, the second AI model may be an AI model that is currently activated by the second device.

[0133] For example, when the second device receives the second information sent by the first device, the second device may use the AI ​​model currently in use, that is, the AI ​​model activated by the second device, to perform an AI operation on the first result in the second information to obtain the second result.

[0134] That is to say, in this example, taking CSI compression as an application example, it is assumed that the second AI model is the AI ​​model that the second device should use. In this way, when the second result does not meet the second condition, it can be determined that the first AI model monitored by the first device does not meet the requirements, or it can be considered that the first AI model monitored by the first device and the second AI model currently being used by the second device do not belong to the same pair. Then, by controlling the correct second AI model, the correct first AI model is found. And the first device is instructed to perform a first operation on the correct first AI model through the first information. For example, the correct first AI model is activated, so that the first device uses the correct first AI model, and the second device uses the correct second AI model to perform subsequent corresponding tasks.

[0135] The present disclosure can utilize an AI model activated by a second device to monitor a first AI model or a first AI function. Based on the monitoring results, the second device can determine whether to perform relevant operations on the first AI model or the first AI function. This allows the first device to select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the AI ​​model or AI function.

[0136] In the AI ​​communication method provided in the embodiment of the present disclosure, FIG6 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG6, the method may further include the following steps:

[0137] In step S51, third information is received.

[0138] In some embodiments, the first device may receive third information, which may be used to indicate the second result.

[0139] For example, the first device receives third information sent by the second device. The third information indicates the second result. The first device can determine the second result obtained by the second device based on the first result based on the third information.

[0140] In step S52 , a first AI model or a first AI function is determined based on the second result.

[0141] In some embodiments, the first device determines a first AI model or a first AI function based on the second result received in S51 .

[0142] That is, the first device may determine, based on the second result, that a first AI model or a first AI function is required to perform the first operation.

[0143] Of course, the first device may also determine the first AI model or first AI function that needs to perform the first operation based on the first result and the second result.

[0144] In step S53, the fourth information is sent.

[0145] In some embodiments, the first device may send fourth information to the second device, wherein the fourth information is used to indicate that the first device is ready to perform a first operation on the first AI model or the first AI function.

[0146] It is understood that, when determining the first AI model or the first AI function, the first device may inform the second device of the determined first AI model or the first AI function, so that the second device can ultimately determine whether to perform the first operation on the first AI model or the first AI function.

[0147] In the present disclosure, a first device notifies a second device of its first AI model or first AI function that it is preparing to perform a first operation, so that the second device can ultimately determine whether to perform the relevant operation on the first AI model or first AI function. This enables the first device to select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0148] In the AI ​​communication method provided in the embodiments of the present disclosure, FIG7 is a flowchart of another AI communication method according to an exemplary embodiment. As shown in FIG7 , in S23, if the first result satisfies the first condition, performing the first operation on the first AI model or the first AI function may further include the following steps:

[0149] In step S61, third information is received.

[0150] In some embodiments, the first device may receive third information sent by the second device, where the third information is used to indicate the second result.

[0151] It can be understood that the implementation process of S61 is the same as that of S51. For details, please refer to the description of each embodiment and its related embodiments in S51, and this disclosure will not go into details here.

[0152] In step S62 , it is determined to perform a first operation on the first AI model or the first AI function based on the second result.

[0153] In some embodiments, the first device may directly determine to perform a first operation on the first AI model or the first AI function based on the second result.

[0154] That is, the first device can determine, based on the second result, that a first AI model or first AI function needs to perform a first operation. Of course, the first device can also determine, based on the first result and the second result, that a first AI model or first AI function needs to perform a first operation. The first device then determines to perform the first operation on the first AI model or first AI function.

[0155] The first device disclosed herein can select a suitable AI model or AI function to perform a corresponding task based on the second result, thereby improving the accuracy of using the AI ​​model or AI function.

[0156] In the AI ​​communication method provided in the embodiment of the present disclosure, FIG8 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG8 , the method may further include the following steps:

[0157] In step S71, the fifth information is sent.

[0158] In some embodiments, the first device may send fifth information to the second device. The fifth information may be used to indicate that the first device performs a first operation on the first AI model or the first AI function.

[0159] For example, when the first device determines to perform the first operation on the first AI model or the first AI function based on the second result and / or the first result, the first device may inform the second device that the first device performs the first operation on the first AI model or the first AI function.

