Implementation method, terminal and base station of ai endogenous

Abstract

An AI endogenous implementation method, a terminal and a base station, wherein the base station comprises a first function surface, used for performing at least one of the following processes by interacting with the terminal: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process. The present application implements a smart endogenous wireless network, and improves the intelligence of the network. The embodiments of the present application can directly take the AI capability as an object of service in the network, can provide high protection and transmission optimization for the AI model parameters, and improve the adaptation degree of the AI and network capabilities.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mobile communication, in particular to an implementation method of artificial intelligence (AI) endogenesis, a terminal and a base station. BACKGROUND

[0002] 5G and AI are gradually integrated into people's lives and bring new economic growth trends. With the rapid development of AI technology, in the future network, the architecture of the network will be more and more integrated with AI technology. In the future 6G era, AI will inevitably become an indispensable part of network architecture.

[0003] With the rapid development of AI technology, in the future network, the architecture of the network will be more and more integrated with AI technology. The existing AI model is trained on the network side, a part of the network or terminal data is collected through the network, AI modeling is performed, and optimization design is performed according to the target to be reached. The above scheme is relatively independent of the intelligent function and the network function. This scheme faces some challenges: 1) more and more data needs to be reported, which occupies more network bandwidth. 2) the wireless capability and the AI efficiency are mutually restricted, and system-level optimization cannot be adopted. 3) the development of the terminal is more and more intelligent, and the current aspect does not fully utilize the intelligence of the terminal, and end-to-end intelligence cannot be formed.

[0004] It can be seen that the combination of the existing network architecture and AI is still in the early stage of research, and many application scenarios have not been considered in the network architecture, such as AI enhancement still adopts an external solution, and a network architecture with intelligent endogenesis has not been proposed. Although the terminal has a certain computing power, the AI capability of the terminal side is not considered when the network is designed. SUMMARY

[0005] At least one embodiment of the present application provides an AI endogenous implementation method, a terminal and a base station, realizes a wireless network with intelligent endogenesis, and improves the intelligence of the network.

[0006] According to one aspect of the present application, at least one embodiment provides a base station, comprising:

[0007] A first function surface is configured to perform at least one of the following processes by interacting with a terminal: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0008] In addition, according to at least one embodiment of the present application, the first function surface comprises a first function layer, and the first function layer is configured to perform at least one of the following processes: an AI task control process, an AI task calculation, a data collection and reporting process, and an AI model transmission and synchronization process.

[0009] The base station further comprises an RRC layer function module, a PDCP layer function module, an RLC layer function module and a MAC layer function module; wherein,

[0010] The RRC layer function module is configured to send information generated by the first function layer to a terminal or a network element on the network side through RRC signaling.

[0011] The PDCP layer function module, the RLC layer function module and the MAC layer function module are configured to transmit RRC signaling generated by the RRC layer function module.

[0012] In addition, according to at least one embodiment of the present application, the information generated and received by the first function layer includes a first data type and a second data type, the first data type is control information for base station and terminal AI model control, including at least one of the following data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data acquisition and preprocessing, and task information for control of AI task evaluation configuration, and the second data is data information for base station and terminal model transmission, data acquisition transmission or data preprocessing after data preprocessing, including at least one of the following data: parameters passed in AI model calculation, data collected or preprocessed by a data acquisition task, and labels.

[0013] In addition, according to at least one embodiment of the present application, the MAC layer function module is further configured to send the parameters passed in AI model calculation generated by the first function layer to a terminal or a network element on the network side.

[0014] In addition, according to at least one embodiment of the present application, the first function layer is further configured to:

[0015] decompose AI task requirements to determine at least one of data acquisition requirements on the base station side and the terminal side, AI model architecture, AI algorithm selection, and AI model evaluation algorithm, and send AI task configuration information to the terminal through RRC signaling or MAC control information to configure AI tasks to be executed by the terminal, wherein the AI task configuration information includes at least one of a segmentation method of an AI model, configuration of an AI model algorithm, configuration of data acquisition requirements, an AI model performance evaluation algorithm, and an executed task identifier;

[0016] receive preprocessed data sent by the terminal to the base station through RRC signaling or MAC CE;

[0017] According to the preprocessed data, the AI model is trained or jointly trained with the terminal to obtain the parameter configuration of the AI model, and the AI model is verified and evaluated by using the collected data and the corresponding labels and the AI model performance evaluation algorithm.

[0018] According to the updated AI model parameters obtained through training, the updated AI model parameters are sent to a terminal and / or a network element at a network side;

[0019] By using the AI model obtained through training, decision information of the AI task is generated, and the generated decision information is sent to the terminal or the network element at the network side.

[0020] In addition, according to at least one embodiment of the present application, the AI task requirement includes a network side task requirement, and / or a terminal side reported task requirement.

[0021] According to another aspect of the present application, at least one embodiment provides a terminal comprising a second functional module;

[0022] The second functional surface is configured to perform at least one of the following processes by interacting with a base station: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0023] In addition, according to at least one embodiment of the present application, the second functional surface comprises a second functional layer configured to perform at least one of the following processes: an AI task control process, an AI task calculation, a data collection and reporting process, and an AI model transmission and synchronization process.

[0024] In addition, according to at least one embodiment of the present application, the information generated and received by the second functional layer includes a third data type and a fourth data type. The third data is control information for AI model control of the base station and the terminal, including at least one of the following data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data collection and preprocessing, and task information for control of AI task evaluation configuration. The fourth data is data information for base station and terminal model transmission, data collection transmission, or data preprocessing, including at least one of the following data: parameters for transmission in AI model calculation, data collected or preprocessed by a data collection task, and labels.

[0025] In addition, according to at least one embodiment of the present application, further comprising: an RRC layer functional module, a PDCP layer functional module, an RLC layer functional module, and a MAC layer functional module; wherein,

[0026] The RRC layer functional module is configured to receive information sent by the base station through RRC signaling and forward the information to the second functional module.

