CN117135650A

CN117135650A - Artificial intelligent model configuration method, device, terminal and network equipment

Info

Publication number: CN117135650A
Application number: CN202210555004.0A
Authority: CN
Inventors: 张嘉真; 左君; 曹昱华
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Communications Ltd Research Institute
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Communications Ltd Research Institute
Priority date: 2022-05-20
Filing date: 2022-05-20
Publication date: 2023-11-28

Abstract

The invention provides an artificial intelligent model configuration method, an artificial intelligent model configuration device, a terminal and network equipment. The artificial intelligence model configuration method is applied to a terminal and comprises the following steps: acquiring artificial intelligent model configuration information sent by network equipment or preconfigured in the terminal, wherein the artificial intelligent model configuration information comprises one or more of the following: model identification, applicable function, applicable scenario, and demand computing power resource information. The invention realizes the configuration of the artificial intelligent model; in addition, the artificial intelligent model configuration information can be used for the terminal and/or the network equipment to quickly determine the artificial intelligent model matched with the current requirement, so that better performance is obtained, and the terminal does not need to wait for the network equipment to issue new artificial intelligent models or new artificial intelligent model configuration information again, so that quick response of new target service is realized.

Description

Artificial intelligent model configuration method, device, terminal and network equipment

Technical Field

The embodiment of the invention relates to the technical field of wireless communication, in particular to an artificial intelligent model configuration method, an artificial intelligent model configuration device, a terminal and network equipment.

Background

Artificial intelligence (Artifical Intelligence, AI for short) is widely applied to mobile communication terminals, which can realize different AI services, such as man-machine interaction, face recognition, shooting beautification, games, etc., according to corresponding artificial intelligence models. Specifically, according to the cooperation degree of the mobile communication terminal and the network device, the cooperation manner may be divided into: 1) The artificial intelligent model may be located at one side of the mobile communication terminal or the network device, and no auxiliary information needs to be interacted; 2) The artificial intelligent model may be located at one side of the mobile communication terminal or the network device, and auxiliary information needs to be interacted at two sides; 3) The artificial intelligent model is deployed on both sides of the mobile communication terminal and the network equipment, and related reasoning is performed by cooperation of the network side and the terminal side.

For cooperative modes 2) and 3), the artificial intelligence model of the mobile communication terminal is typically a typical generic model issued by the network device prior to initiating any AI services. With the change of the AI service or the attribute of the mobile communication terminal, the original artificial intelligent model may not be applicable any more, and the flow of the network device to issue the new artificial intelligent model again is longer, so that the mobile communication terminal and the network device have a relatively difficult quick response to the new AI service.

Disclosure of Invention

The embodiment of the invention provides an artificial intelligence model configuration method, an artificial intelligence model configuration device, an artificial intelligence model configuration terminal and network equipment, which are used for solving the technical problems that the original artificial intelligence model is not applicable any more due to the change of the AI service or the attribute of a mobile communication terminal, and the network equipment has a longer flow for issuing a new artificial intelligence model again, so that the mobile communication terminal and the network equipment are difficult to respond to the new AI service quickly.

In order to solve the technical problems, the invention is realized as follows:

in a first aspect, an embodiment of the present invention provides an artificial intelligence model configuration method, which is applied to a terminal, and includes:

acquiring artificial intelligent model configuration information sent by network equipment or preconfigured in the terminal, wherein the artificial intelligent model configuration information comprises one or more of the following:

model identification, applicable function, applicable scenario, and demand computing power resource information.

Optionally, the demand computing force resource information includes one or more of:

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

Optionally, the artificial intelligence model configuration information is sent through radio resource control signaling.

Optionally, the method further comprises:

and reporting first indication information to the network equipment, wherein the first indication information comprises a model identifier of the artificial intelligent model.

Optionally, the first indication information is sent through at least one of the following:

radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

Optionally, the method further comprises:

and reporting second indicating information to the network equipment, wherein the second indicating information comprises a model identification of the deactivated artificial intelligent model and/or an applicable function of the deactivated artificial intelligent model.

Optionally, the method further comprises:

receiving third indication information sent by the network device, wherein the third indication information comprises a model identification of the activated artificial intelligence model and/or optional applicable functions of the activated artificial intelligence model, and the method further comprises:

receiving fourth indication information sent by the network equipment, wherein the fourth indication information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model;

the fourth indication information and the third indication information are sent through the same type of signaling or through different types of signaling.

In a second aspect, an embodiment of the present invention provides an artificial intelligence model configuration method, which is applied to a network device, and includes:

and sending artificial intelligent model configuration information to the terminal, wherein the artificial intelligent model configuration information comprises one or more of the following:

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

Optionally, the method further comprises:

and receiving first indication information reported by the terminal, wherein the first indication information comprises a model identifier of an artificial intelligent model.

Optionally, the first indication information is sent through at least one of the following signaling:

radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

Optionally, the method further comprises:

and receiving second indicating information reported by the terminal, wherein the second indicating information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model.

Optionally, the method further comprises:

and sending third indication information to the terminal, wherein the third indication information comprises a model identification of the activated artificial intelligence model and/or an applicable function of the activated artificial intelligence model.

Optionally, the method further comprises:

sending fourth indication information to the terminal, wherein the fourth indication information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model;

In a third aspect, an embodiment of the present invention provides a terminal, including a processor, where the processor is configured to:

In a fourth aspect, an embodiment of the present invention provides a network device, including a transceiver, where the transceiver is configured to:

In a fifth aspect, an embodiment of the present invention provides an artificial intelligence model configuration apparatus, which is applied to a terminal, including:

the acquisition module is used for acquiring the artificial intelligent model configuration information sent by the network equipment or preconfigured in the terminal, wherein the artificial intelligent model configuration information comprises one or more of the following:

In a sixth aspect, an embodiment of the present invention provides an artificial intelligence model configuration apparatus, which is applied to a network device, including:

the sending module is used for sending the artificial intelligent model configuration information to the terminal, wherein the artificial intelligent model configuration information comprises one or more of the following:

In a seventh aspect, an embodiment of the present invention provides a terminal, including: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the artificial intelligence model configuration method as described in the first aspect.

In an eighth aspect, an embodiment of the present invention provides a network device, including: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the artificial intelligence model configuration method as described in the second aspect.

In a ninth aspect, an embodiment of the present invention provides a computer readable storage medium, on which a computer program is stored, the computer program implementing the steps of the artificial intelligence model configuration method according to the first aspect when being executed by a processor; alternatively, the computer program when executed by a processor implements the steps of the artificial intelligence model configuration method as described in the second aspect.

