Communication method and communication device

Abstract

The invention discloses a communication method and a communication device, and relates to the technical field of communication. In the method, a center node can obtain request information for training a first AI model from a first radio unit, the center node can determine whether the first AI model can be helped to be trained or not, and when the center node determines that the first AI model can be helped to be trained, the first AI model can be trained. The precision of the first AI model obtained after training of the center node is superior to that of the first AI model obtained after training of the first radio unit. In this way, the precision of the AI model applied by the radio unit can be improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and more specifically, to a communication method and a communication device. Background Art

[0002] In the field of wireless communications, the industry is exploring the use of artificial intelligence (AI) to reconstruct the core functions of the radio access network (RAN), for example, enabling the RAN to have AI capabilities (including training and reasoning capabilities).

[0003] RAN AI capabilities can evolve from the RAN intelligent controller (RIC) to native AI. In other words, RAN AI capabilities can evolve from being deployed only in the RIC to being implemented in distributed nodes such as the central unit (CU), distributed unit (DU), and radio unit (RU). However, when the RU applies an AI model, the accuracy of the model may not meet the RU's requirements. Summary of the Invention

[0004] The present application provides a communication method and a communication device that can support improving the accuracy of the AI model used by the radio unit.

[0005] In a first aspect, a communication method is provided, including: sending a first request message, wherein the first request message requests training of a first AI model; and receiving a response message, wherein the response message indicates whether to agree to train the first AI model.

[0006] The implementation of the solution described in the first aspect may be the first radio unit, a module within the first radio unit (such as a chip system), or a logical node, logic module, or software that implements all or part of the first radio unit's functions, without limitation. For ease of description, the following description uses the first radio unit as an example.

[0007] In the above solution, the first radio unit sends a first request message to the central node and receives a response message from the central node. By establishing an interactive mechanism for exchanging request messages between the first radio unit and the central node, the embodiment of the present application can support improving the accuracy of the AI model used by the radio unit.

[0008] Compared with the solution in which the first radio unit independently trains the first AI model, the first radio unit can request the central node to help train the first AI model through the exchange of information requests between the first radio unit and the central node. The training capability of the central node is better than the training capability of the first radio unit. By having the help of the central node to complete the training of the first AI model, it can avoid the problem of insufficient model accuracy of the first AI model obtained after training due to insufficient training data, uneven data distribution, or a relatively single data source of the first radio unit.

[0009] Through the above-mentioned interaction mechanism, the central node can obtain request information for training the first AI model from the first radio unit, and the central node can determine whether it can help train the first AI model. When the central node determines that it can help train the first AI model, the accuracy of the first AI model obtained after training by the central node will be better than the accuracy of the first AI model obtained after training by the first radio unit.

[0010] Based on the above-mentioned interaction mechanism, the first radio unit can have the opportunity to request the central node to help train the first AI model. Compared with the first radio unit completing the training of the first AI model independently, the training of the first AI model is completed with the help of the central node. This can overcome the problem of insufficient model accuracy of the first AI model obtained after training due to insufficient training data volume, uneven data distribution, or a relatively single data source of the first radio unit, and thus support the improvement of the accuracy of the AI model used by the radio unit.

[0011] In the first aspect, the method also includes: sending first training data, where the first training data is used to train at least one of the common parameters and characteristic parameters of the first AI model, where the common parameters of the first AI model are parameters shared by the first AI model and the second AI model, and the characteristic parameters of the first AI model are parameters that distinguish the first AI model from the second AI model; and receiving a third AI model, where the third AI model is a model obtained after training the first AI model.

[0012] When the response information indicates that the central node agrees to train the first AI model, the first radio unit can report the first training data to the central node. The central node can use the first training data to train the first AI model and send the trained third AI model to the first radio unit. In this way, the first radio unit can obtain the third AI model trained by the central node.

[0013] In the first aspect, receiving the third artificial intelligence model includes: sending a second request message, the second request message requesting to obtain the third AI model; and receiving the third AI model.

[0014] Based on the above information interaction, the first radio unit can obtain the third AI model obtained after training from the central node.

[0015] In the first aspect, before sending the second request information, the method also includes: receiving notification information, which indicates that the training of the first AI model is completed.

[0016] In this way, when the central node completes the training of the first AI model, the central node can indicate to the first radio unit that the training of the first AI model is completed, and the first radio unit can determine that it can request to obtain the third AI model based on this.

[0017] In the first aspect, the method further includes: sending third request information, wherein the third request information requests updating the third AI model; and receiving the updated third AI model.

[0018] Through the above information interaction, the first radio unit can complete the update process of the third AI model and obtain the updated third AI model from the central node.

[0019] In the first aspect, the third model is a model obtained by training the first model based on first training data and second training data, the second training data comes from a second radio unit, and the second radio unit is different from the first radio unit.

[0020] By using training data from other radio units to train the first AI model, the source, quantity, type and distribution of data used for training the first AI model can be broadened, thereby supporting the improvement of the accuracy of the third AI model.

[0021] In a second aspect, a communication method is provided, including: receiving first request information, the first request information requesting training of a first AI model; and sending response information, the response information indicating whether to agree to train the first AI model.

[0022] The execution entity of the solution described in the second aspect can be a central node, a module within the central node (such as a chip system), or a logical node, logic module, or software that can implement all or part of the central node's functions, without limitation. For ease of description, the following description uses the central node as an example.

[0023] In the above solution, the first radio unit sends first request information to the central node and receives response information from the central node.

[0024] Based on the above-mentioned interaction mechanism, the first radio unit can have the opportunity to request the central node to help train the first AI model. Compared with the first radio unit completing the training of the first AI model independently, the training of the first AI model is completed with the help of the central node. This can overcome the problem of insufficient model accuracy of the first AI model obtained after training due to insufficient training data volume, uneven data distribution, or a relatively single data source of the first radio unit, and thus support the improvement of the accuracy of the AI model specifically used by the radio unit.

[0025] In the second aspect, the method also includes: receiving first training data, the first training data is used to train at least one of the common parameters and characteristic parameters of the first AI model, the common parameters of the first AI model are parameters shared by the first AI model and the second AI model, and the characteristic parameters of the first AI model are parameters that distinguish the first AI model from the second AI model; training the first AI model according to the first training data to obtain a third AI model; and sending the third AI model.

[0026] When the response information indicates that the central node agrees to train the first AI model, the first radio unit can report the first training data to the central node. The central node can use the first training data to train the first AI model and send the trained third AI model to the first radio unit. In this way, the first radio unit can obtain the AI model trained by the central node.

[0027] In the second aspect, sending the third AI model includes: receiving second request information, the second request information requesting to obtain the third AI model; and sending the third AI model.

[0028] In the second aspect, before receiving the second request information, the method also includes: sending a notification message indicating that the training of the first AI model is completed.

[0029] In the second aspect, the method further includes: receiving third request information, the third request information requesting to update the third AI model; and sending the updated third AI model.

[0030] In the second aspect, the first AI model is trained according to the first training data to obtain a third model, including: training the first AI model according to the first training data and the second training data to obtain the third model; the second training data comes from the second radio unit, and the second radio unit is different from the first radio unit.

[0031] By using training data from other radio units to train the first AI model, the sources and types of data used for training the first AI model can be broadened, thereby supporting the improvement of the accuracy of the third AI model obtained after training.

[0032] In the second aspect, the first AI model is trained according to the first training data to obtain a third artificial intelligence model, including: determining the common parameters of the first AI model and the characteristic parameters of the first AI model; and training the common parameters of the first AI model and the characteristic parameters of the first AI model respectively according to the first training data.

[0033] By dividing the parameters of the first AI model into common parameters and characteristic parameters and training them separately, the common parameters obtained after training can be applied to other AI models and the training overhead can be reduced. For example, the common parameters only need to be trained once, and the differentiated needs of the first AI model can be met.

[0034] In combination with the method described in any of the first and second aspects, the first request information includes at least one of the following: the name of the first AI model, the parameters to be trained of the first AI model, the identifier of the first radio unit, or the identifier of the cell corresponding to the first radio unit; the first radio unit is a radio unit that applies the first AI model.

[0035] In combination with the method described in any of the first and second aspects, the parameter to be trained is at least one of the characteristic parameters and common parameters of the first AI model.

[0036] In combination with the method of any of the first and second aspects, the response information includes at least one of the following: a name of the first AI model, an identifier of the first radio unit, and confirmation information or non-confirmation information. The confirmation information indicates consent to train the first AI model, and the non-confirmation information indicates disapproval of training the first AI model; the first radio unit is a radio unit that applies the first AI model.

[0037] In combination with the method described in any of the first and second aspects, the first request information includes performance requirement information for the first AI model.

[0038] By reporting the performance requirement information for the first AI model, the central node can train the first AI model in a targeted manner based on the performance requirement information for the first AI model. This can avoid the situation where the first AI model obtained after training cannot meet the performance requirements of the first radio unit, thereby resulting in invalid training, and further increasing the training overhead of the central node for the first AI model.

[0039] In a third aspect, a communication device is provided, which includes: an interface unit for sending a first request message, wherein the first request message requests training of a first AI model; the interface unit is also used to receive a response message, wherein the response message indicates whether to agree to train the first AI model.