[0027] The PDCP layer functional module, the RLC layer functional module, and the MAC layer functional module are configured to transmit RRC signaling sent by the base station to the RRC layer functional module.

[0028] Further, according to at least one embodiment of the present application, the MAC layer function module is further configured to receive parameters passed in AI model calculation generated by the base station and send the parameters to the second function layer.

[0029] Further, according to at least one embodiment of the present application, the second function layer is further configured to:

[0030] send AI task requirements to the base station through the RRC layer function module;

[0031] receive AI task configuration information sent by the base station through RRC signaling or MAC control information, determine AI tasks to be executed, and the AI task configuration information includes at least one of a segmentation method of an AI model, configuration of an AI model algorithm, configuration of data collection requirements, an AI model performance evaluation algorithm, and a task identifier executed by a terminal;

[0032] collect data and pre-process the data according to the configuration of the data collection requirements, and send the pre-processed data to the base station through RRC signaling or MAC CE;

[0033] cooperate with the base station to jointly train an AI model;

[0034] receive updated AI model parameters sent by the base station;

[0035] receive decision information sent by the base station.

[0036] According to another aspect of the present application, at least one embodiment provides an AI endogenous implementation method applied to a base station, the base station comprising a first function surface, and the method comprises:

[0037] interacting with a terminal through the first function surface of the base station, and performing at least one of the following processes: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0038] Further, according to at least one embodiment of the present application, the first function surface comprises a first function layer; the base station further comprises an RRC layer function module, a PDCP layer function module, an RLC layer function module, and a MAC layer function module; and the method further comprises:

[0039] performing at least one of an AI task control process, an AI task calculation, a data collection and reporting process, and an AI model transmission and synchronization process through the first function layer;

[0040] sending information generated by the first function layer to a terminal or a network element on the network side through RRC signaling by means of the RRC layer function module;

[0041] The RRC signaling generated by the RRC layer function module is transmitted through the PDCP layer function module, the RLC layer function module and the MAC layer function module.

[0042] According to another aspect of the present application, at least one embodiment provides an AI-endogenous implementation method applied to a terminal, the terminal comprising a second function surface, the method comprising:

[0043] The terminal interacts with a base station through the second function surface to perform at least one of the following processes: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0044] In addition, according to at least one embodiment of the present application, the second function surface comprises a second function layer; the terminal further comprises an RRC layer function module, a PDCP layer function module, an RLC layer function module and a MAC layer function module; and the method further comprises:

[0045] The second function layer performs at least one of the following processes: an AI task control process, an AI task calculation, a data collection and reporting process, and an AI model transmission and synchronization process.

[0046] The RRC layer function module receives information sent by the base station in the form of RRC signaling and forwards the information to the second function module.

[0047] The PDCP layer function module, the RLC layer function module and the MAC layer function module transmit RRC signaling sent by the base station to the RRC layer function module.

[0048] According to another aspect of the present application, at least one embodiment provides a base station, comprising a processor, a memory and a program stored on the memory and executable on the processor, the program being executed by the processor to implement the steps of the method described above.

[0049] According to another aspect of the present application, at least one embodiment provides a terminal, comprising a processor, a memory and a program stored on the memory and executable on the processor, the program being executed by the processor to implement the steps of the method described above.

[0050] According to another aspect of the present application, at least one embodiment provides a computer-readable storage medium, the computer-readable storage medium storing a program, the program being executed by a processor to implement the steps of the method described above.

[0051] Compared with the prior art, the AI endogenous implementation method, the terminal and the base station provided by the embodiment of the application implement an intelligent endogenous wireless network, and improve the intelligence of the network. The embodiment of the application can directly take AI capability as an object of service in the network, can provide high guarantee and transmission optimization for AI model parameters, and improves the adaptation degree of AI and network capability. BRIEF DESCRIPTION OF DRAWINGS

[0052] 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 accompanying drawings are included to provide a better understanding of the preferred embodiments, and are not to be considered limitations of the application. Furthermore, in the accompanying drawings, like reference numerals refer to same components throughout the several views. In the drawings:

[0053] Figure 1 An application scenario schematic diagram of the embodiment of the application;

[0054] Figure 2 A component structure schematic diagram of the intelligent face provided by the embodiment of the application;

[0055] Figure 3 A flowchart of the interaction of the intelligent face of the base station and the terminal of the embodiment of the application;

[0056] Figure 4 An example diagram of the implementation flow of the end-to-end AI of the embodiment of the application;

[0057] Figure 5 A flowchart of the AI endogenous implementation method provided by the embodiment of the application when applied to the base station side;

[0058] Figure 6 A flowchart of the AI endogenous implementation method provided by the embodiment of the application when applied to the terminal side;

[0059] Figure 7 A structure schematic diagram of the base station provided by the embodiment of the application;

[0060] Figure 8 A structure schematic diagram of the terminal provided by the embodiment of the application. DETAILED DESCRIPTION

[0061] Exemplary embodiments of the present application will be described herein below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0062] The terms “first,” “second,” etc., used in the specification and claims of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus. The terms “and / or” in the specification and claims indicate at least one of the connected objects.

[0063] The technology described herein is not limited to NR systems and Long Time Evolution (LTE) / LTE-Advanced (LTE-A) systems, and can also be used for various wireless communication systems such as Code Division Multiple Access (CDMA), 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 other systems. The terms "system" and "network" are often used interchangeably. A CDMA system can implement a radio technology such as CDMA2000, Universal Terrestrial Radio Access (UTRA), etc. UTRA includes Wideband-CDMA (WCDMA) and other variants of CDMA. A TDMA system can implement a radio technology such as Global System for Mobile Communications (GSM). An OFDMA system can implement a radio technology such as Ultra Mobile Broadband (UMB), Evolution-UTRA (E-UTRA), IEEE 802.21 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, Flash-OFDM, etc. UTRA and E-UTRA are part of Universal Mobile Telecommunication System (UMTS). LTE and LTE-Advanced (e.g., LTE-A) are new releases of UMTS that use E-UTRA. UTRA, E-UTRA, UMTS, LTE, LTE-A, and GSM are described in documents from an organization called the "3rd Generation Partnership Project" (3GPP). CDMA2000 and UMB are described in documents from an organization called the "3rd Generation Partnership Project 2" (3GPP2).The techniques described herein can be used for the systems and radio technologies mentioned above as well as other systems and radio technologies. The description below, however, describes a NR system for purposes of example, and NR terminology is used in much of the description below, although the techniques are applicable beyond NR systems.