Compared with the prior art, the method, the device, the terminal and the network equipment for configuring the artificial intelligent model provided by the embodiment of the invention have the advantages that the terminal can acquire the artificial intelligent model configuration information sent by the network equipment or preconfigured on the terminal, and the artificial intelligent model configuration information comprises one or more of the following items: model identification, applicable functions, applicable scenarios, and demand computing power resource information to enable configuration of an artificial intelligence model. In addition, the artificial intelligent model configuration information can be used for the terminal and/or the network equipment to quickly determine the artificial intelligent model matched with the current requirement, so that better performance is obtained, and the terminal does not need to wait for the network equipment to issue new artificial intelligent models or new artificial intelligent model configuration information again, so that quick response of new target service is realized.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:

fig. 1 is a schematic diagram of a wireless communication system according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 3 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 4 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 5 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

fig. 6 is an exemplary diagram of a MAC-CE provided by an embodiment of the present invention;

fig. 7 is another exemplary diagram of a MAC-CE provided by an embodiment of the present invention;

FIG. 8 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 9 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 10 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 11 is a schematic flow chart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 12 is a flowchart of an artificial intelligence model configuration method according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram of an artificial intelligence model configuration device according to an embodiment of the present invention;

FIG. 14 is a schematic structural diagram of an artificial intelligence model configuration device according to an embodiment of the present invention;

FIG. 15 is a schematic structural diagram of an artificial intelligence model configuration device according to an embodiment of the present invention;

FIG. 16 is a schematic diagram of an artificial intelligence model configuration device according to an embodiment of the present invention;

FIG. 17 is a schematic structural diagram of an artificial intelligence model configuration device according to an embodiment of the present invention;

FIG. 18 is a schematic diagram of an artificial intelligence model configuration device according to an embodiment of the present invention;

fig. 19 is a schematic structural diagram of a terminal according to an embodiment of the present invention;

fig. 20 is a schematic structural diagram of a network device according to an embodiment of the present invention;

fig. 21 is a schematic structural diagram of a terminal according to an embodiment of the present invention;

Fig. 22 is a schematic structural diagram of a network device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The techniques described herein are not limited to NR systems and long term evolution (Long Time Evolution, LTE)/LTE evolution (LTE-Advanced, LTE-a) systems and may also be used for various wireless communication systems such as code division multiple access (Code Division Multiple Access, CDMA), time division multiple access (Time Division Multiple Access, TDMA), frequency division multiple access (Frequency Division Multiple Access, FDMA), orthogonal frequency division multiple access (Orthogonal Frequency Division Multiple Access, OFDMA), single-carrier frequency division multiple access (Single-carrier Frequency-Division Multiple Access, SC-FDMA), and other systems. The terms "system" and "network" are often used interchangeably. A CDMA system may implement radio technologies such as CDMA2000, universal terrestrial radio access (Universal Terrestrial Radio Access, UTRA), and the like. UTRA includes wideband CDMA (Wideband Code Division Multiple Access, WCDMA) and other CDMA variants. TDMA systems may implement radio technologies such as the global system for mobile communications (Global System for Mobile Communication, GSM). OFDMA systems may implement radio technologies such as ultra mobile broadband (UltraMobile Broadband, UMB), evolved UTRA (E-UTRA), IEEE 802.21 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, flash-OFDM, and the like. UTRA and E-UTRA are parts of the universal mobile telecommunications system (Universal Mobile Telecommunications System, UMTS). LTE and higher LTE (e.g., LTE-a) are new UMTS releases that use E-UTRA. UTRA, E-UTRA, UMTS, LTE, LTE-a and GSM are described in the literature from an organization named "third generation partnership project" (3rd Generation Partnership Project,3GPP). CDMA2000 and UMB are described in the literature from an organization named "third generation partnership project 2" (3 GPP 2). The techniques described herein may be used for the systems and radio technologies mentioned above as well as for other systems and radio technologies. However, the following description describes an NR system for purposes of example, and NR terminology is used in much of the description below, although the techniques may also be applied to applications other than NR system applications.

The following description provides examples and does not limit the scope, applicability, or configuration as set forth in the claims. Changes may be made in the function and arrangement of elements discussed without departing from the spirit and scope of the disclosure. Various examples may omit, substitute, or add various procedures or components as appropriate. For example, the described methods may be performed in an order different than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Referring to fig. 1, fig. 1 is a block diagram of a wireless communication system to which an embodiment of the present invention is applicable. The wireless communication system includes a terminal 11 and a network device 12. The terminal 11 may also be referred to as a User terminal or a User Equipment (UE), and the terminal 11 may be a terminal-side Device such as a mobile phone, a tablet Computer (Tablet Personal Computer), a Laptop (Laptop Computer), a personal digital assistant (Personal Digital Assistant, PDA), a mobile internet Device (Mobile Internet Device, MID), a Wearable Device (Wearable Device), or a vehicle-mounted Device, which is not limited to a specific type of the terminal 11 in the embodiments of the present disclosure. The network device 12 may be a base station and/or a core network element, where the base station may be a 5G or later version base station (e.g., a gNB, a 5G NR NB, etc.), or a base station in another communication system (e.g., an eNB, a WLAN access point, or other access points, etc.), where the base station may be referred to as a node B, an evolved node B, an access point, a base transceiver station (Base Transceiver Station, a BTS), a radio base station, a radio transceiver, a basic service set (Basic Service Set, BSS), an extended service set (Extended Service Set, ESS), a node B, an evolved node B (eNB), a home node B, a home evolved node B, a WLAN access point, a WiFi node, or some other suitable terminology in the field, and the base station is not limited to a specific technical vocabulary so long as the same technical effect is achieved, and it should be noted that, in the embodiments of the disclosure, the base station in the NR system is merely an example, but the specific type of the base station is not limited.

The base stations may communicate with the terminal 11 under the control of a base station controller, which may be part of the core network or some base stations in various examples. Some base stations may communicate control information or user data with the core network over a backhaul. In some examples, some of these base stations may communicate with each other directly or indirectly over a backhaul link, which may be a wired or wireless communication link. A wireless communication system may support operation on multiple carriers (waveform signals of different frequencies). A multicarrier transmitter may transmit modulated signals on the multiple carriers simultaneously. For example, each communication link may be a multicarrier signal modulated according to various radio technologies. Each modulated signal may be transmitted on a different carrier and may carry control information (e.g., reference signals, control channels, etc.), overhead information, data, and so on.

The base station may communicate wirelessly with the terminal 11 via one or more access point antennas. Each base station may provide communication coverage for a respective corresponding coverage area. The coverage area of an access point may be partitioned into sectors that form only a portion of that coverage area. A wireless communication system may include different types of base stations (e.g., macro base stations, micro base stations, or pico base stations). The base station may also utilize different radio technologies, such as cellular or WLAN radio access technologies. The base stations may be associated with the same or different access networks or operator deployments. The coverage areas of different base stations, including coverage areas of the same or different types of base stations, coverage areas utilizing the same or different radio technologies, or coverage areas belonging to the same or different access networks, may overlap.

The communication link in the wireless communication system may include an Uplink for carrying Uplink (UL) transmissions (e.g., from the terminal 11 to the network device 12) or a Downlink for carrying Downlink (DL) transmissions (e.g., from the network device 12 to the terminal 11). UL transmissions may also be referred to as reverse link transmissions, while DL transmissions may also be referred to as forward link transmissions. Downlink transmissions may be made using licensed bands, unlicensed bands, or both. Similarly, uplink transmissions may be made using licensed bands, unlicensed bands, or both.