[0040] The above-mentioned communication device can also be used to execute the solution described in the method described in the first aspect and any possible manner of the first aspect, which will not be repeated here.

[0041] In a fourth aspect, a communication device is provided, which includes: an interface unit for receiving a first request message, wherein the first request message requests training of a first AI model; the interface unit is also used to send a response message, wherein the response message indicates whether to agree to train the first AI model.

[0042] The above-mentioned communication device can also be used to execute the solution described in the method described in the aforementioned second aspect and any possible manner of the second aspect, which will not be repeated here.

[0043] In a fifth aspect, a communication device is provided, comprising a processor, wherein the processor is configured to, by executing a computer program or instruction, or by a logic circuit, enable the communication device to execute the method described in the first aspect and any possible manner of the first aspect; or, enable the communication device to execute the method described in the second aspect and any possible manner of the second aspect.

[0044] In a possible implementation, the communication device further includes a memory for storing the computer program or instruction.

[0045] In a possible implementation, the communication device further includes a communication interface, which is used to input and / or output signals.

[0046] In a sixth aspect, a communication device is provided, comprising a logic circuit and an input / output interface, the input / output interface being used to input and / or output signals, the logic circuit being used to execute the method described in the first aspect and any possible manner of the first aspect, or the logic circuit being used to execute the method described in the first aspect and any possible manner of the first aspect, or the logic circuit being used to execute the method described in the second aspect and any possible manner of the second aspect.

[0047] In the seventh aspect, a computer-readable storage medium is provided, on which a computer program or instruction is stored. When the computer program or the instruction is run on a computer, the method described in the first aspect and any possible manner of the first aspect is executed, or the method described in the second aspect and any possible manner of the second aspect is executed.

[0048] In an eighth aspect, a computer program product is provided, comprising instructions, which, when executed on a computer, cause the method described in the first aspect and any possible manner of the first aspect to be executed, or cause the method described in the second aspect and any possible manner of the second aspect to be executed.

[0049] In the ninth aspect, a chip system is provided, which is connected to a memory and is used to read and execute a software program stored in the memory to execute the method described in the first aspect and any possible manner in the first aspect, or to execute the method described in the second aspect and any possible manner in the second aspect.

[0050] In the tenth aspect, a chip system is provided, which includes: a communication interface for communicating with other devices; a processor for enabling a communication device equipped with the chip system to execute the method described in the first aspect and any possible manner in the first aspect, or for enabling a communication device equipped with the chip system to execute the method described in the second aspect and any possible manner in the second aspect.

[0051] In the eleventh aspect, a chip system is provided, which includes a processor, a memory and an input / output port, wherein the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory, so that the processor executes the method described in the first aspect and any possible manner in the first aspect, or so that the processor executes the method described in the second aspect and any possible manner in the second aspect.

[0052] In the twelfth aspect, a chip system is provided, which is applied to an electronic device, and the chip system includes one or more processors, which are used to call computer instructions to enable the electronic device to execute the method described in the above-mentioned first aspect and any possible manner in the first aspect, or, to enable the electronic device to execute the method described in the above-mentioned second aspect and any possible manner in the second aspect, or.

[0053] In the thirteenth aspect, a communication system is provided, including: a first radio unit and a central node: the first radio unit is used to send a first request message to the central node, and the first request message is used to request training of a first artificial intelligence model; the central node is used to receive the first request message, and send a response message to the first radio unit, and the response message indicates whether to agree to train the first artificial intelligence model.

[0054] The first radio unit and the central node mentioned above can also be used to execute the aforementioned method, which will not be described in detail.

[0055] The description of the advantageous effects of any of the third to thirteenth aspects etc. may refer to the description of the advantageous effects of the first and second aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of an application framework of an embodiment of the present application.

[0057] Figure 2 It is a schematic diagram of another application framework of an embodiment of the present application.

[0058] Figure 3 It is a schematic diagram of a communication system applicable to an embodiment of the present application.

[0059] Figure 4 It is a schematic diagram of another communication system applicable to the embodiments of the present application.

[0060] Figure 5 It is a schematic diagram of the interaction flow of a communication method in an embodiment of the present application.

[0061] Figure 6 This is a schematic diagram of the interaction flow of another communication method in an embodiment of the present application.

[0062] Figure 7 This is a schematic diagram of the interaction flow of another communication method according to an embodiment of the present application.

[0063] Figure 8 This is a schematic block diagram of a communication device according to an embodiment of the present application.

[0064] Figure 9 This is a schematic block diagram of another communication device according to an embodiment of the present application. DETAILED DESCRIPTION

[0065] The technical solution in this application will be described below with reference to the accompanying drawings.

[0066] In order to facilitate understanding of the embodiments of the present application, the following points are first explained.

[0067] 1. In this application, unless otherwise specified, "a plurality of or at least two" means two or more.

[0068] 2. In each embodiment of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form a new embodiment based on their internal logical relationships.

[0069] 3. The various numerical numbers involved in this application are only used for the convenience of description and are not used to limit the scope of protection of this application. The size of the serial numbers involved in this application does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic. For example, the terms "first", "second", "third", "fourth" and other various terminology labels (if any) in the specification and claims and drawings of this application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. Among them, the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than what is illustrated or described here.

[0070] At the same time, any embodiment or design described in this application as "exemplary" or "for example" should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner to facilitate understanding.

[0071] 4. The terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus that includes 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 process, method, product or apparatus.

[0072] 5. In this application, "used to indicate" can be understood as "enabling," and "enabling" can include direct enabling and indirect enabling. When describing that certain information is used to enable A, it can include that the information directly enables A or indirectly enables A, and does not necessarily mean that the information contains A.

[0073] The information enabled by the information is called information to be enabled. In the specific implementation process, there are many ways to enable the enabled information, such as but not limited to, directly enabling the information to be enabled, such as the information to be enabled itself or the index of the information to be enabled. The information to be enabled can also be indirectly enabled by enabling other information, wherein there is an association between the other information and the information to be enabled. It is also possible to enable only a part of the information to be enabled, while the other parts of the information to be enabled are known or agreed in advance. For example, it is also possible to enable specific information with the help of the arrangement order of each piece of information agreed in advance (such as specified in the protocol), thereby reducing the enabling overhead to a certain extent. At the same time, it is also possible to identify the common parts of each piece of information and enable them uniformly to reduce the enabling overhead caused by enabling the same information separately.

[0074] 6. In this application, "pre-configuration" may include pre-definition, such as protocol definition. "Pre-definition" may be implemented by pre-storing corresponding codes, tables, or other methods that can be used to indicate relevant information in a device (e.g., including each network element). This application does not limit the specific implementation method.

[0075] 7. "Storage" or "saving" as used in this application may refer to storage in one or more memories. The one or more memories may be provided separately or integrated into an encoder or decoder, a processor, or a communication device. The one or more memories may also be provided in part separately and in part integrated into a decoder, processor, or communication device. The type of memory may be any form of storage medium and is not limited thereto.

[0076] 8. The “protocol” referred to in this application may refer to a standard protocol in the field of communications, such as the fourth generation (4G) network, the fifth generation (5G) network protocol, the new radio (NR) protocol, the 5.5G network protocol, the sixth generation (6 th generation, 6G) network protocols and related protocols used in future communication systems, which are not limited in this application.

[0077] 9. The arrows or boxes indicated by dotted lines in the schematic diagrams in the accompanying drawings of this application specification represent optional steps or optional modules.

[0078] 10. In this application, unless otherwise specified, “ / ” indicates that the objects associated with each other are in an “or” relationship. For example, A / B can mean A or B. “And / or” in this application is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural.

[0079] 11. In this application, indication includes direct indication (also called explicit indication) and implicit indication. Direct indication of information A means including information A. Implicit indication of information A means indicating information A through the correspondence between information A and information B and the direct indication of information B. The correspondence between information A and information B can be predefined, pre-stored, pre-burned, or pre-configured.

[0080] 12. In this application, the use of information C to determine information D includes both situations where information D is determined solely based on information C and situations where information D is determined based on information C and other information. Furthermore, information C can also be used to determine information D indirectly, for example, where information D is determined based on information E, and information E is determined based on information C.

[0081] 13. In this application, "device A sends information A to device B" can be understood as the destination end of the information A or the intermediate network element in the transmission path between the destination end and the device B, which may include sending information to device B directly or indirectly.

[0082] 14. In this application, the phrase "Device B receives information A from Device A" should be understood to mean that the source of information A or an intermediate network element in the transmission path between the source and the device A is Device A, and may include directly or indirectly receiving the information from Device A. Information may undergo necessary processing between the source and destination, such as formatting changes, but the destination can still understand the valid information from the source. Similar expressions in this application should be understood similarly and are not elaborated on here.

[0083] First, a communication system to which the embodiments of the present application are applicable is described.

[0084] The technical solution provided by this application can be applied to various communication systems, such as the fifth generation communication system (5 th generation, 5G), new radio (NR) system, long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD) system, wireless local area network (WLAN) system, satellite communication system and other communication systems, such as the sixth generation communication system (6 th generation, 6G) etc.