[0064] The following description provides examples, and is not limiting of the scope, applicability, or configuration set forth in the claims. Changes can be made in the function and arrangement of elements discussed without departing from the scope and spirit of aspects of the disclosure. Various examples can omit, substitute, or add various procedures or components as appropriate. For instance, the methods described can be performed in an order different from that described, and other steps can be added, omitted, or combined. Also, features described with respect to certain examples can be combined in other examples.

[0065] See Figure 1 , Figure 1 A block diagram of a wireless communication system to which embodiments of the present application can be applied is shown. The wireless communication system includes a terminal 11 and a network device 12. The terminal 11 can also be referred to as a user terminal or a user equipment (UE), and the terminal 11 can be a terminal-side device such as a mobile phone, a tablet personal computer, a laptop computer, a personal digital assistant (PDA), a mobile Internet device (MID), a wearable device, or a vehicle-mounted device, and it should be noted that the specific type of the terminal 11 is not limited in the embodiments of the present application. The network device 12 can be a base station and / or a core network element, and the base station can be a base station of 5G and later versions (for example, a gNB, a 5G NR NB, or the like), or a base station in other communication systems (for example, an eNB, a WLAN access point, or other access points, and the like), and the base station can be referred to as a node B, an evolved node B, an access point, a base transceiver station (BTS), a radio base station, a radio transceiver, a basic service set (BSS), an 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 art, provided that the same technical effect is achieved, and the base station is not limited to a specific technical term, and it should be noted that only the base station in the NR system is taken as an example in the embodiments of the present application, but the specific type of the base station is not limited.

[0066] The base stations can communicate with the terminals 11 under the control of a base station controller, which in various examples can be part of the core network or of some base stations. Some of the base stations can communicate control information or user data with the core network over a backhaul. In some examples, some of these base stations can communicate, either directly or indirectly, with each other over the backhaul, which can be a wired or wireless communication link. The wireless communication system can support operation on multiple carriers (waveform signals of different frequencies). Multi-carrier transmitters can transmit modulated signals simultaneously on the multiple carriers. For example, each communication link can be a multi-carrier signal modulated according to the various radio technologies. Each modulated signal can be sent on a different carrier and can carry control information (e.g., reference signals, control channels, etc.), overhead information, data, etc.

[0067] The base stations can wirelessly communicate with the terminals 11 via one or more access point antennas. Each base station can provide communication coverage for a respective coverage area. The coverage area for an access point can be divided into sectors making up only a portion of the coverage area. The wireless communication system can include base stations of different types (e.g., macro, micro, or pico base stations). The base stations can utilize different radio technologies, such as cellular or WLAN radio access technologies. The base stations can be associated with the same or different access networks or operator deployments. The coverage areas of different base stations, including the coverage areas of base stations of the same or different types, utilizing the same or different radio technologies, or belonging to the same or different access networks, can overlap.

[0068] Communication links in a wireless communication system can include uplinks for carrying Uplink (UL) transmissions (e.g., from a terminal 11 to a network device 12) or downlinks for carrying Downlink (DL) transmissions (e.g., from a network device 12 to a terminal 11). UL transmissions can also be called reverse link transmissions, and DL transmissions can also be called forward link transmissions. Downlink transmissions can be made using a licensed spectrum, an unlicensed spectrum, or both. Similarly, uplink transmissions can be made using a licensed spectrum, an unlicensed spectrum, or both.

[0069] The prior art wireless network includes a data plane and a control plane, without considering the influence of the intelligent plane. In subsequent networks, after considering the capabilities of terminals and networks, the design of the intelligent plane should also be considered. Considering that the definition of network functions and architecture in the existing standard is for the purpose of network control and data transmission, and the end-to-end intelligence is not taken as a service object of the network. Based on this, the embodiment of the application considers the scheme of integrating intelligence with the wireless network, takes the network intelligence as a service object of the network, in order to achieve this purpose, the definition of the intelligent plane is introduced in the embodiment of the application.

[0070] The embodiment of the application aims at network intelligence endogenesis, proposes a network architecture of intelligent endogenesis and a corresponding AI interaction process, and realizes intelligent endogenesis network and improves the intelligence of the network by taking intelligence as part of the air interface transmission on the basis of fully tapping the terminal computing power.

[0071] The embodiment of the application realizes the intelligence of the intelligent face end-to-end throughout the network by introducing the process of wisdom endogenesis. AI is part of the base station, and through the fusion with the wireless architecture, the model algorithm, the interaction process and the wireless communication are organically fused, so as to better serve the end-to-end intelligence.

[0072] The embodiment of the application proposes the definition of the intelligent face, and the main functions of the intelligent face are:

[0073] 1) Control of end-to-end AI tasks, including AI task establishment, task start execution and end, etc.;

[0074] 2) Transmission of AI model and parameters, reporting control of data collection, etc.

[0075] The intelligent face of the embodiment of the application includes a wireless artificial intelligence (Wireless AI, WAI) layer and a communication-related protocol module or protocol layer in the rest of the network. Taking the 5G network architecture as an example, the intelligent face includes WAI, a radio resource control (Radio Resource Control, RRC) layer, a packet data convergence protocol (Packet Data Convergence Protocol, PDCP) layer, a radio link control (Radio Link Control, RLC) layer, a media access control (Media Access Control, MAC) layer and a physical layer (PHY). In this paper, the intelligent face on the base station side is also called the first functional face, and the WAI layer on the base station side is called the first functional layer; the intelligent face on the terminal side is called the second functional face, and the WAI layer on the terminal side is called the second functional layer. The intelligent face is as shown in the figure. Figure 1

[0076] The WAI layer is used for end-to-end AI control, synchronization, calculation and data collection reporting of the base station and the terminal. The functions of the WAI layer and the RRC layer are different. The WAI layer focuses on the configuration of AI capability, the control of AI task and the execution of related calculation, and forms a corresponding decision based on the task and the target. The RRC layer focuses on issuing the decision through the RRC configuration to the terminal.