[0160] The first device of the present disclosure may also inform the second device of the first AI model or first AI function for performing the first operation. The first device may select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0161] In the AI ​​communication method provided by the embodiments of the present disclosure, S22 uses a first AI model or an AI model associated with a first AI function, and performs an AI operation based on a first data set to obtain a first result, which may include: using multiple first AI models or multiple AI models associated with first AI functions, and performing AI operations based on the first data set to obtain multiple first results.

[0162] In some embodiments, when the first device determines a first data set and inputs the first data set into a first AI model or an AI model associated with a first AI function to perform an AI operation, multiple AI models or multiple AI models associated with the first AI function may be used, thereby obtaining multiple first results.

[0163] For example, the same first dataset can be used to monitor different first AI models, such as a first AI model applicable to macrocells and a first AI model applicable to microcells. It is understood that different first AI models can input the same first dataset, but because the models themselves are different, they can each generate corresponding first AI results.

[0164] For another example, using the same first data set, AI models associated with different first AI functions can be monitored separately. For example, an AI model associated with a first AI function for CSI compression, an AI model associated with a first AI function for beam management, and so on. It will be appreciated that AI models associated with different first AI functions can input the same first data set, but because the models themselves are different, corresponding first AI results can be obtained separately.

[0165] Of course, it is clear that the above operations can also be performed using multiple different first data sets. However, for each first data set, multiple first AI models or AI models associated with multiple first AI functions can be monitored.

[0166] The present disclosure can use the same first data set to monitor multiple AI models or AI functions to determine whether to perform relevant operations on the first AI model or the first AI function. This allows the first device to select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0167] In the AI ​​communication method provided in the embodiment of the present disclosure, FIG9 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG9, the method may further include the following steps:

[0168] In step S81, sixth information is received.

[0169] In some embodiments, the first device may receive sixth information sent by the second device. The sixth information is used to instruct the first device to monitor the first AI model or the first AI function. The first device may monitor the first AI model or the first AI function based on the received sixth information.

[0170] For example, the sixth information is a request for instructing the first device to monitor the first AI model or the first AI function.

[0171] In some embodiments, the first device may receive sixth information sent by the second device. The sixth information is used to instruct the first device to assist in testing the first AI model or the first AI function. The first device may assist in testing the first AI model or the first AI function based on the received sixth information. It will be understood that in this embodiment, assisting in testing can be understood as the second device determining to monitor the AI ​​model or AI function deployed on the second device, but considering that the AI ​​model is a bilateral model, the first device is required to assist in testing.

[0172] For example, the sixth information is a request for instructing the first device to assist in testing the first AI model or the first AI function. The request may include an identifier or type corresponding to the first AI model or the first AI function to indicate the first AI model or the first AI function.

[0173] The present disclosure can receive a monitoring request or an assistance test request sent by a second device, thereby monitoring multiple AI models or AI functions, and determining whether to perform relevant operations on a first AI model or a first AI function, so that the first device can select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0174] In the AI ​​communication method provided in the embodiments of the present disclosure, FIG10 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG10 , the method may further include the following steps:

[0175] In step S91, the sixth information is sent.

[0176] In some embodiments, the first device may send sixth information to the second device, wherein the sixth information is used to instruct the second device to monitor the first AI model or the first AI function, so that the second device can monitor the first AI model or the first AI function based on the received sixth information.

[0177] For example, the sixth information is a request for instructing the second device to monitor the first AI model or the first AI function.

[0178] In some embodiments, the first device may send sixth information to the second device. The sixth information is used to instruct the second device to assist in testing the first AI model or the first AI function. This allows the second device to assist in testing the first AI model or the first AI function based on the received sixth information. It will be appreciated that in this embodiment, assisting in testing can be understood as the first device determining to monitor the AI ​​model or AI function deployed on the first device. However, considering that the AI ​​model is a bilateral model, assisting in testing by the second device is required.

[0179] For example, the sixth information is a request for instructing the second device to assist in testing the first AI model or the first AI function. The request may include an identifier or type corresponding to the first AI model or the first AI function to indicate the first AI model or the first AI function.

[0180] It is understood that the sixth information can be used to instruct the peer device to monitor the first AI model or the first AI function. Alternatively, the sixth information can be used to instruct the peer device to assist in testing the first AI model or the first AI function. In other words, when the local device is the first device, the peer device is the second device. When the local device is the second device, the peer device is the first device.