[0085] The technical solution provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC) and Internet of Things (IoT) communication systems or other communication systems.

[0086] A device in the above-mentioned communication system can send signals to another device or receive signals from another device. The signals may include information, signaling, or data. The term "device" may also be replaced by an entity, network entity, network element, communication device, communication module, node, user equipment, mobile device, communication node, etc. The embodiments of this application are described using devices as an example. For example, the communication system may include at least one terminal device and at least one network device. The network device may send downlink signals to the terminal device, and / or the terminal device may send uplink signals to the network device.

[0087] In an embodiment of the present application, the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device.

[0088] The terminal device may be a device that provides voice / data, such as a handheld device or vehicle-mounted device with a wireless connection function. At present, some examples of terminals are: mobile phones, tablet computers, laptop computers, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, wearable devices, terminal devices in 5G networks or future evolved public land mobile communication networks (PLMNs). The terminal equipment in the network (PLMN), etc., is not limited to this in the embodiments of the present application.

[0089] In the embodiment of the present application, the terminal device may also be a wearable device. Wearable devices may also be called wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0090] In the embodiments of the present application, the device for realizing the function of the terminal device can be a terminal device, or a device capable of supporting the terminal device to realize the function, such as a chip system, which can be installed in the terminal device or used in combination with the terminal device. In the embodiments of the present application, the chip system can be composed of a chip, or it can include a chip and other discrete devices. In the embodiments of the present application, only the terminal device is used as an example for description, and the embodiments of the present application are not limited to the solutions of the embodiments of the present application.

[0091] The network device in the embodiment of the present application may be a device for communicating with a terminal device, and the network device may also be referred to as an access network device or a wireless access network device, such as a base station. The network device in the embodiment of the present application may refer to a RAN node (or device) that connects a terminal device to a wireless network. Base station can broadly cover various names as follows, or replace the following names, such as: NodeB, evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmission point (TRP), transmission point (TP), master station, auxiliary station, multi-standard radio (motor slide retainer, MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), CU, DU, RU, positioning node, RIC, etc. The base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof. The base station can also refer to a communication module, modem or chip used to be set in the aforementioned device or apparatus. The base station can also be a mobile switching center, a device that performs base station functions in D2D, V2X, and M2M communications, a network-side device in a 6G network, or a device that performs base station functions in future communication systems. The base station can support networks with the same or different access technologies. Optionally, the RAN node can also be a server, a wearable device, a vehicle or an onboard device. For example, the access network device in V2X technology can be a roadside unit (RSU). The embodiments of this application do not limit the specific technology and specific device form adopted by the network device.

[0092] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move based on the location of the mobile base station. In other examples, a helicopter or drone can be configured to act as a device that communicates with another base station.

[0093] In some deployments, the network device may include a CU or a DU, or a device including a CU and a DU, or a device including a control plane CU node (central unit-control plane (CU-CP)), a user plane CU node (central unit-user plane (CU-UP)), and a DU node. For example, the network device may include a gNB-CU-CP, a gNB-CU-UP, and a gNB-DU.

[0094] In different communication systems, CU (or CU-CP and CU-UP), DU or RU may also have different names, but those skilled in the art can understand their meanings. For example, in an open radio access network (ORAN) system, CU may also be referred to as an open CU (open CU, O-CU), DU may also be referred to as an open DU (open DU, O-DU), CU-CP may also be referred to as O-CU-CP, CU-UP may also be referred to as O-CU-UP, and RU may also be referred to as O-RU. Any of the CU (or CU-CP, CU-UP), DU and RU in this application may be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0095] In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes implementing portions of the base station's functionality. For example, a RAN node can be a CU, DU, CU-CP, CU-UP, or RU. The CU and DU can be separate or included in the same network element, such as the BBU. The RU can be included in a radio frequency device or radio unit, such as the RRU, AAU, or RRH.

[0096] The RAN node may support one or more types of fronthaul interfaces, with different fronthaul interfaces corresponding to DUs and RUs with different functions. If the fronthaul interface between the DU and the RU is a common public radio interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and the RU is another type of interface, relative to the CPRI, some of the downlink and / or uplink baseband functions, such as precoding, digital beamforming (BF), or one or more of inverse fast Fourier transform (IFFT) / cyclic prefix (CP) for downlink, are moved from the DU to the RU for implementation; and for uplink, one or more of digital beamforming (BF), or fast Fourier transform (FFT) / cyclic prefix (CP) removal, are moved from the DU to the RU for implementation.

[0097] In one possible implementation, the interface may be an enhanced common public radio interface (eCPRI). In the eCPRI architecture, the division between the DU and RU is different, corresponding to different types (category, Cat) of eCPRI, such as eCPRI Cat A, B, C, D, E, and F.

[0098] Taking eCPRI Cat A as an example, for downlink transmission, based on layer mapping, the DU is configured to implement layer mapping and one or more functions preceding it (i.e., one or more of coding, rate matching, scrambling, modulation, and layer mapping), while other functions after layer mapping (e.g., RE mapping, BF, or one or more of IFFT / CP) are moved to the RU for implementation. For uplink transmission, based on de-RE mapping, the DU is configured to implement de-mapping and one or more functions preceding it (i.e., one or more of decoding, de-rate matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and de-RE mapping), while other functions after de-mapping (e.g., one or more of digital BF or fast Fourier transform (FFT) / CP removal) are moved to the RU for implementation. It is understood that for the functional description of the DU and RU corresponding to various types of eCPRI, please refer to the eCPRI protocol and will not be repeated here.

[0099] In one possible design, the processing unit for implementing baseband functions in the BBU is called a baseband high layer (BBH) unit, and the processing unit for implementing baseband functions in the RRU / AAU / RRH is called a baseband low layer (BBL) unit.

[0100] In the embodiments of the present application, the device for implementing the functions of the network device can be a network device; it can also be a device that can support the network device to implement the functions, such as a chip system, a hardware circuit, a software module, or a hardware circuit and a software module. The device can be installed in the network device or used in conjunction with the network device. In the embodiments of the present application, only the device for implementing the functions of the network device is used as an example to illustrate, and does not constitute a limitation on the solutions of the embodiments of the present application.

[0101] The network device and / or terminal device can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; it can also be deployed on the water surface; it can also be deployed on aircraft, balloons and satellites in the air. The embodiments of this application do not limit the scenarios in which the network device and the terminal device are located. In addition, the terminal device and the network device can be hardware devices, or they can be software functions running on dedicated hardware, software functions running on general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities including dedicated or general-purpose hardware devices and software functions. This application does not limit the specific forms of the terminal device and the network device.

[0102] In wireless communication networks, the services supported by the networks are becoming increasingly diverse, and therefore the demands they need to meet are becoming increasingly diverse. For example, the network needs to be able to support ultra-high speeds, ultra-low latency, and / or ultra-large connections. This makes network planning, network configuration, and / or resource scheduling increasingly complex. In addition, as network functionality becomes increasingly powerful, such as supporting increasingly high spectrum, supporting advanced multiple input multiple output (MIMO) technology, supporting beamforming, and / or supporting new technologies such as beam management, network energy conservation has become a hot research topic. These new demands, new scenarios, and new features pose unprecedented challenges to network planning, maintenance, and efficient operations. To meet this challenge, artificial intelligence technology can be introduced into wireless communication networks to achieve network intelligence.

[0103] In order to support AI technology in wireless networks, AI nodes may also be introduced into the network.

[0104] Optionally, the AI node can be deployed in one or more of the following locations in the communication system: access network equipment, terminal equipment, or core network equipment. Alternatively, the AI node can be deployed separately, for example, in a location other than any of the above devices, such as a host or cloud server in an over-the-top (OTT) system. The AI node can communicate with other devices in the communication system, such as one or more of the following: network equipment, terminal equipment, or network elements of the core network.

[0105] It is understood that the embodiments of the present application do not limit the number of AI nodes. For example, when there are multiple AI nodes, the multiple AI nodes can be divided based on function, such as different AI nodes are responsible for different functions.

[0106] It is also understood that AI nodes can be independent devices, or integrated into the same device to implement different functions, or can be network elements in hardware devices, or can be software functions running on dedicated hardware, or can be virtualized functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of AI nodes. Among them, AI nodes can be AI network elements or AI modules.

[0107] Figure 1 This is a schematic diagram of an application framework of an embodiment of the present application. Figure 1 As shown, network elements are connected through interfaces (such as NG and Xn) or air interfaces. These network element nodes, such as core network equipment, access network nodes, terminals, or one or more devices in operation administration and maintenance (OAM), are equipped with one or more AI modules (for clarity, Figure 1 Only one is shown in the figure. An access network node can be a single RAN node or include multiple RAN nodes, for example, a CU and a DU. The CU and / or DU can also be configured with one or more AI modules. Optionally, the CU can be split into a CU-CP and a CU-UP. The CU-CP and / or CU-UP are configured with one or more AI models.