[0077] The RRC layer is used for transmitting the decision information and configuration generated by the WAI to the terminal or other network elements through RRC signaling.

[0078] ​The PDCP, RLC, and MAC layers execute the transmission of RRC signaling. The MAC layer can also transmit parameters such as model parameters from the WAI layer, sending this type of data to the network or terminal.

[0079] To ensure the flexibility of end-to-end AI capabilities, the WAI of the access network in this embodiment of the invention can be synchronized with the core network to achieve the effect of overall coordination of AI capabilities.

[0080] Specifically, in this embodiment of the invention, the base station includes a first functional plane (intelligent plane), which is used to perform at least one of the following processes by interacting with the terminal: artificial intelligence (AI) task control process, data collection and reporting process, and AI model transmission and synchronization process.

[0081] Specifically, the first functional plane includes a first functional layer (WAI layer). The first functional layer is used to execute at least one of the following: AI task control process, AI task calculation, data acquisition and reporting process, and AI model transmission and synchronization process.

[0082] In addition, such as Figure 2 As shown, the base station further includes: an RRC layer functional module, a PDCP layer functional module, an RLC layer functional module, and a MAC layer functional module; wherein,

[0083] The RRC layer functional module is used to send the information generated by the first functional layer to the terminal or network element on the network side via RRC signaling.

[0084] The PDCP layer functional module, RLC layer functional module, and MAC layer functional module are used to transmit RRC signaling generated by the RRC layer functional module.

[0085] In this embodiment of the invention, the information generated and received by the first functional layer includes a first data type and a second data type. The first data type is control information for controlling the AI ​​model of the base station and the terminal, including at least one of the following types of data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data acquisition and preprocessing, and task information for control configured for AI task evaluation. The second data is data information transmitted by the base station and the terminal model, transmitted through data acquisition, or processed after data preprocessing, including at least one of the following types of data: parameters transmitted in AI model calculation, data acquired or preprocessed by data acquisition tasks, and tags.

[0086] The MAC layer functional module on the base station side is also used to send the parameters transmitted in the AI ​​model calculation generated by the first functional layer to the terminal or the network element on the network side.

[0087] The first functional layer is further configured to realize an intelligent, intrinsically generated network by performing the following steps:

[0088] a) decompose the AI task requirement, determine at least one of the data collection requirement of the base station side and the terminal side, the AI model architecture, the AI algorithm selection, the AI model evaluation algorithm, and send the AI task configuration information to the terminal through the RRC signaling or the MAC control information, configure the AI task to be executed by the terminal, the AI task configuration information includes at least one of the segmentation method of the AI model, the configuration of the AI model algorithm, the configuration of the data collection requirement, the AI model performance evaluation algorithm, and the task identification of the execution. Here, the AI task requirement includes the task requirement of the network side, and / or the task requirement reported by the terminal side.

[0089] b) receiving the preprocessed data sent by the terminal to the base station through RRC signaling or MAC CE.

[0090] c) according to the preprocessed data, training the AI model or jointly training the AI model with the terminal, obtaining the parameter configuration of the AI model, and verifying and evaluating the AI model by using the collected data and the corresponding label and the AI model performance evaluation algorithm.

[0091] d) according to the updated AI model parameters obtained by training, sending the updated AI model parameters to the terminal and / or the network element of the network side.

[0092] e) using the trained AI model to generate the decision information of the AI task, and sending the generated decision information to the terminal or the network element of the network side.

[0093] In the embodiment of the application, the terminal includes a second functional surface (intelligent surface). The second functional surface is used to interact with the base station to execute at least one of the following processes: artificial intelligence AI task control process, data collection reporting process, AI model transmission and synchronization process.

[0094] Specifically, the second functional surface includes a second functional layer (WAI layer), and the second functional layer is used to execute at least one of the AI task control process, AI task calculation, data collection reporting process, and AI model transmission and synchronization process.

[0095] Here, the information generated and received by the second function layer includes a third data type and a fourth data type, the third data being control information for base station and terminal AI model control, including at least one of the following data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data acquisition and preprocessing, task information for control of AI task evaluation configuration, and the fourth data being data information for base station and terminal model delivery, data acquisition transmission, or data after preprocessing, including at least one of the following data: parameters delivered in AI model calculation, data collected or preprocessed by a data acquisition task, and labels.

[0096] As shown in Figure 2 , the terminal further includes an RRC layer function module, a PDCP layer function module, an RLC layer function module, and a MAC layer function module; wherein,

[0097] The RRC layer function module is configured to receive information sent by the base station in the form of RRC signaling and forward the information to the second function module.

[0098] The PDCP layer function module, the RLC layer function module, and the MAC layer function module are configured to transmit RRC signaling sent by the base station to the RRC layer function module.

[0099] Here, the MAC layer function module is further configured to receive parameters delivered in AI model calculation generated by the base station and send the parameters to the second function layer.

[0100] The second function layer is further configured to implement intelligent endogenous network by performing the following steps:

[0101] a) sending AI task requirements to the base station through the RRC layer function module.

[0102] b) receiving AI task configuration information sent by the base station through RRC signaling or MAC control information, determining AI tasks to be executed, and the AI task configuration information including at least one of the following: AI model segmentation method, AI model algorithm configuration, data acquisition requirement configuration, AI model performance evaluation algorithm, and task identification executed by the terminal.

[0103] c) collecting data and preprocessing according to the data acquisition requirement configuration, and sending the preprocessed data to the base station through RRC signaling or MAC CE.