As described in the background art, the related art terminal can implement different AI services according to the corresponding artificial intelligence model, for example, man-machine interaction, face recognition, photographing beautification, game, etc., but as the AI services or the properties of the terminal change, for example, one or more of the functions and scenes of the AI services, the energy consumption and the available storage space of the terminal change, the original artificial intelligence model may not be applicable any more, and the flow of the network device to issue the new artificial intelligence model again is long, resulting in low AI service response efficiency of the mobile communication terminal and the network device.

In order to solve the above problems, the embodiment of the invention provides an artificial intelligence model configuration method, which can realize the configuration of an artificial intelligence model. Referring to fig. 2, the method for configuring an artificial intelligence model provided by the embodiment of the invention is applied to a terminal, and includes the following steps:

step 21, obtaining artificial intelligent model configuration information sent by a network device or preconfigured in the terminal, wherein the artificial intelligent model configuration information comprises one or more of the following:

In the embodiment of the invention, the terminal can receive the configuration information of the artificial intelligent model sent by the network equipment, wherein the configuration information of the artificial intelligent model comprises one or more attribute parameters of the artificial intelligent model so as to realize the configuration of the artificial intelligent model. Of course, the artificial intelligence model configuration information may also be preconfigured at the terminal.

Specifically, specific attribute parameters of each artificial intelligence model are different for different artificial intelligence models. For example, applicable functions of the artificial intelligence model include CSI (Channel State Information ) feedback, beam management, positioning, etc. Even though the artificial intelligence model has the same function, other properties of the artificial intelligence model are not completely the same, such as different application scenarios, required calculation force resource information, and the like.

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

For different artificial intelligence models, the corresponding demand computing power resource information is different, and the specific expression is that the power consumption and/or the required storage space are different.

It will be appreciated that in other embodiments, the demand force resource information may include the required force of the artificial intelligence model, etc., in addition to the information described above. Correspondingly, the artificial intelligence model configuration information can also comprise performance index requirements of wireless transmission of network equipment and the like besides the information.

The network device sends specific attribute parameters of the artificial intelligent model to the terminal in the form of artificial intelligent model configuration information or the terminal is pre-configured with the artificial intelligent model configuration information, and the artificial intelligent model configuration information can be used as a basis for selecting a proper artificial intelligent model by the terminal so that the terminal selects the artificial intelligent model corresponding to the target service and/or the terminal attribute, and the terminal cooperates with the network device to execute the target service of the selected artificial intelligent model.

Specifically, the artificial intelligence model configuration information is configured through a radio resource control layer (Radio Resource Control, RRC) and sent to the terminal in the form of radio resource control RRC signaling.

Optionally, the artificial intelligence model configuration information is sent in the form of list information.

As shown in Table 1 below, table 1 provides a specific example of one of the artificial intelligence model configuration information:

TABLE 1

As shown in Table 2 below, table 2 provides a specific example of another type of artificial intelligence model configuration information:

TABLE 2

As can be seen from tables 1 and 2, the configuration information of the artificial intelligence model in Table 1 is divided into 9 artificial intelligence models by the application function, the application scene, the power consumption and the required storage space of each artificial intelligence model one by one, and the configuration information of the artificial intelligence model in Table 2 is divided into 3 artificial intelligence models by the application function of each artificial intelligence model, and then divided into 3 artificial intelligence models by the application scene, the power consumption and the required storage space of each artificial intelligence model.

It should be noted that table 1 and table 2 are only examples of embodiments of the present invention, and are not intended to limit the present invention.

In one embodiment, after the terminal receives the configuration information of the artificial intelligence model pre-configured in the terminal and sends or acquires the configuration information of the artificial intelligence model, the terminal can select the artificial intelligence model matching the current requirement according to the attribute parameter of each artificial intelligence model in the configuration information of the artificial intelligence model, so as to obtain better performance. Specifically, referring to fig. 3, the method for configuring an artificial intelligence model according to the embodiment of the present invention further includes the following steps:

step 31, reporting first indication information to the network equipment, wherein the first indication information comprises a model identifier of the selected artificial intelligent model.

Through the steps, the terminal can report the first indication information containing the selected artificial intelligent model identification to the network equipment, so that the network equipment performs the adapting operation after receiving the first indication information, and the terminal is cooperated to execute the target service.

radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

In the embodiment of the present invention, the first indication information is reported to the network device through an uplink channel, for example, the first indication information may be sent through radio resource Control signaling (RRC signaling), or sent through a medium access Control-Control Element (MAC-CE), or sent through a physical uplink Control (Physical Uplink Control Channel, PUCCH) channel or signal, or sent through a physical uplink shared (Physical Uplink Shared Channel, PUSCH) channel or signal, or sent through two or three of radio resource Control RRC signaling, medium access Control MAC-CE, physical uplink Control PUCCH, and physical uplink shared PUSCH.

In one embodiment, referring to fig. 4, the terminal selects an artificial intelligence model for executing a target service according to the artificial intelligence model configuration information, and includes the following steps:

step 41, comparing attribute parameters of each artificial intelligent model according to the artificial intelligent model configuration information, wherein the attribute parameters of the artificial intelligent model comprise one or more of an application function, an application scene and required calculation force resource information;

and 42, selecting an artificial intelligent model corresponding to one or more of the attribute parameters, the demand function of the target service, the demand scene of the target service and the current processing calculation power resource information of the terminal.

In the embodiment of the invention, the terminal selects the artificial intelligent model for executing the target service by comparing the attribute parameters of each artificial intelligent model.

Specifically, the attribute parameters of the artificial intelligent model selected by the terminal need to correspond to one or more of the demand function of the target service, the demand scene of the target service and the current processing calculation force resource information of the terminal. The current processing computing power resource information of the terminal is hardware or network resources required to be occupied when the terminal executes the target task, and may generally include computing power resources of a central processing unit (Central Processing Unit, CPU), computing power resources of a graphics processor (Graphics Processing Unit, GPU), memory resources, network bandwidth resources (channel variation condition), available storage space, current residual energy, and the like.

For example, if the required function of the target service is CSI feedback, an artificial intelligence model whose applicable function is not CSI feedback is excluded, and an artificial intelligence model whose applicable function is CSI feedback is selected.

For another example, taking the demand function of the target service as CSI feedback and the demand scenario as dense cities as an example, the terminal compares the conventional codebook-based normalized mean square error (Normalized Mean Squared Error, NMSE) with the normalized mean square error of different artificial intelligence models, respectively. For the 9 artificial intelligence models in table 1, NMSE with artificial intelligence model identification 0 is the lowest, but its power consumption and required memory space are larger, while NMSE with artificial intelligence model identification 1 is slightly higher, but its power consumption and required memory space are relatively larger and superior to the traditional codebook-based method. Further, the terminal may also select an artificial intelligence model matching the current target service requirement and the terminal attribute in combination with the current processing calculation power resource information of the terminal, for example, for a terminal with more available storage space and current residual energy, an artificial intelligence model with model identification of 0 is selected, for a terminal with less available storage space and current residual energy, an artificial intelligence model with model identification of 1 is selected, then the model identification of the selected artificial intelligence model is reported to the network device through the media access control MAC-CE, and the network device uses the adapted artificial intelligence model in decompressing CSI information according to the model identification of the reported artificial intelligence model.