[0108] The AI module is used to implement the corresponding AI function. The AI modules deployed in different network elements may be the same or different. The model of the AI module can implement different functions according to different parameter configurations. The model of the AI module can be configured based on one or more of the following parameters: structural parameters (such as the number of neural network layers, the width of the neural network, the connection relationship between layers, the weight of the neuron, the activation function of the neuron, or at least one of the bias in the activation function), input parameters (such as the type of input parameters and / or the dimension of the input parameters), or output parameters (such as the type of output parameters and / or the dimension of the output parameters). Among them, the bias in the activation function can also be called the bias of the neural network.

[0109] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or on the same node or device.

[0110] Figure 2 This is a schematic diagram of another application framework of an embodiment of the present application. Figure 2 As shown, the communication system includes RIC. For example, RIC can be Figure 1 AI modules 117 and 118 shown in FIG are used to implement AI-related functions. RIC includes near-real-time RIC (near-RT RIC) and non-real-time RIC (non-RTRIC). Non-real-time RIC mainly processes non-real-time information, such as data that is not sensitive to latency and has a latency of seconds. Real-time RIC mainly processes near-real-time information, such as data that is relatively sensitive to latency and has a latency of tens of milliseconds.

[0111] Near-RT RIC is used for model training and reasoning. For example, it is used to train an AI model and use the AI model for reasoning. Near-RT RIC can obtain network-side and / or terminal-side information from RAN nodes (such as CU, CU-CP, CU-UP, DU and / or RU) and / or terminals. This information can be used as training data or reasoning data. Optionally, Near-RT RIC can submit the reasoning results to the RAN node and / or terminal. Optionally, the reasoning results can be exchanged between the CU and DU, and / or between the DU and RU. For example, Near-RT RIC submits the reasoning results to the DU, and the DU sends it to the RU.

[0112] Non-RT RIC is also used for model training and reasoning. For example, it is used to train AI models and use the model for reasoning. Non-RTRIC can obtain network-side and / or terminal-side information from RAN nodes (such as CU, CU-CP, CU-UP, DU and / or RU) and / or terminals. This information can be used as training data or reasoning data, and the reasoning results can be submitted to the RAN node and / or terminal. Optionally, the reasoning results can be exchanged between the CU and DU, and / or between the DU and RU. For example, Non-RTRIC submits the reasoning results to the DU, and the DU sends it to the RU.

[0113] The Near-RT RIC and Non-RT RIC may also be separately configured as a network element. Optionally, the Near-RT RIC and Non-RT RIC may also be part of other devices. For example, the Near-RT RIC may be configured in a RAN node (e.g., a CU or DU), while the Non-RT RIC may be configured in an OAM, a cloud server, a core network device, or other network devices.

[0114] Figure 3 Schematic diagram of a communication system applicable to the embodiment of the present application. Figure 3 As shown, the communication system includes: a network device 110, a terminal device (such as a terminal device 120 and a terminal device 130), and an AI network element 140. The AI network element 140 is used to perform AI-related operations, such as building a training data set or training an AI model.

[0115] The network device 110 may send data related to the training of the AI model to the AI network element 140, which constructs a training data set and trains the AI model. For example, the data related to the training of the AI model may include data reported by the terminal device. The AI network element 140 may send the results of the operations related to the AI model to the network device 110, and forward them to the terminal device through the network device 110. For example, the results of the operations related to the AI model may include at least one of the following: an AI model that has completed training, an evaluation result or a test result of the model, etc. Exemplarily, a portion of the trained AI model may be deployed on the network device 110, and another portion may be deployed on the terminal device. Alternatively, the trained AI model may be deployed on the network device 110. Alternatively, the trained AI model may be deployed on the terminal device.

[0116] Figure 3 The following description uses the example of a direct connection between AI network element 140 and network device 110. However, AI network element 140 may also be connected to a terminal device. Alternatively, AI network element 140 may be connected to both network device 110 and a terminal device. Alternatively, AI network element 140 may be connected to network device 110 through a third-party network element. This embodiment of the present application does not limit the connection relationship between the AI network element and other network elements.

[0117] The AI network element 140 can also be provided as a module in a network device and / or a terminal device, for example, Figure 3 In the network device 110 or terminal device shown.

[0118] Figure 4 Schematic diagram of another communication system to which the embodiment of the present application is applicable. Figure 4 As shown, the communication system includes: a first network unit, a second network unit, a third network unit, an O-eNB, an O-CU-CP, an O-CU-UP, an O-DU, an O-RU and an O-cloud.

[0119] The above network elements (also referred to as nodes) can be connected to each other. For example, the first network unit is connected to the O-cloud through the O2 interface, the first network unit is connected to the third network unit, O-eNB, O-CU-CP, O-CU-UP, O-DU and O-RU through the O1 interface, the first network unit is connected to the O-RU through the open fronthaul M-Plane interface, the O-DU is connected to the O-RU through the open fronthaul M-Plane interface and the open fronthaul C / U / S-Plane interface, the third network unit is connected to the O-eNB, O-CU-CP, O-CU-UP and O-DU through the E2 interface, the O-CU-CP is connected to the O-DU through the F1-c interface, the O-CU-UP is connected to the O-DU through the F1-u interface, and the O-CU-CP is connected to the O-CU-UP through the E1 interface. Figure 4 For a detailed description of the interface shown, please refer to the existing standards and will not be repeated here.

[0120] As a possible example, the first network unit may be a service management and orchestration framework (SMO), or a network unit with functions similar to those of the SMO, which is not limited.

[0121] In a possible example, the second network unit may be a Non-RT RIC, or a network unit with functions similar to those of the Non-RT RIC, which is not limited.

[0122] As a possible example, the third network unit may be a Near-RT RIC, or a network unit with functions similar to those of a Near-RTRIC, which is not limited.

[0123] Next, some technical concepts involved in this application are briefly described.

[0124] Machine learning (ML): ML is an important technical approach to realizing AI.

[0125] Deep neural networks (DNNs) are a specific implementation of machine learning (ML). According to the universal approximation theorem, neural networks can theoretically approximate any continuous function, enabling them to learn arbitrary mappings. Traditional communication systems require extensive expert knowledge to design communication modules. However, DNN-based deep learning communication systems can automatically discover implicit patterns in massive data sets and establish mapping relationships between data, achieving performance superior to traditional modeling methods.

[0126] According to the network construction method, DNN can be divided into feedforward neural network (FNN), convolutional neural network (CNN) and recurrent neural network (RNN).

[0127] CNN is a neural network specifically designed to process data with a grid-like structure. For example, time series data (discrete sampling along the time axis) and image data (discrete sampling along two dimensions) can both be considered grid-like data. CNNs do not utilize all input information at once for computation. Instead, they use a fixed-size window to intercept a portion of the information for convolution operations, significantly reducing the computational complexity of model parameters. Furthermore, depending on the type of information intercepted by the window (e.g., people and objects in an image represent different types of information), each window can use a different convolution kernel, enabling CNNs to better extract features from the input data.

[0128] RNNs are a type of DNN that utilizes feedback time series information. Their input consists of a new input value at the current moment and their own output value at the previous moment. RNNs are suitable for capturing temporally correlated sequence features and are particularly well-suited for applications such as speech recognition and channel coding.

[0129] The above-mentioned FNN, CNN, and RNN are common neural network structures, which are all constructed based on neurons. As mentioned above, each neuron performs a weighted sum operation on its input values, and the weighted summation result generates an output through a nonlinear function. We call the weights of the weighted summation operation of neurons in the neural network and the nonlinear function the parameters of the neural network. Taking the neuron with max{0,x} as the nonlinear function as an example, The parameters of the neuron to be operated are weights w=[w0,…,w n The weighted sum is biased by b, and the nonlinear function max{0,x}. The parameters of all neurons in a neural network constitute the parameters of the neural network.

[0130] AI Model: An AI model is an algorithm or computer program that implements AI functionality. It represents the mapping between the model's input and output. AI models can be neural networks, linear regression models, decision tree models, support vector machines (SVMs), Bayesian networks, Q-learning models, or other ML models.

[0131] The implementation of an AI model or AI function may be hardware circuitry, software, or a combination of software and hardware, without limitation. Non-limiting examples of software include: program code, program, subroutine, instruction, instruction set, code, code segment, software module, application, or software application.

[0132] exist Figure 4 In the communication architecture shown, when the RU has AI capabilities, the RU can apply the AI model, but the accuracy of the AI model may not meet the RU's requirements.

[0133] In view of this, the present application provides a communication method and a communication device that can support improving the accuracy of the AI model applied by the radio unit.

[0134] The following describes the interaction process of the communication method according to the embodiment of the present application with reference to the accompanying drawings.

[0135] Figure 5 It is a schematic diagram of the interaction flow of a communication method in an embodiment of the present application. Figure 5 The method shown can be performed by the first radio unit and the central node (for example, the central node can be one or more network elements such as the O-DU, O-CU, the third network unit, the second network unit, and the first network unit), or by modules and / or devices (for example, chips or integrated circuits, etc.) with corresponding functions installed in the first radio unit and the central node. The following description takes the first radio unit and the central node as an example. Figure 5 As shown, the method includes:

[0136] S501: A first radio unit sends request information 1 (such as first request information) to a central node.

[0137] Correspondingly, the central node receives the request information 1.

[0138] The request information 1 is used to request training of the AI model 1 (such as the first AI model), or in other words, the request information 1 is used by the first radio unit to request the central node to train the AI model 1.