[0104] d) performing joint training of the AI model in cooperation with the base station.

[0105] e) receiving updated AI model parameters sent by the base station.

[0106] f) receiving decision information sent by the base station.

[0107] Based on the above architecture, the application embodiment also introduces the intelligent interfacial interaction process of the access network side, including the end-to-end AI task control process, data collection and reporting process, AI synchronization process, etc. in the architecture. According to the different transmission contents, WAI can be further divided into a first data type (AI-C) and a second data type (AI-U). AI-C is control information for end-to-end model control, and AI-U is model transmission, data collection transmission or preprocessed data transmission information. The data transmitted in the control tasks of end-to-end AI task control, AI model end-to-end establishment, data collection and preprocessing, and AI task evaluation configuration are called AI-C. The data transmitted in the parameter transmission, data collection and preprocessed parameter processing in AI model calculation, and label transmission are called AI-U.

[0108] Please refer to Figure 3 The interaction of the intelligent surface of the base station and the terminal in the application embodiment mainly includes:

[0109] A) Demand decomposition: send the AI task demand of the terminal or network to the WAI layer of the terminal or network. The WAI layer determines the data collection demand of the base station side and the network side, the AI model architecture, the AI algorithm selection, the AI model evaluation algorithm, etc. according to the AI task demand. See step 301 in Figure 3 .

[0110] B) Task decomposition and configuration: the base station informs the WAI entity of the UE side of the AI task that needs to be performed by the terminal through RRC signaling or MAC CE according to the results of demand decomposition. It may include model algorithm segmentation, model algorithm configuration, data collection configuration, and performance evaluation algorithm. The terminal WAI selects appropriate models and algorithms according to the configuration demand. See steps 302 and 303 in Figure 3 .

[0111] C) Data collection and data preprocessing: the terminal WAI generates preprocessed data according to the received data processing requirements, which may include formatting data format and definition of corresponding labels, and reports it to the base station WAI through RRC signaling or MAC CE. See step 304 in Figure 3 .

[0112] D) Performance evaluation and verification: perform evaluation and verification according to the evaluation method configured by the base station or terminal WAI. For example, the base station WAI training or joint training of the base station and terminal WAI obtains the parameter configuration of the model, and performs verification using the collected data and corresponding labels and the determined evaluation algorithm. See steps 305 and 306 in Figure 3 .

[0113] E) Result delivery: WAI delivers the decision information obtained by AI model to the corresponding network element according to the demand target. The decision involving the terminal is notified to the terminal by RRC signaling or MAC PDU. See steps 307 and 308 in Figure 3 .

[0114] F) For the task that still cannot meet the demand after the result delivery, the demand decomposition is performed again, and the task is solved by other model architecture and AI algorithm. See steps 309 and 310 in Figure 3 .

[0115] Taking scheduling optimization as an example, how to implement the optimization process based on the above architecture is explained, mainly including:

[0116] S1. The network receives the request for scheduling optimization, decomposes the task demand, determines the AI model information, such as CNN data collection (for example, the data volume of each logical information of the user, the data packet size, the channel condition, and the moving speed) and data preprocessing, and configures the terminal.

[0117] S2. The base station sends RRC signaling, carries the AI model information, and notifies the terminal of the data collection configuration information.

[0118] S3. The terminal performs data collection and data preprocessing according to the task configuration.

[0119] S4. The terminal reports the data collection and preprocessing result to the base station.

[0120] S5. The base station performs model inference training, generates the scheduling result, and sends the result to the terminal.

[0121] S6. The terminal evaluates the scheduling performance according to the scheduling result, and if the performance does not meet the demand, continues to report the demand to the base station.

[0122] The above S1 to S5 are repeated until the demand is met or the preset maximum reporting number is reached.

[0123] Please refer to Figure 4 , based on the network architecture of the embodiment of the application, a process example of end-to-end AI is provided, including:

[0124] A) Demand decomposition: the task demand includes two cases: a, the task demand from the network (corresponding to 1 in Figure 4 ), the demand decomposition is performed by the WAI on the base station side, and the model algorithm is confirmed; b, the task demand from the terminal, which is reported to the network side through RRC signaling (corresponding to 1.1-1.4 in Figure 4 ), the demand decomposition is performed by the network side. After the demand decomposition, the base station configures the decomposed task to the terminal through RRC signaling. Corresponding to 2-4 in Figure 4 .

[0125] The requirement decomposition includes task indication of the terminal and the network side, training tasks to be performed by both sides, and the definition of the type, format and label value of the data to be collected by both sides.

[0126] B) Task decomposition and model configuration: the base station side WAI configures the model involving the terminal to the terminal side through RRC signaling. Corresponding to 5-7 in Figure 4

[0127] C) Data collection and data preprocessing: the terminal WAI sends the data to the base station WAI in the agreed format, type and label through RRC signaling or MAC data packet. Corresponding to 8-10 in Figure 4

[0128] D) Model training and performance evaluation: end-to-end AI parameter update is performed through RRC signaling or MAC data packet, and performance evaluation is performed according to the configured evaluation criteria. Corresponding to 11-16 in Figure 4

[0129] E) Result delivery: the result of the WAI, such as air interface parameter update, is sent to the terminal side through RRC signaling or MAC control CE, and the terminal updates the configuration according to the configuration of the base station. Corresponding to 17-18 in Figure 4

[0130] As can be seen from the above, the embodiment of the application provides an interaction process of a base station and a terminal AI endogenous network module under a wisdom endogenous concept. By introducing the definition of an intelligent surface, requirement decomposition, model configuration, data collection, model optimization and result delivery processes, the requirement of data collection and the parameters of the model are transmitted through RRC signaling and MAC data packet, and the AI capability is directly used as an object of network service through the configuration of data collection, which can provide high guarantee and transmission optimization for the AI model parameters and improve the adaptation degree of AI and network capability.