[0181] The present disclosure can receive a monitoring request or an assistance test request sent by a second device, thereby monitoring multiple AI models or AI functions, and determining whether to perform relevant operations on a first AI model or a first AI function, so that the first device can select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0182] In the AI ​​communication method provided by the embodiment of the present disclosure, the first data set is determined based on at least one of the following methods: sending seventh information, where the seventh information is used to instruct the second device to send the first data set to the first device; or obtaining the first data set based on broadcasting.

[0183] In some embodiments, the first device may send seventh information, where the seventh information is used to instruct the second device to send the first data set to the first device.

[0184] For example, when the first device is a terminal and the second device is a network device, the terminal may send seventh information to the network device. The seventh information may instruct the network device to send the first data set to the terminal.

[0185] Of course, for the network device, the first data set may be pre-stored in the network device.

[0186] In some embodiments, the first device may obtain the first data set based on the broadcast.

[0187] For example, when the first device is a terminal and the second device is a network device, the terminal may obtain the first data set based on broadcasting. For example, the network device may send the first data set by broadcasting.

[0188] The present disclosure provides multiple methods for obtaining a first data set, so as to determine whether to perform relevant operations on a first AI model or a first AI function based on the first data set, thereby enabling the first device to select a suitable AI model or AI function to perform a corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0189] In the AI ​​communication method provided by the embodiment of the present disclosure, the first operation includes at least one of the following: an activation operation; a deactivation operation; and an AI switching operation.

[0190] In some embodiments, the first device may determine to activate a first AI model or a first AI function.

[0191] For example, the first device may determine to activate a first AI model or a first AI function.

[0192] In some embodiments, the first device may determine to deactivate the first AI model or the first AI function.

[0193] For example, the first device may determine to deactivate the first AI model or the first AI function.

[0194] In some embodiments, the first device may determine to perform an AI switching operation on the first AI model or the first AI function.

[0195] For example, the first device may determine to switch the currently activated AI model to the first AI model or the first AI function.

[0196] The present disclosure is applicable to performing various operations on a first AI model or a first AI function, so that the first device can select a suitable AI model or AI function to perform a corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0197] In the AI ​​communication method provided in the embodiments of the present disclosure, FIG11 is a flow chart of another AI communication method according to an exemplary embodiment. As shown in FIG11 , the first operation is an AI switching operation, and the first information is used to indicate the first operation to be performed on the first AI model or the first AI function. The method may further include the following steps:

[0198] In step S101 , it is determined whether an AI switching operation is performed on a first AI model or a first AI function.

[0199] In some embodiments, when the first operation is an AI switching operation, the first information received by the first device may be used to indicate that the AI ​​switching operation is to be performed on the first AI model or the first AI function. In this case, the first device may determine whether to perform the AI ​​switching operation on both the first AI model and the first AI function.

[0200] In other words, the second device may decide to perform the AI ​​switching operation on the first AI model or the first AI function. The first device may determine whether to accept the second device's decision based on the second device's decision and in combination with various factors such as its own software and hardware support, the current operating environment support, and the battery level of the first device. In other words, the first device may decide whether to perform the first operation.

[0201] When the first device decides to perform an AI switching operation on the first AI model or the first AI function, it may also inform the second device so that the second device activates or switches to a second AI model associated with the first AI model.

[0202] In the present disclosure, when a second device decides to perform a first operation on a first AI model or a first AI function, the first device can also independently decide whether to execute the operation. This enables the first device to select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0203] In the AI ​​communication method provided by the embodiment of the present disclosure, the first device is a terminal and the second device is a network device.

[0204] In some embodiments, the first device may be a terminal, and the second device may be a network device.

[0205] The present disclosure determines to perform relevant operations on the first AI model or the first AI function, so that the terminal can select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0206] In the AI ​​communication method provided by the embodiment of the present disclosure, the first device is a network device and the second device is a terminal.

[0207] In some embodiments, the first device may be a network device, and the second device may be a terminal.