With the change of the target service or the attribute of the terminal, the original artificial intelligent model may not be applicable any more, and the network device issues a new artificial intelligent model again, or the flow of sending new artificial intelligent model configuration information to the terminal again is longer, so that it is difficult to respond to the new target service quickly.

In order to solve the above problems, after the terminal detects that one or more of a demand function of a target service, a demand scene of the target service, and current processing calculation power resource information of the terminal changes, an artificial intelligent model after updating the target service is selected, a method for configuring the artificial intelligent model according to an embodiment of the present invention further includes the following steps:

and reporting fifth indicating information to the network equipment, wherein the fifth indicating information comprises a model identifier of the updated artificial intelligent model.

In the embodiment of the invention, if the terminal detects that one or more of the demand function of the target service, the demand scene of the target service and the current processing calculation power resource information of the terminal are changed, the attribute parameters of each artificial intelligent model are reevaluated, then the artificial intelligent model after updating the target service is selected to be executed, and then fifth indication information containing the model identification of the artificial intelligent model is reported to the network equipment, and correspondingly, the network equipment is also adapted to use the updated artificial intelligent model to execute the new target service.

Optionally, the fifth indication information and the first indication information are sent through the same type of signaling, or sent through different types of signaling.

It should be noted that, in the above embodiment, under the condition that the existing artificial intelligent model is not added, the terminal does not need to re-receive the artificial intelligent model configuration information sent by the network device, and only needs to use the artificial intelligent model configuration information sent or preconfigured by the network device for the first time to re-evaluate the attribute parameters of each artificial intelligent model. Therefore, compared with the prior art, the terminal of the embodiment of the invention can quickly determine the artificial intelligent model matched with the current requirement, thereby obtaining better performance, and the terminal does not need to wait for network equipment to issue new artificial intelligent models or new artificial intelligent model configuration information again, thereby realizing quick response of new target service.

And when the terminal does not report the model identification of the artificial intelligent model, the terminal and the network equipment adopt a non-artificial intelligent communication mode. In some embodiments, the non-artificial intelligence communication mode may include, but is not limited to, an ENDC mode, an AC mode, a single carrier mode, and the like. The ENDC (eNB NR Dual Connection) mode is a 4G and 5G dual-connection mode, and the terminal establishes communication connection with the 4G network device and the 5G network device at the same time. AC (Access Controller) mode is a plurality of carrier aggregation under the same network access technology, and the terminal establishes communication connection with network devices of different cells under the same macro station. The single carrier mode is to perform signal transmission of single carrier on only one frequency band, and the terminal establishes communication connection with one network device.

In the process of executing the target service, the terminal and the network equipment adopt an artificial intelligent communication mode. If the performance of the selected artificial intelligence model is degraded, it may result in the terminal not being suitable for continuing to execute the target service. If the performance of the selected artificial intelligence model is worse than that of the non-artificial intelligence communication mode, the terminal is not suitable for continuously executing the target service.

To solve the above problems, when determining that the performance of the selected artificial intelligence model is worse than the performance of the non-artificial intelligence communication mode and/or determining that the performance of the selected artificial intelligence model is worse, an artificial intelligence model configuration method according to an embodiment of the present invention further includes the steps of:

In the embodiment of the invention, when the terminal judges that the performance of the selected artificial intelligent model is poorer than that of the non-artificial intelligent communication mode and/or judges that the performance of the selected artificial intelligent model is poorer, the terminal sends second indication information to the network equipment so as to request to fall back to the non-artificial intelligent communication mode and ensure the basic communication performance. And after receiving the second indication information, the network equipment deactivates the corresponding artificial intelligent model.

In particular, the second indication information may include a model identification of the deactivated artificial intelligence model to indicate that the terminal determines the deactivated artificial intelligence model when performance degradation of the artificial intelligence model and/or performance worse than the non-artificial intelligence communication mode occurs, and feeds back the model identification of the deactivated artificial intelligence model to the network device to cause the terminal and the network device to fall back to the non-artificial intelligence communication mode. The second indication information may further include a function of deactivating the artificial intelligence model to indicate that the terminal determines the function of deactivating the artificial intelligence model and feeds back the function of deactivating the artificial intelligence model to the network device to cause the terminal and the network device to fall back to the non-artificial intelligence communication mode when performance of the artificial intelligence model is degraded and/or worse than performance of the non-artificial intelligence communication mode occurs. Of course, the second indication information may also include a model identifier for deactivating the artificial intelligence model and an applicable function for deactivating the artificial intelligence model, so that the terminal and the network device fall back to the non-artificial intelligence communication mode, which is not particularly limited in the present invention.

In one embodiment, the terminal and the network device may each store or obtain in real time criteria for evaluating the artificial intelligence model, and the criteria for evaluating are different for different types of artificial intelligence models. Taking the performance metrics of the artificial intelligence model as an example, there may be several exemplary evaluation criteria:

(1) The performance metric of the regression model typically employs a mean square error (Mean Square Error);

(2) The performance measurement of the classification algorithm generally adopts fault tolerance, precision and the like;

(3) The performance metrics of the clustering algorithm typically employ internal metrics including an accard coefficient (Jaccard Coefficient, JC), an FM Index (Fowlkes and Mallows Index, FMI), a Rand Index (RI), and an F-measure, and external metrics including Compactness (Compactness), segmentation (performance), dunn Index (Dunn Validity Index, DVI), and profile coefficient (Silhouette Coefficient).

Therefore, whether the terminal or the network device can evaluate the current artificial intelligent model according to the existing artificial intelligent model evaluation standard, so as to judge whether the performance of the selected artificial intelligent model is poorer than that of the non-artificial intelligent communication mode or not and judge whether the performance of the selected artificial intelligent model is poorer or not.

In the above embodiment of the present invention, the terminal selects the artificial intelligent model matching the current requirement, and in another embodiment, the network device may also select the artificial intelligent model matching the current requirement. Specifically, referring to fig. 5, an artificial intelligence model configuration method according to an embodiment of the present invention further includes the following steps:

Step 51, receiving third indication information sent by the network device, where the third indication information includes a model identifier of the activated artificial intelligence model and/or an applicable function of the activated artificial intelligence model.

Accordingly, the terminal may select an artificial intelligence model corresponding to the model identification to perform a target service corresponding to the activated artificial intelligence model according to the third indication information.