[0131] Please refer to Figure 5 The AI task control method provided by the embodiment of the application is applied to a base station, and includes the following steps.

[0132] In step 501, the first function surface of the base station is used to interact with a terminal and perform at least one of the following processes: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0133] Through the above steps, the embodiment of the application realizes a wireless network with intelligent endogenesis, and improves the intelligence of the network.

[0134] ​​​​Optionally, the first function surface includes a first function layer; the base station further includes an RRC layer function module, a PDCP layer function module, an RLC layer function module, and a MAC layer function module; and the method further includes:

[0135] At least one of an AI task control flow, AI task calculation, data collection and reporting flow, AI model transmission and synchronization flow is executed through the first function layer.

[0136] Information generated by the first function layer is transmitted to a terminal or a network element on the network side through RRC signaling through the RRC layer function module.

[0137] RRC signaling generated by the RRC layer function module is transmitted through the PDCP layer function module, the RLC layer function module, and the MAC layer function module.

[0138] Optionally, the information generated and received by the first function layer includes a first data type and a second data type, the first data type is control information for AI model control of the base station and the terminal, including at least one of the following data: AI task configuration information, control information for establishing an end-to-end AI model, data collection and preprocessing control information, and task information for controlling AI task evaluation configuration; and the second data is data information transmitted by the base station and the terminal model, data collection transmission, or data preprocessing, including at least one of the following data: parameters transmitted in AI model calculation, data collected or preprocessed by a data collection task, and labels.

[0139] Optionally, the MAC layer function module further transmits the parameters transmitted in AI model calculation generated by the first function layer to the terminal or the network element on the network side.

[0140] Optionally, the first function layer further performs the following steps:

[0141] At least one of data collection requirements, AI model architecture, AI algorithm selection, and AI model evaluation algorithm on the base station side and the terminal side is determined, and AI task configuration information is transmitted to the terminal through RRC signaling or MAC control information to configure the AI task to be executed by the terminal, the AI task configuration information including at least one of AI model segmentation method, AI model algorithm configuration, data collection requirement configuration, AI model performance evaluation algorithm, and executed task identification.

[0142] Preprocessed data transmitted by the terminal to the base station through RRC signaling or MAC CE is received.

[0143] According to the preprocessed data, training or joint training of an AI model with a terminal is performed to obtain an AI model parameter configuration, and the collected data and corresponding labels and an AI model performance evaluation algorithm are used to verify and evaluate the AI model;

[0144] According to the updated AI model parameters obtained by training, the updated AI model parameters are sent to a terminal and / or a network element on a network side;

[0145] The AI model obtained by training is used to generate decision information of an AI task, and the generated decision information is sent to a terminal or a network element on a network side.

[0146] Optionally, the AI task requirement includes a network side task requirement and / or a task requirement reported by a terminal side.

[0147] Please refer to Figure 6 The AI endogenous implementation method provided by the embodiment of the application, when applied to a terminal side, comprises the following steps:

[0148] In step 601, the second function surface of the terminal is used to interact with a base station to perform at least one of the following processes: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0149] Optionally, the second function surface comprises a second function layer, and the terminal further comprises an RRC layer function module, a PDCP layer function module, an RLC layer function module, and a MAC layer function module.

[0150] The second function layer is used to perform at least one of the following processes: an AI task control process, AI task calculation, a data collection and reporting process, and an AI model transmission and synchronization process.

[0151] The RRC layer function module is used to receive information sent by a base station in the form of RRC signaling and forward the information to the second function module.

[0152] The PDCP layer function module, the RLC layer function module, and the MAC layer function module are used to transmit RRC signaling sent by a base station to the RRC layer function module.

[0153] Optionally, the information generated and received by the second functional layer includes a third data type and a fourth data type, the third data is control information for AI model control of the base station and the terminal, including at least one of the following data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data acquisition and preprocessing, and task information for control of AI task evaluation configuration, and the fourth data is data information delivered by the base station and the terminal model, data acquisition transmission or data after preprocessing, including at least one of the following data: parameters delivered in AI model calculation, data collected or preprocessed by a data acquisition task, and labels.

[0154] Optionally, the MAC layer functional module also receives the parameters delivered in AI model calculation generated by the base station and sends them to the second functional layer.

[0155] Optionally, the second functional layer further performs the following steps:

[0156] sending AI task requirements to the base station through the RRC layer functional module;

[0157] receiving AI task configuration information sent by the base station through RRC signaling or MAC control information, determining the AI task to be executed, and the AI task configuration information including at least one of the following: AI model segmentation method, AI model algorithm configuration, data acquisition requirement configuration, AI model performance evaluation algorithm, and task identification executed by the terminal;

[0158] acquiring data and preprocessing according to the data acquisition requirement configuration, and sending the preprocessed data to the base station through RRC signaling or MAC CE;

[0159] cooperating with the base station to perform joint training of the AI model;

[0160] receiving updated AI model parameters sent by the base station;

[0161] receiving decision information sent by the base station.

[0162] The above introduces various methods of embodiments of the application. The following will further provide devices for implementing the above methods.

[0163] Please refer to Figure 7 The embodiment of the application provides a structural diagram of a base station, which includes a processor 701, a transceiver 702, a memory 703 and a bus interface, wherein:

[0164] In the embodiment of the application, the base station further includes a program stored on the memory 703 and executable on the processor 701, and the program is executed by the processor 701 to implement the following steps:

[0165] Interact with the terminal through the first function plane of the base station to perform at least one of the following processes: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0166] It can be understood that the computer program is executed by the processor 701 to realize the above-mentioned Figure 5 The processes of the method embodiments shown above can be realized by the processor 701, and the same technical effects can be achieved. To avoid repetition, they will not be described here.

[0167] In Figure 7 The bus architecture can include any number of interconnected buses and bridges, which are collectively represented by the bus interface, linking various circuits such as one or more processors represented by the processor 701 and the memory represented by the memory 703. The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, and therefore, will not be further described herein. The bus interface provides an interface. The transceiver 702 can be a plurality of elements, i.e., including a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium.