[0208] The present disclosure determines to perform relevant operations on a first AI model or a first AI function, thereby enabling a network device to select a suitable AI model or AI function to perform a corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0209] It should be noted that those skilled in the art will appreciate that the various implementation methods / embodiments involved in the embodiments of the present disclosure can be used in conjunction with the aforementioned embodiments or can be used independently. Whether used alone or in conjunction with the aforementioned embodiments, the implementation principles are similar. In the implementation of the present disclosure, some embodiments are described in terms of implementation methods used together. Of course, those skilled in the art will appreciate that such examples are not limitations on the embodiments of the present disclosure.

[0210] Based on the same concept, the embodiments of the present disclosure also provide an AI communication device and equipment.

[0211] It is understandable that the AI ​​communication devices and equipment provided by the embodiments of the present disclosure include hardware structures and / or software modules corresponding to the execution of each function in order to realize the above functions. In combination with the units and algorithm steps of the various examples disclosed in the embodiments of the present disclosure, the embodiments of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiments of the present disclosure.

[0212] Figure 12 is a schematic diagram of an AI communication device according to an exemplary embodiment. Referring to Figure 12 , the device 200 includes: a processing module 201, configured to determine a first operation to be performed on a first artificial intelligence (AI) model or a first AI function, wherein the first AI model or the first AI function is deployed on a first device.

[0213] The present disclosure determines to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0214] In one embodiment, the processing module 201 is further used to: determine a first data set; use the first AI model or the AI ​​model associated with the first AI function, and perform an AI operation based on the first data set to obtain a first result; and determine to perform a first operation on the first AI model or the first AI function based on the first result.

[0215] The present disclosure determines whether to perform relevant operations on the first AI model or the first AI function through the first data set, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0216] In one embodiment, the processing module 201 is further configured to: determine to perform a first operation on the first AI model or the first AI function when the first result satisfies a first condition.

[0217] The present disclosure determines whether to perform relevant operations on the first AI model or the first AI function when the first result satisfies the first condition, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0218] In one embodiment, the device 200 may further include: a receiving module 202, used to receive first information sent by a second device; wherein the first information is used to indicate that a first operation is performed on the first AI model or the first AI function; or, the first information is used to indicate that the first operation is not performed on the first AI model or the first AI function.

[0219] The present disclosure uses the first information sent by the second device to indicate whether to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0220] In one embodiment, the device 200 also includes: a sending module 203, used to send second information, wherein the second information is used to indicate the first result; the first result satisfies the first condition, including: the second result satisfies the second condition, wherein the second result is obtained by the second device using the second AI model or the AI ​​model associated with the second AI function and based on the first result, and the second AI model or the second AI function is deployed on the second device.

[0221] The present disclosure sends second information to a second device so that the second device determines whether to perform relevant operations on the first AI model or the first AI function based on the second information, thereby enabling the first device to select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0222] In one embodiment, the first AI model and the second AI model are associated; and / or the first AI function and the second AI function are associated.

[0223] The present disclosure is applicable to the case of a bilateral model, so that the second device and the first device can collaboratively process and determine whether to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0224] In one embodiment, the first AI model and the second AI model have the same combination identifier; and / or the first AI function and the second AI function have the same combination identifier.

[0225] The present disclosure is applicable to the case of a bilateral model, so that the second device and the first device can collaboratively process and determine whether to perform relevant operations on the first AI model or the first AI function, so that the first device can select a suitable AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0226] In one embodiment, the first AI model and the second AI model are determined by negotiation between the first device and the second device; or, the first AI function and the second AI function are determined by negotiation between the first device and the second device; or, the first AI model and the second AI model are determined by the first device; or, the first AI function and the second AI function are determined by the first device; or, the first AI model and the second AI model are determined by the second device; or, the first AI function and the second AI function are determined by the second device.

[0227] The present disclosure provides multiple ways to instruct monitoring of a first AI model or a first AI function. Based on the monitoring results, a determination is then made as to whether to perform relevant operations on the first AI model or the first AI function. This allows the first device to select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0228] In one embodiment, the second AI model is an activated AI model of the second device.

[0229] The present disclosure can utilize an AI model activated by a second device to monitor a first AI model or a first AI function. Based on the monitoring results, the second device can determine whether to perform relevant operations on the first AI model or the first AI function. This allows the first device to select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the AI ​​model or AI function.

[0230] In one embodiment, the receiving module 202 is further used to receive third information, where the third information is used to indicate the second result; the processing module 201 is further used to determine the first AI model or the first AI function based on the second result; and the sending module 203 is further used to send fourth information, where the fourth information is used to indicate that the first device is ready to perform a first operation on the first AI model or the first AI function.