In the embodiment of the invention, the network equipment selects the artificial intelligent model matched with the current requirement and instructs the terminal to use the artificial intelligent model selected by the network equipment. After the terminal receives the third indication information comprising the model identification of the activated artificial intelligent model and/or the applicable function of the activated artificial intelligent model, the terminal can select the artificial intelligent model corresponding to the model identification according to the third indication information, and further execute the target service corresponding to the activated artificial intelligent model according to the artificial intelligent model.

Optionally, when the network device determines that one or more of a demand function of the target service, a demand scene of the target service, and current processing calculation power resource information of the terminal changes, the artificial intelligence model configuration method in the embodiment of the invention further includes the following steps:

And receiving sixth indication information sent by the network equipment, wherein the sixth indication information comprises a model identifier and/or an applicable function of the selected updated artificial intelligent model.

Accordingly, the terminal may select an artificial intelligence model corresponding to the updated model identification to perform a target service corresponding to the selected updated artificial intelligence model according to the sixth indication information.

In the embodiment of the invention, if the network equipment detects that one or more of the demand function of the target service, the demand scene of the target service and the current processing calculation power resource information of the terminal are changed, the attribute parameters of each artificial intelligent model are reevaluated, then the artificial intelligent model after updating the target service is selected and executed, and then the model identifier and/or the sixth indication information of the applicable function containing the artificial intelligent model is sent to the terminal, so that the terminal is also adapted to use the updated artificial intelligent model, and thus the new target service is executed.

Optionally, the sixth indication information and the third indication information are sent through the same type of signaling, or sent through different types of signaling.

It should be noted that, in the above embodiment, the network device does not need to send the configuration information of the artificial intelligence model to the terminal again under the condition that the existing artificial intelligence model is not added, and only needs to use the configuration information of the existing artificial intelligence model to re-evaluate the attribute parameters of each artificial intelligence model. Therefore, compared with the prior art, the network equipment of the embodiment of the invention can quickly determine the artificial intelligent model matched with the current requirement, thereby obtaining better performance, and the network equipment does not need to issue new artificial intelligent model configuration information or new artificial intelligent model again, thereby realizing quick response of new target service.

Optionally, when the network device determines that the performance of the selected artificial intelligence model is worse than the performance of the non-artificial intelligence communication mode, and/or determines that the performance of the selected artificial intelligence model is worse, an artificial intelligence model configuration method according to an embodiment of the present invention further includes the following steps:

receiving fourth indication information sent by the network equipment, wherein the fourth indication information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model; the fourth indication information and the third indication information are sent through the same type of signaling or are sent through different types of signaling;

thus, the terminal may deactivate the corresponding artificial intelligence model according to the fourth indication information.

In the embodiment of the invention, when the network equipment judges that the performance of the selected artificial intelligent model is poorer than that of the non-artificial intelligent communication mode and/or judges that the performance of the selected artificial intelligent model is poorer, fourth indication information can be sent to the terminal so as to indicate the terminal to fall back to the non-artificial intelligent communication mode and ensure the basic communication performance. And after receiving the fourth indication information, the terminal deactivates the corresponding artificial intelligent model.

In particular, the fourth indication information may include a model identification of the deactivated artificial intelligence model to indicate that the network device transmits the model identification of the deactivated artificial intelligence model to the terminal when performance degradation of the artificial intelligence model and/or performance worse than the non-artificial intelligence communication mode occurs, and instructs the terminal to deactivate the selected artificial intelligence model so as to enable the terminal and the network device to fall back to the non-artificial intelligence communication mode. The fourth indication may further include a function of deactivating the artificial intelligence model to indicate that the network device transmits the function of deactivating the artificial intelligence model to the terminal upon occurrence of a performance degradation of the artificial intelligence model and/or a performance degradation of the artificial intelligence model compared to the non-artificial intelligence communication mode, instructing the terminal to deactivate the selected artificial intelligence model to bring the terminal and the network device back to the non-artificial intelligence communication mode. Of course, the fourth indication information may also include a model identifier for deactivating the artificial intelligence model and an applicable function for deactivating the artificial intelligence model, so that the terminal and the network device fall back to the non-artificial intelligence communication mode, which is not particularly limited in the present invention.

In the embodiment of the invention, aiming at the artificial intelligent model configuration information in different forms, the third indication information and the fourth indication information can have different indication modes.

By way of example, taking the third indication information and the fourth indication information, and the artificial intelligence model configuration information in two forms of table 1 and table 2 as examples, the following multiple indication manners of the third indication information and the fourth indication information are respectively described.

For table 1, the first indication means: the fourth indication information and the third indication information are transmitted through the same type of signaling (e.g., medium access control MAC-CE).

The first indication manner of the third indication information is:

the model identification of the activated artificial intelligence model is indicated by the media access control MAC-CE.

Correspondingly, the first indication mode of the fourth indication information is as follows:

the deactivation of the model identification of the artificial intelligence model is indicated by the medium access control MAC-CE.

For the first indication mode, the third indication information and the fourth indication information are sent through a media access control (MAC-CE). The media access control MAC-CE activates the artificial intelligence model implicitly, indicating that the applicable function to which the artificial intelligence model corresponds begins to use the AI-based communication mode. By means of an implicit mode, the activation efficiency is high.

In particular, the media access control MAC-CE may indicate a model identification to activate one or more artificial intelligence models, representing one artificial intelligence model to activate a function or multiple artificial intelligence models to activate different functions simultaneously. Illustratively, the media access control MAC-CE may indicate a model identification of one artificial intelligence model that activates the CSI feedback function, e.g., model identification 0, or may indicate model identifications of multiple artificial intelligence models that activate the same CSI feedback function, e.g., model identifications 0, 1, and 2; in addition, the CSI feedback function and the beam management function may be activated simultaneously by the model identification of the artificial intelligence model, such as model identifications 0 and 3, or model identifications 0, 1 and 3, etc.

When the performance of the artificial intelligence model is poor and/or is inferior to that of the non-artificial intelligence communication mode, the media access control (MAC-CE) indicates to deactivate the model identification of one or more artificial intelligence models, and the corresponding functions of the deactivated artificial intelligence models are returned to the non-artificial intelligence communication mode, so that the basic communication performance is ensured.

Illustratively, the media access control MAC-CE may be designed as in fig. 6. Where R is a reserved bit, a=1 denotes activating the artificial intelligence model, and a=0 denotes deactivating the artificial intelligence model. Ci may take 0 or 1, A=1, ci=1 represents an artificial intelligence model that activates index=i, A=0, ci=0 represents an artificial intelligence model that deactivates index=i.

For table 1, the second indication is: the fourth indication information and the third indication information are transmitted through different types of signaling.

The second indication manner of the third indication information is:

the model identification of the activated artificial intelligence model is indicated by a configuration information element IE in radio resource control, RRC, signaling.

Correspondingly, the second indication mode of the fourth indication information is as follows:

For the second indication mode, the third indication information is sent through Radio Resource Control (RRC) signaling, and the fourth indication information is sent through media access control (MAC-CE).