[0168] The processor 701 is responsible for managing the bus architecture and general processing, and the memory 703 can store data used by the processor 701 when performing operations.

[0169] It should be noted that the terminal in this embodiment is a device corresponding to the method shown in the above Figure 5 The implementation modes in the above embodiments are applicable to the terminal embodiment, and the same technical effects can be achieved. In the device, the transceiver 702 and the memory 703, and the transceiver 702 and the processor 701 can be connected through the bus interface, the function of the processor 701 can also be realized by the transceiver 702, and the function of the transceiver 702 can also be realized by the processor 701. It should be noted that the above device provided by the embodiments of the present application can realize all method steps realized by the method embodiments, and can achieve the same technical effects. Here, the same parts and beneficial effects in the method embodiments will not be described in detail.

[0170] In some embodiments of the present application, a computer readable storage medium is also provided, which stores a program that is executed by a processor to realize the following steps:

[0171] Interact with the terminal through the first function plane of the base station to perform at least one of the following processes: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0172] The program is executed by the processor to implement all implementation manners in the AI endogenous implementation method applied to the base station, and achieve the same technical effects.

[0173] Please refer to Figure 8 The terminal provided by the embodiment of the application includes a processor 801, a transceiver 802, a memory 803, a user interface 804 and a bus interface.

[0174] In the embodiment of the application, the terminal further includes a program stored on the memory 803 and executable on the processor 801.

[0175] The processor 801 implements the following steps when executing the program.

[0176] The second function surface of the terminal is used to interact with the base station, and at least one of the following processes is executed: an artificial intelligence (AI) task control process, a data collection and reporting process, and an AI model transmission and synchronization process.

[0177] It can be understood that the computer program is executed by the processor 801 to implement the above Figure 6 The processes of the method embodiment are implemented, and the same technical effects are achieved, and thus details are not repeated here.

[0178] In Figure 8 , the bus architecture can include any number of interconnected buses and bridges, which are linked by various circuits of one or more processors represented by the processor 801 and the memory represented by the memory 803. The bus architecture can also link various other circuits such as peripheral devices, voltage stabilizers and power management circuits, which are well known in the art, and thus further description is not given herein. The bus interface provides an interface. The transceiver 802 can be a plurality of elements, i.e., including a transmitter and a receiver, which provides a unit for communicating with various other devices on a transmission medium. The user interface 804 can also be an interface that can be connected to the required device, including but not limited to a keypad, a display, a speaker, a microphone, a joystick, etc.

[0179] The processor 801 is responsible for managing the bus architecture and general processing, and the memory 803 can store data used by the processor 801 when performing operations.

[0180] It should be noted that the device in this embodiment is the same as the above Figure 6The device corresponding to the method shown above is applicable to the embodiments of this device and can achieve the same technical effect. In this device, the transceiver 802 and the memory 803, as well as the transceiver 802 and the processor 801, can be connected via a bus interface. The functions of the processor 801 can also be implemented by the transceiver 802, and vice versa. It should be noted that the device provided in this embodiment can implement all the method steps of the above method embodiments and achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiments will not be described in detail here.

[0181] In some embodiments of the present invention, a computer-readable storage medium is also provided, on which a program is stored, which, when executed by a processor, performs the following steps:

[0182] Through the second functional surface of the terminal, it interacts with the base station and executes at least one of the following processes: artificial intelligence (AI) task control process, data collection and reporting process, and AI model transmission and synchronization process.

[0183] When executed by the processor, this program can implement all the above-mentioned methods for AI endogenous implementation on the terminal side and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0184] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software 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 implementations should not be considered beyond the scope of this invention.

[0185] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0186] In the embodiments of the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the described embodiments of the apparatus are merely schematic. The units as divided can or can not be physically reallocated, and can or can not be components independent of each other. In some embodiments, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electric, mechanical or other forms.

[0187] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0188] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit.

[0189] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the present application that essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk, and various storage media that can store program codes.

[0190] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A base station, characterized in that, include: The first functional area is used to execute the following processes through interaction with the terminal: artificial intelligence (AI) task control process, data collection and reporting process, and AI model transmission and synchronization process; The first functional plane includes a first functional layer, which is used to execute AI task control process, AI task calculation, data collection and reporting process, and AI model transmission and synchronization process. The first functional layer is also used to decompose the AI ​​task requirements, determine the data collection requirements, AI model architecture, AI algorithm selection, and AI model evaluation algorithm of the base station side and the terminal side, and send AI task configuration information to the terminal through RRC signaling or MAC control information to configure the AI ​​tasks that the terminal needs to execute. The AI ​​task configuration information includes the AI ​​model segmentation method, AI model algorithm configuration, data collection requirement configuration, AI model performance evaluation algorithm, and task identifier to be executed.

2. The base station as described in claim 1, characterized in that, The base station further includes: an RRC layer functional module, a PDCP layer functional module, an RLC layer functional module, and a MAC layer functional module; wherein... The RRC layer functional module is used to send the information generated by the first functional layer to the terminal or network element on the network side via RRC signaling. The PDCP layer functional module, RLC layer functional module, and MAC layer functional module are used to transmit RRC signaling generated by the RRC layer functional module.

3. The base station as described in claim 2, characterized in that, The information generated and received by the first functional layer includes data of a first type and data of a second type. The data of the first type is control information used for the control of the AI ​​model of the base station and the terminal, including at least one of the following types of data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data acquisition and preprocessing, and task information for control configured for AI task evaluation. The data of the second type is data information transmitted by the base station and the terminal model, data acquisition and transmission, or data preprocessing, including at least one of the following types of data: parameters transmitted in AI model calculation, data acquired or preprocessed by data acquisition tasks, and tags.

4. The base station as described in claim 3, characterized in that, The MAC layer functional module is also used to send the parameters transmitted in the AI ​​model calculation generated by the first functional layer to the terminal or network element on the network side.