[0231] In the present disclosure, a first device notifies a second device of its first AI model or first AI function that it is preparing to perform a first operation, so that the second device can ultimately determine whether to perform the relevant operation on the first AI model or first AI function. This enables the first device to select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0232] In one embodiment, when the first result satisfies the first condition, the receiving module 202 is further used to receive third information, where the third information is used to indicate the second result; the processing module 201 is further used to determine a first operation on the first AI model or the first AI function based on the second result.

[0233] The first device disclosed herein can select a suitable AI model or AI function to perform a corresponding task based on the second result, thereby improving the accuracy of using the AI ​​model or AI function.

[0234] In one embodiment, the sending module 203 is further used to: send fifth information, where the fifth information is used to indicate that the first device performs a first operation on the first AI model or the first AI function.

[0235] The first device of the present disclosure may also inform the second device of the first AI model or first AI function for performing the first operation. The first device may select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0236] In one embodiment, the processing module 201 is further configured to: use multiple first AI models or multiple AI models associated with the first AI functions, and perform AI operations based on the first data set to obtain multiple first results.

[0237] The present disclosure can use the same first data set to monitor multiple AI models or AI functions to determine whether to perform relevant operations on the first AI model or the first AI function. This allows the first device to select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0238] In one embodiment, the receiving module 202 is further used to: receive sixth information; wherein the sixth information is used to instruct the first device to monitor the first AI model or the first AI function; or, the sixth information is used to instruct the first device to assist in testing the first AI model or the first AI function.

[0239] The present disclosure can receive a monitoring request or an assistance test request sent by a second device, thereby monitoring multiple AI models or AI functions, and determining whether to perform relevant operations on a first AI model or a first AI function, so that the first device can select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0240] In one embodiment, the sending module 203 is further used to: send sixth information; wherein the sixth information is used to instruct the second device to monitor the first AI model or the first AI function; or, the sixth information is used to instruct the second device to assist in testing the first AI model or the first AI function.

[0241] The present disclosure can receive a monitoring request or an assistance test request sent by a second device, thereby monitoring multiple AI models or AI functions, and determining whether to perform relevant operations on a first AI model or a first AI function, so that the first device can select an appropriate AI model or AI function to perform a corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0242] In one embodiment, the first data set is determined based on at least one of the following methods: sending seventh information, where the seventh information is used to instruct the second device to send the first data set to the first device; or acquiring the first data set based on broadcasting.

[0243] The present disclosure provides multiple methods for obtaining a first data set, so as to determine whether to perform relevant operations on a first AI model or a first AI function based on the first data set, thereby enabling the first device to select a suitable AI model or AI function to perform a corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0244] In one embodiment, the first operation includes at least one of the following: an activation operation; a deactivation operation; and an AI switching operation.

[0245] The present disclosure is applicable to performing various operations on a first AI model or a first AI function, so that the first device can select a suitable AI model or AI function to perform a corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0246] In one embodiment, the first operation is an AI switching operation, the first information is used to indicate that the first operation is performed on the first AI model or the first AI function, and the processing module 201 is further used to determine whether to perform the AI ​​switching operation on the first AI model or the first AI function.

[0247] In the present disclosure, when a second device decides to perform a first operation on a first AI model or a first AI function, the first device can also independently decide whether to execute the operation. This enables the first device to select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of the use of the AI ​​model or AI function.

[0248] In one implementation, the first device is a terminal, and the second device is a network device.

[0249] The present disclosure determines to perform relevant operations on the first AI model or the first AI function, so that the terminal can select an appropriate AI model or AI function to perform the corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0250] In one implementation, the first device is a network device, and the second device is a terminal.

[0251] The present disclosure determines to perform relevant operations on a first AI model or a first AI function, thereby enabling a network device to select a suitable AI model or AI function to perform a corresponding task, thereby improving the accuracy of using the AI ​​model or AI function.

[0252] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0253] Figure 13 is a schematic diagram of an AI communication device according to an exemplary embodiment. For example, device 300 can be any terminal such as a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0254] 13 , device 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 312 , a sensor component 314 , and a communication component 316 .