Illustratively, with CSI feedback and beam management use cases, adding an information element IE to CSI-reportconfig (CSI reporting configuration) indicates the model identity of the active artificial intelligence model, implicitly indicating that the CSI feedback function uses AI-based feedback mode.

When the performance of the artificial intelligence model is degraded and/or worse than that of the non-artificial intelligence communication mode, the media access control (MAC-CE) indicates to deactivate the model identification of the artificial intelligence model, and the corresponding function of the deactivated artificial intelligence model is retracted to the non-artificial intelligence communication mode.

For table 1, the third indication is: the fourth indication information and the third indication information are transmitted through the same type of signaling.

The third indication manner of the third indication information is:

Correspondingly, the third indication mode of the fourth indication information is as follows:

the configuration information element IE that does not include the active artificial intelligence model in the radio resource control RRC signaling indicates a non-AI feedback mode.

When the artificial intelligence model performance is degraded and/or worse than the performance of the non-artificial intelligence communication mode compared to the second indication mode, the third indication mode requires the radio resource control RRC reconfiguration to fall back to the non-artificial intelligence communication mode with a larger latency than the former.

For table 2, the indication manner of the third indication information includes at least one of the following:

the applicable function of the activated artificial intelligence model is indicated through Radio Resource Control (RRC) signaling, and the model identification of the corresponding activated artificial intelligence model is indicated through medium access control (MAC-CE).

Specifically, the radio resource control RRC signaling enables some function based on AI, for example enabling CSI feedback function.

When the performance of the artificial intelligence model is degraded and/or worse than that of the non-artificial intelligence communication mode, the media access control (MAC-CE) indicates to deactivate the model identification of the artificial intelligence model, and the corresponding function of the deactivated artificial intelligence model is retracted to the non-artificial intelligence communication mode. If the media access control MAC-CE does not indicate the activated artificial intelligent model, the minimum artificial intelligent model is identified by the model corresponding to the function by default so as to reduce signaling overhead.

Illustratively, the media access control MAC-CE may be designed as in fig. 7. Where Ci may take 0 or 1, ci=1 denotes an artificial intelligence model that activates index=i of the function index corresponding to, and ci=0 denotes an artificial intelligence model that deactivates index=i of the function index corresponding to.

Referring to fig. 8, the embodiment of the invention also provides an artificial intelligence model configuration method, which is applied to network equipment and comprises the following steps:

step 81, sending artificial intelligent model configuration information to a terminal, wherein the artificial intelligent model configuration information comprises one or more of the following:

In the embodiment of the invention, the network equipment can send the configuration information of the artificial intelligent model to the terminal, wherein the configuration information of the artificial intelligent model comprises one or more attribute parameters of the artificial intelligent model so as to realize the configuration of the artificial intelligent model. Of course, the artificial intelligence model configuration information may also be preconfigured at the terminal.

Specifically, specific attribute parameters of each artificial intelligence model are different for different artificial intelligence models. For example, applicable functions of the artificial intelligence model include CSI (Channel State Information ) feedback, beam management, positioning, etc. Even artificial intelligence models with the same function are not identical in other properties, such as different applicable scenarios, demand force resource information, etc.

Power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

For different artificial intelligence models, the corresponding demand computing power resource information is different, and the corresponding power consumption and/or the required storage space are/is different.

Alternatively, the artificial intelligence model configuration information is sent in the form of a list, such as table 1 and table 2 in the above embodiments.

In one embodiment, after the terminal receives the configuration information of the artificial intelligence model pre-configured in the terminal and sends or acquires the configuration information of the artificial intelligence model, the terminal can select the artificial intelligence model matching the current requirement according to the attribute parameter of each artificial intelligence model in the configuration information of the artificial intelligence model, so as to obtain better performance. Specifically, referring to fig. 9, an artificial intelligence model configuration method according to an embodiment of the present invention further includes the following steps:

and 91, receiving first indication information reported by the terminal, wherein the first indication information comprises a model identifier of the artificial intelligent model.

Accordingly, the network device may select an artificial intelligence model corresponding to the model identification to perform a target service corresponding to the selected artificial intelligence model according to the first indication information.

Through the steps, the network equipment can directly receive the first indication information which is reported by the terminal and contains the selected artificial intelligent model identification, and performs the adapting operation, so that the target service is executed by the cooperative terminal.

radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

In the embodiment of the present invention, the first indication information is reported to the network device through an uplink channel, for example, the first indication information may be sent through radio resource control signaling (RRC signaling), or sent through medium access control (Mediu Access Control-Mediu Access Control, MAC-CE), or sent through a physical uplink control (Physical Uplink Control Channel, PUCCH) channel or signal, or sent through a physical uplink shared (Physical Uplink Shared Channel, PUSCH) channel or signal, or sent through two or three of radio resource control RRC signaling, medium access control MAC-CE, physical uplink control PUCCH channel or signal, and physical uplink shared PUSCH channel or signal.

With the change of the target service or the attribute of the terminal, the original artificial intelligent model may not be applicable any more, and the network device issues a new artificial intelligent model again, or the flow of issuing new artificial intelligent model configuration information again is longer, so that it is difficult to respond to the new target service quickly.

In order to solve the above problems, the method for configuring an artificial intelligence model according to the embodiment of the present invention further includes the following steps:

receiving fifth indication information reported by the terminal, wherein the fifth indication information comprises a model identifier of the updated artificial intelligent model;

accordingly, the network device may select an artificial intelligence model corresponding to the updated model identification according to the fifth indication information to perform a target service corresponding to the selected updated artificial intelligence model.

It should be noted that, in the above embodiment, the network device does not need to send the configuration information of the artificial intelligence model to the terminal again under the condition that the existing artificial intelligence model is not added, and the terminal can re-evaluate the attribute parameters of each artificial intelligence model only by using the configuration information of the artificial intelligence model sent or pre-configured for the first time. Therefore, the terminal of the embodiment of the invention can quickly determine the artificial intelligent model matched with the current requirement, thereby obtaining better performance, and the terminal does not need to wait for network equipment to issue new artificial intelligent models or new artificial intelligent model configuration information again, thereby realizing quick response of new target service.

In order to solve the above problems, referring to fig. 10, an artificial intelligence model configuration method according to an embodiment of the present invention further includes the following steps:

and step 101, receiving second indication information reported by the terminal, wherein the second indication information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model.

Thus, the network device may deactivate the corresponding artificial intelligence model based on the second indication information.

In particular, the second indication information may include a model identification of the deactivated artificial intelligence model to indicate that the terminal determines the deactivated artificial intelligence model when performance degradation of the artificial intelligence model and/or performance worse than the non-artificial intelligence communication mode occurs, and the network device receives the model identification of the deactivated artificial intelligence model fed back by the terminal to enable the terminal and the network device to fall back to the non-artificial intelligence communication mode. The second indication information may further include an applicable function of the deactivated artificial intelligence model to indicate that the terminal determines the applicable function of the deactivated artificial intelligence model when performance degradation of the artificial intelligence model and/or performance degradation of the non-artificial intelligence communication mode occurs, and the network device receives the applicable function of the deactivated artificial intelligence model fed back by the terminal to cause the terminal and the network device to fall back to the non-artificial intelligence communication mode. Of course, the second indication information may also include a model identifier for deactivating the artificial intelligence model and an applicable function for deactivating the artificial intelligence model, so that the terminal and the network device fall back to the non-artificial intelligence communication mode, which is not particularly limited in the present invention.