[0139] AI model 1 can be understood as an AI model that needs to be trained, or in other words, AI model 1 is an original AI model that needs AI training, or in other words, the parameters in AI model 1 are parameters that need AI training. In summary, the first radio unit can request the central node to train AI model 1.

[0140] In one possible implementation, the request information 1 may include at least one of the following:

[0141] The name of AI model 1;

[0142] ◆Parameters to be trained for AI model 1;

[0143] ◆The identification of the first radio unit, or,

[0144] ◆The identifier of the cell corresponding to the first radio unit.

[0145] For example, the request information 1 includes the name of the AI model 1. The central node can determine the AI model requested to be trained by the first radio unit according to the name of the AI model 1.

[0146] For example, request information 1 includes the parameters to be trained for AI model 1. The central node can determine the parameters that need to be trained in AI model 1 based on the parameters to be trained, or in other words, the central node can determine the target parameters to be trained. This eliminates the need for the central node to train all parameters of AI model 1 and allows for targeted AI model training, effectively avoiding wasted power consumption at the central node.

[0147] For example, the request information 1 includes the identifier of the first radio unit. The central node can determine that the AI model 1 is the AI model that the first radio unit needs to apply based on the identifier of the first radio unit, or in other words, the central node can determine that the request information 1 comes from the first radio unit.

[0148] For example, the request information 1 includes the identifier of the cell corresponding to the first radio unit. The central node can determine the cell to which the AI model 1 is specifically applied based on the identifier of the cell corresponding to the first radio unit.

[0149] When the request information 1 includes the identifier of the first radio unit, the identifier of the cell corresponding to the first radio unit, and one or more of the parameters to be trained of the AI model 1, the central node can determine the corresponding AI model based on the association between the identifier of the first radio unit, the identifier of the cell corresponding to the first radio unit, and one or more of the parameters to be trained of the AI model 1, and the AI model 1. For details, see Table 1. The content shown in Table 1 is for example only and is not intended to be a final limitation.

[0150] Table 1

[0151] parameter AI models Radio unit identification 1 AI Model a Radio unit identification 2 AI Model b Cell ID 1 corresponding to the radio unit AI Model c Cell ID 2 corresponding to the radio unit AI Model Parameters to be trained 1 AI Model Parameters to be trained 2 AI Model

[0152] As shown in Table 1:

[0153] ◆The identification of the radio unit 1 and its associated AI model a;

[0154] ◆The identification of the radio unit 2 and its associated AI model b;

[0155] The cell ID 1 corresponding to the radio unit and its associated AI model c;

[0156] The cell ID 2 corresponding to the radio unit and its associated AI model d;

[0157] ◆Parameter 1 to be trained and its associated AI model e;

[0158] ◆Parameter 2 to be trained, and its associated AI model f.

[0159] Table 1 describes the relationship between a single parameter and an AI model, but does not limit the relationship between two or more parameters and an AI model. For example, if request information 1 includes the identifier of the radio unit and the parameters to be trained, the central node can use this information to determine the AI model to which the two are associated.

[0160] When the request information 1 includes the identifier of the radio unit, the identifier of the cell corresponding to the radio unit, and one or more of the parameters to be trained, the central node can determine the corresponding AI model according to the association relationship shown in Table 1.

[0161] For a description of the request information 1, please refer to Table 2 and Table 3. The contents shown in Table 2 and Table 3 are only examples and are not intended to be definitive.

[0162] Table 2

[0163] parameter describe name Used to determine the AI model that needs to be trained Parameters to be trained Used to determine the parameters that need to be trained in the AI model Radio unit identification Used to determine the radio unit proposed for AI model training Cell ID Used to determine the specific application area of the AI obtained after training

[0164] As shown in Table 2:

[0165] ◆Request information 1 includes a name, which is used to identify the AI model that needs to be trained;

[0166] ◆Request information 1 includes parameters to be trained, which is used to determine the parameters that need to be trained in the AI model;

[0167] ◆Request information 1 includes the identification of the radio unit, which is used to determine the radio unit for proposing AI model training;

[0168] ◆Request information 1 includes a cell identifier, which is used to determine the specific cell in which the trained AI model is applied.

[0169] The request information 1 may include one or more of the parameters listed in Table 2, which is not limited.

[0170] Table 3

[0171]

[0172] As shown in Table 3:

[0173] ◆Request information 1 includes the model name, which is used to indicate the AI model that needs to be trained. The model name can be determined based on the content of the corresponding description column;

[0174] ◆Request information 1 includes a network device identifier, which is used to determine the network device to which the radio unit proposing AI model training belongs.

[0175] The network device identifier may be represented as a gNB identifier;

[0176] ◆Request information 1 includes a cell identifier, which is used to indicate the cell to which the trained AI model needs to be applied. The cell indicated by the cell identifier belongs to multiple cells under the network device indicated by the aforementioned network device identifier;

[0177] ◆Request information 1 includes model program parameters (model program), which are used to represent special configurable parameter requirements during model training, for example, the parameters to be trained of AI model 1.

[0178] The request information 1 may include one or more of the parameters listed in Table 3, which is not limited.

[0179] The contents described in Table 2 and Table 3 above may be two different forms of request information 1, but are not limited to other forms.

[0180] In the embodiments of the present application, the parameters of the AI model can be divided into two categories: common parameters and characteristic parameters. Common parameters of an AI model can be understood as parameters shared by two or more AI models, while characteristic parameters of an AI model can be understood as parameters unique to the AI model, which can be used to distinguish different AI models.

[0181] In one possible implementation, the parameters to be trained of the AI model 1 may be one or both of the characteristic parameters of the AI model 1 and the common parameters of the AI model 1.

[0182] For example, the parameters to be trained of AI model 1 may be characteristic parameters of AI model 1. The central node may train only the characteristic parameters of AI model 1 and not the common parameters of AI model 1. This reduces the training overhead of the central node for AI model 1.

[0183] For example, the parameters to be trained for AI model 1 are the public parameters of AI model 1. The central node can train only the public parameters of AI model 1 and not the feature parameters of AI model 1. This can reduce the training overhead of the central node for AI model 1.

[0184] For example, the parameters to be trained of AI model 1 are the public parameters and characteristic parameters of AI model 1. The central node can train the characteristic parameters and public parameters of AI model 1.

[0185] In the embodiment of the present application, the common parameters of AI model 1 can be understood as the parameters shared by AI model 1 and AI model 2, and the characteristic parameters of AI model 1 can be understood as the parameters that distinguish AI model 1 from AI model 2. In other words, the difference or distinction between AI model 1 and AI model 2 can be reflected through the characteristic parameters of AI model 1 and the characteristic parameters of AI model 2.

[0186] By distinguishing between the common parameters and characteristic parameters of AI Model 1, the central node can reduce the training overhead of AI Model 1. For example, if AI Model 2 is a trained AI model and shares common parameters with AI Model 1, the central node does not need to train the common parameters of AI Model 1, but only the characteristic parameters of AI Model 1.

[0187] By indicating or reporting the parameters to be trained of AI model 1, the central node can train AI model 1 in a targeted manner. The central node does not need to train all parameters of AI model 1, which is conducive to reducing the training overhead of the central node for AI model 1.

[0188] In one possible implementation, the request information 1 may also include performance requirement information for the AI model 1.

[0189] The performance requirement information for AI model 1 can be understood as: the performance that the first radio unit needs to meet for the AI model that is actually required to be applied, or the performance parameter information that the AI model obtained after training the AI model 1 of the first radio frequency unit needs to meet, or the requirements put forward by the first radio unit on how the central node trains AI model 1, etc.

[0190] As a possible example, the performance requirement information for AI model 1 may include:

[0191] For example, the accuracy of AI model 1 obtained after training needs to meet the first threshold;

[0192] For another example, it is necessary to use a training data set with a data volume greater than a second threshold to train AI model 1;

[0193] For another example, the AI model 1 obtained after training needs to be suitable for channel detection tasks, etc.

[0194] By reporting the performance requirement information for AI model 1, the central node can train AI model 1 in a targeted manner based on the performance requirement information for AI model 1. This can avoid the situation where the AI model 1 obtained after training fails to meet the performance requirements of the first radio unit, thereby resulting in invalid training, and further increasing the central node's training overhead for AI model 1.

[0195] S502: The central node sends response information 1 to the first radio unit.

[0196] Correspondingly, the first radio unit receives the response information 1.

[0197] Response information 1 corresponds to request information 1, or in other words, response information 1 is used to respond to request information 1, or in other words, response information 1 is a form of expression of a response made by the central node to request information 1. For example, response information 1 is used to indicate to the first radio unit whether the central node agrees to train AI model 1.

[0198] Exemplarily, response information 1 is used to indicate that the central node agrees to train AI model 1.

[0199] As another example, response information 1 is used to indicate that the central node does not agree to train AI model 1.

[0200] In one possible implementation, the response information 1 may include at least one of the following:

[0201] The name of AI model 1;

[0202] ◆The identification of the first radio unit;

[0203] ◆Confirmation information indicating consent to train AI Model 1; or,

[0204] ◆Non-confirmation information, indicating that the AI model 1 is not approved for training.