[0255] The processing component 302 generally controls the overall operation of the device 300, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. The processing component 302 may include one or more processors 320 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 302 may include one or more modules to facilitate interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate interaction between the multimedia component 308 and the processing component 302.

[0256] The memory 304 is configured to store various types of data to support operations on the device 300. Examples of such data include instructions for any application or method operating on the device 300, contact data, phone book data, messages, pictures, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0257] The power component 306 provides power to the various components of the device 300. The power component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 300.

[0258] The multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensor can not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 includes a front camera and / or a rear camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0259] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC) that is configured to receive external audio signals when the device 300 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 304 or transmitted via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.

[0260] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0261] The sensor assembly 314 includes one or more sensors for providing various aspects of the status assessment of the device 300. For example, the sensor assembly 314 can detect the open / closed state of the device 300, the relative positioning of components, such as the display and keypad of the device 300. The sensor assembly 314 can also detect changes in the position of the device 300 or a component of the device 300, the presence or absence of user contact with the device 300, the orientation or acceleration / deceleration of the device 300, and temperature changes of the device 300. The sensor assembly 314 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 314 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 can also include an accelerometer, a gyroscope, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0262] The communication component 316 is configured to facilitate wired or wireless communication between the device 300 and other devices. The device 300 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0263] In an exemplary embodiment, the device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described methods.

[0264] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, which can be executed by the processor 320 of the device 300 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0265] FIG14 is a schematic diagram of another AI communication device according to an exemplary embodiment. For example, device 400 may be provided as a base station or a server. Referring to FIG14 , device 400 includes a processing component 422, which further includes one or more processors, and a memory resource represented by memory 432 for storing instructions executable by processing component 422, such as an application. The application stored in memory 432 may include one or more modules, each corresponding to a set of instructions. In addition, processing component 422 is configured to execute instructions to perform the above method.

[0266] The device 400 may also include a power supply component 426 configured to perform power management of the device 400, a wired or wireless network interface 450 configured to connect the device 400 to a network, and an input / output (I / O) interface 458. The device 400 may operate based on an operating system stored in the memory 432, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0267] The present disclosure can determine the target model to be activated based on multiple first data sets and the test results of the first data sets. It can also implement a method for monitoring inactivated models or inactivated AI functions to ensure that the network or terminal can select the correct AI model for activation.

[0268] It is further understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the related objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0269] It will be further understood that the terms "first," "second," and the like are used to describe various types of information, but such information should not be limited to these terms. These terms are used solely to distinguish information of the same type from one another and do not indicate a particular order or level of importance. In fact, the terms "first," "second," and the like are fully interchangeable. For example, first information could be referred to as second information, and similarly, second information could be referred to as first information without departing from the scope of this disclosure.

[0270] It is further understood that the meanings of words such as "in response to" and "if" involved in this disclosure depend on the context and the actual usage scenarios. For example, the word "in response to" used herein can be interpreted as "at..." or "when..." or "if" or "if".

[0271] It is further understood that although operations are described in a particular order in the drawings in the embodiments of the present disclosure, this should not be construed as requiring that the operations be performed in the particular order shown or in a serial order, or that all of the operations shown be performed to obtain the desired results. In certain circumstances, multitasking and parallel processing may be advantageous.

[0272] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein.

[0273] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the scope of the appended claims.

Claims

1. An artificial intelligence AI communication method, It is characterized in that The method is performed by a first device and includes: Determine to perform a first operation on a first AI model or a first AI function, wherein the first AI model or the first AI function is deployed on the first device.

2. The method according to claim 1, It is characterized in that The method further comprises: Determine a first data set, wherein the first data set is a data set used for testing an AI model; Using the first AI model or the AI ​​model associated with the first AI function, and performing an AI operation based on the first data set to obtain a first result, wherein the AI ​​operation refers to an operation of performing corresponding processing on the AI ​​model to obtain the first result; The determining to perform a first operation on the first AI model or the first AI function includes: Determine, based on the first result, to perform a first operation on the first AI model or the first AI function.

3. The method according to claim 2, It is characterized in that The determining, based on the first result, to perform a first operation on the first AI model or the first AI function includes: When the first result satisfies a first condition, determine to perform a first operation on the first AI model or the first AI function.