In the above embodiment of the present invention, the terminal selects the artificial intelligent model matching the current requirement, and in another embodiment, the network device may also select the artificial intelligent model matching the current requirement. Specifically, referring to fig. 11, after the network device selects an artificial intelligent model for executing the target service according to the artificial intelligent model configuration information, an artificial intelligent model configuration method according to an embodiment of the present invention further includes the following steps:

Step 111, sending third indication information to the terminal, wherein the third indication information comprises a model identification of the activated artificial intelligence model and/or an applicable function of the activated artificial intelligence model.

In the embodiment of the invention, the network equipment selects the artificial intelligent model matched with the current requirement and instructs the terminal to use the artificial intelligent model selected by the network equipment. And after the terminal receives the third indication information comprising the model identification of the activated artificial intelligent model and/or the applicable function of the activated artificial intelligent model, the terminal executes the target service corresponding to the activated artificial intelligent model according to the third indication information.

In one embodiment, the network device selects an artificial intelligence model for executing the target service according to the artificial intelligence model configuration information, and the method comprises the following steps:

comparing attribute parameters of each artificial intelligent model according to the artificial intelligent model configuration information, wherein the attribute parameters of each artificial intelligent model comprise one or more of an application function, an application scene and required calculation force resource information;

and selecting an artificial intelligent model corresponding to one or more of the attribute parameters, the demand function of the target service, the demand scene of the target service and the current processing calculation power resource information of the terminal.

In an embodiment of the invention, the network device selects the artificial intelligence model for executing the target service by comparing the attribute parameters of each artificial intelligence model.

Specifically, the attribute parameters of the artificial intelligent model selected by the network device need to correspond to one or more of the demand function of the target service, the demand scene of the target service and the current processing calculation force resource information of the terminal. The current processing computing power resource information of the terminal is hardware or network resources required to be occupied when the terminal executes the target task, and may generally include computing power resources of a central processing unit (Central Processing Unit, CPU), computing power resources of a graphics processor (Graphics Processing Unit, GPU), memory resources, network bandwidth resources (channel variation condition), available storage space, current residual energy, and the like.

For another example, taking the demand function of the target service as CSI feedback and the demand scenario as dense cities as an example, the terminal compares the conventional codebook-based normalized mean square error (Normalized Mean Squared Error, NMSE) with the normalized mean square error of different artificial intelligence models, respectively. For the 9 artificial intelligence models in table 1, NMSE with artificial intelligence model identification 0 is the lowest, but its power consumption and required memory space are larger, while NMSE with artificial intelligence model identification 1 is slightly higher, but its power consumption and required memory space are relatively larger and superior to the traditional codebook-based method. Further, the network device may also select an artificial intelligence model matching the current target service requirement and the terminal attribute in combination with the current processing calculation power resource information of the terminal, for example, for a terminal with more available storage space and current residual energy, the network device selects an artificial intelligence model with model identification of 0, for a terminal with less available storage space and current residual energy, the network device selects an artificial intelligence model with model identification of 1, and then sends the model identification of the selected artificial intelligence model to the terminal through a media access control MAC-CE, and the terminal uses the adapted artificial intelligence model in decompressing CSI information according to the model identification of the issued artificial intelligence model.

Optionally, when the network device determines that one or more of a demand function of the target service, a demand scene of the target service, and current processing calculation power resource information of the terminal changes, and selects an artificial intelligent model after updating the target service, the method for configuring the artificial intelligent model according to the embodiment of the invention further includes the following steps:

and sending sixth indication information to the terminal, wherein the sixth indication information comprises the model identification and/or the applicable function of the selected updated artificial intelligent model.

Optionally, referring to fig. 12, when the network device determines that the performance of the selected artificial intelligence model is worse than the performance of the non-artificial intelligence communication mode and/or determines that the performance of the selected artificial intelligence model is worse, an artificial intelligence model configuration method according to an embodiment of the present invention further includes the following steps:

step 121, sending fourth indication information to the terminal, wherein the fourth indication information comprises a model identifier for deactivating the artificial intelligence model and/or an applicable function for deactivating the artificial intelligence model;

The first indication manner of the third indication information is:

The second indication manner of the third indication information is:

The third indication manner of the third indication information is:

Referring to fig. 13, an embodiment of the present invention further provides an artificial intelligence model configuration apparatus 130, which is applied to a terminal, and includes:

an obtaining module 131, configured to obtain artificial intelligence model configuration information sent by a network device or preconfigured in the terminal, where the artificial intelligence model configuration information includes one or more of the following:

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

Optionally, referring to fig. 14, the artificial intelligence model configuration apparatus 130 further includes:

a selection module 132, configured to select an artificial intelligence model for executing a target service according to the artificial intelligence model configuration information;

and the sending module 133 is configured to report first indication information to the network device, where the first indication information includes a model identifier of the selected artificial intelligence model.

Optionally, the selecting module 132 is specifically configured to:

and selecting an artificial intelligent model corresponding to one or more of the attribute parameters, the demand function of the target service, the demand scene of the target service and the current processing calculation force resource information of the terminal.

Radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

Optionally, the selecting module 132 is further configured to select an artificial intelligent model after updating the target service when one or more of a demand function of the target service, a demand scenario of the target service, and current processing calculation power resource information of the terminal changes;

the sending module 133 is further configured to report fifth indication information to the network device, where the fifth indication information includes a model identifier of the selected updated artificial intelligence model.

Optionally, the sending module 133 is further configured to:

and when the performance of the selected artificial intelligent model is judged to be poorer than that of the non-artificial intelligent communication mode, and/or the performance of the selected artificial intelligent model is judged to be poorer, reporting second indicating information to the network equipment, wherein the second indicating information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model.

Optionally, the obtaining module 131 is further configured to receive third indication information sent by the network device, where the third indication information includes a model identifier of the activated artificial intelligence model and/or an applicable function of the activated artificial intelligence model;

The selection module 132 is configured to select an artificial intelligence model corresponding to the model identifier to execute a target service corresponding to the activated artificial intelligence model according to the fourth indication information.

Optionally, the obtaining module 131 is further configured to receive sixth indication information sent by the network device, where the sixth indication information includes a model identifier and/or an applicable function of the selected updated artificial intelligence model;

the selecting module 132 is further configured to select, according to the sixth indication information, an artificial intelligence model corresponding to the updated model identifier to execute the target service corresponding to the selected updated artificial intelligence model.