[0205] For example, the response information 1 includes the name of the AI model 1. The first radio unit may determine that the response information 1 is used to respond to the request to train the AI model 1.

[0206] Optionally, the first radio unit can also determine based on this that the central node agrees to train the AI model 1.

[0207] For example, the response information 1 includes the identifier of the first radio unit. The identifier of the first radio unit can also be used by the first radio unit to determine that the response information 1 is used to respond to the request information 1.

[0208] Optionally, the first radio unit can also determine based on this that the central node agrees to train the AI model 1.

[0209] For example, the response information 1 includes confirmation information (such as ACK), so that the first radio unit can determine that the central node agrees to train the AI model 1.

[0210] For example, the response information 1 includes non-acknowledgement information (such as NACK), and the first radio unit can determine that the central node does not agree to train the AI model 1 based on this.

[0211] For a description of the response information 1, please refer to Table 4 and Table 5. The contents shown in Table 4 and Table 5 are only examples and are not intended to be final limitations.

[0212] Table 4

[0213] parameter describe name Used to indicate AI model 1 Confirmation Information Indicates consent to train AI model 1 Radio unit identification Used to indicate the first radio unit

[0214] As shown in Table 4:

[0215] ◆Response information 1 includes a name, which is used to indicate AI model 1. In this way, the first radio unit can determine that response information 1 is a response to the request to train AI model 1;

[0216] ◆Response message 1 includes confirmation information, which indicates consent to train AI model 1;

[0217] ◆Response information 1 includes the identification of the radio unit, which is used to determine the radio unit that proposes AI model training.

[0218] The response information 1 may include one or more of the parameters listed in Table 4, which is not limited.

[0219] Table 5

[0220]

[0221]

[0222] As shown in Table 5:

[0223] ◆Response information 1 includes a model name, which is used to indicate AI model 1. The model name can be determined based on the content of the corresponding description column;

[0224] ◆Response information 1 includes a network device identifier, which is used to determine the network device to which the radio unit proposing the AI model training belongs.

[0225] The network device identifier may be represented as a gNB identifier;

[0226] ◆Response information 1 includes ACK / NACK, which is used to indicate whether to agree to train AI model 1.

[0227] The response information 1 may include one or more of the parameters listed in Table 5, which is not limited.

[0228] In one possible implementation, response information 1 includes an ACK / NACK parameter, which can have different values. For example, a parameter value of 1 indicates that the AI model 1 agrees to be trained; a parameter value of 2 indicates that the AI model 1 does not agree to be trained due to insufficient resources; a parameter value of 3 indicates that the AI model 1 does not agree to be trained due to the lack of support for the AI model 1.

[0229] In one possible implementation, the parameter value used to indicate ACK / NACK in response information 1 is 3. When the first radio unit needs to train another AI model, the first radio unit can continue to send new request information to the central node. When the parameter value used to indicate ACK / NACK in response information 1 is 2, when the first radio unit needs to train another AI model, the first radio unit may not send new request information to the central node.

[0230] It should be noted that the contents described in Table 4 and Table 5 are two different forms of the response information 1, but are not limited to other forms.

[0231] In summary, by establishing a mechanism for interactive information request between the first radio unit and the central node, the embodiment of the present application can support improving the accuracy of the AI model applied by the radio unit.

[0232] Compared with the solution in which the first radio unit independently trains the AI model 1, the first radio unit can request the central node to help train the AI model 1 through the interaction of information request between the first radio unit and the central node. The training capability of the central node is better than the training capability of the first radio unit. By having the help of the central node to complete the training of the AI model 1, it can avoid the problem of insufficient model accuracy of the AI model 1 obtained after training due to insufficient training data, uneven data distribution, or a relatively single data source of the first radio unit.

[0233] Through the above-mentioned interaction mechanism, the central node can obtain the request information for training AI model 1 from the first radio unit, and the central node can determine whether it can help train AI model 1. When the central node determines that it can help train AI model 1, the accuracy of AI model 1 obtained after training by the central node will be better than the accuracy of AI model 1 obtained after training by the first radio unit.

[0234] Based on the above-mentioned interaction mechanism, the first radio unit can have the opportunity to request the central node to help train AI model 1. Compared with the first radio unit completing the training of AI model 1 independently, the training of AI model 1 is completed with the help of the central node. This can overcome the problem of insufficient model accuracy of AI model 1 obtained after training due to insufficient training data of the first radio unit, and thus support the improvement of the accuracy of the AI model specifically applied by the radio unit.

[0235] Combined with the following Figure 6 right Figure 5 The method shown is further described.

[0236] Figure 6 This is a schematic diagram of the interaction flow of another communication method in an embodiment of the present application. Figure 6 The method shown can be performed by RU1 (which can be an example of a first radio unit) and a central node (taking the first network unit as an example), or by modules and / or devices (such as chips or integrated circuits) with corresponding functions installed in RU1 and the first network unit. The following description takes RU1 and the first network unit as an example. Figure 6 As shown, the method includes:

[0237] S601. RU1 sends training data 1 (such as first training data) to a first network unit.

[0238] Correspondingly, the first network unit receives training data 1.

[0239] Training data 1 can be used to train one or both of the common parameters and characteristic parameters of AI model 1. For example, training data 1 is used to train the common parameters of AI model 1, or training data 1 is used to train the characteristic parameters of AI model 1, or training data 1 is used to train both the common parameters and characteristic parameters of AI model 1, without limitation.

[0240] When RU1 sends training data for training the characteristic parameters of AI model 1 to the central node, RU1 can complete the training of the common parameters of AI model 1. When RU1 sends training data for training the common parameters of AI model 1 to the central node, RU1 can complete the training of the characteristic parameters of AI model 1.

[0241] In addition, the common parameters of AI model 1 and the characteristic parameters of AI model 1 can be configured on RU1, or in other words, RU1 can independently determine the common parameters of AI model 1 and the characteristic parameters of AI model 1. The embodiments of the present application do not limit the manner or method by which RU1 determines the common parameters of AI model 1 and the characteristic parameters of AI model 1.

[0242] RU1 can determine whether the first network unit supports training of AI model 1 based on response information 1. When response information 1 is used to indicate that the first network unit agrees to train AI model 1, RU1 can initiate a model training process, such as initiating a process for collecting training data 1, for example, selecting a target user for a target scenario, and performing basic operations such as data collection and preprocessing on the target user to form a sample data format required for model training.

[0243] For a description of RU1 reporting training data 1 to the first network unit, please refer to Table 6. The content shown in Table 6 is only an example and is not a final limitation.

[0244] Table 6

[0245]

[0246] RU1 may report training data 1 to the first network unit in the content or form shown in Table 6.

[0247] S602. The first network unit trains AI model 1 according to training data 1 to obtain AI model 3.

[0248] For the description of the training process of the first network unit for the AI model 1, please refer to the existing process and will not be repeated here.

[0249] In one possible implementation, the first network unit can jointly train multiple AI models.

[0250] For example, a first network unit receives requests for training an AI model from multiple RUs;

[0251] For example, the first network unit receives request information from RU1 requesting to train AI model 1 and request information from RU2 requesting to train AI model 2.

[0252] When the first network unit determines that AI model 1 and AI model 2 can be trained, RU1 and RU2 send training data and training data 2 to the first network unit respectively. The first network unit aggregates the training data from multiple RUs to form a larger training data set, thereby obtaining more sufficient training data to train and optimize the model.

[0253] Exemplarily, the first network unit uses training data 1 and training data 2 to jointly train AI model 1 and AI model 2.

[0254] In one possible implementation, the first network unit can independently complete the training of AI model 1 and AI model 2. In this way, the problem of data transmission and communication costs can be avoided, and the efficiency and accuracy of model training and reasoning can be ensured.

[0255] In one possible implementation, the first network unit can also distribute the training tasks of AI model 1 and AI model 2 to multiple network elements (such as RU, CU, DU, etc.), and multiple network elements can jointly complete the training tasks of AI model 1 and AI model 2. This can improve computing efficiency and speed.

[0256] In one possible implementation, AI model 3 is obtained by the central node training AI model 1 based on training data 1 and training data 2.

[0257] Exemplarily, the first network unit trains AI model 1 based on training data 1 and training data 2. Training data 2 comes from RU 2. In other words, the first network unit obtains multiple training data from multiple RUs and uses the multiple training data to train AI model 1.

[0258] By using training data from other radio units to train AI model 1, the source, quantity, and type of data used for training AI model 1 can be broadened, thereby supporting the improvement of the accuracy of AI model 3.

[0259] In one possible implementation, the first network unit may first determine the common parameters between AI model 1 and AI model 2 and their respective characteristic parameters, and train the common parameters and characteristic parameters between AI model 1 and AI model 2 separately.

[0260] For example, the first network unit trains the common parameters between AI model 1 and AI model 2, and separately trains the characteristic parameters of AI model 1 and the characteristic parameters of AI model 2. When the first network unit needs to send the trained AI model to RU1 and RU2, the trained common parameters and the trained characteristic parameters can be combined and sent to the corresponding RUs.