4. The method according to any one of claims 1 to 3, It is characterized in that The determining to perform a first operation on the first AI model or the first AI function includes: Receive first information sent by a second device; wherein the first information is used to indicate that a first operation is performed on the first AI model or the first AI function; or, the first information is used to indicate that the first operation is not performed on the first AI model or the first AI function.

5. The method according to claim 4, It is characterized in that The method further comprises: Sending second information, wherein the second information is used to indicate the first result; The first result satisfies the first condition including: The second result satisfies the second condition, wherein the second result is obtained by the second device using the second AI model or the AI ​​model associated with the second AI function and based on the first result, and the second AI model or the second AI function is deployed on the second device.

6. The method according to any one of claims 5, It is characterized in that The first AI model and the second AI model have an association relationship; and / or, The first AI function and the second AI function have an associated relationship.

7. The method according to claim 6, It is characterized in that The first AI model and the second AI model have the same combination identifier; and / or, The first AI function and the second AI function have the same combination identifier.

8. The method according to claim 6 or 7, It is characterized in that The first AI model and the second AI model are determined by negotiation between the first device and the second device; or, The first AI function and the second AI function are determined by negotiation between the first device and the second device; or, The first AI model and the second AI model are determined by the first device; or, The first AI function and the second AI function are determined by the first device; or, The first AI model and the second AI model are determined by the second device; or, The first AI function and the second AI function are determined by the second device.

9. The method according to claim 5, It is characterized in that The second AI model is an activated AI model of the second device.

10. The method according to claim 5, It is characterized in that The method further comprises: receiving third information, where the third information is used to indicate the second result; Determine the first AI model or the first AI function based on the second result; Send fourth information, where the fourth information is used to indicate that the first device is ready to perform a first operation on the first AI model or the first AI function.

11. The method according to claim 5, It is characterized in that The performing a first operation on the first AI model or the first AI function when the first result satisfies the first condition includes: receiving third information, where the third information is used to indicate the second result; Determine, based on the second result, to perform a first operation on the first AI model or the first AI function.

12. The method according to claim 11, It is characterized in that The method further comprises: Send fifth information, where the fifth information is used to indicate that the first device performs a first operation on the first AI model or the first AI function.

13. The method according to claim 2 or 3, It is characterized in that The using the first AI model or the AI ​​model associated with the first AI function and performing an AI operation based on the first data set to obtain a first result includes: Use multiple AI models associated with the first AI functions or multiple AI models associated with the first AI functions, and perform AI operations based on the first data set to obtain multiple first results.

14. The method according to any one of claims 1 to 13, It is characterized in that The method further comprises: receiving sixth information; The sixth information is used to instruct the first device to monitor the first AI model or the first AI function; or, the sixth information is used to instruct the first device to assist in testing the first AI model or the first AI function.

15. The method according to any one of claims 1 to 13, It is characterized in that The method further comprises: Send the sixth message; The sixth information is used to instruct the second device to monitor the first AI model or the first AI function; or, the sixth information is used to instruct the second device to assist in testing the first AI model or the first AI function.

16. The method according to claim 2 or 3, It is characterized in that The first data set is determined based on at least one of the following methods: sending seventh information, where the seventh information is used to instruct the second device to send the first data set to the first device; The first data set is obtained based on the broadcast.

17. The method according to any one of claims 1 to 15, It is characterized in that The first operation includes at least one of the following: Activate the operation; Deactivate operation; AI switching operation.

18. The method according to claim 4, It is characterized in that The first operation is an AI switching operation, the first information is used to instruct to perform the first operation on the first AI model or the first AI function, and the method further includes: Determine whether to perform the AI ​​switching operation on the first AI model or the first AI function.

19. The method according to any one of claims 1 to 18, It is characterized in that The first device is a terminal, and the second device is a network device.

20. The method according to any one of claims 1 to 18, It is characterized in that The first device is a network device, and the second device is a terminal.

21. An artificial intelligence (AI) communication device, It is characterized in that The device comprises: A processing module is used to determine to perform a first operation on a first AI model or a first AI function, wherein the first AI model or the first AI function is deployed on a first device.

22. An artificial intelligence (AI) communication device, It is characterized in that include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: execute the method described in any one of claims 1 to 20.

23. A non-transitory computer-readable storage medium, It is characterized in that When the instructions in the storage medium are executed by a processor of the first device, the first device is enabled to execute the method according to any one of claims 1 to 20.

Images