Optionally, referring to fig. 15, the artificial intelligence model configuration apparatus 130 further includes an execution module 134;

the obtaining module 131 is further configured to receive fourth indication information sent by the network device, where the fourth indication information includes a model identifier of the deactivated artificial intelligence model and/or an applicable function of the deactivated artificial intelligence model; the fourth indication information and the third indication information are sent through the same type of signaling or are sent through different types of signaling;

the execution module 134 is further configured to deactivate the corresponding artificial intelligence model according to the fourth indication information.

Referring to fig. 16, an embodiment of the present invention further provides an artificial intelligence model configuration apparatus 160, which is applied to a network device, and includes:

a sending module 161, configured to send artificial intelligence model configuration information to a terminal, where the artificial intelligence model configuration information includes one or more of the following:

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

Optionally, referring to fig. 17, the artificial intelligence model configuration apparatus 160 further includes:

the receiving module 162 is configured to receive first indication information reported by the terminal, where the first indication information includes a model identifier of an artificial intelligent model;

a selection module 163, configured to select an artificial intelligence model corresponding to the model identifier according to the first indication information, to execute a target service corresponding to the selected artificial intelligence model.

Radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

Optionally, the receiving module 162 is further configured to receive fifth indication information reported by the terminal, where the fifth indication information includes a model identifier of the selected updated artificial intelligent model;

the selecting module 163 is further configured to select an artificial intelligence model corresponding to the updated model identifier according to the fifth indication information to execute the target service corresponding to the selected updated artificial intelligence model.

Optionally, referring to fig. 18, the artificial intelligence model configuration apparatus 160 further includes an execution module 164;

the receiving module 162 is further configured to receive second indication information reported by the terminal, where the second indication information includes a model identifier for deactivating the artificial intelligence model and/or an applicable function for deactivating the artificial intelligence model;

the execution module 164 is configured to deactivate the corresponding artificial intelligence model according to the second indication information.

Optionally, the selecting module 163 is configured to select an artificial intelligence model for executing the target service according to the artificial intelligence model configuration information;

The sending module 161 is further configured to send sixth indication information to the terminal, where the sixth indication information includes a model identifier of the activated artificial intelligence model and/or an applicable function of the activated artificial intelligence model.

Optionally, the selecting module 163 is specifically configured to:

Optionally, the selecting module 163 is further configured to select an artificial intelligent model after updating the target service when one or more of the demand function of the target service, the demand scenario of the target service, and the current processing calculation power resource information of the terminal is determined to change;

the sending module 161 is further configured to send sixth indication information to the terminal, where the sixth indication information includes a model identifier and/or an applicable function of the updated artificial intelligence model.

Optionally, the sending module 161 is further configured to:

when the performance of the selected artificial intelligent model is judged to be poorer than that of the non-artificial intelligent communication mode and/or the performance of the selected artificial intelligent model is judged to be poorer, fourth indication information is sent to the terminal, and the fourth indication information comprises a model identifier for deactivating the artificial intelligent model and/or an applicable function for deactivating the artificial intelligent model;

Referring to fig. 19, the embodiment of the present invention further provides a terminal 190, including a processor 191, where the processor 191 is configured to implement each process of the embodiment of the artificial intelligence model configuration method applied to the terminal and achieve the same technical effects, and for avoiding repetition, details are not repeated here. Illustratively, the processor is configured to:

Referring to fig. 20, the embodiment of the present invention further provides a network device 200, which includes a transceiver 201, where the transceiver 201 is configured to implement each process of the embodiment of the artificial intelligence model configuration method applied to the network device, and achieve the same technical effects, and for avoiding repetition, a detailed description is omitted herein. Illustratively, the transceiver 201 is configured to:

Referring to fig. 21, the embodiment of the present invention further provides a terminal 210, which includes a processor 211, a memory 212, and a computer program stored in the memory 212 and capable of running on the processor 211, wherein the computer program realizes each process of the embodiment of the artificial intelligence model configuration method applied to the terminal when being executed by the processor 211, and can achieve the same technical effects, and for avoiding repetition, the description is omitted herein.

Referring to fig. 22, an embodiment of the present invention further provides a network device 220, including a processor 221, a memory 222, and a computer program stored in the memory 222 and capable of running on the processor 221, where the computer program when executed by the processor 221 implements each process of the embodiment of the artificial intelligence model configuration method applied to the network device, and the process can achieve the same technical effects, and is not repeated herein.

The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the processes of the above embodiment of the artificial intelligence model configuration method, and can achieve the same technical effects, so that repetition is avoided, and no further description is given here. Wherein the computer readable storage medium is selected from Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.

The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present invention and the scope of the claims, which are to be protected by the present invention.

Claims

1. An artificial intelligence model configuration method, which is characterized by being applied to a terminal, comprises the following steps:

2. The artificial intelligence model configuration method of claim 1, wherein the demand force resource information includes one or more of:

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

3. The artificial intelligence model configuration method of claim 1, wherein the artificial intelligence model configuration information is transmitted through radio resource control signaling.

4. The artificial intelligence model configuration method of claim 1, further comprising:

5. The artificial intelligence model configuration method of claim 4, wherein the first indication information is transmitted through at least one of:

Radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

6. The artificial intelligence model configuration method of claim 1, further comprising:

7. The artificial intelligence model configuration method of claim 1, further comprising:

and receiving third indication information sent by the network equipment, wherein the third indication information comprises a model identification of the activated artificial intelligence model and/or an applicable function of the activated artificial intelligence model.

8. The artificial intelligence model configuration method of claim 7, further comprising:

9. An artificial intelligence model configuration method, applied to a network device, comprising:

10. The artificial intelligence model configuration method of claim 9, wherein the demand force resource information includes one or more of:

power consumption of the artificial intelligence model;

the required storage space for the artificial intelligence model.

11. The artificial intelligence model configuration method of claim 9, wherein the artificial intelligence model configuration information is transmitted through radio resource control signaling.

12. The artificial intelligence model configuration method of claim 9, further comprising:

13. The artificial intelligence model configuration method of claim 12, wherein the first indication information is transmitted through at least one of:

Radio resource control signaling;

a medium access control-control element;

a physical uplink control channel or signal;

physical uplink shared channels or signals.

14. The artificial intelligence model configuration method of claim 9, further comprising:

15. The artificial intelligence model configuration method of claim 9, further comprising:

16. The artificial intelligence model configuration method of claim 15, further comprising:

17. A terminal comprising a processor, wherein the processor is configured to:

18. A network device comprising a transceiver, wherein the transceiver is configured to:

19. An artificial intelligence model configuration device, which is applied to a terminal, comprising:

20. An artificial intelligence model configuration apparatus, applied to a network device, comprising:

21. A terminal, comprising: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor, implements the steps of the artificial intelligence model configuration method of any one of claims 1 to 8.

22. A network device, comprising: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the artificial intelligence model configuration method of any one of claims 9 to 16.

23. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the artificial intelligence model configuration method according to any one of claims 1 to 8; alternatively, the computer program when executed by a processor implements the steps of an artificial intelligence model configuration method according to any one of claims 9 to 16.