[0261] By dividing the parameters of the AI model into common parameters and feature parameters and training them separately, the common parameters obtained after training can be applied to other AI models and the training overhead can be reduced. For example, the common parameters only need to be trained once, and the differentiated needs of the first AI model can also be met.

[0262] In one possible implementation, the request information 1 can also be used to request the central node to jointly train the AI model 1.

[0263] When the central node determines that joint training of AI model 1 is possible based on request information 1, the central node's training resources, and the central node's training capabilities, the central node can obtain training data from other RUs and use the training data of multiple RUs to train AI model 1. For details, please refer to the aforementioned content on the first network unit training AI model 1 and AI model 2, which will not be repeated here. In this way, the accuracy of AI model 3 can be effectively improved.

[0264] S603. The first network unit sends AI model 3 to RU1.

[0265] Correspondingly, RU1 receives AI model 3.

[0266] Through the above solution, when response information 1 indicates that the central node agrees to train AI model 1, the first radio unit can report training data 1 to the central node. The central node can use training data 1 to train AI model 1 and send the trained AI model 3 to the first radio unit. In this way, the first radio unit can obtain AI model 3 trained by the central node.

[0267] Optionally, the first network unit sending the AI model 3 to RU1 may include:

[0268] S603a. RU1 sends request information 2 to the first network unit.

[0269] Correspondingly, the first network unit receives request information 2. Request information 2 is used to request acquisition of AI model 3.

[0270] For a description of request information 2, please refer to Table 7. The content shown in Table 7 is only an example and is not a final limitation.

[0271] Table 7

[0272] parameter describe Model Name Used to indicate the AI model that needs to be obtained Network device identification gNB identification Cell ID Cell ID Required program parameters Indicates the relevant parameters of the model download request

[0273] As shown in Table 7:

[0274] ◆Request information 2 includes a model name, which is used to indicate the AI model to be obtained;

[0275] ◆ Request information 2 includes a network device identifier, which is used to determine the network device to which the radio unit requesting AI model training belongs. The network device identifier can be represented as a gNB identifier;

[0276] ◆ Request information 2 includes parameters including a cell identifier, which is used to indicate the cell to which the trained AI model needs to be applied. The cell indicated by the cell identifier belongs to multiple cells under the network device indicated by the aforementioned network device identifier;

[0277] ◆Request information 2 includes the program parameters to be downloaded (request programs), which are used to indicate the relevant parameters of the model download request.

[0278] S603b: The first network unit sends AI model 3 to RU1.

[0279] Correspondingly, RU1 receives AI model 3.

[0280] Based on the above information interaction, the first radio unit can obtain the trained AI model from the central node.

[0281] In one possible implementation, after RU1 reports or sends training data 1 to the first network element, it may send request information 2 to the first network element after a period of time. In this way, the signaling interaction overhead between RU1 and the first network element may be reduced.

[0282] In one possible implementation, after RU1 reports or sends training data 1 to the first network unit, it may then send request information 2 to the first network unit after receiving notification information from the first network unit indicating that AI model 1 training is complete. In this way, RU1 can request AI model 3 from the first network unit based on the notification from the first network unit. In this way, when the first network unit completes training of AI model 1, the first network unit can indicate to RU1 that AI model 1 training is complete, and RU1 can therefore determine that it can request AI model 3.

[0283] It should be noted that the central node can extract the trained feature parameters and common parameters of RU1 based on the actual configuration requirements of RU1, and test the extracted trained feature parameters and common parameters. Once they meet RU1's basic test accuracy requirements, the model can be delivered. For example, the central node delivers the trained AI model 3 to RU1. When the aforementioned request information 1 includes performance requirement information corresponding to AI model 1, the central node can test the trained AI model 3 based on the performance requirement information for AI model 1, and only deliver AI model 3 to RU1 when the performance requirement information for AI model 1 is met.

[0284] After receiving AI model 3, RU1 can perform basic management operations on AI model 3, such as format conversion and storage. When AI model 3 needs to be applied, RU1 can load AI model 3, collect the corresponding data required for inference, and perform model inference.

[0285] RU1 can also monitor the performance of AI model 3. When RU1 detects that the model accuracy of AI model 3 has deteriorated to the point where it cannot work, it can communicate with the central node to update the model. For details, see Figure 7 The content shown.

[0286] Through the above technical solution, the embodiment of the present application can support RU1 to obtain the AI model 3 obtained after training by the first network unit.

[0287] Combined with the following Figure 7 right Figure 6 The method shown is further described.

[0288] Figure 7 This is a schematic diagram of the interaction flow of another communication method according to an embodiment of the present application. Figure 7 The method shown can be performed by RU1 and the first network unit, or by modules and / or devices (e.g., chips or integrated circuits) with corresponding functions installed in RU1 and the first network unit. The following description takes RU1 and the first network unit as an example. Figure 7 As shown, the method includes:

[0289] S701. RU1 sends request information 3 to the first network unit.

[0290] Correspondingly, the first network unit receives the request information 3. The request information 3 is used to request to update the AI model 3.

[0291] S702. The first network unit updates and trains the AI model 3 to obtain an updated AI model 3.

[0292] For a description of how the first network unit updates the AI model 3, please refer to the existing process and will not be repeated here.

[0293] During the update process of the AI model 3, RU1 may also send training data for updating the AI model 3 to the first network unit.

[0294] S703. The first network unit sends the updated AI model 3 to RU1.

[0295] Accordingly, RU1 receives the updated AI model 3.

[0296] For example, the first network unit may send all parameters of the updated AI model 3 or the updated AI model 3 itself to RU1, or the first network unit may send updated parameters of the updated AI model 3 to RU1. In summary, RU1 can obtain the updated AI model 3 without limiting the specific implementation method.

[0297] Through the above technical solution, RU1 can obtain the updated AI model 3 from the first network unit.

[0298] Through the above information interaction, RU1 can complete the update process of AI model 3 and obtain the updated AI model 3 from the first network unit.

[0299] Figure 6 and Figure 7 The content shown is described using the first network unit as the central node as an example. The central node can also be DU, CU, second network unit, and third network unit, etc. When the central node is a network element other than the first network unit, the interaction between the central node and the first radio unit can refer to the interaction between the first network unit and RU1, and will not be repeated.

[0300] The following describes the device embodiments corresponding to the method embodiments of the present application. The following only briefly describes the device, and the specific implementation steps and details of the solution can be referred to the method embodiments above.

[0301] To implement the various functions of the method provided herein, the first radio unit and the central node may each include hardware structures and / or software modules, with the aforementioned functions implemented in the form of hardware structures, software modules, or a combination of hardware structures and software modules. Whether a particular function is implemented in the form of hardware structures, software modules, or a combination of hardware structures and software modules depends on the specific application and design constraints of the technical solution.

[0302] Figure 8 8 is a schematic block diagram of a communication device according to an embodiment of the present application. The communication device includes a processor 810 and a communication interface 820, which are interconnected via a bus 830. The communication device may be a central node or a first radio unit.

[0303] Optionally, the communication device may further include a memory 840. The memory 840 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or portable read-only memory (CD-ROM), and is used for related instructions and data.

[0304] The processor 810 may be one or more central processing units (CPUs). In the case where the processor 810 is a CPU, the CPU may be a single-core CPU or a multi-core CPU.

[0305] The processor 810 may be a signal processor, a chip, or other integrated circuit that can implement the method of the present application, or a portion of the circuitry used for processing functions in the aforementioned processor, chip, or integrated circuit.

[0306] The communication interface 820 may also be an input / output interface, which is used for inputting or outputting signals or data, and may also be an input / output circuit.

[0307] When the communication device is a central node, illustratively, the processor 810 is configured to perform the following operations: receive request information 1; send response information 1, etc.

[0308] When the communication device is a first radio unit, illustratively, the processor 810 is configured to perform the following operations: sending request information 1; receiving response information 1, etc.

[0309] The above contents are only for illustrative purposes. When the communication device is a central node or a first radio unit, it will be responsible for executing the methods or steps related to the central node or the first radio unit in the above method embodiments.

[0310] When the communication device is a central node or a first radio unit, the communication interface 820 may also be referred to as a transceiver.

[0311] The above description is merely an example description. For details, please refer to the contents shown in the above method embodiment.

[0312] Figure 8 The implementation of each operation in can also refer to Figures 5 to 7 The corresponding description of the method embodiment shown.

[0313] Figure 9 This is a schematic block diagram of another communication device according to an embodiment of the present application. This communication device can be a central node or a first radio unit, or a chip or module within the central node or the first radio unit, and is configured to implement the methods described in the above embodiments. This communication device includes an interface unit 910 and a processing unit 920. The interface unit 910 and the processing unit 920 are described below as examples.

[0314] The interface unit 910 may include a transmitting unit and a receiving unit. The transmitting unit is used to perform a transmitting operation of the communication device, and the receiving unit is used to perform a receiving operation of the communication device. For ease of description, this embodiment of the application combines the transmitting unit and the receiving unit into a single interface unit. This is described here as a unified description and will not be repeated later.

[0315] When the communication device is a central node, illustratively, the interface unit 910 is used to receive request information 1 and send response information 1, etc. The processing unit 920 is used to execute the content of the central node related to processing, coordination, etc. For example, the processing unit 920 is used to determine the response information 1.