5. The base station as described in claim 3, characterized in that, The first functional layer is also used for: The receiving terminal sends preprocessed data to the base station via RRC signaling or MAC CE; Based on the preprocessed data, the AI ​​model is trained or jointly trained with the terminal to obtain the parameter configuration of the AI ​​model. The AI ​​model is then verified and evaluated using the collected data, corresponding labels, and AI model performance evaluation algorithms. Based on the updated AI model parameters obtained from training, send the updated AI model parameters to the terminal and / or network elements on the network side. The trained AI model is used to generate decision information for AI tasks, and the generated decision information is sent to the terminal or network element on the network side.

6. The base station as described in claim 5, characterized in that, The AI ​​task requirements include task requirements from the network side and / or task requirements reported from the terminal side.

7. A terminal, characterized in that, Including the second functional surface; The second functional plane is used to execute the following processes by interacting with the base station: artificial intelligence (AI) task control process, data collection and reporting process, and AI model transmission and synchronization process; The second functional plane includes a second functional layer, which is used to execute AI task control process, AI task calculation, data acquisition and reporting process, and AI model transmission and synchronization process. The second functional layer is also used to send AI task requirements to the base station through the RRC layer functional module; receive AI task configuration information sent by the base station through RRC signaling or MAC control information, and determine the AI ​​task to be executed. The AI ​​task configuration information includes the AI ​​model segmentation method, the configuration of the AI ​​model algorithm, the configuration of data collection requirements, the AI ​​model performance evaluation algorithm, and the task identifier to be executed by the terminal.

8. The terminal as described in claim 7, characterized in that, The information generated and received by the second functional layer includes data of the third type and data of the fourth type. The data of the third type is control information used for the control of the AI ​​model of the base station and the terminal, including at least one of the following types of data: AI task configuration information, control information for establishing an end-to-end AI model, control information for data acquisition and preprocessing, and task information for control configured for AI task evaluation. The data of the fourth type is data information transmitted by the base station and the terminal model, data acquired and transmitted or data preprocessed, including at least one of the following types of data: parameters transmitted in AI model calculation, data acquired or preprocessed by data acquisition tasks, and tags.

9. The terminal as described in claim 7, characterized in that, It also includes: RRC layer functional modules, PDCP layer functional modules, RLC layer functional modules, and MAC layer functional modules; among which, The RRC layer functional module is used to receive information sent by the base station via RRC signaling and forward it to the second functional module; The PDCP layer functional module, RLC layer functional module, and MAC layer functional module are used to transmit RRC signaling sent by the base station to the RRC layer functional module.

10. The terminal as described in claim 9, characterized in that, The MAC layer functional module is also used to receive parameters transmitted in the AI ​​model calculation generated by the base station and send them to the second functional layer.

11. The terminal as described in claim 8, characterized in that, The second functional layer is also used for: According to the configuration of data collection requirements, the data is collected and preprocessed, and the preprocessed data is sent to the base station via RRC signaling or MAC CE; Joint training of the AI ​​model is performed in collaboration with the base station; Receive the updated AI model parameters sent by the base station; Receive decision information sent by the base station.

12. A method for implementing AI endogenous capabilities, applied to a base station, the base station including a first functional plane, characterized in that, The method includes: Through the first functional surface of the base station, it interacts with the terminal and executes the following processes: artificial intelligence (AI) task control process, data collection and reporting process, and AI model transmission and synchronization process. The first functional surface includes a first functional layer; the method further includes: Through the first functional layer, the AI ​​task control process, AI task calculation, data collection and reporting process, and AI model transmission and synchronization process are executed. The first functional layer also performs the following steps: The AI ​​task requirements are decomposed to determine the data collection requirements, AI model architecture, AI algorithm selection, and AI model evaluation algorithm for both the base station and terminal sides. AI task configuration information is then sent to the terminal via RRC signaling or MAC control information to configure the AI ​​tasks that the terminal needs to execute. The AI ​​task configuration information includes the AI ​​model segmentation method, AI model algorithm configuration, data collection requirement configuration, AI model performance evaluation algorithm, and task identifier to be executed.

13. The method as described in claim 12, characterized in that, The base station further includes: an RRC layer functional module, a PDCP layer functional module, an RLC layer functional module, and a MAC layer functional module; the method further includes: The RRC layer functional module transmits the information generated by the first functional layer to the terminal or network element on the network side via RRC signaling. The RRC signaling generated by the RRC layer function module is transmitted through the PDCP layer function module, RLC layer function module and MAC layer function module.

14. A method for implementing AI endogenous capabilities, applied to a terminal, the terminal including a second functional surface, characterized in that, The method includes: Through the second functional surface of the terminal, it interacts with the base station and executes the following processes: artificial intelligence (AI) task control process, data collection and reporting process, and AI model transmission and synchronization process; The second functional surface includes a second functional layer; the method further includes: The second functional layer executes the AI ​​task control process, AI task calculation, data collection and reporting process, and AI model transmission and synchronization process. The second functional layer also performs the following steps: The AI ​​task requirements are sent to the base station through the RRC layer functional module; The terminal receives AI task configuration information sent by the base station via RRC signaling or MAC control information, determines the AI ​​task to be executed, and the AI ​​task configuration information includes the segmentation method of the AI ​​model, the configuration of the AI ​​model algorithm, the configuration of data collection requirements, the AI ​​model performance evaluation algorithm, and the task identifier to be executed by the terminal.

15. The method as described in claim 14, characterized in that, The terminal further includes an RRC layer functional module, a PDCP layer functional module, an RLC layer functional module, and a MAC layer functional module; the method further includes: The RRC layer functional module receives information sent by the base station via RRC signaling and forwards it to the second functional module. The PDCP layer functional module, RLC layer functional module, and MAC layer functional module transmit RRC signaling sent by the base station to the RRC layer functional module.

16. A base station, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the method as described in any one of claims 12 to 13.

17. A terminal, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, performs the steps of the method as described in any one of claims 14 to 15.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 12 to 15.

Images