[0316] When the communication device is a first radio unit, the interface unit 910 is illustratively configured to perform the following operations: sending request information 1; receiving response information 1; etc. The processing unit 920 is configured to execute the content of the first radio unit related to processing, coordination, etc. For example, the processing unit 920 is configured to determine the request information 1, etc.

[0317] The above contents are only for illustrative purposes. When the communication device is a central node or a first radio unit, it will be responsible for executing the methods or steps related to the central node or the first radio unit in the above method embodiments.

[0318] Optionally, the communication device further includes a storage unit 930, which is used to store a program or code for executing the aforementioned method.

[0319] Figure 8 and Figure 9 The device embodiment shown is for implementing Figures 5 to 7 The content described.

[0320] Figure 8 and Figure 9 The specific execution steps and methods of the device shown can refer to the contents described in the aforementioned method embodiment.

[0321] The present application also provides a chip, including a processor, for calling and executing instructions stored in a memory from the memory, so that a communication device equipped with the chip executes the methods in the above examples.

[0322] The present application also provides another chip, comprising: an input interface, an output interface, and a processor, wherein the input interface, the output interface, and the processor are connected via an internal connection path, and the processor is configured to execute code in a memory. When the code is executed, the processor is configured to execute the methods in the above examples. Optionally, the chip also includes a memory, which is configured to store computer programs or code.

[0323] The present application also provides a processor for coupling with a memory, and for executing the methods and functions involving a network device or a terminal device in any of the above embodiments.

[0324] In another embodiment of the present application, a computer program product including instructions is provided. When the computer program product is run on a computer, the method of the above embodiment is implemented.

[0325] The present application also provides a computer program. When the computer program is executed in a computer, the method of the aforementioned embodiment is implemented.

[0326] In another embodiment of the present application, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a computer, the method described in the above embodiment is implemented.

[0327] Those skilled in the art will appreciate that the units and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0328] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0329] In the several embodiments provided in this application, the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0330] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of the technical solutions of this embodiment.

[0331] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0332] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0333] The above is only a specific implementation of the embodiment of the present application, but the scope of protection of the embodiment of the present application is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the embodiment of the present application, and they should be included in the scope of protection of the embodiment of the present application. Therefore, the scope of protection of the embodiment of the present application should be based on the scope of protection of the claims.

Claims

1. A communication method, characterized in that: The method is applied to a first radio unit, and includes: Sending a first request message, wherein the first request message requests training of a first artificial intelligence model; Receive response information, where the response information indicates whether to agree to train the first artificial intelligence model.

2. The method according to claim 1, characterized in that The response information indicates consent to train the first artificial intelligence model, and the method further includes: Sending first training data, where the first training data is used to train at least one of the common parameters and characteristic parameters of the first artificial intelligence model, The common parameters of the first artificial intelligence model are parameters shared by the first artificial intelligence model and the second artificial intelligence model. The characteristic parameters of the first artificial intelligence model are parameters that distinguish the first artificial intelligence model from the second artificial intelligence model; A third artificial intelligence model is received, where the third artificial intelligence model is a model obtained by training the first artificial intelligence model.

3. The method according to claim 2, characterized in that The receiving of the third artificial intelligence model comprises: Sending a second request message, wherein the second request message requests obtaining the third artificial intelligence model; Receive the third artificial intelligence model.

4. The method according to claim 3, characterized in that Before sending the second request information, the method further includes: Receive notification information, where the notification information indicates that the training of the first artificial intelligence model is completed.

5. The method according to any one of claims 2 to 4, characterized in that The method further comprises: sending a third request message, wherein the third request message requests updating the third artificial intelligence model; Receive the updated third artificial intelligence model.

6. The method according to any one of claims 1 to 5, characterized in that The first request information includes at least one of the following: The name of the first artificial intelligence model, the parameters to be trained of the first artificial intelligence model, the identifier of the first radio unit, or the identifier of the cell corresponding to the first radio unit.

7. The method according to claim 6, characterized in that The parameter to be trained is at least one of a characteristic parameter and a common parameter of the first artificial intelligence model.

8. The method according to any one of claims 1 to 7, characterized in that The response information includes at least one of the following: the name of the first artificial intelligence model, the identifier of the first radio unit, confirmation information or non-confirmation information; The confirmation information indicates consent to train the first artificial intelligence model, The non-confirmation information indicates that the training of the first artificial intelligence model is not agreed to.

9. The method according to any one of claims 1 to 8, characterized in that The first request information includes performance requirement information of the first radio unit for the first artificial intelligence model.

10. The method according to any one of claims 2 to 9, characterized in that The third model is a model obtained by training the first model based on the first training data and the second training data. The second training data comes from a second radio unit, and the second radio unit is different from the first radio unit.

11. A communication method, characterized in that: The method is applied to a central node and includes: Receiving a first request message, wherein the first request message requests training of a first artificial intelligence model; Send a response message, where the response message indicates whether the first artificial intelligence model is agreed to be trained.

12. The method according to claim 11, characterized in that The response information indicates consent to train the first artificial intelligence model, and the method further includes: receiving first training data, where the first training data is used to train at least one of a common parameter and a characteristic parameter of the first artificial intelligence model; The common parameters of the first artificial intelligence model are parameters shared by the first artificial intelligence model and the second artificial intelligence model. The characteristic parameters of the first artificial intelligence model are parameters that distinguish the first artificial intelligence model from the second artificial intelligence model; Training the first artificial intelligence model according to the first training data to obtain a third artificial intelligence model; Send the third artificial intelligence model.

13. The method according to claim 12, characterized in that The sending of the third artificial intelligence model includes: receiving a second request message, wherein the second request message requests obtaining the third artificial intelligence model; Send the third artificial intelligence model.

14. The method according to claim 13, wherein: Before receiving the second request information, the method further includes: Send a notification message, where the notification message indicates that the training of the first artificial intelligence model is completed.

15. The method according to any one of claims 12 to 14, characterized in that The method further comprises: receiving a third request message, wherein the third request message requests updating the third artificial intelligence model; Send the updated third artificial intelligence model.

16. The method according to any one of claims 11 to 15, characterized in that The first request information includes at least one of the following: The name of the first artificial intelligence model, the parameters to be trained of the first artificial intelligence model, the identifier of the first radio unit, or the identifier of the cell corresponding to the first radio unit; The first radio unit is a radio unit that applies the first artificial intelligence model.

17. The method according to claim 16, characterized in that The parameter to be trained is at least one of a characteristic parameter and a common parameter of the first artificial intelligence model.

18. The method according to any one of claims 11 to 17, characterized in that The response information includes at least one of the following: the name of the first artificial intelligence model, the identifier of the first radio unit, confirmation information or non-confirmation information; The confirmation information indicates consent to train the first artificial intelligence model, The non-confirmation information indicates that the training of the first artificial intelligence model is not agreed; The first radio unit is a radio unit to which the first model is applied.

19. The method according to any one of claims 11 to 18, characterized in that The first request information includes performance requirement information for the first artificial intelligence model.

20. The method according to any one of claims 12 to 19, characterized in that The step of training the first artificial intelligence model according to the first training data to obtain a third model includes: Training the first artificial intelligence model according to the first training data and the second training data to obtain the third model; The second training data comes from a second radio unit, and the second radio unit is different from the first radio unit.

21. The method according to any one of claims 12 to 20, characterized in that The step of training the first artificial intelligence model according to the first training data to obtain a third artificial intelligence model includes: Determining common parameters of the first artificial intelligence model and characteristic parameters of the first artificial intelligence model; Based on the first training data, the common parameters of the first artificial intelligence model and the characteristic parameters of the first artificial intelligence model are trained respectively.

22. A communication system, characterized in that: include: First radio unit and central node: The first radio unit is used to send a first request message to the central node, where the first request message is used to request training of a first artificial intelligence model; The central node is used to receive the first request information and send a response information to the first radio unit, where the response information indicates whether the first artificial intelligence model is agreed to be trained.

23. A communication device, characterized in that: comprising a processor configured to, by executing computer programs or instructions, or by executing logic circuits, causing the communication device to perform the method according to any one of claims 1 to 10; or, The communication device is caused to execute the method according to any one of claims 11 to 21.

24. A communication device, characterized in that: It includes a logic circuit and an input / output interface, wherein the input / output interface is used to input and / or output signals. The logic circuit is configured to execute the method according to any one of claims 1 to 10; or The logic circuit is configured to execute the method according to any one of claims 11 to 21.

25. A computer-readable storage medium, characterized in that The computer readable storage medium stores a computer program or instruction. When the computer program or instruction is executed on a computer, causing the method of any one of claims 1 to 10 to be performed; or, The method according to any one of claims 11 to 21 is performed.

26. A computer program product, characterized in that Contains instructions that, when executed on a computer, causing the method of any one of claims 1 to 10 to be performed; or, The method according to any one of claims 11 to 21 is performed.

27. A chip system, characterized in that: The chip system includes a processor, a memory and an input / output port, wherein the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory. so that the processor performs the method according to any one of claims 1 to 10; or, so that the processor executes the method according to any one of claims 11 to 21.

Images